Metal-Ligand Co-operativity: Catalysis and the Pincer-Metal Platform 9783030689162, 3030689166

Citation preview

Topics in Organometallic Chemistry  68

Gerard van Koten Karl Kirchner Marc-Etienne Moret   Editors

Metal-Ligand Co-operativity Catalysis and the Pincer-Metal Platform

68

Topics in Organometallic Chemistry Series Editors Matthias Beller, Leibniz-Institut für Katalyse e.V., Rostock, Germany Pierre H. Dixneuf, Université de Rennes 1, Rennes CX, France Jairton Dupont, UFRGS, Porto Alegre, Brazil Alois Fürstner, Max-Planck-Institut fur Kohlenforschung, Mülheim, Germany Frank Glorius, WWU Münster, Münster, Germany Lukas J. Gooßen, Ruhr-Universität Bochum, Bochum, Germany Steven P. Nolan, Ghent University, Ghent, Belgium Jun Okuda, RWTH Aachen University, Aachen, Germany Luis A. Oro, University of Zaragoza-CSIC, Zaragoza, Spain Michael Willis, University of Oxford, Oxford, UK Qi-Lin Zhou, Nankai University, Tianjin, China

Aims and Scope The series Topics in Organometallic Chemistry presents critical overviews of research results in organometallic chemistry. As our understanding of organometallic structure, properties and mechanisms increases, new ways are opened for the design of organometallic compounds and reactions tailored to the needs of such diverse areas as organic synthesis, medical research, biology and materials science. Thus the scope of coverage includes a broad range of topics of pure and applied organometallic chemistry, where new breakthroughs are being achieved that are of significance to a larger scientific audience. The individual volumes of Topics in Organometallic Chemistry are thematic. Review articles are generally invited by the volume editors. All chapters from Topics in Organometallic Chemistry are published Online First with an individual DOI. In references, Topics in Organometallic Chemistry is abbreviated as Top Organomet Chem and cited as a journal. More information about this series at http://www.springer.com/series/3418

Gerard van Koten • Karl Kirchner • Marc-Etienne Moret Editors

Metal-Ligand Co-operativity Catalysis and the Pincer-Metal Platform

With contributions by J. G. de Vries
D. A. Ekanayake
D. Gelman
H. Guan
B. Guo
E. Hey-Hawkins
K. Hurej
S. Kim
Y.-E. Kim
K. Kirchner
A. Kumar
L. Latos-Grażyński
Y. Lee
D. Milstein
M.-E. Moret
E. Otten
A. Singh
M. R. Tiddens
J. I. van der Vlugt
S. Weber
H.-T. Zhang
M.-T. Zhang

Editors Gerard van Koten Organic Chemistry & Catalysis Utrecht University Utrecht, The Netherlands

Karl Kirchner Institute of Applied Synthetic Chemistry Vienna University of Technology Vienna, Austria

Marc-Etienne Moret Organic Chemistry and Catalysis, Debye Institute for Nanomaterials Science Utrecht University Utrecht, The Netherlands

ISSN 1436-6002 ISSN 1616-8534 (electronic) Topics in Organometallic Chemistry ISBN 978-3-030-68915-5 ISBN 978-3-030-68916-2 (eBook) https://doi.org/10.1007/978-3-030-68916-2 © Springer Nature Switzerland AG 2021 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

Directly after the incipient finding of the PCP-pincer platform, as a monoanionic, tridentate ligand in organometallic chemistry, in the late 1970s, the novel pincermetal compounds found extensive use as catalysts in homogeneous catalysis, opening new pathways in, for example, organic synthesis. Initially, the pincer motif comprised a well-defined Donor-C(anion)-Donor (DCD) ligand architecture consisting of a central, covalent metal-carbon, bond, often from a sp2-C(aryl) donor, and two additional interactions with the neutral donor heteroatom (D or D0 ) of two flanking substituents. In parallel, the full potential of the planar DCD, potentially 6-electron donating pincer motif was explored. The metal, the nature of the donor sites D, and type of central two-electron donor atom, from C to N, Si, etc. was varied. Moreover, it appeared that a modular approach for the synthesis of ligands with architectures, supporting a planar DCD coordination motif, could be followed as well as its subsequent (cyclo-)metallation. Much of this development was spurred by the, often, surprisingly high thermal and kinetic stability of the resulting DCD-pincer metal complexes. This earlier work is partly covered by the “Topics in Organometallic Chemistry Series,” Volume 40 (Organometallic Pincer Chemistry) and Volume 54 (The Privileged Pincer-Metal Platform: Coordination Chemistry and Applications). The greater part of this earlier research involves chemistry in which the pincer ligand acts as innocent ligand: i.e., it is merely the mutual speciation and electronic properties of the three 2e-donor sites of the pincer ligand that is playing its role in affecting the metal properties. Nevertheless, occasionally, it was already observed that incidentally electronic communication along the central metal-(anionic) donor site is playing a distinct role in modulating processes at the metal center. The research presented in this volume showcases research in which the tridentate pincer motif, rather than behaving as an “innocent ligand,” is playing an active role, i.e., metal–ligand cooperation (MLC) occurs involving profound, and often reversible, changes in the binding and electronic situation between the pincer platform and the metal site. A first demonstration of this “non-innocent” behavior of the pincer ligand in the pincer-metal platform was the discovery by Milstein and coworkers of a v

vi

Preface

new mode of reversible de-aromatization/aromatization reactions of the (pyridine) pincer-metal platform. This finding opened novel routes to catalytic bond activation of otherwise unreactive small molecules such as methane, carbon dioxide, and N2O. In much of this chemistry the pincer-metal grouping of the catalytic species forms a template on which the substrate is activated and subsequently transferred into product with regeneration of the pincer catalyst. Recent developments are discussed in the chapter “Recent Advances in the Applications of Metal-Ligand Cooperation via Dearomatization and Aromatization of Pincer Complexes.” The exceptional stability of PCP- and PNP-pincer ligands has inspired researchers in the field to expand the range of central units to more weakly binding and/or reactive units such as amides (R2N–M), phosphide (R2P–M), boranes (R3B305 nm) in the presence of 1 atm of CO2 and 1 equivalent of NEt3 results in the formation of complex 21 in around 50% spectroscopic yield (Φ410 nm ¼ 2.1%). In order to develop a sustainable rWGS process, hydrogen gas was employed as a reductant. Interestingly, photolysis of complex 21 (λ >305 nm) in the presence of

Recent Advances in the Applications of Metal-Ligand Cooperation via. . .

11

Scheme 12 Photodriven rWGS reaction cycle involving metal-ligand cooperation

1 bar of H2 at room temperature results in the selective regeneration of complex 20 (Φ337 nm ¼ 0.5%). The reaction proceeds via heterolysis of H2 via metal-ligand cooperation in a way that the proton gets transferred to the pincer backbone and hydride to the metal. ΔG0298K for this decarbonylation branch of the cycle (21 + H2 ! 20 + CO) was calculated to be +3.0 kJ/mol suggestive of this step to be almost thermoneutral, whereas the DFT computations indicate a strongly endergonic nature for the other part of the cycle which is the reaction of complex 20 with CO2 to form complex 21 and H2O (ΔG0298K ¼ +25.3 kJ/mol).

3 Catalytic Transformation of N2O via Metal-Ligand Cooperation Nitrous oxide (N2O) is a potent greenhouse gas and exhibits 300 times greater warming effect than CO2 [68, 69]. Due to its harmful environmental effects’ catalytic degradation, reduction or transformation of N2O to useful chemicals has drawn significant attraction recently [70, 71]. Several catalytic reactions where N2O is used as O-atom donors, oxidants and N-atom donors have been reported in the literature [71]. Being interested in the catalytic transformations based on (de)hydrogenation reactions, we have recently utilized the tool of metal-ligand cooperation for the catalytic hydrogenation of N2O to N2 and H2O [72]. A related work was reported by Grützmacher where N2O was employed as the hydrogen acceptor for the dehydrogenative coupling of alcohols to esters or acids [73]. A rhodium catalyst

12

A. Kumar and D. Milstein

Scheme 13 Catalytic ‘O’ donor transformation of N2O catalysed by ruthenium pincer complexes

capable of functioning through metal-ligand cooperation was used for this transformation. We discovered that using the ruthenium PNP hydride complex 22 (~2 mol %), almost quantitative hydrogenation of N2O was obtained at 65  C in 48 h [72]. The products, N2 was obtained in 83% yield (by GC) and H2O was detected in 96% yield by the 1H NMR spectroscopy (Scheme 13-1). A plausible catalytic cycle for this transformation is proposed in Scheme 14. A mono oxygen transfer from N2O to the dihydride ruthenium complex 22 results in formation of the hydrido-hydroxo ruthenium complex 23 that reversibly eliminates water through metal-ligand cooperation to form the dearomatized ruthenium complex 24. Complex 24, again via metal-ligand cooperation, reversibly reacts with hydrogen to regenerate the dihydride complex 22. Complexes 23 and 24 also catalyse the hydrogenation of nitrous oxide to N2 and H2O, supporting the mechanism proposed in Scheme 14. Poater [74] and Xie [75] have independently performed DFT calculations on this system to understand the reaction mechanism in detail. Xie performed molecular orbital calculations on complex 22 and N2O to understand the activation of N2O by this complex [75]. Calculations suggest that the activation of N2O occurs through orbital interactions between the LUMO of N2O and the HOMO-6 of

Recent Advances in the Applications of Metal-Ligand Cooperation via. . .

13

Scheme 14 Plausible catalytic cycle for ruthenium-catalysed hydrogenation of N2O

complex 22. On the basis of the HOMO/LUMO gap and NBO analysis, it is suggested that N2O acts as an electrophile in which the central nitrogen atom bears the maximum positive charge. The spherical property of the 1S orbital suggests that the hydride moiety of complex 22 can be easily attacked by an external electrophile such as N2O. Both the calculations by Poater and Xie suggest that H2O plays an important role in lowering the barrier of the transition state during the transformation of complex 22 to complex 23 through a transition state TS-1 shown in Scheme 14. The role of water in facilitating the metal-ligand cooperation in similar systems has also been suggested by us earlier [76, 77]. These complexes were also employed for the homogeneous ‘O’ atom transfer of nitrous oxide into silanes [72]. Using 1 mol% of catalyst 24 and 50 psi of nitrous oxide, silanes were converted to silanols and silyl ethers (Scheme 13-2). In another direction, N2O and CO are both environmentally harmful gases, and therefore their degradation in one step to produce CO2 and N2 has attracted significant attention recently [78, 79]. In this direction, we also reported a highly efficient ruthenium catalyst 25 for the catalytic reaction of N2O and CO to produce N2 and CO2 (Scheme 13-3) [80]. Complex 25 at room temperature slowly converts to complex 28 via activation of the pyridyl C–H bond through metal-ligand cooperation. Complex 28, although much slower, was also found to catalyse the reaction of N2O and CO. A plausible mechanism as depicted in Scheme 15 was proposed based on experimental observations. Nucleophilic attack by hydride ligand on N2O results in the O-atom transfer from N2O to the Ru–H bond forming a Ru-OH complex (26 or 29). Intramolecular nucleophilic attack of the hydroxo ligand in 26 or 29 on the adjacent carbonyl group forms a Ru-COOH complex (27 or 30) that in the presence of CO undergoes beta hydride elimination to regenerate the ruthenium hydride

14

A. Kumar and D. Milstein

Scheme 15 Proposed mechanism for the ruthenium-catalysed reaction of CO with N2O to form CO2 and N2

complex (25 or 28). A DFT calculation was performed by Xie, Fan and co-workers to understand the mechanism in detail [75]. Their calculation also showed three steps in the mechanistic cycle – N2O activation, CO insertion and CO2 release where the activation of N2O was found to be the rate-limiting step. The N2O activation barrier for complex 25 was found to be lower than that for complex 28 explaining the lower catalytic activity of complex 28. Interestingly, the DFT calculations suggested that the CO2 release is not via the beta-hydride elimination path but by a lower energy pathway that goes via protonation of the nitrogen atom of the bipyridine ligand and then proton transfer from ligand to ruthenium.

4 Template Catalysis via Metal-Ligand Cooperation In most reactions which involve metal-ligand cooperation, both the metal and ligand play active roles in bond making and breaking of substrates. Recently, we have discovered catalysis by pincer complexes based on metal-ligand cooperation by aromatization-dearomatization in which most of the catalytic steps occur at the pincer ligand, while the metal has mostly a structural role. This type of metalligand cooperation can greatly facilitate the Michael reaction of nitriles with α,β-unsaturated carbonyl compounds where the role of metal is to support the pincer ligand and preorganize the nitrile and the α,β-unsaturated carbonyl compounds for their reaction under mild conditions [81, 82]. As the pincer complex offers a suitable template for ligand-based reversible C–C bond forming reaction, we have categorized this type of catalytic transformation under ‘template catalysis’. The template catalysis of benzyl nitriles with the α,β-unsaturated carbonyl compounds was first reported using a dearomatized rhenium PNP pincer complex 31 [82], and recently an analogous manganese PNP complex 32 was found to be much more efficient, enabling reaction of simple nitriles at room temperature and in

Recent Advances in the Applications of Metal-Ligand Cooperation via. . .

15

Table 1 Substrate scope for the manganese-catalysed Michael addition of aliphatic nitriles to ethyl acrylatea

Entry 1 2c 3 4 5 6

Nitrile

Time (h) 20 40 6 40 40 12

Conversion (%) 69 90 >99 57 72 >99

Yield (%) 26(21)b 89(83) 93(82) 48(29) 67(48) 94(84)

a

Yields are determined by 1H NMR spectroscopy; isolated yields are given in parentheses The double addition product from the reaction of the formed mono addition product to ethyl acrylate was observed as a side product in 20(17)% yield c MeCN is used as a solvent b

absence of base with low catalytic loading, and has wide substrate scope [81]. Using 0.5 mol% of the dearomatized manganese PNP complex 32, at ambient temperature a variety of aliphatic nitriles were reacted with ethyl acrylate to afford the Michael addition products (Table 1) [81]. Under similar catalytic conditions, several α,β-unsaturated carbonyl compounds, for example, various acrylates, methyl crotonate and cyclohex-2-enone (Table 2, entries 18), were also found to be suitable acceptors. The reaction tolerates ketones (Table 2, entry 2) and fluorinated esters (Table 2, entry 3). Lower product yields were obtained in the case of cyclohex-2-en1-one and trans-methyl crotonate (Table 2, entries 7 and 8). The mechanism of this new template catalysis was investigated by both experiments and computation. Dearomatized PNP pincer complexes of both rhenium (31) and manganese (32) can reversibly activate nitrile to form [1,3]-addition product via C–C and M–N (M ¼ Re or Mn) bond formation (Scheme 16). Thus formed metalketimido complexes can reversibly tautomerize to form metal-enamido complexes if the nitrile moiety has an adjacent methylene group. DFT calculations reveal that in the case of rhenium, formation of enamido complex is either exergonic (for PhCH2CN) or isoergonic (for EtCN), whereas in the case of manganese, the formation of enamido complex via tautomerization is endergonic and is suggested to be assisted by the interaction of traces of water. A mechanism for the Michael addition of nitriles with the carbonyl compounds using the dearomatized manganese pincer complex 32 as a catalyst is depicted in Scheme 17 [81]. The key to this reaction is reversible C–C bond formation of the nitrile with the ligand, as experimentally observed, imparting on the nitrile reactivity modes of enamines. Thus, complex 32 activates the nitrile by forming C–C and M–N bonds via metal-ligand cooperation, generating a ketimido complex 33 that undergoes water-assisted tautomerization to form an enamido complex 34. As

2

26

3

4b

Time (h) 5

2

Acceptor

2

Entry 1

93 (71)

>99

52

86 (84)

>99

58

Yield (%) 93 (71)

Conversion (%) >99

Product

Table 2 Substrate scope for the manganese-catalysed Michael addition of propionitrile to α,β-unsaturated carbonyl compounds (Michael acceptors)a

16 A. Kumar and D. Milstein

5

24

7b

8b

33

>99

18

93 (90)

>99

40

92 (92)

>99

Yields are determined by 1H NMR spectroscopy; isolated yields are given in parentheses No solvent was used

5

6b

b

a

12

5b

Recent Advances in the Applications of Metal-Ligand Cooperation via. . . 17

18

A. Kumar and D. Milstein

Scheme 16 Reversible activation of nitrile by metal-ligand cooperation

Scheme 17 Proposed mechanism for the catalytic Michael addition of nitriles to carbonyl compounds

Recent Advances in the Applications of Metal-Ligand Cooperation via. . .

19

Table 3 Substrate scope of oxa- and aza-Michael addition of unsaturated nitriles

Reaction conditions: crotononitrile or 2-pentenenitrile (2.0 mmol), alcohol or amine (2.0 mL), complex 3 (0.0020 mmol, 0.1 mol%), r.t., 24 h. Yields determined by 1H NMR with mesitylene as the internal standard a Isolated yields b 0.5 mol% complex 3 was used c Benzylamine (1.0 mL) was used

characteristic of enamines, the enamine moiety of complex 34 undergoes C–C bond formation with the α,β-unsaturated carbonyl compound, forming a bound imine (35, 36), which undergoes C–C rupture, generating the Michael-addition product and complex 32. It is noteworthy that Michael addition of simple nitriles to α,β-unsaturated carbonyl compound normally requires a strong base and elevated temperatures. Utilizing the tool of MLC by dearomatization/aromatization, Otten and de Vries reported the useful oxa-Michael addition reaction of alcohols to unsaturated nitriles using a dearomatized Ru-PNN complex under mild conditions [83]. However, much lower yield was observed for the corresponding aza-Michael addition reactions of amines with unsaturated nitriles even at longer reaction time and at higher temperatures. We have recently discovered that the analogous dearomatized Mn-PNN complex 3 can catalyse both oxa- and aza-Michael addition reactions under mild

20

A. Kumar and D. Milstein

and neutral conditions with the catalytic TON even higher than that of ruthenium [84]. Using 0.1 mol% of the manganese pincer complex 3, Michael addition products of a variety of alcohols and amines with the electrophilic unsaturated nitriles were synthesized (Table 3). Based on experimental observations, a plausible mechanism is proposed for the manganese-catalysed oxa and aza-Michael addition reaction (Scheme 18). Complex

Scheme 18 Proposed mechanism for the manganese-catalysed oxa- and aza-Michael addition reaction of unsaturated nitriles (X ¼ O or NH)

Recent Advances in the Applications of Metal-Ligand Cooperation via. . .

21

3 first tautomerizes to form complex 3’. Conjugated unsaturated nitrile I’ reacts with complex 3’ to form ketimido complex C which is in equilibrium with an enamido complex D formed by the reaction of an unconjugated unsaturated nitrile II’ with complex 3’. Nucleophilic addition of a nucleophile (alcohol or amine) to the ketimido complex C results in the formation of a new enamido complex E that can tautomerize to form an imine complex F. Complex F undergoes C–C cleavage reaction to release the product and regenerate the dearomatized complex 3’.

5 Summary and Outlook In conclusion, metal-ligand cooperation operated by the dearomatization/aromatization mode of pincer complexes has allowed facile reversible activation of strong bonds and ultimately led to the development of new catalytic transformations that are atom-economic and environmentally benign. As the pincer ligands in these transformations are non-innocent and the bond activation or catalysis is mediated by a synergy between metal and ligand and not only by the metal, without overall change in the metal oxidation state, it has been possible to utilize several transition metals, especially earth-abundant metals, in place of noble metals making the catalytic processes more sustainable. Using the same concept new ligand designs may achieve appropriate synergy with the main group elements and perform reversible bond activation offering new platform to discover new catalytic reactions free from transition metals. Development of bimetallic pincer complexes with the ability to perform metal-ligand cooperation can also bring new modes of reactivity due to additional possibility of metal-metal cooperation.

References 1. van Koten G (1989) Pure Appl Chem 61:1681–1694 2. van der Vlugt JI, Reek JNH (2009) Angew Chem Int Ed 48:8832–8846 3. Selander N, Szabó KJ (2011) Chem Rev 111:2048–2076 4. Peris E, Crabtree RH (2018) Chem Soc Rev 47:1959–1968 5. Choi J, MacArthur AHR, Brookhart M, Goldman AS (2011) Chem Rev 111:1761–1779 6. Albrecht M, van Koten G (2001) Angew Chem Int Ed 40:3750–3781 7. van der Boom ME, Milstein D (2003) Chem Rev 103:1759–1792 8. Gunanathan C, Milstein D (2014) Chem Rev 114:12024–12087 9. Gunanathan C, Milstein D (2011) Bond activation by metal-ligand cooperation: design of “Green” catalytic reactions based on aromatization-dearomatization of pincer complexes. In: Ikariya T, Shibasaki M (eds) Bifunctional molecular catalysis. Springer, Berlin, pp 55–84 10. Khusnutdinova JR, Milstein D (2015) Angew Chem Int Ed 54:12236–12273 11. Milstein D (2015) PhilosTrans Royal Soc A 373:20140189 12. Gunanathan C, Milstein D (2011) Acc Chem Res 44:588–602 13. Li H, Hall MB (2015) ACS Catal 5:1895–1913 14. Mukherjee A, Milstein D (2018) ACS Catal 8:11435–11469

22

A. Kumar and D. Milstein

15. Schirmer W, Flörke U, Haupt H-J (1987) Zeitschrift für anorganische und allgemeine Chemie 545:83–97 16. Benito-Garagorri D, Kirchner K (2008) Acc Chem Res 41:201–213 17. Deibl N, Kempe R (2016) J Am Chem Soc 138:10786–10789 18. Rösler S, Obenauf J, Kempe R (2015) J Am Chem Soc 137:7998–8001 19. Li H, Zheng B, Huang K-W (2015) Coord Chem Rev 293-294:116–138 20. He L-P, Chen T, Xue D-X, Eddaoudi M, Huang K-W (2012) J Organomet Chem 700:202–206 21. He L-P, Chen T, Gong D, Lai Z, Huang K-W (2012) Organometallics 31:5208–5211 22. Chen T, He L-P, Gong D, Yang L, Miao X, Eppinger J, Huang K-W (2012) Tetrahedron Lett 53:4409–4412 23. Qu S, Dang Y, Song C, Wen M, Huang K-W, Wang Z-X (2014) J Am Chem Soc 136:4974–4991 24. Gonçalves TP, Huang K-W (2017) J Am Chem Soc 139:13442–13449 25. Zhang J, Leitus G, Ben-David Y, Milstein D (2005) J Am Chem Soc 127:10840–10841 26. Gunanathan C, Shimon LJW, Milstein D (2009) J Am Chem Soc 131:3146–3147 27. Balaraman E, Khaskin E, Leitus G, Milstein D (2013) Nat Chem 5:122–125 28. Kumar A, Espinosa-Jalapa NA, Leitus G, Diskin-Posner Y, Avram L, Milstein D (2017) Angew Chem Int Ed 56:14992–14996 29. Gunanathan C, Ben-David Y, Milstein D (2007) Science 317:790–792 30. Gnanaprakasam B, Zhang J, Milstein D (2010) Angew Chem Int Ed 49:1468–1471 31. Mukherjee A, Nerush A, Leitus G, Shimon LJW, Ben David Y, Espinosa Jalapa NA, Milstein D (2016) J Am Chem Soc 138:4298–4301 32. Gunanathan C, Milstein D (2008) Angew Chem Int Ed 47:8661–8664 33. Daw P, Kumar A, Espinosa-Jalapa NA, Ben-David Y, Milstein D (2019) J Am Chem Soc 141:12202–12206 34. Khusnutdinova JR, Ben-David Y, Milstein D (2013) Angew Chem Int Ed 52:6269–6272 35. Khusnutdinova JR, Ben-David Y, Milstein D (2014) J Am Chem Soc 136:2998–3001 36. Langer R, Leitus G, Ben-David Y, Milstein D (2011) Angew Chem Int Ed 50:2120–2124 37. Zhang J, Leitus G, Ben-David Y, Milstein D (2006) Angew Chem Int Ed 45:1113–1115 38. Espinosa-Jalapa NA, Nerush A, Shimon LJW, Leitus G, Avram L, Ben-David Y, Milstein D (2017) Chem A Eur J 23:5934–5938 39. Balaraman E, Gnanaprakasam B, Shimon LJW, Milstein D (2010) J Am Chem Soc 132:16756–16758 40. Garg JA, Chakraborty S, Ben-David Y, Milstein D (2016) Chem Commun 52:5285–5288 41. Kumar A, Janes T, Espinosa-Jalapa NA, Milstein D (2018) Angew Chem Int Ed 57:12076–12080 42. Balaraman E, Gunanathan C, Zhang J, Shimon LJW, Milstein D (2011) Nat Chem 3:609–614 43. Das UK, Kumar A, Ben-David Y, Iron MA, Milstein D (2019) J Am Chem Soc 141 (33):12962–12966 44. Balaraman E, Ben-David Y, Milstein D (2011) Angew Chem Int Ed 50:11702–11705 45. He M, Sun Y, Han B (2013) Angew Chem Int Ed 52:9620–9633 46. Quadrelli EA, Centi G, Duplan J-L, Perathoner S (2011) ChemSusChem 4:1194–1215 47. Olah GA (2005) Angew Chem Int Ed 44:2636–2639 48. Cokoja M, Bruckmeier C, Rieger B, Herrmann WA, Kühn FE (2011) Angew Chem Int Ed 50:8510–8537 49. Braunstein P, Matt D, Dusausoy Y, Fischer J, Mitschler A, Ricard L (1981) J Am Chem Soc 103:5115–5125 50. LeBlanc FA, Berkefeld A, Piers WE, Parvez M (2012) Organometallics 31:810–818 51. Annibale VT, Song D (2012) Chem Commun 48:5416–5418 52. Vogt M, Gargir M, Iron MA, Diskin-Posner Y, Ben-David Y, Milstein D (2012) Chem A Eur J 18:9194–9197 53. Vogt M, Nerush A, Diskin-Posner Y, Ben-David Y, Milstein D (2014) Chem Sci 5:2043–2051

Recent Advances in the Applications of Metal-Ligand Cooperation via. . .

23

54. Rivada-Wheelaghan O, Dauth A, Leitus G, Diskin-Posner Y, Milstein D (2015) Inorg Chem 54:4526–4538 55. Kumar A, Daw P, Espinosa-Jalapa NA, Leitus G, Shimon LJW, Ben-David Y, Milstein D (2019) Dalton Trans 48:14580–14584 56. Chakraborty S, Blacque O, Berke H (2015) Dalton Trans 44:6560–6570 57. Zhang Y, MacIntosh AD, Wong JL, Bielinski EA, Williard PG, Mercado BQ, Hazari N, Bernskoetter WH (2015) Chem Sci 6:4291–4299 58. Singh AK, Singh S, Kumar A (2016) Cat Sci Technol 6:12–40 59. Bernskoetter WH, Hazari N (2017) Acc Chem Res 50:1049–1058 60. Mellmann D, Sponholz P, Junge H, Beller M (2016) Chem Soc Rev 45:3954–3988 61. Feller M, Gellrich U, Anaby A, Diskin-Posner Y, Milstein D (2016) J Am Chem Soc 138:6445–6454 62. Anaby A, Feller M, Ben-David Y, Leitus G, Diskin-Posner Y, Shimon LJW, Milstein D (2016) J Am Chem Soc 138:9941–9950 63. Oren D, Diskin-Posner Y, Avram L, Feller M, Milstein D (2018) Organometallics 37:2217–2221 64. Langer J, Hamza A, Pápai I (2018) Angew Chem Int Ed 57:2455–2458 65. Leitl J, Marquardt M, Coburger P, Scott DJ, Streitferdt V, Gschwind RM, Müller C, Wolf R Angew Chem Int Ed. https://doi.org/10.1002/anie.201909240 66. Wodrich MD, Hu X (2017) Nat Rev Chem 2:0099 67. Schneck F, Schendzielorz F, Hatami N, Finger M, Würtele C, Schneider S (2018) Angew Chem Int Ed 57:14482–14487 68. Ravishankara AR, Daniel JS, Portmann RW (2009) Science 326:123–125 69. Prather MJ (1998) Science 279:1339–1341 70. Severin K (2015) Chem Soc Rev 44:6375–6386 71. Tolman WB (2010) Angew Chem Int Ed 49:1018–1024 72. Zeng R, Feller M, Ben-David Y, Milstein D (2017) J Am Chem Soc 139:5720–5723 73. Gianetti TL, Annen SP, Santiso-Quinones G, Reiher M, Driess M, Grützmacher H (2016) Angew Chem Int Ed 55:1854–1858 74. Luque-Urrutia JA, Poater A (2017) Inorg Chem 56:14383–14387 75. Xie H, Zhang Y, Xiang C, Li Y, Fan T, Lei Q, Fang W (2018) Dalton Trans 47:15324–15330 76. Schwartsburd L, Iron MA, Konstantinovski L, Diskin-Posner Y, Leitus G, Shimon LJW, Milstein D (2010) Organometallics 29:3817–3827 77. Iron MA, Ben-Ari E, Cohen R, Milstein D (2009) Dalton Trans:9433–9439 78. Schwarz H (2017) Cat Sci Technol 7:4302–4314 79. Reeds JP, Yonke BL, Zavalij PY, Sita LR (2011) J Am Chem Soc 133:18602–18605 80. Zeng R, Feller M, Diskin-Posner Y, Shimon LJW, Ben-David Y, Milstein D (2018) J Am Chem Soc 140:7061–7064 81. Nerush A, Vogt M, Gellrich U, Leitus G, Ben-David Y, Milstein D (2016) J Am Chem Soc 138:6985–6997 82. Vogt M, Nerush A, Iron MA, Leitus G, Diskin-Posner Y, Shimon LJW, Ben-David Y, Milstein D (2013) J Am Chem Soc 135:17004–17018 83. Guo B, Zijlstra DS, de Vries JG, Otten E (2018) ChemCatChem 10:2868–2872 84. Tang S, Milstein D (2019) Chem Sci. https://doi.org/10.1039/c9sc03269j

Top Organomet Chem (2021) 68: 25–70 https://doi.org/10.1007/3418_2020_70 # Springer Nature Switzerland AG 2020 Published online: 1 December 2020

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands Martine R. Tiddens and Marc-Etienne Moret

Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 σ-Acceptor Ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Ambiphilic Ligands and the Retrodative Bond Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Metal-Ligand Cooperative Catalysis Employing d10 Complexes of the σ-Acceptor Ligand Diphosphinoborane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Metal-Ligand Cooperative Reactivity at Group 8 and 9 Complexes of the σ-Acceptor Ligand Diphospinoborane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 π-Acceptor Ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Dewar-Chatt-Duncanson Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Anchored Olefin-Metal Complexes: First Steps Towards Metal-Ligand Cooperativity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Metal-Ligand Cooperative Catalysis Induced by Side-On Coordination of a Ketone . . . 3.4 Imine Side-On Coordination: Synthesis and Metal-Ligand Cooperative Reactivity . . . 4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

26 27 27 29 37 42 42 43 49 59 62 64

Abstract Acceptor ligands, which predominantly withdraw electron density from a transition metal center, often engage in weak metal-ligand interactions. These can be stabilized by flanking the acceptor moiety with strongly binding phosphines in a pincer motif, affording more robust complexes in which bond activation and/or bond-forming events can take place while preserving the integrity of the molecule as a whole. This contribution highlights recent developments in this area. Compounds incorporating a borane at the central position are discussed first, followed by compounds incorporating an electrophilic C ¼ E (E ¼ C, O, N) π-bond. In both cases, recent examples highlight the ability of these ligands to (1) respond to electronic changes at the metal by modifying their binding mode and (2) accept a M. R. Tiddens and M.-E. Moret (*) Utrecht University, Organic Chemistry and Catalysis, Debye Institute for Nanomaterials Science, Faculty of Science, Utrecht, The Netherlands e-mail: [email protected]

26

M. R. Tiddens and M.-E. Moret

nucleophilic fragment (e.g., hydride) from substrate molecules. Applications of acceptor pincer ligands as cooperative catalysts are discussed. Keywords Acceptor ligands · Ambiphilic ligands · Bond activation · Cooperative catalysis · Metal-ligand cooperation · Pincer · π-Ligands

1 Introduction The great successes of homogeneous catalysts in terms of stability, activity, and selectivity can be attributed to one’s ability to precisely tune the properties of a transition metal (TM) center by means of ligand design. Traditionally, supporting ligands have been thought of as spectator ligands whose role was to tune the properties of a transition metal and thereby facilitate metal-centered bond activation of substrates. This paradigm is currently challenged by systems displaying metalligand cooperative reactivity, including (1) ligands facilitating bifunctional substrate activation [1–6], (2) redox-active ligands [7–14], and (3) ligands showing hemilabile coordination behavior [15–20]. Here, bond activation and/or bond-forming events involve strong interplay of the metal center and the cooperative ligand, facilitating reaction pathways that would be less accessible by using conventional homogenous catalyst. A prominent early example of catalysts incorporating bifunctional ligands is the BINAP/diamine-Ru system, where the amine ligand functions as a proton relay in the hydrogenation of ketones [21]. Since then, the metal amide/metal-amine interconversion has become one of the preeminent concepts for metal-ligand cooperative systems and has led to many catalytic applications [22, 23]. More broadly, ligands featuring a donor functional group that can transiently accept a proton or another electrophilic fragment now occupy a place of choice in the toolbox of synthetic chemists (and in the present volume). More recently, cooperative ligands featuring an acceptor site for metal-ligand cooperation are emerging as a fertile area of investigations [24–31]. Whereas the classical description of coordination (Werner-type) and organometallic complexes involves ligands donating electron density to the metal, it had long been recognized that the bonding of many ligands (CO, olefins, and other π-ligands) could only be accurately described by including a secondary interaction involving electrons flowing from the metal to the ligand (π-backbonding). In the case of acceptor ligands, this inverse electron flow becomes the dominant bonding interaction: they feature an accessible empty orbital which forms a – generally weak – metal-ligand interaction by effectively withdrawing electron density from the transition metal. On the basis of the symmetry of the accepting orbital, a distinction between σ- or π-acceptor ligands can be made. Acceptor ligands offer opportunities for unusual, cooperative bond activation pathways (Fig. 1). For instance, the accessible empty orbital of an acceptor ligand can act as a hydride relay in the bifunctional activation of E–H bonds or, more

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

27

Fig. 1 Cooperative processes at the weak metalligand interaction of a transition metal center (M) and an acceptor motif (A)

generally, reversibly accept a nucleophilic fragment. Furthermore, reversible coordination of the acceptor moiety can be expected to stabilize reduced intermediates in a catalytic cycle and hence accelerate the reaction. However, these pathways would often cause the acceptor ligand or a derivative to (irreversibly) leave the coordination sphere of the metal, precluding catalysis. To overcome this limitation, an acceptor moiety can be flanked with strongly binding donor groups such as phosphines in the tradition of the original pincer design used to stabilize weaker TM–C bonds [32]. Pincer ligands generally afford robust complexes while leaving enough open space for incoming molecules to approach the reaction center [33–49]. Hence, acceptor pincer ligands can be expected to allow synergistic processes that preserve the integrity of the complex as a whole. This chapter highlights recent examples of metal-ligand cooperation employing σ-acceptor (Sect. 2) and π-acceptor (Sect. 3) pincer ligands featuring P-donor tethers. Stoichiometric and catalytic cooperative processes are discussed, highlighting the unusual bond activation pathways enabled by acceptor pincer ligands. By comparing the reactivity of σ- and π-acceptor moieties, the similarities and differences of the synergistic processes they facilitate are highlighted.

2 σ-Acceptor Ligands 2.1

Ambiphilic Ligands and the Retrodative Bond Model

In his 1995 classification of covalent compounds of the elements [50], Green defines Z-type ligands as ligands that primarily accept electrons from the element they are bound to, i.e., Lewis acids. Transition metal complexes of such σ-acceptor ligands, however, remained merely scientific curiosities for a long time due to the scarcity of stable examples. In the context of coordination chemistry and catalysis, Lewis acids were mostly used as external activators, co-catalysts, or additives. The field emerged as an area of systematic investigation when the group of Hill reported the first fully characterized metallaboratrane in 1999 (Fig. 2, left) [51] in which a trisubstituted borane is the σ-acceptor moiety. A hydrido-borate scorpionate ligand was shown to react with RuII vinyl precursors via B–H addition to form a supported Ru!B interaction. This initial discovery prompted systematic investigations of TM!B bonds supported by scorpionate ligands featuring sulfur- and nitrogen-based buttresses which are covered in several interesting reviews [24, 27, 29].

28

M. R. Tiddens and M.-E. Moret

Fig. 2 Ruthenaboratrane reported by Hill and co-workers in 1999 (left) [51] and the Pt0 complex of Bourissous’ trisphosphinoborane ligand (right) [52]

Fig. 3 Schematic representation of a retrodative bond formation by donation of electron density from a transition metal orbital to the empty p-orbital of a σ-acceptor ligand and molecular orbital diagram for the retrodative bond formation

A second, more versatile approach to the formation of a weak TM!A interaction (A ¼ acceptor) relies on the synthesis of ambiphilic ligands combining Lewis basic site(s) and Lewis-acidic site(s): a field pioneered by Bourissou and co-workers [53]. Figure 2 (right) shows the ambiphilic triphosphine-borane ligand featuring an intramolecular P!B bond, which is in equilibrium with its open form [54]. Coordination of this tetradentate ligand to Pt0 readily affords a cage structure with intrinsic C3 symmetry [52]. In general, ambiphilic ligands offer relatively straightforward and reliable access to complexes featuring a TM!A interaction, and while the most commonly used Lewis-acidic center is boron, ligands featuring heavier group 13 or group 14 elements as σ-acceptor moiety have been reported on. Their coordination chemistry and reactivity have been discussed in recent reviews [24, 26, 31]. As this new research area of σ-acceptor ligands developed, it led to a better understanding of the proposed underlying bonding model for a TM!A interaction. In general, in the coordination of a σ-acceptor ligand, the metal acts primarily as Lewis base. A so-called retrodative bond of σ-character is formed between a filled metal orbital and the accessible empty orbital of the σ-acceptor ligand (Fig. 3). The electron-withdrawing effect of the retrodative bond stabilizes the filled metal orbital. In addition, a second bonding combination is formed between one empty metal

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

29

Scheme 1 Synthesis of L1 with R ¼ iPr or Ph; according to Bourissou and co-workers [56]

orbital of s or p parentage and the σ-acceptor. This low-lying vacant orbital is responsible for an increased Lewis acidity of TM!A complexes [55]. Depending on the number of donor buttresses introduced in the ambiphilic ligand, flexible or more rigid structures are obtained from mono-, bi-, or tripodal ligand frameworks. In particular, the pincer-like bipodal framework offers a good compromise between stability and reactivity: this design strongly anchors the boron atom in the vicinity of the metal center without excessively shielding it from reagent molecules. Hence, the formed weak TM!B interaction bears potential for metalligand cooperative reactivity. Phosphine-tethered borane ligands of type L1 (Scheme 1) were first synthesized by Bourissou and co-workers via a lithium/halogen exchange between o-diphenylphosphino-bromobenzene and dichlorophenylborane (for R ¼ Ph) [56]. The use of L1 type ligands has proven a fruitful strategy to study the TM!B (B ¼ borane) retrodative bond as it potentially coordinates κ3 (P,B,P) to a transition metal center with a retrodative bond between the metal and the boron center. In the following, the coordination chemistry of L1 to late transition metal centers and the reactivity of the resulting complexes are briefly discussed, and illustrative examples of metal-ligand cooperative catalysis are presented.

2.2

Metal-Ligand Cooperative Catalysis Employing d10 Complexes of the σ-Acceptor Ligand Diphosphinoborane

The borane ligand L1 is designed to support a TM!B interaction. Table 1 shows a selection of d10 complexes featuring such a retrodative bond. The tetrahedral complex L1Ni0(THF) was reported by Peters and co-workers to feature a η2(B,Cipso) coordination rather than the expected η1(B) interaction, meaning that the Ni!B interaction is supported by arene coordination (Table 1) [57]. This binding mode is characterized by short TM–B and TM–Cipso distances and a relatively low pyramidalization of the boron atom. The isoelectronic L1CuICl structure also adopts a tetrahedral geometry featuring a similar arene-supported η2(B,Cipso) coordination (Table 1), but a longer TM–B bond distance of Cu–B ¼ 2.396(5) Å [58]. Bourissou and co-workers proposed the

Ni–B: 2.124(2) Å Ni–Cipso: 2.175(2) Å ΣBα: 352 P1 ¼ PPh2

Cu–B: 2.396(5) Å Cu–Cipso: 2.364(3) Å ΣBα: 356 P1 ¼ PPh2

Pd–B: 2.194(3) Å Pd–Cipso: 2.463(3) Å ΣBα: 346 P1 ¼ PPh2

Ag–B: 2.742(3) Å Ag–Cipso: 2.939(3) Å ΣBα: 357 P2 ¼ PiPr2

Au–B: 2.309(8) Å Au–Cipso: 3.099(8) Å ΣBα: 341 P2 ¼ PiPr2

Table 1 The different d10 complexes of L1 adopt different overall geometries with TM!B retrodative bond interactions of different strengths (THF ¼ tetrahydrofuran, ΣBα ¼ sum of angles around boron) [57–61]

30 M. R. Tiddens and M.-E. Moret

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

31

presence of a three-center Cu–B–Cipso interaction on the basis of Kohn-Sham orbitals obtained by density functional theory (DFT) calculations. Natural bond orbital (NBO) analysis showed that among the various donor-acceptor interactions involving this triangle, a Cu!B donating interaction of similar magnitude as the η1(B) interactions found in related complexes is present [24]. Additionally, natural population analysis confirms a net charge transfer from Cu to B, reinforcing the description of the ligand as an acceptor ligand. Moving to the second and third row transition metal centers, L1 coordinates to AuI in an initially unexpected yet most pincerlike manner [61]. A square planar geometry around the tetracoordinated AuI is observed, featuring trans-diphosphine coordination of L1 and a chloride co-ligand trans to the η1(B)-coordinated borane (Table 1). A strong Au!B interaction is observed as evident from a short bond distance (2.309(8) Å) and strong pyramidalization of the borane center (ΣBα ¼ 341 ). Frontier orbital analysis shows a B–Au–Cl three-center interaction; however, the charge depletion at gold and charge increase at boron are not large enough to be considered a 2e oxidation of the gold center to AuIII. In addition, 197Au Mössbauer spectroscopy supported the classification of L1AuICl as a 16 VE AuI complex. The distinct η1(B) and arene-supported η2(B,C) coordination modes can be considered two extremes of L1 coordination to d10 transition metal centers. This becomes apparent upon the evaluation of the coordination of L1 to Pd0 [59]. Depending on the donor strength and steric requirements of the phosphine tethers, the Pd complexes of L1 adopt a strongly distorted square planar geometry (P1 ¼ PPh2, Table 1) with at most a weak Pd–Cipso interaction (Pd–C ¼ 2.463(3) Å) or a T-shape geometry (P3 ¼ PCy2, Cy ¼ cyclohexyl, no co-ligand) [62]. Both complexes feature a strong Pd!B interaction as evident from short Pd–B bond distances (Pd–B ¼ 2.194(3) Å for P1 ¼ PPh2 and Pd–B ¼ 2.243(2) Å for P3 ¼ PCy2) and a significant pyramidalization of the boron atom (ΣBα ¼ 346 for P1 ¼ PPh2 and ΣBα ¼ 341 for P3 ¼ PCy2). In contrast with other d10 analogues, the silver(I) complex L1AgI(I) exhibits a very weak TM!B interaction (Table 1) [60]. Rather than κ3 (P,B,P), a trigonal planar κ2 (P,P) coordination geometry was proposed based on the sum of angles of 356.5 in the P2AgI plane. Competing B–F bond formation prevents the coordination of L1 to AgF, showing that halide abstraction can hamper coordination of σ-acceptor ligands. This study of L1 coordination to d10 transition metals nicely illustrates how acceptor ligands can give rise to structures that challenge our understanding of the bonding and geometry of transition metal complexes. More generally, the series of complexes shown in Table 1 exposes the coordination flexibility of the L1 platform, demonstrating most importantly that the TM–B interaction is not enforced by the pincer architecture but rather a possibility among several accessible geometries. Inagaki and co-workers hypothesized that the electron-depleting nature of L1 would amplify the intrinsic alkynophilicity of a gold cationic center which, in catalysis, can be utilized for a more effective activation of alkynes towards nucleophilic attack. For this purpose, cationic AuI complexes of L1 were synthesized (Scheme 2) [63]. The synthesis of an L1Au+ fragment from L1AuICl by direct

32

M. R. Tiddens and M.-E. Moret

Scheme 2 Synthesis of the cationic [L1Au][SbF6] complex by indirect halide abstraction; 1,5-COD ¼ 1,5-cyclooctadiene [63]

halide abstraction proved difficult. Therefore, a dinuclear species stabilized by a 1,5-cyclooctadiene (1,5-COD) bridge, [(L1Au)2(COD)][SbF6]2 (Scheme 2, middle), was obtained first by using Ag[SbF6] in the presence of 1,5-COD. During crystallization, the 1,5-COD co-ligand dissociates to afford the mononuclear [L1Au][SbF6] species, which was characterized by X-ray crystallography. The Au!B bond is significantly weakened by halide abstraction as evident from an elongated Au–B distance of 2.52(1) Å in [L1Au][SbF6] vs 2.335(5) Å in L1AuICl. In addition, the sum of C–B–C angles increases from 344 to 355.1 , indicating a boron hybridization close to sp2. The other bond distances and angles do not undergo major changes, showing that the electron density at the Au+ center has a direct influence on the TM!B bond strength. The cationic gold species was tested as catalyst for the cycloisomerization of enynes, in which alkyne activation by coordination to the gold center is followed by an intramolecular nucleophilic attack. In their recent “digest paper” [64], Inagaki and co-workers discuss seven examples in which the presence of a retrodative bond between the gold cation and the σ-acceptor L1 leads to a higher catalytic activity and selectivity. Table 2 shows the [2+2] cycloaddition of 1,8-enynes as an example of these comparative studies. The dinuclear species [(L1Au)2(COD)][SbF6]2 was used as precatalyst. Under the optimized reaction conditions of 2 mol% [Au+] in 1,2-dichloroethane (DCE) at room temperature for 24 h, a seven-membered ring was selectively formed in moderate to good yields depending on the substitution. Directly compared to other phosphine-stabilized gold cations such as [(PPh3)2Au] [SbF6], [(PPh3)Au][SbF6], [(XPhos)Au][SbF6], or [(Xantphos)Au][SbF6] (Table 2), the [L1Au]2(COD)[SbF6]2 species shows superior catalytic activity indicating that the TM!B interaction has a beneficial effect on the reactivity of the gold cation. While no in-depth mechanistic study was conducted, it is assumed that a σ-acceptor trans to the triple bond induces an electron push-pull charge transfer across the alkyne–Au!B coordination plane by donation of more electron density into the Au!B bond. This results in a stronger activation of the triple bond and subsequently facilitates nucleophilic attack by the olefin. Overall, this study serves as an example of metal-ligand cooperative catalysis in which the weak and responsive TM!B interaction is utilized to enhance the Lewis acidity of the transition metal center. Peters and co-workers demonstrated how the accessible empty orbital of L1 can be used as hydride relay in bifunctional dihydrogen (H2) activation and catalytic reduction of olefins [57, 65]. The initial L1Ni0(THF) complex (Table 1) appeared unreactive towards H2, suggesting the cleavage of the η2(B,Cipso) coordination to be

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

33

Table 2 Comparative study of the [Au+] catalyzed [2+2] cycloaddition of 1,8-enynes; DCE ¼ dichloroethane, Ar ¼ argon atmosphere [63]

[Au+]

Substrate R ¼ CO2Me R ¼ CH2OPiv

Yield 62% 85%

[PPh3–Au–PPh3]SbF6

R ¼ CO2Me R ¼ CH2OPiv R ¼ CO2Me R ¼ CH2OPiv R ¼ CO2Me R ¼ CH2OPiv

No reaction No reaction 16% 51% No reaction No reaction

R ¼ CO2Me R ¼ CH2OPiv

39% 55%

[PPh3–Au]SbF6

difficult. The exchange of the phosphine substituents in L1 from phenyl to isopropyl, however, enabled the isolation of a dinitrogen complex L1Ni0(N2), which upon addition of H2 gas exchanges ligands to form a nonclassical Ni–(H2) adduct (Scheme 3) [65]. Over the course of several hours, the formation of a bridging borohydride– Ni–hydride complex was observed by NMR. During H2 activation, the nature of the Ni!B interaction changes from a modest perturbation exerted by the empty p(B) orbital on the d10(Ni) center bearing a σ-bound H2 ligand to the interaction of an anionic borohydride ligand stabilizing the mononuclear NiII–H species. While cis homolytic H2 activation via a cisdihydride intermediate cannot be fully ruled out, computational studies conclude that synergetic heterolytic H2 activation is the most likely mechanism [66]. No high-

34

M. R. Tiddens and M.-E. Moret

Scheme 3 Synthesis of a nonclassical Ni–(H2) adduct by N2/H2 exchange at L1Ni0(N2) and bifunctional H2 activation across the Ni!B interaction to form a bridging borohydride–Ni–hydride species; P2 ¼ PiPr2 [65]

Scheme 4 Bifunctional activation of the Si–H bond in diphenylsilane across the Ni!B interaction enables the catalytic hydrosilylation of benzaldehyde; P1 ¼ PPh2 [67]

energy penalty seems to be associated with breaking the Ni!B interaction in this process. Experimentally, H2 activation was shown to be more facile using a slightly altered L1Ni-system where a mesityl group replaces the phenyl group on boron (L1MesNi0, Scheme 4) [57]. Hereby, steric bulk likely weakens the Ni!B interaction, which becomes a η3(B,C,C) coordination involving the ipso and ortho carbon of the mesityl substituent. The resulting L1MesNi0 complex undergoes facile and instantaneous reaction with H2 to form the bridging borohydride–Ni–hydride species. Using this L1MesNi0 complex, catalytic styrene hydrogenation under very mild conditions (4 atm H2) was observed, constituting an example of bifunctional H2 activation involving a σ-acceptor ligand in a catalytic reaction. Additionally, stoichiometric reaction of L1MesNi0 with diphenylsilane (H2SiPh2) shows the formation of a bridging borohydride–Ni–(SiHPh2) species resulting from the bifunctional activation of the Si–H bond over the Ni!B interaction (Scheme 4) [67]. The solid-state structure of the bridging borohydride–Ni–(SiHPh2) supports the structural analysis of the bridging borohydride–Ni–hydride species, which was so far based on NMR analysis alone. Si–H bond activation leads to a NiII center which adopts a distorted square planar geometry. The η3(B,C,C) interaction is broken as the mesityl group decoordinates to accommodate the bridging hydride. Furthermore, a Ni–Si bond distance of 2.2435(7) Å is found supporting the Ni–silyl characterization. Under mild catalytic conditions, the L1MesNi0 complex is a competent catalyst in the hydrosilylation of benzaldehydes with H2SiPh2 (Scheme 4). Here as well, it is proposed that the borane σ-acceptor functions as hydride relay in this metal-ligand cooperative catalytic transformation.

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

35

Scheme 5 Bifunctional activation of allyl acetate across the Pd!B interaction (left) [59] and conversion of a σ-acceptor borane ligand into a σ-donor boryl ligand by phenyl group transfer (right); P1 ¼ PPh2 [68]

Tauchert and co-workers studied oxidative addition at the L1Pd0(2,6-lutidine) complex [59]. No reactivity towards bromobenzene was observed. However, a reaction with iodobenzene leads to the classic oxidative addition product which over time undergoes a reductive elimination of the phenyl-substituent on boron and the phenyl-ligand on PdII leading to a PBP–PdII boryl pincer complex (Scheme 5, right) [68]. The reaction of this complex with phenyl lithium in the presence of 2,6-lutidine gives back the starting L1Pd0(2,6-lutidine) complex. Overall, this phenyl group transfer enables the reversible conversion of a σ-acceptor borane ligand to a σ-donor boryl moiety, which is one promising strategy to access transition metal boryl complexes [69]. The Pd!B retrodative bond depletes electron density at the Pd center, thereby impeding activation of strong σ-bonds via classical oxidative addition. Therefore, bifunctional C–O bond activation across the Pd!B interaction was attempted by Tauchert and co-workers in a reaction of the L1Pd0(2,6-lutidine) complex with allyl acetate (Scheme 5, left) [59]. Here, a Pd-allyl complex and a new B–OAc bond are formed. C–O bond activation is thought to be favored by the formation of a new strong B–O bond, showing that the σ-acceptor borane can function as relay for other groups than hydrides. The allyl acetate activation is reversible, and the equilibrium can be shifted by addition of 2,6,-lutidine (Scheme 5, left). This reactivity was applied in the catalytic allylic substitution reaction of allyl acetate with diethylamine. However, an accelerating effect of added tetrabutylammonium acetate suggests that species featuring a strong Pd!B interaction may be inactive, the extra acetate source breaking the Pd!B bond and thereby enhancing Pd-centered catalytic conversion. Recently, Kameo and Bourissou reported a different cooperative approach to facilitate the activation of strong σ-bonds, specifically of aromatic C–Cl bonds, using L1Pd0(PPh3). While L1 coordination results in the depletion of electron density at the Pd center of L1Pd0(PPh3), a more electron-rich Pd species is formed in a subsequent reaction with potassium hydride (KH). Here, a hydride insertion into the Pd!B bond forms a B–H–Pd bridge in the overall anionic Pd complex K[L1-H-Pd(PPh3)] [70]. This hydride is positioned at the apical position in an overall trigonal-pyramidal geometry at the Pd center (Fig. 4).

36

M. R. Tiddens and M.-E. Moret

B1 H1 P1

Pd1

P2

P3

Fig. 4 X-ray crystal structure of K[L1-H-Pd(PPh3)] showing the bridging borohydride motif (thermal ellipsoids at 50% probability). The [K([2.2.2]-cryptand)] cation, hydrogen atoms (except the borohydride), and phenyl groups on the phosphorus atoms (except for the bound carbon atom) are omitted for clarity [70]

Cl

R

H

5 mol% L1Pd0(PPh3) H-COOK [2.2.2]-cryptand

N

H

H

92 %

H

B R

H

F

17 examples 78-99 %

97 %

P1

Pd P1 PPh3

L1Pd0(PPh3)

Fig. 5 Catalytic hydrodechlorination of (hetero)aryl chlorides (reaction conditions: 60–100 C, 48–72 h); P1 ¼ PPh2 [70]

The electron-rich [L1-H-Pd(PPh3)]K is reactive towards various C–Cl bonds, which was used in the catalytic hydrodechlorination of (hetero)aryl chlorides (Fig. 5). In this system, potassium formate is used as hydride source. High yields and a high functional group tolerance were observed for heteroarene substrates. Lower yields were obtained for substrates featuring electron-donating substituents para to the C–Cl bond, in line with the general trend of oxidative addition being more difficult when the C–Cl bond is less polarized. Based on computational work, a catalytic cycle for hydrodechlorination was proposed (Fig. 6). In contrast with the general Pd-catalyzed C–C cross-coupling mechanism, which consists of a sequence of oxidative addition, transmetalation, and reductive elimination, this reaction starts by reaction of L1Pd0(PPh3) with KH to form the anionic Pd0 borate [L1-H-Pd(PPh3)]2, which then undergoes oxidative addition of the C–Cl bond and elimination of KCl to form a proposed Pd–Ar

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

37

Fig. 6 Proposed catalytic cycle for the catalytic hydrodechlorination of (hetero)aryl chlorides; P1 ¼ PPh2 [70]

intermediate. The following B-to-Pd hydride transfer and reductive elimination of the C–H bond are exergonic and most likely facilitated by the formation of a TM!B interaction.

2.3

Metal-Ligand Cooperative Reactivity at Group 8 and 9 Complexes of the σ-Acceptor Ligand Diphospinoborane

As low spin d8 transition metal complexes tend to adopt a square planar geometry, their filled dz2 orbital might act as Lewis base to an apical σ-acceptor moiety in an overall square pyramidal complex. Such a TM!B interaction is observed in the 16VE L1RhICl(DMAP) complex featuring a 4-dimethylaminopyridine (DMAP) co-ligand (Fig. 7) [56]. Indeed, the Rh center adopts a square pyramidal geometry with the boron atom in the apical position to maximize the orbital overlap between the full dz2(Rh) orbital and the empty p(B) orbital. NBO calculations find a two-center two-electron (2c2e) bond between Rh and B. A strong TM!B interaction is evident from the pyramidalized boron center (ΣBα ¼ 340.2 ) and a short Rh– B distance (2.295(5) Å). In addition, the 11B NMR signal shifts upfield to 19.4 ppm from 43 ppm in L1, indicating a four-coordinate boron atom. The geometry of L1RhICl(DMPA) is representative for low spin d8 TM complexes of L1 (Fig. 7). Bourissou and co-workers evaluated the variety in TM!B interaction strength in RhI, PtII, and PdII complexes of L1 specifically. Based on 11B NMR and X-ray data, the study revealed the TM!B interaction to become significantly weaker when going from RhI to PtII to PdII. Therefore, this series of d8 complexes of L1 illustrates

38

M. R. Tiddens and M.-E. Moret

Fig. 7 Square pyramidal d8 complexes of L1; DMAP ¼ 4-dimethylaminopyridine, P2 ¼ PiPr2

P2 B P2 P2

RCOOM Rh Cl

P2 P2

B Rh

Ph

Ph O O

Rh

B R

O

O P2 R

P2 P2

B Rh OTf

Scheme 6 Reaction of L1RhICl with bidentate oxygenous ligands (RCOOM ¼ KOAc, CsOPiv). The phenyl group transfer from B to Rh is an equilibrium which is proposed to go through an intermediate species featuring a η2(B,Cipso) coordination mode of L1 similar to the isolated species (box) from a reaction of L1RhICl with TMSOTf; P2 ¼ PiPr2 [72]

how the coordination strength of the σ-acceptor borane to the transition metal center is a continuum tuned by the intrinsic Lewis basicity of the metal center. The Lewis acidity of the central boron atom in L1 is quenched by coordination to a transition metal center as was experimentally shown by Britovsek and co-workers. They attempted bifunctional C–O bond activation across a Rh!B interaction in the reaction of square pyramidal [L1RhI(CO)2][SbF6] species with methyl acetate [71]. In contrast with the related L1Pd0(2,6-lutidine) complex (Scheme 5, Sect. 2.2), [L1RhI(CO)2][SbF6] is unreactive towards neutral oxygen-containing substrates, which was attributed to the strong Rh!B interaction. Ozerov and co-workers later reported on the reactivity of L1RhICl species with anionic oxygen-containing substrates (Scheme 6). A borane-to-boryl interconversion by phenyl transfer from the ligand to the transition metal center was observed upon the reaction with alkali-metal carboxylates [72]. Two isomers are observed in equilibrium upon a reaction of L1RhICl with potassium acetate (KOAc) or cesium pivalate (CsOPiv). The initial replacement of chloride results in a Rh species featuring a Rh!B interaction and a κ2-carboxylate co-ligand. The second species features a terminal phenyl group on Rh and a sp3-hybridized borate ligand with a carboxylate bridge between Rh and B. This bridging interaction is not present in the product of the reaction of L1RhICl with trimethylsilyl triflate (TMSOTf). Instead, a η2(B,C) coordination of L1 is observed (Scheme 6, box). Most likely, the lower Lewis basicity of triflate does not allow for B–O adduct formation. A related species featuring a η2(B,C) interaction was proposed to be an intermediate along the reaction pathway of the phenyl group transfer process. Across the periodic table, the bridging borohydride (B–H–TM) motif seems to be broadly accessible to transition metal complexes of L1. Kameo and Nakazawa

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

O +

OH

0.5 mol% L1-H-RhI(CO)(PPh3) 20 h, 70

39

OH +

oC

O

B P1

90% yield

P1

H Rh PPh3 CO

L1-H-RhI(CO)(PPh3)

Scheme 7 Reaction scheme of a transfer hydrogenation catalyzed by the L1-H-RhI(CO)(PPh3) featuring a bridging borohydride motif; P1 ¼ PPh2 [73]

Scheme 8 Reduction of L1FeIBr by one electron results in the case of P1 ¼ PPh2 in a L1Fe0 complex featuring a η7(B,Ph) coordination (left), whereas P2 ¼ PiPr2 leads to a dinuclear N2 bridged L1Fe(μ-1,2-N2)FeL1 complex featuring one η3(B,Cipso,Cortho) and one η2(B,Cipso) coordination mode of L1 (right) [74]

reported a L1-H-RhI(CO)(PPh3) species (Scheme 7) synthesized from the reaction of L1 with Rh(H)(CO)(PPh3)3 [73]. Structurally, the B–H–Rh species resembles the anionic [L1-H-Pd(PPh3)]K species (Sect. 2.2, Fig. 6) [70]: in the solid state, the Rh center adopts a trigonal-bipyramidal geometry with a hydride at the apical position. This hydride is part of an overall three-center two-electron (3c2e) B–H–Rh interaction. This bonding interaction was further analyzed by NBO comparative analysis of the free ligand (L1), the Rh(H)(CO)(PPh3)3 precursor, and the L1-H-RhI(CO)(PPh3) species. While the charge on boron increases and the charge on Rh decreases as expected upon coordination of L1 to the Rh precursor, the B–H– Rh interaction seems to minimally influence the charge on the hydride, suggesting that the σ-acceptor effectively withdraws electron density from the transition metal center via the B–H–TM motif. L1-H-RhI(CO)(PPh3) is a catalyst for the transfer hydrogenation of propiophenone with isopropanol (Scheme 7) [73]. Though the mechanism of this catalytic reaction was not further studied, L1-H-RhI(CO)(PPh3) outcompetes the boron-free Rh(H)(CO)(PPh3)3 species in catalytic activity significantly (90% yield versus 29% yield) indicating a positive influence of the L1 coordination. First-row d8 transition metal complexes of L1 were reported by Peters and co-workers as they synthesized the L1FeIBr complex from in situ reduction of FeBr2 in the presence of L1, showing the propensity of borane ligands to stabilize FeI species [74]. An extra one-electron reduction led to the formation of the dinuclear N2-bridged complex L1Fe(μ-1,2-N2)FeL1 (P2 ¼ PiPr2, Scheme 8, right) or a η7(B,Ph)-coordinated monomeric L1Fe0 species (P1 ¼ PPh2, Scheme 8, left). The peculiar η7(B,Ph) mode is achieved by distorting the B–Cipso bond in L1. The higher hapticity in the diamagnetic Fe complex is maintained in solution according

40

M. R. Tiddens and M.-E. Moret

Scheme 9 Nβ functionalization by reaction of L1Fe(μ-1,2-N2)FeL1 with (a) 1,2-bis (chlorodimethylsilyl)ethane and 2.1 equivalent Na/Hg to form L1Fe(N2bse) and Nα functionalization upon a subsequent hydrosilylation with (b) PhSiH3, P2 ¼ PiPr2 [74]

Scheme 10 Bifunctional H2 activation across the Fe!B bond (left) and iron dicarbyne synthesis by oxygen atom functionalization with trimethylsilyl triflate (TMSOTf). Addition of 1 atm H2 leads to the formation of an olefin product: P2 ¼ PiPr2, K ¼ potassium [75]

to the upfield shift of aryl resonances in 1H NMR. In the dinuclear species, the pseudotetrahedral Fe centers are inequivalent as Feα coordinates L1 η3(B,Cipso, Cortho), whereas the Feβ shows a η2(B,Cipso) coordination. In solution, however, the Fe centers are equivalent, leading to assumption that the η3(B,C,C) interaction is highly flexible, which makes these Fe species ideal starting points of further studies into metal-ligand cooperative reactivity. The Fe-bound N2 molecule was functionalized at the Nβ position in the reaction of L1Fe(μ-1,2-N2)FeL1 with 1,2-bis(chlorodimethylsilyl)ethane and 2.1 equivalents of Na/Hg to form the iron-aminoimide complex L1Fe(N2bse) (Scheme 9, middle) [74]. The double silylation of Nβ results in a pseudotetrahedral d6 Fe-aminoimide complex featuring a η3(B,C,C) interaction as well as a Fe  N triple bond (Fe–N, 1.6607(5) Å) and a reduced N–N bond (N–N bond distance average, 1.326 Å). As the TM!B interaction was shown to activate H–E bonds (E ¼ H, Si) in a bifunctional manner (Sect. 2.2), L1Fe(N2bse) was reacted with phenylsilane (PhSiH3) in an attempt to hydrosilylate the Fe  N triple bond. A facile reaction results in silylation at Nα, whereas the hydride is incorporated into a B–H–Fe motif (Scheme 9, right). The iron hydrazido species features a N–N bond distance of 1.492(4) Å indicating a single bond. Hence, in the overall two-step reduction of a N2 triple bond to a single bond, the σ-acceptor ligand L1 acts both as a stabilizing ligand for an electron-rich Fe0 center and as a hydride acceptor in the bifunctional hydrosilylation of the Fe  N triple bond. Upon addition of 1 atm CO to L1Fe(μ-1,2-N2)FeL1, the mononuclear iron dicarbonyl species L1Fe(CO)2 was formed (Scheme 10, middle) [75]. The iron dicarbonyl species features a Fe!B retrodative bond and an interaction between

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

41

Scheme 11 Bifunctional C–H bond activation (left) and reactivity of L1Co0(N2) with E–H (E ¼ O, S, Si) bonds (right), (a) benzoquinoline and (b) phenol or thiophenol; P2 ¼ PiPr2 [76]

iron and one phenylene linker. Facile H2 activation by L1Fe(CO)2 with 1 atm H2 leads to the formation of a bridging B–H–Fe motif as well as a Fe–H bond (Scheme 10, left). X-ray analysis reveals cis-dihydride stereochemistry. Double oxygen atom functionalization was observed in a reaction of L1Fe(CO)2 with TMSOTf under strongly reducing conditions (excess potassium; Scheme 10). The disilylation results in a structurally unique iron dicarbyne complex in which the Fe!B interaction is replaced by a stabilizing (FeCcarbyne)!B interaction. This interpretation of the bonding situation is based on a relatively long Fe–B distance (2.593(1) Å) and shorter Ccarbyne–B distance (1.862(1) Å) obtained from the X-ray crystal structure. Additionally, the boron atom is pyramidalized (ΣBα ¼ 328 ), indicating borate character. Facile C–C bond formation is observed upon the addition of 1 atm H2 to the iron dicarbyne species, affording the Z-olefin product (Me3SiO)CH¼CH(OSiMe3) and an unidentified paramagnetic Fe-containing product (Scheme 10, right). Analogous to L1Fe(μ-1,2-N2)FeL1, Peters and co-workers synthesized the cobalt species, L1Co0(N2) featuring a terminal N2 ligand and a η2(B,C) coordination of the extended σ-acceptor motif. This d9 complex of L1 was tested for a series of bifunctional E–H bond activations in parallel with the iron analogue, generally displaying similar reactions. Activation of benzoquinoline affords a bridging borohydride species (B–H–Co) with new Co–C and Co–N bonds (Scheme 11, left), where the heterocyclic N-atom acts as a directing group. A similar product is formed by N–H bond activation of 8-aminoquinoline. In both cases, this fifth N-donor ligand is thought to have a stabilizing effect on the formed CoII species. The reaction of L1Co0(N2) with phenol (E ¼ O) and thiophenol (E ¼ S) leads to the formation of terminal Co-phenolate and Co-phenylthiolate complexes as well as 0.5 equivalent of H2 gas (Scheme 11, right). The complexes still feature the η2(B,C) coordination mode, suggesting that the bridging borohydride species (B–H–Co) is not stable for these four-coordinated CoII species. In contrast, L1Co0(N2) activates the Si–H bond of Ph2SiH2 to form a bridging B–H–Co motif (Scheme 11, right). This bifunctional bond activation is similar to the observed structurally related Fe and Ni complexes [67, 74]. The formation of the B–H–Co motif is reversible and

42

M. R. Tiddens and M.-E. Moret

was applied to the catalytic hydrosilylation of benzaldehydes, alkyl aldehydes and aryl and alkyl ketones, where L1Co0(N2) generally outcompetes the structurally related Ni system [67].

3 π-Acceptor Ligands 3.1

Dewar-Chatt-Duncanson Model

It is remarkable that, while Zeises’ salt K[PtCl3(C2H4)]H2O was reported in 1827 as the first organometallic complex [77, 78], it took more than 100 years to explain its olefin coordination. This complication originated from the lack of a binding model to fully interpret the observed data. The Dewar-Chatt-Duncanson (DCD) bonding model, which is widely used today to explain olefin coordination, was proposed in the 1960s by Michael J. S. Dewar, Joseph Chatt, and Leonard A. Duncanson. This model involves two important orbital interactions between the η2(C,C)-bound olefin and the transition metal center. First, the π-electrons of the olefin double bond form a σ-bond with the transition metal (Fig. 8, left). Additionally, a filled d-orbital backdonates electron density into the π* orbital of the double bond (Fig. 8, middle). Olefin coordination to a transition metal center can be described as two resonance extremes: the π-adduct (Fig. 8, I) and the metallacycle coordination (Fig. 8, II). Formally, the oxidation state of the metal is increased by two in the metallacycle extreme. In cases where π-backdonation is the dominating interaction, the ligand effectively accepts electron density from the transition metal making it an acceptor ligand with a low-lying π* orbital as the characteristic accessible empty orbital. The DCD model was originally proposed for metal-bound olefin coordination but can also be applied to other π-ligands such as side-bound ketones and imines. The synthesis, coordination chemistry and metal-ligand cooperative reactivity of transition metal pincer complexes featuring these π-acceptors are discussed in the next sections.

Fig. 8 Bonding description of a metal-bound olefin ligand in two orbital interactions (left) and the two resonance extremes of the DCD model (right)

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

3.2

43

Anchored Olefin-Metal Complexes: First Steps Towards Metal-Ligand Cooperativity

In general, metal-bound olefins insert readily into a M–Y bond resulting in a metalalkyl species. Thereby the motif formally accepts either a nucleophilic or an electrophilic fragment (Y ¼ Nu or E+). Such steps are often part of a catalytic cycle in which the olefin is one of the substrates and subsequently leaves the coordination sphere of the metal as a product molecule. They could potentially also be applied in the context of metal-ligand cooperative catalysis, if the reactive olefin were anchored to the metal in a pincer-type ligand design and thus forced to remain in the coordination sphere. Figure 9 (right) schematically shows the envisioned bifunctional activation of a X–Y bond across the anchored olefin-metal interaction. In addition, the weak interaction between an olefin motif and a transition metal may be reversibly disrupted, stabilizing reactive intermediates that require different coordination environments at the metal. A pincer ligand with an olefin as central binding moiety would then act as a hemilabile ligand (Fig. 9, left). In this section, the synthesis and reactivity of pincer complexes featuring an anchored olefin motif are discussed. Here, the reversible β-hydride insertion/elimination process is central as it represents a first step towards metal-ligand cooperative reactivity using this type of π-acceptor ligands. Rigid o-phenylene linkers have been abundantly used to anchor a central σ-acceptor motif in the proximity of transition metal centers (Sects. 2.2 and 2.3). Iluc and co-workers used this approach to bring a central ethyl-group into close proximity of a PdII center (Scheme 12, left) [79]. Heat-induced C–H activation and dehydrohalogenation generates a square planar PCP PdII complex (Scheme 12, middle) [80]. A second dehydrohalogenation step with potassium bis(trimethylsilyl)amide (KHMDS) leads to a Pd0 complex featuring a metal-bound olefin motif (Scheme 12, right). Bifunctional activity

Hemilability Mn L

L

Mn

Mn+2

XY

Y Mn+2 X

Fig. 9 Resonance extremes of a metal-bound olefin motif and their prototypical cooperative reactivity

Scheme 12 Synthesis of a metal-bound olefin motif in the coordination sphere of Pd0 [79]

44

M. R. Tiddens and M.-E. Moret

Scheme 13 The interconversion of Ru=C to Ru-olefin involves α- and β-hydride elimination and insertion processes [81]

Scheme 14 The Rh-olefin complex shows N2-dependent β-hydride insertion and elimination [83]

X-ray crystallography suggests a strong interaction between Pd0 and the bound olefin as an elongated C¼C bond length of 1.398(3) Å vs 1.34 Å (for a typical C¼C bond) is observed, indicating significant π-backdonation. The Pd-olefin complex was also identified as the product of a side reaction of a Pd-bound nucleophilic carbene incorporated in a PCP pincer structure with CH2Cl2, in which a formal CH2-group transfer to the nucleophilic carbon atom occurs. A perhaps less unexpected connection between a metal-bound olefin motif and a carbene complex is found in isomerization of an aliphatic PCP Ru carbene species (Ru=C) to a Ru-olefin species featuring a 1,2-connected olefin motif (Scheme 13) reported by Gusev and co-workers [81]. As postulated by Shaw and co-workers [82], the transformation is thought to go through a Ru-alkyl intermediate formed by initial hydrogenation and α-hydride insertion of the carbene complex. Subsequent β-hydride elimination and H2 release form the metal-bound olefin motif. The displayed reversible transformations make systems with labile hydrogen atoms present in α- and β-position to the metal promising candidates for investigations into metal-ligand cooperative processes. Specifically, the process of β-hydride elimination/insertion was studied by Milstein and co-workers using a metal-bound olefin motif with a 1,1-disubstitution pattern. Here, the olefin double bond reversibly inserts into a Rh–H bond (Scheme 14) [83]. An aliphatic PCP pincer complex of Rh was shown to react with sodium hydride (NaH), resulting in formal HCl elimination. Interestingly, a subsequent β-hydride elimination is observed, resulting in the formation of a metal-bound olefin motif in Rh-olefin. This Rh-olefin complex is in a fast equilibrium with the corresponding alkyl complex via olefin insertion/β-hydride elimination. Free N2 traps the olefin insertion product by coordination to Rh to form Rh-alkyl. A kinetic study revealed N2 dissociation to be the rate-limiting step in the conversion from

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

45

Scheme 15 Reversible H2 activation and β-hydride insertion at the Ir-olefin complex [84]

Rh-alkyl to Rh-olefin. This equilibrium is a remarkable example of direct trans migration via a concerted, highly organized transition state. The insertion is thought to be possible due to the unique geometry of Rh-olefin, in which the square planar geometry is distorted in order to bend the olefin towards the trans-hydride already in the ground state. Overall, this system demonstrates that facile and reversible β-hydride elimination/insertion is possible even with the natural trans configuration imposed by the pincer structure and shows the potential of π-complexes to act as a transient hydride storage moiety in cooperative processes. In a related study by Wendt and co-workers, Ir-olefin (Scheme 15) was synthesized starting from Ir-alkyl, an aliphatic PCP pincer complex featuring a methylsubstituted cyclohexyl ring [84]. In the reaction, dehydrogenation was induced upon heating, leading to the formation of a metal-bound olefin motif. In this process, the C–C bond length decreases from 1.553(4) Å in Ir-alkyl to 1.438(15) Å in Ir-olefin. This distance being between the typical ranges for C–C single and double bonds indicates strong π-backdonation from the electron-rich Ir center to the olefin motif. In the presence of H2, Ir-olefin is in equilibrium with the corresponding IrIIIdihydride complex. Furthermore, upon heating, a stable IrIII-trihydride complex (Scheme 15, right) can be obtained by formal hydrogen iodide (HI) elimination with NaOtBu in the presence of a H2 atmosphere. The IrIII-trihydride species does not release a H2 molecule or undergo β-insertion. In contrast, a hydride in the IrIIIdihydride complex slowly inserts into the olefin double bond, reinstating the initial Ir-alkyl complex. The shuffling between metal-olefin and metal-alkyl species by means of a reversible β-hydride insertion/elimination process enables the cooperative activation of small molecules as was presented by Wendt and co-workers. They used the slightly different Ir complex (Ir-Ph, Scheme 16) featuring a coordinated, internal C¼C bond [85]. A comparably strong metal-olefin interaction is indicated by the C¼C bond elongation to 1.425(7) Å according to X-ray crystal structure determination. Ir-Ph readily activates H2 to form the corresponding IrIII-trihydride complex (Ir-(H)3). A subsequent, reversible H2 addition coupled to β-insertion generates an equilibrium between the two Ir species Ir-(H)3 and its corresponding insertion product Ir-(H)4. The latter features a central Csp3 donor atom and is characterized as a tetrahydride by NMR spectroscopy. In addition, the described β-hydride insertion/elimination process is observed in the reversible CO2 addition to Ir-(H)3 (Scheme 16, bottom). This reactivity constitutes an interesting example of metal-ligand cooperativity where insertion of CO2 into the Ir–H bond is coupled to a β-hydride insertion to form the IrIII-formate species (Ir-OC(O)H) [86].

46

M. R. Tiddens and M.-E. Moret

Scheme 16 Metal-ligand cooperative reactivity of Ir-(H)3 with H2 and CO2 [85, 86]

Scheme 17 Overview of the different coordinative interactions between a transition metal center and L2 [87–89]

The discussed examples show that a π-acceptor olefin ligand in the central position of a pincer ligand can reversibly accept a hydride ligand from the metal it is bound to, opening up possibilities for bifunctional substrate activation as depicted in Fig. 9 (right). The utility of an olefin as a hemilabile moiety (Fig. 9, left) is more apparent in the chemistry of a related ligand family in which the central olefin is connected to the phenylene linkers in a 1,2 fashion instead of the 1,1-connectivity discussed so far. A systematic analysis of the coordination of such ligands to different transition metal centers in different oxidation states was conducted by Iluc and co-workers. The used olefin ligand (L2, Scheme 17) is designed to bind in a κ3(P,C¼C,P) fashion with an η2 coordination of the olefin. Indeed, the pincerlike coordination of L2 is observed for electron-rich transition metals of groups 8, 9, and 10, while no coordination of the central olefin motif is observed for more electronpoor transition metal centers such as FeII and CoII (Scheme 17) [87–91]. In addition,

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

47

Scheme 18 Synthesis of a vinyl-NiII complex formation from L2 and a NiII source and reduction of the vinyl-NiII complex with a hydride source to form the electron-rich Ni0 center [88]

Scheme 19 Reaction of L2Ni0 with MeI to form a cationic methyl NiII complex [88]

halide abstraction at L2FeIICl2 and L2CoIICl2 with NaB(ArF)4 leads to a weak olefin coordination (Scheme 17, right) best described as a π-adduct with weak π-backdonation. Integration of an olefin in a chelating ligand such as L2 enables a flexibility in coordination strength of the central motif that is usually not associated with pincer ligands. The distinction between weak and strong olefin coordination is based on M–C bond distances and C¼C bond elongation. For instance, the C¼C bond length in free L2 is 1.330(4) Å, weak η2 (C,C) coordination of the motif in [L2CoIICl][B(ArF)4] leads to an elongation of the bond to 1.397(6) Å, and strong coordination as in L2CoICl elongates the C¼C bond even more to 1.442(5) Å. Therefore, L2 functions as an adaptive π-acceptor ligand since it stabilizes transition metal centers in different oxidation states by means of withdrawing electron density to various extents. Upon coordination of L2 to NiCl2(dme), formal proton abstraction from the C¼C bond and overall HCl elimination yields a vinyl-NiII complex (Scheme 18) [88]. The formation of similar vinyl pincer complexes is also observed upon coordination of L2 to precursors of PdII and PtII [89]. The vinyl-NiII complex can react with a hydride source (Li[HBEt3]) to initially form a vinyl-Ni-H species. Over time, reductive elimination of the C–H bond leads to the formation of L2Ni0 featuring the metal-bound olefin motif. L2Ni0 can be synthesized directly in a reaction of Ni (COD)2 with L2 [88]. This Ni0 complex adopts a pseudotrigonal-planar geometry in which the trans olefin motif is twisted out of the P–Ni–P plane. An elongated C¼C bond distance of 1.406(5) Å indicates significant π-backdonation from the electronrich Ni0 center to the olefin motif. More importantly, this shows that L2 gives access to electron-rich transition metal centers of low oxidation state. η2 coordination of the olefin motif is also observed in Ni species of higher oxidation state. In a reaction of L2Ni0 with methyl iodine (MeI), a cationic methyl nickel complex ([L2NiIIMe]I, Scheme 19) is formed. An elongation of the C–C bond distance to 1.383(3) Å indicates weaker π-backdonation in this NiII cation than in L2Ni0.

48

M. R. Tiddens and M.-E. Moret

Scheme 20 Protonation of L2MePd0 with HCl affords a vinyl Pd species (left) and synthesis of a PdII species by oxidative addition of H2SiPh2 to L2MePd0 (right) [92]

Scheme 21 Reversible deprotonation of one methyl group in L2MePdCl2 [92]

L2 coordination to (dba)2Pd results in a dimeric product [88]. However, an analogue L2 possessing two methyl substituents on a central Z-olefin (L2Me) affords a monomeric L2MePd0 product (Scheme 20) for which η2 coordination of the olefin motif was verified by X-ray diffraction crystallography [92]. Oxidative addition of the Si–H bond in H2SiPh2 to L2MePd0 results in the formation of a distorted square planar PdII species in which the olefin motif has decoordinated (Scheme 20, right). Hence, L2Me acts as a hemilabile ligand, adapting to the electronic requirements of the Pd center in both oxidation states. In a reaction of L2MePd0 with HCl, initial oxidative addition to form the PdII species is observed similar to the reaction with H2SiPh2. However, upon heating the sample for 2 h at 80 C, a vinyl PdII pincer complex is obtained most likely by β-hydride insertion (Scheme 20, left). Though attempts to deprotonate the vinyl Pd species with a base failed, the transformation in Scheme 20 suggests a promising hydrogen acceptor capability of the π-acceptor olefin motif in L2MePd0. Similar to this, early work by Bennet and co-workers showed the synthesis of a vinyl Rh complex by formal protonation of the olefin backbone in L2RhI(CO)Cl with HCl leading to the formation of a new M–C σ bond [93]. Dehydrohalogenation of the PdII complex L2MePdCl2 was observed in a reaction with benzyl potassium (PhCH2K). The η1-allyl PdII product (Pd-η1-allyl, Scheme 21) contains a terminal olefin motif which was formed by deprotonation of one methyl group as evident from a new set of doublet of doublet signals at 4.85 ppm and 4.63 ppm in 1H NMR. In addition, an X-ray crystal structure of this asymmetric species confirms the double bond character of the new motif (C–C bond distance, 1.366(5) Å; Fig. 10). Facile protonation of Pd-η1-allyl with HCl cleanly forms L2MePdCl2 again, showing the deprotonation to be reversible. Overall, a variety of stoichiometric processes involving metal-ligand cooperativity at the central olefin position of pincer ligands have been discussed. Specifically, these include several examples of facile and reversible β-hydride

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands Fig. 10 X-ray crystal structure of Pd-η1-allyl showing the new C2¼C4 double bond motif (thermal ellipsoids at 50% probability). Hydrogen atoms and iPr-groups on the phosphorus atoms (except for the olefin-bound carbon atoms) are omitted for clarity [92]

49

C4 P1

C2

C1 C3

Pd1 Cl1 P2

elimination/migratory insertion in which the olefin transiently stores a hydride equivalent. In addition, metal-centered reactivity can be facilitated by (de)coordination of the olefin moiety. Such processes have the potential to become part of catalytic cycles in further investigations. In the next section, the role of a π-acceptor ketone motif in metal-ligand cooperative catalysis is highlighted.

3.3

Metal-Ligand Cooperative Catalysis Induced by Side-On Coordination of a Ketone

Due to the electronegativity of oxygen, both the π(C, O) and the π*(C, O) orbital of a ketone are lower in energy than those of an olefin. This can be anticipated to render side-on bound ketones both weaker donors and stronger acceptors than olefins. In addition, the lone pairs on the oxygen atom in the ketone motif offer an additional position for reactivity and metal-ligand cooperativity. However, free ketones preferentially coordinate end-on η1(O) to most transition metal centers, while side-on η2(C,O) coordination is required for a ketone motif to act as a π-acceptor ligand (Fig. 11). Incorporation of the ketone motif into a rigid pincer design featuring o-phenylene linkers brings the motif into close proximity of the transition metal center in a pre-oriented geometry favoring side-on binding. The phosphine-tethered ketone ligand L3 (Fig. 11, box) was first reported by Ding and co-workers, who used its Ru complexes in the catalytic hydrogenation of ketones [94]. While itself achiral, L3 was proposed to enhance enantioselectivity by mechanically transferring chiral information from a chiral diamine ligand onto the Ru-bound substrate. In addition,

50

M. R. Tiddens and M.-E. Moret

O

M

O

O

M

end-on η 1(O) sigma-donor

O

Mn

Mn+2

PR2 O

side-on η 2(C,O) π-acceptor

PR2

L3

Fig. 11 End-on η1(O) and side-on η2 (C,O) coordination modes of ketones to transition metal center; the phosphine-tethered ketone ligands L3 (box)

Bifunctional activity

Hemilability O

Mn L

L -L

O

Mn

O

Mn+2

+ H2

HO

M H

Fig. 12 Resonance extremes of a η2 (C,O)-coordinated ketone motif and their prototypical cooperative reactivity

Fig. 13 The phosphine-tethered ketone ligand L2 and its coordination chemistry to Ni0, NiI, and NiII [95]

side-on coordination of the ketone motif was determined by X-ray crystallography, and this carbonyl coordination to the catalytically active RuII species was suggested to be essential for high yields and selectivity. In the following, the coordination chemistry of L3 to late transition metal centers and metal-ligand cooperative catalysis using these L3-TM complexes is presented. First, the hemilabile coordination behavior (Fig. 12, left) of L3 is discussed, as well as its implications for catalysis. Second, the ability of the ketone motif to act a hydride relay is examined in the context of bifunctional H2 activation (Fig. 12, right). Moret and co-workers studied the coordination of L3 to a redox series of Ni (Ni0, I Ni , and NiII) [95]. The ligand binds in a κ3(P,C¼O,P) fashion with η2(C,O) coordination of the ketone motif to the electron-rich Ni0 and NiI centers but adopts a κ2(P,P) mode with the electron-poor, high-spin NiII center, thereby adapting its coordination mode to the electronic structure of nickel (Fig. 13). In addition, NBO analysis on optimized geometries of all three Ni species indicated significant charge

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

51

Fig. 14 Resonance structures of MI ketone complexes [96]

transfer from the Ni center to the ketone, supporting its description as an acceptor moiety. This coordination behavior of L3 is rather general for late first-row transition metal centers. The ketone motif does not coordinate to electron-poor FeII, CoII, and NiII [96], but side-on coordination is detected in electron-rich transition metal complexes of L3 (Ni0 [95], NiI, FeI, CoI [97], Pd0 [98], RhI [99], and RuII [94]). Interestingly, in isostructural NiI, CoI, and FeI complexes of L3, an increase in the amount of charge transfer upon binding (longer C–O distance) correlates with a longer M–C and a shorter M–O bond distance [96]. The opposite would be expected from the increase of π-backdonation into the primarily carbon-centered π* orbital. This apparent discrepancy was rationalized by proposing a (minor) contribution of a third resonance structure involving a ketyl radical interacting with MII in addition to the resonance extremes of the DCD model (Fig. 14). Though small, the increasing contribution of this ketyl resonance structure in the trend from NiI to FeI would account for a stronger ionic M–O bond and a weaker M–C bond while maintaining an increasing electron donation to the motif. The consequences of the observed hemilability of L3 were investigated using the Ni-catalyzed alkyne cyclotrimerization reaction as a benchmark reaction. Under optimized conditions, the Ni complex L3Ni0(BPI) (BPI ¼ benzophenone imine, a labile co-ligand) converts terminal alkynes selectively into the corresponding 1,2,4substituted trimerization products (Table 3, entry 1) [100]. The catalysis was tested for six substrates (R ¼ Ph, CO2Me, CH2OMe, CO2Et, 4-F-C6H4, 4-OMe-C6H4) showing a higher yield for an electron-withdrawing substrate. In all investigated cases, at most very small amounts of cyclooctatetraene (COT) by-products are formed. The activity of L3Ni0(BPI) was compared to the performance of Ni complexes featuring a pincer-type trisphosphine (PPP) or a bidentate diphosphine ether (POP) supporting ligand (Table 3, entries 2 and 3). L3Ni0(BPI) outcompetes these systems in catalytic activity and selectivity indicating an advantage of a π-acceptor motif for this reaction. To further rationalize the role of the ketone moiety, a mechanistic study relying on experimental and computational data was conducted, and a catalytic cycle was proposed, which is shown in Fig. 15. For the computational work, acetylene was used as a model substrate (R ¼ H), and phenyl substituents replaced the p-tolyl substituents on the P-donor moieties. In the stable, 18 VE L3Ni0(BPI) precatalyst, the ketone moiety masks a coordination site by η2(C,O) coordination as evident from a characteristic chemical shift of the carbonyl triplet signal at 119.0 ppm in 13C NMR. A downfield shift of this

52

M. R. Tiddens and M.-E. Moret

Table 3 Comparative study of the Ni-catalyzed alkyne cyclotrimerization reaction (P4 ¼ P (p-tolyl)2, P1 ¼ PPh2) [100]

Ni-catalyst 1)

R¼ Ph CO2Me

Yield 1,2,4trimer (a) 86.9 90.2

2)

Ph CO2Me

3.1 24.5

3)

Ph CO2Me

2.8 65.0

Yield 1,3,5trimer (b) 3.2 6.3

Yield COTs (c) 0 2.5

Ratio a/b/c 97:3:0 91:7:2

1.9 2.1

0 7.5

62:38:0 72:6:22

0.2 12.3

0 6.5

94:6:0 77:15:8

carbonyl 13C NMR signal to 202.7 ppm indicates that the occupied site is readily freed up to allow alkyne binding (step 1). This ligand exchange results in a Ni0 alkyne complex, which is also the in situ observed resting state of the catalyst. Geometry optimization of the bis(acetylene) Ni complex indicates that the ketone moiety remains decoordinated during the second alkyne uptake (step 2). Remarkably, the oxidative coupling step (step 3) is facilitated by concomitant ketone η2(C,O) coordination, which stabilizes the resulting NiII metallacyclopentadiene intermediate. In the calculated trigonal-bipyramidal structure, the two P-atoms occupy the axial positions, and elongation of the C–O bond indicates a strong interaction of the NiII center with the η2(C,O)-coordinated ketone motif. This is presumably a result of strong σ-donation by the C-atoms of the metallacyclopentadiene into the d-orbital that backdonates into the π*(C,O) orbital, which is parallel to the equatorial plane. As the oxidative coupling step is widely acknowledged to be the rate-determining step of the strongly exothermic

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

53

Fig. 15 Proposed catalytic cycle for the L3Ni0(BPI) catalyzed cyclotrimerization of terminal alkynes, P4 ¼ P(p-tolyl)2 [100, 101]

cyclotrimerization reaction, stabilization of the intermediate directly following the rate-determining step accelerates the overall reaction. Consecutive alkyne coordination and migratory insertion processes lead to the formation of a nickelacycloheptatriene intermediate (step 4) in which the carbonyl has decoordinated to accommodate alkyne coordination [101]. Finally, reductive elimination to obtain the trimerization product (step 5) and ligand exchange with an incoming alkyne substrate (step 6) close the catalytic cycle. Interestingly, the Ni-trimer adduct formation (step 5) is thought to be accelerated by facile ketone coordination as evident from a small activation free energy (ΔGo,{ ¼ +0.8 kcal/mol) [101]. The saturated NiII complex is less likely to insert a fourth equivalent of alkyne to form COTs, which accounts for the increased selectivity of L3Ni0(BPI) for cyclotrimerization products. Overall, the adaptive coordination behavior of the π-acceptor ketone ligand along the reaction coordinate of the alkyne cyclotrimerization reaction explains the enhanced catalytic activity and selectivity of L3Ni0(BPI), making this approach promising for future catalyst development. A pincer ligand featuring a π-binding central unit can also be synthesized in the coordination sphere of a transition metal. In this vein, Iluc and co-workers demonstrated the synthesis of a η2(C,E)-coordinated chalcogen ketones (R2C ¼ E, E ¼ S, Se, Te) in the coordination sphere of a PdII pincer featuring a nucleophilic carbene at the central position (Scheme 22) [98]. The Pd-carbene compound reacted with elemental sulfur, selenium, or tellurium to form new C ¼ E bonds. In contrast, the

54

M. R. Tiddens and M.-E. Moret

Scheme 22 Conversion of a nucleophilic PdII carbene complex to various η2(C,E)-coordinated chalcogenoketone Pd complexes by formal O, S, Se, and Te atom transfer [98]

Scheme 23 Synthesis of the Co carbene pincer Co=C by reductive deoxygenation proposed to occur via a metal-ligand cooperative process [97]

complex did not react with O2; instead, half of an equivalent of nitrobenzene quantitatively yielded the ketone complex (Scheme 22, left). Based on X-ray crystallography, the extent of C ¼ E multiple bonding in the complex decreases from oxygen to tellurium. Significant double bond character is observed for E ¼ O, S, only residual π-bonding is found for E ¼ Se, and the C–Te bond length is typical for a sp3-C–Te single bond. The 13C NMR data shows a similar trend, where the coordinated C¼O bond gives rise to the most downfield resonance (160.4 ppm) and the others shift upfield following the decreasing electronegativity of the chalcogens (S: 114.6 ppm; Se: 112.5 ppm; Te: 102.4 ppm). Further reactivity studies of these new π-ligands incorporating heavier chalcogens would certainly be of interest. The abovementioned conversion of a carbene species into a η2(C,O)-coordinated ketone by formal oxygen atom transfer can also be reversed. The group of Young investigated the reductive deoxygenation of group 9 ketone complexes to form the corresponding carbene species. The five-coordinate, cationic [L3CoI(PMe3)2] [BArF4] complex featuring a η2(C,O)-coordinated ketone moiety was synthesized by coordination of L3 to [Co(PMe3)4][BArF4]. In the presence of H2 gas, reductive deoxygenation of [L3CoI(PMe3)2][BArF4] to form the PCP Co carbene species Co=C is observed (Scheme 23) [97]. The proposed reaction pathway involves homolytic H2 activation to form the dihydride species Co-(H)2, a subsequent insertion of the ketone double bond into one Co–H bond to yield the hydroxylalkyl cobalt-hydride intermediate Co(H)-OH and H2O elimination to obtain the final carbene product Co=C. The two intermediates, Co-(H)2 and Co(H)-OH, are

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

55

Scheme 24 Synthesis of the iridium carbene pincer Ir=C by initial coordination of A to the Ir precursor ([IrCl(COD)]2) and subsequent cooperative H2O elimination, P1 ¼ PPh2 [102]

Scheme 25 Synthesis of Rh¼C from an α-hydroxylalkyl RhI complex (Rh(H)-OH) by reaction with (a) LiHMDS and (b) [H(OEt2)2][BArF4] [99, 103]

proposed for this transformation based on mechanistic studies performed on the corresponding Ir-[102] and Rh-systems [99, 103]. First, the synthesis of the PCP Ir carbene species Ir=C by dehydration of the alcohol ligand (A, Scheme 24) is considered [102]. Here, in the reaction of A with 0.5 equivalent of [IrCl(COD)]2, an α-hydroxylalkyl IrIII complex (Ir(H)-OH) was characterized by in situ low-temperature NMR spectroscopy. The complex is in equilibrium with a η2(C,O) keto Ir dihydride species (Ir-(H)2) as evident from a 13C NMR signal at 132.1 ppm. This indicates that a β-hydride insertion/elimination process reversibly converts Ir-(H)2 into Ir(H)-OH in which the ketone motif in Ir-(H)2 can be seen as a hydride relay. Ir-(H)2 was also observed in situ in the reaction of L3, [IrCl(COD)]2 and H2. Though the subsequent H2O elimination to form Ir=C is not a clean reaction as the carbene species is further reduced by excess H2 gas, the second synthesis route of Ir-(H)2 establishes a connection between L3 and Ir=C. The stepwise synthesis of the PCP Rh carbene species Rh=C proceeds via an isolable α-hydroxylalkyl RhIII hydride species (Rh(H)-OH; Scheme 25) [99]. Rh (H)-OH is synthesized by reaction of A with [RhCl(COD)(PPh3)]. The X-ray structure shows that upon C–H activation, the ligand adopts a mer configuration (Fig. 16). Interestingly, the hydroxyl hydrogen forms a hydrogen bridge to the chloride co-ligand, suggesting a relatively high acidity of this proton. Indeed, the 1 H NMR signal for the hydroxyl proton, located at 7.57 ppm, disappears upon the addition of D2O. Moreover, HCl elimination to form an α-hydroxylalkyl RhI complex (Rh-OH) is observed upon treatment of Rh(H)-OH with LiHMDS. Rh-OH can also be synthesized from L3 and the [RhH(PPh3)4] precursor. Upon protonation of Rh-OH with Brookhart’s acid, Rh=C is immediately formed by H2O elimination [103]. Most likely, the cationic Rh species [Rh(H)-OH]+ is an

56

M. R. Tiddens and M.-E. Moret

P3

O1 H11

Cl1

C1

H12 Rh1 P1

P2

Fig. 16 X-ray crystal structure of Rh(H)-OH showing the α-hydroxylalkyl group (thermal ellipsoids at 50% probability). Hydrogen atoms (except H11 and H12) and phenyl groups on the phosphorus atoms (except for the bound carbon atom) are omitted for clarity [99]

Scheme 26 Cooperative deoxygenation of N2O by Ir = C to form the η2 (C,O)-coordinated ketone Ir complex. Subsequent cooperative deoxygenation involves H2 activation, β-hydride insertion, α-hydroxyl group migration, and H2O elimination to reestablish Ir=C [104–106]

intermediate in this reaction. More generally, α-hydroxylalkyl metal-hydride species are proposed as intermediate in the reductive deoxygenation of L3 to form the corresponding carbene species. A possible mechanism for H2O elimination as last step in the overall reductive deoxygenation reaction has been suggested by Piers and co-workers. For this study, a related Ir carbene pincer compound affords a stoichiometric cycle for the deoxygenation of N2O with H2 [104]. Scheme 26 shows the different transformations starting from a reaction of the Ir carbene complex (Ir=C) with N2O to form the η2(C, O) ketone complex (Ir(C=O)). Subsequent reduction with H2 results in the elimination of the oxygen atom in the form of H2O. The reaction of Ir(C=O) with H2 affords the adduct Ir-(H)2 that exists as a cis isomer (depicted in Scheme 26) and a trans isomer (not depicted), the former being the kinetic product and the latter the thermodynamic product of the reaction. When a H2/D2 gas mixture is used, H/D scrambling to obtain Ir-(H)2, Ir-(D)2, Ir-(HD), and

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

57

Fig. 17 Catalytic deoxygenation of amine- and pyridine N-oxides [107]

Ir-(DH) is mediated by the kinetic cis isomer [105], which was proposed to proceed via a reversible β-hydride insertion into the ketone bond followed by activation of a second hydrogen molecule by the resulting hydroxylalkyl/hydride compound Ir(H)-OH. At high temperatures, the α-hydroxyl group in Ir(H)-OH is thought to migrate to the Ir center forming Ir(OH)=C [105, 106]. A high-energy barrier is expected for this step, which offers an explanation for the high reaction temperature (>100 C) required for the H2O elimination process. The subsequent reductive elimination of H2O from Ir(OH)=C occurs rapidly. The cooperative manner in which the ketone motif operates in these examples of oxygenation and (reductive) deoxygenation demonstrates how π-acceptors can diversify the reactivity pathways of a transition metal complex. The stoichiometric examples have inspired the use of Rh=C as catalyst in the deoxygenation of amine and pyridine N-oxides to form amines and pyridines (Fig. 17) [107]. Isopropanol (iPrOH) proved to be a good hydrogen source compared to H2 or SiHEt3 since it prevents overreduction. Under optimized conditions, a range of amine and pyridine N-oxides were converted into the corresponding amines and pyridines with moderate to excellent yields. Alkyl- and arylamine N-oxides are generally converted in high yields to their desired products. In addition, high yields of quinoline and substituted pyridine products are obtained with a tolerance for electron-withdrawing and electron-donating substituents. Based on stoichiometric reactions, a mechanism for the catalytic transformation of amine N-oxide to amine was proposed (Fig. 18). In the first step, the Rh=C deoxygenates the trimethylamine N-oxide (ONEt3) substrate to form the triethylamine (Et3N) product as well as the Rh-ketone species (Rh(C=O)). In a second step, Rh(C=O) is reductively deoxygenated by a reaction with iPrOH to close the catalytic cycle. Under the same catalytic conditions, Rh(C=O) was also used as catalyst for the deoxygenation of ONEt3, yielding 62% of NEt3 (vs 98% for Rh=C). The lower productivity can be ascribed to the required deoxygenation of Rh (C=O) to Rh=C prior to the first catalytic turn over. Overall, formal oxygen atom transfer reactions interconverting a transition metalcarbene complex and a η2(C,O) bound ketone complex were employed in stoichiometric and catalytic deoxygenation reactions. These examples establish a proof of concept involving a side-on coordinated ketone motif as hydride relay. The discovery of this novel metal-ligand cooperative mode is promising for future development of homogeneous catalysts.

58

M. R. Tiddens and M.-E. Moret

Fig. 18 Proposed catalytic cycle for the deoxygenation of amine and pyridine N-oxides [107]

Scheme 27 Synthesis of Co=C species by direct oxygen atom transfer [108]

Recently, Young and co-workers reported on the synthesis of Co carbene species by direct oxygen atom transfer (Scheme 27) [108]. Two Co ketone complexes were synthesized incorporating either one bidentate dppm (1,1-bis(diphenylphosphino)methane) or two monodentate PMe3 co-ligands. Upon heating the [L3CoI(PMe3)2] [BArF4] complex, a carbene motif is formed in the product Co=C (1) as the oxygen atom migrates to one of the pincer flanking phosphine groups. The mechanism of this oxide transfer was investigated by DFT, where, surprisingly, the first step involves one pincer flanking phosphine group decoordinating from the Co center. The subsequent oxygen atom transfer step is exergonic and proceeds through a single transition state featuring relative short Co–C and P–O bond distances, indicating concomitant formation of the M¼C and P–O bonds. It was additionally hypothesized that a bidentate phosphine co-ligand could function as a sacrificial oxygen acceptor via hemilabile dissociation. Indeed, in the presence of a second equivalent of dppm, the desired Co=C (2) species is obtained from [L3CoI(dppm)] [BArF4] by direct oxygen atom transfer to dppm.

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

3.4

59

Imine Side-On Coordination: Synthesis and Metal-Ligand Cooperative Reactivity

Imines can also act as π-acceptor ligands when coordinating side-on to a transition metal center. As for ketones, the π(C, N) and the π*(C, N) orbitals of an imine are generally lower in energy compared to olefins, making a side-bound imine motif a stronger π-acceptor and a weaker donor ligand. Furthermore, the lone pair on the nitrogen atom in the imine motif represents an extra position for additional reactivity and metal-ligand cooperativity. While there is an abundance of examples showing η1(N) coordination of imines to transition metal (Fig. 19, left), η2(C,N) side-on coordination of imines (Fig. 19, right) to a transition metal is less frequently observed. Incorporation of the imine motif into a rigid pincer ligand design can be used to encourage η2(C,N) coordination, enabling the study of imine motifs as π-acceptor ligands. The phosphine-tethered imine ligand (L4, [109] Scheme 28) can access two distinct binding modes. A η1(N)-coordination of L4 to electron-poor transition metal centers such as CoII [110, 111], NiII [110, 112], and PdII [111, 112] is observed (Scheme 28, left), while a side-on η2(C,N) coordination to Ni0 is preferred (Scheme 28, right). X-ray diffraction analysis of the L4Ni0(PPh3) species shows an elongated C–N bond suggesting substantial metallacycle character of the M–C–N interaction. In 13C NMR spectra, the characteristic signal of the imine carbon shifts significantly from 160 ppm in free L4 to 84 ppm in L4Ni0(PPh3), indicating the rehybridization of the imine motif from sp2 to sp3. Metal-ligand cooperative processes employing π-acceptor imine ligands are fairly unexplored and L4 has attractive properties for such investigations. For instance, L4 is suited to electronically stabilize electron-rich transition metal centers of low oxidation states by coordinating η2(C,N) to the metal center. Moreover, L4 coordinates as an adaptive ligand, changing its hapticity according to the electronic properties of the metal center (Fig. 20, left). In addition, bifunctional substrate

Fig. 19 End-on η1(N) and side-on η2 (C,N) coordination modes of an imine to a transition metal center

Scheme 28 L4 coordinates end-on η1(N) to NiII and side-on η2 (C,N) to Ni0 [113]

60

M. R. Tiddens and M.-E. Moret

adaptive coordination N

Mn+2

- 2e-

Bifunctional activity N

Mn

N

Mn+2

XY

N Y

Mn+2 X

Fig. 20 Resonance extremes of a η2 (C,N)-coordinated imine and their prototypical cooperative reactivity

activation at the metal-imine interaction could be imagined, where the imine double bond inserts into a M–Y bond (Fig. 20, right). Upon coordination of L4 to Ni0 in the presence of 1 atm CO gas, the oxidative coupling of two imine motifs is observed (Scheme 29) [113]. A dimeric species of mixed valence is formed, suggesting that CO traps a reactive [Ni0] complex of L4. Therefore, the PPh3 co-ligand plays an important role in stabilizing a reactive, monomeric species. By comparison, the related olefin complex L2Ni0 (Scheme 18) [88] features iPr-substituents on the phosphine linkers which give sufficient steric encumbrance to obtain monomeric species. In the case of the sterically less encumbered reactive [Ni0] species, however, dimerization and redox processes take place instead. Bifunctional activation of a Si–H bond was observed upon reaction of L4Ni0(PPh3) with Ph2SiH2, resulting in hydrosilylation of the imine bond [114]. The hydrosilazane product (Scheme 30) was initially characterized by multinuclear NMR spectroscopy, revealing, interestingly, that the remaining Si–H bond is σ-coordinated to the Ni0 center. This η2(Si–H) coordination to the Ni center was confirmed in an X-ray crystal structure of a structurally analogous hydrosilazane compound resulting from the reaction of L4Ni0(PPh3) and phenyl-methylsilane (PhMeSiH2; Fig. 21). DFT calculations found a transition state for the Si–H bond activation in which the oxidative addition of the Si–H bond and the β-hydride insertion into the imine double bond to proceed in a concerted step (Scheme 30, middle). Therefore, a ligand-to-ligand hydride transfer mechanism is suggested, illustrating the ability of the imine ligand to facilitate bond activation processes by acting as a hydride acceptor moiety. A relatively weak N–Si interaction as indicated by a relatively long N–Si distance (2.3266(5) Å) prompted an investigation into the reactivity of the hydrosilazane complex by means of silane scrambling experiments. Treatment of L4Ni0(PPh3) with deuterated diphenylsilane (Ph2SiD2) established a C–D bond in the ligand backbone, which does not exchange with the Si–H bonds of added hydrosilanes, indicating that hydrosilylation is irreversible. When exposed to phenyl-methylsilane (PhMeSiH2; Scheme 31), the N–(SiPh2D) fragment is partially exchanged for N– (SiMePhH) with concomitant formation of Ph2SiHD, indicating facile and reversible cleavage of the N–Si bond. Hence, the system appears to convert from an initial stoichiometric hydride acceptor to a more reactive silyl reservoir by formal hydrosilylation of the π-acceptor imine motif. This suggests an intriguing strategy

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

61

Scheme 29 Reaction of L4 with a Ni0 precursor in the presence of CO gas leads to a dimeric species which is schematically drawn to emphasize the new C–C bond [113]

Scheme 30 Bifunctional Si–H bond activation by a L4Ni0(PPh3) complex is thought to proceed through a concerted transition state via ligand-to-ligand hydride transfer [114]

N1 Si1 C7 H1 Ni1 P2

P1 P3

Fig. 21 X-ray crystal structure of the hydrosilazane complex showing the σ-coordinated Si-H bond (thermal ellipsoids at 50% probability). Hydrogen atoms (except the hydride) and phenyl groups on the phosphorus atoms (except for the bound carbon atoms) are omitted for clarity [114]

Scheme 31 Schematic representation of the silane scrambling experiment [114]

62

M. R. Tiddens and M.-E. Moret HH N

Ts

PiPr2 Ni NH3

NH3

PiPr2

PhINTs (iPr)2P

Ni Br

PiPr2

N

PiPr2 Ni Br PiPr2

counter anion: [SbF6]

Scheme 32 Formation of a metalloaziridine Ni complex by reaction of a Ni carbene with NH3 (left) and formation of a η2(C,N) bound imine motif by reaction of the Ni carbene with (tosylimino)phenyl-λ3-iodane (PhINTs) [115]

to generate reactive species by cooperative cleavage of element-hydrogen bonds using the L3Ni0(PPh3) or related acceptor pincer systems. Apart from the presently discussed example, metal-ligand cooperative systems employing a π-acceptor imine pincer ligand are scarce. This is likely due to the propensity of imine ligands to form η1(N) complexes, requiring subtle ligand design. As mentioned before, side-on coordinated ketone motifs can be synthesized in the coordination sphere of a transition metal by formal oxygen atom transfer processes to a carbene complex. Similar reactivity was observed by Piers and co-workers who reported on a Ni carbene complex which upon reaction with (tosylimino)phenyl-λ3iodane (PhINTs) forms a new imine bond (Scheme 32, right) [115]. The imine double bond has a bond length of 1.354(4) Å, indicating a strong coordination of the imine to Ni. The coordination resembles the metallacycle extreme of the DCD model. While reaction with PhINTs leads to a new π-complex, reaction of the Ni carbene with ammonia (NH3) leads to a protonated metalloaziridine (Scheme 32, left). The product features a new C–N single bond (1.440(4) Å) which is with a Ni–N distance of 2.050(3) Å in close proximity of the Ni center. This product formally arises from coordination of NH3 to the carbene followed by deprotonation by a second equivalent of NH3 and release of [NH4]Br. A third NH3 equivalent then occupies the empty coordination side at Ni. Contrary to the reported work on reductive deoxygenation of η2 (C,O)-coordinated ketone complexes (Sect. 3.3), a reversible transformation between a carbene complex and the corresponding η2 (C,N)-coordinated imine complex has not been established so far. Future research might be aimed at investigating this possibility as well as the general proposed ability of the imine to act as hydride relay.

4 Concluding Remarks The majority of pincer complexes, featuring a strong donor ligand in the central position, behave as a rigid ligand framework. In contrast, the incorporation of a central σ- or π-acceptor motif affords more flexible pincer ligands, giving access to a variety of different binding modes originating from the – generally weak – metal-

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

63

acceptor ligand interaction. Besides the different symmetry of the accepting orbital, a significant difference between σ- and π-acceptors is the necessary presence, in the latter, of a (weakly) donating π orbital that contributes to the bonding to the transition metal. As a consequence, σ-acceptor motifs are likely more electron-withdrawing than π-acceptors and hence may have a stronger impact on the electronic properties (e.g., Lewis acidity) of the transition metal. On the other hand, π-acceptors open a wide spectrum of opportunities for metal-ligand cooperation due to the accessibility of related structures such as vinyl- (C¼C) and carbene species (C ¼ E, E ¼ C, N, O) and to the presence of additional lone pairs (π-ketone and π-imine). In recent years, the field of acceptor pincer ligands has expanded rapidly from reports on unique coordination behavior and studies on stoichiometric cooperative reactivity to catalytic examples demonstrating the value and potential of the different classes of cooperative acceptor ligands for homogeneous catalysis. The “inverted” polarity of the metal-acceptor interaction manifests itself in two main classes of cooperative processes. First, the accessible empty orbital can reversibly accept electron density (hemilability), increasing the range of electronic structures accessible to a given transition metal center. This adaptive coordination can be expanded further with the participation of neighboring groups, such as boron-bound aromatic residues or with alternative (donor) binding modes such as the η1(N) mode for imines. This flexibility allows both σ- and π-acceptor ligands to stabilize transition metal centers in a range of formal oxidation states and possibly facilitate (formal) redox processes. In particular, the hemilabile coordination behavior of a π-acceptor ketone motif (L3) has an accelerating effect in the alkyne cyclotrimerization reaction catalyzed by L3Ni0(BPI) [100, 101]. Second, the empty orbital can reversibly accept a nucleophilic fragment, which is frequently a hydride (bifunctional activity). Distinct reactivity pathways occur upon hydride uptake by either σ- or π-acceptor ligands: the former often acts as hydride acceptor by hydride insertion to form a bridging L1-H-TM motif, while the latter undergoes β-hydride insertion to reduce the π-bond. Both of these processes have been observed in the stoichiometric activation of H–H and E–H (E ¼ Si, C, N, O, etc.) bonds. Furthermore, such hydride insertions are a crucial step in two examples of cooperative catalysis discussed in this chapter, namely, the catalytic hydrodechlorination of (hetero)aryl chlorides by the σ-acceptor pincer complex L1Pd0(PPh3) [70] and reductive deoxygenation of amine and pyridine N-oxides catalyzed by the π-acceptor complex [L3RhI(PPh3)][BArF4] [107]. In this way, both cooperative hydride uptake mechanisms have their specific impact. Future work exploiting the ability of acceptor ligands to transiently accept a hydride or other nucleophiles for substrate activation and catalysis is eagerly awaited. In general, the correlation between acceptor pincer ligand coordination and cooperative (catalytic) reactivity of the metal complex constitutes an exciting area for discovery of bond activation processes and catalytic reactions using metal-ligand cooperation.

64

M. R. Tiddens and M.-E. Moret

Acknowledgment This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 715060).

References 1. Karunananda MK, Mankad NP (2017) Cooperative strategies for catalytic hydrogenation of unsaturated hydrocarbons. ACS Catal 7:6110–6119 2. Khusnutdinova JR, Milstein D (2015) Metal–ligand cooperation. Angew Chem Int Ed 54:12236–12273 3. Balaraman E, Gunanathan C, Zhang J, Shimon LJW, Milstein D (2011) Efficient hydrogenation of organic carbonates, carbamates and formates indicates alternative routes to methanol based on CO2 and CO. Nat Chem 3:609–614 4. Semproni SP, Milsmann C, Chirik PJ (2014) Four-coordinate cobalt pincer complexes: electronic structure studies and ligand modification by Homolytic and Heterolytic pathways. J Am Chem Soc 136:9211–9224 5. Shvo Y, Czarkie D, Rahamim Y, Chodosh DF (1986) A new group of ruthenium complexes: structure and catalysis. J Am Chem Soc 108:7400–7402 6. Langer R, Leitus G, Ben-David Y, Milstein D (2011) Efficient hydrogenation of ketones catalyzed by an iron pincer complex. Angew Chem Int Ed 50:2120–2124 7. Chirik PJ (2011) Preface: forum on redox-active ligands. Inorg Chem 50:9737–9740 8. Chirik PJ, Wieghardt K (2010) Radical ligands confer nobility on base-metal catalysts. Science 327:794 9. Arevalo R, Chirik PJ (2019) Enabling two-electron pathways with Iron and cobalt: from ligand design to catalytic applications. J Am Chem Soc 141:9106–9123 10. Dzik WI, van der Vlugt JI, Reek JNH, de Bruin B (2011) Ligands that store and release electrons during catalysis. Angew Chem Int Ed 50:3356–3358 11. Luca OR, Crabtree RH (2013) Redox-active ligands in catalysis. Chem Soc Rev 42:1440–1459 12. Myers TW, Berben LA (2013) Aluminum–ligand cooperative N–H bond activation and an example of dehydrogenative coupling. J Am Chem Soc 135:9988–9990 13. Myers TW, Kazem N, Stoll S, Britt RD, Shanmugam M, Berben LA (2011) A redox series of aluminum complexes: characterization of four oxidation states including a ligand Biradical state stabilized via exchange coupling. J Am Chem Soc 133:8662–8672 14. van der Vlugt JI (2012) Cooperative catalysis with first-row late transition metals. Eur J Inorg Chem 2012:363–375 15. Braunstein P, Naud F (2001) Hemilability of hybrid ligands and the coordination chemistry of oxazoline-based systems. Angew Chem Int Ed 40:680–699 16. Jeffrey JC, Rauchfuss TB (1979) Metal complexes of hemilabile ligands. Reactivity and structure of dichlorobis(o-(diphenylphosphino)anisole)ruthenium(II). Inorg Chem 18:2658–2666 17. Adams GM, Weller AS (2018) POP-type ligands: variable coordination and hemilabile behaviour. Coord Chem Rev 355:150–172 18. Lindner R, van den Bosch B, Lutz M, Reek JNH, van der Vlugt JI (2011) Tunable hemilabile ligands for adaptive transition metal complexes. Organometallics 30:499–510 19. van der Vlugt JI, Pidko EA, Vogt D et al (2008) T-shaped cationic CuI complexes with hemilabile PNP-type ligands. Inorg Chem 47:4442–4444 20. van der Vlugt JI, Pidko EA, Vogt D, Lutz M, Spek AL, Meetsma A (2009) CuI complexes with a noninnocent PNP ligand: selective dearomatization and electrophilic addition reactivity. Inorg Chem 48:7513–7515

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

65

21. Noyori R (2002) Asymmetric catalysis: science and opportunities (Nobel lecture). Angew Chem Int Ed 41:2008–2022 22. Lawrence MAW, Green K-A, Nelson PN, Lorraine SC (2018) Review: pincer ligands – tunable, versatile and applicable. Polyhedron 143:11–27 23. Hetterscheid DGH, van der Vlugt JI, de Bruin B, Reek JNH (2009) Water splitting by cooperative catalysis. Angew Chem Int Ed 48:8178–8181 24. Bouhadir G, Bourissou D (2017) Coordination of Lewis acids to transition metals: Z-type ligands. In: Mingos DMP (ed) The chemical bond III: 100 years old and getting stronger. Springer, Cham, pp 141–201 25. Amgoune A, Bourissou D (2011) σ-acceptor, Z-type ligands for transition metals. Chem Commun 47:859–871 26. You D, Gabbaï FP (2019) Tunable σ-accepting, Z-type ligands for organometallic catalysis. Trends Chem 1:485–496 27. Owen GR (2016) Functional group migrations between boron and metal centres within transition metal–borane and –boryl complexes and cleavage of H–H, E–H and E–E0 bonds. Chem Commun 52:10712–10726 28. Braunschweig H, Dewhurst RD (2011) Transition metals as Lewis bases: “Z-type” boron ligands and metal-to-boron dative bonding. Dalton Trans 40:549–558 29. Kameo H, Nakazawa H (2013) Recent developments in the coordination chemistry of multidentate ligands featuring a boron moiety. Chem Asian J 8:1720–1734 30. Verhoeven DGA, Moret M-E (2016) Metal–ligand cooperation at tethered π-ligands. Dalton Trans 45:15762–15778 31. Bouhadir G, Bourissou D (2016) Complexes of ambiphilic ligands: reactivity and catalytic applications. Chem Soc Rev 45:1065–1079 32. van Koten G, Timmer K, Noltes JG, Spek AL (1978) A novel type of Pt–C interaction and a model for the final stage in reductive elimination processes involving C–C coupling at Pt; synthesis and molecular geometry of [1,N,N0 -η-2,6-bis{(dimethylamino)methyl}-toluene] iodoplatinum(II) tetrafluoroborate. J Chem Soc Chem Commun:250–252 33. Choi J, MacArthur AHR, Brookhart M, Goldman AS (2011) Dehydrogenation and related reactions catalyzed by iridium pincer complexes. Chem Rev 111:1761–1779 34. Albrecht M, van Koten G (2001) Platinum group organometallics based on “pincer” complexes: sensors, switches, and catalysts. Angew Chem Int Ed 40:3750–3781 35. Kirchner K (2015) Pincer and pincer-type complexes applications in organic synthesis and catalysis. Edited by Kálmán J. Szabó and Ola F. Wendt. Angew Chem Int Ed 54:4706–4707 36. Allen KE, Heinekey DM, Goldman AS, Goldberg KI (2013) Alkane dehydrogenation by C–H activation at iridium(III). Organometallics 32:1579–1582 37. Tanaka R, Yamashita M, Nozaki K (2009) Catalytic hydrogenation of carbon dioxide using Ir (III)pincer complexes. J Am Chem Soc 131:14168–14169 38. Martín M, Sola E (2020) Chapter two – recent advances in the chemistry of group 9 – pincer organometallics. In: Pérez PJ (ed) Advances in organometallic chemistry. Academic Press, pp 79–193 39. Morales-Morales D (2018) Pincer compounds chemistry and applications. Elsevier, Amsterdam 40. Li H, Gonçalves TP, Lupp D, Huang K-W (2019) PN3(P)-pincer complexes: cooperative catalysis and beyond. ACS Catal 9:1619–1629 41. van Koten G, Milstein D (2013) Organometallic pincer chemistry. Springer, Berlin 42. Asay M, Morales-Morales D (2015) Non-symmetric pincer ligands: complexes and applications in catalysis. Dalton Trans 44:17432–17447 43. Lyaskovskyy V, de Bruin B (2012) Redox non-innocent ligands: versatile new tools to control catalytic reactions. ACS Catal 2:270–279 44. Gunanathan C, Milstein D (2011) Metal–ligand cooperation by aromatization– dearomatization: a new paradigm in bond activation and “Green” catalysis. Acc Chem Res 44:588–602

66

M. R. Tiddens and M.-E. Moret

45. Schneider S, Meiners J, Askevold B (2012) Cooperative aliphatic PNP Amido pincer ligands – versatile building blocks for coordination chemistry and catalysis. Eur J Inorg Chem 2012:412–429 46. Alig L, Fritz M, Schneider S (2019) First-row transition metal (De)hydrogenation catalysis based on functional pincer ligands. Chem Rev 119:2681–2751 47. van der Vlugt JI, Reek JNH (2009) Neutral tridentate PNP ligands and their hybrid analogues: versatile non-innocent scaffolds for homogeneous catalysis. Angew Chem Int Ed 48:8832–8846 48. Budweg S, Junge K, Beller M (2020) Catalytic oxidations by dehydrogenation of alkanes, alcohols and amines with defined (non)-noble metal pincer complexes. Cat Sci Technol 10:3825–3842 49. Tondreau AM, Atienza CCH, Weller KJ, Nye SA, Lewis KM, Delis JGP, Chirik PJ (2012) Iron catalysts for selective anti-Markovnikov alkene hydrosilylation using tertiary silanes. Science 335:567 50. Green MLH (1995) A new approach to the formal classification of covalent compounds of the elements. J Organomet Chem 500:127–148 51. Hill AF, Owen GR, White AJP, Williams DJ (1999) The sting of the scorpion: a metallaboratrane. Angew Chem Int Ed 38:2759–2761 52. Bontemps S, Bouhadir G, Gu W, Mercy M, Chen CH, Foxman BM, Maron L, Ozerov OV, Bourissou D (2008) Metallaboratranes derived from a triphosphanyl–borane: intrinsic C3 symmetry supported by a Z-type ligand. Angew Chem Int Ed 47:1481–1484 53. Bourissou D, Bouhadir G (2016) Ambiphilic ligands: unusual coordination and reactivity arising from Lewis acid moieties. In: Stradiotto M, Lundgren RJ (eds) Ligand design in metal chemistry: reactivity and catalysis. Wiley, New York, pp 237–269 54. Bontemps S, Bouhadir G, Dyer PW, Miqueu K, Bourissou D (2007) Quasi-Thermoneutral P ! B interactions within Di- and tri-phosphine boranes. Inorg Chem 46:5149–5151 55. Sircoglou M, Bontemps S, Bouhadir G et al (2008) Group 10 and 11 metal boratranes (Ni, Pd, Pt, CuCl, AgCl, AuCl, and Au+) derived from a triphosphineborane. J Am Chem Soc 130:16729–16738 56. Bontemps S, Gornitzka H, Bouhadir G, Saffon N, Miqueu K, Gu W, Mercy M, Chen CH, Foxman BM, Maron L, Ozerov OV, Bourissou D (2006) Rhodium(I) complexes of a PBP ambiphilic ligand: evidence for a metal!borane interaction. Angew Chem Int Ed 45:1611–1614 57. Harman WH, Peters JC (2012) Reversible H2 addition across a nickel–borane unit as a promising strategy for catalysis. J Am Chem Soc 134:5080–5082 58. Sircoglou M, Bontemps S, Mercy M, Miqueu K, Ladeira S, Saffon N, Maron L, Bouhadir G, Bourissou D (2010) Copper(I) complexes derived from mono- and diphosphino-boranes: Cu!B interactions supported by arene coordination. Inorg Chem 49:3983–3990 59. Schindler T, Lux M, Peters M, Scharf LT, Osseili H, Maron L, Tauchert ME (2015) Synthesis and reactivity of palladium complexes featuring a diphosphinoborane ligand. Organometallics 34:1978–1984 60. Cao Y, Shih W-C, Bhuvanesh N, Ozerov OV (2019) Silver halide complexes of a borane/bis (phosphine) ligand. Dalton Trans 48:9959–9961 61. Sircoglou M, Bontemps S, Mercy M, Saffon N, Takahashi M, Bouhadir G, Maron L, Bourissou D (2007) Transition-metal complexes featuring Z-type ligands: agreement or discrepancy between geometry and dn configuration? Angew Chem Int Ed 46:8583–8586 62. Tauchert ME, Steinhoff P (2016) A T-shape diphosphinoborane palladium(0) complex. Beilstein J Org Chem 12:1573–1576 63. Inagaki F, Matsumoto C, Okada Y, Maruyama N, Mukai C (2015) Air-stable cationic gold (I) catalyst featuring a Z-type ligand: promoting enyne cyclizations. Angew Chem Int Ed 54:818–822 64. Murakami R, Inagaki F (2019) Recent topics of gold catalyst featuring Z-type ligands. Tetrahedron Lett 60:151231

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

67

65. Harman WH, Lin T-P, Peters JC (2014) A d10 Ni–(H2) adduct as an intermediate in H-H oxidative addition across a Ni-B bond. Angew Chem Int Ed 53:1081–1086 66. Li Y, Hou C, Jiang J, Zhang Z, Zhao C, Page AJ, Ke Z (2016) General H2 activation modes for Lewis acid–transition metal bifunctional catalysts. ACS Catal 6:1655–1662 67. MacMillan SN, Hill Harman W, Peters JC (2014) Facile Si–H bond activation and hydrosilylation catalysis mediated by a nickel–borane complex. Chem Sci 5:590–597 68. Schuhknecht D, Ritter F, Tauchert ME (2016) Isolation and properties of a palladium PBP pincer complex featuring an ambiphilic boryl site. Chem Commun 52:11823–11826 69. Shih W-C, Gu W, MacInnis MC, Timpa SD, Bhuvanesh N, Zhou J, Ozerov OV (2016) Facile insertion of Rh and Ir into a boron–phenyl bond, leading to Boryl/Bis(phosphine) PBP pincer complexes. J Am Chem Soc 138:2086–2089 70. Kameo H, Yamamoto J, Asada A, Nakazawa H, Matsuzaka H, Bourissou D (2019) Palladium–borane cooperation: evidence for an anionic pathway and its application to catalytic hydro-/deutero-dechlorination. Angew Chem Int Ed 58:18783–18787 71. Conifer CM, Law DJ, Sunley GJ, White AJP, Britovsek GJP (2011) Lewis acids and Lewis acid-functionalized ligands in rhodium-catalyzed methyl acetate carbonylation. Organometallics 30:4060–4066 72. Shih W-C, Gu W, MacInnis MC, Herbert DE, Ozerov OV (2017) Boryl/borane interconversion and diversity of binding modes of oxygenous ligands in PBP pincer complexes of rhodium. Organometallics 36:1718–1726 73. Kameo H, Nakazawa H (2012) Synthesis of a rhodium complex featuring the Rh–H–B linkage via a hydride migration from rhodium to borane: study on the electronic deviation induced by the presence of the boron moiety. Organometallics 31:7476–7484 74. Suess DLM, Peters JC (2013) H–H and Si–H bond addition to FeNNR2 intermediates derived from N2. J Am Chem Soc 135:4938–4941 75. Suess DLM, Peters JC (2013) A CO-derived iron dicarbyne that releases olefin upon hydrogenation. J Am Chem Soc 135:12580–12583 76. Nesbit MA, Suess DLM, Peters JC (2015) E–H bond activations and hydrosilylation catalysis with iron and cobalt metalloboranes. Organometallics 34:4741–4752 77. Zeise WC (1827) Poggendorffs Annalen der Physik 9:632–633 78. Mingos DMP (2001) A historical perspective on Dewar’s landmark contribution to organometallic chemistry. J Organomet Chem 635:1–8 79. Comanescu CC, Vyushkova M, Iluc VM (2015) Palladium carbene complexes as persistent radicals. Chem Sci 6:4570–4579 80. Comanescu CC, Iluc VM (2014) Synthesis and reactivity of a nucleophilic palladium (II) carbene. Organometallics 33:6059–6064 81. Gusev DG, Lough AJ (2002) Double CH activation on osmium and ruthenium centers: carbene vs olefin products. Organometallics 21:2601–2603 82. Crocker C, Errington RJ, McDonald WS, Odell KJ, Shaw BL, Goodfellow RJ (1979) Rapid reversible fission of a C–H bond in a metal complex: X-ray crystal structure of [RhHCl (But2PCH2CH2CHCH2CH2PBut2)]. J Chem Soc Chem Commun:498–499 83. Vigalok A, Kraatz H-B, Konstantinovsky L, Milstein D (1997) Evidence for direct trans insertion in a Hydrido-olefin rhodium complex – free nitrogen as a trap in a migratory insertion process. Chem Eur J 3:253–260 84. Jonasson KJ, Polukeev AV, Marcos R, Ahlquist MSG, Wendt OF (2015) Reversible α-hydrogen and α-alkyl elimination in PC(sp3)P pincer complexes of iridium. Angew Chem Int Ed 54:9372–9375 85. Polukeev AV, Marcos R, Ahlquist MSG, Wendt OF (2015) Formation of a C–C double bond from two aliphatic carbons. Multiple C–H activations in an iridium pincer complex. Chem Sci 6:2060–2067 86. Polukeev AV, Wendt OF (2015) Iridium pincer complexes with an olefin backbone. Organometallics 34:4262–4271

68

M. R. Tiddens and M.-E. Moret

87. Barrett BJ, Iluc VM (2014) Coordination of a hemilabile pincer ligand with an olefinic backbone to mid-to-late transition metals. Inorg Chem 53:7248–7259 88. Barrett BJ, Iluc VM (2014) Group 10 metal complexes supported by pincer ligands with an olefinic backbone. Organometallics 33:2565–2574 89. Bennett MA, Clark PW (1976) Tridentate chelate π-bonded complexes of rhodium(I), iridium (I), and iridium(III) and chelate σ-bonded complexes of nickel(II), palladium(II), and platinum (II) formed by intramolecular hydrogen abstraction reactions. J Organomet Chem 110:367–381 90. Baratta W, Herdtweck E, Martinuzzi P, Rigo P (2001) Carboncarbon double bond formation from two o-methyl groups in osmium phosphine complexes. Organometallics 20:305–308 91. Campos J, Ortega-Moreno L, Conejero S, Peloso R, López-Serrano J, Maya C, Carmona E (2015) Reactivity of cationic Agostic and Carbene structures derived from platinum (II) metallacycles. Chem Eur J 21:8883–8896 92. Barrett BJ, Iluc VM (2017) Metal-ligand cooperation between palladium and a diphosphine ligand with an olefinic backbone. Inorg Chim Acta 460:35–42 93. Bennett MA, Johnson RN, Tomkins IB (1976) Additions to metal atom and to coordinated ligand in complexes of rhodium(I) and iridium(I) formed by a tridentate olefinic ditertiary phosphine: chelate olefin complexes and σ-alkyls of rhodium(III) and iridium(III). J Organomet Chem 118:205–232 94. Jing Q, Sandoval CA, Wang Z, Ding K (2006) Complete chiral induction from Enantiopure 1,2-diamines to benzophenone-based achiral Bisphosphane ligands in Noyori-type RuII catalysts. Eur J Org Chem 2006:3606–3616 95. Saes BWH, Verhoeven DGA, Lutz M, Klein Gebbink RJM, Moret ME (2015) Coordination of a diphosphine–ketone ligand to Ni(0), Ni(I), and Ni(II): reduction-induced coordination. Organometallics 34:2710–2713 96. Verhoeven DGA, van Wiggen MAC, Kwakernaak J, Lutz M, Klein Gebbink RJM, Moret ME (2018) Periodic trends in the binding of a phosphine-tethered ketone ligand to Fe, Co, Ni, and Cu. Chem Eur J 24:5163–5172 97. Sung S, Wang Q, Krämer T, Young RD (2018) Synthesis and reactivity of a PCcarbeneP cobalt(i) complex: the missing link in the cobalt PXP pincer series (X ¼ B, C, N). Chem Sci 9:8234–8241 98. Rothstein PE, Comanescu CC, Iluc VM (2017) Formation of palladium η2-bound Chalcogenoketones across a Pd+C bond. Chem Eur J 23:16948–16952 99. Sung S, Boon JK, Lee JJC, Rajabi NA, Macgregor SA, Krämer T, Young RD (2017) Convergent (De)hydrogenative pathways via a rhodium α-hydroxylalkyl complex. Organometallics 36:1609–1617 100. Orsino AF, Gutiérrez del Campo M, Lutz M, Moret M-E (2019) Enhanced catalytic activity of nickel complexes of an adaptive diphosphine–benzophenone ligand in alkyne cyclotrimerization. ACS Catal 9:2458–2481 101. Orsino AF, Moret M-E (2020) Nickel-catalyzed alkyne cyclotrimerization assisted by a hemilabile acceptor ligand: a computational study. Organometallics 39:1998–2010 102. Sung S, Young RD (2017) Facile generation of iridium PCcarbeneP pincer complexes via water elimination from an alcohol proligand. Dalton Trans 46:15407–15414 103. Sung S, Joachim T, Krämer T, Young RD (2017) Protonolysis of an α-hydroxyl ligand for the generation of a PCcarbeneP pincer complex and subsequent reactivity studies. Organometallics 36:3117–3124 104. Doyle LE, Piers WE, Borau-Garcia J (2015) Ligand cooperation in the formal hydrogenation of N2O using a PCsp2P iridium pincer complex. J Am Chem Soc 137:2187–2190 105. Doyle LE, Piers WE, Borau-Garcia J, Sgro MJ, Spasyuk DM (2016) Mechanistic studies on the addition of hydrogen to iridaepoxide complexes with subsequent elimination of water. Chem Sci 7:921–931 106. Doyle LE, Piers WE, Bi DW (2017) Cationic PCP iridaepoxide and carbene complexes for facile water elimination and activation processes. Dalton Trans 46:4346–4354

Metal-Ligand Cooperation at Phosphine-Based Acceptor Pincer Ligands

69

107. Tinnermann H, Sung S, Cala BA, Gill HJ, Young RD (2020) Catalytic deoxygenation of amine and pyridine N-oxides using rhodium PCcarbeneP pincer complexes. Organometallics 39:797 108. Sung S, Tinnermann H, Krämer T, Young RD (2019) Direct oxide transfer from an η2-keto ligand to generate a cobalt PCcarbeneP(O) pincer complex. Dalton Trans 48:9920–9924 109. Ainscough EW, Brodie AM, Buckley PD, Burrell AK, Kennedy SMF, Waters JM (2000) Synthesis, structure and kinetics of group 6 metal carbonyl complexes containing a new ‘P2N’ mixed donor multidentate ligand. J Chem Soc Dalton Trans:2663–2671 110. Hou J, Sun WH, Zhang S, Ma H, Deng Y, Lu X (2006) Synthesis and characterization of tridentate nickel complexes bearing P^N^N and P^N^P ligands and their catalytic property in ethylene oligomerization. Organometallics 25:236–244 111. Chen L, Ai P, Gu J, Jie S, Li BG (2012) Stereospecific polymerization of 1,3-butadiene catalyzed by cobalt complexes bearing N-containing diphosphine PNP ligands. J Organomet Chem 716:55–61 112. Doherty S, Knight JG, Scanlan TH, Elsegood MRJ, Clegg W (2002) Iminophosphines: synthesis, formation of 2,3-dihydro-1H-benzo[1,3]azaphosphol-3-ium salts and N-(pyridin2-yl)-2-diphenylphosphinoylaniline, coordination chemistry and applications in platinum group catalyzed Suzuki coupling reactions and hydrosilylations. J Organomet Chem 650:231–248 113. Verhoeven DGA, Negenman HA, Orsino AF, Lutz M, Moret ME (2018) Versatile coordination and C–C coupling of diphosphine-tethered imine ligands with Ni(II) and Ni(0). Inorg Chem 57:10846–10856 114. Verhoeven DGA, Orsino AF, Bienenmann RLM, Lutz M, Moret ME (2020) Cooperative Si– H addition to side-on Ni(0)-imine complexes forms reactive hydrosilazane complexes. Organometallics 39:623–629 115. LaPierre EA, Piers WE, Gendy C (2018) Redox-state dependent activation of silanes and ammonia with reverse polarity (PCcarbeneP)Ni complexes: electrophilic vs. nucleophilic carbenes. Dalton Trans 47:16789–16797

Top Organomet Chem (2021) 68: 71–94 https://doi.org/10.1007/3418_2020_69 # Springer Nature Switzerland AG 2020, corrected publication 2020 Published online: 26 November 2020

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems Seji Kim, Yeong-Eun Kim, and Yunho Lee

Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Electron and Proton Transfer Between Metal and Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 C-group Transfer Reactions of a M-P Moiety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 O-(S-)group Transfer Reactions of a M-P Moiety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 N-group Transfer Reactions of a M-P Moiety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

72 75 84 86 89 90 91

Abstract A metal-bound phosphorus atom can actively participate in various organometallic reactions as an electron reservoir and/or a group transfer site, because a phosphorus atom can adopt several distinct forms such as cationic phosphenium, anionic phosphide, neutral phosphine, or a phosphinyl radical, thus making it more versatile. Furthermore, the ability of the phosphorus atom to have various oxidation states promotes the phosphorus atom to be cooperatively engaged in the reaction occurring at a metal center. By using such properties of phosphorus, metal-ligand cooperative (MLC) reactions occurring at a P-M moiety embedded in several transition metal pincer systems are discussed in this chapter. Keywords Group transfer · Metal-ligand cooperativity · Phosphorus · Pincer ligand · Redox active

The original version of this chapter was revised. A correction to this chapter can be found at https://doi.org/10.1007/3418_2020_74

S. Kim, Y.-E. Kim, and Y. Lee (*) Department of Chemistry, Seoul National University, Seoul, South Korea e-mail: [email protected]

72

S. Kim et al.

1 Introduction Phosphorus-containing ligands have been widely utilized in various organometallic reactions and transition metal catalyses. The profound impact of various tertiary phosphines as a neutral donor has been well-recognized, due to their excellent character, versatile enough to play important roles in transition metal catalytic processes, such as hydrogenation, hydroformylation, and polymerization. To control the electronic and steric properties of phosphorus-containing ligands, various substituents have been synthetically incorporated. An electronic factor can be easily managed by employing alkyl and aryl groups, as well as heteroatoms to give a P-E bond (E ¼ O, N, or Si). A steric property nicely defined by “cone angle” suggested by Tolman in 1970 [1] can be controlled by utilizing substituents of different sizes. With their relative ease of synthetic preparation, various alkyl and aryl groups have been employed to control the steric and electronic properties of transition metal complexes, which allows to achieve high stereoselectivity. Their advantageous effects have been well perceived from a wide range of organometallic studies. One of the advantages of working with phosphorus-containing organometallic complexes is the convenience of using 31P nuclear magnetic resonance (NMR) spectroscopy. Having the phosphorus nuclear spin of ½ with high sensitivity, its NMR spectroscopic data can be conveniently collected and utilized not only to characterize various diamagnetic species but also to study kinetics in order to obtain a mechanistic understanding of organometallic reactions [2]. Various mono- and multidentate ligands having phosphorus atom(s) as a donor moiety have been synthesized and widely employed. In particular, tridentate pincer systems are an important class of ligand to produce four-coordinate square planar metal complexes, in which various catalytic reactions effectively occur at the site trans to a central moiety. By incorporating heteroatom(s) in central and/or side arm(s) along with P donor(s), the local geometry about a metal center can be finely tuned, and thus it is effective in controlling the electronic structure of a metal complex and regulating the energy of the metal-based frontier molecular orbitals (FMOs). Thus, various phosphorus-containing pincer ligands have been designed and prepared for the last several decades. Most of phosphorus-containing ligands are conventionally categorized as spectator ligands, which do not directly interact with substrates. Although they are tightly connected to the metal center, there is no formal redox change or bond formation/cleavage occurring at a P site during chemical reactions. In contrast, actor ligands cooperatively assist the adjacent metal by providing electrons, protons, and/or functional groups, and thus they are actively involved in chemical transformations. Within phosphoruscontaining pincer systems, several examples reveal metal-ligand cooperativity (MLC), which has been recently recognized as a promising way to expand the role of transition metals in organometallic catalysis. To be actively engaged in chemical reactions, several redox noninnocent ligands have been prepared by employing π-conjugated systems such as porphyrin [3], pyridine [4–6], catecholate [7, 8], aminophenolate [9, 10], and salen [11] as cooperative sites. For example, the Milstein group reported various reactions employing phosphorus-containing

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

73

Scheme 1 Four-coordinate metal complexes supported by phosphorus-containing ligands revealing metal-ligand cooperative activity

pincer-type ligands based on pyridine, acridine, and phenanthroline scaffolds [12]. In their system, a noninnocent moiety of ligands is a π-conjugated system of mostly pyridine and its analogues. There are few phosphorus-containing ligands revealing MLC. In this chapter, various metal-ligand cooperative sites embedded in a tridentate pincer scaffold operating with anionic phosphide ligands and their analogous will be primarily discussed. Electron-rich anionic phosphide ligands are relatively uncommon, while a wide range of neutral alkyl and aryl phosphine ligands are generally employed in organometallic chemistry. A diaryl phosphide PPP ligand (PPP– ¼ bis (2-di-iso-propylphosphinophenyl)phosphide; Scheme 1) reveals an unusual coordination chemistry, especially in their redox activity and group transfer reactions. Recently, the Lee group reported the phosphide-based metal-ligand cooperativity by employing a (PPP)Ni scaffold, in which various group transfer reactions occur at a phosphide-nickel moiety, vide infra. Since a phosphide oxide moiety acts as an ambidentate donor, its cooperativity was also studied with two pincer ligands: {o-iPr2P-(C6H4)}2P–(O) (DPPO ¼ bis(2-di-iso-propylphosphinophenyl)phosphide oxide) and {o-O-(C6H4)}2P–(O) (OPO ¼ bis(2-phenoxyl)phosphinite) (Scheme 1). Besides alkyl and aryl phosphide ligands, a N-heterocyclic phosphenium cation (NHP+ ¼ 1,3-bis(2-diphenylphosphinophenyl)-1,3,2-diazaphosphenium), which has a fairly different donor and acceptor character relative to that of NHCs, has been recently employed in a phosphine pincer ligand, DPNHP (Scheme 1). Electrophilic N-heterocyclic phosphenium cations (NHPs) are fairly different from other conventional spectator phosphine ligands. Having the weak σ-donor and strong π-acceptor properties due to the formal positive charge and isotropic s-character of the lone pair orbital on phosphorus [13], NHPs can be reasonable actor ligands revealing MLC. The Jones group reported Pd and Pt complexes having an NHP coordination and presented a convincing analogy between NHPs and NO+, a wellknown redox-active ligand, as shown in Fig. 1 [14]. An NHP ligand adopts either planar or pyramidal coordination modes about a phosphorus center, which is defined by the angle between the N-P-N plane and a P-M bond vector, which are analogous to the linear and bent binding modes of a nitrosyl ligand. Various coordination chemistries with NHPs have been explored and well-established including new preparative methods and reactivity. A phosphorus atom embedded in the three nitrogen atoms in order to influence the donor property was recently employed as a new P moiety in the NPN pincer ligand (NPN ¼ bis(2-pyridylamidophenyl)amido phosphine P(N(o-N(2-pyridyl)

74

S. Kim et al.

Fig. 1 Interconversions of NO–/NO+ and NHP–/NHP+ along with their coordination modes

Fig. 2 Comparison of structural properties based on single-crystal XRD structures

C6H4)2) (Scheme 1). Owing to their resonance stabilization and the structural flexibility to adopt either planar or pyramidal geometry, the potential noninnocent character of a new phosphorus-containing ligand has been recently recognized. A central anionic phosphide moiety within a pincer system may have a higher chance of being involved in a metal-ligand cooperative reaction relative to an analogous amide group. This may be related to the orbital overlapping between a pz orbital of phosphorus and orbitals of neighboring atoms or groups. For example, an anionic diphosphinoamide ligand (PNP– ¼ N[2-PiPr2-4-Me-C6H3]2–), widely utilized in preparing square planar complexes of group ten metals, tends to form a stable planar nickel(II) complex, as shown in Fig. 2 [15]. The central nitrogen atom

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

75

Fig. 3 Metallophosphoranes

is nicely sited within the plane defined by two carbon atoms and a metal ion, while a P atom of a PPP ligand is located noticeably away from the plane: θ ¼ 1 vs. 34 found in analogous nickel and palladium complexes showing significantly distorted structures. A similar structural feature can be seen from phosphide-containing ligands such as DPNHP. Such structural distortion can affect the reactivity of a metal ion, which can be coupled with a redox change and/or bond formation and cleavage occurring at a P site. Since phosphorus is prone to reveal hypervalency, another class of anionic phosphorus-containing ligands should be mentioned. Analogous to the pentavalent phosphoranes (PR5), metallophosphoranes (LnMPR4; Fig. 3) have been known, in which the phosphorus ligand is considered to be an anionic phosphoranide moiety (PR4–) [16]. After the first crystallographically characterized metallophosphorane was reported in 1981 by Riess and coworkers [17], only ~30 structures possessing a phosphoranide moiety are reported [16]. Various group- and halide-transfer reactions occurring at a M-P moiety are known to involve the formation of metallophosphoranes as intermediates or possible transition states. Although the activity of metallophosphorane displays metal-ligand cooperativity, none of them are related to the pincer ligand system; thus, this chapter does not discuss metallophosphoranes.

2 Electron and Proton Transfer Between Metal and Phosphorus Having an anionic phosphide moiety, a diphosphinophosphide ligand (PPP) is widely utilized in various research groups. The Peters group reported a dicopper complex possessing two bridging phosphido moieties, {(PPP)Cu}2 (1; Fig. 4), which reveals two reversible redox events at –0.42 V and –1.02 V vs. Fc+/Fc in THF [18]. Upon the sequential chemical oxidation with [FeCp2][BArF4] (Cp ¼ cyclopentadienyl), the diamond core of a CuICuI species drastically changes its geometry with the contracted Cu-Cu distances and narrowed Cu-Pμ-Cu angles, according to their X-ray crystallographic data, as shown in Fig. 4 [18]. A PPP ligand of 1 assists a diamond core to retain the geometry about each copper center between square planar and tetrahedral in the three different oxidation states [18]. The X-band electron paramagnetic resonance (EPR) spectroscopic data, collected at 10 K, on the 1e– oxidized species of 1, [{(PPP)Cu}2][BArF4] (2), reveals an

76

S. Kim et al.

Fig. 4 Oxidation of {(PPP)Cu}2 (1) by [FeCp2][BArF4] generates its cationic species [{(PPP)Cu}2][BArF4] (2) and a dicationic species [{(PPP)Cu}2][BArF4]2 (3)

isotropic S ¼ 1/2 signal with a hyperfine splitting originated from the core with two copper centers and the phosphide bridges. This result suggests that the unpaired electron is fairly delocalized within the diamond core of 2. According to the computational studies using density functional theory (DFT), the bridging phosphorus atoms show significant changes in the charge population on the oxidized species of 1 [19]. Multifrequency EPR studies support that the corresponding oxidation occurs at a PPP ligand with a substantial degree [20]. Thus, the central P atoms are actively involved in the redox events of the diamond core of 1. An analogous ligand, a κ1-P(V)-phosphide oxide donor (–P(O)R1R2), should be mentioned. Having highly polar canonical structure P+-O–, an ambidentate P¼O moiety reveals linkage isomerism through the coordination of a P or O donor. In addition, an oxygen atom of a P¼O moiety can be participated in the acid-base reactions, which makes it as an actor ligand. The conversion of an X-type phosphide oxide ligand to a L-type phosphine ligand by protonation is firstly reported by the Bourissou group [21]. The reaction of a diphosphine-phosphine oxide (DPPO ¼ {o-iPr2P-(C6H4)}2P(O)Ph) ligand with Pd(0) results in a Ph-P(O) bond cleavage to generate a Pd(II) complex (DPPO)Pd(Ph) (4) via the direct oxidative addition of a P-C bond involving a three-center P,Cipso,Pd transition state; see Fig. 5 [ 21, 22]. Alternatively, 4 can be generated from the metalation with a phosphonium salt of DPPO. Upon protonation with trifluoromethanesulfonic acid (HOTf), a tertiary phosphenium salt, R3PH+X–, was generated, which is reasonably stable enough to manage. The resulting salt was oxidatively added to Pd2(dba)3 to give a cationic palladium(II) monohydride species possessing a phosphine oxide coordination, κPO(P)P, displaying a 1H NMR peak at –16 ppm; see Fig. 5. The corresponding hydride at a palladium(II) complex was deprotonated by a base, 1,8-diazabicyclo[5.4.0]undec-7ene (DBU). The product 4 having a phosphide oxide coordination, κPP(O)P, was obtained via the formation of a transient Pd(0) species, which leads to an oxidative addition of a P-CPh bond; see Fig. 5. The anionic

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

77

Fig. 5 Metalation of a DPPO ligand and its cooperativity occurring at a P-Pd site

phosphide oxide moiety of 4 is selectively protonated by HOTf to transform into a neutral hydroxyphosphine donor, and the reverse reaction is possible by a deprotonation of a P-OH group with DBU. During acid and base reaction occurring at a P¼O moiety, phosphorus changes its oxidation state between P(IV) and P(III) while a palladium center remains intact. Although this example presents a possibility of a P¼O moiety to be used in acid-base reaction, such reaction is not yet to be applied to assist a palladium center during the transformation of any substrate. More recently, the C-H bond activation mediated by organometallic complexes via the reduction-coupled oxo activation (ROA) principle suggested by the Goddard III group was theoretically explored. This principle involves a proton-/electrondecoupled hydrogen abstraction mechanism occurring at a main group elementoxo moiety with a reducible transition metal. In particular, the coordination of a P¼O moiety to a high-valent vanadium ion was evaluated by the density functional theory (DFT) calculations. The C-H bond activation of various alkanes, such as ethane, propane, and n-butane, can occur at a V-P(O) site, due to the fact that the hydrogen atom abstraction is amenably coupled with the reduction of V(V). Compound (OPO)V(Cl)2 (7, OPO ¼ bis(2-phenoxyl)phosphinite) can activate a C-H bond via the interaction of an anionic P¼O bond with substrate. The formation of a P-OH moiety in 7 via a hydrogen atom transfer (HAT) is subsequently coupled with the reduction of a high-valent V(V) ion to V(IV), which is expected to reveal high reactivity; see Fig. 6 [23, 24]. This ROA principle provides a new interesting strategy to utilize a P¼O/P-OH moiety for transition metal-based C-H activation. An NHP moiety in a diphosphine pincer ligand, DPNHP, shows a redoxnoninnocent character established with group ten transition metals. When a cationic phosphenium ligand (DPNHP)(PF6) is treated with (COD)PtCl2 (COD ¼ 1,5-cyclooctadiene), the electrophilic nature of the central phosphorus atom abstracts chloride from a PtCl2 fragment to form a [{(DPNHP)Cl}PtCl]+ (9; Fig. 7)

78

S. Kim et al.

Fig. 6 The reduction-coupled oxo activation (ROA) mechanism of a (OPO)V scaffold

Fig. 7 Metalation of a DPNHP ligand with platinum and palladium and an example of metalligand cooperative reaction

[25]. Interestingly, its metalation with a zerovalent platinum precursor, Pt(PPh3)4, results in the formation of a four-coordinate platinum complex, {(DPNHP)Pt (PPh3)}{PF6} (10) (Fig. 7) [26]. The coupling constant of the central P bound to a Pt center is 1JPt-P ¼ 445 Hz, which is noticeably smaller than those of other NHP+-Pt (0) complexes (6162–7354 Hz), but comparable to the reported coupling constant of an anionic phosphide coordinated to a Pt(II) center (648–2749 Hz) [26]. The central phosphorus atom adopts a pyramidal geometry, while the platinum center reveals a square planar geometry, as expected. Thus, the monomeric platinum complex is described as a NHP– phosphido platinum(II) having a P-Pt distance of 2.2535(6) Å. Similarly, the reaction of a cationic DPNHP+ ligand with a palladium(0) precursor also produces a phosphido-Pd(II) complex {(DPNHP)Pd(PPh3)}{PF6} (11). In this reaction, additional trimethyl phosphine was needed; otherwise, a Pd(0) dimer bridged by an NHP+ phosphenium ligand is generated, as shown in Fig. 7. The short P-Pd bond length of 2.162(2) Å in the bimetallic complex suggests a double bond character between a central P atom of a NHP+ ligand and a Pd(0) center [26]. Thus, upon addition of PMe3, an NHP moiety in a DPNHP ligand shows a conversion from NHP+ to NHP– accompanying the oxidation of the metal center. Incorporation of aryl rings to an NHP-based pincer system can assist the

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

79

Fig. 8 Redox activity of a DPNHP ligand in nickel complexes

delocalization of the lone pair electrons of the nitrogen atoms through the aromatic rings and thus stabilize the anionic phosphide coordination [26]. Since the central phosphide donor of the anionic NHP moiety has a reasonable basicity, (DPNHP)Pt (II) species, 10 reveals a cooperative behavior, when HX was added. During its reaction with HCl or PhSH, phosphide accepts a proton, while an anionic ligand coordinates to a Pt(II) ion to give a product {(DPNHP)H}Pt(II)X (12, X ¼ Cl or 13, X ¼ SPh) [27]. Treatment of (DPNHP)(PF6) with Ni(COD)2 also generates a bimetallic species, {(DPNHP)Ni}2{PF6}2 (14; Fig. 8) [28]. Relative to an analogous palladium complex, 14 reveals asymmetric interaction between two zerovalent nickel ions and bridging phosphenium ligands. The P-Ni bond length of 2.0437 Å is clearly shorter than the other Ni-P distance of 2.2491(5) Å. The short P-Ni distance and a planar geometry indicate that 14 possesses an NHP+ phosphenium moiety embedded within a dimer species [28]. This dicationic dinickel species shows two reversible reduction waves at –0.81 and –1.15 V vs. Fc+/Fc in THF. From one- or two-electron reduction, 14 maintains its dimeric scaffold. From the reduction with a Ni(0) powder, a monocationic mixed valence dinickel species, {(DPNHP)Ni}2{PF6} (15), was generated showing a sharp isotropic EPR signal at g ¼ 2.04, consistent with a S ¼ 1/2 state. Based on the density functional theory (DFT) calculations, the optimized structure of this monocationic species is similar to that of a dicationic complex 14, and its Mulliken spin density on two nickel centers (0.60 and 0.08) indicates the

80

S. Kim et al.

corresponding reduction occurs at one of the two nickel(0) ions. A two-electron reduced dinickel species, {(DPNHP)Ni}2 (16), is structurally different from that of 15. Although it is also a dimer, compound 16 has a symmetrical core having the Ni-Ni bond length of 2.5144(5) Å and a P-Ni distance of 2.1191(7) Å, which are in the ranges of reported phosphido-bridged NiI complexes (Ni-Ni ¼ 2.37–2.56 Å and Ni-P ¼ 2.16–2.26 Å). Thus, a neutral dinickel compound 16 can be assigned as a Ni (I) dimer bridged by two phosphido NHP– ligands. This is clearly an interesting transformation of a dinickel(0) species 14 induced by the chemical reduction, which involves two electron reduction of a phosphenium ion to a bridging phosphide moiety accompanying a single-electron oxidation of two nickel ions. The reverse one- or two-electron oxidations of a neutral dinickel species 16 result in a formation of a mono- and dicationic species. Compound 15 can be also generated from the 1:1 mixture of 14 and 16. A reaction of a neutral (DPNHP)Cl ligand with Na{Co(CO)4} results in the formation of a cobalt(I) compound (DPNHP)Co(CO)2 (17; Fig. 9) [29]. The pyramidal geometry of a NHP moiety and the relatively longer Co-P bond distance (2.2386(6) Å) imply the presence of a lone pair on the central phosphorus atom. Oxidation of 17 by treating with trimethylamine N-oxide leads to oxidation of a central phosphorus atom to give an unusual metal-bound NHP phosphinito species, (DPNHP¼O)Co(CO)2 (18) [29]. A cooperative behavior of a central phosphorus atom of a DPNHP ligand was also presented with a neutral cobalt(II) compound {(DPNHP)Cl}Co(Cl)2 (19), which was synthesized from the metalation of (DPNHP)Cl with CoCl2 [30]. Upon reduction of 19 by using three equivalents of KC8, a phosphido cobalt(I) species, (DPNHP)Co(PMe3) (20), was formed. The DFT-optimized structure of 20 displays a pyramidal geometry of the central phosphorus coordinated to a distorted square planar cobalt center. An NHP– phosphido

Fig. 9 Syntheses of cobalt complexes using a DPNHP ligand and the reactivity of a (DPNHP)Co scaffold

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

81

Fig. 10 Syntheses of rhodium complexes using a DPNHP ligand and the reactivity of a (DPNHP) Rh scaffold

donor having a nonbonding lone pair can participate in the activation of H2. While a cobalt(I) hydride species is generated, a new P-H bond also forms displaying a P signal at 134 ppm with a P-H coupling (1JP-H ¼ 320 Hz) in its 31P NMR spectrum, which is largely upfield shifted from the signal of 17 appearing 251 ppm. During the heterolysis of H2 occurring at a Co-P moiety, a phosphide moiety clearly acts as an internal base to give {(DPNHP)H}Co(H)(PMe3) (21). Similar to the cobalt complexes supported by a DPNHP ligand, a series of analogous rhodium complexes are recently reported as shown in Fig. 10 [31]. Depending on the reduction conditions, three different types of (DPNHP)Rh (I) species were synthesized from (DPNHP-Cl)Rh(Cl) (22) prepared from the reaction of a chlorophosphine ligand (DPNHP)Cl with [Rh(COD)Cl]2. While the reduction of 22 with methyldiphenylphosphine in THF generates a mononuclear Rh (I) complex (DPNHP)Rh(PMePh2) (23), its reduction with tert-butyl isocyanide produces a dimer {(DPNHP)Rh(CNtBu)}2 (24) having an asymmetric diamond core. The same product can be prepared from the exposure of 23 toward CNtBu. Without any additional ligand, the corresponding reduction produced {(DPNHP)Rh (CNtBu)}2 (25), suggesting that in the absence of a π-acidic ligand, two DPNHP ligands span two rhodium centers to form a metal-metal bond with the Rh-Rh distance as 2.6712(5) Å, which is clearly shorter than that of the isocyanide adduct, 24 (3.708 Å). Due to the different electronic property, 24 reacts with thiophenol to give {(DPNHP-H)Rh(CNtBu)(SPh)} (27), while 25 does not react with any exogenous ligand including thiophenol. The similar heterolytic cleavage of a S-H bond occurs with 23 to generate a Rh(I) thiolate complex{(DPNHP-H)Rh(SPh)} (26). According to 31P NMR spectroscopic data, both complexes 26 and 27 clearly possess a new P-H bond revealing a doublet with large coupling constants of 1JP-H ¼ 368 Hz at 106 ppm and 1JP-H ¼ 506 Hz at 119 ppm, respectively, ensuring that both reactions undergo via the protonation of the central P atom. Comparing with cobalt congeners, these (DPNHP)Rh(I) species are, however, significantly less

82

S. Kim et al.

Fig. 11 Conversion of PIII to PV by the reaction with ammonia borane and its catalytic cycle for the hydrogenation of azobenzene

reactive toward H2 and phenol, suggesting that the cooperativity of a DPNHP ligand with group 9 metals can be effectively accomplished with first row transition metal. A reversible phosphorus-based two-electron redox cycle without metal interaction was previously studied with 5-aza-2,8-dioxa-3,7-di-tert-butyl-1phosphabicyclo[3.3.0]octa-3,6-diene (ADPO; Fig. 11), which was initially prepared by the Arduengo group [32]. More recently, the Radosevich group reported the unusual T-shaped phosphorous(III) compound operates as hydrogen acceptor during the reaction with ammonia borane to generate a dihydridophosphorane PV compound (Fig. 11) [33]. Hydrogenation of a phosphorus compound PIII was monitored by 31P NMR spectroscopy revealing a dramatic change in the 31P chemical shift from 187 ppm for PIII to –43.7 ppm for PV. Due to two apparently equivalent hydrogen atoms at the phosphorus center, the corresponding signal appears as a triplet of triplet with a coupling constant of 1JP-H ¼ 670 Hz, while a long-range 3JP-H coupling with the remote vinylic hydrogen (3JP-H ¼ 34 Hz) provides an additional splitting. The formation of a five-coordinate PV species from a three-coordinate PIII compound exhibits the ability of the central phosphorus atom as a redox-active moiety. Thus, the dihydridophosphorane PV compound was employed in conversion of an unsaturated organic substrate, such as azobenzene to 1,2-diphenylhydrazine, as depicted in Fig. 11. Compared to normal organic phosphines or phosphorus atoms of frustrated Lewis pair (FLP) operating as an electron donor, the recovery of the threecoordinate PIII compound from PV shows a distinct electrophilic reactivity of the high-valent phosphorus atom. The reversible redox change of the central phosphorus atom between PIII/PV was nicely employed in the catalytic hydrogenation of azobenzene with the phosphorus-based redox-active system, which is welldemonstrated by the Radosevich group as a unique example in phosphine-based redox catalysis.

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

83

Fig. 12 Qualitative frontier molecular orbital diagrams depicting the electronic structure arising from nontrigonal perturbation of a three-coordinate phosphine [34]

Similar to transition metal, the three-coordinate PIII compound cleaves the N-H bond of RNH2 via oxidative addition (Fig. 11) [34]. According to the DFT analysis, the reaction enthalpy for the proton transfer from an amine to PIII (ΔH ¼ 87.7 kcal/ mol) is unfavorable. The presence of low-lying LUMO (–1.0 eV) and electron affinity (EA ¼ –2.2 eV) of PIII suggest that PIII initially forms a P-N bond with amine, which is subsequently followed by P-H bond formation [34]. The DFT analysis on P(NH2)3 as a model compound of PIII reveals that a lowest energy of LUMO is obtained, when the compound has a planarized structure with C2v symmetry relative to other structures having C3v or Cs symmetry (Fig. 12). Consequently, the energy gap between HOMO and LUMO in frontier orbitals of the nontrigonal phosphorus diminishes. This electronic feature allows the central phosphorus atom to reveal the “biphilic” reactivity acting as both donor and acceptor, vide infra [35]. The redox-active tri-coordinate phosphorous compound, P(N(ortho-N(2-pyridyl) C6H4)2, attributed by the nontrigonal geometry can be employed in the transition metal complexes and operated as a cooperative site. When a nontrigonal phosphorous triamide center embedded in a tridentate NPN ligand was inserted into a ruthenium(II)-hydride bond in Ru(H)(Cl)(PPh3)3, a stable metallophosphorane complex, (NPHN)RuCl(CO)(PPh3) (28; Fig. 13), was generated [36]. The 31P peak for a central phosphorus atom of 28 appears at –12.3 ppm split into a doublet due to a coupling with the adjacent hydrogen atom displaying a large coupling constant of 1 JP-H ¼ 535 Hz. An indicative vibration for a P-H bond at 2,226 cm–1 in its IR

84

S. Kim et al.

Fig. 13 Insertion of ruthenium hydride to a tri-coordinate phosphorous ligand and the ligand-based hydride abstraction/addition reactivity

Fig. 14 Phenyl group abstraction from tetraphenylborate anion occurring at the phosphenium center

spectrum supports that 28 possesses a P-H bond. The phosphorus moiety of 28 adopts a distorted trigonal bipyramidal geometry (τ ¼ 0.50) having both H and Ru in its equatorial position and two nitrogen atoms occupying axial positions (∠N1-PN2 ¼ 170.74(7) ). Interestingly, when 28 was treated with a hydride acceptor such as triphenylcarbenium hexafluorophosphate (Ph3CPF6), the P-bound hydride was eliminated to generate (NPN)Ru(II) species 29 under ambient conditions. As a reverse reaction, 29 was reacted with NaBH4 resulting in the formation of 28. Thus, the nontrigonal phosphorus center displays both hydride donor and acceptor activity because of its biphilic character.

3 C-group Transfer Reactions of a M-P Moiety A DPNHP ligand shows redox-active behaviors involving electron/proton transfer or oxidation as described earlier, and it also displays C-group transfer. The phosphenium center of a cationic DPNHP ligand with a non-coordinating anion, BPh4–, shows an electrophilic reactivity toward a phenyl group of BPh4–, when it is exposed to CuCl, as shown in Fig. 14 [37]. As a result, a copper(I) complex (30)

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

85

Fig. 15 Migrations of mesityl and C5H5– moieties to the central P atom of a DPNHP ligand

having a coordination of a neutral phosphine ligand is synthesized, which involves the formation of a P-C bond (1.824(2) Å) with a phenyl group migrated from BPh4– at room temperature. The electrophilic reactivity of the central phosphorus atom of an NHP moiety is also presented in a neutral (DPNHP)Cl ligand. A treatment of (DPNHP)Cl with mesitylcopper(I) affords an aryl phosphine formation in {(DPNHP)Mes}CuCl (31; Fig. 15) [37]. The central 31P peak of 31 appears at 67.9 ppm, which is similar to that of 71.2 ppm for 30, but relatively upfield shifted from the signals of 147.9 ppm for (DPNHP)Cl and ~90 ppm for a phosphenium ligand, indicating that 31 contains a neutral aryl phosphine coordination to a cuprous ion. A reaction of a (DPNHP)Cl ligand with nickelocene results in P-C bond coupling, which occurs between the central phosphorus atom and a cyclopentadienyl (Cp–) group, as shown in Fig. 15 [38]. In the presence of PPh3, a Cp– ligand of nickelocene migrates to an NHP moiety to form {(DPNHP)Cp}Ni(II)Cl2 (32) displaying a peak at 98.1 ppm in 31P NMR spectrum similar with 30 and 31. One of the chloride anions is presumed to come from solvent, CH2Cl2. Interestingly, the loss of the symmetry at the η5-Cp ring in 32 was detected according to the 1H NMR spectrum showing 6.26, 5.63, 5.56, and 2.59 ppm in 1:1:1:2 integral ratio, which are remarkably different with the η5-Cp ring showing a singlet. Thus, the Cp ring in 32 exists only as the vinylic isomer, which can be generated from the sigmatropic rearrangement, not other isomers such as an ion pair resonance of a Cp– and a P+ or an allylic isomer. The coordination of the metal center, which destabilizes the ion pair resonance form, is inferred to result in this selectivity.

86

S. Kim et al.

4 O-(S-)group Transfer Reactions of a M-P Moiety Various group transfer reactions occurring at a phosphide-metal moiety involving a new type of metal-ligand cooperativity have been recently investigated by the Lee group [39–41]. In a (PPP)Ni scaffold (PPP– ¼ –P{2-PiPr2-C6H4}2) shown in Fig. 16, the central phosphide moiety of a phosphine-based tridentate PPP ligand cooperatively assists small molecule activation occurring at a nickel center. The initial cooperativity was recognized from the reaction of (PPP)NiII-L with a π-acidic ligand leading to the formation of a nickel(0) complex. When CO or tBuNC coordinates to a square planar Ni(II) center of (PPP)Ni-R (33, R ¼ OMe or OPh), alkoxide group transfer occurs, which involves a cleavage of a Ni-O bond followed by a P-O bond formation to generate a phosphinite moiety [39]. As a result, a fourcoordinate nickel(0) species (34–37; Fig. 15) was cleanly synthesized. During this transformation, nickel’s geometry converts from square planar (τ4 ¼ 0.16–0.32) to tetrahedral (τ4 ¼ 0.75–0.81) according to their X-ray crystallographic data. Interestingly, noticeable changes in the 31P NMR chemical shifts and coupling constants were also accompanied depending on the geometry of nickel complexes. A 31P peak for the central phosphorus atoms of a phosphide Ni(II) species appears at 97–101 ppm with a relatively small coupling constant (JP-P ¼ 5–10 Hz), while a neutral phosphorus atom of a nickel(0) phosphinite species displays a peak at 147–155 Hz with JP–P ¼ 60–72 Hz. While the reverse reaction was difficult to be achieved with alkoxides, the reversible transformation of a Ni(II) thiolate species to (PPSArP)Ni(CO) (36) was successfully demonstrated with a thiolato ligand [40]. A nickel(II) thiolato species, (PPP)Ni(SAr) (33-SAr, Ar ¼ phenyl and C6H4-para-

Fig. 16 Phosphido-nickel(II)/phosphinite-nickel(0) conversion induced by a π-acidic ligand

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

87

NMe2), shows a thiolato group transfer involving P-S bond formation between the central phosphorus atom and a thiolate group, which is also induced by the coordination of a π-acidic CO ligand. According to the preliminary kinetic experimental data, this reaction reveals a pseudo-first-order decay with respect to the consumption of 33-SAr under the excess CO(g) conditions, indicating an intramolecular reaction pathway. When the reaction mixture was exposed to vacuum, a CO ligand was decoordinated from a Ni(0) center allowing to regenerate a Ni(II)-SAr species. This result suggests the conversion of a Ni(II) species to a Ni(0) species is fully reversible. Finally, cooperative reactivity of (PPSArP)Ni(CO) (36) occurring between P and Ni was recognized, when 36 was treated with triphenylmethyl chloride resulting in the formation of (PPP)Ni-Cl and Ph3CSPh as shown in Fig. 16. The corresponding reaction involves the MLC of a (PPP)Ni scaffold, because the corresponding thiolato group transfer occurs at a P site via the cleavage of a P-S bond to give a newly formed C-S bond and concomitantly two-electron oxidation occurs at a nickel (0) center [40]. We also found that the metal-ligand cooperative group transfer can compete with migratory insertion reaction [41]. When a nickel(II)-isopropoxide species reacts with a π-acidic ligand, two different products were detected, presumably due to the polarization of CX bond (X ¼ O or N). When tert-butyl isocyanide coordinates to 33-iPrO, a iPrO group transfers to the phosphorus atom of a PPP ligand to generate a phosphinite nickel(0) species (37). Interestingly, with CO(g), migratory insertion of the iPrO group occurs to generate an alkoxy carbonyl nickel (II) species, (PPP)Ni(C(O)OiPr) (38; Fig. 16) [41]. The cooperativity of a PPP ligand accompanying a two-electron redox change of a nickel ion was employed in the activation of CO2 (Fig. 17) [39]. When a nickel(0)N2 species, (PPOMeP)Ni(N2) (39), was exposed to CO2(g), the methoxy group of a

Fig. 17 Methoxy group transfer reaction of a (PPP)Ni scaffold involving CO2 conversion mediated by MLC

88

S. Kim et al.

Fig. 18 X-ray structure of (PPMeP)Ni((η2-CO2)

Fig. 19 Effective electron exchange between Co and P coupled with group transfer occurring in a (PPP)Co scaffold

phosphinite moiety was transferred to the carbon atom of CO2 to give a Ni(II)methylcarbonate species (PPP)Ni(OC(O)OMe) (41). This reaction leads to the formation of a nickel(0)-CO2 adduct, (PPOMeP)Ni(η2-CO2) (40), as an intermediate species, which was verified by the comparison of 31P NMR data with a structurally characterized nickel(0)-CO2 adduct (PPMeP)Ni(η2–CO2) [42] as shown in Fig. 18. Thus, the transformation of 39–41 occurs through a stepwise pathway involving initial coordination of CO2 to Ni(0). The corresponding methoxy group transfer is nicely coupled with two-electron oxidation of nickel to give a square planar nickel (II) carbonato complex. Finally, treating 41 with methyl iodide generates a nickel (II) iodo complex, which involves the methylation of a central P atom of a PPP ligand and the generation of dimethyl carbonate (DMC). Recently, another cooperativity of a metal-phosphide moiety has been established by using a (PPP)Co scaffold revealing an open-shell type reactivities. Under basic conditions with triethylamine, the reaction of a (PPP)H ligand with CoBr2 results in the formation of a P-P bond coupled dicobalt(I) complex, (P2P-PP2)(CoBr)2 (42; Fig. 19) [43]. This reaction occurs via initial formation of a five-coordinate cobalt

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

89

(II) dibromide species having a P-H bond. Upon treating with NEt3, (PPHP)Co(Br)2 was converted to a P-P bond-coupled dimer 42, which probably undergoes dehalogenation coupled with the reduction of cobalt(II) (Fig. 19). Consequently, a phosphinyl radical appears as a form of a predicted intermediate species, (PPP) CoBr. The DFT analysis on the intermediate species displays reasonable effective spin densities found at the central phosphorus atoms in both a ferromagnetic- and an antiferromagnetic-coupled state, thus supporting a phosphinyl radical coordination at a cobalt(I) center. This result suggests that both P- and Co-based frontier orbitals are similar in energy, and thus inner-sphere electron transfer can effectively occur between P and Co depending on the coordination environment of a (PPP)Co scaffold. When a phenoxide group is introduced to 42, a phenoxide dicobalt (I) species (43) forms leading to interestingly transformation [43]. By treating with tert-butyl isocyanides to the solution of 43, two monomeric cobalt(I) species are generated in a 1:1 ratio; a phosphinite cobalt(I) phenoxide (PPOPhP)Co(OPh) (44) and a diamagnetic bis-isocyanide species, (PPP)Co(CNtBu)2 (45). Upon coordination of a π-acidic ligand, one phenolate migrates from a cobalt(I) center to a P-P moiety to form a P-O bond, while a phosphide moiety is regenerated to give 45. Compared to the cooperativity of a (PPP)Ni scaffold typically undergoing a two-electron process by using a Ni(II/0) couple, a (PPP)Co scaffold prefers a single-electron process occurring at a central phosphide moiety of a PPP ligand, while a cobalt ion remains in its +1 oxidation state.

5 N-group Transfer Reactions of a M-P Moiety To a (PPP)Ni scaffold, introducing a redox-noninnocent ligand, such as anilide, allows to broaden the scope of metal-ligand cooperativity between P and Ni [44]. Similar to other alkoxide group transfer, the cooperative transformation of a ditolylamido nickel(II) species (46) to (PPNTol2P)Ni(CO) (47) displays when 46 is exposed to CO(g). Thus, not only P-O and P-S bonds, a new P-N bond is successfully generated from the reaction of 46 with CO via the metal-ligand cooperative transformation operating at a P-Ni moiety, as shown in Fig. 20. Interestingly, when CO(g) is charged to the solution of a phosphide nickel(II) anilido complex (PPP)Ni (NHMes) (46, Mes ¼ 2,4,6-trimethylphenyl), a P-P-coupled dinuclear nickel(0) species, (P2P-PP2){Ni(CO)}2 (48), is generated along with the formation of isocyanate and amine (Fig. 20) [44]. This reaction involves multiple single-electron transfer processes; two electrons are provided from both anilide and phosphide ligands to reduce a nickel(II) ion to give 48. When light-induced one-electron transfer occurs from anilide, an open-shell phosphide-Ni(I) specie should be generated. Corresponding inner-sphere electron transfer leads to form a N radical, which is responsible for H-atom abstraction. According to the DFT evaluations, a significant spin density is found at a nickel ion in a (PPP)Ni moiety. Interestingly, when a CO ligand coordinates to a corresponding nickel(I) center, umpolung-type inversion of spin density between Ni and P occurs to give a nickel(0) mono-carbonyl species

90

S. Kim et al.

Fig. 20 Amido group transfer reactions and P-P bond formation/cleavage of a (PPP)Ni(II) anilido species

having a substantial spin density at the central P atom. Thus, the carbonylation of a phosphide-Ni(I) specie produces a dinuclear nickel(0)-CO species 48 proceeded by radical coupling of a P•-Ni0-CO species. The UV-Vis spectrum of a resulting dinickel(0) complex 48 displays a unique absorption at 536 nm, which is related to the transition from a σ-bonding to an antibonding orbital of a P-P bond, according to time-dependent (TD) DFT calculations. Upon white LED light irradiation, a P-P bond is cleaved to form (PPP•)Ni(0)(CO) (49). Frozen-solution X-band EPR data shows a signal at g ~2.015 with hyperfine couplings (A ~ 110 and 50 G) with two different phosphorus atoms measured at 20 K, supporting the formation of a phosphorus-based radical. The P radical generated by visible-light irradiation can be also captured by the chemical treatment using a 2,4,6-tri-tert-butylphenoxyl radical to give (PPOArP)Ni(CO) (50, Ar ¼ 2,4,6-tri-tert-butylphenyl). Various σ-bonds, such as N-H, N-N, and O-H, were activated by the light-induced reaction of a P-P bond containing nickel(0) dimer complex, 48 to generate mononuclear Ni (0) species (51–53). These results suggest that a P moiety of a PPP ligand has a dual role to serve as a redox-active site and a reaction center.

6 Conclusion Cooperative reactivities of phosphorus-containing transition metal pincer systems possessing an anionic phosphido-, phosphinito-, or NHP-based phosphido moiety are highlighted in this chapter. As a reliable electron/proton and group transfer site, the phosphorus atom actively participates in a range of chemical reactions. This is possible because a phosphorus atom can adopt several distinct forms such as cationic phosphenium, anionic phosphide, neutral phosphine, or a phosphinyl radical, thus making it more versatile. Furthermore, the ability of the phosphorus atom to have diverse oxidation states promotes the phosphorus atom to be cooperatively engaged

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

91

in various organometallic reactions occurring at a metal center. The redox-coupled geometric alteration of the phosphorus atom located at the center of pincer systems enables a new way to operate the metal-ligand cooperativity. A phosphorus-based orbital can be found in the frontier molecular orbitals of first row transition metal complexes, thus suggesting that electron exchange can occur within a P-M moiety during a chemical reaction. By utilizing such cooperation between a phosphorus atom and a metal center in pincer ligand systems, the role of transition metal complexes in valuable catalytic reactions can be further extended. Acknowledgments This work was supported by C1 Gas Refinery Program (NRF2015M3D3 A1A01064880).

References 1. Tolman CA (1970) Phosphorus ligand exchange equilibria on zerovalent nickel. A dominant role for steric effects. J Am Chem Soc 92:2956–2965 2. Quin LD, Verkade JG (eds) (1994) Phosphorus-31 NMR spectral properties in compound characterization and structural analysis. VCH Publishers, New York 3. Rittle J, Green MT (2010) Cytochrome P450 compound I: capture, characterization, and C–H bond activation kinetics. Science 330:933–937 4. Yamada H, Siems WF, Koike T, Hurst JK (2004) Mechanisms of water oxidation catalyzed by the cis,cis-[(bpy)2Ru(OH2)]2O4+ ion. J Am Chem Soc 126:9786–9795 5. Hurst JK, Cape JL, Clark AE, Das S, Qin C (2008) Mechanisms of water oxidation catalyzed by ruthenium diimine complexes. Inorg Chem 47:1753–1764 6. Cape JL, Lymar SV, Lightbody T, Hurst JK (2009) Characterization of intermediary redox states of the water oxidation catalyst, [Ru(bpy)2(OH2)]2O4+. Inorg Chem 48:4400–4410 7. Kobayashi K, Ohtsu H, Wada T, Kato T, Tanaka K (2003) Characterization of a stable ruthenium complex with an oxyl radical. J Am Chem Soc 125:6729–6739 8. Lippert CA, Arnstein SA, Sherrill D, Soper JD (2010) Redox-active ligands facilitate bimetallic O2 homolysis at five-coordinate oxorhenium(V) centers. J Am Chem Soc 132:3879–3892 9. Dzik WI, van der Vlugt JI, Reek JNH, de Bruin B (2011) Ligands that store and release electrons during catalysis. Angew Chem Int Ed 50:3356–3358 10. Lippert CA, Hardcastle KI, Soper JD (2011) Harnessing redox-active ligands for low-barrier radical addition at oxorhenium complexes. Inorg Chem 50:9864–9878 11. Chaudhuri P, Hess M, Müller J, Hildenbrand K, Bill E, Weyhermüller T, Wieghardt K (1999) Aerobic oxidation of primary alcohols (including methanol) by copper(II)– and zinc(II)– phenoxyl radical catalysts. J Am Chem Soc 121:9599–9610 12. Khusnutdinova JR, Milstein D (2015) Metal–ligand cooperation. Angew Chem Int Ed 54:12236–12273 13. Gudat D (1997) Cationic low coordinated phosphorus compounds as ligand: recent development. Coord Chem Rev 163:71–106 14. Caputo CA, Jennings MC, Tuononen HM, Jones ND (2009) Phospha-Fischer carbenes: synthesis, structure, bonding, and reactions of Pd(0)– and Pt(0)–phosphenium complexes. Organometallics 28:990–1000 15. Adhikari D, Mossin S, Basuli F, Dible FR, Chipara M, Fan H, Hoffman JC, Meyer K, Mindiola DJ (2008) A dinuclear Ni(I) system having a diradical Ni2N2 diamond core resting state: synthetic, structural, spectroscopic elucidation, and reductive bond splitting reactions. Inorg Chem 47:10479–10490

92

S. Kim et al.

16. Goodman J, Macgregor SA (2010) Metallophosphoranes: the hidden face of transition metal– phosphine complexes. Coord Chem Rev 254:1295–1306 17. Wachter J, Mentzen BF, Riess JG (1981) Synthesis and crystal structure of a (η5-C5H5)(CO)2molybdenum bicyclophosphoraide: the first transition metal complex with a “R4P–”-ligand. Angew Chem Int Ed 20:284 18. Mankad NP, Rivard E, Harkins SB, Peters JC (2005) Structural snapshots of a flexible Cu2P2 core that accommodates the oxidation states CuICuI, Cu1.5Cu1.5, and CuIICuII. J Am Chem Soc 127:16032–16033 19. Rhee YM, Head-Gordon M (2008) A delicate electronic balance between metal and ligand in [Cu–P–Cu–P] diamondoids: oxidation state dependent plasticity and the formation of a singlet diradicaloid. J Am Chem Soc 130:3878–3887 20. Mankad NP, Harkins SB, Antholine WE, Peters JC (2009) Multifrequency EPR studies of [Cu1.5Cu1.5]+ for Cu2(μ-NR2)2 and Cu2(μ-PR2)2 diamond cores. Inorg Chem 48:7026–7032 21. Derrah EJ, Ladeira S, Bouhadir G, Miqueu K, Bourissou D (2011) Original phenyl–P(O) bond cleavage at palladium(0): a combined experimental and computational study. Chem Commun 47:8611–8613 22. Derrah EJ, Martin C, Mallet-Ladeira S, Miqueu K, Bouhadir G, Bourissou D (2013) Chelating assistance of P–C and P–H bond activation at palladium and nickel: straightforward access to diverse pincer complexes from a diphosphine–phosphine oxide. Organometallics 32:1121–1128 23. Cheng M-J, Fu R, Goddard III WA (2014) Design and validation of non-metal oxo complexes for C–H activation. Chem Commun 50:1748–1750 24. Fu R, Goddard III WA, Chem M-J, Nielsen RJ (2017) Computational design of a pincer phosphinito vanadium ((OPO)V) propane monoxygenation homogeneous catalyst based on the reduction-coupled oxo activation (ROA) mechanism. ACS Catal 7:356–364 25. Day GS, Pan B, Kellenberger DL, Foxman BM, Thomas CM (2011) Guilty as charged: non-innocent behavior by a pincer ligand featuring a central cationic phosphenium donor. Chem Commun 47:3634–3636 26. Pan B, Xu Z, Bezpalko MW, Foxman BM, Thomas CM (2012) N-heterocyclic phosphenium ligands as sterically and electronically-tunable isolobal analogues of nitrosyls. Inorg Chem 51:4170–4179 27. Pan B, Bezpalko MW, Foxman BM, Thomas CM (2012) Heterolytic addition of E–H bonds across Pt–P bonds in Pt N-heterocyclic phosphenium/phosphido complexes. Dalton Trans 41:9083–9090 28. Evers-McGregor DA, Bezpalko MW, Foxman BM, Thomas CM (2016) N-heterocyclic phosphenium and phosphido nickel complexes supported by a pincer ligand framework. Dalton Trans 45:1918–1929 29. Pan B, Bezpalko MW, Foxman BM, Thomas CM (2011) Coordination of an N-heterocyclic phosphenium containing pincer ligand to a Co(CO)2 fragment allows oxidation to form an unusual N-heterocyclic phosphinito species. Organometallics 30:5560–5563 30. Poitras AM, Knight SE, Bezpalko MW, Foxman BM, Thomas CM (2018) Addition of H2 across a cobalt–phosphorus bond. Angew Chem Int Ed 57:1497–1500 31. Hatizis GP, Oliemuller LK, Dickie DA, Thomas CM (2020) N-heterocyclic phosphide complexes of rhodium supported by a rigid pincer ligand. Eur J Inorg Chem 2020:2873–2881 32. Culley SA, Arduengo AJ (1984) Synthesis and structure of the first 10-P-3 species. J Am Chem Soc 106:1164 33. Dunn NL, Ha M, Radosevich AT (2012) Main group redox catalyst: reversible PIII/PV redox cycling at a phosphorus platform. J Am Chem Soc 134:11330–11333 34. McCarthy SM, Lin Y-C, Devarajan D, Chang JW, Yennawar HP, Rioux RM, Ess DH, Radosevich AT (2014) Intermolecular N–H oxidative addition of ammonia, alkylamines, and arylamines to a planar σ 3-phosphorus compound via an entropy-controlled electrophilic mechanism. J Am Chem Soc 136:4640–4650 35. Lee K, Blake AV, Tanushi A, McCarthy SM, Kim D, Loria SM, Donahye CM, Spielvogel KD, Keith JM, Daly SR, Radosevich AT (2019) Validating the biphilic hypothesis of nontrigonal phosphorus(III) compounds. Angew Chem Int Ed 58:6993–6998

Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems

93

36. Tanushi A, Radosevich AT (2018) Insertion of a nontrigonal phosphorus ligand into a transition metal–hydride: direct access to a metallohydrophosphorane. J Am Chem Soc 140:8114–8118 37. Knight SE, Bezpalko MW, Foxman BM, Thomas CM (2014) Coordination of N-heterocyclic phosphine- and phosphenium-containing pincer ligands to copper(I): evidence for reactive electrophilic metal–phosphenium intermediates. Inorg Chim Acta 422:181–187 38. Evers-McGregor DA, Bezpalko MW, Foxman BM, Thomas CM (2017) Exploring the coordination chemistry of N-heterocyclic phosphenium/phosphido ligands using nickelocene as a synthon. Inorg Chim Acta 460:17–21 39. Kim Y-E, Kim O, Oh S, Kim S, Kim J, Han SW, Lee Y (2015) Phosphinite-Ni(0) mediated formation of a phosphide-Ni(II)-OCOOMe species via uncommon metal–ligand cooperation. J Am Chem Soc 137:4280–4283 40. Oh S, Lee Y (2016) Reversible intramolecular P–S bond formation coupled with a Ni(0)/Ni (II) redox process. Organometallics 35:1586–1592 41. Oh S, Kim S, Lee D, Gwak J, Lee Y (2016) Alkoxide migration at a nickel(II) center induced by a π-acidic ligand: migratory insertion versus metal–ligand cooperation. Inorg Chem 55:12863–12871 42. Kim Y-E, Kim J, Lee Y (2014) Formation of a nickel carbon dioxide adduct and its transformation mediated by a Lewis acid. Chem Commun 50:11458–11462 43. Kim S, Lee Y (2020) Metal-ligand cooperativity of a Co-P moiety. Inorg Chem Front 7:1172–1181 44. Kim Y-E, Lee Y (2018) A P–P bond as a redox reservoir and an active reaction site. Angew Chem Int Ed 57:14159–14163

Top Organomet Chem (2021) 68: 95–134 https://doi.org/10.1007/3418_2020_65 # Springer Nature Switzerland AG 2020 Published online: 25 December 2020

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary Coordination Sphere Ajeet Singh, Evamarie Hey-Hawkins, and Dmitri Gelman

Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Genesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Carbometalated Pincer Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 C(sp2)-Based Pincer Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 C(sp3)-Based Pincer Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 N-Based Pincer Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 N(sp2)-Based Pincer Complexes Possessing Lewis Acid Functionalities . . . . . . . . . . . 3.2 N(sp2)-Based Pincer Complexes Possessing Hydrogen Bonding Functionalities . . . 3.3 Binuclear Reactivity of N(sp2)-Based Pincer Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

96 99 100 100 105 113 113 120 123 126 127

Abstract Pincer complexes represent a family of potent compounds having a tremendous number of manifold applications in organometallic chemistry, synthesis, catalysis, materials science, and bioinorganic chemistry. This chapter overviews the recent developments in the chemistry and catalytic applications of pincer complexes incorporating secondary coordination sphere or an appended functionality that can interact with the catalytic center or can modulate its reactivity via secondary substrate-catalyst interactions. Combining the concepts of modular and stable pincer A. Singh Institute of Chemistry, The Hebrew University, Jerusalem, Israel E. Hey-Hawkins (*) Institute of Inorganic Chemistry, Faculty of Chemistry and Mineralogy, Leipzig University, Leipzig, Germany e-mail: [email protected] D. Gelman (*) Institute of Chemistry, The Hebrew University, Jerusalem, Israel Peoples’ Friendship University of Russia (RUDN University), Moscow, Russia e-mail: [email protected]

96

A. Singh et al.

ligands and secondary interactions provides excellent opportunities for fine-tuning the properties of a coordinated metal center and, consequently, attracts considerable interest. Keywords Ligand-metal cooperation · Pendant functional group · Pincer complexes · Secondary interactions

Abbreviations BArF4 B(sia)2 BBN BnOH BPin COD Cp* DME FA KC8 kPa MOM NHC RNA RT TOF TON TS Ts

Tetrakis(pentafluorophenyl)borate Diisoamylborane 9-Borabicyclo[3.3.1]nonyl Benzyl alcohol 4,4,5,5-Tetramethyl-1,2,2-dioxoboryl 1,5-Cyclooctadiene Pentamethylcyclopentadienyl Dimethoxyethane Formic acid Potassium graphite Kilopascal Methoxymethyl N-heterocyclic carbenes Ribonucleic acid Room temperature Turnover frequency Turnover number Transition state Toluenesulfonyl

1 Introduction Tridentate, meridionally coordinating DXD (Fig. 1) pincer ligands have become a useful and versatile tool for modern coordination and organometallic chemistry [1– 8]. Moreover, their corresponding pincer metal complexes have been successfully employed for a broad spectrum of applications including catalysis, materials, supramolecular chemistry, sensing, and medicine [9–14]. A distinctive and advantageous feature of such ligands is their unique DXD η3-mer binding mode, which enhances the integrity of generally labile carbon-metal or heteroatom-metal bonds through the Fig. 1 Schematic representation of the pincer complexes

D X M D

M = transition or main group metal D = neutral donor such as N, P, As, O, S or NHC X = neutral or anionic donor, such as C, N, P, S or NHC

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

97

Fig. 2 Aromatic versus aliphatic pincer complexes

formation of the kinetically and thermodynamically stable C-M bond-sharing bicyclic motif (Fig. 1) [15, 16]. A great deal of interest and the subsequent development of pincer ligands were in (hetero)aromatic complexes. The most common type of pincer platforms comprised either a central, formally anionic aryl moiety or a neutral heteroaromatic unit – both donate two electrons to the metal. Direct communication between the metal-centered dxz orbital and the π orbitals of the (hetero)aromatic backbone, along with the possibility to fine-tune the electronic properties of the metal site by modulating the electronic nature of the ring, makes them attractive candidates for practical applications [17–21]. Non-aromatic pincer complexes are far less common and have received only limited attention compared with their arene-based cousins, mainly due to the lower availability of the precursors and ligands. On the other hand, the conformational flexibility of the all-aliphatic frames, along with the presence of labile α- and β-hydrogen, reduces their stability, compared with the aromatic counterparts [22– 25], although, as typical for all pincer complexes, the η3-mer coordination mode is still translated into carbon-metal and heteroatom-metal bonds that are stable to such an extent that they can be handled under a non-inert atmosphere, survive exposure to reactive species, and even be post-synthetically modified (Fig. 2). It has been generally postulated that pincer complexes represent an appealing example of compounds possessing the “just right” balance between stability and reactivity that suits a broad spectrum of catalytic applications. However, many of them benefit primarily from high thermal, redox, and chemical stability, rather than their reactivity [26–28]. Thus, paradoxically, the lack of coordination flexibility in the carbometalated pincer ligands appears to be their most significant limitation in catalysis because certain reactive intermediates of different geometries fail to form, at least when it comes to more sophisticated catalytic cycles. A coordination switch via metal-amide/metal-amine interconversion [29–31] in the aliphatic disilylamido PN(sp3)P pincer complexes (1–2), discovered by Frysuk in 1983 [32], and the heteroaromatic PNN pincer catalysts (3–4) by Milstein in 2005 [33] led to an increased number of publications describing new metal-ligand cooperating reactivity patterns in heteroaromatic [34–38] and aliphatic pincer systems (Scheme 1) [39–42]. These studies also fueled extensive research in the field of carbometalated pincer complexes, although similar coordination switches in the carbon-based DCD-type pincer ligands are less obvious. For example, during their pioneering studies, Shaw and co-workers postulated the susceptibility of transition metal pincer complexes, bearing all-aliphatic ligands toward α- and β-hydride elimination, to form isomeric carbene or olefin chelate compounds [23, 43–45]. This reversible interplay between

98

A. Singh et al.

Me2Si Me2Si

PPh2 N Ir

H H

+ H2 - H2

PPh2 1

PPh2 H N Ir H H Me2Si H PPh2 2

NEt2 H N Ru CO

Me2Si

PtBu2

H NEt2

+ H2 - H2 H

3

D

N Ru CO H PtBu2 4

D TM Z

X Y

-X-Z +X-Z

D

TM

Y D

Scheme 1 Metal-ligand cooperativity patterns in heteroatom-based pincer complexes

PtBu2 H Rh Cl H

PtBu2 7

Hα -Hβ

Hβ

PtBu2 H M Cl

PtBu2 -H-Hα

PtBu2 5: M = Rh 6: M = Ir

Ir

Cl

PtBu2 8

Scheme 2 Coordination flexibility in all-aliphatic pincer complexes

Fig. 3 Design of the pincer complexes possessing appended functionality

C(sp3)-carbometalated and α- or β-eliminated species has shown that the pincer complexes have the ability to accommodate more exotic coordination states, thus holding promise to open new practical reactivity patterns (Scheme 2). There are several excellent recent reviews dealing with these aspects of pincer reactivity and their applications [25, 46–48]. Designing bifunctional catalysts possessing a secondary coordination sphere or an appended functionality that can interact with the catalytic center or modulate its reactivity via secondary substrate-catalyst interactions has emerged as another way to diversify reactivity patterns of the pincer catalysts (Fig. 3) [49–52]. This relatively new development in the chemistry of pincer complexes stems from the idea that the primary coordination sphere is not the only factor that contributes to the properties of transition metal catalysts, but metals can also interact with other molecules forming outer-sphere intermediates. Thus, in this review, we will limit ourselves to some aspects of design, reactivity, and catalysis using multifunctional pincer complexes where coordination versatility and reactivity are governed by an appended functional group in the secondary coordination environment of a pincer-derived ligand.

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

1.1

99

Genesis

The term “secondary interactions” in the context of transition metal catalysis was coined by Ito and Sawamura to describe enzyme-like catalyst-substrate interactions that occur outside the primary coordination sphere of the metal site [53]. At the outset, secondary interactions have been mentioned in the literature only sporadically and mainly referred to asymmetric synthesis [54–60]; however, more systematic and comprehensive studies, aiming at “teaching” organometallic pincer catalysts “new tricks,” appeared after Crabtree and Klein Gebbink independently provided the first landmark examples of the interplay between the primary and secondary coordination spheres leading to a non-classical pincer reactivity. The first important example reported by Crabtree and Brudvig comprised a di-manganese terpyridine-based pincer catalyst 9 equipped with a peripheral molecular recognition site derived from Kemp’s triacid. The attractive interactions at the secondary hydrogen bonding site directed functionalized substrates toward the catalytic Mn(μ-O)2Mn core in such a way that they modified the usual selectivity for oxidation [61]. Thus, oxidation of ibuprofen using 9 as a catalyst and peroxomonosulfate as an oxidant provided >98% regioselectivity at the distant benzylic position with up to 700 turnovers (Fig. 4). It has been mentioned that prototypical pincer ligands are well-suited for structural modification; however, divergent installation of a secondary coordination sphere or an appended functionality within reasonable vicinity from the primary site is not straightforward. For example, owing to the planar shape of the (hetero)aromatic backbones, pincer sidearms are the most suitable sites in order to avoid the employment of cumbersome linkers (Fig. 5). In 2006 Klein Gebbink and co-workers synthesized a series of chiral non-racemic palladium complexes 10–11 possessing aromatic mono-anionic NCN-pincer ligands

Fig. 4 Di-manganese terpyridine-based pincer catalyst with a peripheral molecular recognition site

X

D M D

FG X

D M D

Fig. 5 Modification of the (hetero)aromatic pincer complexes

FG

100

A. Singh et al.

N

BF4R

Pd N

N Pd

R

10:R = CO2Me 11: R = CO2Bz

N 12

Ph Ph OH OH Ph Ph

BF4O Ph

H

+

NC CO2Me

10-12

Ph O

CO2Me N

10: cis/trans = 35/65; ee% (cis) =12 11: cis/trans = 38/62; ee% (cis) =16 12: cis/trans = 63/37; ee% (cis) =40

Scheme 3 Hydrogen bond-directed aldol condensation between aromatic aldehydes and methyl α-isocyanoacetate

containing L-proline-derived donors [62]. According to the initial hypothesis, the stereogenic center at the proline ring, together with the bulkiness of the corresponding substituent, can efficiently control stereogenicity at the flipping nitrogen atoms upon their coordination to the palladium center. In the course of these studies, their catalytic potential was tested in the aldol condensation between aromatic aldehydes and methyl α-isocyanoacetate. However, the regio- and enantioselectivity obtained with the cationic derivatives 10–11 were poor, apparently due to the flexibility of the chiral pocket created by the proline derivatives around the catalytic center [63]. To increase the steric bulk close to the metal, pyrrolidinyl N-donor moieties with diphenylhydroxymethyl functional groups were installed in the NCN-pincer ligand framework [64]. Crystallographic characterization of the new complex 12 showed that the two diphenylmethyl groups point away from the metal center, hence making the O-donor atoms predisposed for coordination with the metal and hydroxyl hydrogens for intramolecular hydrogen bonding. The catalytic tests using the same benchmark reaction revealed increased regioselectivity, as well as stereoselectivity for the cis-oxazoline products in the presence of 12, in contrast with the regioselectivity observed for 10–11. The assistance of the hydrogen bond-donating hydroxyl groups was suggested to play an essential role in the stereoselective outcome of the reaction through stabilization of the hypothetic palladium-bound enolate intermediate, which could induce a preferential arrangement for attack of the aldehyde (Scheme 3). Although hydrogen bond-assisted enhancement of enantioselectivity had already been known at that time [65, 66], these works became a landmark in the design of carbometalated pincer complexes equipped with pendant functional groups at the secondary coordination sphere.

2 Carbometalated Pincer Complexes 2.1

C(sp2)-Based Pincer Complexes

In 2017, Huang disclosed an exciting example on NCP-type Ir pincer complex 13 (Scheme 4) as a very active but poorly regio- and stereoselective catalyst for alkene

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

101

Scheme 4 The NCP-type Ir pincer catalysts for alkene isomerization

Scheme 5 Synthesis of the complex 20 possessing agostic functionality

isomerizations under mild conditions [67]. It was hypothesized that increasing the steric bulk at the pyridine site of the pincer catalyst might significantly improve the selectivity of the reaction. The synthetic strategy leading to these ligands is relatively modular because it relies on robust Suzuki coupling, followed by a series of trivial functional group modifications. For example, the cross-coupling between 2-bromo-6-(tert-butyl)pyridine (15) and 3-(hydroxymethyl)phenylboronic acids (16), followed by a two-step modification of the hydroxymethyl sidearm in the corresponding 2-(tert-butyl)-6{3-[(di-tert-butylphosphino)methyl]phenyl}pyridine (17), leads to the desired 18 in good overall yield. The oxidative carbometalation of the latter with [IrCl(COD)]2 leads to the formation of PCN iridium hydride 19 [68]. Its successive reaction with NaOtBu in toluene leads to the creation of a new complex 20, where the iridium center lacks the coordination of other ligands except for the agostic interaction with the appended tert-butyl group. Although X-ray crystallographic analysis of this agostic complex was not provided, the spectroscopic data are consistent with the suggested structure (Scheme 5). The new agostic complex 20 was evaluated for the isomerization of linear terminal alkenes. It was expected that the agostic bonds would dissociate readily to generate the catalytically active 14-electron species. For example, the catalytic isomerization of 1-octene using 20 as a catalyst occurred at RT, resulting in the formation of 94% (E)-2-octene (E/Z ¼ 51:1) in 1 h. High conversions and high regio- and stereoselectivities were obtained for a broad spectrum of substrates. For example, industrially valuable linear C6-, C7-, C10-, and C16-alkenes were converted into (E)-2-alkenes with high selectivities. Importantly, this catalyst system was tolerant of various functional groups. Thus, terminal alkenes bearing free and

102

A. Singh et al.

Scheme 6 Representative isomerization of alkenes catalyzed by 20

Scheme 7 Synthesis of the pincer ligand bearing appended aza-15-crown-5 functionality

protected hydroxyl as well as carbonyl, ester, and halide groups underwent selective isomerization (Scheme 6). Analogous pincer complexes possessing substituents not capable of agostic interactions were reported nearly inactive or not selective under the described reaction conditions; therefore, it is suggested that the reaction is operated by a mechanism in which the appended sidearm on the backbone acts as a shuttle. Aiming at more tunable catalysis, Miller et al. chose another strategy to adjust the properties of the pincer catalysts by means of an additional crown ether-based hemilabile sidearm. Over the years, much information about crown ether-cation interactions has been gained; therefore, a catalyst crafted with an appended crown ether function would offer a rare possibility to tune the hemilability of the donor via allosteric interactions. Moreover, the controlled hemilability may also directly control the primary coordination sphere, thus fine-tuning a balance between stability and reactivity. The first NCOP pincer ligand bearing an aza-15-crown-5 ether macrocycle as the amine donor (22) was prepared in ca. 50% yield by amination of 3-bromomethylphenol (21) with aza-15-crown-5 with subsequent phosphination with chlorodiisopropylphosphine [69]; advantageously, aza-crown ethers of different ring sizes as well as phosphine derivatives having various electronic and steric properties can be installed using only slightly modified protocols (Scheme 7) [70]. Since it is an aromatic PCN pincer ligand, 22 can accommodate a variety of late transition metals via oxidative or electrophilic C(sp2)-H activation (Scheme 8). Depending on the reaction conditions, the aza-crown-based pincer ligand can adopt tridentate, tetradentate, or pentadentate coordination modes to satisfy the ligancy of the metal. The interconversion between the modes was found to be highly reversible owing to weak dative Ir-O bonds that were theoretically estimated (the

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

103

Scheme 8 Coordination modes of the ligand 22

Scheme 9 Lewis basecontrolled coordination modes

O Cl PiPr2 N 23

O NaBArF4

Ir H O O O O

N 25

O CH3CN vacuum

N 26

H

BArF4 PiPr2

Ir L O O O O

L= CH3CN

H

PiPr2 Ir O O

BArF4

O O

O CH3CN Keq = 34 at 25 °C

27

N O O

H

PiPr2 Ir L

BArF4

LO O

Scheme 10 Cationcontrolled coordination modes

calculated activation barrier ranged between +26.7 and +55.6 kJ mol1) [69]. For example, a pentadentate coordination mode forms by halide abstraction from hydrido chloride 23 to produce the cationic 25. Addition of a weak donor, such as acetonitrile, can displace the oxygen donor back to the tetradentate species 26 and even farther to the partial formation of tridentate 27 in the presence of a large excess of acetonitrile. Of course, in general, the reversibility depends on the donor ability of the substrate and the nature of the metal [70, 71], but in this particular case, pentadentate 25 can be regenerated upon applying vacuum (Scheme 9). More importantly, the coordination modes and the substrate-complex equilibrium can be controlled by employing cation-crown ether interactions. For example, it was demonstrated that the addition of two equivalents of CH3CN sharply shifts the equilibrium to the right unlike what was described in the previous case (Scheme 10) [69]. Moreover, the nature of the alkali metal cation and its concentration can regulate the rate of chemical reactions by allosteric interactions in pincer catalysts bearing aza-crown ethers as an appended function (Scheme 11). Monitoring the rate of the H-D exchange in 25 during the reaction with molecular D2 as a function of the alkali metal cation concentrations provided proof of the principle. This straightforward

104

A. Singh et al.

Scheme 11 Allosteric interaction-controlled reactivity

O N 25

H Ir

PiPr2 O

O

O O

O MBArF4 D2 M = Li, Na

N

D

PiPr2 Ir O

O

27-M-D

BArF4

M O O

+

H-D

no additives: kobs = 1.2x10-6 sec-1 0.3 equiv. Na+: kobs = 2.4x10-5 sec-1 0.4 equiv. Li+: kobs = 2.8x10-4 sec-1

Scheme 12 Cationregulated isomerization of alkenes

25 (1 mol%) MBArF4 no additives: TOF = 1.8 h-1 1 mol% K+: TOF = 1.8 h-1 2 mol% Na+: TOF = 2.8 h-1 4 mol% Li+: TOF = 2000 h-1

Scheme 13 Switching off the catalytic activity of 27-Li

experiment demonstrated that the rate constant of the reaction in the presence of Na+ is one order of magnitude higher than without it. Furthermore, the addition of Li+ accelerated the reaction by two orders of magnitude, stressing the fact that the choice of the alkali metal cation and the amount of salt in solution control the substrate’s access to a metal center, as well as the overall rate of the chemical transformations [69, 70, 72]. These principles have found practical applications in catalysis when the 1,000fold acceleration of the isomerization of allylbenzene to β-methylstyrene, catalyzed by 25, was achieved after the addition of lithium cations, whereas virtually no acceleration of the process took place after addition of Na+ or K+ (Scheme 12) [73]. The cation-modulated reactivity can be used as a tool, not only for speeding up but also for slowing down the reactions upon demand. For instance, the addition of the chloride anions to the active form of the catalyst (27-Li) results in the precipitation of LiCl, and consequently, the catalytic activity ceases (Scheme 13). This feature makes NCOP pincer catalysts bearing an appended aza-crown function capable of performing more advanced tasks, such as turning “on” and “off” catalytic activity to promote alternative reactions or synthesize complex materials from a mixture of different building blocks [73].

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

2.2

105

C(sp3)-Based Pincer Complexes

As was mentioned, the structural simplicity of the planar C(sp2)-based pincer skeleton limits the installation of the appended functionality to the peripheral donor group. In this context, more complex and three-dimensional aliphatic pincer complexes provide more flexibility for designing the secondary coordination environment in the vicinity of the metal center. This is especially important in cases of octahedral and trigonal bipyramidal complexes (Fig. 6). A coordinationally more flexible PC(sp3)P ligand (31), which incorporates a tethered hemilabile anisole group, was designed by Iluc and co-workers. The new PC(sp3)P scaffold was synthesized starting with the reaction of 2,20 -dibromobenzophenone (28) with anisyllithium (29) to form bis-(2-bromophenyl)-2-anisylmethanol (30), followed by successive dehydroxylation and phosphination (Scheme 14). Usually, bis [2-(diphenylphosphino)phenyl] methanes and similar ligands coordinate transition metals to form eight-membered metalacycles adopting a rigid boatlike conformation, where the endo C(sp3)-bound hydrogen approaches the metal center and, therefore, creates an ideal situation for the metalation of the methylene bridge [74, 75]. However, an attempted metalation of 31 with [IrCl(COD)]2 was not particularly facile, and the desired pincer complex 32 formed only after 6 days of heating. The solid-state molecular structure displays a distorted octahedral geometry around the iridium center. The chloride ligand occupies the trans position with respect to the metalated sp3 carbon, whereas the hydride ligand is found trans to the anisole substituent (Scheme 15). Tetradentate 32 can be easily converted to dihydride complex 33 with the same coordination environment. The reactivity studies showed that the appended methoxy group could be displaced by carbon monoxide or external phosphine; however, an attempted reductive elimination of hydrogen from 33 failed, leaving not much hope for promising catalytic activity (Scheme 16).

FG D X M D

D X M D

FG

FG D X M D

FG = appended functionality

Fig. 6 Modification of aliphatic pincer complexes

O

Br

OMe

+

PiPr2

MeO

Li

Br 28

Br

MeO

Br 29

Scheme 14 Synthesis of the PC(sp3)P ligand 31

30

PiPr2 31

106

A. Singh et al. iPr iPr P

PiPr2 [IrCl(COD)] 2

MeO

PiPr2

C

6 days, C6H6 reflux

P iPr

31

iPr P

iPr O CH3 NaAlH4 Ir

O CH3

C

Cl

P

iPr H

iPr

32

Ir iPr

H

H

33

Scheme 15 Coordination chemistry of the PC(sp3)P ligand 31

iPr

no reaction

base

iPr P C P iPr iPr 33

O CH3 Ir H

H

iPr

H3C O CO

C P iPr

iPr P

CO Ir

iPr

H

H

34

Scheme 16 Reactivity of 33

Scheme 17 One-step Diels-Alder cycloaddition approach for the synthesis of three-dimensional PC(sp3)P pincer complexes

After considering structural limitations in designing functional carbometalated pincer complexes, Gelman and co-workers initiated a research project devoted to designing a modular family of three-dimensional PC(sp3)P pincer complexes based on the dibenzobarrelene scaffold [48, 76]. The unique topology of these molecules allows, in principle, constructing multifunctional catalysts possessing primary and secondary coordination spheres in reasonable proximity, in such a way that communication between them might affect the catalytic site in a controllable fashion. We envisioned a straightforward synthesis of these ligands based on the one-step DielsAlder cycloaddition of the desired dienophile to a common, readily available, and structurally simple anthracene-based precursor that represents a highly modular and divergent approach to a variety of three-dimensional platforms equipped with custom-tailored primary and secondary coordination spheres (Scheme 17). Indeed, a set of bifunctional ligands of this type was prepared using a Diels-Alder click of 1,8-bis(phosphino)anthracenes with a variety of dienophiles. For instance, the amine-functionalized (37) [77] was created via a two-step protocol starting from the quantitative Diels-Alder cycloaddition of the known 1,8-bis(diphenylphosphino)anthracene (35) and fumaronitrile, followed by a reduction of the adduct (Scheme 18).

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

107

Scheme 18 Synthesis of the bifunctional ligand 37

Fig. 7 Representative examples of the bifunctional PC(sp3)P pincer ligands mol%) OH

Scheme 19 Metalation of the ligand 39

Using only minor modifications, other ligands possessing different functional groups, such as hydroxyl- (38 and 42) [78, 79], alkoxyl- (39–41) [79, 80], and carboxyl (43) [81], have been prepared, characterized, and studied (Fig. 7). Reacting 39 with [IrCl(COD)]2 in toluene results in oxidative carbometalation of the ligand and the formation of 44 (Scheme 19). The ligand in this complex coordinates the iridium center in a tetradentate fashion, exhibiting almost an octahedral environment with relatively long O-Ir bonds of ca. 2.387 Å [80]. The preferred tetradentate chelation of 39 can be explained by the rigidity of the ligand as well as by its unique shape making the O-donor atom predisposed for coordination to the metal.

108

A. Singh et al.

The long distance, however, indicates the highly hemilabile nature of the corresponding bond that was confirmed experimentally by a reversible interaction with different donors. For example, exposure of 44 to excess of acetonitrile leads to complete displacement of the hemilabile sidearm, consequently forming octahedral 45. However, the tetradentate coordination mode (44) may be regenerated after applying vacuum or simply refluxing in a solvent different from acetonitrile. Even stronger donors, such as CO, coordinate to Ir in 44 in a highly reversible fashion, which is in sharp contrast to the prototypical iridium triptycene-based PC(sp3)P complex 47, possessing no hemilabile sidearm, which is known to bind carbon monoxide irreversibly (Scheme 20). The presence of the hemilabile sidearm was also found to accelerate reductive elimination from the present complexes. For example, when 44 was treated with NaOtBu under atmospheric H2 in toluene-d8 at 50 C for 30 min, dihydride complex 50 was formed (which can only occur after successive reductive elimination of HCl from 44 and an oxidative addition of H2 to the apparent iridium(I) intermediate 49) (Scheme 21). The same reaction with 47 requires prolonged heating. Another study, performed by the same group, aimed to demonstrate the participation of appended functionality in the activation/formation of chemical bonds. To explore this possibility, a series of iridium and ruthenium complexes possessing different sidearms were synthesized using the previously described methodology (Fig. 8). Complexes 51 and 52 exhibit the spontaneous extrusion of molecular hydrogen, which originates from intramolecular iridium hydride-carboxylic/hydroxyl proton interactions (Scheme 22, top) [77, 78]. Although we cannot completely rule out an intermolecular pathway, isolation, and full structural assignment of the suggested intermediates, the alkoxide-iridium species (55) and the carboxylate-iridium species OMe

OMe +L

MeO

reflux, - L Ir

iPr2P Cl

H

PiPr2

44

+ CO

MeO L

iPr2PCl

Ir

H

PiPr2

iPr2P Cl

Ir

H

OC

PiPr2

iPr2P Cl

45: L = MeCN 46: L = CO

Ir

H

PiPr2

48

47

Scheme 20 Hemilabile sidearm-controlled coordination

OMe MeO

iPr2P Cl

Ir H 44

PiPr2

OMe

OMe

NaOtBu toluene-d8 50 oC

MeO

iPr2P

Ir

H2

PiPr2

49

Scheme 21 Hemilabile sidearm-controlled reductive elimination

MeO

iPr2P

Ir H

H 50

PiPr2

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

109

Fig. 8 The iridium and ruthenium complexes possessing different sidearms

H H

HO

60 oC, -H2

H Ph2P

H

OH

+ H2

Ir

HO

O H

OH

O

RT, -H2

PPh2

Cl 55

51 OH

OH

Ir

Ph2P

PPh2

Cl

H

O

H

O H

O

OH

H Ph2P Cl 52

Ir

Ir PPh2

Cl 56

H H

H2N

NH2 2 H+A-

H Ph2P

Ph2P

PPh2

Ir PPh2

Cl 53

AH3N

H H

ANH3 -H2

H

Ph2P Cl

H

Ir PPh2 57

H

HN

Ph2P Cl

ANH3

Ir 58

A-

PPh2

Scheme 22 Ligand-metal cooperation in the bifunctional PC(sp3)P pincer complexes

(56) strongly support this hypothesis. Hydrogen formation, however, proceeds at a different rate according to 1H NMR spectroscopy: whereas the hydride signals of 51 disappear after 30 min of heating to 60 C, with 52 this process occurs at room temperature. This reactivity trend is most likely dictated by the acidity of the corresponding protons in 51 and 52, as well as the thermodynamic stability of the resulting 55 and 56. Expectedly, a reversed process followed the opposite trend: when 55 and 56 were pressurized with 500 kPa of hydrogen at 100 C in CDCl3 side by side, the parent compound 51 was regenerated after 3 h, whereas only ca. 5% of 52 re-formed. It was also demonstrated that the regeneration of the hydride species is possible using the usual hydrogen surrogates such as alcohols, formic acid derivatives, as well as hydrides [82–84]. On the other hand, amine-containing 53 is stable in solution, and no hydrogen formation was observed unless external acids (e.g., p-toluenesulfonic, acetic, or formic acid) are added. In this case, the hydride signal quickly disappears, followed by hydrogen liberation independent of the strength of the acid. It is therefore assumed that H2 originates from intramolecular (rather than intermolecular)

110

A. Singh et al.

protonolysis of the Ir-H bond with the aid of the protonated sidearm in 57, leading to the formation of a cationic intermediate 58 stabilized by amine chelation (Scheme 22, bottom). This reactivity was put into practice in catalysis. Thus, oxidation of 1-phenylethanol in the presence of 0.1 mol% of 51 in p-xylene with heating under reflux and in an N2 atmosphere led to the formation of acetophenone as the sole product after 10 h. Similar activity was displayed by compound 54; however, none of the iridium hydride complexes that lack acidic sidearms (e.g., 44 or 47) were active under these reaction conditions. The functionalized complex 51 proved to be a very prominent catalyst for dehydrogenating secondary alcohols to ketones as well as for primary alcohols to esters and lactones (Scheme 23) [78]. Mechanistically, these transformations are facilitated by the presence of the appended functional group and proceed through the following sequence of elementary steps: a) an H2-forming step, leading to the formation of the arm-closed iridium species 55; b) a ligand exchange step, leading to the arm-open iridium alkoxide species; and c) the regeneration of the Ir-H catalyst 51 by β-hydride elimination with subsequent formation of the oxidized product (Scheme 24). Analogous ruthenium complexes exhibited similar reactivity. For example, selective and efficient homo- and cross-coupling of alcohols, resulting in the formation of

Scheme 23 Acceptorless dehydrogenation of alcohols using 51 as a catalyst

Scheme 24 Plausible mechanism of the acceptorless dehydrogenation of alcohols by 51

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

111

Scheme 25 Bifunctional ruthenium complex 54 as a catalyst for the cross-coupling of alcohols

the corresponding ketones in very good to excellent yield, was realized using 54 as a catalyst [83, 85]. Excellent yields (87–94%) of the corresponding cross-coupled products were obtained between differently substituted aromatic alcohols, depending on whether they are electron-rich, electron-deficient, or aliphatic, regardless of their enolizable nature (Scheme 25). A new series of structure-reactivity relationship studies on the decomposition of formic acid in different formulations was performed using complexes bearing different pendant functional groups (51 and 53) [77]. All of these catalysts were found to be active at 70 C using the constant 1:2000 catalyst/FA ratio degrading formic acid to hydrogen and carbon dioxide without contamination of carbon monoxide, albeit with a different efficiency. Thus, when the reaction was performed in a dimethoxyethane solution of HCO2H/Et3N azeotrope, an initial TOF of 13,710 h1 was achieved by 51, showing 3.18  105 turnovers (after repetitious injections of FA). Under the same conditions, 53 was about one order of magnitude less reactive. An opposite reactivity trend was observed under amine-free conditions. Thus, when amine was excluded from the formulation, amine-containing 53 exhibited superior activity over hydroxyl-containing 51 with almost doubled TON and TOF: 4.98  105 versus 1.89  105 and 18,390 h1 versus 9,460 h1. Furthermore, it was found that 53 was active even in neat formic acid, which is comparable to or exceeds the performance of the state-of-the-art catalysts operating under amine-free conditions and in the absence of exotic additives [86–89]. A plausible sequence of mechanistic events for the decomposition of FA by 53 was suggested (Scheme 26). Protonation of precatalyst 53 was proposed to take place in the presence of diluted or neat formic acid forming an actual catalytic species (57). Obviously, protonation of the amine groups is essential and facilitates solubilization of the catalyst. Next, activated 57 starts a turnover by releasing dihydrogen via intramolecular protonolysis of the Ir-H bond, along with the subsequent formation of a cationic 16-electron intermediate 58, as demonstrated in Scheme 22. Unlike the traditional and generally accepted mechanism, where protonolysis of such metal-hydride intermediates by FA may be mediated intermolecularly by water or external amines [90–93], it is believed that an intramolecular interaction with a highly acidic pendant ammonium group greatly facilitates hydrogen liberation from 57. The formation of a coordinatively saturated iridium formate complex 59 proceeds via 1,2-addition of a molecule of formic

112

A. Singh et al. NH2 H 2N H Ph2P

2 HCO2H

Ir PPh2

Cl 53

CO2

H Ir

Ph2P CO2

O H Ph2P

H3 N

H N O 3 O H Ir Ph2P Cl

Ir PPh2 Cl H-bound 60

PPh2

Cl

fast

57

slow

NH3 2 HCO2

O

NH3 2 HCO2

H3N

NH3

2 HCO2-

NH3

HN

PPh2

Cl

fast

59

HCO2-

Ir

Ph2P

PPh2

H2

58 HCO2H

Scheme 26 Plausible mechanism for the decomposition of FA by 53

NH3 2 HCO2

H2N H Ph2P

Ir

TS57

2 HCO2-

H N O 2 H O

H

Cl

NH3

PPh2

H Ph2P

Ir Cl

PPh2

TS59

Fig. 9 Secondary interaction-stabilized transition states

acid across the coordination Ir-N bond of 58. Finally, extrusion of CO2 from 59 and the regeneration of 57 complete the catalytic cycle. However, taking into account the steric hindrance of the iridium center in 59, a coordination switch from O-bound formate (59) to H-bound formate (60) prior to CO2 elimination was hypothesized. Quantum chemical calculations confirmed this assumption and validated the suggested mechanism. The computations confirmed that the first step in the cycle proceeds via polar 1,2-Ir-H. . . .H3N-elimination of dihydrogen from 57 via TS57 located only 45.2 kJ mol1 above 57, resulting in the formation of the cationic iridium intermediate 58. This step is chelation assisted, which makes it exergonic by 49.7 kJ mol1. Finally, the rate-determining, carbon dioxide extrusion step from coordinatively saturated 59 was found to proceed via TS59, which is located 140.0 kJ mol1 above the Ir-formate intermediate 59 with the cleaved Ir-O bond and non-classical O2C. . .H. . .Ir contact. The stabilizing interactions between the positively charged ammonium site and the negatively charged carboxylate make the β-H elimination of CO2 at the coordinatively saturated metal center possible via an intramolecular outer-sphere mechanism (Fig. 9).

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

113

3 N-Based Pincer Complexes Increasing attention has recently been paid to pincer-type complexes with a central amine group, due to the versatility of their coordination chemistry and a more versatile synthesis of the N-based scaffolds. In principle, the amine can be bound to alkyl, alkenyl, or aryl groups as well as heteroatom-containing groups such as silanes. The donor groups may incorporate phosphines, carbenes, heterocycles, and heteroatoms, which makes N-based pincer ligands attractive candidates for the installation of the secondary coordination sphere.

3.1

N(sp2)-Based Pincer Complexes Possessing Lewis Acid Functionalities

A pyridine moiety is highly robust; therefore, it serves as an accessible scaffold for constructing pincer ligands. In 2013, Szymczak et al. combined Lewis acid/base functionalities and a transition metal in a bifunctional catalyst capable of activating inert molecules (A–B) by synergistic interactions of the filled d orbitals on the metal with antibonding orbitals of A–B at the primary coordination sphere, along with destabilization of the bonding orbitals by interactions with a Lewis acid/base pair at the secondary coordination sphere (Scheme 27). To realize this idea, a robust terpyridine-based scaffold was modified to incorporate a morpholine/pinacol boronic ester as a Lewis acid/base pair (61) and was then metalated by reaction with vanadium(III) chloride (Scheme 28) [94]. The crystal structure of the pincer complex 62 featured a distorted octahedral geometry around vanadium, but with an apparent distortion of the axial chloride ligand toward the boronic ester group, as illustrated by a Cl-V-N angle of 162.0 and a short B-Cl bond of 2.41 Å, consistent with B-Cl interaction.

Scheme 27 Lewis acid/base pair-assisted activation of molecules

Scheme 28 Coordination chemistry of 61

114

A. Singh et al. O

O N Cl N

Cl V

O O B N Cl

N

N2H4, Et3N Et3NHCl

HO H O H N N N Cl B V N N Cl N 63

62

Scheme 29 Secondary interaction-assisted activation of hydrazine tBu

tBu NH N

N

Br NaH

64

N N N

B Br

Fe Br

N N

B

tBu 65

N N

B

N

N N tBu

tBu N N

9-BBN

N

N NH tBu

tBu N N

B

tBu 66

67

Scheme 30 Synthetic approach to the synthesis of Lewis acid-functionalized pincer complexes

Addition of hydrazine to 62, in the presence of NEt3, results in the formation of paramagnetic 63 (Scheme 29). The solid-state structure of 63 revealed an unusual η2 coordination mode of the deprotonated hydrazine molecule: the vanadium center is spanned by both Lewis acidic and Lewis basic sites at the secondary coordination sphere. The relevance of this bifunctional system to reductive N-N bond cleavage was demonstrated by treating 63 with stoichiometric amounts of [CoCp*2] that furnished 0.6  0.2 equivalents of ammonia, which is consistent with reduction-induced N-N bond cleavage. Aiming at the challenging disproportionation of hydrazine to ammonia and N2, the same group re-designed the bifunctional pyridine-based pincer complex to include two appended Lewis acidic sites that could enable an activation-reduction sequence involving inert substrates. The second model, ligand 66, was made by modifying a previously reported 2,6-bis(5-tert-butyl-1H-pyrazole-3-yl)pyridine (64) [95] by reacting it with allyl bromide, followed by the hydroboration of 65 with 9-BBN, yielding the desired scaffold 66. Metalation of 66 with FeBr2 in CH2Cl2 successfully provided the second-generation catalysts (67) [96] (Scheme 30). Exposing 67 to hydrazine proved that nitrogen-based substrates interact with the Lewis acids located in the secondary coordination sphere of the catalysts. It was found that the corresponding 68 and 69 form, upon reaction, with one or two equivalents of N2H4, respectively (Scheme 31). An attempted reduction of 68 with two equivalents of KC8 indeed resulted in N-N bond cleavage, accompanied by forming an unusual bis(amido)iron complex (70) (Scheme 32). Analysis of single crystals of 70 revealed a nearly square-pyramidal geometry about iron and the formation of two new amido ligands with the Fe-NH2 distance of

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

NH2 H 2 N B B N H2 H2N Br Br tBu

N N

Fe N

B 2 N2H4 THF

N N

tBu

Br

tBu

N N

69

N2H4 THF

Br Fe N

H2 N B N H2 Br Br

B

B

N N

tBu

tBu

115

N N

Fe N

N N

tBu

68

67

Scheme 31 Lewis acid-assisted fixation of hydrazine

B

H2 N Br

tBu

N N

Fe N

N B H2 Br N N

B

B KC8 THF tBu

NH2 H2N tBu

68

N N

Fe N

N N

tBu

70

Scheme 32 Lewis acid-assisted activation of hydrazine

ca. 2.075 Å, which is markedly longer than in other Fe-NH2 complexes reported earlier because of coordination of the amido lone pairs to the Lewis acidic appended functional groups (the B-N distance is ca. 1.633 Å). It is worth noting that the suggested platform can be classified as modular, since every conceivable type of donor could be incorporated into the ligands, and they can be functionalized with numerous appended boron-based Lewis acidic motifs in the secondary coordination sphere fitting most metal centers for different applications. For example, a series of carbene CNC iron pincer complexes crafted with a variety of Lewis acidic sidearms have been prepared for structure-reactivity studies targeting the elucidation of parameters affecting the reductive cleavage of hydrazine [97]. Here, 2,6-bis(N-allylimidazolium)pyridine (71) was used as a common structural precursor that was hydroborated with a variety of borane precursors after the successful metalation (Scheme 33). The studies showed that the nature of the secondary coordination sphere has a strong influence on the ability of the molecule to capture hydrazine: weakly acidic BPin is incapable of complexing hydrazine, whereas sterically demanding B(sia)2 Lewis acids favor intermolecular coordination. Only the moderately acidic and sterically accessible BBN fragment was found capable of hydrazine cleavage. In a series of interesting papers, Szymczak and co-workers described the (2-pyridylimino)isoindolate scaffold as another modular NNN pincer ligand allowing communication between the primary and the secondary Lewis acid coordination toward cooperative activation of inert bonds. The synthesis of this platform takes advantage of the straightforward condensation between, for example, 2-bromo-6-aminopyridine (77) and phthalonitrile (78) to form bis(20 -bromo-60 iminopyridyl)isoindoline (79). Nucleophilic substitution at 79 with sodium benzyl alcoholate, followed by BBr3-mediated ether hydrolysis, yielded the NNN pincer

116

A. Singh et al.

Scheme 33 Synthetic approach to Lewis acid-functionalized pincer complexes 74–76

Scheme 34 Synthesis of the pincer ligands 81 and 82

ligand 81, which possesses hydroxyl groups in the proximity to the primary coordination sphere. Other variants, such as methyl-substituted 82, can be prepared in one step (Scheme 34). Complexation studies revealed that 81 is smoothly metalated with RuCl2(PPh3)3 in the presence of stoichiometric amounts of NaOCH3 and forms ruthenium (II) pincer complex 83 [98]. Tridentate 83 is readily dehydrohalogenated with

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

117

Scheme 35 Coordination chemistry of the ligand 81

PPh3 N N N

N N

Ru

H

O PPh3 84

O

HBPin -H2, -PPh3

PPh3 N N N N Ru O H N B O O O O B O 85

Scheme 36 Plausible intermediate state of B-H heterolysis by 84

another equivalent of NaOCH3 and produces tetradentate 84 by intramolecular coordination to a deprotonated hydroxyl function stabilized by a weak hydrogen bond between the hydroxypyridine and pyridonate arms (Scheme 35). As was demonstrated in previous studies on bifunctional complexes possessing appended hydroxyl groups, an interplay between alkoxide and alkoxyl coordination can facilitate bond activation/formation processes [99–101]. Therefore, the catalytic potential of the coordination switching to 84 was evaluated by exposing it to H-BPin, which resulted in hydrogen evolution and the formation of a new compound (85) bearing two pinacolyl boronate groups (Scheme 36) [98]. The solid-state and NMR data indicate the presence of a weak B-H interaction, suggesting that 85 may represent a plausible intermediate state of B-H heterolysis and Lewis acidity of the boron centers. Based on this observation, compound 84 was subjected to catalytic nitrile hydroboration with HBPin, and it was found to be a prominent catalyst after being activated by deprotonation. The active species 86 exhibited full conversion of a wide range of benzonitriles into the corresponding benzylamines with TOF as high as 1.19  102 s1 (Scheme 37). The plausible mechanism for the nitrile hydroboration reaction shown in Scheme 38 suggests that the real catalytic species is the bifunctional pincer complex 87 possessing a Lewis acidic secondary coordination sphere [98]. The catalytic cycle starts with the activation of the B-H bond that takes place in the reaction of 86 with HBPin with the loss of phosphine (87), as described in Scheme 38. The H and BPin transfer across the nitrile substrate proceeds following outer-sphere activation by the pendant Lewis acid groups (89 and 91). After demonstrating the critical role of the in situ-formed secondary coordination in cooperative substrate activation, a structurally well-defined catalyst was designed

118

A. Singh et al. PPh3 N N

N

N

PPh3 N

Ru N

KN(SiMe3)2 18-crown-6 Benzene

O H

O

N

N

N

Ru

PPh3 86

84 CN 86 (5 mol%) HBPin, 45 oC C6H6

N(BPin)2

HCl

R

NH2 R

NH2 94%

K

O

PPh3

R

O

N

NH2 MeO

91%

NH2 44%

Scheme 37 Catalytic nitrile hydroboration using 86 as a catalyst

Scheme 38 Plausible mechanism for the nitrile hydroboration

CF3

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

2

N

N

PPh3 N

N PPh3N Ru

N

119

N

CH3 NaOtBu

N

N

N

Ru H2C

N

CH2 Ru

Cl

N

92 93

N

N

N PPh3

Scheme 39 Unusual base-mediated C-H activation of the pendant methyl groups in 92

PPh3 N N

N 0.5

N

PPh3 N

N Ru H2C

N

CH2 Ru N 93

N

N

9-BBN -H2

N PPh3

N

N

N

PPh3 N

Ru N

CH B

94

H2 (1 atm) CO (1 atm) C6H6

N

N

N

Ru N

H

CO

CH2

B

95

Scheme 40 Lewis acid-assisted H2 activation

[102]. Studying the coordination and organometallic chemistry of the visually functionless ligand 82, Tseng et al. prepared a ruthenium complex 92 and discovered an unusual base-mediated C-H activation of the pendant methyl groups leading to dimerization of 92 to form 93 (Scheme 39). The reaction of 93 with 9-borabicyclo[3.3.1]-nonane (9-BBN) afforded 94 via extrusion of hydrogen from the B-H and CH2 groups (Scheme 40). Unlike the aromatization-dearomatization mechanism described for pyridine-based pincer ligands [103], the aromaticity in 94 was retained. Remarkably, a rare Ru-(η2-BC) interaction and a weak donating Ru-B interaction might be suggested, based on the Ru-C and Ru-B distance (2.521 and 2.592 Å, respectively) according to X-ray analysis. The noninnocent behavior of the appended Lewis acid was probed by the hydrogen activation experiment, which yielded a new product 95 in the presence of stabilizing CO. Its solid-state analysis revealed a 1,2-addition mode of H-H across the Ru-C bond, proving that hydrogen cleavage proceeded with help from the pendent boron Lewis acid (Scheme 40). Cooperative substrate activation was demonstrated not only in stoichiometric but also in catalytic reactions. Thus, comparison of the catalytic performance displayed by 92 and 94 for alkyne hydrogenation revealed that 94 promotes highly selective and fast semi-hydrogenation of diphenylacetylene to Z-stilbene, whereas 92 leads to a mixture of stereoisomers as well as an over-hydrogenated product (Scheme 41). Remarkably, also the addition of Lewis basic additives such as N, N-diethylpropargylamine decreased both the conversion and selectivity, thus supporting the hypothesis that incorporation of a secondary Lewis acidic site, such as 9-BBN, imparted a new reactivity based on metal-appended group cooperation [102].

120

A. Singh et al. Cat. (1 mol%) H2 (2 atm) C6H6, 80 oC

Catalyst 92

No additives

Conv. 65%

Sel. 31:19:15

Catalyst 94

No additives

Conv. 100%

Sel. 98:2:0

Catalyst 94

amine

Conv. 50%

Sel. 39:10:1

Scheme 41 Catalytic alkyne semi-hydrogenation catalyzed by 92 and 94

O HN NH EtO2C

N

N H

Base N OH

O

O P

N

Base

N

O O

H

Zn N

NH2

HN

N

NH N

OH

O OH

HN HN

O

N

N H

O OH H

O

O O P O O NH

H2N 96

base

N

H Base H

97

OH OH

2+ N

O O

N

N H2O

N

Zn

N

OH2 98

Fig. 10 RNA hydrolysis assisted by H bonding interactions

3.2

N(sp2)-Based Pincer Complexes Possessing Hydrogen Bonding Functionalities

Hydrogen bonding interactions by means of Brønsted acids or bases with metal complexes provided an opportunity for hydrogen transfer to/from metal-coordinated substrates. In 1993, Anslyn and co-workers discovered a metal-free enzyme mimic (96) formed by incorporating active site functional groups of nucleases that exhibited very high RNA hydrolysis rates. Figure 10 (left) shows the organocatalytic entity in which imidazole acts as a shuttle to deliver the hydroxyl group to the phosphodiester linkage, whereas the phosphorane transition state is stabilized by hydrogen bonding from 96 [104]. A decade later, the same group offered a zinc-based pincer catalyst 97 by appending guanidinium groups on the terpyridine-like framework (Fig. 10, center) [105]. The reactivity of this bisguanidinium cleft was examined in RNA hydrolysis. The rate of phosphoadenine ester hydrolysis with a guanidinium unit was about 3,300 times greater than that of analogous 98 with an appended methyl unit (Fig. 10, right) [105]. The similar non-covalent interactions between the zinc-bound hydroxo ligand and the phosphate group during hydrolysis helped to approach the natural RNA hydrolysis rates in artificial systems.

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

121

Exploring this concept, Gilbertson and co-workers took advantage of the subsequent condensation of 2,6-diacetylpyridine with aniline and the N, N-dialkylethane-1,2-diamine of choice to synthesize a series of modular tetradentate pyridine-2,6-diimine-based pincer ligands mimicking active sites of iron-containing metalloenzymes such as lipoxygenases, nitrile hydratases, acid phosphatases, and dioxygenases (Scheme 42) [106–113]. Solid-state investigation of the zinc (99) and iron (100) complexes bearing these functional pincer ligands revealed a secondary coordination sphere H bonding between MII-Cl and the protonated pendant diisopropylamine base (Scheme 43) [107]. The hydrogen atom involved in this interaction was located on the electron density map of both complexes, providing an accurate N-HCl distance of 2.17 Å in 101 and 2.18 Å in 102. Since metalloenzymes are recognized for their ability to tune the redox states of the active site, as well as the protonation state of the surrounding residues in the secondary coordination sphere, cooperation between the redox activity and the protonation state of the functionalized scaffolds was probed. To this end, 100 was reduced under a CO atmosphere with sodium amalgam producing the reduced complex 103, which can be successfully protonated to yield 104 quantitatively (Scheme 44). It was found that both complexes 103 and 104 undergo a quasireversible one-electron oxidation of the ligand scaffold, albeit at different potentials. For example, for the neutral form 103, the oxidation of the ligand occurs at E1/2 ¼ 0.590 V, whereas upon protonation, the ligand becomes more difficult to oxidize, and the process takes place at E1/2 ¼ 0.485 V for 104. A series of structure-reactivity experiments, comparing the initial rates of nitrite reductions, revealed a strong dependence of the reaction rate on the pKa value of the

Scheme 42 General approach to the tetradentate pyridine-2,6-diimine-based pincer ligands possessing an appended amine functionality

Scheme 43 Ligand-metal interactions induced by the appended amine functionality

122

A. Singh et al. PF6-

N iPr

N iPr

N Fe Cl Cl

iPr

NaHg

N

CO

iPr 100

iPr

N N N Fe OC CO iPr N iPr iPr 103

NH4PF6 MeCN/H2O

N N N Fe OC CO iPr N H iPr iPr 104

iPr

Scheme 44 Reactivity of the pyridine-2,6-diimine-based pincer complex 100

Fig. 11 The pyrrolidine-functionalized 105 and morpholine-functionalized complex 106

Br

Br N

N N

7 mol% Pd(OAc)2 14% DPPF NaOtBu toluene, 110 oC

OtBu

OtBu

N

N

HCOOH

N

108

107 OH

OH

N

N

O

O NH

HN

N

N

109

109'

Scheme 45 Synthetic approach to bifunctional pincer ligand 109

pendant amines and their steric bulk. Thus, when triethylammonium chloride (pKa ¼ 18.8 in CH3CN) was used as the acid source, an initial rate of 2.30  108 M s1 was observed for the morpholine-based 106 (pKa ¼ 17.1 in acetonitrile). However, a twofold increase in the initial rate (3.97  108 M s1) was found for the bulky 103 (pKa ¼ 19.2 in acetonitrile) and threefold (7.07  108 M s1) in the pyrrolidine-functionalized 105 (pKa ¼ 18.3 in acetonitrile). These results can be explained by the lower pKa value of the pendant shuttle in 106 (Fig. 11) [111–113]. Terpyridine ligands containing hydroxyl groups at the 6- and 60 -positions, designed by Szymczak, are ideally suited for creating hydrogen bonding interactions between the secondary coordination sphere environment and metal-bound substrates. Ligand 109 was synthesized by converting 6,60 -dibromo terpyridine (107) to 6,60 -di-tert-butoxy terpyridine (108) using a catalyst derived from Pd(OAc)2 and 1,10 -bis(diphenylphosphino)ferrocene (DPPF), followed by acid-mediated dealkylation (Scheme 45) [114].

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . . Scheme 46 Chemoselective transfer hydrogenation of unsaturated ketones catalyzed by 110

OH

OH

N

N N

O RuCl2(PPh3)3

123 H N

109 O

Cl Ru L N

H L

O

+ PF 6

N

110 L = PPh3 110, KtOBu

OH

iPrOH 95%

Metalation of 109 with RuCl2(PPh3)3 in methanol in the presence of NH4PF6 provides octahedral ruthenium(II) complex 110 with trans-located PPh3 ligands perpendicular to the plane of the molecule and close-contact interactions between the Ru-bound chloride and the hydroxide sidearm, consistent with intramolecular hydrogen bonding. Ruthenium(II) complex 110 was capable of efficiently catalyzing the transfer hydrogenation of a variety of ketones in the presence of KOtBu in 2-propanol with great chemoselectivity in the presence of substituted alkenes, which is rarely observed in conventional catalysts (Scheme 46). The mechanistic rationale for the chemoselective transfer hydrogenation of carbonyls in the presence of double bonds assumes the assistance of the secondary coordination sphere. Thus, the catalytic cycle starts the formation of the hydride complex 111 via a halide-isopropoxide exchange/β-hydride elimination sequence. Following hydride formation, the attack on the substrate proceeds through an associative transition state 112 in which the carbonyl group is activated with the aid of interaction with potassium alkoxide tether to form 113 – a strong cation effect on the reaction rate supports this assumption. Protonolysis of 113 with the substrate alcohol, followed by β-hydride elimination, regenerates the catalytically active 111 (Scheme 47) [115].

3.3

Binuclear Reactivity of N(sp2)-Based Pincer Complexes

At least one-third of studied enzymes contain metals as cofactors responsible for catalyzing a broad spectrum of reactions. In the case of binuclear metalloenzymes, the catalytic activity is underlined by the interplay between the closely positioned metal centers that provide many advantages in terms of charge delocalization, lower activation barriers, and cooperative activation of substrates. During the last decades, tremendous efforts have been invested into mimicking various metalloenzymes by synthesizing structurally well-defined polynuclear complexes, in particular, heterobimetallic ones. Although considerable progress has been made in this field, accessing heterobimetallic compounds in a controlled fashion without contamination with statistical mixtures, oligomers, nanoparticles, and other complications remains challenging [116]. One commonly accepted strategy for directing and stabilizing

124 Scheme 47 Plausible mechanism of the chemoselective transfer hydrogenation of carbonyls in the presence of double bonds

A. Singh et al.

PPh3 N

R

R'

PPh3 N

H

R' R

O

PPh3

rate determining step

N

Ru

112

O

N

O

H

N

Ru

N

CO

O K

O

CO

O K

N

N

Ru

N

O

K

111

O

CO

H

R' R

O

113 PPh3 N O

N O K

N

Ru

OH O

CO

OH

H

R

R'

O 114

Fig. 12 Design of the macrocyclic pincer ligand mimicking CODH catalytic site

O C

Ni

S Fe

S

O O Fe

S

Fe

HN 117

Fe S

O

N NH

N

N N

heterobimetallic complexes is to create distinct coordination sites for selective binding of hard and soft metal centers; however, this alone does not guarantee success. In this context, the creation of distinctive environments within tridentate pincer-type motifs may significantly amplify selectivity and help to obtain well-defined systems in a highly efficient manner. It was demonstrated that the bond-sharing metallobicyclic motif in transition metal pincer complexes provides additional stabilization of the metal; therefore, they are less prone to reversible dissociation and ligand redistribution than the corresponding chelate complexes. Similarly, main group metals will also benefit from multidentate pincer-like chelation. In 2010, Holm and co-workers sought to mimic the catalytic site of carbon monoxide hydrogenases (CODH) that catalyze reversible two-electron CO to CO2 oxidation [117, 118]. These enzymes have a Ni-Fe binuclear core and, according to the recent structural studies, contain a common structural motif consisting of a cubic NiFe3S4 cluster with an additional appended iron sidearm (Fig. 12) [119, 120]. Considering the inaccessibility of the suitable binucleating platform, a

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . . NH2 H2N N

O2N

O2N

N

O

N N O

O

H2N

NH2 N

N Cl

1) NaBH4 2)H2CO, HOAc 3) Pd/C/N2H4

NO2

115

125

O Cl

O

N NH

HN

N N

N

116

N N

117 (17%) O

O

N NH

HN 119

N

N X

N

118: X = NCH3 119: X = S

Scheme 48 Synthetic approach to macrocyclic pincer ligands

macrocyclic NNN-pincer ligand 117 was engaged in reproducing the face-to-face bimetallic core. The prototypical macrocycle 117 was synthesized from 3-nitrobenzaldehyde in an overall 17% yield via intermediates 115–116 by the sequence of trivial reactions (Scheme 48). A slight modification of the bridge produces other members of the family of these CODH mimics possessing aza-crown (118) and thioazacrown motifs (119). Nickel was installed in the pincer site by complexation of 117 with Ni(OTf)2 under basic conditions to yield hydroxo complex 120, which was readily converted to cyanide complex 121 by exposure to a potent cyanide ligand. NNN-pincer coordination of nickel(II) results in a conformational change in the orientation of the phenyl rings to have mirror symmetry. Three NiII-FeII complexes, with cyanide (122), hydroxy (123), and carbonate bridges (124), were prepared by insertion of FeII in 121 (122) and in 120 (123 and 124). In all these species, NiII was found in a planar coordination environment with respect to the NNN pincer site, while the geometry of the FeIIN3ClL units was between trigonal pyramidal and square pyramidal (Scheme 49). The cooperative reactivity of such systems was demonstrated by Tolman and co-workers [121] who employed this binucleating platform for the design of artificial methane monooxygenase enzyme-mimicking catalysts [122]. Serendipitously, an attempted synthesis of the mixed-valent dicopper(I,II) complex 127 leads to the formation of only one isolable product 126 in 20% yield (Scheme 50). An X-ray crystal structure of 126 revealed 2-hydroxytetrahydrofuran in deprotonated form, bridging the CuII and NaI centers. Tetrahydrofuran (solvent) oxidation was

126

A. Singh et al.

Scheme 49 Coordination chemistry of the macrocyclic pincer ligands

Scheme 50 The design of artificial methane monooxygenase enzyme-mimicking catalysts

suspected. To evaluate if the O atom in the product originates from the hydroxide in 125, the reaction was repeated with an 18O-labeled analog unequivocally proving the relevance of the bimetallic species to oxidation catalysis. It was also demonstrated that the very distinctive donor arrays offered by this design allow synthesis and full characterization of a series of copper-cooper, copper-palladium, and copperplatinum bimetallic complexes (127–129) for further structure-reactivity studies (Scheme 50) [123].

4 Conclusions As evident by the consistently high number of publications on the chemistry of pincer complexes in the past 35 years, they remain in the domain of very intense academic activity owing to their high chemical and thermal stability as well as their

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

127

singular structural features. The well-defined structure and tunability of pincer compounds often offer improved and, in some cases, unique catalytic properties when compared to conventional systems. However, after many years of very fruitful research in this area, further progress toward new, efficient catalysis should not remain limited to routine structure-reactivity screening – it requires the development of new reaction schemes and reactivity patterns. Designing catalysts possessing noninnocent ligands is a very new development in organometallic chemistry. This trend also appears as a fresh direction in the chemistry of pincer complexes, resulting in many publications concerning ligand-metal cooperating pincer catalysts operating via diverse mechanisms. This chapter focused on recent progress in the chemistry and catalytic applications of pincer compounds incorporating the secondary coordination sphere or an appended functionality that can interact with the catalytic center or can modulate its reactivity via secondary substrate-catalyst interactions. We overviewed the intriguing reactivity of such multifunctional catalytic systems while trying to stress the need for new platforms that combine the concepts of modular and stable pincer ligands and outer-sphere reactivity, thus providing excellent opportunities for creating challenging targetoriented catalysis. Acknowledgments This research was supported by GIF (German-Israeli Foundation for research and development) Grant N I-1508-302.5/2019. This work was supported by the RUDN University Program “5-100.” We are grateful for all stimulating discussion with many colleagues and friends and in particular for all work and insights provided by current and past group members.

References 1. van Koten G (2013) Pincer ligands as powerful tools for catalysis in organic synthesis. J Organomet Chem 730:156–164. https://doi.org/10.1016/j.jorganchem.2012.12.035 2. Deng Q-H, Melen RL, Gade LH (2014) Anionic chiral tridentate N-donor pincer ligands in asymmetric catalysis. Acc Chem Res 47(10):3162–3173. https://doi.org/10.1021/ar5002457 3. Asay M, Morales-Morales D (2015) Non-symmetric pincer ligands: complexes and applications in catalysis. Dalton Trans 44(40):17432–17447. https://doi.org/10.1039/C5DT02295A 4. Andrew RE, Gonzalez-Sebastian L, Chaplin AB (2016) NHC-based pincer ligands: carbenes with a bite. Dalton Trans 45(4):1299–1305. https://doi.org/10.1039/C5DT04429D 5. Melen RL, Gade LH (2016) New chemistry with anionic NNN pincer ligands. Top Organomet Chem 54:179–208. https://doi.org/10.1007/3418_2015_114 6. Lawrence MAW, Green K-A, Nelson PN, Lorraine SC (2018) Review: pincer ligands-tunable, versatile and applicable. Polyhedron 143:11–27. https://doi.org/10.1016/j.poly.2017.08.017 7. Maser L, Vondung L, Langer R (2018) The ABC in pincer chemistry - from amine- to borylene- and carbon-based pincer-ligands. Polyhedron 143:28–42. https://doi.org/10.1016/j. poly.2017.09.009 8. Alig L, Fritz M, Schneider S (2019) First-row transition metal (De)hydrogenation catalysis based on functional pincer ligands. Chem Rev 119(4):2681–2751. https://doi.org/10.1021/acs. chemrev.8b00555 9. Albrecht M, van Koten G (2001) Platinum group organometallics based on "pincer" complexes: sensors, switches, and catalysts. Angew Chem Int Ed 40(20):3750–3781. https://doi. org/10.1002/1521-3773(20011015)40:203.0.CO;2-6

128

A. Singh et al.

10. Wieczorek B, Dijkstra HP, Egmond MR, Klein Gebbink RJM, van Koten G (2009) Incorporating ECE-pincer metal complexes as functional building blocks in semisynthetic metalloenzymes, supramolecular polypeptide hybrids, tamoxifen derivatives, biomarkers and sensors. J Organomet Chem 694(6):812–822. https://doi.org/10.1016/j.jorganchem.2008.12. 010 11. Suijkerbuijk BMJM, Schamhart DJ, Kooijman H, Spek AL, van Koten G, Klein Gebbink RJM (2010) Mono(NCN-pincer palladium)-metalloporphyrin catalysts: evidence for supramolecular bimetallic catalysis. Dalton Trans 39:6198–6216. https://doi.org/10.1039/b925236n 12. Jurgens S, Kuhn FE, Casini A (2018) Cyclometalated complexes of platinum and gold with biological properties: state-of-the-art and future perspectives. Curr Med Chem 25(4):437–461. https://doi.org/10.2174/0929867324666170529125229 13. Junge K, Papa V, Beller M (2019) Cobalt-pincer complexes in catalysis. Chem Eur J 25 (1):122–143. https://doi.org/10.1002/chem.201803016 14. Liu J-K, Gong J-F, Song M-P (2019) Chiral palladium pincer complexes for asymmetric catalytic reactions. Org Biomol Chem 17(25):6069–6098. https://doi.org/10.1039/ c9ob00401g 15. Dehand J, Pfeffer M (1976) Cyclometallated compounds. Coord Chem Rev 18(3):327–352. https://doi.org/10.1016/s0010-8545(00)80431-2 16. Shaw BL (1975) Formation of large rings, internal metalation reactions, and internal entropy effects. J Am Chem Soc 97(13):3856–3857. https://doi.org/10.1021/ja00846a072 17. Riehl JF, Jean Y, Eisenstein O, Pelissier M (1992) Theoretical study of the structures of electron-deficient d6 ML5 Pi complexes. Importance of a .Pi.-donating ligand. Organometallics 11(2):729–737. https://doi.org/10.1021/om00038a035 18. van de Kuil LA, Luitjes H, Grove DM, Zwikker JW, van der Linden JGM, Roelofsen AM, Jenneskens LW, Drenth W, van Koten G (1994) Electronic tuning of arylnickel(II) complexes by para substitution of the terdentate monoanionic 2,6-bis[(dimethylamino)methyl]phenyl ligand. Organometallics 13:468–477. https://doi.org/10.1021/om00014a020 19. Mohammad HAY, Grimm JC, Eichele K, Mack H-G, Speiser B, Novak F, Quintanilla MG, Kaska WC, Mayer HA (2002) CH oxidative addition with a (PCP)Ir(III)-pincer complex. Organometallics 21(26):5775–5784. https://doi.org/10.1021/om020621w 20. Batema GD, Lutz M, Spek AL, van Walree CA, de Mello DC, Meijerink A, Havenith RWA, Perez-Moreno J, Clays K, Buechel M, van Dijken A, Bryce DL, van Klink GPM, van Koten G (2008) Substituted 4,40 -stilbenoid NCN-pincer platinum(II) complexes. Luminescence and tuning of the electronic and NLO properties and the application in an OLED. Organometallics 27(8):1690–1701. https://doi.org/10.1021/om700352z 21. Roddick DM (2013) Tuning of PCP pincer ligand electronic and steric properties. Top Organomet Chem 40:49–88. https://doi.org/10.1007/978-3-642-31081-2_3 22. Crocker C, Empsall HD, Errington RJ, Hyde EM, McDonald WS, Markham R, Norton MC, Shaw BL, Weeks B (1982) Transition metal–carbon bonds. Part 52. Large ring and cyclometallated complexes formed from But2PCH2CH2CHRCH2CH2PBut2(R¼ H or Me) and IrCl3, or [Ir2Cl4(cyclo-octene)4] : crystal structures of the cyclometallated hydride, and the carbene complex [IrCl [IrHCl(But2PCH2CH2CHCH2CH2PBut2)], (But2PCH2CH2CCH2CH2PBut2)]. J Chem Soc Dalton Trans 7:1217–1224 23. Errington RJ, McDonald WS, Shaw BL (1982) Transition metal-carbon bonds. Part 54. Complexes of palladium, platinum, rhodium, and iridium with (Me3C)2PCH2CHMe(CH2)3P (CMe3)2. Crystal structures of [2-methyl-1,5-bis(di-tert-butylphosphino)pent-3-yl-C3 PP'] chloropalladium(II) and -chlorohydridoiridium(III). J Chem Soc Dalton Trans:1829–1835 24. Vigalok A, Milstein D (2000) Direct synthesis of thermally stable PCP-type rhodium carbenes. Organometallics 19(11):2061–2064. https://doi.org/10.1021/om990764r 25. Polukeev AV, Wendt OF (2018) Iridium complexes with aliphatic, non-innocent pincer ligands. J Organomet Chem 867:33–50. https://doi.org/10.1016/j.jorganchem.2017.12.009 26. O'Reilly ME, Veige AS (2014) Trianionic pincer and pincer-type metal complexes and catalysts. Chem Soc Rev 43(17):6325–6369. https://doi.org/10.1039/C4CS00111G

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

129

27. Chakraborty S, Bhattacharya P, Dai H, Guan H (2015) Nickel and Iron pincer complexes as catalysts for the reduction of carbonyl compounds. Acc Chem Res 48(7):1995–2003. https:// doi.org/10.1021/acs.accounts.5b00055 28. Kumar A, Bhatti TM, Goldman AS (2017) Dehydrogenation of alkanes and aliphatic groups by pincer-ligated metal complexes. Chem Rev 117(19):12357–12384. https://doi.org/10.1021/ acs.chemrev.7b00247 29. Noyori R, Yamakawa M, Hashiguchi S (2001) Metal-ligand bifunctional catalysis: a nonclassical mechanism for asymmetric hydrogen transfer between alcohols and carbonyl compounds. J Org Chem 66(24):7931–7944. https://doi.org/10.1021/jo010721w 30. Clapham SE, Hadzovic A, Morris RH (2004) Mechanisms of the H2-hydrogenation and transfer hydrogenation of polar bonds catalyzed by ruthenium hydride complexes. Coord Chem Rev 248(21–24):2201–2237. https://doi.org/10.1016/j.ccr.2004.04.007 31. Ikariya T, Murata K, Noyori R (2006) Bifunctional transition metal-based molecular catalysts for asymmetric syntheses. Org Biomol Chem 4(3):393–406 32. Fryzuk MD, MacNeil PA (1983) Amides of rhodium and iridium stabilized as hybrid multidentate ligands. Organometallics 2:355–356. https://doi.org/10.1021/om00074a030 33. Zhang J, Leitus G, Ben-David Y, Milstein D (2005) Facile conversion of alcohols into esters and dihydrogen catalyzed by new ruthenium complexes. J Am Chem Soc 127 (31):10840–10841. https://doi.org/10.1021/ja052862b 34. Deeken S, Motz G, Kempe R (2007) How common are true aminopyridinato complexes? Z Anorg Allg Chem 633(2):320–325. https://doi.org/10.1002/zaac.200600269 35. Gunanathan C, Gnanaprakasam B, Iron MA, Shimon LJW, Milstein D (2010) "Long-range" metal-ligand cooperation in H2 activation and ammonia-promoted hydride transfer with a ruthenium acridine pincer complex. J Am Chem Soc 132(42):14763–14765. https://doi.org/ 10.1021/ja107770y 36. Li H, Zheng B, Huang K-W (2015) A new class of PN3-pincer ligands for metal-ligand cooperative catalysis. Coord Chem Rev 293-294:116–138. https://doi.org/10.1016/j.ccr.2014. 11.010 37. Goncalves TP, Huang K-W (2017) Metal-ligand cooperative reactivity in the (pseudo)dearomatized PNx(P) systems: the influence of the zwitterionic form in dearomatized pincer complexes. J Am Chem Soc 139(38):13442–13449. https://doi.org/10.1021/jacs.7b06305 38. Mastalir M, Glatz M, Pittenauer E, Allmaier G, Kirchner K (2019) Rhenium-catalyzed dehydrogenative coupling of alcohols and amines to afford nitrogen-containing aromatics and more. Org Lett 21(4):1116–1120. https://doi.org/10.1021/acs.orglett.9b00034 39. Clarke ZE, Maragh PT, Dasgupta TP, Gusev DG, Lough AJ, Abdur-Rashid K (2006) A family of active iridium catalysts for transfer hydrogenation of ketones. Organometallics 25 (17):4113–4117. https://doi.org/10.1021/om060049z 40. Vasudevan KV, Scott BL, Hanson SK (2012) Alkene hydrogenation catalyzed by nickel hydride complexes of an aliphatic PNP pincer ligand. Eur J Inorg Chem 2012 (30):4898–4906. https://doi.org/10.1002/ejic.201200758 41. Schneider S, Meiners J, Askevold B (2012) Cooperative aliphatic PNP amido pincer ligands versatile building blocks for coordination chemistry and catalysis. Eur J Inorg Chem 2012 (3):412–429. https://doi.org/10.1002/ejic.201100880 42. Wang Z, Chen X, Liu B, Q-b L, Solan GA, Yang X, Sun W-H (2017) Cooperative interplay between a flexible PNN-Ru(II) complex and a NaBH4 additive in the efficient catalytic hydrogenation of esters. Cat Sci Technol 7(6):1297–1304. https://doi.org/10.1039/ C6CY02413K 43. Empsall HD, Hyde EM, Markham R, McDonald WS, Norton MC, Shaw BL, Weeks B (1977) Synthesis and X-ray structure of an unusual iridium ylide or carbene complex. J Chem Soc Chem Commun 17:589–590 44. Crocker C, Errington RJ, McDonald WS, Odell KJ, Shaw BL, Goodfellow RJ (1979) Rapid reversible fission of a carbon-hydrogen bond in a metal complex: X-ray crystal structure of

130

A. Singh et al.

cyclometalated rhodium complex. J Chem Soc Chem Commun 11:498–499. https://doi.org/ 10.1039/DT9820001217 45. Bernard LS (1980) Some steric, conformational and entropy effects of tertiary phosphine ligands. J Organomet Chem 200(1):307–318. https://doi.org/10.1016/s0022-328x(00)88652-0 46. Leis W, Mayer HA, Kaska WC (2008) Cycloheptatrienyl, alkyl and aryl PCP-pincer complexes: ligand backbone effects and metal reactivity. Coord Chem Rev 252 (15–17):1787–1797 47. Winter AM, Eichele K, Mack H-G, Kaska WC, Mayer HA (2008) Prototropic rearrangements in cycloheptatrienyl PCP pincer iridium complexes. Dalton Trans:527–532. https://doi.org/10. 1039/b712712j 48. Gelman D, Romm R (2013) PC(sp3)P transition metal pincer complexes: properties and catalytic applications. Top Organomet Chem 40:289–318. https://doi.org/10.1007/978-3642-31081-2_10 49. Moore CM, Szymczak NK (2014) Appended functionality in pincer ligands. Wiley, pp 71–94. https://doi.org/10.1002/9783527681303.ch3 50. McDonald AR, Dijkstra HP (2016) Tethered pincer complexes as recyclable homogeneous catalysts. Top Organomet Chem 54:335–369. https://doi.org/10.1007/3418_2015_124 51. Hale LVA, Szymczak NK (2018) Hydrogen transfer catalysis beyond the primary coordination sphere. ACS Catal 8(7):6446–6461. https://doi.org/10.1021/acscatal.7b04216 52. Balakrishna MS (2018) Unusual and rare pincer ligands: synthesis, metallation, reactivity and catalytic studies. Polyhedron 143:2–10. https://doi.org/10.1016/j.poly.2017.07.026 53. Sawamura M, Ito Y (1992) Catalytic asymmetric synthesis by means of secondary interaction between chiral ligands and substrates. Chem Rev 92(5):857–871. https://doi.org/10.1021/ cr00013a005 54. McLain SJ (1983) Organometallic crown ethers. 1. Metal-acyl binding to a crown ether-held cation. J Am Chem Soc 105(20):6355–6357. https://doi.org/10.1021/ja00358a051 55. Komarov IV, Gorichko MV, Kornilov MY (1997) Synthesis of chiral functionalized phosphine ligands based on camphor skeleton. Tetrahedron Asymmetry 8(3):435–445. https://doi. org/10.1016/S0957-4166(97)00003-7 56. Landis CR, Sawyer RA, Somsook E (2000) Synthesis and characterization of a chiral, Aza-15Crown-5-functionalized ferrocenyldiphosphine ligand for asymmetric catalysis. Organometallics 19(6):994–1002. https://doi.org/10.1021/om990391u 57. Malkov AV, Mariani A, MacDougall KN, Kocovsky P (2004) Role of noncovalent interactions in the Enantioselective reduction of aromatic Ketimines with Trichlorosilane. Org Lett 6 (13):2253–2256. https://doi.org/10.1021/ol049213+ 58. Albano VG, Bandini M, Melucci M, Monari M, Piccinelli F, Tommasi S, Umani-Ronchi A (2005) Novel chiral diamino-oligothiophenes as valuable ligands in Pd-catalyzed allylic alkylations. On the "primary" role of "secondary" interactions in asymmetric catalysis. Adv Synth Catal 347(11–13):1507–1512. https://doi.org/10.1002/adsc.200505109 59. Gibson SE, Rudd M (2007) The role of secondary interactions in the asymmetric palladiumcatalyzed hydrosilylation of olefins with monophosphine ligands. Adv Synth Catal 349 (6):781–795. https://doi.org/10.1002/adsc.200600458 60. Andrushko V, Andrushko N, Boerner A (2008) P-ligands with additional functional groups concepts of secondary interactions and hemilability. Wiley, pp 886–914 61. Das S, Incarvito CD, Crabtree RH, Brudvig GW (2006) Molecular recognition in the selective oxygenation of saturated C-H bonds by a dimanganese catalyst. Science 312 (5782):1941–1943. https://doi.org/10.1126/science.1127899 62. Gosiewska S, Herreras Martinez S, Lutz M, Spek AL, van Koten G, Klein Gebbink RJM (2006) Diastereopure cationic NCN-pincer palladium complexes with square planar η4-N,C,N, O coordination. Eur J Inorg Chem 2006(22):4600–4607. https://doi.org/10.1002/ejic. 200600508 63. Gosiewska S, Huis M, de Pater JJ, Bruijnincx PC, Lutz M, Spek AL, van Koten G, Gebbink RJK (2006) Novel enantiopure non-C2-symmetric NCN-pincer palladium complexes with

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

131

L-proline chiral auxiliaries: mer η3-N,C,N versus square planar η4-N,C,N,O coordination. Tetrahedron Asymmetry 17:674–686. https://doi.org/10.1016/j.tetasy.2005.12.040 64. Gosiewska S, Martinez Herreras S, Lutz M, Spek AL, Havenith RWA, van Klink GPM, van Koten G, Klein Gebbink RJM (2008) Synthesis, structure, and catalytic performance of diastereopure five-coordinated NCN-pincer palladium(II) complexes bearing bulky amino acid substituents. Organometallics 27(11):2549–2559. https://doi.org/10.1021/om701261q 65. Holz J, Sturmer R, Schmidt U, Drexler H-J, Heller D, Krimmer H-P, Borner A (2001) Synthesis of chiral 2,5-bis(oxymethyl)-functionalized bis(phospholanes) and their application in Rh- and Ru-catalyzed enantioselective hydrogenations. Eur J Org Chem 24:4615–4624. https://doi.org/10.1002/1099-0690(200112)2001:243.0.CO;2-N 66. Borner A (2001) The effect of internal hydroxy groups in chiral diphosphane rhodium(I) catalysts on the asymmetric hydrogenation of functionalized olefins. Eur J Inorg Chem 2:327–337. https://doi.org/10.1002/1099-0682(200102)2001:23.0. CO;2-A 67. Jia X, Zhang L, Qin C, Leng X, Huang Z (2014) Iridium complexes of new NCP pincer ligands: catalytic alkane dehydrogenation and alkene isomerization. Chem Commun 50 (75):11056–11059. https://doi.org/10.1039/C3CC46851H 68. Wang Y, Qin C, Jia X, Leng X, Huang Z (2017) An Agostic iridium pincer complex as a highly efficient and selective catalyst for monoisomerization of 1-alkenes to trans-2-alkenes. Angew Chem Int Ed 56(6):1614–1618. https://doi.org/10.1002/anie.201611007 69. Kita MR, Miller AJM (2014) Cation-modulated reactivity of iridium hydride pincer-crown ether complexes. J Am Chem Soc 136(41):14519–14529. https://doi.org/10.1021/ja507324s 70. Smith JB, Kerr SH, White PS, Miller AJM (2017) Thermodynamic studies of cationmacrocycle interactions in nickel pincer-crown ether complexes enable switchable ligation. Organometallics 36(16):3094–3103. https://doi.org/10.1021/acs.organomet.7b00431 71. Camp AM, Kita MR, Grajeda J, White PS, Dickie DA, Miller AJM (2017) Mapping the binding modes of hemilabile pincer-crown ether ligands in solution using diamagnetic anisotropic effects on NMR chemical shift. Inorg Chem 56(18):11141–11150. https://doi.org/10. 1021/acs.inorgchem.7b01485 72. Grajeda J, Kita MR, Gregor LC, White PS, Miller AJM (2016) Diverse cation-promoted reactivity of iridium carbonyl pincer-crown ether complexes. Organometallics 35(3):306–316. https://doi.org/10.1021/acs.organomet.5b00786 73. Kita MR, Miller AJM (2017) An ion-responsive pincer-crown ether catalyst system for rapid and switchable olefin isomerization. Angew Chem Int Ed 56(20):5498–5502. https://doi.org/ 10.1002/anie.201701006 74. Lesueur W, Solari E, Floriani C, Chiesi-Villa A, Rizzoli C (1997) A bidentate bisphosphine functioning in intramolecular aliphatic metalation and as an NMR spectroscopic probe for the metal coordination environment. Inorg Chem 36:3354–3362. https://doi.org/10.1021/ ic9700819 75. Weng W, Parkin S, Ozerov OV (2006) Double C-H activation results in ruthenium complexes of a neutral PCP ligand with a central carbene moiety. Organometallics 25:5345–5354. https:// doi.org/10.1021/om060464r 76. Gelman D, Musa S (2012) Coordination versatility of sp3-hybridized pincer ligands toward ligand-metal cooperative catalysis. ACS Catal 2:2456–2466. https://doi.org/10.1021/ cs3005083 77. Cohen S, Borin V, Schapiro I, Musa S, De-Botton S, Belkova NV, Gelman D (2017) Ir(III)PC(sp3)P bifunctional catalysts for production of H2 by dehydrogenation of formic acid: experimental and theoretical study. ACS Catal 7(12):8139–8146. https://doi.org/10.1021/ acscatal.7b02482 78. Musa S, Shaposhnikov I, Cohen S, Gelman D (2011) Ligand-metal cooperation in PCP pincer complexes: rational design and catalytic activity in acceptorless dehydrogenation of alcohols. Angew Chem Int Ed 50:3533–3537. https://doi.org/10.1002/anie.201007367

132

A. Singh et al.

79. De-Botton S, Romm R, Bensoussan G, Hitrik M, Musa S, Gelman D (2016) Coordination versatility of p-hydroquinone-functionalized dibenzobarrelene-based PC(sp3)P pincer ligands. Dalton Trans 45(40):16040–16046. https://doi.org/10.1039/C6DT02201D 80. De-Botton S, Cohen S, Gelman D (2018) Iridium PC(sp3)P pincer complexes with Hemilabile pendant arms: synthesis, characterization, and catalytic activity. Organometallics 37 (8):1324–1330. https://doi.org/10.1021/acs.organomet.8b00105 81. Musa S, Shpruhman A, Gelman D (2012) New PC(sp3)P pincer complexes of platinum and palladium. J Organomet Chem 699:92–95. https://doi.org/10.1016/j.jorganchem.2011.11.007 82. Musa S, Ghosh A, Vaccaro L, Ackermann L, Gelman D (2015) Efficient E-selective transfer semihydrogenation of alkynes by means of ligand-metal cooperating ruthenium catalyst. Adv Synth Catal 357(10):2351–2357. https://doi.org/10.1002/adsc.201500372 83. Musa S, Fronton S, Vaccaro L, Gelman D (2013) Bifunctional ruthenium(II) PCP pincer complexes and their catalytic activity in acceptorless dehydrogenative reactions. Organometallics 32(10):3069–3073. https://doi.org/10.1021/om400285r 84. Musa S, Filippov OA, Belkova NV, Shubina ES, Silantyev GA, Ackermann L, Gelman D (2013) Ligand-metal cooperating PC(sp3)P pincer complexes as catalysts in olefin hydroformylation. Chem Eur J 19(50):16906–16909. https://doi.org/10.1002/chem. 201303311 85. Musa S, Ackermann L, Gelman D (2013) Dehydrogenative cross-coupling of primary and secondary alcohols. Adv Synth Catal 355(14–15):3077–3080. https://doi.org/10.1002/adsc. 201300656 86. Bertini F, Glatz M, Stoger B, Peruzzini M, Veiros LF, Kirchner K, Gonsalvi L (2019) Carbon dioxide reduction to methanol catalyzed by Mn(I) PNP pincer complexes under mild reaction conditions. ACS Catal 9(1):632–639. https://doi.org/10.1021/acscatal.8b04106 87. Bernskoetter WH, Hazari N (2017) Reversible hydrogenation of carbon dioxide to formic acid and methanol: Lewis acid enhancement of base metal catalysts. Acc Chem Res 50 (4):1049–1058. https://doi.org/10.1021/acs.accounts.7b00039 88. Celaje JJA, Lu Z, Kedzie EA, Terrile NJ, Lo JN, Williams TJ (2016) A prolific catalyst for dehydrogenation of neat formic acid. Nat Commun 7:11308. https://doi.org/10.1038/ ncomms11308 89. Boddien A, Mellmann D, Gaertner F, Jackstell R, Junge H, Dyson PJ, Laurenczy G, Ludwig R, Beller M (2011) Efficient dehydrogenation of formic acid using an Iron catalyst. Science 333(6050):1733–1736. https://doi.org/10.1126/science.1206613 90. Fukuzumi S, Yamada Y, Suenobu T, Ohkubo K, Kotani H (2011) Catalytic mechanisms of hydrogen evolution with homogeneous and heterogeneous catalysts. Energy Environ Sci 4 (8):2754–2766. https://doi.org/10.1039/c1ee01551f 91. Luo Q, Beller M, Jiao H (2013) Formic acid dehydrogenation on surfaces - a review of computational aspects. J Theor Comput Chem 12(7):1330001/1330001–1330001/1330028. https://doi.org/10.1142/s0219633613300012 92. Yang X (2013) Mechanistic insights into iron catalyzed dehydrogenation of formic acid: β-hydride elimination vs. direct hydride transfer. Dalton Trans 42:11987–11991. https://doi. org/10.1039/c3dt50908g 93. Jongbloed LS, de Bruin B, Reek JNH, Lutz M, van der Vlugt JI (2016) Reversible cyclometalation at RhI as a motif for metal-ligand bifunctional bond activation and base-free formic acid dehydrogenation. Cat Sci Technol 6(5):1320–1327. https://doi.org/10.1039/ C5CY01505G 94. Tutusaus O, Ni C, Szymczak NK (2013) A transition metal Lewis acid/base triad system for cooperative substrate binding. J Am Chem Soc 135(9):3403–3406. https://doi.org/10.1021/ ja400962h 95. Yoshinari A, Tazawa A, Kuwata S, Ikariya T (2012) Synthesis, structures, and Reactivities of pincer-type ruthenium complexes bearing two proton-responsive Pyrazole arms. Chem Asian J 7(6):1417–1425. https://doi.org/10.1002/asia.201200014

Cooperative Reactivity by Pincer-Type Complexes Possessing Secondary. . .

133

96. Kiernicki JJ, Zeller M, Szymczak NK (2017) Hydrazine capture and N-N bond cleavage at Iron enabled by flexible appended Lewis acids. J Am Chem Soc 139(50):18194–18197. https://doi.org/10.1021/jacs.7b11465 97. Kiernicki JJ, Zeller M, Szymczak NK (2019) Requirements for Lewis acid-mediated capture and N-N bond cleavage of hydrazine at Iron. Inorg Chem 58(2):1147–1154. https://doi.org/10. 1021/acs.inorgchem.8b02433 98. Geri JB, Szymczak NK (2015) A proton-switchable bifunctional ruthenium complex that catalyzes nitrile hydroboration. J Am Chem Soc 137(40):12808–12814. https://doi.org/10. 1021/jacs.5b08406 99. Fujita K-I, Tanino N, Yamaguchi R (2007) Ligand-promoted dehydrogenation of alcohols catalyzed by CpIr complexes. A new catalytic system for oxidant-free oxidation of alcohols. Org Lett 9(1):109–111. https://doi.org/10.1021/ol062806v 100. Kawahara R, Fujita K-I, Yamaguchi R (2012) Cooperative catalysis by iridium complexes with a bipyridonate ligand: versatile dehydrogenative oxidation of alcohols and reversible dehydrogenation-hydrogenation between 2-propanol and acetone. Angew Chem Int Ed 51 (51):12790–12794. https://doi.org/10.1002/anie.201206987 101. Saha B, Wahidur Rahaman SM, Daw P, Sengupta G, Bera JK (2014) Metal–ligand cooperation on a diruthenium platform: selective imine formation through acceptorless dehydrogenative coupling of alcohols with amines. Chem Eur J 20(21):6542–6551. https:// doi.org/10.1002/chem.201304403 102. Tseng K-NT, Kampf JW, Szymczak NK (2016) Modular attachment of appended boron Lewis acids to a ruthenium pincer catalyst: metal-ligand cooperativity enables selective alkyne hydrogenation. J Am Chem Soc 138(33):10378–10381. https://doi.org/10.1021/jacs.6b03972 103. Zell T, Milstein D (2015) Hydrogenation and dehydrogenation Iron pincer catalysts capable of metal-ligand cooperation by aromatization/dearomatization. Acc Chem Res 48(7):1979–1994. https://doi.org/10.1021/acs.accounts.5b00027 104. Smith J, Ariga K, Anslyn EV (1993) Enhanced imidazole-catalyzed RNA cleavage induced by a bis-alkylguanidinium receptor. J Am Chem Soc 115(1):362–364. https://doi.org/10.1021/ ja00054a062 105. Aït-Haddou H, Sumaoka J, Wiskur SL, Folmer-Andersen JF, Anslyn EV (2002) Remarkable cooperativity between a ZnII ion and guanidinium/ammonium groups in the hydrolysis of RNA. Angew Chem Int Ed 41(21):4013–4016. https://doi.org/10.1002/1521-3773(20021104) 41:213.0.CO;2-Y 106. Kendall AJ, Zakharov LN, Gilbertson JD (2010) Synthesis and stabilization of a monomeric Iron(II) hydroxo complex via intramolecular hydrogen bonding in the secondary coordination sphere. Inorg Chem 49(19):8656–8658. https://doi.org/10.1021/ic101408e 107. Delgado M, Sommer SK, Swanson SP, Berger RF, Seda T, Zakharov LN, Gilbertson JD (2015) Probing the protonation state and the redox-active sites of pendant base iron(II) and zinc(II) pyridinediimine complexes. Inorg Chem 54(15):7239–7248. https://doi.org/10.1021/ acs.inorgchem.5b00633 108. Delgado M, Ziegler JM, Seda T, Zakharov LN, Gilbertson JD (2016) Pyridinediimine Iron complexes with pendant redox-inactive metals located in the secondary coordination sphere. Inorg Chem 55(2):555–557. https://doi.org/10.1021/acs.inorgchem.5b02544 109. Hartle MD, Delgado M, Gilbertson JD, Pluth MD (2016) Stabilization of a Zn(II) hydrosulfido complex utilizing a hydrogen-bond accepting ligand. Chem Commun 52(49):7680–7682. https://doi.org/10.1039/C6CC01373B 110. Kwon YM, Delgado M, Zakharov LN, Seda T, Gilbertson JD (2016) Nitrite reduction by a pyridinediimine complex with a proton-responsive secondary coordination sphere. Chem Commun 52(73):11016–11019. https://doi.org/10.1039/C6CC05962G 111. Delgado M, Gilbertson JD (2017) Ligand-based reduction of nitrate to nitric oxide utilizing a proton-responsive secondary coordination sphere. Chem Commun 53(81):11249–11252. https://doi.org/10.1039/C7CC06541H

134

A. Singh et al.

112. Burns KT, Marks WR, Cheung PM, Seda T, Zakharov LN, Gilbertson JD (2018) Uncoupled redox-inactive Lewis acids in the secondary coordination sphere entice ligand-based nitrite reduction. Inorg Chem 57(16):9601–9610. https://doi.org/10.1021/acs.inorgchem.8b00032 113. Cheung PM, Burns KT, Kwon YM, Deshaye MY, Aguayo KJ, Oswald VF, Seda T, Zakharov LN, Kowalczyk T, Gilbertson JD (2018) Hemilabile proton relays and redox activity lead to {FeNO}x and significant rate enhancements in NO2- reduction. J Am Chem Soc 140 (49):17040–17050. https://doi.org/10.1021/jacs.8b08520 114. Moore CM, Szymczak NK (2013) 6,60 -Dihydroxy terpyridine: a proton-responsive bifunctional ligand and its application in catalytic transfer hydrogenation of ketones. Chem Commun 49(4):400–402. https://doi.org/10.1039/C2CC36927C 115. Moore CM, Bark B, Szymczak NK (2016) Simple ligand modifications with pendent OH groups dramatically impact the activity and selectivity of ruthenium catalysts for transfer hydrogenation: the importance of alkali metals. ACS Catal 6(3):1981–1990. https://doi.org/10. 1021/acscatal.6b00229 116. Gade LH (2000) Highly polar metal–metal bonds in “early–late” heterodimetallic complexes. Angew Chem Int Ed 39(15):2658–2678. https://doi.org/10.1002/1521-3773(20000804) 39:153.0.CO;2-C 117. Huang D, Holm RH (2010) Reactions of the terminal NiII-OH group in substitution and electrophilic reactions with carbon dioxide and other substrates: structural definition of binding modes in an intramolecular NiIIFeII bridged site. J Am Chem Soc 132 (13):4693–4701. https://doi.org/10.1021/ja1003125 118. Zhang X, Huang D, Chen Y-S, Holm RH (2012) Synthesis of binucleating macrocycles and their nickel(II) hydroxo- and cyano-bridged complexes with divalent ions: anatomical variation of ligand features. Inorg Chem 51(20):11017–11029. https://doi.org/10.1021/ic301506x 119. Dobbek H, Svetlitchnyi V, Gremer L, Huber R, Meyer O (2001) Crystal structure of a carbon monoxide dehydrogenase reveals a [Ni-4Fe-5S] cluster. Science 293(5533):1281–1285. https://doi.org/10.1126/science.1061500 120. Can M, Armstrong FA, Ragsdale SW (2014) Structure, function, and mechanism of the nickel metalloenzymes, CO dehydrogenase, and acetyl-CoA synthase. Chem Rev 114 (8):4149–4174. https://doi.org/10.1021/cr400461p 121. Halvagar MR, Tolman WB (2013) Isolation of a 2-Hydroxytetrahydrofuran complex from copper-promoted hydroxylation of THF. Inorg Chem 52(15):8306–8308. https://doi.org/10. 1021/ic401446s 122. Haack P, Limberg C (2014) Molecular CuII-O-CuII complexes: still waters run deep. Angew Chem Int Ed 53(17):4282–4293. https://doi.org/10.1002/anie.201309505 123. Halvagar MR, Neisen B, Tolman WB (2013) Copper-, palladium-, and platinum-containing complexes of an asymmetric dinucleating ligand. Inorg Chem 52(2):793–799. https://doi.org/ 10.1021/ic301914u

Top Organomet Chem (2021) 68: 135–180 https://doi.org/10.1007/3418_2020_68 # Springer Nature Switzerland AG 2020 Published online: 29 October 2020

Redox-Active Pincer Ligands Jarl Ivar van der Vlugt

Contents 1 General Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Pincer Ligands That Undergo Reductive Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Bis(imino)pyridine as Redox-Active Pincer Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 2,20 ,200 -Terpyridine and Substituted Derivatives Thereof . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Other Bis-Azadiene-Based Pincer Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Miscellaneous Pincer Systems Amenable to Ligand-Centred Reduction . . . . . . . . . . . . 3 Pincer Ligands That Undergo Oxidative Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Trianionic Pincer Ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Dianionic Pincer Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Monoanionic Pincer Ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Neutral Pincer Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

136 138 138 145 148 153 155 155 161 166 171 172 173

Abstract This review aims to provide a comprehensive overview of the emerging field of redox-active pincer ligands. As such, following a short general introduction on ligand-centred redox activity, the recent literature is discussed in two separate sections. The first deals with ligand platforms that predominantly display reductive chemistry from the ‘parent’ scaffold; the second focusses more on systems that readily undergo oxidative chemistry. Fundamental stoichiometric reactivity as well as, where explored, catalytic applications will be discussed. This combined survey hopefully inspires the exploration and uncovering of many new avenues within the realm of redox-active pincer chemistry. Keywords Homogeneous catalysis · Pincer ligand · Redox-active ligand · Singleelectron transfer · Transition metals

J. I. van der Vlugt (*) Bioinspired Coordination Chemistry and Catalysis Group, Institute of Chemistry Carl von Ossietzky University Oldenburg, Oldenburg, Germany e-mail: [email protected]

136

J. I. van der Vlugt

1 General Introduction Traditionally, when considering the design of a metal complex for a specific desired (stoichiometric or catalytic) transformation, the metal centre (with an oxidation state commonly ranging from 0 to VI for 1st and 2nd row mid-to-late transition metals) is the locus for chemical bond activation and follow-up reactivity, with one or more ligands bound to this metal centre acting as spectators that stabilize, tune and/or sterically restrict the coordination sphere around it in order to invoke selective substrate coordination, activation and transformation. Redox-chemistry in these complexes is typically confined to the metal centre, because the energy input required to induce the transfer of an electron to (reduction) or from (oxidation) a spectator, or ‘redox-inert’, ligand is much higher than the cost associated with changing the oxidation state of the metal. Often, the ability of a metal to change its oxidation state whilst mediating a chemical conversion is essential for catalytic processes. Generally speaking, catalytic processes can be distinguished in reductive, oxidative or redox-neutral processes. Many (industrially relevant) reactions, even when formally redox-neutral, involve two-electron redox steps and are thus linked to, e.g. 2nd and 3rd row (noble) transition metals. Base metals generally prefer to undergo one-electron redox events, but controlled (‘metal-mediated’) odd-electron pathways are still less commonly encountered, although there is a great interest in the use of abundant, cheap and non-toxic materials. Moreover, the development of unprecedented types of reactivity – which could facilitate or significantly shorten the synthesis of highly desired molecules – is also sought after. Certain types of ligands have shown to be able to do both by working in synergy with the metal and thereby expanding upon a metal’s ‘common’ reactivity. Apart from ligand platforms that actively participate in bond breaking or making processes (coined reactive ligands or metal-ligand cooperative systems) [1–5], a second class of ligands can be defined that show reversible redox-chemistry. Organic ligands that feature energetically well-accessible, low-lying π-donor (relevant for ‘oxidation’) or π*-acceptor (key during ‘reduction’) orbitals can undergo reversible shuttling between at least two well-defined redox states whilst being bound to a transition metal or main group element (and commonly also in free form) [6–12]. Conjugation generally strengthens the overall extend of this phenomenon, but this is not necessarily required. When this redox-shuttling takes place without changes to the oxidation state of the metal(loid), the ligand is termed redox-active (Fig. 1). In cases where the redox event leads to a more diffuse and ambiguous overall electronic structure, due to strong electronic coupling between a ligand and a metal centre, the term redox noninnocence is generally deemed more appropriate [13, 14], although relatively extensive crossover usage of the terms in the literature is encountered. Several types of metalloenzymes exploit ligand-centred redox activity as a means to break down multi-electron reactions into several single-electron steps, thereby avoiding high energetic penalties (large overpotentials) and enabling the overall transformation to occur near thermodynamic potential.

Redox-Active Pincer Ligands

137

Fig. 1 Redox-active vs. redox-noninnocent ligand reactivity

This overview will primarily be focussed on the former category, i.e. the redoxactive ligands. Within the realm of pincer chemistry, several archetypical designs have surfaced that qualify as redox-active pincer systems. In this context, it is of importance to separate those systems where the main skeletal backbone of the ligand (i.e. part of the pincer framework itself) is redox-active vs. cases where substituents on side arm or flanking groups bear the locus of the redox activity (e.g. a dangling ferrocene unit). Although the latter are of interest in their own right, only designs featuring redox-active pincer backbones will be covered herein. In contrast to, e.g. ferrocene, which is accessible only in a neutral and mono-oxidized form (two-state system), many of these redox-active organic systems actually offer (at least) three different oxidation states, generically referred to (at least in this overview) as the two-electron reduced (dianionic), one-electron reduced/oxidized (monoanionic ligand radical) and two-electron oxidized form (neutral). The following sections will provide a comprehensive (but certainly not exhaustive) overview of what can be considered the main redox-active pincer ligand classes to be encountered in the literature. Whenever deemed relevant, illustrative experimental data related to stoichiometric reactivity and/or catalytic activity are included to demonstrate the potential of the ligand type. This chapter is divided into two main sections, with the first section describing systems that can generally be considered to undergo or display mainly, reductive ‘ligand centered’ chemistry, i.e. the organic fragment undergoing one (or more) redox-change(s) from the parent neutral form into (a) more anionically charged derivative(s) by either electrochemical or chemical reduction, and the second section describing systems that can be considered reactive toward ‚oxidation’ from a parent charged (mono-, di- or trianionic) state. This chapter is meant as an introduction and global overview of the various (sub)classes of redoxactive pincers and their chemistry and relevant stoichiometric or catalytic reactivity. The main observations in support of the postulated electronic structure are provided, but concrete experimental data are only sparingly included. Hence, for complete description and detailed information, the reader is referred to the original contributions.

138

J. I. van der Vlugt

2 Pincer Ligands That Undergo Reductive Chemistry There are various bis(1,4-diazadiene) platforms that exhibit fascinating ligandcentred reduction chemistry [15, 16]. This section is dedicated to several illustrious examples from within this (still expanding) subclass of redox-active pincer systems.

2.1

Bis(imino)pyridine as Redox-Active Pincer Platform

Undoubtedly the most well-established within this category, bis(imino)pyridines (generally denoted as pdi for pyridine-diimine) were first demonstrated to be redox-active in 2000, although basic coordination chemistry has been developed since the 1970s [17] and interesting catalytic applications emerged in the 1990s [18– 21]. When not bound to a (transition) metal, pdi ligands typically show reversible redox-chemistry at very highly cathodic potentials below 2.5 V (vs. Fc/Fc+) (Fig. 2). In seminal work [22], Wieghardt and co-workers detailed the reductive chemistry of a series of homoleptic [M(pdi)2]2+ complexes (M ¼ first row metal), wherein each pdi fragment can store one unpaired electron (as determined using controlled potential coulometry), although the electrochemical reduction occurs in a stepwise fashion, i.e. with the intermediacy of a ligand-mixed valent [M(pdi•) (pdi)]+ species and, at more negative potential, formation of neutral diradical [M (pdi•)2] species. As a prototypical example of pure ligand-centred redox, the homoleptic complex Zn(pdi)2 undergoes two ligand-centred reductions at 1.39 and 1.67 V vs. Fc/Fc+ in acetonitrile. The ligand-centred redox was verified by combining X-ray crystallography, NMR, UV-vis and EPR spectroscopy, UV-vis spectroelectrochemistry, SQUID and solution state magnetochemistry as well as Mössbauer spectroscopy (for Fe congener only). This was quickly followed up by a theoretical study by Budzelaar and co-workers, demonstrating that a biradical description, in which ligand radical anions are antiferromagnetically coupled to the metal centre, is significant in most cases [23]. The typical potential window for these pdi reductions is between 1 and 2 V vs. Fc/Fc+. However, care must be taken to account for metal-centred redoxchemistry as well, as demonstrated for the Co-congener, which displays metalcentred reduction for the homoleptic Co(pdi)22+ system, and only one ligand-centred reduction within the solvent window of acetonitrile near 2 V. Herbert and co-workers very recently extended the family of Fe-based homoleptic pdi series Fig. 2 General redoxchemistry demonstrated by bis(imino)pyridine (pdi) pincer ligands

Redox-Active Pincer Ligands

139

Fig. 3 General strategies to modify the redox-chemistry of pdi frameworks

with different substituents in the para-position of the imino-flanked aryl groups, each derivative showing two reversible 1ereductions between1.0 and2.0 V, with ~0.3 V differences between para-F and para-CN [24]. In general, three strategies are well-developed to vary and tune the redox properties and chemical reactivity of the pdi platform (Fig. 3), involving substitution at (1) the 4-position of the central pyridine core, (2) the imine carbon atom and (3) the imine N atom. Modifications of the para-position of the pyridine ring have been shown to dramatically alter the reduction potential by up to 0.5 V [25]. Substitution of this position may also negate (ir)reversible radical-type C–C bond formation occurring on the backbone [26]. Changes at the imine carbon have substantially less influence on the overall redox properties (~0.1 V difference between methyl and phenyl, with the latter more easily reduced), although manipulation of this position (e.g. by introduction of a phenyl group) may induce undesired deactivation under catalytically relevant conditions via arene coordination [27]. Variation of the imine nitrogen substituents from aryl to alkyl influences the electron density of the scaffold, which could impact the reduction potential, leading to cathodic shifts of up to 200 mV [28]. The corresponding aldimine derivative (i.e. with a hydrogen on the imino carbon), which is far less commonly applied, tends to show more metal-centred radical density relative to the ketimine derivatives, more follow-up reactivity (e.g. disproportionation, hemilability) and also less straightforward formation of, e.g. dinitrogen adducts under chemically reducing conditions (e.g. reduction of metal dihalide precursors with Na/Hg under an N2 atmosphere), although reactivity is dependent on the metal (mainly Fe and Ni examined with bis(aldimino)pyridine) [29–32]. Reversible C–C bond formation on the β-carbon of a methimine-based pdi fragment after deprotonation at this position with a strong base and under an N2 atmosphere was reported for the complex CoIICl(pdi•) [32]. Related chemistry has also been reported for Mn-based systems [33]. Recent studies by Budzelaar have revealed that alkylation of pdi on the imino carbon atoms from ZrBn4, to generate a dianionic diamido pincer bound to ZrIVBn2, likely is mediated by the intrinsic radical character of the pdi backbone [34]. The Gilbertson group recently decorated one of the imino-N atoms with a crown ether moiety to introduce redox-inactive main group cations within the second coordination sphere of a pdi-bound transition metal [36]. This also enabled finetuning of the redox-chemistry exhibited by the ligand, although the shifts were

140

J. I. van der Vlugt

Fig. 4 Reductive chemistry of the complex FeCl2(pdi)

modest. The corresponding Na+ encapsulated bimetallic species showed enhanced activity (based on an initial rate analysis) for stoichiometric nitrite reduction, with NO2 being the anion associated with the sodium cation [37]. Base metals have dominated the coordination chemistry of this pdi ligand type, which likely relates to the weak field nature of the donor atoms involved, inducing intermediate- or high-spin states that are easily accommodated by these metals [38] as well as to the emergence of nonnoble metal chemistry for catalysis. Hence, the high-lying, filled orbitals of the reduced pdi ligand can act as electron reservoir, providing access to reducing equivalents required for overall two-substrate activation. The Lewis acidity of the metal is also tuned by ligand-centred redox and the concomitant spin state changes induced thereby, which enhances substrate binding. Heteroleptic systems are of more relevance for follow-up substrate coordination and activation. Much focus has been given to the elucidation of the reaction chemistry and electronic structure of, e.g. dihalide, (di)alkyl and (bis)dinitrogen adducts, in particular with Fe, Co and Ni. One-electron reduction of FeCl2(pdi) results in the formation of paramagnetic high-spin FeII bearing a ligand-centred radical that couples antiferromagnetically with one of the four Fe-centred unpaired electrons to give an overall S ¼ 3/2 spin state (Fig. 4) [39]. Substitution of halide for alkyl does not change this electronic structure picture. Subsequent reduction with, e.g. sodium under an atmosphere of N2 leads to a second ligand reduction event to generate a triplet ligand diradical dianion, concomitant with a spin state change at iron to intermediate-spin FeII (as a result of the overall stronger ligand field induced by reduction), which leads to the creation of partially filled metal-centred orbitals that could allow for substrate coordination and activation. This reduced FeII(pdi2•) (N2) species is the dominant species in solution, but under judicious conditions, the five-coordinate bis(dinitrogen) adduct Fe(pdi)(N2)2 can also be isolated [40]. Strikingly, this derivative is characterized as showing redox noninnocence, with highly covalent metal-pdi binding similar to that observed in Fe(CO)2(pdi). Hence, for these latter two compounds, the oxidation state of the iron is best described as a resonance hybrid between Fe0 and FeII. The same ambiguous situation also exists for dinitrogen-bridged dimeric [{Fe(N2)(pdi)}2(μ-N2)] complexes that can be isolated when the steric bulk at the imino aryl rings is reduced [35]. Hence, in these cases the bis(imino)pyridine ligand is best perceived as a strong π-acceptor ligand.

Redox-Active Pincer Ligands

141

Fig. 5 C–C bond activation via combined two-electron transfer shared by metal and pdi ligand

Fig. 6 Formation of an intermediate-spin FeIII metallacyclic intermediate by cycloaddition of a diyne substrate

The mono-nitrogen adduct FeII(pdi2•)(N2) reacts with the strained C–C bond of biphenylene, resulting in overall oxidative addition ‘at iron’ to give a ferracyclic derivative (Fig. 5) [41]. The combined spectroscopic, magnetic and computational data support that the two-electron oxidation of this intermediate-spin FeII dinitrogen complex with a ligand diradical generates an FeIII centre with a pdi ligand radical. Hence, cooperative redox reactivity by both the metal and the ligand, each being oxidized by one electron, appears to be operative. Although typically one-electron reduction of each of the imine side arms is considered for most chemistry, further ligand-centred reduction is possible to create a trianionic monoradical platform with one 1,4-ethylenediamido binding pocket and a flanking carbon-centred iminyl radical [42–45]. The group of Chirik explored the potential role of the ligand-centred redoxchemistry in the context of catalysis, primarily for C–C bond formation [46, 47]. By integrating the bis(imino)pyridine framework in a tricyclic tetrahydroacridyl backbone allowed for convincing evidence for the involvement of both the Fe metal and ligand-centred redox was provided to explain the oxidative cyclization of an internal diyne or α,ω-diene substrate to provide a metallacyclic intermediate (Fig. 6) [48]. Follow-up work revealed two-electron donation from the pdi-derived ligand to be operative in the intermolecular cross cycloadditions of alkenes and dienes (Fig. 7) [49]. However, the intermolecular [2+2] cycloaddition of substituted alkenes

142

J. I. van der Vlugt

Fig. 7 Two-electron donation from the pdi ligand allows for C-C crosscoupling between an alkene and a diene

followed an Fe-only redox pathway, as suggested from extensive mechanistic studies published recently [50]. Upon reduction of the corresponding heteroleptic high-spin cobalt-dichloride complex [Co(Cl)2(pdi)], the spin state of the formed species strongly depends on the substitution pattern at the imine nitrogens. Aryl substituents lead to the low-spin CoII antiferromagnetically coupled to a ligand radical, giving the overall S ¼ 0 square planar d7 complex CoIICl(pdi•). Imino-alkyl substituted systems display spin crossover to an overall S ¼ 1 state with high-spin CoII at high temperature [51, 52]. Dialkyl-cobalt(II) species with pdi are unknown, but the mono-alkyl derivatives are easily accessible and are best described as low-spin CoII with a pdicentred radical. The corresponding cationic [CoI(N2)(pdi)]+ starting material has low-spin CoI with a neutral pdi ligand [53]. Single or two-electron reduction which occurs at the ligand, with cobalt remaining in the +I oxidation state, leads to the formation of [CoI(N2)(pdi•)] and [CoI(N2)(pdi2•)], the latter containing a dianionic diradical ligand fragment. The neutral congener has been reported to undergo binuclear oxidative addition with aryl and alkyl halides, involving parapyridine substitution [54]. For related Co catalysis on C–C bond formation as described for Fe (vide supra), the redox-active ligand radical does not seem to actively participate in the redox-chemistry but does tune the reactivity of the Co centre [55]. The group of Roşca recently developed bis(imino)pyrazine-based analogs of pdi and its coordination to Fe0 precursor Fe(bda)(CO)3 (bda ¼ benzylideneacetone) [56]. The resulting diamagnetic complex predominantly showed spectroscopic features that support an Fe0 oxidation state, which was also supported by DFT calculations. Although the electrochemical reduction at 2.28 V was postulated to be ligand-centred, the main chemical pathway for reaction discussed was methylation of the free pyrazine N-donor. The corresponding cationic complex was susceptible to chemical reduction by cobaltocene, which furnished a dinuclear Fe2 species via C–C bond formation at the meta-position of the pyrazine ring. The Tomson group utilized dinucleating bis(imino)pyridines with alkyl linkers to connect imino groups, originally prepared by Drew and Nelson in 1982 [57], which presented a way to create Fe2, Co2 and Ni2 species for inter alia elegant nitride and dinitrogen chemistry [58–61]. The role of ligand reduction on the nature of the metal-metal interaction was studied in detail for the cobalt-cobalt derivative (Fig. 8) [62]. It was convincingly shown that increased levels of pdi2-based reduction lead to

Redox-Active Pincer Ligands

143

Fig. 8 Formation of dinuclear Co2 species with flexible dinucleating pdi-type ligand and effect of ligand oxidation state on cobalt–cobalt bond

elongation and ultimately rupture of the Co-Co present in the ligand-mixed valent parent Co2(μ-Cl)(pdi2•) species, based on both experimental data (X-ray crystallography, electronic absorption spectroscopy) and DFT calculations. Although less pronounced, the same phenomenon occurred with the Ni congener. No direct discussion on the possible involvement or role of the ligand-centred radical character on the displayed chemistry with nitrogen based substrates bound in-between both metal centres is provided. The chemistry of ketimino-derived pdi with nickel is relatively underexplored. Budzelaar and co-workers reported that the square planar complex Ni(CH3)(pdi) most likely contains low-spin NiII bound to a ligand radical anion [52]. The corresponding dinitrogen adduct, obtained upon reduction, is isoelectronic with [CoI(N2)(pdi•)], and accordingly a low-spin NiII with a closed-shell ligand dianion is assumed to be the best description for the electronic structure. Ligand dimerization has been observed as well for this species. Srivastava, Kumar and co-workers applied a six-coordinate NiII-pdi complex for solvent-free N-alkylation and dehydrogenative coupling catalysis but without discussing any ligand effect [63]. Trovitch developed the chemistry of manganese with pdi-based pincer ligands, albeit with two additional PPh2 donor fragments, each connected to one imino arm, to generate a potential pentadentate ligand, and applied these systems in the hydrosilylation of ketones [64]. The high-spin MnCl2-species bearing this pdi_diphosphine ligand can be converted to the derivative that was initially

144

J. I. van der Vlugt

Fig. 9 Oxidative addition of PhN¼NPh on a V(III)pdi-dimer

described as low-spin MnII with a dianionic closed-shell pdi2 ligand, based on EPR spectroscopy by double reduction of the pdi-skeleton. Subsequent mechanistic investigations, including also ester hydrosilylation mediated by the Mn-hydride derivative that was formulated as a doubly reduced closed-shell pdi2– ligand bound to low-spin MnIII (by virtue of the strong ligand field induced by the hydride and both phosphines) and supported by DFT calculations, led to reformulation of the non-hydride species Mn(pdi_diphosphine) responsible for ketone hydrosilylation as an intermediate-spin MnII centre that is antiferromagnetically coupled to a triplet pdi2•- diradical backbone [65]. This assignment also best corresponds to the obtained structural parameters from X-ray crystallography. Chirik reported a paramagnetic and NMR silent dinuclear vanadium-pdi complex with a terminal bridging N2 ligand and a bulky version of pdi (Fig. 9) [66]. Crystallographic evidence obtained for the N–N bond of the bridging fragment together with the metric parameters observed for the pdi ligands support two-electron reduction of the latter to a closed-shell dianionic counterpart, as well as N2 reduction to N22, providing two vanadium(III) centres, as previously also suggested by Gambarotta and Budzelaar for a related species [67]. This species was able to cleave the N¼N bond in azobenzenes to give diimido species VV(NPh)2(pdi•), which is enabled by ligand-to-vanadium single-electron transfer (as well as one electron from the N22 fragment that evolves as molecular nitrogen). Subsequent investigations with the same dinuclear species revealed that the combined reduction equivalents stored in the pdi and N2 ligands enable oxidative N–N bond addition of 1,2-diarylhydrazines to give mononuclear vanadium species with two anilido fragments (Fig. 10) [68]. Crossover experiments using hydrazines with different arene substituents established that both anilido moieties stem from one single hydrazine substrate, supporting direct N-N oxidative addition. Combined experimental and theoretical analyses of the electronic structure of the product supported that the oxidation state at vanadium remains +III during this transformation. Interconversion between the bis(imido) and bis(anilido) species via two sequential hydrogen atom addition steps was also demonstrated. The mixed anilido-imido intermediate contained a VV centre supported by the dianionic pdi2 framework, highlighting the intricate involvement of the redox-active ligand in this

Redox-Active Pincer Ligands

145

Fig. 10 Ligand-mediated oxidative N–N bond addition of 1,2-diarylhydrazines

process. A very related electronic structure was reported for the dinitrogen-bridged Mo2 derivative, but reactivity with ammonia resulted in N–H and C–N bond cleavage for one of the imino fragment, leading to a bridging parent imido species [69]. The Herbert group very recently interrogated both PI and PIII complexes with the same bulky pdi derivative and observed that the low oxidation state derivative shows reversible reduction at 1.38 V vs. Fc/Fc+, likely related to backbone-based redoxchemistry [70]. Berry and co-workers investigated dinuclear Ru complexes with pdi and a bridging dinitrogen ligand [71]. The size of the imino substituents influenced electronic communication between both one-electron reduced pdi ligand radicals and therefore the overall spin state of the system.

2.2

2,20 ,200 -Terpyridine and Substituted Derivatives Thereof

Although omnipresent in research related to photo-, magneto- and supramolecular chemistry as well as in catalysis [72], the exploitation of the redox activity of 2,20 ,200 -terpyridine (tpy)-based systems is strikingly underrepresented. This is particularly noteworthy in connection to the fact that their bidentate 2,20 -bipyridinebased counterparts are well-established and frequently employed for their redox activity, e.g. in electrocatalytic CO2 reduction. Recent studies using tpy as locus for redox activity have emerged, nonetheless. The free tpy molecule undergoes a reversible reduction at 2.55 V vs. Fc/Fc+ to generate the radical anion tpy•. Scanning more negatively results in a second, but irreversible, reduction at 3.05 V [73, 74]. Wieghardt and co-workers have thoroughly investigated the electronic structure of a series of homoleptic M(tpy)2 complexes with varying overall charge. Redox-chemistry almost invariably occurs at the tpy ligand, with the formation of mixed valent [M(tpy•)(tpy)] as well as diradical [M(tpy•)2] species disclosed [75, 76]. Remarkable ‘outlier’ in this regard proved to be the complex Co(tpy)2, where metal-centred redox was observed [77].

146

J. I. van der Vlugt

Fig. 11 Ligand vs. metalcentred reduction within Ni-tpy complexes, depending on nature of co-ligand

More recently, heteroleptic versions were also investigated, particularly for M (Cp2)(tpy) complexes based on early transition metals [78, 79]. The accompanying variations in the different bonds within conjugated 1,4-azadiene fragments can be captured using a structural parameter Δ, which relates C–C and C–N bond lengths (either experimentally determined or computationally assessed) to the level of ligand reduction. Sarkar and co-workers showed that the introduction of electron-donating dimethylamino groups on the periphery of the tpy backbone may induce ligand dissociation during reductive redox-chemistry in homoleptic Ru-bis(tpy) complexes [80]. Dechambenoit very recently reported a ‘fused’ dinucleating bis-terpyridinetype ligand and the resulting magnetochemistry of its bis-CoII-complex [81]. Vicic demonstrated that the choice of co-ligand in square planar NiII(tpy)(X)2 complexes dramatically influences the fate of the ligand (and metal) redox state upon singleelectron reduction, switching between ligand radical-NiII in Ni(tpy•)(Me) to neutral ligand-NiI in Ni(tpy)(Br) (Fig. 11) [82, 83]. This strict distinction is, e.g. relevant in the context of C-C cross-coupling chemistry [84–86]. Chirik reported an Fe(tpy)-bisalkyl complex that was best characterized as highspin FeIII(tpy•)(R)2, with antiferromagnetic coupling occurring between one of the five unpaired electrons residing on iron with the single electron (radical) within the tpy framework [87]. This complex was prepared by ligand exchange from FeII(Py)2(CH2SiMe3)2 with free tpy and thus formally involves metal-to-ligand single-electron transfer to generate the above-described electronic structure, which is quite remarkable given the strongly cathodic reduction potential required to reduce tpy to its radical anion. Rather than a ‘classic’ redox event occurring in, e.g. cyclic voltammetry, this may be considered an extreme case of metal-to-ligand backbonding. The electronic structure was assigned based on a combined Mössbauer, DFT, XRD and magnetochemical analysis of this complex. DFT analysis revealed the broken-symmetry solution (with redox-active tpy) to be 1.9 kcal/ mol lower in energy than the spin-unrestricted solution corresponding to redoxinnocent behaviour of tpy. The latter would imply high-spin FeII, but this assignment was not supported by the observed isomer shift and quadrupole splitting obtained by

Redox-Active Pincer Ligands

147

Mössbauer spectroscopy. The computational analysis also pointed to the central pyridine ring in the tpy framework bearing most of the ligand-based spin density, which also supported the crystallographically determined bond length perturbations within the tpy skeleton. Strong overall covalency between the metal and ligand radical orbitals lends support for strong pi-backbonding. Furthermore, application of these systems in the catalytic hydrosilylation of olefins with tertiary silanes was also reported. Alongside the results obtained by Chirik, Nakazawa and co-workers developed complexes with substituents on both the flanking pyridine rings of tpy for the same catalytic purpose [88]. Strikingly different reactivity was reported for the reaction between the high-spin CoII(CH2SiMe3)2(Py)2 precursor and tpy. Unlike for the Fe case, where both alkyl ligands were preserved, Chirik reported that pyridine ligand displacement at cobalt occurs concomitantly with ejection of an alkyl radical from Co to generate the square planar complex CoII(tpy•)(R), very similar as was observed in case of the Ni-dimethyl precursor before by Vicic (vide supra) [89]. Antiferromagnetic coupling between the ligand-centred and the Co-centred unpaired electron resulted in an overall low-spin configuration for this system. The fate of the alkyl radical nor the possible intermediacy of a CoI(tpy)(CH2SiMe3) species that undergoes metal-toligand single-electron transfer was not further commented on. This complex was also applied in the context of catalytic alkene hydroaddition, showing remarkable Markovnikov selectivity for the hydroboration of styrene, but although the pincer character of the tpy ligand was discussed in the context of the suppressed competitive deactivation pathways or substrate inhibition for diene or arene containing substrates, the potential role of the ligand redox activity was not elaborated. For this particular type of catalytic species, the ligand acting as an ‘electron reservoir’ may tune the spin state/oxidation state at the respective metal centre (iron or cobalt) and thus tune Lewis acidity and reactivity of the metal ion and modify back-electron transfer to acceptor ligands such as alkenes and dienes. Follow-up research by Chirik revealed that with cobalt dichloride or cobalt bis (acetate), the tpy ligand behaves as a redox-innocent pincer platform, giving rise to, e.g. high-spin five-coordinated CoII(OAc)2(Artpy) [90]. This species was shown to undergo significant rearrangement by ligand scrambling upon reaction with pinacolborane, resulting in inter alia six-coordinate Co(Artpy)2. This species, which was also independently synthesized, showed temperature-dependent EPR signatures corresponding to S ¼ ½ (10 K) or S ¼ 3/2 (r.t.), which could relate to ligand redox-chemistry. Not surprisingly, this coordinatively saturated species proved inactive for hydroboration catalysis but likely relates to a deactivation pathway promoted by pinacolborane under catalytic conditions. In the context of dinitrogen activation chemistry with earth-abundant base metals, Chirik reported on the (oxidative and) reductive (electro)chemistry of a dinuclear [{Mo(Phtpy)}2(μ2-N2)] dication featuring a linear bridging N2 fragment [91]. This chemistry was intended to circumvent chemical reactivity observed for related Mo2(μ2-N2) species featuring a bis(imino)pyridine pincer platform (vide infra), where the ligand undergoes irreversible bond cleavage chemistry upon reaction with ammonia (or hydrazine) [69]. Precursor to this dinitrogen-bridged complex

148

J. I. van der Vlugt

Fig. 12 Redox-chemistry of Mo(tpy) complex

was the species MoCl(Phtpy)(PPh2Me)2, which was described as the metal-centred radical MoI based on EPR and magnetic susceptibility data as well as DFT computations. The tpy unit did show structural deviations from the neutral unbound tpy state, indicative of metal-to-ligand π-backbonding. Chemical reduction of this MoI species under N2 atmosphere proved unsuccessful, but chloride abstraction with NaBArF under N2 led to the isolation of diamagnetic [{Mo(Phtpy)}2(μ2-N2)]2+ (Fig. 12). Using cyclic voltammetry, two reversible reduction events (as well as two oxidation events) were observed for this species, and chemical reduction with one or two equivalents led to the singly reduced monoradical species and the doubly reduced diradical derivative, wherein the radical is localized within the tpy fragment(s), most notably in the central pyridine ring(s). Species Y is diamagnetic, proposedly due to antiferromagnetic coupling of both ligand-centred radicals through strong orbital coupling over the π-symmetric tpy-Mo-N-N-Mo-tpy vector. Recently, this Mo dimer was also investigated using ultrafast coherence spectroscopy [92]. Very recently, the relevance of the redox activity of a terpyridine ligand, working in conjunction with a bidentate NHC-containing co-ligand within a Ru complex, to modulate the overpotential (tpy) and kinetics (NHC) during the electroreduction of CO2 was presented [93]. Murugesu, Brusso and co-workers detailed Fe complexes of the tpy analog 5-bis(2-pyridyl)-1,2,4,6-thiatriazinyl radical, which displays a neutral seven π-electron ring system flanked by two pyridyl groups but not formally ligand-only redox-chemistry but rather a diffuse redox-noninnocent interplay between Fe and ligand [94].

2.3

Other Bis-Azadiene-Based Pincer Platforms

The Goswami group reported on the ortho-amination of redox-active bidentate arylazopyridine scaffolds [95] when bound to PdII-dichloride and in the presence of cobaltocene (and an amine), wherein the aryl moiety undergoes functionalization, resulting in the in situ formation of pincer-like platforms with the redox-active azo-fragment now taking centre stage rather than being a flanking functionality. The as-isolated species are ligand-centred radical PdIICl(NNN) complexes that show

Redox-Active Pincer Ligands

149

Fig. 13 Reductive redoxchemistry of a bis (arylazo)pyridine ligand whilst coordinated to Fe

redox potentials of 0.7 and +0.25 V for reduction and oxidation, respectively, of the azo radical [96]. The corresponding ligand radical complexes are intensely coloured, with multiple transitions occurring in the visible spectrum, many involving intra-ligand electron transfer as deduced from TD-DFT calculations. The mechanism for the ligand functionalization proposedly involves initial reduction of the azo-fragment to generate an anionic PdII species wherein the chloride co-ligands are labilized. Exchange for the amine thus enables subsequent intra-ligand electron transfer to create an aminyl radical (following N–H or N–C bond cleavage) in the coordination sphere of PdII. This activates the aryl ring via a radical coupling step, creating a Pd-aryl fragment that subsequently undergoes C–N bond coupling, with electron-redistribution to regenerate the azo radical anion. This finding, together with the extensive work already performed on arylazopyridines, and the analogy to bis(imino)pyridine guided the chemistry of bis(phenylazo)pyridine pincers, which have moderately low-lying and energetically favourable π* orbitals, located at the azo-fragments. In 2014, 2,6-bis(phenylazo)pyridine was reported as the first example of this type of symmetric pincer, including both homo- and heteroleptic FeII complexes thereof (Fig. 13) [97]. The free ligand already displays two reversible reductions at 1.47 and 1.79 V vs. Fc/Fc+ in acetonitrile, with the ligand radical anion NNN• shows a single-line EPR spectrum at g ¼ 2.001 and is available for isolation by bulk electrolysis. These potentials are strongly anodically shifted compared to the pdi platform, indicating that there are subtle yet significant differences between these two ligand classes. The corresponding FeCl2(NNN) complex was characterized as high-spin FeII with a neutral NNN ligand, based on X-ray crystallography, magnetic data (effective magnetic moment  5.06 μB) and Mössbauer analysis. For the FeCl2 species, only the first (ligand-centred) reduction at 0.08 V is reversible. Reaction of two equivalents of the neutral NNN ligand with hydrated Fe(ClO4)2 led to the isolation of homoleptic species [Fe(NNN)2](ClO4). This complex was best described as low-spin FeII, again derived from combined XRD (elongated N–N bonds), Mössbauer spectroscopic and magnetic susceptibility (μeff  1.65 μB) and a nearly isotropic EPR spectrum with only a small metal contribution at g¼ 1.968. This implies that one ligand is present as a ligand-centred radical anion. This complex displays three successive reversible one-electron ligand-centred reductions at 0.18, 0.88 and1.2 V, with no apparent change in the metal ion oxidation state. The electronic structures of the monocationic homoleptic complex [Fe(NNN)(NNN•)]+ and its mono-oxidized and further reduced forms were also generated by bulk electrolysis and thereafter characterized with various spectroscopic techniques and supporting density functional theory (DFT).

150

J. I. van der Vlugt

Fig. 14 Mechanism of Ni-mediated but factually ligand-centred alcohol dehydrogenation

Following this initial report, the Ni derivative was employed for the efficient catalytic aerobic oxidation of alcohols [98]. Strikingly, the mechanism underlying the mode of operation of this system was determined to be exclusively ligandcentred, involving hydrogen atom transfer via quantum mechanical tunnelling, as supported by a DFT study (Fig. 14). Hence, at least one of the azo-fragments undergoes full hydrogenation to the –NH-NH stage via hydrogen transfer from the substrate, prior to reoxidation by O2. The presence of a co-catalytic amount of

Redox-Active Pincer Ligands

151

KOtBu, albeit not necessary, improves the overall yield. Furthermore, zinc dust is essential for the reaction to proceed, which implies a preactivation of the NiII(Cl)2(NNN) system to generate a ligand-centred radical within one of the azo-fragments. Alcohol deprotonation and alcoholate coordination set up for ratelimiting intramolecular H atom transfer from the substrate to the azo radical side arm to release the oxidized aldehyde or ketone product, aided by single-electron transfer from the ketyl radical substrate to the other azo side arm. This latter fragment becomes reoxidized by an equiv of O2, releasing superoxide. A second equiv of substrate then undergoes proton transfer to generate one fully hydrogenated hydrazido side arm in the ligand and another Ni-bound alcoholate. Ligand reoxidation with external dioxygen completes the cycle. Hence, the Ni only serves to bring the substrate in close proximity to the ligand-centred locus of reactivity. The temperature-dependent KIE of 12–17 measured between 283 and 300 K implies a contribution from quantum mechanical tunnelling. Given that the nickel merely serves as substrate docking station, the corresponding Zn complex also proved catalytically active for the same reaction, with the ligand even able to act as four-electron-four-proton reservoir [99]. The Ni derivative was also reported as a catalyst for the tandem aerobic synthesis of azines from aromatic alcohols and hydrazine, again with ligand redox controlling the reaction [100]. Using Ru analogs with this bis(arylazo)pyridine NNN pincer and bearing different kinds of ancillary ligands, ranging from electron-donating to π-accepting, Goswami and co-workers determined that these have profound effect on the responsivity of the azo moiety and thus on the reactivity thereof [101]. Iron complexes of the strongly related phenanthroline-derived monoazo-NNN systems were also developed, their electronic structure determined and their activity for alcohol oxidation established. However, due to the absence of a second redoxactive azo-arm, FeII/FI redox was implied as part of the mechanism for ketone release in this particular situation [102]. This same ligand scaffold was also recently used for Ni by the group of Paul, demonstrating redox-induced interconversion and hemilability in homoleptic complexes [103] and for Co-catalysed coupling of alcohols and 2-aminobenzamide to furnish quinazolin – 4(3H)-ones [104]. Crabtree and co-workers reported on a Ni system with bis(naphthyridinyl)pyridine as pincer ligand, incorporating two pendant bases, for electrocatalytic CO2 reduction [105]. No precise information on the precise electronic structure is provided, although the authors speculate about ligand-centred redox. When the flanking imino groups within the 1,4-azadiene arrangement are included within a ring, several additional motifs become available. The group of Caulton investigated the electrochemistry of bis(tetrazinyl)pyridine (btzp)-based complexes of chromium and molybdenum [106]. Reaction of the precursor species M(CO)3(NCMe)3 with free ligand resulted in formation of homoleptic M(btzp)2 complexes. A combined spectroscopic, X-ray crystallographic and computational study revealed that the electronic structures of the respective complexes was best described as M2+ with two-ligand radical anions, by virtue of the energetically accessible π*-orbitals within the tetrazine fragments. As such, extensive metal-toligand electron transfer occurs upon coordination. However, this in itself is not

152

J. I. van der Vlugt

enough to qualify this as a redox-active ligand. Cyclic voltammetry signified multiple reversible redox events for both M ¼ Cr and Mo. For chromium, full reversibility over the potential window of +0.4 to 2.7 V was observed, with the first reduction occurring at E 0/1 ¼ 0.831 V. The fact that this Cr-comple displays two additional reversible reductions is attributed to a combination of (1) significant delocalization of charge and spin density, (2) coordinative saturation of the metal, coupled to (3) strong ligand chelation effects, which inhibit (irreversible) bond cleavage and formation processes. The molybdenum analog displayed very similar reductive redox-chemistry, with slightly more positive reduction potentials, which also hints at ligand- rather than metal-centred reductions, given the propensity of 2nd row metals to be less prone to reduction relative to 1st row congeners. DFT calculations suggest more metal-ligand covalency for Mo, resulting in more delocalization of electrons between the π systems of both ligands. The major spin density is found in the flanking tetrazine rings, but the central pyridine ring is also involved in the ligand-centred electron storage. The overall picture contrasts with that observed for M(tpy)-based redox-chemistry. Apart from tetrazine, oxazoline is another valuable functionality in this context, as this could allow for the introduction of chiral entities. So-called Pybox ligands, i.e. pyridine-bisoxazolines, have been studied to a very limited extent relative to other azadiene-based NNN scaffolds. Chirik reported that the Fe(CH2SiMe3)2(NNNox) system was inactive for the catalytic hydrosilylation of 1-octene, in contrast to the bis(imino)pyridine analog and the terpy derivative (vide supra) [87]. Earlier research on this system demonstrated activity in ketone hydrosilylation [107]. Transformation of the corresponding dichloride species into the bis(dinitrogen) analog under reductive conditions was not accessible with Pybox as ligand, forming homoleptic paramagnetic Fe(NNNox)2 instead, with an S ¼ 1 spin state (μeff ¼ 3.0 μB). The corresponding bis(carbonyl) species [Fe(CO)2(NNNox)] showed IR bands that were shifted by Δν 40 cm1 to lower wavenumber relative to the pdi derivative, suggesting that the Pybox side arms provide a more electron-rich Fe centre. The electronic structure for the bis-alkyl complex proved very similar to that of the bis(imino)pyridine analog, with a broken-symmetry high-spin FeIII solution and a ligand-centred radical strongly favoured by DFT calculations and experimental Mössbauer data. A well-characterized square planar [Ni(Pybox)(Ph)]BArF4 system was applied and mechanistically studied by the group of Fu for Negishi-type arylations of propargylic bromides [108]. Electrochemical reduction showed two reversible redox events at 1.37 and 2.36 V vs. Fc/Fc+ in THF. Chemical reduction with decamethylcobaltocene led to the isolable, formally nickel(I) derivative. The metric parameters and solid-state structure of this reduced species showed high similarity with the nickel(II) cation and the EPR spectrum contained a near-axial signal centred at g ¼ 2.00 and coupling with one 14N nucleus in only one direction. On the basis of these observations, the authors postulate that the best electronic description for this reduced species is a NiII with a ligand-centred anion, most likely located on the pyridine ring, similarly as postulated by Vicic for a Ni(tpy)(Me) species (vide supra).

Redox-Active Pincer Ligands

153

Fig. 15 Pd complexes bearing a hemilabile and redox-active diiminopyrrole pincer ligand

The Anderson group developed a new diiminopyrrole NNN pincer ligand that proved to show hemilabile, proton-responsive as well as redox-active character in the coordination sphere of Pd [109]. Starting from the square planar species PdCl (NNN), with all three N-donors bound to palladium, reduction of this species with sodium naphthalenide led to a mixture of products, with one being structurally characterized as a Pd dimer with a Pd–Pd bond, indicating PdI oxidation states. To circumvent this dimerization, addition of two equiv trimethylphosphine to PdCl (NNN) led to PdCl(NNN)(PMe3)2 wherein only the central pyrrolide N is still bound to Pd. Reduction of elemental sodium in THF led to a complex formulated as Pd(L )NNN•), with L either being a THF solvent molecule or a PMe3 (Fig. 15). EPR spectroscopy in solution at r.t. confirmed the paramagnetic nature of this species and suggested the presence of a ligand-centred radical, given the g value of 2.003. The 105Pd coupling is 55 MHz, which, although quite large for a ligandbased radical PdII species, is still significantly smaller than generally observed for a PdI analog.

2.4

Miscellaneous Pincer Systems Amenable to Ligand-Centred Reduction

Milstein and co-workers reported on the redox-noninnocent nature of a neutral acridine-based PNP pincer 4,5-bis(diphenylphosphino)acridine, which was previously utilized for metal-ligand cooperative bond activation with Fe and Mn for acceptorless dehydrogenative coupling catalysis, wherein the central ring underwent reversible protonation para to the nitrogen donor [110–113]. Reduction with Na/Hg product a ligand radical anion in the coordination sphere of Fe, Co and Ni, with spin density heavily localized on the same –CH position undergoing protonation but instead now leading to C–C bond dimerization of two-ligand-radical fragments [114]. It was also shown that a Mn complex bearing a slightly enlarged donor space (4,5-bis(diphenylphosphinomethyl)acridine) revealed the same C–C bond formation. No information on the reversibility of this C–C bond formation has been detailed to date. Chirik evaluated the redox-chemistry, electronic structure and catalytic applicability of Fe and Co complexes of a bis(carbene)pyridine pincer first reported by Danopoulos and co-workers [115]. Starting from the high-spin FeII-dibromide complex [Fe(Br)2(CNC)], reduction with sodium amalgam under N2 atmosphere

154

J. I. van der Vlugt

Fig. 16 Radical reactivity on a CNC pincer backbone in the coordination sphere of Co

led to the bis-dinitrogen adduct [Fe(CNC)(N2)2], which was crystallographically characterized [116]. With small methyl substituents of the flanking NHC-fragment, dimerization occurred in the solid state (with one linear bridging N2 ligand), but solution state IR spectroscopy indicated only mononuclear species to be present in toluene at r.t. These reduced Fe species proved effective for the hydrogenation of unfunctionalized, hindered alkenes. A combined spectroscopic and computational study revealed that in the coordination sphere of Fe, this CNC framework behaves as a redox-noninnocent ligand but not strictly as a redox-active one [117]. This means that extensive π-backbonding results in an electronic structure that is a hybrid of Fe0 and FeII but without concrete radical spin density located within the CNC scaffold. The same picture arose upon substitution of the N2 ligands for the diene N,N-diallyltert-butylamine, to which this Fe species proved catalytically inactive, whilst it was previously shown to undergo 2π + 2π cycloaddition mediated by Fe species with ligand-centred radical-based character. The corresponding CoII-methyl complex with this bis(arylimidazol-2-ylidene) pyridine CNC ligand was shown to be one of the most catalytically active Co systems for the hydrogenation of sterically hindered, unactivated alkenes. The corresponding cobalt hydride complex, [Co(iPrCNC)(H)], which is generated from the methyl derivative under an atmosphere of H2, underwent migration of the metal hydride to the 4-position of the pyridine ring to give [Co(4-H2-iPrCNC)(N2)] (Fig. 16). Similar alkyl migration also occurred when 1,1-diphenylethylene was reacted with this Co-H species [118]. These observations and a combined X-ray structural, spectroscopic and computational investigation led the Chirik group to conclude that on Co, this CNC ligand platform does qualify as redox-active under reductive conditions, given the definitive evidence for a bis(arylimidazol-2-ylidene) pyridine radical in the coordination sphere of Co. Spin density calculations established that the radicals were localized on the pyridine ring, accounting for the observed reactivity. Several bipyridine-based pincers, often featuring a flanking phosphine, have been utilized for, e.g. electrochemical CO2 reduction [119, 120]. Despite the 2,20 -bipyridine being redox-active, extensive studies into the electronic structure have not been reported to date.

Redox-Active Pincer Ligands

155

3 Pincer Ligands That Undergo Oxidative Chemistry Two overarching or underlying motifs that guide oxidative ligand-centred redoxreactivity are the 1,2-aminophenol and ortho-phenylene diamine frameworks [121], which bear strong similarity with 1,2-catechol [122, 123] as well as 1,2-dithiolene scaffolds [124–126] that are well-established redox-active moieties in their own right (albeit not in the context of pincer chemistry). Upon deprotonation and coordination of the 1,2-aminophenol (and sometimes even as free organic fragment), this motif readily allows for reversible oxidation from the dianionic amidophenolato ap to monoanionic monoradical iminosemiquinonato isq to neutral iminobenzoquinone ibq state (Fig. 17). Apart from this well-established platform, which may be extended to afford different types of pincers depending on the donor properties of the flanking entity, diarylamide has proven a fruitful platform to extend to redox-active pincer systems. This section will discuss pincer designs based on the highest possible charge available for coordination to a transition metal.

3.1

Trianionic Pincer Ligands

Integration of this 1,2-aminophenol motif into a tridentate arrangement with a deprotonable flanking donor group does not alter this redox-chemistry. Coupling these ‘catechol-type’ systems with a flanking donor group that readily undergoes deprotonation has led to the development and application of symmetric redox-active ONO, SNS and NNN ligand platforms [127]. The group of Heyduk made groundbreaking contributions to the field of redoxactive pincers from 2008 onward, but his earlier contributions on complexes featuring two bidentate catechol-type redox-active ligands within a metal coordination sphere have paved the way since 2005. The group has extensively worked on group transfer reactions on d0 metal centres [128, 129]. The switch from bidentate to tridentate pincer-like ligand skeletons allows for more stable complexes, a more open coordination sphere for substrate activation and bond formation and conjugation over both aromatic rings of the ligand. The major achievement from this group’s body of work is the realization that redox-active (pincer) ligands can act as two-electron storage reservoirs, particularly when combined with d electron-deficient group 4 metals in their highest oxidation state. As such, two-electron-type transformations including substrate activation and bond formation (e.g. oxidative addition, reductive elimination) can occur by ligand-to-metal two-electron transfer. Fig. 17 General redox states of 1,2-aminophenolbased pincer systems (and derivatives thereof)

156

J. I. van der Vlugt

Fig. 18 Oxidative addition of Cl2 at a d0 ZrIV metal centre mediated by two-electron donation from the NNN ligand

This has enabled for instance stoichiometric nitrene transfer reactions on TaV [130], as well as the formal oxidative addition of Cl2 (in the form of PhICl2) onto ZrIV, and even the radical addition of a chlorine atom from this iodobenzene dichloride reagent on a d0 metal was facilitated by ligand redox (Fig. 18) [131]. Similar ligand-to-metal single-electron transfer to facilitate Cl atom addition was also reported for TaV with the NNN platform [132]. The presence of the ligand radical was probed by UV-vis spectroscopy, CV and EPR spectroscopy, showing an eight-line pattern centred at g ¼ 1.897 at r.t., consistent with a single unpaired electron coupling to the I ¼ 7/2 tantalum centre and no (resolved) coupling to either of the nitrogen atoms or aromatic hydrogens. The combined data support a ligandcentred radical but with significant delocalization of the ligand valence orbitals onto the Ta centre. Follow-up work focussed on tuning the electronic and steric parameters of this NNN platform, whilst coordinated to TaV, by introduction of substituents at the para-positions of the aromatic rings, relative to the amido nitrogen, which led to a series of species that spanned 270 mV for the oxidation potentials, whilst no significant changes to the structural or spectroscopic properties of the various tantalum complexes could be discerned [133]. Furthermore, using the complex [ZrIVCl(CNtBu)2(NNN)] as starting point, with NNN being the fully deprotonated form of bis(2-isopropylamino-4methoxyphenylamine, nitrene transfer from an organoazide donor to an isocyanide acceptor led to the formation of carbodiimides (Fig. 19) [131]. Initial dissociation of one of the isocyanides to free a coordination space and follow-up reaction with one equivalent of p-tert-butylphenyl azide produced a Zr-imido species. The NNN ligand undergoes two-electron oxidation to afford the required electron density at Zr to accommodate this dianionic ligand. The imido group acts as nucleophile toward the electrophilic carbon of the coordinated isocyanide substrate, forming a three-membered Zr-C-N metalacycle that transforms, via formal reductive elimination of the C¼N bond that is facilitated by two-electron transfer to reinstall the NNNap ligand oxidation state, to N-bound diimide that undergoes exchange with fresh isocyanide. The Baik group reported a DFT computational study on the mechanism of this Zr-mediated nitrogen group transfer reaction that supported the mechanism proposed by Heyduk [134]. The ligand-to-substrate overall two-electron transfer proved

Redox-Active Pincer Ligands

157

Fig. 19 Catalytic nitrene transfer by [ZrCl(CNtBu)2((NNN)] complex, yielding carbodiimide 4-tBuC6H4N¼N¼NtBu

energetically favoured over a more ‘common’ metal-to-substrate electron transfer pathway, likely due to the d0 metal centre. However, the precise pathway involved two sequential and non-concerted ligand-to-azide single-electron transfer steps. The first electron transfer (resulting in the NNN ligand going from ap to isq oxidation state) produces a one-electron reduced azido fragment that binds side-on via the α and β nitrogen atoms (with α being the nitrogen that bears the organic substituent). N2 loss is realized only upon further NNN ligand oxidation to the ibq state, resulting in the zirconium-imido species. Strikingly, the initially formed isomer is the transone, with the imido opposite the isocyanide. Prior to the productive C–N bond forming step, isomerization to a cis-conformer is required. Heyduk and co-workers also carried out a study on the isostructural vanadium, niobium and tantalum complexes of the strongly related ONO ligand bis(3,5-di-tertbutyl-2-phenol)amine to evaluate the impact of the metal ion on redox activity of the ligand platform [135]. The 2nd and 3rd row congeners were both best described as MV with a trianionic ONOap ligand, but for vanadium, the electronic description VIV(ONOisq) was more appropriate, based on an X-ray crystallographic study.

158

J. I. van der Vlugt

Fig. 20 Redox-active ligand facilitated reductive elimination of a disulfide from Fe(III)

Electrochemical ligand-centred oxidation thus led to a species best formulated as having a VIV(ONOibq) configuration. DFT and TD-DFT calculations were used to probe the electronic structure of the complexes and help verify the different electronic structures stemming from changes to the coordinated metal ion. Five- (with phosphine co-ligands) and six-coordinate (pyridine as co-ligands) RhIII complexes with this ONO platform were investigated [136]. The nature of the co-ligand strongly affected the electrochemistry of the overall system. For the tris(pyridine) complex [Rh(ONO)(Py)3], the first oxidation to convert this species into the corresponding cation is observed at E½1 ¼ 0.91 V, whereas for [Rh(ONO) (PPh3)2], this event is shifted anodically by 640 mV to E½1 ¼ 0.27 V. Using mixed methylphenylphosphines or trimethylphosphine leads to less significant shifting. However, the second oxidation event proved insensitive to both the coordination geometry around Rh and to nature of the ancillary ligand, with a redox potential E½2 between +0.07 and +0.19 V. The combined experimental evidence, also including NMR and UV-vis spectroscopy, leads to an overall S ¼ 0 spin state with no unpaired electrons. DFT calculations have been used to clarify the precise interplay between the ligand and metal oxidation state. For most reported Rh complexes, RhIII with a trianionic ONOap ligand is deemed the best description. However, for some of the phosphine complexes, the ONO ligand and the rhodium centre both significantly contribute to the π-bonding HOMO and the π-antibonding LUMO, giving rise to unique spectroscopic properties, which led the authors to propose that RhII(ONOsq) (PR3)2 is a valid limiting resonance structure, emphasizing that the metal and ligand share the HOMO electron pair. Rather than viewing this as a diradical, the covalent nature of the ligand-metal interaction prohibits the description of this complex as having an open-shell, singlet-biradical electron configuration. Reductive coupling of thiols to disulfide was achieved using an wellcharacterized [Fe(ONOibq)(NSiMe3)2] complex (Fig. 20) [137]. The crystal structure of the latter species contained alternating C–C and C¼C bonds, in the ligand backbone, as deduced from the metric parameters, as well as two C¼O bonds, which supports the ibq oxidation state being retained upon coordination of the ligand precursor ([K]ONO) to the FeIII starting material. EPR and Mössbauer spectroscopy provided evidence for a high-spin S ¼ 5/2 FeIII centre. Treatment of this species with two equiv of tert-butylthiol in the presence of pyridine led to the formation of bis(trimethyl)amine, tert-butyldisulfide and the complex [Fe(ONOap)(Py)3]. Hence, whilst the oxidation state of Fe remains +III before and after reaction, the ligand has

Redox-Active Pincer Ligands

159

Fig. 21 Redox activity in a tripyrrindione pincer ligand whilst coordinated to PdII

undergone a two-electron reduction. The authors were unable to isolate or observe the disulfide intermediate, but they did formulate three potential mechanisms for the reductive coupling, presuming at least one (or two) of the two thiolate fragments initially bind(s) to iron, following proton transfer to the -N(SiMe3)2 fragment: (1) a thiol-thiolate coupling, (2) a bimetallic pathway and (3) ligand-assisted reductive elimination. More recently, both the ONO and the related SNS [138] platforms were utilized to generate (hetero)dinuclear complexes, taking advantage of the fact that phenolate and thiolate fragments can readily bridge two metal centres. This has led to a series of ligand-mixed valent Fe-Fe and Fe-Zn species with interesting magnetic properties [139] as well as a W-Ni system, best formulated as WV-NiI and with a direct metal– metal bond and two bridging thiolates that proved active for the electrocatalytic reduction of protons [140]. The Fe or W centre, respectively, is bound to two SNS ligands, effectively creating a metalloligand fragment. Whilst the electrochemical data suggest that one-electron reduction of the dinuclear system precedes protonation, the locus of protonation and subsequent hydrogen evolution remains unclear. Two distinct possibilities are (1) proton reduction at the nickel centre, in analogy with enzymatic and biomimetic NiFe-hydrogenase systems as well as monometallic Ni analogs or (2) tungsten-centred proton reduction. Also W-Pd and W-Pt species were described, but these featured only one rather than two bridging thiolates, as well as a tungsten group 10 metal–metal bond [141]. Electrochemical and computational data show that the frontier orbitals in these systems are predominantly localized on the W(SNS)2 redox-active metalloligand. A mononuclear Ni(SNS) complex showed very interesting H atom transfer reactivity, which would not only be an alternative pathway of generating an SNSisq ligand-centred radical within the coordination sphere of, in this case, a nickel centre but also could merge the fields of proton-responsive and redox-active ligands [142]. Trinuclear clusters featuring the Mo(SNS)-metalloligand were also recently disclosed [143]. The group of Thomas reported on the electrochemistry of homoleptic nine-coordinate lanthanide complexes M(ONO)3, with M being Eu, Gd, Yb and Lu [144]. The group of Tomat recently explored the coordination chemistry of linear tripyrroles, which are ubiquitous in nature as precursors for the biosynthesis of cyclic tetrapyrroles (e.g. porphyrins) or as metabolites appearing during degradation thereof [145]. One particularly interesting motif in this respect is the tripyrrin-1,14dione, which can act as a trianionic pincer ligand upon double deprotonation of both –NH groups. Although perhaps not too surprising, given the known redox activity of porphyrins, Tomat described that the square planar Pd(NNN)(OH2) complex (Fig. 21) with a tripyrrin ligand indeed showed well-behaved

160

J. I. van der Vlugt

Fig. 22 Reactivity of a dihydrazonopyrrole pincer ligand in the coordination sphere of Ni

ligand-based chemo- and electro-oxidation chemistry, with all three redox states of the ligand X-ray crystallographically characterized and the ligand radical species exhibiting a single line centred at g ¼ 2.003 in the EPR spectrum [146]. The majority of the spin density as well as the SOMO are localized on the central pyrrole as well as on the C-C π system of the two pyrrolidone flanking rings. Additionally, the dione fragments allow for H bonding to the ancillary ligand. Follow-up work including also the structurally related CuII(NNN•)(OH2) analog revealed that in solution, the neutral tripyrrindione complexes dimerize, based on EPR and UV-vis spectroscopy. Variable-temperature measurements using both EPR and absorption techniques allowed for the determination of the thermodynamic parameters of this π-dimerization, which resembles the process occurring with porphyrin radical cations. The inferred electronic structure, featuring coupling of ligand-based electronic spins in the π-dimers, was supported by DFT calculations [147]. The group of Anderson recently designed a dihydrazonopyrrole platform that can be considered related to that of known diiminopyrrole ligands, utilized inter alia by the Schaper group for lactide polymerization catalysis [148, 149]. Introduction of the hydrazone fragments dramatically increases the steric hindrance but also enlarges the chelate size and provides for an extended conjugated backbone for this NNN platform, which potentially could act as a trianionic ligand upon full deprotonation of all three –NH groups [150]. Reaction of this ligand with only two equiv of KH as strong base and addition of NiCl2(dme) as well as free PMe3 led to two NiII species (Fig. 22). The first contains the asymmetric dianionic NNN ligand, with one hydrazone side arm still protonated, but the second was crystallographically characterized as fully square planar C2 symmetric binding with two six-membered chelate rings. This species proved to be paramagnetic, and EPR spectroscopy confirmed the locus for this to be the ligand side arm. Computational analysis revealed that the spin density mainly resides on the ligand. Chemical oxidation and reduction led to isolation of the corresponding cation and anion, respectively. The neutral species could be converted to the asymmetrically binding monoprotonated analog by reaction with H2, lending support to formation of this radical species by oxidation and H atom loss from the asymmetric derivative. The authors also reported on versatile ligand-based reactivity towards small molecules. Follow-up work related to this initial Ni chemistry led, upon reaction of the pyridine adduct Ni(NNN•)(Py) with decamethylcobaltocene as reducing agent, to the isolation of an anionic T-shaped NiII(NNN) complex with a vacant coordination site and the NNN ligand fully trianionic in charge (Fig. 23) [151]. This latter species

Redox-Active Pincer Ligands

161

Fig. 23 Generation and reactivity of a T-shaped anionic NiII complex with a dianionic NNN ligand

could be reoxidized using AgBF4 in the absence of coordinating donors to the corresponding T-shaped Ni(NNN•). This allowed the authors a platform for metalligand cooperative activation of water. Very recently, the same group reported on the generation and oxidative reactivity of a NiII-superoxo complex, by reaction of a very similar but more bulky T-shaped ligand-centred radical precursor, via ligand-to substrate single-electron transfer, which paved the way for stoichiometric O atom transfer reactions [152].

3.2

Dianionic Pincer Systems

Inspired by the work from the Heyduk group as well as by pioneering work in the area of 1,2-catechol-type redox-active ligands by Pierpont, Kaim and Wieghardt, the van der Vlugt group developed hybrid redox-active pincers based on the same 1,2-aminophenol framework as employed in, e.g. the ONO pincer, but with pyridine or phosphine as neutral flanking heterodonor groups to provide overall dianionic platforms [153–155]. These have proven very versatile for the integration of, e.g. PdII, a metal ion not commonly associated with single-electron chemistry. Both the NNO and the PNO ligand platform readily coordinate to PdII in either the fully protonated L form, in a monoanionic LH form, with the acidic phenol converted into the phenolato or as a doubly deprotonated and monoanionic iminosemiquinonato ligand radical form LISQ, giving access to crystalline paramagnetic square planar PdCl(LISQ) complexes with discrete well-behaved ligand-based electrochemical responses according to cyclic voltammetry (Fig. 24). Starting from the latter species, well-behaved and fully reversible one-electron ligand-centred reduction was observed at 0.8 (PNO) or 1.1 V vs. Fc/Fc+ (NNO), whilst one-electron oxidation occurred at 0.1 (NNO) or +0.1 V vs. Fc/Fc+ (PNO), showing a subtle but significant and somewhat counter-intuitive dependency of the redox properties on the peripheral donor group. The respective EPR spectra for the

162

J. I. van der Vlugt

Fig. 24 Reactivity of NNO and PNO ligands bearing a redox-active o-aminophenol fragment in the coordination sphere of PdII

Pd(NNOISQ)Cl and Pd(PNOISQ)Cl complexes at r.t. reveal hyperfine coupling interactions to the two hydrogens present on the NO ring as well as to the iminonitrogen and the Pd centre present, with an additional coupling to either the pyridylN or the phosphine-P. The calculated spin density distribution is almost completely (~90–95%) localized on the redox-active ligand scaffold in these systems. Chemical reduction of the bench-stable Pd(NNOisq)Cl complex with CoCp2 led to the air-sensitive but isolable and crystallographically characterized [Pd(NNOap) Cl] derivative, which can be considered as having one electron stored within the NNO platform. This electron can be transferred to a Pd-bound substrate, as was shown by the activation and follow-up cyclization (Fig. 25) of 1-azido-4phenylbutane, an organic azide with a weak benzylic C(sp3)-H bond [153]. Upon in situ reduction of Pd(NNOisq)Cl and exchange of the chloride, thermal azide activation leads to release of N2 and generation of a Pd-bound nitrene fragment bearing a low-lying empty p orbital at the nitrene nitrogen centre. The thus-formed donor-acceptor electron pair undergoes single-electron transfer from the electronrich NNOAP platform to the hypovalent –NR fragment to create a diradical species that is best described as an open-shell singlet, based on DFT calculations. This effectively generates a nitrogen-based substrate radical bound to PdII that is amenable to C-H insertion reactivity, as demonstrated by the formation of the N-Bocpyrrolidine adduct from the benchmark substrate 4-phenylbutyl-1-azide via two-step H atom transfer and rebound, as supported experimentally by KIE data and DFT computations. Although initially only stoichiometric reactivity was obtained, removal of crystal lattice solvent (chloroform) enabled up to three turnovers for this Pd-mediated radical C-H amination, demonstrating the potential for ligand-tosubstrate single-electron transfer for radical-type catalysis at redox-inert metal centres [155].

Redox-Active Pincer Ligands

163

Fig. 25 Generation of a nitrene radical in the coordination sphere of PdII via ligand-to-substrate single-electron transfer and follow-up C-H amination reactivity tBu

tBu tBu

tBu

O N Pd Cl P Ph

Ph

TIPF6

tBu

O

tBu PhS SPh

N Pd P Ph

Ph

O

Ph Ph P Pd

N Pd

S Ph

P Ph

N

tBu

O tBu

Ph

Fig. 26 Ligand-to-substrate single-electron transfer to homolytically cleave disulfides and generate ligand-mixed valent dinuclear Pd species

The same group also reported that upon abstraction of the halide co-ligand from Pd(PNOAP)Cl using redox-inert Tl-salt in the presence of a disulfide led to ligand-todisulfide single-electron transfer to induce stoichiometric homolytic S–S bond cleavage (Fig. 26) [154]. Using a 1:1 ratio of Pd complex and diphenyl disulfide led to formation of a free thiyl radical after S-S cleavage, as PhSSPh was detected from reaction of the thiyl radical with the benzene solvent by GC-MS analysis. Use of di(tert-butyl)disulfide allowed for spectroscopic observation of the corresponding disulfide adduct by NMR spectroscopy. The initially formed palladium-thiolate species dimerizes with a second equiv of reduced Pd(PNOAP) complex, resulting in the crystallographically characterized ligand-mixed valent dinuclear species (PNOISQ)Pd-SPh-Pd(PNOAP). Clean formation of the dimeric species was obtained when using a 4:1 ratio of Pd complex to disulfide substrate. Related to these mononuclear complexes, homodinuclear palladium complexes and close analogs thereof were also reported, together with the magnetochemistry and spectroelectrochemistry of their ligand radical binding pockets [156, 157]. Moreover, a route was devised to prepared hetero di- and trinuclear d8-d10 complexes, some of which show remarkable electrocatalytic reactivity that appears to be modulated by the ligand oxidation state and the intramolecular metal-metal distance that is tuned by this ligand-based oxidation state change [158]. The van der Vlugt group also reported that the reverse, i.e. substrate-to-ligand single-electron transfer, is feasible with these hybrid redox-active aminophenolderived pincer systems (Fig. 27) [159]. In situ reaction of NaN3 with either cis- or trans-RuIII(Cl)2(PPh3)(NNOisq) led to the formation of a crystallographically characterized trinuclear species bearing two bridging nitrido ligands and three fully reduced NNOap fragments. Using isotope-labelled 15N-azide strongly suggests

164

J. I. van der Vlugt

Fig. 27 Formation of a trinuclear ruthenium complex with two bridging nitrides via single-electron transfer from a nitride ligand to the redox-active ligand

multiple events of substrate-to-ligand single-electron transfer, as GC analysis indicates all three isotopomers of N2 are formed as by-product. There appears to be much potential to extrapolate this concept of activating substrates via a substrate-to-ligand single-electron transfer to catalytic applications. Most recently, the well-defined complex FeIII(Cl)2(NNOisq) was reported [160]. The combined X-ray crystallographic, Mössbauer and UV-vis spectroscopy, SQUID and solution state magnetometry and DFT-based spin density calculations confirmed the high-spin overall S ¼ 2 spin state for this FeCl2(NNOisq) system. Only quasi-reversible reduction and oxidation events were observed by CV in dichloromethane at 0.74 and +0.51 V vs. Fc/Fc+, respectively. Chemical oxidation using a silver salt led to chloride abstraction but no stable species could be isolated. Chemical reduction using cobaltocene led to a disproportionation reaction, generating homoleptic FeII(NNOisq)2. The latter proved catalytically inactive, whereas the dichloride species was shown to be a precatalyst for the intramolecular C(sp3)-H amination of a broad range of organoazides, with TON’s of up to 620 at 0.1 mol% catalyst loading (Fig. 28). Kinetic data reveal the rate-limiting step to include di-tertbutyl dicarbonate (Boc2O; first order measured) but not the azide substrate (zero order in azide observed), which is very atypical for azide C-H amination catalysis. The group of Soper showed that a bis(phenolato)-NHC ligand, first reported in 2009 [161] and previously proposed by the groups of Bellemin-Laponnaz and Dagorne and independently by the Bercaw group to have redox activity 162–164], displays well-behaved redox-chemistry in the coordination sphere of cobalt, leading to a low-spin CoII species with a dianionic, closed-shell bisphenolate ligand [165]. Cyclic voltammetry demonstrated the existence of three redox states, accessible via quasi-reversible events at 0.32, 0.30 and 0.77 V, respectively, which both the metal and the OCO ligand taking part in the redox-chemistry (Fig. 29). Hence, the authors conclude that the combined data, including X-ray crystallography, provide the first unequivocal and structural evidence for OCO-ligand-based redox activity. Notably, the Heyduk group reported that the related tetradentate ONNO

Redox-Active Pincer Ligands

165

Fig. 28 FeIII(Cl)2(NNOisq) is a precatalyst for the intramolecular C-H amination of aliphatic azides

Fig. 29 Redox activity within an OCNHCO platform bound to Co

ligand system also displays redox activity [166]. Varying the degree of unsaturation in the NHC backbone was found to modify the ligand-based oxidation potentials by up to 400 mV, related to the varying degree of π-back donation character. This Co (OCO) platform was utilized for the photoinduced activation of a CoIII–CF3 bond to afford arene C-H trifluoromethylation, although the low-spin character at Co likely prevents any direct ligand redox role [167]. The Hochloch group recently developed a related triazole-derived redox-active OCO-pincer featuring a mesoionic (‘abnormal’) carbene in the backbone [168]. The corresponding TiIV complex showed two

166

J. I. van der Vlugt

reversible oxidation processes at +0.46 V and +0.93 V vs. Fc/Fc+ by CV, which were both assigned to occur at the ligand, based on similarity with the NHC-based OCO system. Although originally starting from a bis(arylazo)pyrrolide scaffold, which was expected to exhibit only ligand-centred reduction chemistry, Pramanik and coworkers established that coordination to RhIIICl3 led to oxygenation of one of the side arm aryl groups, converting the NNN pincer into a preferential dianionic NNO pincer, wherein the phenolate fragment displayed ligand-centred oxidation chemistry [169].

3.3

Monoanionic Pincer Ligands

The Iluc group reported on the formation of a nucleophilic carbene, embedded within a tridentate PCP pincer arrangement, in the coordination sphere of PdII, providing complex [Pd(PCP)(PMe3)] with formally a dianionic ligand, although the alternative description as a zwitterionic C-Pd+ appears more appropriate, given the single bond character of the Pd–C bond (Fig. 30) [170]. This ligand was previously used by Piers for Ir chemistry and later shown to also support metalligand cooperative reactivity with Ni [171–173]. DFT calculations supported experimental data showing that the HOMO is indeed largely C-centred, allowing, e.g. protonation and methylation at this pivotal carbon position of the pincer platform. Upon reaction with either molecular bromine or iodine, oxidation of the carbene carbon leads to formation of the mononuclear species [PdPCP)(X)], which for X ¼ iodide is in equilibrium with the dimeric form in which Ccarbene-Cphenyl coupling has taken place [174]. The mononuclear chlorido derivative, with an effective magnetic moment μeff of 1.86 μB, was also accessible from reaction with dichloromethane via halogen atom transfer. This reaction also generated another Pd species, originating from ‘methylene’ transfer, with a vinylic fragment incorporated at the PCP backbone that is side-on bound to Pd. Crystallographic data provide the first bona fide

Fig. 30 A carbon-centred radical in a PCcarbeneP platform in the coordination sphere of PdII and follow-up reactivity

Redox-Active Pincer Ligands

167

Fig. 31 Double ligand-to-substrate single-electron transfer to generate a diradical dinuclear complex

examples of palladium radical carbene complexes, with g values close to 2 for all halide complexes. The DFT-calculated spin density plot showed ~65% radical character at the central carbon atom of the PCP scaffold. This system displays stoichiometric H atom transfer reactivity with a variety of donors bearing weak C– H bonds. More recently, the same group also reported the PtII congener of this carbene radical ligand [175]. This species showed an effective magnetic moment μeff of 1.27 μΒ in solution (Evans’ method) as well as an EPR signal in line with a ligandcentred radical, with a g value of 2.0105 and hyperfine coupling with 195Pt nucleus of 92 G, ruling out a platinum centred radical. Unlike what was found with Pd, dimerization for the iodido-containing variant was not observed with Pt. Although in CV only quasi-reversible one-electron events were observed with the Pd-complex, chemical oxidation successfully led to isolation of the cationic carbene complex (Pd (PCP)I]BArF4, which was also crystallographically characterized. The C in this complex was shown to have electrophilic character, and thus a formal Umpolung has taken place at this pivotal donor [176]. The carbene radical was also shown to be able to cleave S–S bonds in a homolytic fashion. Oxidation of the nucleophilic neutral carbene complex [Pd(PCP)(PMe3)] bearing a PCP variant containing tert-butyl groups on both phenyl rings that are para to the carbene carbon provided the cation ligand radical-containing analogous PdII species a EPR doublet at g ¼ 2.020, from coupling with the 31P nucleus of trans-bound PMe3 and hyperfine coupling to all six aromatic hydrogens of the PCP backbone [177]. However, reaction of the neutral carbene species with either 1,8-naphthylene disulfide or 9,10-anthracenedione afforded an overall biradical species, with a 1,8-naphthylene dithiolate or 9.10-anthracenediolate bridging both Pd centres (Fig. 31). Hence, two ligand-to-substrate single-electron transfer processes have occurred. For the bis-thiolate species (μeff ¼ 3.51 μB, suggesting 2  S ¼ ½ or overall S ¼ 1), EPR spectroscopy provided a signal at g ¼ 2.0210 with hyperfine coupling to three pairs of equivalent protons and a 105Pd nucleus. These observations suggest two isolated Pd-carbene radical fragments, pointing to a lack of interaction between the two carbene radical centres. Originally prepared by the groups of Liang [178] and Kaska and Mayer in 2003 [179], diarylamino-based PNP pincers have rapidly gained a prominent place within the pincer community. Mostly this is due to the combination of a strong π-donor trans to a metal site available for substrate binding (e.g. in the case of square planar

168

J. I. van der Vlugt

Fig. 32 N-centred one-electron oxidation within a monoanionic PNP pincer platform

complexes) with bulky peripheral phosphine fragments and the relative rigidity of the diarylamino skeleton. However, given their diarylamino-based skeleton, which itself is known to undergo oxidation, these PNP ligands may also engage in reversible one-electron oxidation processes at nitrogen upon deprotonation and coordination of the central amido functionality to a (transition) metal. The first evidence for ligand-centred redox-chemistry was shown in 2008 by the group of Mindiola (Fig. 32) [180]. Electrochemical oxidation of green [NiCl(MePNPiPr)] using cyclic voltammetry showed a reversible anodic wave at -0.06 V vs. Fc/Fc+. Following this initial report, van der Vlugt and co-workers showed that the corresponding Ni(PNP)-azido complex displayed similar reactivity, with a redox potential E½ of 0.03 V vs. Fc/Fc+ (in THF) [181]. Chemical oxidation of [NiCl(MePNPiPr)] using ferrocenium triflate resulted in isolation of the purple square planar NiII complex [NiCl((MePNPiPr)]OTf, which was crystallographically characterized. Almost no structural changes were noticeable relative to the starting complex, strongly indicating that Ni had remained its +II oxidation state. X-Band EPR spectroscopy in solution at r.t. gave an isotropic signal with a g value of 2.0238, pointing toward a ligand-centred radical, and additional hyperfine interactions with the nitrogen and both P atoms as well as six aromatic H atoms from the ligand backbone and DFT calculations, showing ~60% ligandcentred spin density, also supported this. A room temperature EPR spectrum of the solid sample revealed substantial spin density to reside on the Ni centre as well. The amido p orbital dominates in the HOMO of these complexes, leading to a large contribution of this orbital in the SOMO of the oxidized complex. Additionally, there is substantial delocalization of the spin density over the phenyl rings of the ligand backbone. A combined UV-vis and multi-edge X-ray absorption spectroscopy (XAS) further corroborated the electronic structure of [NiCl((MePNPiPr)]OTf – the latter is considered a direct experimental way of deduce ligand-based redoxchemistry, as it enables probing of the redox-active orbitals. Hence, the authors combined Ni L-, Cl K- and P K-edge XAS measurements to determine changes in the electronic structure on going from neutral [NiCl((MePNPiPr)] to one-electron oxidized [NiCl((MePNPiPr)]OTf. The Ni LIII-edge spectra for both species reveal only minor differences, which substantiates the claim that the nickel is only mildly affected by the redox process. The related CoCl(MePNPiPr) complex and close analogs thereof, such as the azido-derivative, are presumed to feature very similar ligand-centred one-electron oxidation behaviour, although this has not been completely substantiated to date [182]. In sharp contrast, reduction chemistry occurs exclusively at Co or at Ni, resulting in dinuclear complexes with amido-bridging PNP ligands, as detailed in separate studies by the Mindiola group [183, 184].

Redox-Active Pincer Ligands

169

Fig. 33 Generation and follow-up C–C bond forming reactivity of a PNP-centred radical species

The aromatic rings of diarylamino-based ligands are susceptible toward radical reactivity at the ortho- and para-positions with respect to the amido functionality upon N coordination to a transition metal. Strikingly, substitution of the sensitive para-positions, as in the ditolylamino-based PNP systems, does not prevents all radical reactivity, as highlighted by work from the Ozerov and Nocera groups (Fig. 33) [185]. They prepared the neutral species [M(CO)3(MePNPiPr)] (M ¼ Mn, Re) as well as the one-electron oxidized cationic derivatives by chemical oxidation using AgOTf. EPR spectroscopy again indicates ligand-centred redox-chemistry. For the cationic Mn complex, the X-band EPR spectrum of the frozen solution showed a broad isotropic signal at 4.2 K. The solution spectrum displayed hyperfine interactions, dominated by coupling of the electronic spin with the 55Mn nucleus (I ¼ 5/2, 100%), as well as partially resolved coupling with the 14N (I ¼ 1, 99.63%) nucleus, whilst expected coupling with 31P and 1H nuclei was not resolved. The Re analog only showed 185, 187Re hyperfine coupling at r.t. Despite these experimental observations, the fact that the observed g values of 2.004 (Mn) and 2.013 (Re) are very close to the value for the free electron (ge ¼ 2.0023) and the well-resolved spectra at room temperature, which indicates slow spin relaxation rates on the EPR time scale, strongly suggest ligand-centred radical character – the alternative electronic description as MnII and ReII would lead to larger g value shifts (Δg) as well as different values for the metal hyperfine coupling constants as observed for these species. This assignment is further supported by DFT and TD-DFT calculations – overall about 50% spin density is localized at the amido nitrogen, with the remainder almost equally distributed over the two tolyl rings. These one-electron oxidized species were subsequently shown to undergo radical-type C–C bond formation at the ligand with allyltributyltin as the allyl radical transfer reagent. Peters and co-workers reported on dinuclear Cu complexes with bridging diphenylamine-based PNP ligands that feature Cu ions in various oxidation states [186, 187]. This group also reported one isolated case of ligand backbone redoxchemistry at the non-substituted para-position that culminated in C–C bond formation [188]. Mindiola reported a dimeric Ag complex [189], but has not detailed any ligand-based redox-chemistry thereof. The group of van der Vlugt recently described the first Au complexes with these diarylamido-bisphosphine-based PNP pincers [190, 191]. Apart from very interesting coordination chemistry, enabling both mononuclear AuI and dinuclear AuI-AuI, AuI-AuIII and AuIII-AuIII species, also ligand-centred oxidation at the ortho position relative to the amido nitrogen was reported. However, no reversible (electro)chemical redox-chemistry was observed for these species either.

170

J. I. van der Vlugt

The related PNP-platform bearing non-arene C2 bridgeheads between the central amino-N and the flanking phosphine-P donors [192, 193] display redox activity, whereby the reversibility as well as the locus of the oxidation depends both on the metal present (Co or Ni) and on the level of dehydrogenation of the C2 bridges, as detailed by the group of Schneider. The redox-chemistry of the Ni-di(vinylphosphino)amido derivative was characterized in detail using XAS, EPR, CV, XRD, spectroelectrochemistry and DFT [194]. On top of this ligandcentred redox-chemistry, this platform also allows for proton and H atom transfer reactivity. Strikingly, the Co analog showed exclusive oxidation (and reduction) chemistry at the metal centre, not the ligand backbone, resulting in the isolation of a rare square planar CoIII complex that was crystallographically characterized [195]. Diarylamido-monophosphine analogs as well all-nitrogen diarylamido-bisamino analogs have also been established as redox-active ligand frameworks. Ozerov and co-workers prepared a series of PNZ ligands, with Z being either a differently substituted phosphorus-based donor or an imine donor [196]. A series of group 10 metal complexes (M ¼ Pd, Pt, some Ni) was prepared, complemented by the corresponding Rh(CO) complexes and the electronic properties scrutinized by CV as well as IR spectroscopy (for the Rh-carbonyl species). Quasi-reversible one-electron redox events were observed across the whole series, with several observations supporting ligand-centred redox, including (1) redox potentials were mainly affected by substituents conjugated with the diarylamido π system but remote from the metal centre; (2) only small changes were observed for the E½ values across the various metals; and (3) structural changes in the diarylamido backbone largely effected the redox potential. The IR spectroscopic study of the various Rh-carbonyl complexes allowed for insight in the donor strength of the pincer ligands, and although there did not appear to be a clear relationship between νCO as measured by IR and the E½ obtained by CV across the whole ligand scope (i.e. the most electron-donating ligand is not the most easily oxidized), some trends could be identified within smaller subsets of ligands wherein only substituents at specific sites were varied. The Fiedler group reported on dinuclear cobalt complexes bearing a diarylamido as bridging moiety to span two binding pockets consisting of Schiff-base type salicylaldimine chelates that also feature redox activity in themselves, overall creating a pentadentate donor sphere accommodating two metals [197]. As a result, several electrochemical oxidative events are apparent by CV. Ozerov developed monophosphinemonoimine-diarylamido ligand scaffolds that could form dinuclear palladium complexes, each bound to a redox-active pincer unit [198]. Depending on the length of the –(CH2)n linker between both imine units, electronic communication between both redox-active pockets upon oxidation by CV was either present (short) or absent (long). Vicic et al. examined the (redox-) chemistry of the Ni(CF3)(NNN) complex, with NNN being Nickamine originally successfully developed by Hu for Ni-catalysed cross-coupling of alkyl halides [199]. The trifluoromethyl complex Ni(CF3) (NNNMe) showed a quasi-reversible oxidation wave in CV at +0.80 V vs. Ag/Ag+ in THF, likely coupled to follow-up chemistry, as scan rate variations led to unsymmetrical current flow for the oxidation vs. the re-reduction process

Redox-Active Pincer Ligands

171

[200]. The electrochemical oxidation potential for this CF3 complex is equal in magnitude to that of its precursor, NiCl(NNNMe), which indicates that the locus of the redox is the ligand. Strikingly, the latter complex was reported to show irreversible oxidation behaviour in THF with NBu4PF6 as electrolyte [201]. DFT calculations support the proposition that the oxidation is ligand-centred, with the HOMO of the neutral complex being largely ligand-based and centred on the amido nitrogen and upon oxidation 41% spin density residing on the amido nitrogen in the cation [Ni (CF3)(NNNMe)]+. The Vicic group also reported on a series of Cu and Ni complexes with bis(oxazolinyl)amino (boxam) ligands [202]. For the Ni species, a dependency of oxidation potential on the co-ligand X in the series Ni(X)(boxam) was observed, with Eox ranging from 0.17 V for X ¼ Cl to +0.42 V for X ¼ CF3 (referenced against Fc/Fc+) and with varying levels of reversibility. Despite the likely intermediacy of substrate radical species in, e.g. the alkyl halide cross-coupling, with potential intermediacy of a NiIII species, no evidence for the redox involvement of these ligands in any catalytic process has been reported to date. The group of Gardinier has developed the chemistry of diarylamido systems with flanking pyrazole units featuring neutral β nitrogen donors for high-valent RhIII dichloride [203] as well as ReI-tricarbonyl [204] complexes. Furthermore, one homoleptic GaIII complex was reported, which exhibited weak electronic communication between both ligand fragments, proposedly via superexchange [205]. A series of homoleptic NiII-bis-ligand species were reported that had different parasubstituents in the 4 and 40 positions of the diarylamido backbone to further elucidate the effects that tailor the redox-chemistry, such as the (weak) electronic communication between ligands in ligand-mixed valent mono-oxidized derivatives, as deduced from the intervalence charge transfer (IVCT) transition [206]. In all cases, the combination of spectroscopic, electrochemical and computational data suggest quasi-reversible but localized redox-chemistry to occur strictly in the diarylamido backbone. Also a dinucleating analog was prepared and coordinated to Re(CO)3 as a bis-anionic fragment [207].

3.4

Neutral Pincer Systems

Although not strictly reporting on ligand-centred radical species, Agapie reported on bisphosphino-arene pincer-like PareneP species within a variety of mononuclear and dinuclear complexes. The central arene fragment was shown to show redox activity, for instance, in the installment of a dinuclear M2-polycarbonyl (M ¼ Fe, Co) core [208]. X-ray crystallographic determination of the solid-state structures revealed features consistent with reduction of the arene ring to a bis-allyl motif, concomitant with formal oxidation of the M2 core by two electrons to yield MI-MI fragments. More recently, based on this original PareneP motif, the group converted the central arene motif into a capped 1,2-catechol derivative that still proved amenable to partial, i.e. η2 or η4 arene coordination, to, e.g. Mo [209]. Reaction of the siloxane or catecholborane capped derivatives with O2 led to deprotection and oxidation of

172

J. I. van der Vlugt

the catecholate fragment to generate the corresponding quinone that showed η2C-Carene coordination to Mo. Follow-up work revealed that the redox-active hydroquinone-type motif Mo(CO)3(PareneOH2P) can be transformed into the quinone derivative PareneO2P through loss of two protons and four electrons to form Mo(CO)(Cl)2(PareneO2P) in a series of steps [210]. Proton-coupled electron transfer reactivity for this system was also established and thermochemical analysis of the –OH bonds established that Mo coordination weakens the bond dissociation free energy (BDFE) relative to the free ligand. The coordination chemistry and reactivity of the closely related 1,4-catechol-based pincer ligand design was explored with Pd [211]. Reaction of the free, fully protonated ligand with PdIICl2(cod) and subsequent halide abstraction provided cationic [PdCl(PareneOH2P)]OTf with a Pd-η2-arene interaction. In situ reduction of the Pd-dichlorido analog with a Ni0 species led to Pd0 buttressed by the phosphine donors only and no arene binding observable. However, reaction with the palladium precursor Pd acetate led to formation of the 1,4-quinone derivative bound in an all-carbon η4-fashion to Pd0. The interconversion from the former to the latter species was achieved by reaction with O2, which was studied in detail using time-resolved and temperature-dependent UV-vis spectroscopy and kinetics, as well as NO and N2O. The same group also reported on the related syn-9,10-anthracenediyl-linked bisphenol that supported a ZrIV(bisphenyl) complex that underwent photoinduced release of bibenzyl [212]. Metric data obtained from X-ray crystallography indicated that the anthracene unit has a lower degree of aromaticity in the central ring consistent with a two-electron reduction, thus resulting in η4 bonding of the central ring to Zr. Although formally considered a ZrIV complex, this system has two electrons stored within the anthracene unit, which sets it up for anthracene-based redox-chemistry. Analogous to known ZrII reactivity, the oxidative cyclometalation of alkynes was observed by coupling of two equiv of diphenyl acetylene or phenyl acetylene, giving zirconacyclopentadiene species. These complexes were able to insert CO into the metallacycle to obtain tetraphenylcyclopentadienone. Stoichiometric heterocoupling between phenyl acetylene and p-tolunitrile was also observed. This latter reaction could also be performed in a catalytic fashion using 5 mol% complex at 105 C to produce pyrimidines (i.e. two TolCN react with one PhCCH), suggesting that some reductive elimination step from the formed azazirconacyclic intermediate is also feasible to regenerate the ‘masked ZrII’ species. Mechanistic studies were employed to shed light on the likely order of events leading to the pyrimidine product.

4 Concluding Remarks In summary, ligand-centred redox activity has emerged as a recent addition to the portfolio when considering not only the design but also the desired reactivity inferred upon or by a pincer ligand, in connection with a (main group or) transition metal in its binding pocket. Several types of designs feature prominently in this

Redox-Active Pincer Ligands

173

respect, allowing for well-defined reductive or oxidative chemistry, depending on the type of backbone and redox-active fragments incorporated. The reversible storage of electrons or ‘holes’ (i.e. oxidizing equivalents) into an organic (pincer) backbone allows not only for transformations that would otherwise be very difficult or impossible but modulating the ligand oxidation state and, in effect the ligand field strength associated with it, also impacts, e.g. the Lewis acidity of or the induced spin state at the metal centre. These strategies have already been utilized for both selected stoichiometric and catalytic applications. It is deemed only a matter of time before new ligand designs will appear that may not only prove useful to investigate questions related to electronic structure of metal complexes but also be useful for other (catalytic) transformations, perhaps even excluding the need for transition metals altogether.

References 1. Li H, Gonçalves TP, Lupp D, Huang K-W (2019) ACS Catal 9:1619–1629 2. Verhoeven DGA, Moret M-E (2016) Dalton Trans 45:15762–15778 3. Khusnutdinova JR, Milstein D (2015) Angew Chem Int Ed 54:12236–12273 4. van der Vlugt JI (2012) Eur J Inorg Chem:363–375 5. van der Vlugt JI, Reek JNH (2009) Angew Chem Int Ed 48:8832–8846 6. Khan FF, Chowdhury AD, Lahiri GK (2020) Eur J Inorg Chem:11381–11146 7. van der Vlugt JI (2019) Chem A Eur J 25:2651–2662 8. Luca OR, Crabtree RH (2013) Chem Soc Rev 42:1440–1459 9. Praneeth VKK, Ringenberg MR, Ward Angew TR (2012) Chem Int Ed 51:10228–10234 10. Lyaskovskyy V, de Bruin B (2012) ACS Catal 2:270–279 11. Kaim W (2012) Eur J Inorg Chem:343–348 12. Chaudhuri P, Nazari Verani C, Bill E, Bothe E, Weyhermüller T, Wieghardt K (2001) J Am Chem Soc 123:2213–2223 13. Jørgensen CK (1966) Coord Chem Rev 1:164–178 14. Kaim W, Schwederski B (2010) Coord Chem Rev 254:1580–1588 15. Nikolaevskaya EN, Druzhkov NO, Syroeshkin MA, Egorov MP (2020) Coord Chem Rev 417:213353.1–213353.18 16. Caulton KG (2012) Eur J Inorg Chem:435–443 17. Alyea EC, Ferguson G, Restivo RJ (1975) Inorg Chem 14:2491–2495 18. Small BL, Brookhart M, Bennett AMA (1998) J Am Chem Soc 120:4049–4050 19. Britovsek GJP, Gibson VC, Kimberley BS, Maddox PJ, McTavish SJ, Solan GA, White AJP, Williams DJ (1998) Chem Commun:849–850 20. Gibson VC, Redshaw C, Solan GA (2007) Chem Rev 107:1745–1776 21. Small BL (2015) Acc Chem Res 48:2599–2611 22. de Bruin B, Bill E, Bothe E, Weyhermüller T, Wieghardt K (2000) Inorg Chem 39:2936–2947 23. Budzelaar PHM, de Bruin B, Gal AW, Wieghardt K, van Lenthe JH (2001) Inorg Chem 40:4649–4655 24. Braun JD, Gray PA, Sidhu BK, Nemez DB, Herbert DE (2020) Dalton Trans. https://doi.org/ 10.1039/d0dt00543f 25. Darmon JM, Turner ZR, Lobkovsky E, Chirik PJ (2012) Organometallics 31:2275–2285 26. Sandoval JJ, Melero C, Palma P, Álvarez E, Rodríguez-Delgado A, CÁmpora J (2016) Organometallics 35:3336–3343

174

J. I. van der Vlugt

27. Archer AM, Bouwkamp MW, Cortez M-P, Lobkovsky E, Chirik PJ (2006) Organometallics 25:4269–4279 28. Wile BM, Trovitch RJ, Bart SC, Tondreau AM, Lobkovsky E, Milsmann C, Bill E, Wieghardt K, Chirik PJ (2009) Inorg Chem 48:4190–4200 29. Russell SK, Milsmann C, Lobkovsky E, Weyhermüller T, Chirik PJ (2011) Inorg Chem 50:3159–3169 30. Reed BR, Stoian SA, Lord RL, Groysman S (2015) Chem Commun 51:6496–6499 31. Reed BR, Yousif M, Lord RL, McKinnon M, Rochford J, Groysman S (2017) Organometallics 36:582–593 32. Hojilla Atienza CC, Milsmann C, Semproni SP, Turner ZR, Chirik PJ (2013) Inorg Chem 52:5403–5417 33. Sugiyama H, Aharonian G, Gambarotta S, Yap GPA, Budzelaar PHM (2002) J Am Chem Soc 124:12268–12274 34. Rahimi N, Herbert DE, Budzelaar PHM (2019) Eur J Inorg Chem:780–786 35. Russell SK, Darmon JM, Lobkovsky E, Chirik PJ (2010) Inorg Chem 49:2782–2792 36. Delgado M, Ziegler JM, Seda T, Zakharov LN, Gilbertson JD (2016) Inorg Chem 55:555–557 37. Burns KT, Marks WR, Man Cheung P, Seda T, Zakharov LN, Gilbertson JD (2018) Inorg Chem 57:9601–9610 38. Chirik PJ (2017) Angew Chem Int Ed 56:5170–5181 39. Bart SC, Chłopek K, Bill E, Bouwkamp MW, Lobkovsky E, Neese F, Wieghardt K, Chirik PJ (2006) J Am Chem Soc 128:13901–13912 40. Bart SC, Lobkovsky E, Chirik PJ (2004) J Am Chem Soc 126:13794–13807 41. Darmon JM, Stieber SCE, Sylvester KT, Fern Ández I, Lobkovsky E, Semproni SP, Bill E, Wieghardt K, DeBeer S, Chirik PJ (2012) J Am Chem Soc 134:17125–17137 42. Reardon D, Conan F, Gambarotta S, Yap G, Wang Q (1999) J Am Chem Soc 121:9318–9325 43. Enright D, Gambarotta S, Yap GPA, Budzelaar PHM (2002) Angew Chem Int Ed 41:3873–3876 44. Knijnenburg Q, Gambarotta S, Budzelaar PHM (2006) Dalton Trans:5442–5448 45. Tondreau AM, Stieber SCE, Milsmann C, Lobkovsky E, Weyhermüller T, Semproni SP, Chirik PJ (2013) Inorg Chem 52:635–646 46. Arevalo R, Chirik PJ (2019) J Am Chem Soc 141:9106–9123 47. Blanchard S, Derat E, Desage-El Murr M, Fensterbank L, Malacria M, Mouriès-Mansuy V (2012) Eur J Inorg Chem:376–389 48. Hoyt JM, Sylvester KT, Semproni SP, Chirik PJ (2013) J Am Chem Soc 135:4862–4877 49. Hoyt JM, Schmidt VA, Tondreau AM, Chirik PJ (2015) Science 349:960–963 50. Joannou MV, Hoyt JM, Chirik PJ (2020) J Am Chem Soc 142:5314–5330 51. Bowman AC, Milsmann C, Bill E, Lobkovsky E, Weyhermüller T, Wieghardt K, Chirik PJ (2010) Inorg Chem 49:6110–6123 52. Zhu D, Thapa I, Korobkov I, Gambarotta S, Budzelaar PHM (2011) Inorg Chem 50:9879–9887 53. Bowman AC, Milsmann C, Hojilla Atienza CC, Lobkovsky E, Wieghardt K, Chirik PJ (2010) J Am Chem Soc 132:1676–1684 54. Zhu D, Korobkov I, Budzelaar PHM (2012) Organometallics 31:3958–3971 55. Schmidt VA, Hoyt JM, Margulieux GW, Chirik PJ (2015) J Am Chem Soc 137:7903–7914 56. Regenauer NI, Settele S, Bill E, Wadepohl H, Roşca D-A (2020) Inorg Chem 59:2604–2612 57. Drew MG, Nelson J, Esho F, McKee V, Nelson SM (1982) J Chem Soc Dalton Trans:1837–1843 58. Zhang S, Cui P, Liu T, Wang Q, Longo TJ, Thierer LM, Manor BC, Gau MR, Carroll PJ, Papaefthymiou GC, Tomson N (2020) Angew Chem Int Ed. https://doi.org/10.1002/anie. 202006391 59. Liu T, Gau MR, Tomson NC (2020) J Am Chem Soc 142:8142–8146 60. Zhang S, Wang Q, Thierer LM, Weberg AB, Gau MR, Carroll PJ, Tomson NC (2019) Inorg Chem 58:12234–12244

Redox-Active Pincer Ligands

175

61. Cui P, Wang Q, McCollom SP, Manor BC, Carroll PJ, Tomson NC (2017) Angew Chem Int Ed 56:15979–15983 62. Wang Q, Zhang S, Cui P, Weberg AB, Thierer LM, Manor BC, Gau MR, Carroll PJ, Tomson NC (2020) Inorg Chem 59:4200–4214 63. Arora V, Dutta M, Das K, Das B, Srivastava HK, Kumar A (2020) Organometallics. https:// doi.org/10.1021/acs.organomet.0c00233 64. Mukhopadhyay TK, Flores M, Groy TL, Trovitch RJ (2014) J Am Chem Soc 16:882–885 65. Mukhopadhyay TK, Rock CL, Hong M, Ashley DC, Groy TL, Baik M-H, Trovitch RJ (2017) J Am Chem Soc 139:4901–4915 66. Milsmann C, Turner Z, Semproni SP, Chirik PJ (2012) Angew Chem Int Ed 51:5386–5390 67. Vidyaratne I, Gambarotta S, Korobkov I, Budzelaar PHM (2005) Inorg Chem 44:1187–1189 68. Milsmann C, Semproni SP, Chirik PJ (2014) J Am Chem Soc 136:12099–12107 69. Margulieux GW, Turner ZR, Chirik PJ (2014) Angew Chem Int Ed 53:14211–14215 70. Gray PA, Braun JD, Rahimi N, Herbert DE (2020) Eur J Inorg Chem. https://doi.org/10.1002/ ejic.202000250asap 71. Noss ME, Wieder NL, Carroll PJ, Zdilla MJ, Berry DH (2018) Inorg Chem 57:7036–7043 72. Winter A, Schubert U (2020) ChemCatChem 12:2890–2941 73. Hamacher C, Hurkes N, Kaiser A, Klein A, Schüren A (2009) Inorg Chem 48:9947–9951 74. Morris DE, Hanck KW, Dearmond MK (1983) J Electroanal Chem 149:115–130 75. Scarborough CC, Lancaster KM, DeBeer S, Weyhermüller T, Sproules S, Wieghardt K (2012) Inorg Chem 51:3718–3732 76. England J, Scarborough CC, Weyhermüller T, Sproules S, Wieghardt K (2012) Eur J Inorg Chem 363:4605–4621 77. England J, Bill E, Weyhermüller T, Neese F, Atanasov M, Wieghardt K (2015) Inorg Chem 54:12002–12018 78. Wang M, Weyhermüller T, Bill E, Ye S, Wieghardt K (2016) Inorg Chem 55:5019–5039 79. Wang M, Römelt C, Weyhermüller T, Wieghardt K (2019) Inorg Chem 58:121–132 80. Klein J, Stuckmann A, Sobottka S, Suntrup L, van der Meer M, Hommes P, Reissig H-U, Sarkar B (2017) Chem A Eur J 23:12314–12325 81. Ma X, Suturina EA, De S, Négrier P, Rouzières M, Clérac R, Dechambenoit P (2018) Angew Chem Int Ed 57:7841–7845 82. Ciszewski JT, Mikhaylov DY, Holin KV, Kadirov MK, Budnikova YH, Sinyashin O, Vicic DA (2011) Inorg Chem 50:8630–8635 83. Jones GD, Martin JL, McFarland C, Allen OR, Hall RE, Haley AD, Brandon RJ, Konovalova T, Desrochers PJ, Pulay P, Vicic DA (2006) J Am Chem Soc 128:13175–13183 84. Zhang C-P, Wang H, Klein A, Biewer C, Stirnat K, Yamaguchi Y, Xu L, Gomez-Benitez V, Vicic DA (2013) J Am Chem Soc 135:8141–8144 85. Jones GD, McFarland C, Anderson TJ, Vicic DA (2005) Chem Commun:4211–4213 86. Anderson TJ, Jones GD, Vicic DA (2004) J Am Chem Soc 126:8100–8101 87. Tondreau AM, Hojilla Atienza CC, Darmon JM, Milsmann C, Hoyt HM, Weller KJ, Nye SA, Lewis KM, Boyer J, Delis JGP, Lobkovsky E, Chirik PJ (2012) Organometallics 31:4886–4893 88. Kamata K, Suzuki A, Nakai Y, Nakazawa H (2012) Organometallics 31:3825–3828 89. Palmer WN, Diao T, Pappas I, Chirik PJ (2015) ACS Catal 5:622–626 90. Léonard NG, Bezdek MJ, Chirik PJ (2017) Organometallics 36:142–150 91. Bezdek MJ, Guo S, Chirik PJ (2016) Inorg Chem 55:3117–3127 92. Rafiq S, Bezdek MJ, Chirik PJ, Scholes GD (2019) Chem 5:401–416 93. Gonell S, Assaf EA, Duffee KD, Schauer CK, Miller AJM (2020) J Am Chem Soc. https://doi. org/10.1021/jacs.0c02912 94. Harriman KLM, Leitch AA, Stoian SA, Habib F, Kneebone JL, Gorelsky SI, Korobkov I, Desgreniers S, Neidig ML, Hill S, Murugesu M, Brusso JL (2015) Dalton Trans 44:10516–10523 95. Samanta S, Ghosh P, Goswami S (2012) Dalton Trans 41:2213–2226

176

J. I. van der Vlugt

96. Sengupta D, Chowdhury NS, Samanta S, Ghosh P, Seth SK, Demeshko S, Meyer F, Goswami S (2015) Inorg Chem 54:11465–11476 97. Ghosh P, Samanta S, Roy SK, Demeshko S, Meyer F, Goswami S (2014) Inorg Chem 53:4678–4686 98. Sengupta D, Bhattacharjee R, Pramanick R, Rath SP, Chowdhury NS, Datta A, Goswami S (2016) Inorg Chem 55:9602–9610 99. Pramanick R, Bhattacharjee R, Sengupta D, Datta A, Goswami S (2018) Inorg Chem 57:6816–6824 100. Chakraborty M, Sengupta D, Saha T, Goswami S (2018) J Org Chem 83:7771–7778 101. Rath SP, Sengupta D, Ghosh P, Bhattacharjee R, Chakraborty M, Samanta S, Datta A, Goswami S (2018) Inorg Chem 57:11995–12009 102. Sinha S, Das S, Sikari R, Parua S, Brandaõ P, Demeshko S, Meyer F, Paul ND (2017) Inorg Chem 56:14084–14100 103. Das S, Sinha S, Jash U, Sikari R, Saha A, Barman SK, Brandão P, Paul ND (2018) Inorg Chem 57:5830–5841 104. Sinha S, Das S, Mondal R, Mandal S, Paul ND (2020) Dalton Trans. https://doi.org/10.1039/ d0dt00394h 105. Luca OR, Konezny SJ, Hunsinger GB, Müller P, Takase MK, Crabtree RH (2014) Polyhedron 82:2–6 106. Maciulis NA, Schaugaard RN, Losovyj Y, Chen C-H, Pink M, Caulton KG (2018) Inorg Chem 57:12671–12682 107. Tondreau AM, Darmon JM, Wile BM, Floyd SK, Lobkovsky E, Chirik PJ (2009) Organometallics 28:3928–3940 108. Schley ND, Fu GC (2014) J Am Chem Soc 136:16588–16593 109. McNeece AJ, Chang M-C, Filatov AS, Anderson JS (2018) Inorg Chem 57:7044–7050 110. Srimani D, Diskin-Posner Y, Ben-David Y, Milstein D (2013) Angew Chem Int Ed 52:14131–14134 111. Rohmann K, Kothe J, Haenel MW, Englert U, Hölscher M, Leitner W (2016) Angew Chem Int Ed 55:8966–8969 112. Daw P, Kumar A, Angel Espinosa-Jalapa N, Diskin-Posner Y, Ben-David Y, Milstein D (2018) ACS Catal 8:7734–7741 113. Daw P, Kumar A, Angel Espinosa-Jalapa N, Ben-David Y, Milstein D (2019) J Am Chem Soc 141:12202–12206 114. Daw P, Kumar A, Oren D, Espinosa-Jalapa NA, Srimani D, Diskin-Posner Y, Leitus G, Shimon LJW, Carmieli R, Ben-David Y, Milstein D (2020) Organometallics 39:279–285 115. Danopoulos AA, Wright JA, Motherwell WB (2005) Chem Commun:784–786 116. Yu RP, Darmon JM, Hoyt JM, Margulieux GW, Turner ZR, Chirik PJ (2012) ACS Catal 2:1760–1764 117. Darmon JM, Yu RP, Semproni SP, Turner ZR, Stieber SCE, DeBeer S, Chirik PJ (2014) Organometallics 33:5424–5433 118. Yu RP, Darmon JM, Milsmann C, Margulieux GW, Stieber SCE, DeBeer S, Chirik PJ (2013) J Am Chem Soc 135:13168–13184 119. Talukdar K, Issa A, Jurss JW (2019) Front Chem 7:330.1–330.11 120. Min S, Rasul S, Li H, Grills DC, Takanabe K, Li L-J, Huang K-W (2016) ChemPlusChem 81:166–171 121. Broere DLJ, Plessius R, van der Vlugt JI (2015) Chem Soc Rev 44:6886–6915 122. Ershova IV, Pisnukov AV (2020) Russ J Coord Chem 46:154–177 123. Pierpont CG (2011) Inorg Chem 50:9766–9772 124. Eisenberg R, Gray HB (2011) Inorg Chem 50:9741–9751 125. Sproules S, Wieghardt K (2011) Coord Chem Rev 255:837–860 126. Sproules S, Wieghardt K (2010) Coord Chem Rev 254:1358–1382 127. O’Reilly ME, Veige AS (2014) Chem Soc Rev 43:6325–6369 128. Munhá RF, Zarkesh RA, Heyduk AF (2013) Dalton Trans 42:3751–3766

Redox-Active Pincer Ligands

177

129. Heyduk AF, Zarkesh RA, Nguyen AI (2011) Inorg Chem 50:9849–9863 130. Zarkesh RA, Ziller JW, Heyduk AF (2008) Angew Chem Int Ed 47:4715–4718 131. Nguyen AI, Zarkesh RA, Lacy DC, Thorson MK, Heyduk AF (2011) Chem Sci 2:166–169 132. Nguyen AI, Blackmore KJ, Carter SM, Zarkesh RA, Heyduk AF (2009) J Am Chem Soc 131:3307–3316 133. MunhÁ RF, Zarkesh RA, Heyduk AF (2013) Inorg Chem 52:11244–11255 134. Ghosh S, Baik M-H (2015) Chem A Eur J 21:1780–1789 135. Hananouchi S, Krull BT, Ziller JW, Furche F, Heyduk AF (2014) Dalton Trans 43:17991–18000 136. Szigethy G, Shaffer DW, Heyduk AF (2012) Inorg Chem 51:12606–12618 137. Wong JL, Sánchez RH, Logan JG, Zarkesh RA, Ziller JW, Heyduk AF (2013) Chem Sci 4:1906–1910 138. Shaffer DW, Szigethy G, Ziller JW, Heyduk AF (2013) Inorg Chem 52:2110–2118 139. Wong JL, Higgins RF, Bhowmick I, Cao DX, Szigethy G, Ziller JW, Shores MP, Heyduk AF (2016) Chem Sci 7:1594–1599 140. Rosenkoetter KE, Ziller JW, Heyduk AF (2016) Inorg Chem 55:6794–6798 141. Rosenkoetter KE, Ziller JW, Heyduk AF (2017) Dalton Trans 46:5503–5507 142. Rosenkoetter KE, Wojnar MK, Charette BJ, Ziller JW, Heyduk AF (2018) Inorg Chem 57:9728–9737 143. Wojnar MK, Ziller JW, Heyduk AF (2017) Eur J Inorg Chem:5571–5575 144. Mouchel Dit Leguerrier D, Barré R, Bryden M, Imbert D, Philouze C, Jarjayes O, Luneau D, Molloy JK, Thomas F (2020) Dalton Trans. https://doi.org/10.1039/d0dt01388a 145. Tomat E (2016) Comm Inorg Chem 36:327–342 146. Gautam R, Loughrey JJ, Astashkin AV, Shearer J, Tomat E (2015) Angew Chem Int Ed 54:14894–14897 147. Gautam R, Astashkin AV, Chang TM, Shearer J, Tomat E (2017) Inorg Chem 56:6755–6762 148. Daneshmand P, Fortun S, Schaper F (2017) Organometallics 36:3860–3877 149. Daneshmand P, Michalsky I, Aguiar PM, Schaper F (2018) Dalton Trans 47:16279–16291 150. Chang M-C, McNeece AJ, Hill EA, Filatov AS, Anderson JS (2018) Chem A Eur J 24:8001–8008 151. Chang M-C, Jesse KA, Filatov AS, Anderson JS (2019) Chem Sci 10:1360–1367 152. McNeece AJ, Jesse KA, Xie J, Filatov AS, Anderson JS (2020) J Am Chem Soc. https://doi. org/10.1021/jacs.0c03244 153. Broere DLJ, de Bruin B, Reek JNH, Lutz M, Dechert S, van der Vlugt JI (2014) J Am Chem Soc 136:11574–11577 154. Broere DLJ, Metz LL, de Bruin B, Reek JNH, Siegler MA, van der Vlugt JI (2015) Angew Chem Int Ed 54:1516–1520 155. Broere DLJ, van Leest NP, de Bruin B, Siegler MA, van der Vlugt JI (2016) Inorg Chem 55:8603–8611 156. Broere DLJ, Demeshko S, de Bruin B, Pidko EA, Reek JNH, Siegler MA, Lutz M, van der Vlugt JI (2015) Chem A Eur J 21:5879–5886 157. Broere DLJ, Plessius R, Tory J, Demeshko S, de Bruin B, Siegler MA, Hartl F, van der Vlugt JI (2016) Chem A Eur J 22:13965–13975 158. Broere DLJ, Modder DK, Blokker E, Siegler MA, van der Vlugt JI (2016) Angew Chem Int Ed 55:2406–2410 159. Bagh B, Broere DLJ, Siegler MA, van der Vlugt JI (2016) Angew Chem Int Ed 55:8381–8385 160. Bagh B, Broere DLJ, Sinha V, Kuijpers PF, van Leest NP, de Bruin B, Demeshko S, Siegler MA, van der Vlugt JI (2017) J Am Chem Soc 139:5117–5124 161. Bellemin-Laponnaz S, Welter R, Brelot L, Dagorne S (2009) J Organomet Chem 694:604–606 162. Borré E, Dahm G, Aliprandi A, Mauro M, Dagorne S, Bellemin-Laponnaz S (2014) Organometallics 33:4374–4384 163. Romain C, Choua S, Collin J-P, Heinrich M, Bailly C, Karmazin-Brelot L, BelleminLaponnaz S, Dagorne S (2014) Inorg Chem 53:7371–7376

178

J. I. van der Vlugt

164. Weinberg DR, Hazari N, Labinger JA, Bercaw JE (2010) Organometallics 29:89–100 165. Harris CF, Bayless MB, van Leest NP, Bruch QJ, Livesay BN, Bacsa J, Hardcastle KI, Shores MP, de Bruin B, Soper JD (2017) Inorg Chem 56:12421–12435 166. Blackmore KJ, Lal N, Ziller JW, Heyduk AF (2009) Eur J Inorg Chem:735–743 167. Harris CF, Kuehner CS, Bacsa J, Soper JD (2018) Angew Chem Int Ed 57:1311–1315 168. Baltrun M, Watt FA, Schoch R, Wölper C, Neuba AG, Hohloch S (2019) Dalton Trans 48:14611–14625 169. Ghorui T, Roy S, Pramanik S, Pramanik K (2016) Dalton Trans 45:5720–5729 170. Comanescu CC, Iluc VM (2014) Organometallics 33:6059–6064 171. Burford RJ, Piers WE, Parvez M (2012) Organometallics 31:2949–2952 172. Gutsulyak DV, Piers WE, Borau-Garcia J, Parvez M (2013) J Am Chem Soc 134:11776–11779 173. LaPierre EA, Piers WE, Spasyuk DM, Bi DW (2016) Chem Commun 52:1361–1364 174. Comanescu CC, Vyushkovab M, Iluc VM (2015) Chem Sci 6:4570–4579 175. Deziel AP, Hoffbauer MR, Iluc VM (2019) Organometallics 38:879–887 176. Cui P, Iluc VM (2015) Chem Sci 6:7343–7354 177. Cui P, Hoffbauer MR, Vyushkova M, Iluc VM (2019) Dalton Trans 48:9663–9668 178. Liang L-C, Lin J-M, Hung C-H (2003) Organometallics 22:3007–3009 179. Winter AM, Eichele K, Mack H-G, Potuznik S, Mayer HA, Kaska WC (2003) J Organomet Chem 682:149–154 180. Adhikari D, Mossin S, Basuli F, Huffman JC, Szilagyi RK, Meyer K, Mindiola DJ (2008) J Am Chem Soc 130:3676–3682 181. Vreeken V, Siegler MA, de Bruin B, Reek JNH, Lutz M, van der Vlugt JI (2015) Angew Chem Int Ed 54:7055–7059 182. Vreeken V, Baij L, de Bruin B, Siegler MA, van der Vlugt JI (2017) Dalton Trans 46:7145–7149 183. Fout AR, Basuli F, Fan H, Tomaszewski J, Huffman JC, Baik M-H, Mindiola DJ (2006) Angew Chem Int Ed 45:3291–3295 184. Adhikari D, Mossin S, Basuli F, Dible BR, Chipara M, Fan H, Huffman JC, Meyer K, Mindiola DJ (2008) Inorg Chem 47:10479–10490 185. Radosevich AT, Melnick JG, Stoian SA, Bacciu D, Chen C-H, Foxman BM, Ozerov OV, Nocera DG (2009) Inorg Chem 48:9214–9221 186. Mankad NP, Harkins SB, Antholine WE, Peters JC (2009) Inorg Chem 48:7026–7032 187. Harkins SB, Peters JC (2005) J Am Chem Soc 127:2030–2031 188. Harkins SB, Mankad NP, Miller AJM, Szilagyi RK, Peters JC (2008) J Am Chem Soc 130:3478–3485 189. DeMott JC, Basuli F, Kilgore UHJ, Foxman BM, Huffman JC, Ozerov OV, Mindiola DJ (2007) Inorg Chem 46:6271–6276 190. Vreeken V, Broere DLJ, Jans ACH, Lankelma M, Reek JNH, Siegler MA, van der Vlugt JI (2016) Angew Chem Int Ed 55:10042–10046 191. Vreeken V, Siegler MA, van der Vlugt JI (2017) Chem A Eur J 23:5585 192. Alig L, Fritz M, Schneider S (2019) Chem Rev 119:2681–2751 193. Schneider S, Meiners J, Askevold B (2012) Eur J Inorg Chem:412–429 194. Schneck F, Finger M, Tromp M, Schneider S (2017) Chem A Eur J 23:33–37 195. Lagaditis PO, Schluschaß B, Demeshko S, Würtele C, Schneider S (2016) Inorg Chem 55:4529–4536 196. Davidson JJ, DeMott JC, Douvris C, Fafard CM, Bhuvanesh N, Chen C-H, Herbert DE, Lee C-I, McCulloch BJ, Foxman BM, Ozerov OV (2015) Inorg Chem 54:2916–2935 197. Wang D, Lindeman SV, Fiedler AT (2015) Inorg Chem 54:8744–8754 198. Herbert DE, Ozerov OV (2011) Organometallics 30:6641–6654 199. Csok Z, Vechorkin O, Harkins SB, Scopelliti R, Hu X (2008) J Am Chem Soc 130:8156–8157 200. Madhira VN, Ren P, Vechorkin O, Hu X, Vicic DA (2012) Dalton Trans 41:7915–7919 201. Vechorkin O, Csok Z, Scopelliti R, Hu X (2009) Chem A Eur J 15:3889–3899

Redox-Active Pincer Ligands

179

202. Kieltsch I, Dubinina GG, Hamacher C, Kaiser A, Torres-Nieto J, Hutchison JM, Klein A, Budnikova Y, Vicic DA (2010) Organometallics 29:1451–1456 203. Wanniarachchi S, Liddle BJ, Kizer B, Hewage JS, Lindeman SV, Gardinier JR (2012) Inorg Chem 51:10572–10580 204. Wanniarachchi S, Liddle BJ, Toussaint J, Lindeman SV, Bennett B, Gardinier JR (2010) Dalton Trans 39:3167–3169 205. Liddle BJ, Wanniarachchi S, Hewage JS, Lindeman SV, Bennett B, Gardinier JR (2012) Inorg Chem 51:12720–12728 206. Hewage JS, Wanniarachchi S, Morin TJ, Liddle BJ, Banaszynski M, Lindeman SV, Bennett B, Gardinier JR (2014) Inorg Chem 53:10070–10084 207. Gardinier JR, Hewage JS, Bennett B, Wang D, Lindeman SV (2018) Organometallics 37:989–1000 208. Horak KT, Velian A, Day MW, Agapie T (2014) Chem Commun 50:4427–4429 209. Henthorn JT, Lin S, Agapie T (2015) J Am Chem Soc 137:1458–1464 210. Henthorn JT, Agapie T (2016) Inorg Chem 55:5337–5342 211. Horak KT, Agapie T (2016) J Am Chem Soc 138:3443–3452 212. Low CH, Rosenberg JN, Lopez MA, Agapie T (2018) J Am Chem Soc 140:11906–11910

Top Organomet Chem (2021) 68: 181–226 https://doi.org/10.1007/3418_2020_61 # Springer Nature Switzerland AG 2020 Published online: 1 December 2020

A Pincer Motif Etched into a meta-Benziporphyrin Frame Karolina Hurej and Lechosław Latos-Grażyński

Contents 1 The Convolution of Pincer Ligands and m-Benziporphyrins: Outlook . . . . . . . . . . . . . . . . . . . . 2 The Synthesis of m-Benziporphyrin and Aza-m-Benziporphyrins: Toward a Pincer-Like Macrocyclic Ligand Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 m-Benziporphyrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 m-Benziporphodimethenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 meso-Alkylidenyl-m-Benziporphyrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 22-Alkyl-m-Benziporphyrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Aza-m-Benziporphyrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 m-Benziporphyrins: Transformations of m-Phenylene Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Inner Core Transformations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Perimeter Modifications: Hydroxy and Alkoxy Substituents . . . . . . . . . . . . . . . . . . . . . . . . 4 m-Benziporphyrins and Aza-m-Benziporphyrins: Coordination Chemistry . . . . . . . . . . . . . . . 4.1 Coordination Motifs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Inner Core Reactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 C–H Bond Activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 C–C Bond Activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Contraction of m-Phenylene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Contraction: General Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

182 186 186 188 189 190 191 197 197 199 202 202 208 210 213 214 218 219 222

Abstract The incorporation of a meta-phenylene moiety into a β-alkylated or mesotetraarylporphyrin framework resulted in the formation of m-benziporphyrins. Their molecular design preserves all the essential virtues of the original tetrapyrrolic architecture of regular porphyrin, including the perfect match between the ionic radii of the inserted metal cation and the size of the macrocyclic (CNNN) core, and steric protection provided by thoughtfully chosen β-alkyl or meso-aryl substituents.

K. Hurej and L. Latos-Grażyński (*) Department of Chemistry, University of Wrocław, Wrocław, Poland e-mail: [email protected]; [email protected]; http://llg.chem.uni.wroc.pl/

182

K. Hurej and L. Latos-Grażyński

The analogous incorporation of a pyridine moiety yielded a class of pyriporphyrins. The organometallic derivatives of m-benziporphyrins and pyriporphyrins are reminiscent of the large family of pincer ligand complexes, in which the metal–arene bond is supported by two amine or phosphine arms. This chapter relates the chemistry of the pincer ligand and that of the m-benziporphyrin (aza-mbenziporphyrin) emphasizing the fact that both groups have been structurally and functionally based on common dominators, i.e., on m-phenylene (aza-m-phenylene) rings. Keywords Carbaporphyrin · Contraction · Metalloporphyrins · Pincer ligand · Porphyrin

1 The Convolution of Pincer Ligands and m-Benziporphyrins: Outlook The multiple biochemical roles played by iron protoporphyrin IX have inspired many diverse studies of fundamental scientific significance, triggering a search for suitable porphyrins and metalloporphyrins prearranged to act as biomimetic models. The activation of dioxygen, oxygen atom transfer, and the direct functionalization of C–H bonds via metal-catalyzed atom/group transfer reactions are ground-breaking, attractive areas of exploration. All of them require porphyrin or metalloporphyrin participation. In general, advances in metalloporphyrin chemistry are evidently convoluted with developments in porphyrin synthetic methods. In light of these, it is quite clear why chemists in sophisticated explorations in principle mostly apply meso-arylsubstituted porphyrin 1 and its complexes 1-M as their fundamental macrocycle. Evidently, these molecules preserve the essential features of natural porphyrins. Fortunately, they can be easily synthesized in impressive varieties and readily manipulated by virtue of the appropriate external substitution, which in addition provides the required steric protection, taking on the shape of a porphyrinic cauldron to enable, for example, control over catalytic stereoselectivity. Along the road, a unique class of core-modified porphyrins emerged, founded on the replacement of a single pyrrole ring in the parent meso-aryl-substituted porphyrin 1 by a carbocycle or heterocarbocyle [1–11], including meso-tetraaryl-2-aza-21carbaporphyrin (N-confused porphyrin) 2 [12, 13] and meso-tetraaryl 21-carbaporphyrin 3 (Scheme 1) [14]. The well-defined subclass of carbaporphyrinoids has been exemplified here by meso-tetraaryl-m-benziporphyrin 5 and alternatively by its appropriate complexes 5-M, which are the fundamental structures for this review [2, 5]. The isomeric meso-tetraaryl-p-benziporphyrin 4 will be only briefly addressed particularly in relation to contraction processes [4, 15– 20]. The related 3-aza and 22-aza derivatives, i.e., meso-tetraaryl-22-aza-m-

A Pincer Motif Etched into a meta-Benziporphyrin Frame

Ar

Ar N

H N

N

Ar

N H

Ar

NH

N

183

Ar

Ar NH

HN N

Ar

1

H N

3 Ar

Ar N

N H N Ar

Ar

Ar

4 Ar

Ar

2

N

N

Ar

Ar

Ar

Ar

Ar

Ar

N

Ar

Ar

HN

5

N Ar

Ar

N

Ar

Ar

N N

N

N

H N Ar

Ar

6

N

N

H N

N H N

Ar

Ar

7

8

Scheme 1 meso-Tetraarylporphyrin 1 and derivatives: meso-tetraaryl-2-aza-21-carbaporphyrin (N-confused porphyrin) 2, meso-tetraaryl-21-carbaporphyrin 3 and their m-phenylene analogues: meso-tetraaryl-p-benziporphyrin 4, meso-tetraaryl-m-benziporphyrin 5, meso-tetraaryl-22-aza-mbenziporphyrin (meso-tetraarylpyriporphyrin) 6, meso-tetraaryl-3-aza-m-benziporphyrin 7, and meso-diaryl-3-aza-m-benziporphyrin 8

benziporphyrin (meso-tetraarylpyriporphyrin) 3 and meso-tetraaryl-3-aza-mbenziporphyrin (N-confused meso-tetraarylpyriporphyrin) 4, are of interest as well [21, 22]. The occurrence of a carbon atom in the inner cavity of 5, 7, 8 provides the suitable (CNNN) surrounding, which enables creating a specific interaction of embedded mphenylene with inserted diamagnetic or paramagnetic metal cations [4, 9]. Importantly, the properties of organometallic complexes are controlled by suitable choices of electronic factors and steric protection provided by the cautiously selected external substitution of 5. It has been revealed that the coordination chemistry of m-benziporphyrins takes several routes, which differ in the role of the arene moiety [2, 16, 23–26]. In the context of the general topic assigned to this volume, it is essential to recall that the organometallic derivatives of m-benziporphyrin resemble the well-developed family of pincer ligand complexes, in which the metal–arene bond is supported by two amine, sulfide, or phosphine arms 9–11 (Scheme 2) [27– 30]. Transition metal complexes containing the pincer-type ligands have attracted scientifically well-justified interest [31–44]. The peculiar properties of the metal centers have been efficiently enforced by pincer ligands. In organometallic complexes containing a direct transition metal– carbon bond, chelation leads to the formation of metallacycles, which provide the

184

K. Hurej and L. Latos-Grażyński

N

N

M N

5-M O

O N

M

S

N

M

P

S

X

X

9

M

P

X

11

10 +

X S O

Pd X O

12

II

S O

H

N

III

Cu

N

H

H

N

II

Cu

13-Cu(III)

H

N

N R

N

R

13-Cu(II)

Scheme 2 Reminiscence of pincer ligand complexes to metallo-m-benziporphyrins

effective stabilization of the M–C bond. Modifications of various ligand parameters resulted in the fine-tuning of the steric and the electronic properties of the metal center preserving the essential coordination frame. Extensive exploration has provided direct access to fundamental comprehension of a variety of reactions in organometallic chemistry. A range of applications (e.g., alkane dehydrogenation catalysts, the mechanistic elucidation of fundamental transformations C–C bond activation, the construction of the first metallodendrimers for sustainable homogeneous catalysis, sensors) have also been developed [45–47]. Essential concepts – nicely explored for pyridine- or acridine-based ruthenium, platinum, or iridium pincer complexes – have emerged. Milstein and coworkers consider this “a new paradigm in bond activation and ‘green’ catalysis” [35, 36]. In these catalytic systems, pincer ligands cooperate by means of the aromatization–dearomatization of the built-in heteroaromatic moiety. Significantly, bond activation takes place with no formal change in the metal oxidation state affording the activation of select single bonds (H–H, C–H, O–H, and N–H). For instance, a unique water splitting process was catalyzed by a pyridine-based pincer ruthenium complex [48]. To put this review into proper perspective, one has to note that organometallic chemistry in a macrocyclic environment is still evidently limited to carbaporphyrinoids. The rare examples are Loeb’s organopalladium crown ether complexes 13 [49] and copper(n) triazamacrocyclic complexes 13-Cu(III) and

A Pincer Motif Etched into a meta-Benziporphyrin Frame

185 3

Z

2

X N

M N

N

β position meso position X = N, CH; Z = N, CH M = Fe, Pd, Rh

4

21

6 22

18 19

N 25 16

8

23 N

H 24 N

14

9 11

13

Scheme 3 Controlled alterations (left) and numbering of m-benziporphyrin

13-Cu(II) (classified also as copper macrocyclic pincer complexes or macrocyclic arene complexes) [27, 50], which exemplify some aspects of organometallic chemistry in the macrocyclic environment as well as reveal the potential enclosed in this area. For instance, aryl-copper(III)-halide complexes 13-Cu(III) closely resemble elusive intermediates invoked in catalytic reactions, such as Ullmann cross-coupling [51]. A detailed study covered the key mechanistic aspects of a mild C–H activation process conducted by triazamacrocyclic copper(II) complexes revealing the intermediate locating an aromatic C–H bond in close proximity to metal center 13-Cu(II). The copper-catalyzed aerobic oxidative functionalization of an arene C–H bond has been detected as well [52]. Thus, the fusion of macrocyclic motifs with the NCN or NNN functionality (Schemes 2 and 3) offers a promising platform for studying the activation of C–H and C–C bonds. Eventually, the requested properties were tuned by the appropriate core modifications and meso- or β-substitutions (Scheme 3). The present chapter summarizes published work on m-benziporphyrins and aza-m-benziporphyrins 5–8 resorting, however, only to arbitrarily chosen most representative examples. Such a selection makes it possible to appreciate the common virtues of pincer ligand and carbaporphyrin chemistry. In particular, the intention is to outline the impact on organometallic chemistry in rationally confined environments. Specifically, the synthetic strategy and physicochemical characterization of m-benziporphyrin have been illustrated. We have also focused on the coordinating properties of m-benziporphyrins and their specific transformation triggered by the imposed metal cation–m-phenylene interactions. The presentation is split into three complementary parts. The first section covers the synthesis of β-alkylated and meso-tetraaryl-substituted m-benziporphyrins, the second explores the modifications of the m-phenylene moiety, whereas the final part describes the coordinating properties of m-benziporphyrins and addresses some issues of peculiar metal-triggered inner core reactivity.

186

K. Hurej and L. Latos-Grażyński

2 The Synthesis of m-Benziporphyrin and Aza-m-Benziporphyrins: Toward a Pincer-Like Macrocyclic Ligand Library 2.1 2.1.1

m-Benziporphyrins β-Substituted m-Benziporphyrin

R

R

R

R

R

R

The first example of β-substituted m-benziporphyrin 14 was reported by Berlin and Breitmaier in 1994 [1]. In their work, evidently pioneering for carbaporphyrinoid chemistry, they applied the [3 + 1] methodology. Thus, β-substituted m-benziporphyrin 16 was obtained by the condensation of isophtalaldehyde 14 with tripyrrane 15, using HBr as a catalyst with a 5.9% yield [1]. Over time, this method was optimized and HBr was substituted by TFA and chloranil by DDQ, increasing the yield to 28% (Scheme 4) [53]. An alternative synthetic route of β-alkylated m-benziporphyrins starting from isophthaloyl chloride was also developed [54]. This macrocycle is nonaromatic. It belongs to a group of carbaporphyrinoids, in which the local aromaticity of the built-in m-phenylene unit dominates, interrupting the macrocyclic π-delocalization pathway [3, 4, 9, 53, 54]. However, as an effect of protonation, the positive charge located on the six-membered ring of the dicationic 16-H22+ prompts the appearance of a weak diatropic ring current, as reflected by the appropriate 1H NMR pattern (Scheme 5) [7, 9, 53, 54]. It is consistent with the significant participation of the aromatic contributor 160 -H22+ in the description of the electronic structure (in this chapter, aromatic or antiaromatic π-delocalization routes are marked by bold black lines).

Scheme 4 Synthesis of βalkylated m-benziporphyrin 16

CHO OHC

1. H+ 2. [O]

14 + HOOC HOOC

HN NH HN

15

N

H N

16

N

A Pincer Motif Etched into a meta-Benziporphyrin Frame

187

TFA N

H N

N

NH

H N

HN

NH

16'-H22+

16

H N

HN

16''-H22+

Scheme 5 Protonation of β-substituted m-benziporphyrin 16

2.1.2

meso-Tetraaryl-m-Benziporphyrin R

R

R

R

meso-Tetraaryl-m-benziporphyrin 5 was obtained (2001) using the stochastic method [2]. The synthesis involved the condensation of dicarbinol 18, formed from isophthalaldehyde 17, with pyrrole and the corresponding aryl aldehyde (Scheme 6). The choice of meso-aryl substituents determined the yield, which oscillated in the 15–17% range. The syntheses of derivatives substituted at the external m-phenylene position were subsequently reported [55]. meso-Tetraaryl-m-benziporphyrin 5, similarly as the β-substituted contender 16, displays the 1H NMR pattern, which is characteristic of nonaromatic porphyrinoids. The m-phenylene units preserve a benzene-like electronic structure, being efficiently isolated from the π-electron delocalization of the accompanying tripyrrolic brace (Scheme 7). Thus, the absence of macrocyclic aromaticity in m-benziporphyrins reflects a mismatch between the porphyrin-like and benzene-like π-delocalization

pirol, Ph

CHO PhMgBr OHC

17

Ph

ArCHO, BF3. Et2O

Ph N

OH OH 18

Ph

H N

Ar

Ar

5

Scheme 6 Synthesis of meso-tetraaryl-m-benziporphyrin 5

N

188

K. Hurej and L. Latos-Grażyński

Ar

Ar

Ar N

H N

H+

N

NH

Ar

Ar

Ar

H N

Ar

HN

Ar NH

Ar

Ar

HN Ar

Ar

5'-H22+

5

H N

5"-H22+

Scheme 7 Protonation of meso-substituted m-benziporphyrin 5

Ar

Ar N

H N

alcohol, acid or hydride

N

Ar

HN

NH N

Ar

Ar

R

Ar

Ar

Ar

5

19 R= OH, OMe, H

Scheme 8 m-Benziphlorin 19 formation

modes [2, 3, 5]. However, after the addition of acid, a dication forms, exhibiting weak diatropic properties consistent with the 500 -H22+ electronic structure. Macrocycle 5 is puckered, acquiring a canted m-phenylene conformation with respect to the N3 plane. Characteristically, the 1H NMR spectra of meso-tetraaryl-mbenziporphyrin 2 reflect features of an effectively planar structure indicating the conformational equilibrium between two energetically equivalent conformers, differentiated by the direction of the phenylene ring canting [56]. m-Benziporphyrin 5 easily undergoes a conversion to phlorin 19 as a result of water, alcohol, or dihydrogen additions (Scheme 8) [2]. Using the [3 + 1] methodology, 24-hetero-m-benziporphyrins, belonging to the β-substituted or meso-substituted series, were obtained with heteroatoms: oxygen or sulfur [57, 58]. Their diprotonation converts nonaromatic 24-hetero-mbenziporphyrins into the dicationic diatropic species.

2.2

m-Benziporphodimethenes

R2

R1

R1 R2

R3

R3

A Pincer Motif Etched into a meta-Benziporphyrin Frame Scheme 9 Synthesis of m-benziporphodimethene 21

189

R

R R OH

Ph OH

pyrrole, ArCHO, BF3.Et2O

R

N

20

R

R

H N

N Ar

Ar

R = Ph, Me

R

21

m-Benziporphodimethenes 21 are m-benziporphyrin analogues but with two sp3hybridized meso-carbon atoms adjacent to the m-phenylene unit. The original synthesis was based on the reaction between the proper dicarbinol 20 (R ¼ Aryl), pyrrole, and arylaldehyde (Scheme 9) [16]. Subsequently, the methyl derivatives (R ¼ Me) were explored [59–61]. It was also documented that the tetraethyl derivative can potentially be used as long-wavelength zinc(II)-specific chemosensors [59]. The zinc(II) complexes of 21 (R ¼ Ph) acted as an interesting building block capable of self-assembling and continually growing into multidimensional nanostructure arrays [62].

2.3

meso-Alkylidenyl-m-Benziporphyrins R1

R3

R1

R3

Another derivate of m-benziporphyrin – a macrocycle with an exocyclic double bond – has been reported by Lee and coworkers [11, 63]. The first step of 23 synthetic pathways is based on the reaction bis-malonate derivative of isophthalaldehyde with pyrrole giving 22. Next, benzitripyrrane 22 undergoes condensation with pyrrole and pentafluorobenzaldehyde resulting in mesoalkylidene benziporphyrin 23, as well as expanded macrocycles, such as benzipentaphyrin and dibenzioctaphyrin (Scheme 10). Benzitripyrrane 22 also reacts with furan or thiophene dicarbinol 24, generating meso-alkylidene heterobenziporphyrins: oxabenziporphyrin 25-O and thiabenziporphyrin 25-S, respectively (Scheme 11).

190

K. Hurej and L. Latos-Grażyński EtO2C EtO2C EtO2C

EtO2C CO2Et 1. pyrrole, ArCHO, TFA 2. DDQ

CO2Et N

HN NH

CO2Et

EtO2C

22

H N

N Ar

Ar

23

Scheme 10 Synthesis of meso-alkylidenyl-m-benziporphyrin 23 [64–66]

EtO2C EtO2C EtO2C

X = S or O

EtO2C CO2Et

1. TFA 2. DDQ

CO2Et

22 HO Ar

X

HO

Ar

Ar

Ar

X

N

N

HN NH

CO2Et

EtO2C

25 24

Scheme 11 Synthesis of meso-alkylidenyl-24-hetero-m-benziporphyrin 25 [67]

2.4

22-Alkyl-m-Benziporphyrins R1

R1 R

R3

R3

5,20-Diphenyl-10,15-ditolyl-22-methyl-m-benziporphyrin 29-Me was formed in a several-step process starting from 2,6-dicyanotoluene 26-Me [24]. The first step involved the reduction of two nitriles to imines and hydrolysis to form 27-Me (Scheme 12). The subsequent reaction with phenylmagnesium bromide afforded the crucial diol 28-Me. The precursor 28-Me was used in the typical stochastic condensation with pyrrole and p-tolylaldehyde, giving 29-Me with a 22% yield. The conditions of the last two steps were analogous to the synthetic method applied for m-benziporphyrin [2]. The higher yield for 29-Me as compared to 5 can be due to the lack of phlorin formation during this condensation. In the synthesis of 22-ethyl-m-benziporphyrin 29-Et, 2-ethylisophthalonitrile 26-Et was used instead of 2-methylisophthalonitrile 26-Me [25]. Initially, isophthalonitrile 26-H reacted with ethyl iodide yielding 26-Et almost quantitatively. The further procedure followed the analogous path already elaborated for 22-methyl-m-benziporphyrin 29-Me (Scheme 12).

A Pincer Motif Etched into a meta-Benziporphyrin Frame

191

Ph

Ph a N

R

26-R

N

c

b O

O

HO

R

OH Ph

27-R

R

28-R

N

Ph Tol

R H N

N Tol

29-R

Scheme 12 Synthesis of 22-alkyl-m-benziporphyrins 29-R: (a) DIBAH, 4 h, 183 K, (b) PhMgBr, 1 h, 293 K, (c) pyrrole, ArCHO, BF3 Et2O, DDQ,1 h. R ¼ Me, Et

2.5

Aza-m-Benziporphyrins

Aza-m-benziporphyrins – pyriporphyrins [68] – are structural analogues of mbenziporphyrin, in which the properly oriented pyridine motif replaces the benzene unit affording three isomeric structures: 22-aza-m-benziporphyrin 6 [22], 3-aza-mbenziporphyrin 7 [21], and 2-aza-m-benziporphyrin 8 [69]. 7 and 8 preserve the carbaporphyrinoid features conserving the characteristic (CNNN) set of inner core donor atoms, whereas 6 contains the (NNNN) porphyrin-like donor set.

2.5.1

3-Aza-m-Benziporphyrin N R

R

R

R

The condensation of 3,5-bis[2-pyrrolyl(aryl)methyl] pyridine (pyritripyrrane) 30, pyrrole, and aryl aldehyde in dichloromethane, catalyzed by TFA and followed by oxidation with DDQ, resulted in the formation of meso-tetraryl-3-aza-mbenziporphyrin 7 [21]. The strict [3 + 1] methodology was applied to afford mesotetraaryl-24-thia-3-aza-m-benziporphyrin (N-confused 21-thiapyriporphyrin) 7-S (Scheme 13). The condensation of pyritripyrrane 30 with 2,5-bis [arylhydroxymethyl]thiophene 24 in dichloromethane catalyzed by TFA and followed by oxidation with DDQ afforded 7-S. The yields varied in the 2.5–18% range depending on the meso substitution. The protonation of 7 led directly to the dicationic species 7-H22+(Scheme 14). The feasible monocations 7-HA+ or 7-HB+ were not detected in the course of the 1H NMR titration. The tricationic structure 7-H33+ was observed at a very high concentration of acid.

192

K. Hurej and L. Latos-Grażyński N

N

Ar1 HN

Ar1

+

NH

Ar1

Ar2

S

N

N

HO

HO

30

Ar1

1. TFA 2. DDQ

S Ar2

Ar2

24

Ar2

7-S

Scheme 13 Synthesis of meso-tetraryl-24-thia-3-aza-m-benziporphyrin 7-S

N Ar1 NH

N Ar1

Ar1 N

H N

H

+

N

H N

Ar2

Ar2

7-HA

N H Ar2

Ar2

Ar1

Ar1

7

Ar1

Ar1

Ar1

Ar1

H+ NH Ar1

H+

H N

HN

NH Ar2

Ar2

7-H22+

HN

NH

H N

N

+

N

+

H+

H N

HN Ar2

Ar2

7-H33+

N Ar2

Ar2

7-HB+

Scheme 14 The protonation of meso-substituted 3-aza-m-benziporphyrin 7

Unexpectedly, the protonation of N-confused 21-thiapyriporphyrin 7-S did not follow the track detected for 7. Thus, for 6,21-diphenyl-11,16-bis( p-tolyl)-24-thia3-aza-m-benziporphyrin 7-S, the addition of acid afforded 6-hydroxy-24-thia-3-azam-benziphlorin 32 (Scheme 15). The process was reversed by neutralization using the nitrogen base. The outcome of adding water is regioselective and depends strongly on the macrocyclic structure. Namely, the presence of bulky mesityl substituents at the meso position adjacent to the m-pyridine ring inhibited a nucleophilic attack, leading to 32. Thus, the hydroxylation of 6,21-dimesityl-11,16-bis( p-tolyl)24-thia-3-aza-m-benziporphyrin 7-S occurred at the perimeter of the pyridine, affording the aromatic 2-hydroxy-24-thia-3-aza-m-benzichlorin 31-OH or 2-ethoxy-24-thia-3-aza-m-benzichlorin 31-OEt when ethanol was added (Scheme 15). The clear diatropicity of 31, detected by 1H NMR, is consistent with 18π-electron delocalization pathways, warranting the observation of the macrocyclic ring current effect in the 1H NMR spectra.

A Pincer Motif Etched into a meta-Benziporphyrin Frame

RO Mes H

H N

193 H N

N Mes

Ar

Ar

Ph

+

H

HN

NH

Tol

Tol

31

OH

H

N

N

TEA

S

Ph

+

S

Tol

Tol Ar = Mes

HN

NH

TEA

S

Tol

Tol Ar = Ph

7-S

32

R = H or Et

Scheme 15 Reactivity of 24-thia-3-aza-m-benziporphyrin 7-S in acidic conditions

2.5.2

2-Aza-m-Benziporphyrins

R

N

R

R1

R2

The synthetic route yielding 2-aza-m-benziporphyrin 8 was based on the McDonaldtype [3 + 1] condensation of the pyridine analogue of tripyrrane 33 with 3,4-diethyl1H-pyrrole-2,5-dicarbaldehyde 34 (Scheme 16) [69]. An interesting observation has been reported by Lash during the synthesis of β-substituted 2-aza-m-benziporphyrin. The condensation of pyridine-2,4dicarbaldehyde 35 and tripyrrane 36 did not result in the isolation of the targeted β-alkylated 2-aza-m-benziporphyrin. Stirring the reaction mixture containing the transient 37 with phenyl chloroformate allowed for the isolation of 3-aza-dihydrom-benziporphyrin (dihydropyriporphyrin) 38 with an external PhO2C protecting group, which incorporates the tautomeric aromatic structure of 37 (Scheme 17) [69].

Scheme 16 Synthesis of 2-aza-m-benziporphyrin 8

Ph

N

1. TFA 2. DDQ

HN

Ph NH

+ HN

33

CHO

N

Ph N

Ph

H N

OHC

34 8

N

194

K. Hurej and L. Latos-Grażyński O

N OHC

N

CHO

35 + HOOC HOOC

TFA

PhO

HN

NH

N

N

37

38

NH HN

36

PhOCOCl

HN

NH

HN

N

Scheme 17 Synthesis of 3-aza-dihydro-m-benziporphyrin

2.5.3

22-Aza-m-Benziporphyrin (Pyriporphyrin)

N

R

R R

R

R

R

In the very first attempt to obtain β-alkylated pyriporphyrin (22-aza-mbenziporphyrin) 42, carried out by Berlin and Breitmaier in 1994, pyridine-2,6dicarbaldehyde 39 reacted with tripyrrane 40 efficiently providing β-alkylated 3-aza-dihydro-m-benziporphyrin (dihydropyriporphyrin, pyriporphyrinogen) 41 [70]. The subsequent intensive oxidation of 41 with p-chloranil in excess did not result in the formation of the pursued β-alkylated 22-aza-m-benziporphyrin 42. Instead of 42, nonaromatic pyriporphyrinone 43 was formed as the single isolable macrocyclic product [70]. In contrast, the gentle oxidation of 41 resulted in the oxidative coupling of 41, generating the dimeric structure 44, in which two macrocyclic subunits are bridged by the C(sp3)meso–C(sp3)meso single bond (Scheme 18).

2.5.4

meso-Tetraaryl-22-Aza-m-Benziporphyrin R

R N

R

R

A Pincer Motif Etched into a meta-Benziporphyrin Frame

OHC

CHO

N

39 + ROOC

N

TFA

HN

NH

HN

ROOC

N

NH HN

40

41

p-chloranil (eqimolar)

p-chloranil (excess)

O

N N

195

H N

N N

NH

N

42

HN

N

HN

N

HN

NH

N

NH

N

44

43

Scheme 18 Synthesis of β-alkylated pyriporphyrin

Ar1 Ar1

N NH

1. EtMgBr 2. PhCOCl

Ar1

HN

45

Ar1

N

Ar1

N

N Ar2

1. NaBH4 2. TFA 3. TEA/DDQ

O

O

46

N Ar2

Ar1

N H N

N Ar2

Ar2

6

Scheme 19 Synthesis of pyriporphyrin 6

meso-Tetraaryl-22-aza-m-benziporphyrin 6 could be considered the straightforward structural homologue of meso-tetraarylporphyrin 1, in which the pyridine motif replaces one of the pyrrole rings [22]. The synthesis strategy was based on [3 + 1]-type condensation and gave a 5.5% yield (Scheme 19). In contrast to the synthesis of 3-aza-m-benziporphyrin 7 [21], the benzoylation of the pyritripyrrane 45 to form diol 46 was necessary. The introduction of bulky meso-aryl groups adjacently to the incorporated pyridine ring was of importance making it possible to increase the stability of 45 during the condensation. Two stages of the protonation of 6 were demonstrated in the 1H NMR experiment. The addition of the first proton resulted in the formation of two symmetrical tautomers, in which one of them 6-H+ shows macrocyclic diatropicity (Scheme 20).

196

K. Hurej and L. Latos-Grażyński

Ar1

Ar1

N N

H N

Ar1

N

N Ar2

Ar2

H N

Ar1

Ar1

N H

H+

N

N Ar2

Ar2

N

H N

Ar1

N HN

NH N Ar2

Ar2

6-1-H+

6

Ar1

Ar1

N H

Ar2

Ar2

6-1'-H+

6-2-H+

Scheme 20 Monoprotonation of pyriporphyrin 6

2.5.5

meso-Alkylidenyl-24-Hetero-22-Aza-m-Benziporphyrin R1

R1

N

R3

R3

Synthetic routes yielding meso-alkylidenyl-24-thia-22-aza-m-benziporphyrin and meso-alkylidenyl-24-oxa-22-aza-m-benziporphyrin have been reported by Lee and coworkers [71]. Both of them were synthesized via the [3 + 1] strategy, starting from the pyritripyrrane analogues 47 and appropriate dicarbinol 24 and using TFA as a catalyst (Scheme 21). Oxapyriporphyrin 48-O was obtained with a 19% yield. 48-O

EtO2C EtO2C

N

EtO2C

NH

X = S or O

EtO2C

CO2Et

1. TFA 2. DDQ

HN

HO Ar

EtO2C

HN

Ar

CO2Et

N H

TFA

HN

NH X

Ar

Ar

Ar

Ar

24

CO2Et

EtO2C TEA

X

HO X

CO2Et

N NH

47

EtO2C

CO2Et

48

49

X = S or O

X=O

+ EtO2C

CO2Et

EtO2C

CO2Et

N N

N X

Ar

Ar

50

for X = S

Scheme 21 Synthesis of meso-alkylidenyl-dihydro-22-aza-24-hetero-m-benziporphyrin 48, its endocyclic isomer 49, and 22-aza-24-hetero-m-benziporphyrin 50 [71]

A Pincer Motif Etched into a meta-Benziporphyrin Frame

197

easily undergoes reversible protonation with trifluoroacetic acid, resulting in the tricationic form 49. Interestingly, the condensation of 47 and 24-S gave two tautomers of thiapyriporphyrin with a 10% (48-S) and 2% (50) yield [71].

3 m-Benziporphyrins: Transformations of m-Phenylene Unit 3.1

Inner Core Transformations R

R X

R

R

m-Benziporphyrins are prone to undergo a unique transformation when stimulated, for instance, by the proper choice of an inserted metal cation. Nonetheless, to this day, a rather limited amount of examples of such reactivity have been reported in the literature [4, 9, 72]. Thus, an attempt to incorporate silver(III) into m-benziporphyrin resulted in regioselective acetoxylation or pyridination affording 22-acetoxy-mbenziporphyrin 51 [2] and 22-pyridiniumyl-m-benziporphyrin 52 [73], respectively, depending on the choice of the neutralizing agent (acetate or pyridine) (Scheme 22). The first transformation involved the reaction of m-benziporphyrin 5 with silver (I) acetate [2]. In the second case, silver(I) tetrafluoroborate was used as the metal source and simultaneously as the oxidant, whereas the pyridine acted also as a nucleophile [73]. The 1H NMR studies documented the perpendicular orientation of the pyridinium group with respect to the phenylene ring of 22-pyridiniumyl-mbenziporphyrin 52.

Ph

Ph N

N H N

Ph

Ph

Ph N

N

H N Ph

52

AgBF4, pyridine reflux

Ph

Ph

Ph N

N

OAc H N

Ph

5

AgOAc, reflux

Ph

N Ph

51

Scheme 22 Regioselective acetoxylation and pyridination of m-benziporphyrin

198

K. Hurej and L. Latos-Grażyński

The acidic hydrolysis of 22-acetoxy-m-benziporphyrin 51 produced a protonated form of 22-hydroxy-m-benziporphyrin 53-H2+ (Scheme 23) [74, 75]. Eventually, after the neutralization of 53-H2+ to form 53, an equilibrium mixture of two tautomers, 53 and 54, was detected using variable-temperature 1H NMR spectroscopy [75]. The tautomer 53 embeds the phenol unit and – as expected for a typical mbenziporphyrin structure – exhibits nonaromatic behavior, preserving the local phenol aromaticity. The alternative tautomer 54 contains a keto group allowing for a 20π-electron delocalization pathway consistent with the experimentally detected macrocyclic antiaromaticity [75]. The suggested mechanism of 22-pyridiniumyl-m-benziporphyrin 52 formation implies the formation of transient silver(III) m-benziporphyrin 5-Ag(III) and the axial coordination of the pyridine to be followed by an intramolecular redox reaction and the transfer of the pyridine to the C(22) center (Scheme 24) [73]. The phenylpyridinium unit of 52 acquires a specific biphenyl-like geometry, in which the pyridinium group is canted with respect to the m-phenylene moiety.

Ar

Ar

Ar N

OAc H N

HCl

N

Ar

NH

NaOH

OH HN H N

N

53-H22+

51

OH

Ar O

NH

N

Ar

Ar

Ar

Ar

H N

Ar

Ar

Ar

Ar

N Ar

Ar

Ar

54

53

Scheme 23 Protonation of 22-hydroxy-m-benziporphyrin

Ar

Ar

Ar

H N Ar

Ar py

Ag(I) N

Ar

Ar N

N

III

Ag

N

N

N Ar

5-Ag(III)

Ar N

H N Ar

Ar CDCl3, Δ

Ar N

HN

N H N

Ar

55

Ar

5-Ag(III)py

Ar N

N

Ar

Ar

5

Ag N

Ar

HN

Ar

N Ar

52

Scheme 24 Formation of 22-pyridiniumyl-m-benziporphyrin 52 and its transformations

A Pincer Motif Etched into a meta-Benziporphyrin Frame

199

The pyridinium species 52 underwent quantitative isomerization to 55 (Scheme 24) [73]. Presumably, the placement of a pyridinium moiety in the center of mbenziporphyrin triggered the peculiar rearrangement of 52, albeit observed solely on the 1H NMR scale. Eventually, m-benziphlorin 55 was formed with the 22-pyridiniumyl ring fused to the adjacent meso position. Formally, this resulted in the incorporation of a new heterocyclic unit, 4a-azafluorene, into the porphyrinoid skeleton.

3.2

Perimeter Modifications: Hydroxy and Alkoxy Substituents X1

X2

Essential perimeter modifications of the m-phenylene ring are addressed to C(2) or C (2) and C(4) positions via hydroxy or alkoxy substitution, affording mesosubstituted 58 and β-substituted 60 derivatives [76–79]. The representative synthetic route to the meso-substituted macrocycle 58 is shown in Scheme 25 [76]. However,

HO OHC

CHO

56

PhMgBr

Ar

HO Ar

pyrrole, ArCHO, BF3.Et2O

HO Ar N

OH OH

Ar

57

Ar

O

N

H N

NH Ar

Ar

Ar

H N

Ar

Ar

58

59 H+

H+ HO Ar

Ar

NH

H N

58'-H22+

Ar

NH Ar

Ar

O

HO Ar

HN

HN

H N

Ar NH

HN Ar

Ar

Ar

58''-H22+

Scheme 25 Synthesis of meso-tetraaryl-2-hydroxy-m-benziporphyrin 59

H N

HN Ar

Ar

59-H22+

200

K. Hurej and L. Latos-Grażyński RO

N

HO

Pd

N

1. Pd(II) 2. RX

N

O

N

H N

60-Pd

60

N

O Pd(II)

HN

NH

N

Pd

N

N

N

61

61-Pd

R = H, n-Bu, Ac, Ts

Scheme 26 Tautomers 58 and 59 of β-alkylated 2-hydroxy-m-benziporphyrin and relevant palladium(II) complexes

it was demonstrated that neutral 2-hydroxy-m-benziporphyrins 58 and 60 exist solely as keto forms (2-oxybenziporphyrins) 59 and 61 (Schemes 25 and 26) [76, 77, 79]. The diprotonation of 58 stabilizes the 2-hydroxy-m-benziporphyrin skeleton 58-H22+. Nonetheless, despite the disrupting phenol moiety, the diatropicity, albeit diminished, was evidently preserved. This reflects the significance of the 580 -H22+ resonance contributor [76]. Both tautomers, 60 and 61, were trapped by the coordination of palladium(II) and generating initially 61-Pd, which converts into 60-Pd in the course of protonation or proper alkylation (Scheme 26) [79]. The 1H NMR data provide evidence for the macrocyclic aromaticity of 61-Pd. The protonation, alkylation, or acylation of the external oxygen atom switches the molecule to the less aromatic state 61-Pd, which was manifested by a significant reduction of the macrocyclic ring current, limiting the porphyrinoid diatropicity [79]. The impact of methoxy substitution on the macrocyclic aromaticity of m-benziporphyrin was clearly manifested in the dicationic form of 2,4-dimetoxym-benziporphyrin 62-H22+(Scheme 27), which was confirmed by the relevant parameters of the 1H NMR spectra [75]. meso-Tetraaryl-2,4-dimethoxy-mbenziporphyrin acts in a similar fashion [80, 81]. In direct analogy to 2-oxybenziporphyrin 61, the synthesis of aromatic 2-oxypyriporphyrin (22-aza-2-oxy-m-benziporphyrin) 64 was also elaborated, using the McDonald-type [3 + 1] strategy as shown in Scheme 28, starting with 3-hydroxypyridine-2,6-dicarbaldehyde 63 and tripyrrane 40 [53].

OMe

MeO

OMe

MeO

OMe

MeO

OMe

MeO

TFA N

H N

62

N

N

H N

62'

N

NH

H N

HN

62-H22+

Scheme 27 Protonation of β-alkylated 2,4-dimethoxy-m-benziporphyrin 62

NH

H N

HN

62'-H22+

A Pincer Motif Etched into a meta-Benziporphyrin Frame Scheme 28 Synthesis of 2-oxypyriporphyrin 64

201 O

HO OHC

N

CHO

1. TFA 2. DDQ

63 + ROOC HN

ROOC

N HN

NH N

NH HN

40

64

Alternatively, β-alkylated 2-oxypyriporphyrin 66 was synthesized by a rather atypical procedure which involved the ring expansion step of dihydroxychlorin 65 (Scheme 29) [82]. A spectacular approach to synthesize the derivatives of meso-tetraaryl-22-aza-3oxa-m-benziporphyrin 68 involves the controlled expansion of a pyrrole ring [68]. Pandey and coworkers showed that the reaction of 2,3-dioxo-mesotetraphenylchlorin 67 with an excess of diazomethane creates the pyridinone frame of 68 (Scheme 30) [83].

O

OH

HO

N

N NaIO4–silica

HN

NH

HN

NH

N

N

65

66

Scheme 29 Synthesis of β-alkylated 2-oxypyriporphyrin via ring expansion

O

MeO

O

O Ar

Ar

Ar N NH

CH2N2

HN Ar

68

HN

NH N

N Ar

Ar N

Ar2

Ar

69

Scheme 30 Expansion of pyrrole ring to form 22-aza-3-oxa-m-benziporphyrin 68

202

K. Hurej and L. Latos-Grażyński

4 m-Benziporphyrins and Aza-m-Benziporphyrins: Coordination Chemistry 4.1

Coordination Motifs

M

Carbaporphyrinoids provide a unique macrocyclic platform which is perfectly suited to exploring organometallic chemistry confined to an atypical macrocyclic environment [4, 8, 9, 72, 84]. Their remarkable structural flexibility is reflected by the adaptation of various coordination modes inside a macrocyclic frame, with or without the formation of a direct metal–carbon bond. The set of organometallic derivatives formed by a combination of metal cations and appropriately tuned carbaporphyrinoids may serve as a perfect (chemically and structurally) “chemical flask” formed by a porphyrin-like surrounding. Atypical oxidation states of metal ions can also be trapped in organometallic settings [4, 8]. Thus, dynamic developments of the synthetic routes of carbaporphyrinoids allow for an exploration of a porphyrin-like or porphyrin-unlike coordination chemistry [4, 5, 8, 68, 84–88]. As shown above, m-benziporphyrins 5, 16 and aza-m-benziporphyrin derivatives 6, 7, 8 are members of the porphyrinoid family with benzene or pyridine moieties built into the porphyrinic framework. The relatively accessible synthetic routes to building a library of β-alkylated and meso-tetraaryl-substituted m-benziporphyrins have been outlined above. Such an overview permits the conclusion that mbenziporphyrins can be considered as highly attractive ligand candidates for the exploration of organometallic chemistry inside the carbaporphyrinoid core. Following organic syntheses, the coordination chemistry of m-benziporphyrins and their analogues has been systematically pursued. The insertion of the following metal cations was attempted: cadmium(II) [16, 59, 61, 74], copper(II) [26], gold(III) [24], iron(II) [26, 89], mercury [16, 59, 61], nickel(II) [16, 56, 78, 90], palladium (II) [2, 57, 79, 90], platinum(II) [2], rhodium(III) [23, 25], silver(III) [61, 91], and zinc(II) [16, 62, 74], albeit with consequences defined by the very nature of the applied cation. In the course of coordination, m-benziporphyrins can acquire two fundamental structures, 5-M and 5-MX (Scheme 31). In a very illustrative case, the diamagnetic, square-planar complex 5-Ni is readily and reversibly transformed by the addition of gaseous HCl into the side-on coordinated paramagnetic 5-NiCl [16]. In solution, the preferred macrocyclic conformation of 5-NiCl has the m-phenylene ring tilted away

A Pincer Motif Etched into a meta-Benziporphyrin Frame Scheme 31 Fundamental structures 5-Ni-and 5-NiCl involved in acid-controlled equilibrium

203

Ar

Ar

Ar N

Ni

HCl

N

N

Ni

base, Δ

N Ph

Ar Cl N

N

Ph

Ar

Ar

5-Ni

5-NiCl

from the nickel(II) center with the inner C–H bonds and the axial Ni–Cl bond in an anti-orientation. As demonstrated by the analysis of the line width of the 5-NiCl 1H NMR spectra, the trace amounts of syn conformers rapidly equilibrate with the respective anti forms [56]. A similar acid-controlled rearrangement was noticed for palladium(II) mesoalkylidenyl m-benziporphyrin [66]. Other meso-alkylidene m-benziporphyrin complexes contain nickel(II) or silver(I) ions (Scheme 32) [66]. In the presence of bulky exocyclic double-bond paramagnetic modified high-spin nickel(II), m-benziporphyrin 23-NiCl was formed. The diamagnetic silver(I) complex 23-Ag was stabilized via the coordination of two pyridine molecules as axial ligands. The relevant X-ray-determined molecular structures of 5-Pd and 5-NiCl exemplifying in-plane [57] and side-on [16] coordination motifs are shown in Fig. 1. The representative molecular structures for aza-m-benziporphyrin complexes 3-Zn [22] and 4-Fe [89] are presented in Fig. 2. EtO2C

CO2Et

EtO2C

CO2Et Cl N

Ni

Ar = C6F5

N

N

EtO2C

CO2Et Pd(II) N

H N

23

EtO2C

CO2Et

EtO2C

CO2Et

N

N Ar

Ar

23-NCl EtO2C

CO2Et

EtO2C

Ar

Ar

Ni(II)

Pd

CO2Et X

TFA N

N

Pd

N

N

N

23-Pd-X

23-Pd EtO2C

Ar

Ar

Ar

Ar

Ag(I)

CO2Et

EtO2C

Ar = p-Tol or C6F5

X = OCOCF3

CO2Et

EtO2C

CO2Et L N

Ag

L = 2,6-dimethylpyridine

N

Ar = C6F5

N L Ar

Ar

23-Ag

Scheme 32 Nickel(II), palladium(II), and silver(I) meso-alkylidenyl m-benziporphyrins – formation and protonation [66]

204

K. Hurej and L. Latos-Grażyński

Fig. 1 Crystal structures of (a) palladium(II) m-benziporphyrin 70-Pd and (b) chloronickel(II) mbenziporphyrin 5-NiCl (side views with aryl groups omitted for clarity). Adapted from Refs. [16, 57], respectively

Fig. 2 Crystal structures of (a) zinc(II) 3-aza-m-benziporphyrin 7-Zn and (b) iron(III) 22-aza-mbenziporphyrin 6-Fe (both side views with aryl groups omitted for clarity). Adapted from Refs. [22, 89], respectively

A Pincer Motif Etched into a meta-Benziporphyrin Frame

Ar N

N

Ar

N N

N

M

N

Ar

Ar

M

Ar

Ar

N

N

N

Ar

Ar

Ar

5-M

Ar

Ar

N

N

M

N

N Ar

M = Zn, Fe

RO

M N

6-M

M = Au, Ni, Pd, Rh, Pt

205

7-M

14-M

M = Fe

M = Ni, Pd

O

O N

N

N

M

N

N

N

M

N

N

60-M

N

N

61-M

M = Pd

M

64-M

M = Pd, Ag

M = Ni, Cu, Zn

R MeO

O Ar

Ar

N N

M N

N

N

M

OMe

Ar

Ar N

N

N

M

N

N Ar

67-M

70-M

M = Ni

M = Pd

Ar

71-M M = Ni, Pd

Scheme 33 Regular (CNNN) and (NNNN) coordination modes reported for m-benziporphyrin and aza-m-benziporphyrins

In the first type of coordination mode, the metal cation is typically coordinated by an equatorial [CNNN] donor set forming a direct metal–carbon σ-bond (Scheme 33). Alternatively, m-benziporphyrin acts as a monoanionic ligand coordinating through three equatorial nitrogen donors. Nevertheless, the specific construction of mbenziporphyrin imposes a structure promoting a side-on interaction with mphenylene that results in a remarkable agostic interaction of the metal ion with the C(21)–H unit of the m-phenylene ring. In such structures, the interaction between the C–H unit and the metal cation was experimentally identified either by the remarkable change of the 1H and 13C NMR chemical shifts of the C(22)–H fragment or the scalar M–C(22) and M–H(22) coupling (M ¼ 111/113Cd, 199Hg) [16]. The remarkable downfield positions of the H(22) resonance in the 1H NMR spectra of the paramagnetic nickel(II) 5-NiCl or iron(II) 5-FeCl complexes reflect the structurally imposed

206

K. Hurej and L. Latos-Grażyński

constraints introducing agostic interactions [16, 26]. The appropriate shifts are 380 ppm and 450 ppm (293 K) for 5-NiCl and 5-FeCl, respectively. The equatorial [CNNN] coordination mode seems to be a common denominator, generally accepted for m-benziporphyrins and aza-m-benziporphyrins encountered for the attempted coordination of gold(III) (5-Au [24]), iron(III) (7-Fe [89]), nickel (II) (5-Ni [16], 14-Ni [90], 71-Ni [78]), palladium(II) (5-Pd [2], 14-Pd [90], 23-Pd [66], 61-Pd [79], 60-Pd [79], 70-Pd [57]), platinum (II) (5-Pt [2]), rhodium(III) (5-Rh [23]), and silver(III) (61-Ag [91]), gathered in Scheme 33. Furthermore, 22-aza-benziporphyrins acquire a planar macrocyclic arrangement imposed by the coordination of four nitrogen equatorial donors, i.e., iron(III) (6-Fe [22]), copper (64-Cu(II) [53]), nickel(II) (64-Ni [53], 52-Ni [92, 93]), and zinc(II) (6-Zn [22], 64-Zn [53]). A side-on interaction, sometimes enforced by inner substitution, was reported for copper(II) (5-Cl)-CuX [26]), cadmium(II) (5-CdX [16], 21-CdX [16, 61]), iron (II) (5-FeX [26], 7-FeX [94]), mercury(II) (5-HgX [16], 21-HgX [59, 61]), nickel (II) (5-NiX [16, 56], 21-NiX [16, 56], 23-NiX [66]), palladium (23-PdX [66]), silver (I) (21-AgX [61]), or zinc(II) (5-ZnX [16], 21-ZnX [16, 62]) (Schemes 31, 32, and 34). Metallo-m-benziporphyrins can serve as perspective building blocks to construct intricate ensembles. Thus, the external N-coordination of pyridine in 7-FeX provided a route for the construction of diiron species, whereas the pyridine moiety serves as a unique bridge (Scheme 35a) [74]. In the solid, the pairs of 7-FeX molecules are linearly arranged in the head-to-head fashion revealing the adjacency of the two perimeter nitrogen atoms linked by the NHN hydrogen bond) (7-Fe)H (7-Fe) (Scheme 35b) [89]. A unique molecular arrangement was encountered in the dimeric structure of copper(II) 22-chloro-m-benziporphyrin) [(5-Cl)-Cu)] Cu2Cl2[(5-Cl)-Cu)]. Here, two (5-Cl)-Cu subunits (side-on coordination of copper(II)) are linked by bridging the [Cu2Cl4]2anion (Scheme 35c) [26]. Moreover, the silver(I) porphodimethene 21-Ag forms a dimeric structure with two subunits assembled in the head-to-tail fashion. They are connected by the atypical η2 π-coordination of silver(I) to the β-pyrrolic C¼C bond of the adjacent building block (Scheme 35d) [61]. N Ar

Ar N M X N

Ar

Ar

Ar N X

N Ar

R M

Ar N

N

M

X

N

Ar

Ar

Ar

5-MX

(5-R)-MX

M = Cd, Fe, Hg, Ni, Zn

M = Cu R = Cl M = Cd, Ni, Zn, R = CH3COO

R

R R

N M X N

N

N

Ar

R

Ar

7-MX M = Fe

Scheme 34 Side-on coordination motifs of m-benziporphyrin complexes

Ar

N Ar

21-MX M = Ag, Cd, Hg, Ni, Zn

A Pincer Motif Etched into a meta-Benziporphyrin Frame

Ar

Ar

Ar N X N FeII

N

N FeIIX3

Ar

C)

Ar

Ar N

N N

Fe

H

N

Fe

N

B)

Ar

Cl

Ar

Ar

D)

Ar

Cl

Cu

N

Cl Cu

N

N

Ar

Ar

Ar

Ar Ar N

N

Me Me

7-Fe

Ar

(5-Cl)-Cu-X

N N N

Ag

Ar

7-Fe

Cl

N

N

N

Ar

N

Me Me

Ar

Ar

N

Cu

Cu

(5-Cl)-Cu-X

(7-FeX)-Fe Ar

Ar Cl

Cl

N

A)

207

Ar Ar

21-Ag

Me Me

Ag N

Me Me

21-Ag

Scheme 35 m-Benziporphyrin complexes as building blocks: (a) (7-FeX)-Fe [94], (b) (7-Fe)H(7-Fe) [89], (c) [(5-Cl)-Cu)]Cu2Cl4[(5-Cl)-Cu)] [26], (d) (21-Ag)2 [61]

N Ar

Ar N

O

Ar

Ar R

O

O M

N

N

N Ar

M

N

N

53-M M = Fe, Ni, Pd

Ar

N

N

N Ar

M

Ar

(7-O)-Fe M = Fe

(61-Me)-M M = Pd

Scheme 36 Coordination motifs – inner core modifications (hydroxylation and C-alkylation)

Interestingly, the regioselective alkylation of the originally trigonal coordinating C(22) atom of palladium(II) 2-oxy-m-benziporphyrin 61-Pd resulted in the strongly distorted tetrahedral of the formed (61-Me)Pd (Scheme 36) [79]. The formal incorporation of an oxygen atom into the M–C(22) bond embedded in m-benziporphyrin complexes eventually results in the introduction of a phenolate donor to the coordination core. This changed the original (CNNN) surrounding into the ((C)ONNN) one as seen for iron(III) ((7-O)-Fe [94], (53-Fe [74]), nickel(II) (53-Ni [74]), and palladium(II) (53-Pd [74]). The canted position of the phenolate unit is shown in the X-ray-determined molecular structure presented in Fig. 3. A peculiar and uncommon coordination motif was observed after the coordination of rhodium(III) or gold(III) ions to 22-alkyl-m-benziporphyrin (Scheme 37) [23– 25]. The insertion of rhodium(III) into the structurally prearranged 22-methyl-mbenziporphyrin afforded aromatic rhodium(III) 22-(μ-methylene-m-benziporphyrin) 72-Rh [25]. The formation of a methylene bridge between the rhodium(III) center and the C(22) inner carbon atom was firmly established. In an analogous process, 22-ethyl-m-benziporphyrin yielded rhodium(III) 22-(μ-ethylidene-m-benziporphyrin) 73-Rh [25].

208

K. Hurej and L. Latos-Grażyński

Fig. 3 Crystal structures of palladium(II) 22-hydroksy-m-benziporphyrin 53-Pd (side view with aryl groups omitted for clarity). Adapted from Ref. [74]

Ar

Ar

Ar

Ar

Ar

N Rh Cl N

N

72-Rh

N Ar

Ar

Cl

Ar

Ar

CH3

H Rh N

73-Rh

N

N Ar

Ar

Cl

Ar

Ar

CHO

H Rh N

74-Rh

N

N Ar

Ar

Au N

N Ar

(29-Me)-Au

Scheme 37 22-R-m-Benziporphyrins – insertion of rhodium(III) and gold(III) ions

A similar structure was originally detected for rhodium(III) 22-(μ-acetaldehyde)m-benziporphyrin 74-Rh in the course of investigations on the reactivity of the rhodium(III) m-benziporphyrin system [23]. The appropriated DFT studied on 73-Rh, 74-Rh, and 75-Rh revealed that the environment of the coordinated C (22) carbon atom was modestly distorted from planarity, sufficiently to limit the [6]annulene aromaticity promoting macrocyclic diatropicity, which was confirmed by experimental and calculated 1H NMR chemical shifts [23, 25]. The reaction of 22-methyl-m-benziporphyrin 29-Me with sodium tetrachloroaurate(III) resulted in a quantitative transformation into the very reactive aromatic gold(III) 22-methyl-mbenziporphyrin (29-Me)-Au [24]. In contrast to rhodium(III), the μ-methylene bridge was not formed.

4.2

Inner Core Reactivity

The structural electronic properties and reactivity of the metallocarbaporphyrinoids are directly related to the mode of metal–carbon interaction [4, 5, 16, 72]. In the context of a variety of feasible metal–carbon bonding modes, warranted by specific

A Pincer Motif Etched into a meta-Benziporphyrin Frame

209

molecular structures, metallocarbaporphyrinoids provide the appropriate environment for intensive spectroscopic, synthetic, and structural exploration targeting organometallic reactivity [9, 72]. In a more specified approach, m-benziporphyrins and aza-m-benziporphyrins enable exploring some intriguing aspects of organometallic chemistry which engage the m-phenylene or aza-m-phenylene units. Accordingly, one can investigate C–H bond activation or C–M reactivity, identify novel structural motifs, and – most importantly – gain some insight into rare examples of benzene cleavage and contraction [95–103].

4.2.1

Overview

In the following section, the core reactivity of m-benziporphyrins and aza-mbenziporphyrins has been subjectively and arbitrarily divided into eight types distinguished by the principal type of transformation. Nonetheless, in several instances the process reflects a complexity of two or more inseparable steps – then the dominant one was applied as the criterion: • • • • • • • •

Oxidative nucleophilic processes (Scheme 22) [2, 73] Macrocyclic fusion (Scheme 24) [73] Various substitutions: chlorination (Scheme 38) [26] Various substitutions: alkylation (Scheme 39) [79] Oxidation and oxygenation (Scheme 40) [94] C–H bond activation (Scheme 41) [23] C–C bond activation (Scheme 44) [25] Contraction of m-phenylene (Scheme 45) [23]

It should be noted that some of reactions have already been addressed in the previous sections; thus, here they have been very briefly recalled and some summarized in the form of Schemes, aiming to systemize the presentation. Various Substitutions: Chlorination Scheme 38 Chlorination of m-benziporphyrin [26]

5 Ar

5-Cl Ar

Ar N H N

Ar

CuCl2

N

N

Cu2Cl4

N

N Ar

5

Ar Cl Cu

Ar

Ar

2

[(5-Cl)-Cu)]Cu2Cl4[(5-Cl)-Cu)]

210

K. Hurej and L. Latos-Grażyński

Various Substitutions: Alkylation Scheme 39 Inner alkylation of 2-oxy-mbenziporphyrin [79]

61-Pd

(61-R)-Pd O

O

N

Pd

R N Pd

RI

N

N

N

N

61-Pd

(61-R)-Pd

Oxygenation 7

(7-O)-M

N Ar

N Ar

Ar X N FeII N

O2

N

N Ar

Ar O

III

Fe

N

N Ar

7-FeX

Ar

Ar

(7-O)-Fe

Scheme 40 Iron(II)-promoted oxygenation of 3-aza-m-benziporphyrin 7 to form the frame of 22-hydroxy-3-aza-m-benziporphyrin 4-OH [94]

4.3

C–H Bond Activation

The formation of a μ-acetaldehyde bridge in the rhodium(III) m-benziporphyrin can be considered an enlightening example of C–H bond activation in a carbaporphyrinoid surrounding (Scheme 41) [23]. The nature of this transformation and, above all, the source of the μ-acetaldehyde bridge in 75-Rh remained quite a puzzle for the chemistry confined m-benziporphyrin cavity [23]. In fact, such a transformation did not have any precedent in the chemistry of metallocarbaporphyrinoids [72]. Two hypotheses were initially outlined involving (a) a carbonyl ligand and/or (b) solvent participations. The first one examined a feasible transformation of the axial carbonyl ligand. In particular, the reductive coupling of two 5-Rh(CO) molecules was considered. The process could resemble the one-electron activation of CO by a

A Pincer Motif Etched into a meta-Benziporphyrin Frame 13CO

211

silica gel, CH2Cl2 H

13

5-Rh( CO) Ph

Ph N

Rh

MeO Tol

N

NaOMe

N Tol

Ph

Ph

Tol

N Rh Cl N

5-Rh(MeO)

CH2Cl2

CO N Tol

N

Cl

Tol

N

Rh N

+

Cl Ph H

Ph

+ Ph CHO

Ph CH2Cl2

N Ar

Tol

N

N Ar

74-Rh

75-Rh

5-Rh(CO)

Rh

Scheme 41 Reactivity of rhodium(III) m-benziporphyrin 5-Rh(CO)

rhodium(II) porphyrin bimetalloradical complex and concerted reactions of two (RhCO) units [104], forming a novel two-carbon fragment to be finally incorporated into the Rh(III)–C(22) bond. To check this hypothesis, a carbonyl ligand enriched in 13C (13CO) was introduced yielding 5-Rh(13CO) – a unique marker suitable for 13C NMR spectroscopy. 5-Rh(13CO) was subjected to aromatization in typical 5-Rh(CO) ! 74-Rh conditions. The identity 13C NMR spectra generated in the reactions of 5-Rh(13CO) and 5-Rh(CO) excluded the CO involvement in the generation of the methylene bridge (Scheme 41) [23]. Finally, a series of solvents were checked. The prolonged heating of 5-Rh(CO) in dichloromethane afforded 75-Rh – a rhodium(III) m-benziporphyrin complex substituted with vinyl chloride attached in the C(22) position. Eventually, 75-Rh converted to 74-Rh (Scheme 41). Subsequently, the analogous reaction but in CD2Cl2 was tested, producing the derivative deuterated at the chlorovinyl substituent unambiguously identifying dichloromethane as a long pursued substrate [23]. Significantly, apart from dichloromethane other solvents undergo C–H activation with 5-Rh(13CO). Thus, during the insertions of rhodium(III) in aromatic solvents (benzene, toluene), six-coordinate complexes of rhodium (III) m-benziporphyrins with axially coordinated phenyl 5-Rh(Ph) and benzyl 5-Rh(Bz) were identified (Scheme 42) [105]. The relevant DFT optimized models are shown in Fig. 4.

Ph

Ph N

Rh N

Tol

Ph

Ph

CO N

N Tol

5-Rh(Ph)

Tol

Rh

CO N

N

Tol

5-Rh(Bz)

Scheme 42 Reactivity of rhodium(III) m-benziporphyrin in aromatic solvents [105]

212

K. Hurej and L. Latos-Grażyński

Fig. 4 DFT-optimized molecular models of (a) 5-Rh(Ph) and (b) 5-Rh(Bz) (side view with aryl groups omitted for clarity). Adapted from Ref. [105]

In addition to the typical solvents, rhodium(III) m-benziporphyrin reacts with aldehydes, such as butanal, giving the alkanoyl five-coordinated complex 5-Rh (RCO) (Scheme 43). The exposure of the solution to UV radiation (λ ¼ 365 nm) in the presence of dioxygen leads to the regioselective activation of the Rh–C(22) bond and, consequently, to the incorporation of an oxygen atom. As a result, σ-alkanoyl rhodium(III) 22-hydroxy-m-benziporphyrin 53-Rh(RCO) was obtained (Scheme 43) [105]. Ph

Ph N Cl Tol

CO Rh N N

Tol

5-Rh(CO)

Ph

Ph butanal, 0,5 h

N Tol

Rh

N

N

Ph

Ph O

O N Rh

365 nm

Tol

5-Rh(RCO)

Scheme 43 Alkanolyation of rhodium(III) m-benziporphyrin

Tol

N

O

N Tol

53-Rh(RCO)

A Pincer Motif Etched into a meta-Benziporphyrin Frame

4.4

213

C–C Bond Activation

The reversible cleavage of the C–C bond in rhodium(III) 22-(μ-ethylidene-mbenziporphyrin 73-Rh (Scheme 44) was detected, providing an unusual example of C–C bond activation [25]. Rhodium(III) 22-(μ-ethylidene-m-benziporphyrin) 73-Rh was reduced to 76-Rh under strict anaerobic conditions. One-electron reduction led to an electronic structure, described by a set of two main resonance contributors, (P•)Rh(III) $ (P)Rh (II) (P ¼ m-benziporphyrin), however, with the dominant features of (P•)Rh(III) [25]. The fingerprint 1H NMR pattern consistent with the suggested electronic structure of 76 is presented in Fig. 5 revealing the enormous spread of (150 to 200 ppm) resonances [25]. The subsequent reduction of the paramagnetic 76-Rh instigated ethyl migration from carbon(22) to rhodium(III), affording diamagnetic rhodium(III) mbenziporphyrin 78 with the apically coordinated σ-ethyl ligand. The 73-Rh ! 78Rh ! 73-Rh conversion is a ground-breaking example of reversible C(sp2)–C(sp3) bond cleavage in a carbaporphyrin environment, which evidently resembles the transformations reported for rhodium pincer ligand complexes [32, 106–109].

Ph CH3

Ph III

Tol

N Rh Cl N

N Tol

Ph CH3

Ph -

Zn ,

H+

N

Tol

N

I

N

Rh N

Tol

Tol

77-Rh

-[H2]

Ph CH3 III

Rh

H

Tol

+ e-

76-Rh

Ph N

N

N

Tol

73-Rh +H+

III

Rh

Ph CH3

Ph

N

N

80-Rh

Ph CH3

Ph N

Tol

Tol

I

Rh N

79-Rh

Ph CH3

Ph

N

N Tol

Tol

III

Rh N

N

Ph

Ph EtMgBr

Tol

78-Rh

Scheme 44 Reactivity of rhodium(III) 22-(μ-ethylidene)-m-benziporphyrin 73

Tol

CO III N Rh N Cl N

Tol

5-Rh(CO)

214

K. Hurej and L. Latos-Grażyński

Fig. 5 1H NMR spectrum of 76-Rh (C6D6, 325 K). Adapted from Ref. [25]. The relative intensity of the +180 to 160 ppm and 100 to 120 ppm regions in traces was increased five times relative to the inner part of the spectrum

4.5

Contraction of m-Phenylene

In basic conditions, rhodium(III) 22-(μ-acetaldehyde)-m-benziporphyrin 74-Rh, formed from rhodium(III) m-benziporphyrin 5-Rh(CO), undergoes a fundamental transformation of the built-in m-phenylene moiety to create rhodium(III) 21-(μ-acetaldehyde)-21-carbaporphyrins 81-Rh, 82-Rh, and 83-Rh (Scheme 45). The porphyrinoid coordination core supports the formyl-substituted rhodacyclopropane motif. In the course of contraction, a perimeter carbon atom (C(2) or C(3)) of 5-Rh (CO) is completely extruded (81-Rh) or conserved as a formyl substituent of the emerged cyclopentene ring in 82-Rh and 83-Rh [23]. The X-ray-determined molecular structure of 81-Rh resembles the geometry of regular metalloporphyrins modified by the insertion of a single carbon atom into an M–N bond (Fig. 6) [110]. In fact, the structural features of 81-Rh bear some similarity to the molecular geometry of rhodium(III) μ-methylene-21carbaporphyrin generated via the contraction of rhodium(III) p-benziporphyrin 4-RhCO [18]. Thus, the geometry of 81-Rh demonstrates some tetrahedral distortion around the C(21) carbon atom and, in consequence, implies the Rh(III)η2-C (21)C(25) bonding mode consistent with the formation of the rhodacyclopropane ring. The postulated mechanism of contraction involves an OH nucleophilic attack at the C(3) position forming (74-OH)-Rh. The subsequent rearrangement generated the bicyclo[3.1.0]hexane frame built into the transient 84. In the third stage, the extrusion of the perimeter carbon atom took place (Scheme 46).

A Pincer Motif Etched into a meta-Benziporphyrin Frame

215 Ph CHO

Ph H N

Rh N

Ar

Ph

Tol

N Rh Cl N

CO N

CH2Cl2

Tol

5-Rh(CO)

Ph CHO

Ph N Ar

Rh N

74-Rh

N

Ar

81-Rh

+ Ph

N

CHO Al2O3, CH2Cl2

Ph CHO

Ph H N

Rh

Ar

N

N

Ar

Ar

82-Rh OHC Ph CHO

Ph H N Rh Ar

N

N

Ar

83-Rh

Scheme 45 Contraction of rhodium(III) m-benziporphyrin 5-(RhCO) to rhodium(III) 21-carbaporphyrins 81-Rh, 82-Rh, and 83-Rh

Fig. 6 X-ray-determined molecular crystal structure of 81-Rh. Adapted from Ref. [23]

Rhodium(III) 22-(μ-ethylidene)-m-benziporphyrin 73-Rh also undergoes the contraction of m-phenylene, giving a mixture of 21-(μ-ethylidene)-21carbaporphyrins 85-Rh, 86-Rh, and 87-Rh (Scheme 47). They contain the rhodacyclopropane motif substituted by a methyl group [25].

216

K. Hurej and L. Latos-Grażyński HO H

+ Ar CHO

Ar N

N

Rh N

Ar

Al2O3

N

Ar

H

H

Ar

Ar CHO N

Ar

N

Rh N

Ar

84-t1 -

H

Ar

N

N

OH

H Ar CHO

Rh

Ar

(74-OH)-Rh

Ar N

N

OH

H

N

Rh

Ar

74-Rh H

Ar CHO

Ar

Ar

84-t2

HO H C

-[H2]

HCHO

CHO Ar CHO

Ar N Ar

Rh

N

N

Ar CHO

Ar N

Ar

Rh

N

N

Ar

81-Rh

Ar

82-Rh

Scheme 46 Postulated mechanism of the contraction of m-phenylene to cyclopentadiene

CHO Ph

Ph

CH3

H N Tol

Cl

Rh N

73-Rh

H

Al2O3

N

N Tol

Ph CH3

Ph

Tol

Rh N

85-Rh

H

+

N Tol

N Tol

OHC

Ph CH3

Ph Rh N

86-Rh

H N Rh

+

N Tol

Ph CH3

Ph

Tol

N

N

Tol

87-Rh

Scheme 47 Formation of 21-(μ-ethylidene)-21-carbaporphyrins 85-Rh, 86-Rh, and 87-Rh

Rhodium(III) 21-(μ-ethylidene)-21-carbaporphyrin 85-Rh is reactive in solution and in solids. After a few hours, it converts into the corresponding green-brown compound 90-Rh [25]. The protonation of the internal C(25) atom in 85-Rh leads to a transient form – rhodium(III) 21-ethyl-21-carbaporphyrin 88-Rh. The oxidation of 88-Rh generates 21-ethylidene-21-carbaporphyrin 89-Rh, which formally contains an embedded 6-methylfulvene motif. The isomerization of 89-Rh produces rhodium(III)

A Pincer Motif Etched into a meta-Benziporphyrin Frame

217

21-vinyl-21-carbaporphyrin 90-Rh, creating the 21-vinyl unit from the 21-ethylidene one (Scheme 48) [25]. Gold(III) 22-methyl-m-benziporphyrin (29-Me)-Au has also been found to be prone to contraction [24]. The reaction of 22-methyl-m-benziporphyrin 29-Me with sodium tetrachloroaurate(III) in benzene gave a mixture of gold(III) 21-methyl-mbenziporphyrins 91-Au and 92-Au with an almost quantitative yield (95%) (Scheme 43) [24]. The insertion of gold(III) conducted in dichloromethane afforded solely (29-Me)-Au, which readily transformed into 91-Au and 92-Au in the course of column chromatography (Scheme 49). 92-Au converts to the weakly aromatic ketone derivative 93-Au, which is the result of the formal substitution of a formyl group with a hydroxyl group. The resulting species undergoes keto-enol tautomerization 94-Au–95-Au (Scheme 50) [24].

+ Ph

Ph

CH3 N

Rh

N

N

Tol

Ph

Ph +H

CH3

+

Tol

N

-[H2]

N

Tol

N

Rh

Tol

88-Rh

Ph

Ph

CH3 Cl

N

Tol

85-Rh

N

Rh

Ph

Ph

CH2

base

N

N

N

Tol

Tol

N

Rh

Cl

89-Rh

Tol

90-Rh

Scheme 48 Acid–base reactivity of rhodium(III) 21-(μ-ethylidene)-21-carbaporphyrins 85-Rh

2+ Ph

Ph

. Na[AuCl4] 2H2O

N

N H N

Tol

Tol

N

Au

21-Me

Al2O3

N

Au

N

Tol

+

Ph N

Tol

Tol

91-Au

(29-Me)-Au

N

Au N

Tol

Tol

+

Ph

Ph

N

N

Tol

Ph

N

OHC

+

Ph

Ph

92-Au

Scheme 49 Reactivity of gold(III) m-benziporphyrin (29-Me)-Au

OHC Ph N

HO

O

Ph Au

Ph OH-

N

N

N

92-Au

Au

H+

N

93-Au

Ph N

Au

Ph

N

Ph N

N Tol

Tol

O

Ph

N Tol

Tol

Ph

N

N Tol

Tol

Au

Tol

Tol

94-Au

Scheme 50 Formation of gold(III) 2-oxy-21-carbaporphyrin and its tautomerization

95-Au

218

4.6

K. Hurej and L. Latos-Grażyński

Contraction: General Remarks

At this point, it is significant to evoke that two isomeric carbaporphyrinoids, pbenziporphyrin 4 and m-benziporphyrin 5, acted as the suitable macrocyclic templates to create 21-carbaporphyrin complexes. These isomers undergo conceptually related rearrangements of the inner core ( p-benziporphyrin) or perimeter (mbenziporphyrin) extrusion of a single carbon atom (Scheme 51), which is convoluted with the contraction of p- or m-phenylene rings to a cyclopentadiene unit. In macrocyclic terms, the subunit contractions ultimately create the identical molecular target of 21-carbaporphyrin 3 [17, 18, 23–25, 111]. Thus, at this stage, the investigation of meso-tetraaryl-21-carbaporphyrin 3 was initiated, but solely as a macrocyclic ligand in 3-M complexes formed in the course of the palladium(II), gold(III), and rhodium(III) benziporphyrin rearrangements (Scheme 51). Eventually, an efficient protocol leading to free-base meso-tetraaryl-21carbaporphyrin 3 was elaborated [14]. Consequently, two routes to palladium (II) meso-tetraaryl-21-carbaporphyrin are available (Scheme 52). Originally 3-Pd was generated through the rather demanding contraction of palladium(II) pbenziporphyrin [17]. In due course, it was proven that the identical species can be produced by the classical insertion of palladium(II) into a prearranged

Ar

Ar N

Rh

Ar

N

N

III Rh

N Ar

X N

N

N Ar

N

Rh N

Ar

Ar

Ar

Ar

Ar

Ar

Ar

rhodium(III) m-benziporphyrin

rhodium(III) 21-carbaporphyrin

rhodium(III) p-benziporphyrin

5-Rh

3-Rh

4-Rh

Scheme 51 Contraction of isomeric rhodium(III) benziporphyrins to form the common structural target – a rhodium(III) 21-carbaporphyrin

Ar

Ar N Ar

Pd N

4-Pd

OH

N Ar

Ar

Ar H

-

N

Pd

Pd(II)

N

HN

NH

N

N Ar

Ar

Ar

Ar

Ar

Ar

3-Pd

Scheme 52 Routes to form palladium(II) 21-carbaporphyrin [14, 17]

3

A Pincer Motif Etched into a meta-Benziporphyrin Frame

219

21-carbaporphyrin 3 (Scheme 52). The molecular design of 3 conserves the virtues of meso-tetraarylporphyrin 1, including the perfect match between the ionic radii of an inserted metal and the size of the macrocyclic (CNNN) core opening the route to original organometallic chemistry in well-defined macrocyclic vessels.

5 Conclusion and Perspectives In general, the exploration of carbaporphyrinoids enabled addressing several fundamental issues of chemistry, such as the aromaticity of molecules of the Möbius topology [112, 113], organometallic copper(II) compounds [114, 115], the contraction of the benzene ring to cyclopentadiene [17], or the inclusion of a d-electron subunit in a π-electron conjugation pathway [4, 116]. Previous studies have shown that metalloporphyrins act as effective catalysts for oxygenation or C–H bond functionalization. Their activity and selectivity strongly depend on the central metal and on the stereoelectronic features of porphyrin ligands [117, 118]. Metallocarbaporphyrin catalytic activity was correspondingly probed in selected processes, including epoxidation and cyclopropanation. Noteworthy, in the cyclopropanation of styrene with the use of rhodium(III) [119] or cobalt(II) [120] N-confused porphyrin, the selectivity of the catalyst outperforms the porphyrins and corroles, reflecting the influence of electronic factors (Scheme 53). A potentially stimulating approach, which relates the chemistry of the pincer ligand and m-benziporphyrin (aza-m-benziporphyrin) chemistry, has been presented in this chapter. Essentially, both groups have been structurally and functionally based on common dominators, i.e., on m-phenylene (aza-m-phenylene) rings. We strongly believe that such an analogy could and should be fruitfully extended to a larger number of pincer ligand–carbaporphyrinoid couples, providing stimulus for the creation of original organometallic chemistry in inventively constructed surroundings. Thus, to emphasize this point, several metallocarbaporphyrinoids have been shown in Scheme 54 using “the pincer ligand chemists’ perspective,” i.e., palladium(II) m-benziporphyrin 5-Pd [2], iron(III) 2-aza-21-carbaporphyrin (N-confused porphyrin) 3-Fe [121], silver(III) 2-oxa-21-carbaporphyrin 96-Ag [122], cadmium(III) 2-thia-21-carbaporphrin 97-Cd [123], gold(III) 21-carbaporphyrin 3-Au [111], palladium(II) vacataporphyrin 98-Pd [124], ruthenium(II) p-benziporphyrin 4-Ru [125], copper(II) thiaethyneporphyrin 99-Cu [126], ruthenocenothiaporphyrin 100-Ru [127], cobalt(II) azuliporphyrin 101-Co [128], O +

H

O N2

Et

[cat]

H COOEt

toluene, 80oC

Scheme 53 Cobalt(II) N-confused porphyrin-mediated cyclopropanation of styrene [120]

220

K. Hurej and L. Latos-Grażyński

N

Pd

N

N

5-Pd N

O

S Cl

Br N

Fe

N

N

N

Cd

N

N

2-Fe

96-Ag

97-Cd

N

Au

N

N

3-Au

Cl Pd

N

N

N

Co

N

N

105-Pd

Au

N

N

Cl N

Pd

S

100-Ru

N

Ag N

N

N

106-Pd

N

HN

NH

99-Cu

102-Au

N

N

Cu SH

N

101-Co

N

N

CO

4-Ru

N

Cl N Pd

N

Ru N

98-Pd

N

Ag

N

N

Cl N

N

N

Ag

N

N

103-Ag

104-Pd

N

R P R N N

Pd N

107-Pd

108-P

Scheme 54 Carbaporphyrinoids and metallocarbaporphyrinoids from the “pincer ligand perspective” (pincer ligand pattern in bold blue, perimeter substitution omitted for clarity)

A Pincer Motif Etched into a meta-Benziporphyrin Frame

221

gold(III) benzocarbaporphyrin 102-Au [129], silver(III) tropoporphyrin 103-Ag [91], palladium(II) 1,3-naphthiporphyrin 104-Pd [90], palladium (II) 1,4-naphthiporphyrin 105-Pd [130], palladium(II) meso-anthriporphyrin 106Pd [131], palladium(II) pyreniporphyrin 107-Pd [132], and phosphorus(V) phenathriporphyrin 108-P [133]. On the other hand, several expanded and contracted porphyrins which incorporate pincer motifs containing a meta-phenylene or pyridine moiety (Scheme 55), such as subpyriporphyrin 109 [134], dithia-m-benzisapphyrin 110 [139], the dithia analogue of meso-alkylidenyl-m-benzisapphyrins 111 [138], m-benzihexaphyrins (1.0.0.1.0.0) 112 [135, 136], and dipyrihexaphyrin (1.1.1.1.1.1) 113 [137], can be potentially applied as promising ligands in this area, as exemplified by in-plane thorium(IV), uranium(IV), and neptunium(IV) dipyriamethyrin complexes 114-Ac [140, 141].

Ar Ar

N N

R

Ar

R

R

R

Ar N

N

HN Ar

S

N

N Ar

S

Ar

S

Ar

S

Ar

109

111

110 CO2Et

Me

HN

N

Me

N

NH

Me CO2Et

EtO2C

CO2Et

N N

N

N

N

Ar

Ar

Ar

Ar Me

EtO2C

CO2Et

EtO2C

EtO2C

N

EtO2C CO2Et

112

CO2Et EtO2C

113 N N Ar

XN Ar

Ac N

X

N N

114-Ac

Ac = Th(IV), U(IV), Np(IV)

Scheme 55 Select examples of contracted and expanded porphyrinoids, containing a metaphenylene or pyridine moiety [134–141]

222

K. Hurej and L. Latos-Grażyński

In conclusion, novel ligands are expected to be constructed fusing the benefits of pincer ligand and carbaporphyrin strategies to provide unconventional insight into organometallic chemistry unlimited by specific coordinating confinement.

References 1. Berlin K, Breitmaier E (1994) Angew Chem Int Ed Engl 33:1246 2. Stępień M, Latos-Grażyński L (2001) Chem A Eur J 7:5113 3. Stępień M, Latos-Grażyński L (2009) Aromaticity and tautomerism in porphyrins and porphyrinoids. In: Krygowski TM (ed) Aromaticity in heterocyclic compounds. Springer, Berlin, p 83. Chap 3 4. Pawlicki M, Latos-Grażyński L (2010) Carbaporphyrinoids – synthesis and coordination properties. In: Kadish KM, Smith KM, Guilard R (eds) Handbook of porphyrin science: with applications to chemistry, physics, materials science, engineering, biology and medicine. World Scientific Publishing, Singapore, p 104. Chap 8 5. Stępień M, Latos-Grażyński L (2005) Acc Chem Res 38:88 6. Latos-Grażyński L, Rachlewicz K, Wojaczyński J (1999) Coord Chem Rev 190–192:109 7. Lash TD (2015) Org Biomol Chem 13:7846 8. Lash TD (2014) Chem Asian J 9:682 9. Lash TD (2017) Chem Rev 117:2313 10. Berlicka A, Latos-Grażyński L (2020) J Porphyrins Phtalocyanines 24:1 11. Lee CH, Roznyatovskiy V, Hong SJ, Sessler JL (2011) Porphyrins and porphyrin analogs bearing double bonds at one or more meso-positions. In: Kadish KM, Smith KM, Guilard R (eds) Handbook of porphyrin science: with applications to chemistry, physics, materials science, engineering, biology and medicine. World Scientific Publishing, Singapore, p 197. Chap 13 12. Chmielewski PJ, Latos-Grażyński L, Rachlewicz K, Głowiak T (1994) Angew Chem Int Ed Engl 33:779 13. Furuta H, Asano T, Ogawa T (1994) J Am Chem Soc 116:767 14. Garbicz M, Latos-Grażyński L (2019) Angew Chem Int Ed 58:6089 15. Stępień M, Latos-Grażyński L (2002) J Am Chem Soc 124:3838 16. Stępień M, Latos-Grażyński L, Szterenberg L, Panek J, Latajka Z (2004) J Am Chem Soc 126:4566 17. Szyszko B, Latos-Grażyński L, Szterenberg L (2011) Angew Chem Int Ed 50:6587 18. Idec A, Szterenberg L, Latos-Grażyński L (2015) Chem A Eur J 21:12481 19. Idec A, Skonieczny J, Latos-Grażyński L, Pawlicki M (2016) Eur J Org Chem 2016:3691 20. Idec A, Pawlicki M, Latos-Grażyński L (2019) Chem A Eur J 25:200 21. Myśliborski R, Latos-Grażyński L (2005) Eur J Org Chem:5039 22. Myśliborski R, Latos-Grażyński L, Szterenberg L (2006) Eur J Org Chem:3064 23. Hurej K, Pawlicki M, Szterenberg L, Latos-Grażyński L (2016) Angew Chem Int Ed 55:1427 24. Hurej K, Pawlicki M, Latos-Grażyński L (2017) Chem A Eur J 23:2059 25. Hurej K, Pawlicki M, Latos-Grażyński L (2018) Chem A Eur J 23:115 26. Hung CH, Chang FC, Lin CY, Rachlewicz K, Stępień M, Latos-Grażyński L, Lee GH, Peng SM (2004) Inorg Chem 43:4118 27. Casitas A, Canta M, Sola M, Costas M, Ribas X (2011) J Am Chem Soc 133:19386 28. Bergbreiter DE, Osburn PL, Liu YS (1999) J Am Chem Soc 121:9531 29. Kraatz HB, Milstein D (1995) J Organomet Chem 488:223 30. Canty AJ, Minchin NJ, Skelton BW, White AH (1987) J Chem Soc Dalton Trans:1477 31. Albrecht M, van Koten G (2001) Angew Chem Int Ed 40:3750 32. van der Boom ME, Milstein D (2003) Chem Rev 103:1759

A Pincer Motif Etched into a meta-Benziporphyrin Frame

223

33. Benito-Garagorri D, Kirchner K (2008) Acc Chem Res 41:201 34. Whited MT, Grubbs RH (2009) Acc Chem Res 42:1607 35. Choi J, MacArthur AH, Brookhart M, Goldman AS (2011) Chem Rev 111:1761 36. Gunanathan C, Milstein D (2011) Acc Chem Res 44:588 37. van Koten G, Gebbink R (2011) Dalton Trans 40:8731 38. van Koten G (2013) The Monoanionic ECE-pincer ligand: a versatile privileged ligand platform-general considerations. Topics in organometallic chemistry. Springer, Cham, p 1 39. Lawrence MAW, Green KA, Nelson PN, Lorraine SC (2018) Polyhedron 143:11 40. Gorgas N, Kirchner K (2018) Acc Chem Res 51:1558 41. van Koten G (2019) NCN-pincer organometallics in bioorganometallic chemistry. Advances in bioorganometallic chemistry. Elsevier, Amsterdam, p 113. Chap 6 42. Valdes H, Rufino-Felipe E, Morales-Morales D (2019) J Organomet Chem:898 43. Alig L, Fritz M, Schneider S (2019) Chem Rev 119:2681 44. Haque A, Xu LL, Al-Balushi RA, Al-Suti MK, Ilmi R, Guo ZL, Khan MS, Wong WY, Raithby PR (2019) Chem Soc Rev 48:5547 45. Li J, Siegler M, Lutz M, Spek AL, Gebbink RJ, van Koten G (2010) Adv Synth Catal 352:2474 46. Bonnet S, van Lenthe JH, van Dam HJ, van Koten G, Gebbink RJ (2011) Dalton Trans 40:2542 47. Gagliardo M, Snelders DJ, Chase PA, Gebbink RJ, van Klink GR, van Koten G (2007) Angew Chem Int Ed 46:8558 48. Kohl SW, Weiner L, Schwartsburd L, Konstantinovski L, Shimon LJ, Ben-David Y, Iron MA, Milstein D (2009) Science 324:74 49. Kickham JE, Loeb SJ (1995) Inorg Chem 34:5656 50. Font M, Ribas X (2016) Pincerlike cyclic systems for unraveling fundamental coinage metal redox processes. Topics in organometallic chemistry. Springer, Cham, p 269 51. Casitas A, King AE, Parella T, Costas M, Stahl SS, Ribas X (2010) Chem Sci 1:326 52. Ribas X, Calle C, Poater A, Casitas A, Gomez L, Xifra R, Parella T, Benet-Buchholz J, Schweiger A, Mitrikas G, Sola M, Llobet A, Stack T (2010) J Am Chem Soc 132:12299 53. Lash TD, Chaney ST, Richter DT (1998) J Org Chem 63:9076 54. Darrow WT, Lash TD (2017) J Porphyrins Phtalocyanines 21:532 55. Lash TD, Yant VR (2009) Tetrahedron 65:9527 56. Stępień M, Latos-Grażyński L, Szterenberg L (2004) Inorg Chem 43:6654 57. Lash TD, Toney AM, Castans KM, Ferrence GM (2013) J Org Chem 78:9143 58. Fosu SC, Ferrence GM, Lash TD (2014) J Org Chem 79:11061 59. Hung CH, Chang GF, Kumar A, Lin GF, Luo LY, Ching WM, Diau EWG (2008) Chem Commun:978 60. Chang GF, Kumar A, Ching WM, Chu HW, Hung CH (2009) Chem Asian J 4:164 61. Chang GF, Wang CH, Lu HC, Kan LS, Chao I, Chen WH, Kumar A, Lo LY, dela Rosa MAC, Hung CH (2011) Chem A Eur J 17:11332 62. Huang CS, Li YL, Yang JE, Cheng N, Liu HB, Li YJ (2010) Chem Commun 46:3161 63. Singh K, Abebayehu A, Mulugeta E, Sareen D, Lee CH (2016) J Porphyr Phthalocyanines 20:21 64. Jeong SD, Nowak-Krol A, Kim Y, Kim SJ, Gryko DT, Lee CH (2010) Chem Commun 46:8737 65. Park D, Jeong SD, Ishida M, Lee CH (2014) Chem Commun 50:9277 66. Abebayehu A, Park D, Hwang S, Dutta R, Lee CH (2016) Dalton Trans 45:3093 67. Jeong SD, Hong SJ, Park KJ, Ham S, Sessler JL, Lynch V, Lee CH (2007) J Org Chem 72:6232 68. Brückner C, Akhigbe J, Samankumara LP (2014) Porphyrin analogs containing non-pyrrolic heterocycles. In: Kadish KM, Smith KM, Guilard R (eds) Handbook of porphyrin science: with applications to chemistry, physics, materials science, engineering, biology and medicine. World Scientific Publishing, Singapore, p 4. Chap 164

224

K. Hurej and L. Latos-Grażyński

69. Lash TD, Pokharel K, Serling JM, Yant VR, Ferrence GM (2007) Org Lett 9:2863 70. Berlin K, Breitmaier E (1994) Angew Chem Int Ed Engl 33:219 71. Sim EK, Jeong SD, Yoon DW, Hong SJ, Kang Y, Lee CH (2006) Org Lett 8:3355 72. Szyszko B, Latos-Grażyński L (2015) Chem Soc Rev 44:3588 73. Stępień M, Latos-Grażyński L (2003) Org Lett 5:3379 74. Stępień M, Latos-Grażyński L (2003) Inorg Chem 42:6183 75. Stępień M, Latos-Grażyński L, Szterenberg L (2007) J Org Chem 72:2259 76. El-Beck JA, Lash TD (2006) Org Lett 8:5263 77. Lash TD (1995) Angew Chem Int Ed Engl 34:2533 78. Richter DT, Lash TD (2001) Tetrahedron 57:3657 79. Stępień M, Latos-Grażyński L, Lash TD, Szterenberg L (2001) Inorg Chem 40:6892 80. Lash TD, Szymanski JT, Ferrence GM (2007) J Org Chem 72:6481 81. Szymanski JT, Lash TD (2003) Tetrahedron Lett 44:8613 82. Ryppa C, Niedzwiedzki D, Morozowich NL, Srikanth R, Zeller M, Frank HA, Bruckner C (2009) Chem A Eur J 15:5749 83. Kozyrev AN, Alderfer JL, Dougherty TJ, Pandey RK (1999) Angew Chem Int Ed 38:126 84. Toganoh M, Furuta H (2010) Synthesis and metal coordination of N-confused and N-fused porphyrinoids. In: Kadish KM, Smith KM, Guilard R (eds) Handbook of porphyrin science: with applications to chemistry, physics, materials science, engineering, biology and medicine. World Scientific Publishing, Singapore, p 295. Chap 8 85. Chmielewski PJ, Latos-Grażyński L (2005) Coord Chem Rev 249:2510 86. Pacholska-Dudziak E, Latos-Grażyński L (2007) Eur J Inorg Chem:2594 87. Ishizuka T, Ikeda S, Toganoh M, Yoshida I, Ishikawa Y, Osuka A, Furuta H (2008) Tetrahedron 64:4037 88. Toganoh M, Furuta H (2012) Chem Commun 48:937 89. Myśliborski R, Rachlewicz K, Latos-Grażyński L (2007) J Porphyrins Phtalocyanines 11:172 90. Lash TD, Young AM, Rasmussen JM, Ferrence GM (2011) J Org Chem 76:5636 91. Lash TD, Rasmussen JM, Bergman KM, Colby DA (2004) Org Lett 6:549 92. Adams KR, Bonnett R, Burke PJ, Salgado A, Valles MA (1993) J Chem Soc Chem Commun:1860 93. Adams KR, Bonnett R, Burke PJ, Salgado A, Valles MA (1997) J Chem Soc Perkin Trans:11769 94. Myśliborski R, Rachlewicz K, Latos-Grażyński L (2006) Inorg Chem 45:7828 95. Ellis D, McKay D, Macgregor SA, Rosair GM, Welch AJ (2010) Angew Chem Int Ed 49:4943 96. Kaplan L, Wendling LA, Wilzbach KE (1971) J Am Chem Soc 93:3819 97. Krempner C, Reinke H, Wustrack R (2007) Inorg Chem Comm 10:239 98. Pope RM, VanOrden SL, Cooper BT, Buckner SW (1992) Organometallics 11:2001 99. Sattler A, Parkin G (2009) Nature 463:523 100. Hu S, Shima T, Hou Z (2014) Nature 512:413 101. Kvaskoff D, Lueerssen H, Bednarek P, Wentrup C (2014) J Am Chem Soc 136:15203 102. Jakoobi M, Sergeev AG (2019) Chem Asian J 14:2181 103. Hicks J, Vasko P, Goicoechea JM, Aldridge S (2019) J Am Chem Soc 141:11000 104. Zhang XX, Parks GF, Wayland BB (1997) J Am Chem Soc 119:7938 105. Hurej K (2017) PhD thesis, University of Wrocław 106. Frech CM, Shimon LJ, Milstein D (2007) Chem A Eur J 13:7501 107. Frech CM, Milstein D (2006) J Am Chem Soc 128:12434 108. Gandelman M, Konstantinovski L, Rozenberg H, Milstein D (2003) Chem A Eur J 9:2595 109. Vigalok A, Rybtchinski B, Shimon LJW, Ben-David Y, Milstein D (1999) Organometallics 18:895 110. Chevrier B, Weiss R, Chottard JC, Lange M, Mansuy D (1981) J Am Chem Soc 103:2899 111. Szyszko B, Kupietz K, Szterenberg L, Latos-Grażyński L (2014) Chem A Eur J 20:1376 112. Stępień M, Latos-Grażyński L, Sprutta N, Chwalisz P, Szterenberg L (2007) Angew Chem Int Ed 46:7869

A Pincer Motif Etched into a meta-Benziporphyrin Frame

225

113. Stępień M, Szyszko B, Latos-Grażyński L (2010) J Am Chem Soc 132:3140 114. Grzegorzek N, Pawlicki M, Szterenberg L, Latos-Grażyński L (2009) J Am Chem Soc 131:7224 115. Pawlicki M, Kańska I, Latos-Grażyński L (2007) Inorg Chem 46:6575 116. Stępień M, Sprutta N, Latos-Grażyński L (2011) Angew Chem Int Ed 50:4288 117. Singh R, Mukherjee A (2019) ACS Catal 9:3604 118. Che CM, Lo VKY, Zhou CY, Huang JS (2011) Chem Soc Rev 40:1950 119. Niino T, Toganoh M, Andrioletti B, Furuta H (2006) Chem Commun:4335 120. Fields KB, Engle JT, Sripothongnak S, Kim C, Zhang XP, Ziegler CJ (2011) Chem Commun 47:749 121. Rachlewicz K, Wang SL, Ko JL, Hung CH, Latos-Grażyński L (2004) J Am Chem Soc 126:4420 122. Pawlicki M, Latos-Grażyński L (2005) J Org Chem 70:9123 123. Chmielewski MJ, Pawlicki M, Sprutta N, Szterenberg L, Latos-Grażyński L (2006) Inorg Chem 45:8664 124. Pacholska-Dudziak E, Skonieczny J, Pawlicki M, Szterenberg L, Ciunik Z, Latos-Grażyński L (2008) J Am Chem Soc 130:6182 125. Idec A, Pawlicki M, Latos-Grażyński L (2017) Inorg Chem 56:10337 126. Grzegorzek N, Nojman E, Szterenberg L, Latos-Grażyński L (2013) Inorg Chem 52:2599 127. Grocka I, Latos-Grażyński L, Stępień M (2013) Angew Chem Int Ed 52:1044 128. Białek MJ, Chmielewski PJ, Latos-Grażyński L (2019) Chem A Eur J 25:14536 129. Lash TD, Colby DA, Szczepura LF (2004) Inorg Chem 43:5258 130. Szyszko B, Latos-Grażyński L (2011) Organometallics 30:4354 131. Szyszko B, Latos-Grażyński L, Szterenberg L (2012) Chem Commun 48:5004 132. Gao RX, AbuSalim DI, Lash TD (2017) J Org Chem 82:6680 133. Szyszko B, Szterenberg L, Białońska A, Latos-Grażyński L (2015) Angew Chem Int Ed 54:4932 134. Myśliborski R, Latos-Grażyński L, Szterenberg L, Lis T (2006) Angew Chem Int Ed 45:3670 135. Carré FH, Corriu RJP, Bolin G, Moreau JJE, Vernhet C (1993) Organometallics 12:2478 136. Corriu RJP, Bolin G, Moreau JJE, Vernhet C (1991) J Chem Soc Chem Commun:211 137. Abebayehu A, Dutta R, Lee CH (2016) Chem A Eur J 22:13850 138. Jeong SD, Sessler JL, Lynch V, Lee CH (2008) J Am Chem Soc 130:390 139. Sengupta R, Thorat KG, Ravikanth M (2018) J Org Chem 83:11794 140. Brewster JT, He Q, Anguera G, Moore MD, Ke XS, Lynch VM, Sessler JL (2017) Chem Commun 53:4981 141. Brewster JT, Mangel DN, Gaunt AJ, Saunders DP, Zafar H, Lynch VM, Boreen MA, Garner ME, Goodwin CAP, Settineri NS, Arnold J, Sessler JL (2019) J Am Chem Soc 141:17867

Top Organomet Chem (2021) 68: 227–262 https://doi.org/10.1007/3418_2020_66 # Springer Nature Switzerland AG 2020 Published online: 30 October 2020

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed Hydrogenation/ Dehydrogenation Reactions Stefan Weber and Karl Kirchner

Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Hydrogenation Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Hydrogenation of Aldehydes and Ketones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Hydrogenation of Esters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Hydrogenation of Amides, Imines, Nitriles, and Heterocycles . . . . . . . . . . . . . . . . . . . . . . 2.4 Reduction of Carbon Dioxide, Carbonates, and Carbamates . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Hydrogenation of Alkenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Dehydrogenation and Coupling Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Acceptorless Dehydrogenative Coupling (ADC) Reactions . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Synthesis of Aldimines, Cyclic Imides, and Amides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Synthesis of Esters and Functionalization of Nitriles and Alkanes . . . . . . . . . . . . . . . . . . 3.4 Synthesis and Derivatization of Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Hydrogen-Borrowing Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Alkylation of Amines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7 Alkylation of Alcohols and Ketones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.8 Alkylation of Nitriles and Sulfonamide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.9 Upgrading of Ethanol into 1-Butanol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

228 229 230 233 235 239 242 243 243 243 246 248 249 250 252 257 257 258 259

Abstract The usage of earth-abundant metals as catalysts for chemical synthesis in order to install more sustainable reactions is a major goal in modern synthesis. Within the last few years, well-defined manganese complexes appeared in academic research and were proven to be a powerful player in the field of benign oxidation and reduction reactions. Hydrogenation of polarized double bonds such as aldehydes, ketones, esters, amides, and nitriles, but also carbon-carbon double bonds, can efficiently be achieved by well-defined manganese complexes. In the case of

S. Weber and K. Kirchner (*) Institute of Applied Synthetic Chemistry, Vienna University of Technology, Vienna, Austria e-mail: [email protected]

228

S. Weber and K. Kirchner

oxidation reactions, typical condensation reactions such as aldol condensation or Michael addition may be carried out with alcohols as starting material by in situ oxidation to carbonyl moieties, employing finely tuned manganese complexes. In this book chapter, we describe the development of the emerging field of manganesecatalyzed hydrogenation/dehydrogenation reactions in conjunction with metalligand cooperation processes. Keywords Bidentate ligands · Dehydrogenation reactions · Homogeneous catalysis · Hydrogenation reactions · Manganese complexes · Pincer ligands

1 Introduction The development of sustainable synthesis is one of the major goals in modern chemistry. Within this context, the use of catalysts in order to substitute procedures employing reagents is highly favorable. This leads to more cost-efficient and environmentally benign processes. In the last decades, a broad variety of different catalytic systems were implemented in modern organic synthesis. Most of these catalysts are based on precious metals such as palladium, platinum, ruthenium, or rhodium. These systems often show high reactivity and stability. However, the amount of precious metals is limited, since the production from natural resources is limited. Although recycling systems were developed over the last decades, a supply risk may occur in the future. Therefore, the substitution of precious metals by inexpensive and earth-abundant elements seems to be highly favorable [1]. Interestingly, the use of manganese(I) [1–3] complexes for organic synthesis was neglected until 2016. Manganese is the third most abundant metal in the earth’s crust. Apart from that, a broad variety of well-defined manganese(I) complexes can be easily synthesized from the air- and moisture-stable, commercially available manganese(I) pentacarbonyl halide [Mn(CO)5X] (X ¼ Cl, Br) precursors. Due to the presence of carbonyl ligands, most of the synthesized complexes are diamagnetic d6-low-spin systems, which allow characterization not only via infrared analysis but also via NMR spectroscopy. Apart from that, the strongly bonded carbonyl ligands stabilize the complex in this oxidation state and typically lead to air- and moisturestable complexes [4–11]. Very shortly after the first seminal reports in (de)hydrogenation reactions, a growing number of organic and organometallic chemists used a broad variety of well-defined manganese(I) systems for sustainable oxidation and reduction reactions [12, 13]. Many of these catalytic reactions involve metal-ligand cooperation (MLC). Within this book chapter, we describe the role of well-defined manganese complexes in catalytic hydrogenation and dehydrogenation reactions. Since most of the systems strongly rely on the possibility of metal-ligand cooperativity, we will focus on the

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

229

Chart 1 Commonly used chelate ligands capable of metal-ligand cooperation (MLC)

role of this behavior in manganese(I) catalysis. Chart 1 displays an overview of frequently used ligands in the field of manganese-catalyzed (de)hydrogenation reactions. In context, a broad variety of aliphatic or aromatic pincer ligands with different linkers are used. Apart from that, bidentate ligands are also frequently employed.

2 Hydrogenation Reactions The reduction of unsaturated compounds, such as polar carbon-heteroatom multiple bonds (e.g., carbonyl derivatives, nitriles, and carbon dioxide) but also non-polar carbon-carbon multiple bonds, is of great interest in organic synthesis. The use of catalytic systems in combination with hydrogen gas as cheap and benign hydrogen source is very interesting due to the high atom efficiency with respect to reagents. Within this context, a broad variety of different, well-defined manganese(I) catalysts were introduced in the last few years.

230

2.1

S. Weber and K. Kirchner

Hydrogenation of Aldehydes and Ketones

Beller and coworkers were the first ones to report the hydrogenation of ketones and aldehyde using a well-defined manganese(I) catalyst in 2016. Within this context, a series of manganese(I) dicarbonyl as well as manganese(II) complexes, containing aliphatic PNP pincer ligands, were synthesized and characterized. Notably, only manganese(I) complexes were active for the hydrogenation of carbonyl compounds, whereas the related manganese(II) complex did not show any reactivity for this transformation. Upon addition of tBuONa as base, a broad variety of aldehydes and ketones could be reduced to the corresponding alcohols employing catalyst Mn1. This procedure displayed high chemoselectivity, whereas esters, amides, and non-conjugated and conjugated C-C double and triple bonds were not reduced. The general reaction scheme is depicted in Scheme 1. Mechanistic investigation revealed the formation of a five-coordinated manganese dicarbonyl intermediate containing a deprotonated pincer ligand. Treatment of this complex with hydrogen gas gave rise to a hydride complex via a typical metal-ligand cooperation process [14]. Shortly after that, Kempe and coworkers reported on a triazine-based PNP-supported manganese(I) complex for the hydrogenation of carbonyl compounds (Scheme 2). Within this seminal work, a broad variety of ketones and

Scheme 1 Hydrogenation of aldehydes and ketones catalyzed by Mn1

Scheme 2 Hydrogenation of aldehydes and ketones catalyzed by M2

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

231

Scheme 3 Hydrogenation of ketones catalyzed by Mn3

Scheme 4 Additive-free, highly selective hydrogenation of aldehydes catalyzed by Mn4

aldehydes were reduced under comparatively mild conditions and short reaction times with a catalyst loading of merely 0.1 mol% Mn2 for most substrates [15]. In 2017, Sortais and coworkers utilized a cationic manganese(I) tricarbonyl PNP pincer catalysts based on a 2,6-diaminopyridine scaffold for the hydrogenation of ketones. This protocol, however, required relatively harsh reaction conditions (10 mol% of tBuOK, 130 C) and 5 mol% of pre-catalyst (Scheme 3) [16]. In contrast, the Kirchner group described a highly chemoselective hydrogenation of aldehydes using a manganese hydride complex based on the same PNP pincer ligand system (Scheme 4). This additive-free protocol operated at room temperature with a catalyst loading as low as 0.05 mol%. If three equivalents of DBU (DBU ¼ 1,8-diazabicyclo(5.4.0)undec-7-ene) relative to the catalyst were added, turnover numbers of up to 10,400 were achieved. Furthermore, a remarkable chemoselectivity was observed, whereas acetals and conjugated and non-conjugated C-C double bonds and even ketones were not reduced. The investigation of different linkers (NH-, NMe-, and CH2-) revealed that no reaction took place when metal-ligand cooperation was blocked as in the case of NMe-linkers. If the NH-linker was substituted by a methylene group, only a poor reactivity was observed, underlining the importance of the NH-linker motive for this reaction [17]. Sortais and coworkers reported on the use of PN-based manganese(I) tricarbonyl complexes with CH2- and NH-linkers for the hydrogenation of ketones and aldehydes. Within this work, a superior performance of the NH-based systems was presented. Under optimized conditions, as depicted in Scheme 5, only 0.5 mol% Mn5 and 2 mol% of KHMDS were needed under mild reaction conditions (50 C) [18].

232

S. Weber and K. Kirchner

Scheme 5 Hydrogenation of aldehydes and ketones by PN-supported Mn5

Scheme 6 Hydrogenation of ketones catalyzed by Mn6

Recently, Sortais and coworkers reported on the use of a manganese (I) tricarbonyl complex featuring a PC bidentate ligand with NHC moiety which exhibited an interesting MLC reaction. When Mn6 was treated with KHMDS, deprotonation of the methylene linker was observed leading to the formation of a phosphonium ylide. This complex reacted at room temperature with hydrogen gas or CO2 yielding the corresponding hydride and formate complexes. Probing the catalytic performance of Mn6 revealed high reactivity for the hydrogenation of ketones with catalyst loadings of 0.1–1 mol% (Scheme 6) [19]. Since the reduction of prochiral substrates, such as ketones, results in the formation of a chiral product, several manganese-catalyzed enantioselective hydrogenations were developed. An overview on several enantioselective procedures for the hydrogenation of ketones is depicted in Scheme 7. Clarke and coworkers were the first ones describing a well-defined manganese (I) complex for the enantioselective reduction of ketones using hydrogen gas. A cationic manganese(I) tricarbonyl complex supported by a PNN ligand gave up to 97% ee under mild reaction conditions (50 C) with only 1 mol% catalyst loading of Mn7 [20]. Shortly after that, the group of Beller reported on the use of an aliphatic PNP system containing a chiral phospholane motive (Mn8). The hydrogenation of ketones took place at 40 C with a catalyst loading of 1 mol%. It is noteworthy to say that higher enantioselectivity was achieved for aliphatic systems, which is quite unusual, whereas the enantioselectivity for aromatic systems was moderate [21]. More recently, Han, Ding, and coworkers reported on the use of a PNN pincer system containing a chiral phospholane system in order to induce chirality. Remarkably, this system operated at room temperature. Within this context, a very broad

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

233

Scheme 7 Enantioselective hydrogenation of ketones catalyzed by Mn7, Mn8, and Mn9

variety of different aromatic ketones could be converted to the corresponding alcohols with up to 98% ee and turnover numbers up to 9,800, if a catalytic amount of hexafluoro isopropanol (HFIP) was added [22].

2.2

Hydrogenation of Esters

Beller and coworkers were the first ones to report manganese-catalyzed hydrogenation of esters to alcohols in 2016. For this purpose, aliphatic pincer PNP ligands were used. The investigation of the substituents on the phosphorous donors revealed moderate reactivity for bulky groups such as isopropyl or cyclohexyl. In contradiction to that, ethyl groups on the donors resulted in high reactivity.

234

S. Weber and K. Kirchner

Scheme 8 Hydrogenation of esters catalyzed by the cationic complex Mn10

Scheme 9 Hydrogenation of esters catalyzed by Mn7 and Mn7’

For comparison, the cationic tricarbonyl complex Mn10, where the ligand is coordinated in a facial rather than in a meridional fashion, gave identical yields. Investigation of the substrate scope revealed high reactivity for aromatic and aliphatic esters as well as for lactones, yielding diols (Scheme 8) [23]. Shortly after that, Clarke and coworkers reported on the use of a chiral, cationic tricarbonyl complex Mn7 for the hydrogenation of esters to primary alcohols. Within this seminal work, 1 mol% of catalyst was sufficient for the reduction of aromatic and aliphatic esters [20]. Further improvement with the racemate Mn7’ was reported in 2018. The catalyst loading could be lowered to 0.1 mol% for most substrates. Furthermore, the reaction temperature could be decreased to 50 C. K2CO3 was introduced as a weaker base in comparison to tBuOK in the previous work (Scheme 9). In addition to that, hydrogenation of α-chiral esters with essentially no loss of enantiopurity was reported within this context [24].

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

235

Scheme 10 Reductive cleavage of esters catalyzed by Mn11

Scheme 11 Hydrogenation of esters catalyzed by PN-supported Mn 12

Milstein and coworkers introduced a PNN-based system for the hydrogenation of esters as depicted in Scheme 10. This procedure operated at 100 C with addition of KH as base. Mechanistic investigation revealed the presence of two different hydride complexes upon treatment of the deprotonated species with hydrogen gas. The formed complexes were characterized via nuclear Overhauser-enhanced 1H-NMR analysis. In one species the hydride is syn to the proton of the amine group, whereas the hydride is anti to the amine proton of the other isomer. The two species seemed to be in equilibrium with one another during catalysis [25]. Pidko and coworkers introduced a neutral tricarbonyl complex supported by a relatively simple aliphatic PN bidentate ligand for the reductive cleavage of esters as depicted in Scheme 11. A catalyst loading of only 0.2 mol% was sufficient for the hydrogenation of aromatic and aliphatic esters. Unfortunately, 75 mol% of tBuOK were required to achieve high reactivity [26].

2.3

Hydrogenation of Amides, Imines, Nitriles, and Heterocycles

Beller and coworkers were the first ones to report on the hydrogenation of amides yielding alcohols and amines in 2017. A manganese(I) complex featuring a PNN ligand was used which coordinated in a facial rather than in a meridional pincer-like manner (Scheme 12). A broad variety of activated and unactivated aromatic amides were hydrogenated to the corresponding amines and primary alcohols [27].

236

S. Weber and K. Kirchner

Scheme 12 Hydrogenation of amides yielding alcohols and amines catalyzed by Mn13

Scheme 13 Lewis acid-assisted hydrogenation of amides to amines mediated by Mn14

Scheme 14 Hydrogenation of in situ formed aldimines to amines catalyzed by Mn5

In 2018, Milstein and coworkers introduced a protocol for the hydrogenation of amides to amines. In this seminal work, a lutidine-based PNP pincer ligand turned out to be the best candidate for this transformation as depicted in Scheme 13. Addition of B(C6F5)3 as Lewis acid increased the reactivity of the system drastically involving coordination on the oxygen of the amide group and the hemiaminal intermediate [28]. An interesting pathway for reductive amination was reported by Sortais and coworkers in 2018. After condensation of aldehyde and amine, the so-formed imine was hydrogenated to the corresponding amine as shown in Scheme 14. A broad variety of aromatic and aliphatic in situ formed imines were successfully reduced by employing Mn5 as pre-catalyst. This procedure showed high functional group tolerance, whereas esters, amides, nitriles, and ketones were not hydrogenated. However, this concept was limited to in situ formed aldimines [29].

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

237

Scheme 15 Chemoselective hydrogenation of aldimines and ketimines catalyzed by Mn15

Scheme 16 Hydrogenation of N-heterocycles catalyzed by Mn13

Recently, Kempe and coworkers reported on the use of a triazine-based PNP dicarbonyl complex (Mn15) for the hydrogenation of aldimines and ketimines. This methodology showed high functional group tolerance, whereas esters, nitriles, alkenes, nitro groups, and even ketones were not reduced. The reaction proceeded at 50 C with a catalyst loading as low as 0.4 mol% at 20 bar H2 pressure as depicted in Scheme 15. Detailed mechanistic investigations revealed that a series of manganate species were involved during the catalysis [30]. Lan, Liu, and coworkers reported on the first hydrogenation of N-heterocycles. Within this context, a broad variety of different aliphatic PNP- and PNN-supported complexes were investigated. Detailed IR analysis of the corresponding amido complexes allowed the ranking of different donation groups regarding electron donation. The combination of single crystal analysis for several amido complexes and DFT calculations resulted in detailed insight on the steric hindrance of the investigated donating groups. The gained knowledge on the steric and electronic parameters of aliphatic pincer-based catalyst was applied in hydrogenation of several carbonyl groups, including ketones, esters, and amides, whereas an imidazole-based PNN ligand (Mn13) showed the highest reactivity due to high electron-donating properties and low steric hindrance. Apart from that, a broad variety of different six-membered N-heterocycles were reported (Scheme 16). Mechanistic studies revealed a stepwise hydrogenation to 1,2-dihydroquinolines followed by isomerization to 3,4-dihydroquinolines [31].

238

S. Weber and K. Kirchner

In 2016, the Beller group reported the hydrogenation of nitriles using a welldefined manganese catalyst. Within this context, an aliphatic PNP pincer ligand with i Pr groups connected with the phosphorus showed the highest reactivity. Mechanistic investigations strongly indicated an outer-sphere mechanism supported by typical MLC [14]. Thus far, the seminal work of Beller and coworkers remains the only example for the hydrogenation of nitriles via a typical MLC upon manganese-based systems. Apart from that, our group was the first ones to report on the hydrogenation of nitriles and ketones, using a bisphosphine ligand which is not capable of MLC [32]. Shortly after that, the group of Garcia reported on the hydrogenation of nitriles, using a similar system. They used 2-BuOH as solvent resulting in hydrogenation utilizing dihydrogen but also transfer hydrogenation where the solvent acted as hydrogen source [33]. An overview of manganese-based hydrogen of nitriles is given in Scheme 17. Very recently, our group reported on the first additive-free hydrogenation of nitriles to primary amines. Substitution of the bromide ligand via a methyl group resulted in a surprisingly stable alkyl complex (Mn18), which can be stored on the bench for at least 1 month without any notable decomposition. The reaction mechanism was investigated by DFT calculations. Activation of the pre-catalyst was

Scheme 17 Hydrogenation of nitriles catalyzed by Mn1, Mn16, and Mn17

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

239

Scheme 18 Additive-free hydrogenation of nitriles via an inner-sphere mechanism

achieved by migratory insertion reaction, yielding an acyl ligand and a vacant coordination side. After complexation of hydrogen gas, proton abstraction of the strongly basic acyl ligand from the coordinated dihydrogen resulted in the formation of a hydride complex and a very loosely bonded aldehyde. Upon release of the coordinated aldehyde (or alternatively reduction to the corresponding alcohol), a vacant side was created, allowing coordination of nitrile substrate and reduction via an inner-sphere mechanism (Scheme 18) [34].

2.4

Reduction of Carbon Dioxide, Carbonates, and Carbamates

Reduction of carbon dioxide can be achieved in a direct way, yielding formates, formaldehyde, or methanol. In cooperation with Gonsalvi and coworkers, our group was the first one to report manganese-catalyzed direct hydrogenation of CO2 to formate. Turnover numbers of up to 10,000 could be achieved. Interestingly, when LiOTf was used as co-catalyst, the reactivity could drastically be increased, and turnover numbers greater than 30,000 could be achieved (Scheme 19) [35]. Nervi, Khusnutdinova, and coworkers reported on a modified bipyridine-based tricarbonyl complex for the hydrogenation of CO2 to formate with turnover numbers up to 6,250, using DBU as base. Interestingly under slightly more forcing conditions and diethylamine as base, the corresponding formamide was formed with turnover numbers up to 588 as depicted in Scheme 20 [36]. Shortly after that, the group of Prakash reported on the N-formylation of primary and secondary amines utilizing CO2 as formylation agent yielding formamides.

240

S. Weber and K. Kirchner

Scheme 19 Hydrogenation of CO2 to formate catalyzed by Mn4

Scheme 20 Hydrogenation of CO2 to formate, amides, and methanol catalyzed by Mn1 and Mn19

Apart from that, the hydrogenation of CO2 to methanol was reported. The stepwise reduction preceded via formamide intermediates. The consecutive hydrogenation of the amide intermediates resulted in the formation of methanol and generation of amines (Scheme 20). Turnover numbers of up to 36 were reported [37]. Apart from direct CO2 hydrogenation, carbon dioxide-derived substances such as (cyclic) carbonates, carbamates, or urea derivatives may be used in an indirect route of CO2 reduction to methanol. In fact, Milstein and coworkers were the first ones to

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

241

Scheme 21 Hydrogenation of carbonates catalyzed by Mn1, Mn11, and Mn20

report on the hydrogenation of (cyclic) carbonates to methanol and alcohols (or diols for cyclic carbonates) as shown in Scheme 21. The catalyst Mn11 was used bearing a PNN pincer ligand. A broad variety of different carbonates were efficiently converted using 2 mol% of catalyst and 4 mol% KH as base at 110 C. Furthermore, the depolymerization of poly(propylene carbonate) yields propylene diol, propylene carbonate, and methanol [38]. Very shortly after this discovery, Cavallo, El-Sepelgy, Rueping, and coworkers reported on the hydrogenation of cyclic carbonates using a cationic PNN tricarbonyl complex [39]. Independently from the abovementioned seminal works, Leitner and coworkers reported on the hydrogenation of cyclic carbonates to diols and methanol using the PNP pincer-supported complex Mn1 [40]. An overview of manganese-catalyzed hydrogenation of carbonates is given in Scheme 21. Recently, the group of Milstein reported the hydrogenation of carbamates yielding amines and alcohols including methanol. The utilization of a bipyridine-based PNN pincer ligand afforded the best results. Apart from that, the uncommon

242

S. Weber and K. Kirchner

Scheme 22 Hydrogenation of carbamates and urea derivatives catalyzed by Mn21

Scheme 23 Additive-free hydrogenation of alkenes via inner-sphere mechanism

hydrogenation of symmetric urea derivatives gave primary amines and methanol in excellent yields as displayed in Scheme 22 [41].

2.5

Hydrogenation of Alkenes

The Kirchner group was the first one to report on the hydrogenation of olefins using a well-defined manganese catalyst in 2019. In contrast to the commonly employed MLC mode allowing outer-sphere-type hydrogenations, the novel alkyl complexes employed an inner-sphere mechanism which was initiated by migratory insertion of the alkyl group into the carbonyl ligand. Within this context, a broad variety of different mono- and disubstituted alkenes were efficiently reduced. Remarkably, the hydrogenation of mono- and 1,1-disubstituted alkenes was performed at room temperature. Apart from that, high functional group tolerance was achieved, while alcohols, esters, acetals, and even anhydrides were tolerated (Scheme 23) [42].

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

243

3 Dehydrogenation and Coupling Reactions 3.1

Acceptorless Dehydrogenative Coupling (ADC) Reactions

An efficient way of using alcohols as carbon source for alkylation of amines is oxidation of alcohol moieties to the corresponding carbonyl motifs, thereby releasing hydrogen gas. The so-formed carbonyl group forms an imine with the primary amine and water. A general pattern of ADC reaction is shown in Scheme 24. A broad variety of different well-defined manganese complexes were employed for the acceptorless dehydrogenative coupling of alcohols and amines within the last years. Aliphatic and aromatic PNP pincer ligands were frequently used as tridentate ligands. Most manganese complexes contained two carbonyl co-ligands. Bromide or hydrides were often used as anionic ligands. Furthermore, deprotonated ligands which are typically anionic gave rise to fivefold coordinated Mn(I) complexes.

3.2

Synthesis of Aldimines, Cyclic Imides, and Amides

In 2016, Milstein and coworkers were the first ones to report the manganesecatalyzed ADC of aromatic amines with benzylic alcohols, yielding imines. Hydrogen gas and water were the only by-products of this transformation. Within this context, a coordinatively unsaturated complex (Mn23) bearing a deprotonated PNP ligand and two carbonyl ligands was utilized. Remarkably, no base was needed for this transformation. Investigation of the reaction mechanism proved the presence of two different alkoxide-based complexes and one hydride species (Scheme 25) [43]. In the same year, our group reported the coupling of alcohols with amines. Within this seminal work, iron and manganese based upon the same PNP ligand containing NH-linkers were compared. Interestingly, the iron-based system gave rise to the

Scheme 24 General reaction scheme of acceptorless dehydrogenative coupling (ADC)

Scheme 25 Synthesis of aldimines via ADC catalyzed by Mn23

244

S. Weber and K. Kirchner

Scheme 26 Synthesis of aldimines and secondary amines catalyzed by Mn4 and Fe1

formation of secondary amines, whereas the manganese compound Mn4 selectively yielded aldimines (Scheme 26). Mechanistic studies, based on DFT calculation, revealed the reason for the selectivity of these systems. In case of Fe1, apart from the deprotonated PNP ligand, a hydride was present as second anionic ligand, which inserted into the C-N bond, forming the amine product. Since Mn4 was only capable of bearing one anionic ligand, being the deprotonated ligand, only carbonyls were coordinated to the metal center. Therefore, no insertion in the coordinated imine was possible [44]. In 2019, Madsen and coworkers investigated the use of Mn(III)-salen-based systems for ADC of amines and primary alcohols. For this system, Ca3N2 was used as base in order to active the pre-catalyst which very likely operated via MLC [45]. Shortly after that a porphyrin-supported manganese(III) complex Mn25 was reported which was active also in the absence of base. The catalyst loading could be decreased to 2 mol%; however, high reaction temperatures in refluxing mesitylene were required (Scheme 27) [46]. The group of Milstein reported the synthesis of cyclic imides by ADC of diols and primary amines as it can be seen in Scheme 28. Within this context, the performance of two lutidine-based PNN pincer complexes containing either a secondary or a tertiary amine as donors was investigated. If Mn11 was treated with KH, deprotonation of the amine was observed, whereas deprotonation and dearomatization were achieved in the case of a tertiary amine as donor. Interestingly, Mn11 showed higher reactivity for the ADC of diols and amines to form cyclic imides than the PNN-based system, containing a tertiary amine. The reason for the better performance of Mn11 was assigned to a hemilabile behavior of the PNN ligand. Further mechanistic investigations revealed a stepwise oxidation and condensation cascade via formation of lactones and ring opening yielding amides or direct amide formation [47]. An additive-free N-formylation protocol of primary amines employing methanol as formylation agent yielding N-formamides was reported by the group of Milstein in 2017 (Scheme 29). For this purpose, a PNP-supported dicarbonyl complex

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

Scheme 27 Synthesis of aldimines catalyzed by Mn24 and Mn25

Scheme 28 Synthesis of cyclic imides catalyzed by Mn11

Scheme 29 Synthesis of amides catalyzed by Mn26 and Mn27

245

246

S. Weber and K. Kirchner

containing a hydride ligand was employed. Mn26 showed interesting behavior upon heating. In contradiction to many other pincer-based systems, the benzyl carbon was deprotonated, rather than the NH functionality, forming a tetradentate ligand. Investigation of this species revealed lower catalytic activity in comparison to Mn26 [47]. With the help of DFT calculations, a typical MLC mode via a fivefold coordinated amido intermediate was suggested for the N-formylation of amines with methanol. The proposed reaction mechanism included the oxidation of methanol to formaldehyde, formation of a hemiacetal, and finally oxidation to the desired formamides [48]. In 2019 the group of Milstein reported the synthesis of secondary amides via ADC of primary alcohols with ammonia employing Mn27. In order to achieve high selectivity, stoichiometric amounts of KH were used. Interestingly, the use of catalytic amounts of base mainly resulted in the formation of aldimines [49].

3.3

Synthesis of Esters and Functionalization of Nitriles and Alkanes

In 2017, the group of Gauvin employed a fivefold coordinated aliphatic PNP-supported amido complex for the ADC of primary alcohols yielding esters (Scheme 30). In this context, Mn28 efficiently converted benzylic and aliphatic alcohols to the corresponding esters with a catalyst loading as low as 0.6 mol% [50]. Liu and coworkers employed Mn1 for dual-deoxygenative coupling of alcohols in 2018. Within this context the homo-coupling of 2-aryl-ethanols yielding diphenylpropene derivatives was reported. Interestingly, sodium formate was formed as by-product, leading to the loss of one carbon fragment during the coupling process. In addition to that, cross-coupling of 2-aryl-ethanols with primary alcohols, using an excess of the primary alcohol as coupling partner, followed by hydrogenation using a heterogeneous nickel catalyst yielded saturated systems (Scheme 31) [51]. Milstein and coworkers discovered in 2017 a base-free procedure for α-olefination of nitriles using alcohols as building blocks (Scheme 32). Mechanistic investigation revealed that the role of Mn26 was beyond the oxidation of alcohols to aldehydes. In fact, the condensation of nitriles with aldehydes was accelerated in

Scheme 30 Manganese-catalyzed ester synthesis via oxidative homo-coupling of alcohols

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

247

Scheme 31 Oxidative coupling of alcohols catalyzed by Mn1

Scheme 32 α-Olefination of nitriles via ADC employing Mn26

Scheme 33 Olefination of various heteroarenes catalyzed by Mn(I) complexes

presence of Mn26 indicating that this catalyst serves as a template for the condensation step [52, 53]. A different type of olefination reaction was independently reported by the groups of Kempe [54] and Maji [55] within a very short period of time (Scheme 33). In contrast to the seminal work of Kempe, Maji and coworkers employed [Mn(CO)5Br] and L12 as precursors rather than a well-defined complex. Mechanistic studies revealed that this manganese catalyst is able to perform AAD of the primary alcohol but also increased the rate of the condensation step indicating that the manganese complex serves as platform for C-C bond formation.

248

3.4

S. Weber and K. Kirchner

Synthesis and Derivatization of Heterocycles

In 2016, our group reported for the first time the synthesis of pyrimidines and quinolines via ADC catalyzed by a Mn4. Within this context a broad variety of different substituted quinolines and pyrimidines were successfully synthesized [56]. Shortly after that, the group of Kempe reported a four-component reaction yielding pyrimidines. Consecutive addition of different primary alcohols resulted in high selectivity [57]. Furthermore, an elegant way of synthesizing substituted pyrroles via the oxidative coupling of amino alcohols with secondary alcohols was reported by the same group [58]. An overview for the synthesis of various heterocycles via multicomponent reactions is given in Scheme 34. In 2017, the Kirchner group reported the selective aminomethylation of different aromatic compounds, using methanol in a Mannich-type reaction (Scheme 35), employing Mn4 as catalyst [59]. The synthesis of pyrazines via oxidative homo-coupling of β-amino alcohols catalyzed by Mn27 was reported by Milstein and coworkers in 2018. Apart from that, quinoxalines could be synthesized from condensation of 1,2-diamoinobenzene and 1,2-diols, which were in situ oxidized to the corresponding carbonyl compounds (Scheme 36) [60]. The synthesis of benzimidazoles (Scheme 15, Mn29), quinoxalines, pyrazines, benzothiazoles, and quinolines (Scheme 16, Mn29’) mediated by cationic manganese tricarbonyl complexes containing facial coordinating NNS ligands was reported by the group of Srimani (Scheme 37) [61, 62].

Scheme 34 Synthesis quinolines, pyrimidines, and pyrroles catalyzed by Mn4 and Mn15

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

249

Scheme 35 Functionalization of aromatic compounds catalyzed by Mn2

Scheme 36 Synthesis of pyrazines and quinoxalines catalyzed by Mn27

Scheme 37 Synthesis of various heterocycles catalyzed by Mn29 and Mn29’

3.5

Hydrogen-Borrowing Reactions

Apart from ADC reactions where hydrogen is extruded during the reaction, hydrogen-borrowing (HB) or hydrogen autotransfer (HAT) offers an interesting alternative in the utilization of alcohols as alkylation agents. Within this concept

250

S. Weber and K. Kirchner

Scheme 38 General reaction pattern of hydrogen-borrowing (HB) or hydrogen autotransfer (HAT) reactions

Scheme 39 Alkylation of amines via HB catalyzed by Mn1 and Mn30

the in situ generated hydrogen is stored on the metal center in cooperation with the ligand. The hydrogen can be used for the reduction of the in situ generated unsaturated compounds (e.g., imines or α,β-unsaturated carbonyls). A general reaction pattern is depicted in Scheme 38.

3.6

Alkylation of Amines

Beller and coworkers were the first ones to apply a HB procedure for the selective mono-alkylation of aniline derivatives with primary alcohols with a well-defined manganese catalyst (Scheme 39). Within this seminal work, they compared aliphatic PNP pincer ligands containing iPr or Cy substituents with NNN-derived sytems. These studies demonstrated that Mn1 featuring iPr substituents at the phosphorus atoms was the most active pre-catalyst. In addition, the importance of the low oxidation of the manganese(I) center was underlined, since the analogous Mn (II) complex showed almost no reactivity [63]. Very recently, Ke and coworkers reported on the alkylation of amines with primary alcohols employing Mn30 as catalyst. Notably, this procedure operated at room temperature. Apart from that, the bis-NHC-supported complex was not capable of metal-ligand cooperation. Mechanistic studies based on IR studies and DFT calculations suggested that an outer-sphere mechanism involving a tricarbonyl

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

251

Scheme 40 Mono-methylation of anilines catalyzed by Mn3 and Mn31

species bearing an alkoxide ligand is more likely than an inner-shell mechanism requiring loss of one CO ligand [64]. Since a general procedure for the mono-methylation of amines is of great interest, Beller and coworkers presented an interesting synthetic route employing methanol as methylation agent in 2017. The introduced catalytic system operated with a catalyst loading as low as 1 mol% of Mn31. This procedure allowed the selective methylation of a broad variety of anilines, whereas functional groups such as ketones, esters, amides, and alkenes were well tolerated (Scheme 40) [65]. In the same year, the group of Sortais used the cationic tricarbonyl complex Mn3 for the methylation of anilines. Although higher catalyst loadings and higher reaction temperatures were required, the amount of base could significantly be reduced to 20 mol% [66]. Balaraman and coworker investigated the role of different chelating ligands in combination with Mn(CO)5Br for the alkylation of anilines. Within this context, NNN-based ligand L13 as well as the saline-based ligand L14, both capable of MLC, revealed high reactivity (Scheme 41). However, moderate to good conversion could be achieved using dppb as bisphosphine ligand. This type of ligand is not able to operate via a typical metal-ligand cooperation mode. Unfortunately, the reason for product formation with this type of ligand remained unclear [67]. In 2019 Hultzsch and coworkers reported the alkylation of amines with primary and secondary alcohols. Within this context an in situ generated Mn(I) complex containing L10 was employed [68]. Recently, the group of Madsen reported the uncommon alkylation of secondary amines with primary alcohols, using a porphyrin-based Mn(III) system Mn25. Unfortunately, the mechanism of this unusual reaction stayed unclear within this context [46]. A highly interesting, base-switchable synthesis of amines or imines was introduced by Kempe and coworkers in 2018. The formed product was highly dependent on the use of either tBuOK or tBuONa as base. In the case of potassiumbased alkoxide, amines were generated with a selectivity of >98%, whereas the

252

S. Weber and K. Kirchner

Scheme 41 Alkylation of amines catalyzed by in situ generated catalyst and Mn25

Scheme 42 Base-dependent alkylation of amines catalyzed by Mn15 yielding amines or imines

sodium analogues resulted in the formation of imines. The reaction pattern is given in Scheme 42. The reason of this behavior was explained via an in situ generated manganate complex during the reaction. The formed potassium-manganate reacted about 40 times faster with the formed imine than the sodium-manganate, forming the amine product [69].

3.7

Alkylation of Alcohols and Ketones

Besides C-N bond formation reactions, the construction of C-C bonds is of great interest in organic chemistry. Within this context, the β-alkylation of secondary

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

253

alcohols as well as the α-alkylation of ketones can be achieved via manganesecatalyzed hydrogen autotransfer reactions. Yu and coworkers used an NN-based tricarbonyl complex containing a methoxide ligand for β-alkylation of secondary alcohols. A broad variety of aromatic and aliphatic secondary alcohols could be alkylated using Mn32. The synthetic importance of the introduced protocol could be demonstrated by selective β-alkylation of two steroid derivatives [70]. El-Sepelgy, Rueping, and coworkers compared the reactivity of different PNNand PNP-based pincer complexes for the β-alkylation of secondary alcohols using primary alcohols as alkylation agent, whereas Mn33 showed the highest reactivity of the given reaction. Apart from that, the coupling of two aliphatic alcohols was reported [71]. An interesting way of synthesizing cycloalkenes from diols and secondary alcohols or ketones was reported by the group of Leitner in 2019 employing Mn1. This procedure allowed the formation of cycloalkanes with a ring size of five to seven. However, a large excess of diols as alkylation agents had been used in order to suppress the formation of lactones as side products [72]. Very recently, the same group reported on the methylation of primary and secondary alcohols, employing methanol as carbon source. In the case of secondary alcohols as substrate, demethylation could be achieved [73]. An overview of the abovementioned reaction is given in Scheme 43. Beller and coworkers employed manganese-catalyzed HAT for the α-alkylation of ketones with primary alcohols using an aliphatic PNP-supported complex (Mn1) as it can be seen in Scheme 44. This procedure allowed the use of a broad variety of different aromatic and aliphatic ketones. In addition to that, benzylic as well as aliphatic alcohols could serve as carbon source. The potential use of the established system for fast derivatization of hormones was demonstrated, whereas estrone and testosterone derivatives could selectively be alkylated in the α-position [74]. The group of Milstein compared the performance of Mn11, Mn23, and Mn26 for the alkylation of ketones with primary alcohols, whereas Mn26 showed slightly higher reactivity for the desired reaction. In contradiction to other procedures, the installed protocol operated with an equimolar amount of ketone substrate and alcohol which lead to high atom efficiency. The β-alkylation of secondary alcohols with primary alcohols and consecutive oxidation resulted in the formation of a ketone group [75]. The investigation of NN and NNN chelating ligands for the phosphine-free α-alkylation of ketones was reported by Maji and coworkers in 2018. Upon the investigated ligands, the hydrozone-based NNN ligand revealed the highest reactivity in combination with [Mn(CO)5Br] as precursor. Unfortunately, the complexion of L12 with the precursor had to be done prior to the catalytic reaction resulting a time-consuming additional step [76]. An ADC of ketones with primary alcohols to yield α,β-unsaturated ketones was introduced by Guanathan employing Mn15’. Within this context a catalyst loading as low as 0.3 mol% with the use of Cs2CO3 as weak base was reported [77].

254

S. Weber and K. Kirchner

Scheme 43 β-alkylation of alcohols catalyzed by Mn(I) catalysts utilizing HB reactions

The group of Sortais introduced a procedure for α-methylation of various ketones using methanol as building block. Substrates, containing carbon on the α-position, were selectively mono-methylated. Remarkably, in the absence of a substituent at the α-position, double methylation could be achieved (Scheme 45) [78].

3.7.1

Alkylation of Amides and Esters

Milstein and coworkers employed Mn26 for the α-alkylation of amides and esters with primary alcohols. High catalyst loading (5 mol%), an over-stoichiometric amount of base, and four equivalents of esters or two equivalents of amide were necessary to achieve good yields [75]. El-Sepelgy, Rueping, and coworkers investigated the role of PNP- and PNN-supported systems for the alkylation of amides and esters. Mn33 showed

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

255

Scheme 44 Alkylation and alkenylation of ketones via HB or ADC using Mn(I) catalysts

Scheme 45 α-Methylation of ketones catalyzed by Mn3

high reactivity for the given transformations, whereas low conversion was detected when a PNP ligand was employed [79]. The group of Balaraman tested a series of aliphatic PNP-derived complexes for the alkylation of esters and amides employing a HB procedure. Interestingly, when

256

S. Weber and K. Kirchner

Scheme 46 Alkylation of esters and amides catalyzed by Mn(I) catalysts

cyclohexyl groups on the phosphorous donors were used, high reactivity could be achieved. If isopropyl groups were used, slightly lower yields were detected, whereas phenyl groups performed moderately. Under optimized conditions, a catalyst loading as low as 0.5 mol% with reduced equivalent of esters and amides and base and lower reaction temperatures could be achieved in comparison to other procedures [80]. The group of Sortais reported on the methylation of benzylic ester using methanol as alkylation agent, employing Mn3 [78]. An overview of esters and amide functionalization is given in Scheme 46.

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

3.8

257

Alkylation of Nitriles and Sulfonamide

The group of Maji was the first ones to utilize a HB protocol for the selective alkylation of nitriles. Within this context 2 mol% of an in situ generated complex and L15 as bidentate ligand were used. Interestingly, although comparingly high reaction temperatures were applied, only 20 mol% of base were required for this transformation [81]. Deuterium-labeling experiments indicated that a manganese hydride complex was generated under the given reaction conditions. The so-formed complex was thought to reduce the C-C bond via MLC. El-Sepelgy, Rueping, and coworkers reported an optimized procedure for the α-alkylation of nitriles. Within this context, the catalyst loading could be decreased to 1 mol% when Mn1# was employed. The comparison of Mn1# with a cationic pyridine-based PNN and a cationic aromatic PNP ligand revealed that the PNN-based system showed slightly lower reactivity, whereas low product formation could be achieved for the lutidine-based complex. In addition to that, C2CO3 was introduced as weak base (Scheme 47) [82]. Sortais and coworkers reported the first methylation procedure for sulfonamides with a small substrate scope using Mn3 as depicted in Scheme 48 [66]. The group of Morrill employed Mn1 for the alkylation of a broad variety of different sulfonamides. It should be noted that only 10 mol% of K2CO3 were employed as weak base. In addition to that, an equimolar amount of sulfonamide and alcohol was used for this procedure [83].

3.9

Upgrading of Ethanol into 1-Butanol

The groups of Liu [84] and Jones [85] independently employed Mn1 in the Guerbet reaction for the conversion of ethanol to 1-butanol (Scheme 49). Within this context, Liu and coworkers could report impressive catalyst loading in ppm ranges with

Scheme 47 Alkylation of nitriles catalyzed by an in situ generated complex and Mn1#

258

S. Weber and K. Kirchner

Scheme 48 Alkylation of sulfonamides catalyzed by Mn14 and Mn9

Scheme 49 Upgrading of ethanol employing the Guerbet reaction mediated by Mn1

selectivity of up to 92%. The role of the NH motive was underlined during this work. Methylation of the nitrogen blocked this position for MLC reactions, which almost shuttled down the catalytic activity of the used system.

4 Conclusion Sustainable (de)hydrogenation reactions employing well-defined manganese complexes became an important and fast-growing field within the last few years. The fact that most of the catalysts are air- and moisture-stable enables easy access to a wide range of researchers beyond specialized experts. Apart from the high stability of many manganese(I)-based systems, the broad applicability in hydrogenation and dehydrogenation in conjunction with subsequent condensation reactions renders these catalysts powerful players in the field of homogeneous catalysis. So far, a large majority of manganese-based catalysts rely strongly on metal-ligand cooperation involving as yet mostly pincer and pincer-type complexes. This approach showed remarkable reactivity for several applications but also revealed limitations, especially for the functionalization of unpolarized functional groups such as C-C

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

259

multiple bonds. Further research in order to introduce different concepts, e.g., innersphere reaction pathways, would broaden the applicability of manganese-based systems. The inspiring discoveries within the last few years led to remarkable findings in manganese-catalyzed (de)hydrogenation reactions. Regarding the large growing number of research articles within this flowering field, manganese catalysis is just at its beginning, and many breakthroughs may be expected in the near future.

References 1. Bullock RM (2013) Science 342:1054–1055 2. Christianson DW (1997) Prog Biophys Mol Biol 67:217 3. Wieghardt K (1989) Angew Chem Int Ed 28:1153 4. Carney JR, Dillon BR, Thomas SP (2016) Eur J Org Chem:3912–3929 5. Garbe M, Junge K, Beller M (2017) Eur J Org Chem:4344–4362 6. Maji B, Barman M (2017) Synthesis 49:3377–3393 7. Filoenko GA, van Putten R, Hensen EJM, Pidko EA (2018) Chem Soc Rev 47:1484–1515 8. Gorgas N, Kirchner K (2018) Acc Chem Res 51:1558–1569 9. Zell T, Langer R (2018) Chem Catal Chem 10:1930–1940 10. Mukherjee A, Milstein D (2018) ACS Catal 8:11435–11469 11. Guanathan C, Milstein D (2013) Science 341:1229712 12. de Vries JG (2006) The handbook of homogeneous hydrogenation. Wiley-VCH Verlag GmbH, Weinheim 13. Khusnutdinova JR, Milstein D (2015) Angew Chem Int Ed 54:12236–12273 14. Elangovan S, Topf C, Fischer S, Jiao H, Spannenberg A, Baumann W, Ludwig R, Junge K, Beller M (2016) J Am Chem Soc 138:8809–8814 15. Kallmeier F, Irrgang T, Dietel T, Kempe R (2016) Angew Chem Int Ed 55:11806–11809 16. Bruneau-Voisine A, Wang D, Roisnel T, Darcel C, Sortais J-B (2017) Catal Commun 92:1–4 17. Glatz M, Stöger B, Himmelbauer D, Veiros LF, Kirchner K (2018) ACS Catal 8:4009–4016 18. Wie D, Bruneau-Voisine A, Chauvin T, Dorcet V, Roisnel T, Valyaev DA, Lugan N, Sortais J-B (2018) Adv Syn Catal 360:676–681 19. Buhaibeh R, Filippov OA, Bruneau-Voisine A, Willot J, Duhayon C, Valyaev DA, Lugan N, Canac Y, Sortais J-B (2019) Angew Chem Int Ed 58:6727–6731 20. Widegren MB, Harkness GJ, Slawin AMZ, Cordes DB, Clarke ML (2017) Angew Chem Int Ed 56:5825–5828 21. Garbe M, Junge K, Walker S, Wei Z, Jiao H, Spennenberg A, Bachmann S, Scalone M, Beller M (2017) Angew Chem Int Ed 56:11237–11241 22. Zhang L, Tang Y, Han Z, Ding K (2019) Angew Chem Int Ed 58:4973–4977 23. Elangovan S, Garbe M, Jiao H, Spannenberg A, Junge K, Beller M (2016) Angew Chem Int Ed 55:15364–15368 24. Widegren MB, Clarke ML (2018) Org Lett 20:2654–2658 25. Espinosa-Jalapa NA, Nerush A, Shimon LJW, Leitus G, Avram L, Ben-David Y, Milstein D (2017) Chem Eur J 23:5934–5938 26. van Putten R, Uslamin EA, Garbe M, Liu C, Gonzalez-de-Castro A, Lutz M, Junge K, Hensen EJM, Beller M, Lefort L, Pidko EA (2017) Angew Chem Int Ed 56:7531–7534 27. Papa V, Cabrero-Antonino JR, Alberico E, Spannenberg A, Junge K, Junge H, Beller M (2017) Chem Sci 8:3576–3585 28. Zou Y-Q, Chakraborty S, Nerusch A, Oren D, Diskin-Posner Y, Ben-David Y, Milstein D (2018) ACS Catal 8:8014–8019

260

S. Weber and K. Kirchner

29. Wei D, Bruneau-Voisine A, Valyaev DA, Lugan N, Sortais J-B (2018) Chem Commun 54:4302–4305 30. Freitag F, Irrgang T, Kempe R (2019) J Am Chem Soc 141:11677–11685 31. Wang Y, Zhu L, Shao Z, Li G, Lan G, Liu Q (2019) J Am Chem Soc 141:17337–17349 32. Weber S, Stöger B, Kirchner K (2018) Org Lett 20:7212–7215 33. Guarduno JA, Garcia JJ (2019) ACS Catal 9:392–401 34. Weber S, Veiros LF, Kirchner K (2019) Adv Syn Catal 361:5412–5420 35. Bertini F, Glatz M, Gorgas N, Stöger B, Peruzzini M, Veiros LF, Kirchner K, Gonsalvi L (2017) Chem Sci 8:5024–5029 36. Dubey A, Nencini L, Fayzullin RR, Nervi C, Khusnutdinova JR (2017) ACS Catal 7:3864–3868 37. Kar S, Goeppert A, Kothandaraman J, Prakash GKS (2017) ACS Catal 7:6347–6351 38. Kumar A, Janes T, Espinosa-Jalapa NA, Milstein D (2018) Angew Chem Int Ed 57:12076–12080 39. Zubar V, Lebedev Y, Azofra LM, Cavallo L, El-Sepergy O, Rueping M (2018) Angew Chem Int Ed 57:13439–13443 40. Kaithal A, Hölscher M, Leitner W (2018) Angew Chem Int Ed 57:13449–13453 41. Das UK, Kumar A, Ben-David Y, Iron MA, Milstein D (2019) J Am Chem Soc 141:12962–12966 42. Weber S, Stöger B, Veiros LF, Kirchner K (2019) ACS Catal 9:9715–9720 43. Mukherjee A, Nerush A, Leitus G, Shimon LJW, Ben-David Y, Espinosa-Jalapa NA, Milstein D (2016) J Am Chem Soc 138:4298–4301 44. Mastalir M, Glatz M, Gorgas N, Stöger B, Pittenauer E, Allmaier G, Veiros LF, Kirchner K (2016) Chem Eur J 22:12316–12320 45. Samuelsen SV, Santilli C, Ahlquist MSG, Madsen R (2019) Chem Sci 10:1150–1157 46. Azizi K, Akrami R, Madsen R (2019) Chem Eur J 25:6439–6446 47. Espinosa-Jalapa NA, Kumar A, Milstein D (2017) J Am Chem Soc 139:11722–11725 48. Chakraborty S, Gellich U, Diskin-Posner Y, Leitus G, Avram L, Milstein D (2017) Angew Chem Int Ed 56:4229–4233 49. Daw P, Kumar A, Espinosa-Jalapa NA, Ben-David Y, Milstein D (2019) J Am Chem Soc 141:12202–12206 50. Nguyen DH, Trivelli X, Capet F, Paul J-F, Dumeignik F, Gauvin RM (2017) ACS Catal 7:2022–2032 51. Wang Z, Zhang KM, Liu Q (2018) Angew Chem Int Ed 57:15143–15147 52. Chakraborty S, Das UK, Ben-David Y, Milstein D (2017) J Am Chem Soc 139:11710–11713 53. Luque-Urrutia JA, Sola M, Milstein D (2019) J Am Chem Soc 141:2398–2403 54. Zhang G, Irrgang T, Dietel T, Kallmeier F, Kempe R (2018) Angew Chem Int Ed 57:9131–9135 55. Barman MK, Waiba S, Maji B (2018) Angew Chem Int Ed 57:9126–9130 56. Mastalir M, Glatz M, Pittenerauer E, Allmaier G, Kirchner K (2016) J Am Chem Soc 138:15543–15546 57. Deibl N, Kempe R (2017) Angew Chem Int Ed 56:1663–1666 58. Kallmeier F, Dudziec B, Irrgang T, Kempe R (2017) Angew Chem Int Ed 56:7261–7265 59. Mastalir M, Pittenauer E, Allmaier G, Kirchner K (2017) J Am Chem Soc 139:8812–8815 60. Daw P, Kumar A, Espinosa-Jalapa NA, Diskin-Posner Y, Ben-David D, Milstein D (2018) ACS Catal 8:7734–7741 61. Das K, Mondal A, Srimani D (2018) J Org Chem 83:9553–9560 62. Das K, Mondal A, Srimani D (2018) Chem Commun 54:10582–10585 63. Elangovan S, Neumann J, Sortais J-B, Junge K, Darcel C, Beller M (2016) Nat Commun 7:12641 64. Huang M, Li Y, Liu J, Shu S, Liu Y, Ke Z (2019) Chem Commun 55:6213–6216 65. Neumann J, Elangovan S, Spannenberg A, Junge K, Beller M (2017) Chem Eur J 23:5410–5413

The Role of Metal-Ligand Cooperation in Manganese(I)-Catalyzed. . .

261

66. Bruneau-Voisine A, Wang D, Dorcet V, Roisnel T, Darcel C, Sortais J-B (2017) J Catal 347:57–62 67. Landge VG, Mondal A, Kumar V, Nandakumar A, Balaraman E (2018) Org Biomol Chem 16:8175–8180 68. Homberg L, Roller A, Hultzsch KC (2019) Org Lett 21:3142–3147 69. Fertig R, Irrgang T, Freitag F, Zander J, Kempe R (2018) ACS Catal 8:8525–8530 70. Liu T, Wang L, Wu K, Yu Z (2018) ACS Catal 8:7201–7207 71. El-Spelgy O, Matabor E, Brzozowska A, Rueping M (2019) Chem Sustain Chem 12:309–3102 72. Kaithal A, Gracia LL, Camp C, Quadrelli EA, Leitner W (2018) J Am Chem Soc 141:17487–17492 73. Kaithal A, van Bonn P, Hölscher M, Leitner W (2020) Angew Chem Int Ed 59:215–220 74. Peña-López M, Piehl P, Elangovan S, Neumann H, Beller M (2016) Angew Chem Int Ed 55:14967–14971 75. Chakraborty S, Daw P, Ben-David Y, Milstein D (2018) ACS Catal 8:10300–10305 76. Barman MK, Jana A, Maji B (2018) Adv Syn Catal 360:3233–3238 77. Gawil SS, Pandia BK, Guanathan C (2019) Org Lett 21:3842–3847 78. Bruneau-Voisine A, Pallova L, Bastin S, Cesar V, Sortais J-B (2019) Chem Commun 55:314–317 79. Jang YK, Krückel T, Rueping M, El-Sepelgy O (2018) Org Lett 20:7779–7783 80. Rana J, Gupta V, Balaraman E (2019) Dalton Trans 48:7049–7099 81. Jana A, Reddy CB, Maji B (2018) ACS Catal 8:9229–9231 82. Borghs JC, Tran MA, Sklyaruk J, Rueping M, El-Sepelgy O (2019) J Org Chem 84:7927–7935 83. Reed-Berendt BG, Morill LC (2019) J Org Chem 84:3715–3724 84. Fu S, Shao Z, Wang Y, Liu Q (2017) J Am Chem Soc 139:11941–11948 85. Kulkarni NV, Brennessel WW, Jones WD (2017) ACS Catal 8:997–1002

Top Organomet Chem (2021) 68: 263–320 https://doi.org/10.1007/3418_2020_63 # Springer Nature Switzerland AG 2020 Published online: 29 December 2020

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a HN(CH2CH2PR2)2 Ligand Dewmi A. Ekanayake and Hairong Guan

Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Ligand Synthesis and Coordination Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Group 8 Metal Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Ruthenium Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Iron Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Osmium Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Group 9 Metal Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Rhodium Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Cobalt Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Iridium Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Group 10 Metal Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Group 6 Metal Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Nitrosyl Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Bis(Carbonyl) Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Group 7 Metal Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Manganese Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Rhenium Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Summary and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

264 267 269 269 285 296 297 297 298 302 305 306 306 308 309 309 314 315 316

Abstract This chapter first provides a brief background of how hydrogenation mechanisms have evolved over the years leading to the blossom of catalytic systems with metal-ligand cooperativity. The main body of the chapter focuses specifically on complexes supported by ligands of the type HN(CH2CH2PR2)2. The discussion of hydrogenation systems is organized based on the central metals including Ru, Fe, Os, Rh, Co, Ir, Ni, Pd, Mo, W, Mn, and Re (in that particular order). Substrates involved in these hydrogenation reactions include olefins, aldehydes, ketones, esters,

D. A. Ekanayake and H. Guan (*) Department of Chemistry, University of Cincinnati, Cincinnati, OH, USA e-mail: [email protected]

264

D. A. Ekanayake and H. Guan

amides, epoxides, nitriles, imines, N-heterocycles, CO2 (to formate or methanol), silyl formates, CO (to ethylene glycol or methanol), and cyclic carbonates. When appropriate, the presence or the lack of metal-ligand cooperativity in these catalytic systems is highlighted. Keywords CO2 reduction · Hydride · Hydrogenation · Metal-ligand cooperativity · Pincer complexes

1 Introduction The development of well-defined transition metal-based catalysts for hydrogenation reactions has been an active research area for almost half a century [1–4]. Early efforts were focused on catalytic hydrogenation of C¼C (or C  C) bonds. The generalized and simplified reaction mechanism involves oxidative addition of H2 and coordination of the C¼C bond to the metal (Scheme 1, Cycle A). These two steps can occur in either order, as exemplified by Wilkinson’s RhCl(PPh3)3 catalyst for hydrogenating olefins (H2 first) [5] and Halpern’s [(CHIRAPHOS)Rh(solvent)2]+ catalyst for hydrogenating α-aminoacrylic acid derivatives (C¼C bond first) [6]. In any case, subsequent C¼C insertion into the metal-hydrogen bond followed by reductive elimination of the hydrogenation product completes the catalytic cycle. Hydrogenation reactions can also be catalyzed by a monohydride such as RuHCl(PPh3)3, whose mechanism (Scheme 1, Cycle B) usually features hydrogenolysis of a metal alkyl intermediate generated from C¼C insertion [7]. Catalytic hydrogenation of C¼O bonds in aldehydes and ketones, especially those without a neighboring heteroatom to assist carbonyl coordination, was

Scheme 1 Generalized mechanisms for catalytic hydrogenation of C¼C bonds

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

265

Scheme 2 Hydride transfer pathways

Fig. 1 Representative hydrogenation (pre)catalysts (acidic and hydridic hydrogens are highlighted)

developed much later. Noyori attributed the difficulty to the preferred coordination mode adopted by the carbonyl group [8]. Unlike olefinic substrates, simple aldehydes and ketones often coordinate to metals via the oxygen lone pair instead of the π system [9], which places the carbonyl carbon far away from the hydride to be delivered (Scheme 2, Pathway A). To overcome this issue, Noyori proposed to design catalysts with an acidic hydrogen strategically situated in the ligand scaffold so that it can protonate or form a hydrogen bond with the carbonyl oxygen, forcing an η2-coordination mode for the C¼O bond (Scheme 2, Pathway B). Alternatively, in an outer-sphere mechanism, the hydrogen-bonded substrate is brought to the close proximity of the hydride ligand for the desired hydride transfer. The concept of metal-ligand cooperativity described above has significantly advanced the field of homogeneous hydrogenation. In particular, the E–HO interaction illustrated in Pathway B (Scheme 2) potentially activates the carbonyl group and deemphasizes the role that the metal needs to play. It is therefore not a coincidence that the past decade has witnessed a rapid development of hydrogenation catalysts targeting more challenging substrates such as esters [10, 11] and amides [12, 13] and/or focusing on first-row transition metals including iron [14, 15] and cobalt [16]. Some of these hydrogenation (pre)catalysts as well as the earlier ones developed by Shvo [17] and Noyori [18, 19] are highlighted in Fig. 1. The vast majority of metal-ligand bifunctional catalysts used for hydrogenation reactions contain at least one NH or NH2 donor, which can be preinstalled prior to

266

D. A. Ekanayake and H. Guan

Scheme 3 Catalyst activation strategies

Scheme 4 Simplified catalytic cycle and transition state

complexation or formed under hydrogenation conditions (e.g., hydrogenation of ligand C¼N bonds) [20]. Although occasionally it is possible to synthesize the H– M–N–H-type complex first [21], most catalytic systems generate this active species in situ from various precatalysts (Scheme 3). Effective catalyst activation strategies include (1) removal of HX (X ¼ Cl, Br, etc.) by a strong base followed by H2 activation [11], (2) hydrogenolysis of a metal alkyl species [16], and (3) unmasking the hydride from the corresponding borohydride complex with heating or in the presence of a BH3 scavenger [22]. It had been hypothesized that hydrogenation of C¼O bonds catalyzed by H–M– N–H-type complexes would proceed via a concerted H+/H transfer to the substrate followed by heterolytic cleavage of H2 by the resulting amido species (Scheme 4) [23]. The lost catalytic activity in replacing NH with an NMe donor group is usually an indication of metal-ligand bifunctional catalysis [24]. However, DFT calculations [25] and kinetic studies [26] suggest that the mechanism is more nuanced than initially thought. For example, the delivery of H+/H to the substrate can be asynchronous, and the alcohol product can serve as a proton shuttle for H2 activation. Furthermore, the metal-bound NH functionality may merely play the role of stabilizing the transition states (through hydrogen bonding interactions) rather than participating in H+ transfer [27]. There are also a number of hydrogenation systems in which alkylation of the NH functionality still results in an active catalyst [28]. Nevertheless, the success of employing H–M–N–H-type complexes as hydrogenation catalysts is evident and likely to provide the momentum to develop new catalysts featuring this particular structural motif.

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

267

This chapter focuses specifically on complexes supported by ligands of the type HN(CH2CH2PR2)2 (RPNHP for short), which are arguably among the most extensively studied hydrogenation catalysts in recent years [29]. Our discussion starts with how these ligands are made and how they are used to prepare the PNP-type complexes. The subsequent overview of hydrogenation catalysis is organized based on the metals, starting from the more popular group 8 elements, transitioning to those in groups 9 and 10, and concluding with mid-transition metals.

2 Ligand Synthesis and Coordination Modes The more frequently used RPNHP ligands (R ¼ iPr, Cy, Ad or 1-adamantyl, tBu) are commercially available in the neat form or as a THF solution, whereas PhPNHP is typically sold as a hydrochloride salt. If needed, they can be synthesized from [H2N (CH2CH2Cl)2]Cl in one or few steps, depending on the properties of the phosphorus substituents (Scheme 5). Synthesis of PhPNHP or other aryl-substituted ligands is readily accomplished by refluxing [H2N(CH2CH2Cl)2]Cl with the corresponding secondary phosphine in the presence of KOtBu [30–32]. Introducing alkyl groups as the phosphorus substituents requires nitrogen protection with a trimethylsilyl group prior to the addition of a lithium dialkylphosphide for the nucleophilic substitution reaction [33–36]. Hydrolysis of the resulting Me3SiN(CH2CH2PR2)2 restores the NH moiety, which is occasionally performed in the presence of nBu4NF [37] or a 2 M solution of H2SO4 [35] to promote the N–Si bond cleavage. For purification purpose, the crude products are sometimes protonated by a dilute aqueous HCl solution to yield the hydrochloride salts as precipitates [30, 32, 34], and the free R PNHP ligands are released following the treatment with NaOH or KOH. Chiral RPNHP ligands are also known in the literature (Scheme 6). Chirality has been introduced through the use of a phosphide derived from (2S,5S)-2,5-dimethyl1-phenylphospholane [38] or an enantiomerically pure secondary phosphine-borane H3B•PH(R)Me (R ¼ tBu, Cy) [39]. In the latter case, lithiation of H3B•PH(R)Me and the subsequent nucleophilic substitution reaction are stereospecific, resulting in stereo-retention at the phosphorus center. In contrast, the in situ generated Li

Cl

H2 Cl N

Cl

Ar2PH, KOtBu THF, reflux

Me3SiCl, Et3N DMSO, reflux SiMe3 Cl

N

R2P

N

Scheme 5 Synthesis of achiral RPNHP ligands

H N

PAr2

Ar = Ph, 4-MeC6H4, 3,5Me2C6H3, 3,5-(CF3)2C6H3

SiMe3

R2PLi THF Cl reflux

Ar2P

PR2

H2O-THF reflux

H N R2 P PR2 (R = Me, Et, iPr, Cy, Ad, tBu)

268

D. A. Ekanayake and H. Guan

PLi

SiMe3 Cl

N

Cl

THF BH3

THF

NaHCO3 (aq.) Me R

nBuLi

TBAF THF H N

P BH3

THF, 0 oC

R P H Me

H N

P

P

R P H3B

H N

Me

HBF4 OEt2 Me CH Cl 2 2

R

P

R P

(R = tBu, Cy)

Me

SiMe3 BH3

Me P Ph

I OH NaH THF 0 oC

N

I

THF

NaHCO3 (aq.)

H N

Ph Me

P BH3

Me P

Ph

H 3B

HBF4 OEt2 CH2Cl2 Ph Me P

H N

P

Me Ph

Scheme 6 Synthesis of chiral RPNHP ligands

[H3BPPhMe] is configurationally unstable. Synthesis of the corresponding chiral R PNHP ligand thus relies on the use of (SP)-(1-hydroxyethyl) methylphenylphosphine-borane as a masked secondary phosphine-borane and Me3SiN(CH2CH2I)2 as a more reactive electrophile to minimize the chance for racemization [39]. The borane-protected RPNHP ligands can be handled in air, and the removal of BH3 by HBF4•OEt2 is often carried out right before complexation. The RPNHP ligands or their deprotonated form [N(CH2CH2PR2)2] (abbreviated here as RPNP) have been employed to make complexes of virtually every metal in groups 4–11 [40]. The coordination chemistry of these ligands is rich, exhibiting a variety of modes including κ 1-N [41], κ2-P,N [42], κ 2-P,P [43], κ3-P,N,P, and μ2-P,P [44]. As far as hydrogenation catalysts are concerned, the κ3-P,N,P coordination mode is most relevant, because it not only provides an entry to the H–M–N–H species but also stabilizes the metal complexes. As tridentate ligands, RPNHP or R PNP can adopt a meridional or facial configuration, depending on the phosphorus substituents, metals, and ancillary ligands. To illustrate this point, Fig. 2 summarizes the solid-state structures of (RPNHP)FeX2 [45–47] and (RPNHP)CoX2 [48–53] known to date. The solution structures of (iPrPNHP)FeCl2 probed by Mössbauer and magnetic circular dichroism spectroscopy also suggest that these PNP-type ligands are flexible in binding with metals [45].

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

269

Fig. 2 Solid-state structures of (RPNHP)MX2 studied by X-ray crystallography

3 Group 8 Metal Systems 3.1 3.1.1

Ruthenium Catalysts Synthesis of (Pre)catalysts

Synthetic routes to hydrogenation (pre)catalysts involving ruthenium-based PNP-type complexes are summarized in Scheme 7. Complex (PhPNHP)RuHCl (CO) was first developed by Takasago International Corporation with a trademark name of Ru-MACHO [32, 54]. It was originally isolated as a mixture of syn and anti (referring to the relative configuration of NH and RuH ) isomers from the reaction of Ph PNHP with RuHCl(CO)(PPh3)3 performed in refluxing toluene, although minor modifications to the procedures could lead to the anti isomer only [55, 56]. The presence of two isomers is deemed to be unimportant for the hydrogenation reactions because catalyst activation by a base (Scheme 3) removes the NH hydrogen. The synthetic approach has been successfully extended to other RPNHP ligands [55–57] including the one bearing chiral phospholane rings [58]. Substitution of the chloride in Ru-MACHO, iPrRuHCl, and CyRuHCl for BH4 and H has been accomplished through the addition of NaBH4 [32, 56, 59] and NaBEt3H [56, 60], respectively. The latter reaction with the sterically crowded tBuRuHCl, however, produces a five-coordinate ruthenium hydride, likely due to a facile H2 elimination from the initial product tBuRuH2 [56]. The phenyl analog PhRuH2 synthesized from the NaBEt3H method has a low purity because of rapid decomposition [56]. It can alternatively be synthesized from Ru-MACHO and KOtBu (or KN(SiMe3)2) under H2, though contaminated with ~5% of Ru-MACHO [21]. Other ruthenium precursors have been used to prepare PNP-type hydrogenation (pre)catalysts. The reaction of Ru(COD)(2-methylallyl)2 with tBuPNHP under 5 bar of H2 produces a mixture of tBuRuH2(H2) and tBuRuH(H2) (Scheme 7, Method B) [61]. Pure tBuRuH(H2) can be obtained by stirring the mixture under argon, and its reaction with a primary alcohol also affords tBuRuH as a result of alcohol dehydrogenation and decarbonylation [62]. Using ( p-cymene)RuCl2(NHC) (NHC ¼ 1,3dimethylimidazol-2-ylidene) as the ruthenium source provides an opportunity to

270

D. A. Ekanayake and H. Guan

Scheme 7 Synthetic routes to ruthenium-based hydrogenation (pre)catalysts

incorporate an N-heterocyclic carbene into the catalyst structure. As illustrated in Scheme 7 (Method C), its reaction with a RPNHP ligand can lead to a neutral or cationic pincer complex depending on the solvent used [63].

3.1.2

Hydrogenation of Esters, Ketones, and Their Derivatives

Both Ru-MACHO and Ru-MACHO-BH are commercially available, and among the complexes shown in Scheme 7, they are the most extensively studied ones for catalytic hydrogenation reactions. In 2012, Takasago International Corporation reported that Ru-MACHO mixed with NaOMe was effective for hydrogenation of esters to alcohols (Eq. 1) [54]. The catalytic system is amenable to benzyloxy,

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

271

piperidinyl, or l-menthoxy group at the α-position but problematic with methoxy or dimethylamino group at the β-position (e.g., MeOCH2CH2CO2Me and Me2NCH2CH2CO2Me). Most remarkably, hydrogenation of methyl (R)-lactate can be performed at room temperature on a multiton scale with minimal erosion to the optical purity (Eq. 2). ð1Þ

ð2Þ

In a subsequent report [21], Ikariya demonstrated that Ru-MACHO was efficient in catalyzing hydrogenation of α-difluorinated esters with turnover numbers (TONs) as high as 20,000 (Eq. 3). Functional groups tolerated in this transformation include C¼C bonds (terminal or internal), α-pyridyl, and α-thienyl rings. In addition to Ru-MACHO, PhRuH2 and trans-(PhPNHP)RuCl2(CO) are also capable of catalyzing the hydrogenation reactions, although the dichloride complex displays a lower reactivity. For certain substrates (R0 ¼ H, F, Cl, CF3), the hydrogenation process can be stopped at the hemiacetal stage, and in general the selectivity for R0 CF2CH (OH)OR00 is improved by lowering the H2 pressure, temperature, and/or the amount of NaOMe. α-Monofluorinated esters can also be hydrogenated under the catalytic conditions; however, the fluorinated primary alcohol products partially undergo cyclization to form epoxides. In a closely related study [64], Lazzari and Cassani showed similar results with RfCO2Me (Rf ¼ C3F7 or C5F11), which led to the isolation of highly fluorinated primary alcohols (Eq. 4).

ð3Þ

ð4Þ

Another application of the ruthenium-catalyzed ester hydrogenation reactions is in the synthesis of the fragrance hydroxyambran (or 2-cyclododecylpropan-1-ol) [65]. As shown in Scheme 8, hydrogenation of the isomeric mixture of esters with 10% Pd/C provides ethyl 2-cyclododecylpropanoate by saturating all C¼C bonds.

272

D. A. Ekanayake and H. Guan

Scheme 8 Synthesis of hydroxyambran via hydrogenation reactions

The second step for ester hydrogenation can be catalyzed at 180 C in toluene by Ru-MACHO (0.2 mol%) in conjunction with NaOMe (2 mol%) or at 150 C in diglyme by Ru-MACHO-BH (1 mol%) alone, both under 50 bar H2. The homogeneous ester hydrogenation can be performed first, although hydrogenation of the resulting mixture of alcohols with 10% Pd/C is plagued by deoxygenation. In collaboration with Procter & Gamble Company, we studied the hydrogenation of fatty acid methyl esters (FAMEs) under neat conditions [66]. Starting from FAMEs derived from coconut oil, fatty alcohols can be obtained in high yields when the hydrogenation reaction is catalyzed at 135 C by Ru-MACHO (0.07–1.1 mol%, nNaOMe/nRu ~9, 35.5–52.7 bar H2) or Ru-MACHO-BH (0.13–1.0 mol%, 35.5–69.9 bar H2). The catalytic reaction with Ru-MACHO and NaOMe has also been conducted on the kilogram scale with a TON of 1860. Direct hydrogenation of coconut oil to fatty alcohols is feasible under base-free conditions, which involve Ru-MACHO-BH (2.6–2.8 wt%) operating at 135 C under 52.7 bar H2. Hydrogenation of FAMEs containing C¼C bonds is more challenging, likely due to the presence of peroxide impurities. Dumeignil and Gauvin recently developed a purification procedure involving 18 h of treatment of FAMEs with basic alumina followed by drying with 3 Å molecular sieves for 48 h [67]. The prepurified FAMEs can undergo smooth hydrogenation to fatty alcohols catalyzed by Ru-MACHO-BH or iPrRuHBH4 (Eq. 5). Depressurizing the system and then reheating the reaction mixture to 130 C under N2 offers a one-pot, two-step synthesis of wax esters. It should be noted that, compared to Ru-MACHO-BH, the isopropyl derivative iPrRuHBH4 shows slightly higher catalytic activity in both hydrogenation and dehydrogenation steps and substantially higher overall selectivity for wax esters.

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

273

Scheme 9 Complete and partial hydrogenation of diethyl oxalate

ð5Þ In 2016, Ogata and Kayaki developed a series of NHC-ligated ruthenium PNP-type catalysts for ester hydrogenation that operate under milder conditions [63]. In particular, complexes PhRuCl2(NHC) and ArRuCl2(NHC) outperform Ru-MACHO in hydrogenating methyl benzoate at 80 C under 10 bar H2. Further optimization of the catalytic conditions showed that in the presence of KOtBu or NaOMe, PhRuCl2(NHC) was active at 50 C even under a balloon pressure of H2, converting various esters to the corresponding alcohols (Eq. 6). ð6Þ Dialkyl oxalates belong to a special class of esters for which hydrogenation of the first carbonyl group significantly impacts the reactivity of the remaining carbonyl group. A 2013 report by Beller demonstrated that hydrogenation of diethyl oxalate with Ru-MACHO or Ru-MACHO-BH in the presence of NaOEt yielded ethylene glycol exclusively (Scheme 9) [68]. Interestingly, replacing the catalyst with iPr RuHCl or iPrRuH2 under otherwise the same conditions afforded ethyl glycolate only. Further investigation of Ru-MACHO-BH under base-free conditions suggested that the hydrogenation process could stop at the glycolate stage under a lower temperature (60 C) and after a shorter reaction time (1 h). These results imply that the second hydrogenation step is more difficult. As another example of Ru-MACHO-BH differentiating the reactivity of two ester functionalities, MeOCOCH2CO2tBu was subjected to similar hydrogenation conditions (0.54 mol % [Ru], 5.4 mol% NaOEt, 60 bar H2, 100 C, in THF, 3 h), resulting in a partial hydrogenation product with the sterically more hindered carbonyl group left

274

D. A. Ekanayake and H. Guan

Scheme 10 Hydrogenation of α-keto esters

unreacted (i.e., HOCH2CH2CO2tBu as the product) [68]. A closely related substrate is the oxamate illustrated in Eq. 7. The hydrogenation reaction was carried out under more demanding conditions, which unsurprisingly led to complete hydrogenation to ethylene glycol [69].

ð7Þ

Given the higher electrophilicity of the carbonyl carbons, ketones should be more readily hydrogenated than esters. Thus, for molecules containing both ketone and ester functionalities, it is possible to fine-tune the reaction conditions so that one or both carbonyl groups are hydrogenated. This was demonstrated by Tang and Xiao in their study of Ru-MACHO-catalyzed hydrogenation of α-keto esters [70]. Using NaHCO3 as the base additive paired with relatively low H2 pressure (10 bar) and temperature (25 C) leads to α-hydroxy esters almost exclusively (Scheme 10). In contrast, using a stronger base NaOtBu and raising the H2 pressure to 50 bar and temperature to 80 C result in 1,2-diols with high selectivity (86–100%). Selective hydrogenation of γ-keto esters, in principle, could generate γ-hydroxy esters in an analogous way, although the base additive required for catalyst activation also promotes intramolecular transesterification. Very recently, Paixão and Nielsen reported such conversion with TONs of up to 7,400 by employing Ru-MACHO as the precatalyst and NaOEt as the base (Eq. 8) [71]. Under similar conditions, the related ruthenium complexes including Ru-MACHO-BH, PhRuH2, and the commercially available iPrRuHCl also catalyze the hydrogenation of ethyl levulinate to γ-valerolactone, albeit less effectively.

ð8Þ

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

275

Scheme 11 Hydrogenation of α-chiral hemiacetals and aldehydes

Incorporating a chiral RPNHP ligand into the catalyst structure should allow ketones to be hydrogenated enantioselectively. In a recent report, Junge and Beller tested the catalytic activities of R*RuHCl (syn/anti mixture or pure anti isomer) in hydrogenation of acetophenone and cyclohexyl methyl ketone (Eq. 9) [58]. While the conversion is quantitative, the enantioselectivity is low, suggesting room for improvement in future ligand screening.

ð9Þ

Since hemiacetals and aldehydes are intermediates during ester hydrogenation, they can be readily reduced to alcohols under the hydrogenation conditions optimized for esters. Obviously, many other transition metal complexes can also catalyze this process. Employing Ru-MACHO-BH as the hydrogenation precatalyst has some advantage due to the fact that a base additive is not needed, which is ideal for base-sensitive substrates. In exploring precursors to the new antibiotic nemonoxacin, Clarke used this specific ruthenium complex to catalyze the hydrogenation of a hemiacetal and an aldehyde made from asymmetric hydroformylation reactions [72]. Under the conditions outlined in Scheme 11, the alcohol products are obtained with retention of stereochemistry. It is interesting to note that the ester functionality is intact during the hydrogenation process. In addition to esters, ketones, hemiacetals, and aldehydes, amides have also been explored as substrates for the ruthenium-catalyzed hydrogenation reactions, although the conditions are much harsher. In 2018, Tu reported the hydrogenation of lactams to amino alcohols catalyzed by Ru-MACHO-BH (Eq. 10) [73]. The high

276

D. A. Ekanayake and H. Guan

temperature of 150 C is critical to the success of the hydrogenation process. According to the catalyst activation mechanism (Scheme 3), a base additive is normally not needed for Ru-MACHO-BH to be catalytically active. In fact, Ru-MACHO-BH does show some catalytic activity for hydrogenating N-phenyl2-pyrrolidone. However, the addition of K3PO4 significantly enhances the catalytic efficiency (96% vs. 64% yield). The NHC-ligated complex PhRuCl2(NHC) displays slightly lower activity (82% yield), whereas the methylated PNP pincer complexes (PhPNMeP)RuHCl(CO) and (PhPNMeP)RuH(BH4)(CO) (PhPNMeP ¼ MeN (CH2CH2PPh2)2) are completely inactive, suggesting the importance of the NH moiety. Under similar conditions, hydrogenation of unprotected lactams (e.g., caprolactam and azocan-2-one) and oxazolidinones (e.g., 3-phenyloxazolidin-2-one) is also possible, providing the corresponding amino alcohols in high yields.

ð10Þ

3.1.3

Hydrogenation of Other Bonds

Substrates that can be hydrogenated with the aforementioned ruthenium catalysts go beyond those containing carbonyl groups. Very recently, Gunanathan showed that Ru-MACHO along with KOtBu was effective and selective for the hydrogenation of epoxides to secondary alcohols (Eq. 11) [74]. This transformation proceeds via direct hydrogen transfer from the presumed active species PhRuH2 rather than by a two-step process involving epoxide-to-ketone isomerization followed by ketone hydrogenation. Functional groups compatible with the catalytic conditions are very similar to those observed in ester hydrogenation, except that herein terminal C¼C bonds are also hydrogenated. Hydrogenation of chiral epoxide R-glycidol, however, gives a complex mixture, perhaps due to the interference by KOtBu. Another limitation of the catalytic system is that internal epoxides resist hydrogenation.

ð11Þ

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

277

The strategy of using the ruthenium-based PNP pincer complexes for hydrogenation reactions has been extended to nitrile reduction. Two reports on this topic appeared in 2015, but featured different RPNHP ligands. Both Ru-MACHO and Ru-MACHO-BH were shown by Beller to catalyze the hydrogenation of nitriles to primary amines, although the former required KOtBu to activate the catalyst [75]. One of the challenges for nitrile hydrogenation is selectivity, as the intermediates can be trapped by the initially produced primary amines, which lead to secondary amines, secondary imines, and/or tertiary amines as by-products. Under the conditions summarized in Eq. 12, a variety of aliphatic and aromatic nitriles are converted to primary amines with high selectivity. Lowering the temperature or catalyst loading or hydrogenating long-chain nitriles such as dodecanenitrile, however, erodes selectivity for the primary amines. The catalytic system exhibits high functional group tolerance including the preservation of ester functionalities. Substrates that fail to react include furan-2-carbonitrile, 2-methyl-3-butenenitrile, and 6-bromohexanenitrile. Prechtl focused on the study of ruthenium complexes supported by the more bulky ligand tBuPNHP. Hydrogenation of benzonitrile and p-tolunitrile catalyzed by tBuRuH2(H2)/tBuRuH(H2) or tBuRuH was optimized to favor the secondary imines (Eq. 13), although hydrogenation of p-bromobenzonitrile suffered from moderate yield and low selectivity, and hydrogenation of heptyl cyanide catalyzed by tBuRuH2(H2)/tBuRuH(H2) afforded predominantly octylamine [76]. Under similar catalytic conditions, externally added amines can trap the primary imine intermediates, leading to efficient hydrogenative coupling of nitriles to secondary imines (Eq. 14). Finally, switching the solvent from toluene to iPrOH and raising the temperature from 50 C to 90 C render tBuRuH more selective for the formation of primary amines (Eq. 15). However, varying amounts of R0 CH2N¼CMe2 (0–29%) were also observed due to dehydrogenation of the solvent i PrOH to acetone.

ð12Þ

ð13Þ

278

D. A. Ekanayake and H. Guan

ð14Þ

ð15Þ For the reactions shown in Eqs. 13 and 14, a small amount of secondary amines was often detected, suggesting that it is possible to develop ruthenium-catalyzed hydrogenation of imines. In 2017, Tang reported such process with an objective to develop a diastereoselective route to convert chiral α-ketimino esters to chiral aryl glycine derivatives [77]. As illustrated in Eq. 16, at 25–40 C, Ru-MACHO in combination with NaOEt is effective for the C¼N bond hydrogenation, which provides N-tert-butylsulfinyl-protected α-amino esters with high diastereoselectivity.

ð16Þ

3.1.4

Hydrogenation Reactions Related to CO2 or CO Reduction

Homogeneous hydrogenation of CO2 or CO to liquid fuels such as methanol has been subject to extensive studies in recent years, which, to some degree, is propelled by the development of PNP-type hydrogenation catalysts. Conversion of CO2 to methanol is formally a six-electron reduction process, and each hydrogenation event reduces formal oxidation state of the carbon by two. Based on this analysis, reduction of CO to methanol is formally a four-electron reduction process. Conversion of CO2 or CO to oxalate or ethylene glycol requires odd number of electrons to be transferred (Fig. 3), which usually involves a radical intermediate or a process separate from hydrogenation. For a more systematic discussion, hydrogenation reactions described in this section are organized based on how formal oxidation state of the carbon changes: (A) +4 to +2, (B) +2 to 2, (C) +2 to +3 to 1, and (D) +4 to 2. Hydrogenation of CO2 to formic acid in organic solvents is an endergonic process (ΔG0298 ¼ +32.8 kJ mol1). The thermodynamics can be improved by performing

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

279

Fig. 3 Compounds relevant to CO2 or CO reduction

the reaction in water (ΔG0298 ¼ 4.0 kJ mol1) and/or adding a base to convert formic acid to a formate salt [78]. Direct hydrogenation of bicarbonate to formate is also thermodynamically favorable. For PNP-type catalytic systems, Beller reported in 2014 that transfer hydrogenation of HCO3 (or CO2) to HCO2 with MeOH was efficiently catalyzed by Ru-MACHO or iPrRuHCl in an alkaline solution [79]. It was noted that hydrogen pressure was built up during the reaction, consistent with catalytic methanol dehydrogenation. To confirm that the in situ generated H2 was responsible for bicarbonate reduction, Ru-MACHO was tested as a hydrogenation catalyst for NaHCO3, which, under the conditions outlined in Eq. 17, afforded HCO2Na in 71% yield. A more recent report by Treigerman showed that this hydrogenation process could be conducted at 70 C in iPrOH-H2O mix solvent and the catalyst could be reused at least three times with an overnight rest of the catalyst between two consecutive runs [80]. Czaun, Prakash, and Olah carried out a more detailed study of Ru-MACHO- and Ru-MACHO-BH-catalyzed hydrogenation of bicarbonate (Eq. 18) as well as hydrogenation of CO2 assisted by a hydroxide (Eq. 19) or a carbonate (Eq. 20) [81]. The reverse reaction, dehydrogenation of formate, was also catalyzed by Ru-MACHO or Ru-MACHO-BH. To demonstrate the reversible hydrogen storage in formate salts, Ru-MACHO-BH was employed to catalyze CO2 hydrogenation (75 bar, pH2: pCO2 ¼ 3: 1) in the presence of NaOH followed by dehydrogenation under an atmospheric pressure, a process that was repeated at 70 C for six times without a significant loss of the catalytic activity. Interestingly, the NH moiety is not needed here; (PhPNMeP)RuHCl(CO) catalyzes the hydrogenation and the dehydrogenation reactions with a comparable or better efficiency than Ru-MACHO and Ru-MACHO-BH. ð17Þ

280

D. A. Ekanayake and H. Guan

Scheme 12 One-pot, two-step hydrogenation of CO2 to methanol

ð18Þ

ð19Þ

ð20Þ Hydrogenation of CO2 along with a primary or secondary amine to generate a formamide is also a formally two-electron reduction process (+4 to +2 for the change in formal oxidation state of the carbon). In 2015, Ding reported that in the presence of KOtBu (0.1 mol%), Ru-MACHO, iPrRuHCl, CyRuHCl, AdRuHCl, tBuRuHCl, and the methylated complex (PhPNMeP)RuHCl(CO) were all effective in catalyzing N-formylation of morpholine under H2 and CO2 (35.5 bar each, 0.1 mol% [Ru], 120 C, in THF) [82]. Using Me2NH as the amine (also as a base) and lowering the catalyst loading of Ru-MACHO to 0.000093 mol% produced DMF with TONs of up to 599,000. The catalyst showed remarkably high stability under the catalytic conditions. With a catalyst loading of 0.002 mol%, Ru-MACHO was reused 11 times without the concern for a brief exposure to air between runs. Further hydrogenation of formamides to methanol (formal oxidation state change from +2 to 2) is possible but needs to be performed under a higher temperature and in the presence of KOtBu. As illustrated in Scheme 12, N-formylation of morpholine followed by hydrogenation of the resulting formamide in the same reactor produces methanol in 36% yield along with the unreacted formamide. Another example of changing formal oxidation state of the carbon from +2 to 2 involves catalytic hydrogenation of silyl formates to methanol. A recent report by Hong showed that silyl formates were first prepared from silanes and CO2 catalyzed by Rh2(OAc)4-K2CO3 or RuCl3•H2O [83]. The subsequent hydrogenation reactions can be catalyzed by Ru-MACHO combined with KOtBu but more efficiently by Ru-MACHO-BH, which does not require a base. Under the optimized conditions (Eq. 21), various silyl formates (R0 3Si ¼ trialkyl, aryldialkyl, and alkyldiaryl groups) are converted to methanol and the corresponding silanols. Hydrogenation of silyl formates bearing an electron-donating aryl group (e.g., R0 3Si ¼ Me2( p-MeOC6H4)Si

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

281

or Me2( p-MeC6H4)Si) under 10 bar H2 is complicated by the formation of R0 3SiOMe and R0 3SiOSiR0 3 as the by-products. The selectivity for methanol and silanols can, however, be improved by raising the H2 pressure to 80 bar or by adding 0.1 equiv. of methanol. For the latter strategy, methanol attacks silyl formates to yield silanols and methyl formate, which is in turn readily hydrogenated to 2 equiv. of methanol under the catalytic conditions. O H

Ru-MACHO-BH (0.1-5 mol%)

+ H2 (10-80 bar) CH3OH + R'3SiOH THF, 150 oC, 12 h OSiR'3

ð21Þ

When K3PO4 is used as the catalyst, secondary amines can react with CO (30 bar) at 140 C to give formamides. This reaction coupled with formamide hydrogenation provides an indirect route of hydrogenation of CO to methanol. The challenge lies in the fact that the carbonylation step is favored by an alcoholic solvent, whereas the hydrogenation step is favored by a relatively nonpolar solvent such as toluene. To solve this problem, Prakash designed a one-pot, two-step process in which carbonylation of piperidine or diethylenetriamine (DETA) was carried out in ethanol first [84]. A ruthenium catalyst (Ru-MACHO or Ru-MACHO-BH), toluene, and H2 were then added to the reactor, and following hydrogenation, methanol was produced in 75–80% yield. Direct hydrogenation of CO to methanol was made possible by using DETA as the amine and toluene-EtOH (1:1) as the mix solvent (Eq. 22). The reaction was performed in a closed system, providing methanol in 59% yield (or a TON of 539 based on the amount of Ru-MACHO-BH used) along with formamides in 15% yield.

ð22Þ

Similarly, palladium-catalyzed oxidative carbonylation of piperidine provides an oxamide that can be hydrogenated to ethylene glycol, representing an indirect method of hydrogenating CO to ethylene glycol. The overall process involves changes of formal oxidation state of the carbon from +2 to +3 and then to 1. To this end, Li and Beller reported in 2016 that oxidative carbonylation of piperidine was best catalyzed by Pd(acac)2-P(o-tol)3 using compressed air as the source of oxidant [85]. Hydrogenation of the resulting oxamide is affected by Ru-MACHO or Ru-MACHO-BH (0.1–1 mol% loading, in toluene) at 160 C under 60 bar H2 using KOtBu (2–10 mol%) as the additive. Combining these two steps in one reactor is difficult, and the exchange of solvents and a filtration through silica gel are required after the formation of the oxamide (Scheme 13).

282

D. A. Ekanayake and H. Guan

Scheme 13 Piperidine-mediated conversion of CO to ethylene glycol

For the hydrogenation of CO2 to methanol (with a change of the carbon formal oxidation state from +4 to 2), one strategy is to use cyclic carbonates as surrogates for CO2, which can bypass formic acid (incompatible with metal hydrides) or formate salts (thermodynamic sinks). This was successfully demonstrated in 2012 by Ding who studied ruthenium-catalyzed hydrogenation of ethylene carbonate [55]. Among the precatalysts screened, Ru-MACHO performs significantly better than the analogous complexes bearing alkyl groups as the phosphorus substituents (i.e., iPrRuHCl, CyRuHCl, AdRuHCl, and tBuRuHCl) with TONs as high as 87,000. In this case, the NH moiety is crucial for the catalysis because the methylated complex (PhPNMeP)RuHCl(CO) fails to hydrogenate ethylene carbonate. Under the optimized conditions (Eq. 23), various cyclic carbonates are converted to diols and methanol in almost quantitative yields. The hydrogenation strategy was further applied to poly(propylene carbonate) with an Mw of 1,000,698, giving 1,2-propylene glycol and methanol in high yield (Eq. 24). Hydrogenation of (R)-propylene carbonate under similar conditions generates racemic 1,2-propylene glycol, presumably due to the reversibility of the hydrogenation process.

ð23Þ

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

283

Scheme 14 Ruthenium-catalyzed hydrogenation of CO2 in the presence of Me2NH

ð24Þ

The seminal work by Sanford in 2015 demonstrated that direct catalytic hydrogenation of CO2 to methanol could be accomplished via tandem catalysis of Me2NH promoted by Ru-MACHO-BH (Scheme 14) [86]. The proposed mechanism involves equilibrium between CO2 and dimethylammonium dimethylcarbamate (DMC), which can be hydrogenated to formic acid (trapped as dimethylammonium formate or DMFA) and DMF, respectively. The most challenging step is the hydrogenation of DMF to methanol, a process requiring temperatures as high as 155 C. Under such conditions, the ruthenium catalyst also starts to decompose. To maximize the yield for methanol, a temperature ramp strategy was developed so that a sufficient amount of DMF and DMFA could be accumulated at 95 C. The subsequent hydrogenation carried out at 155 C provides methanol with TONs of up to 550 and DMF-DMFA with combined TONs of up to 1870. In addition to Me2NH, polyamines can also be employed to assist CO2 hydrogenation. Olah and Prakash reported in 2016 that pentaethylenehexamine (PEHA) combined with a catalytic amount of Ru-MACHO or Ru-MACHO-BH promoted the hydrogenation of CO2 to methanol in an etherate solvent (e.g., THF, 1,4-dioxane, diglyme, or triglyme) [87]. After extensive optimization of the reaction, it was determined that with this new catalytic system, the temperature ramp strategy and the addition of K3PO4 were unnecessary. At 155 C under 75 bar H2/CO2 (3: 1 or 9: 1), methanol was obtained with TONs of up to 1,200 and the catalyst was reused five times with 75% of the initial activity retained. CO2 can also be captured from simulated air (400 ppm of CO2 in 80% N2 and 20% O2) by an aqueous solution of

284

D. A. Ekanayake and H. Guan CO2 R'3N + H2 O

captured CO 2 Ru-MACHO-BH (catalyst) 2-MeTHF (solvent)

catalyst 2-MeTHF HCO2 H O 2 HNR'3

N N (DABCO)

R'3 N =

H 2 (50 bar) 55 o C, 5 h

catalyst 2-MeTHF HCO3 HNR'3 H2O R'2NCO2

H 2 (80 bar) 145 oC, 72 h

catalyst 2-MeTHF MeOH HCO2 MeOH HNR'3 H2O HCONR'2

R'3 N = H 2N

H N (PEHA)

4 NH 2

Scheme 15 CO2 capture and the subsequent hydrogenation reaction in a biphasic mixture

PEHA and then subjected to hydrogenation conditions (155 C, 50 bar H2, Ru-MACHO-BH as the catalyst, 55 h), which provides methanol in 79% yield. Additional improvements to the catalytic system include hydrogenation of the captured CO2 (from pure CO2 or simulated air) using various polyamines in a biphasic mixture of water and 2-methyltetrahydrofuran (2-MeTHF). This allows an easy separation of the catalyst (in 2-MeTHF layer) from the hydrogenation products (in water layer). Depending on the temperature applied, the hydrogenation product can be a formate salt [88] or predominantly methanol [89] (Scheme 15). Both processes have shown excellent recyclability of the catalyst (4–5 runs). The polyamines play important roles in determining the yield and selectivity of the hydrogenation process. For hydrogenation of the captured CO2 to formate (in 1,4-dioxane, 50 C, 50 or 80 bar H2), 1,4-diazabicyclo[2.2.2]octane (DABCO), 1,1,3,3-tetramethylguanidine (TMG), and 1,8-diazabicycloundec-7-ene (DBU) outperform PEHA and branched polyethyleneimines (BPEI, Mw ¼ 800) in terms of the formate yield [88]. For hydrogenation of the captured CO2 to methanol (in 2-MeTHF, 145 C, 70 bar H2), PEHA gives a higher methanol yield than BPEI (Mw ¼ 800 or 25,000), linear polyethyleneimines (LPEI, Mw ¼ 2,500 or 100,000), and poly(allylamine) (PAA, Mw ¼ 10,000). The latter three polyamines also produce more formate and formamide as the by-products [89]. In a related study, Kayaki also used BPEI (Mn ¼ 600) and LPEI (Mn ¼ 2,500, 5,000, 25,000, or 250,000) to assist CO2 hydrogenation, although the reactions were carried out in THF only [90]. At 100 C under 100 bar H2 and 100 bar CO2, ruthenium complexes including Ru-MACHO, Ru-MACHO-BH, CyRuHCl, and PhRuCl2(NHC) were shown to be similarly effective in converting CO2 and the polymers to N-formylated PEI with 67–90% CHO content. Further hydrogenation of the N-formylated PEI to methanol or direct hydrogenation of CO2 to methanol assisted by BPEI or LPEI is best catalyzed by Ru-MACHO-BH at 140–160 C under 80 bar H2/CO2 (3:1 or 7:1).

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

285

O H

H CO

R'2NH2 Ph2 H P CO Ru PPh2 N

OH

H2

H O

H2

R'2NH

Ph2 H CO H P Ru PPh2 N

CO resting state H2

CO Ph2 H H Ph2 CO R'2NH2 R'2NH H P CO H P Ru Ru PPh 2 N PPh2 N H H H PhRuH 2

Scheme 16 Involvement of the cationic bis(carbonyl) hydride species during the catalytic hydrogenation reaction

The nature of the phosphorus substituents also plays critical roles in determining the catalytic efficiency. Although Ru-MACHO, iPrRuHCl, and CyRuHCl (in the presence of K3PO4) all prove to be active precatalysts for the hydrogenation of formamides to methanol [91, 92], for polyamine-assisted CO2 hydrogenation, Ru-MACHO (or Ru-MACHO-BH) appears to be the best choice for maximizing methanol yield [89, 92]. A recent mechanistic study by Prakash offered very insightful information about why the phenyl groups are beneficial for the hydrogenation reaction [92]. Evidently, during CO2 to methanol conversion, a small amount of CO (~0.2%) is generated, which poisons those ruthenium catalysts bearing alkyl substituents. In fact, during PEHA-assisted CO2 hydrogenation, the resting state of the catalyst was identified as a cationic bis(carbonyl) hydride complex (Scheme 16). For the phenyl derivative, the CO is more labile due to weaker donation from the phosphorus atoms, allowing [(PhPNHPh)Ru(CO)2H]+ to reenter the catalytic cycle by forming the active species PhRuH2. Such process is less favorable for the alkyl derivatives.

3.2 3.2.1

Iron Catalysts Synthesis of (Pre)catalysts

The recent surge in developing base metal catalysis has prompted many research groups to design iron-based hydrogenation catalysts. A logical extension of the work shown in the previous section would be replacing ruthenium with iron, although the

286

D. A. Ekanayake and H. Guan

Scheme 17 Synthetic routes to iron-based hydrogenation (pre)catalysts

chemistry of the ruthenium PNP-type complexes cannot be simply extrapolated to the iron systems. This is already reflected by how the iron-based (pre)catalysts are made (Scheme 17). First of all, iron analogs of the ruthenium precursors RuHCl(CO) (PPh3)3 and Ru(COD)(2-methylallyl)2 do not exist. While the catalysis community enjoys the use of Ru-MACHO and Ru-MACHO-BH, both of which are

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

287

commercially available, Fe-MACHO remains elusive, and Fe-MACHO-BH has a very limited lifetime in solution [93]. Nevertheless, in 2013, Beller first reported the synthesis of iPrFeHBH4, which involved the treatment of trans-(iPrPNHP) FeBr2(CO) (made from iPrPNHP and FeBr2(THF)2 under 1 bar CO) with excess NaBH4 in EtOH (Scheme 17, Method A) [94]. Reducing the amount of NaBH4 to 1 equiv. led to the isolation of iPrFeHBr [95], which was alternatively prepared in THF from trans-(iPrPNHP)FeBr2(CO) using NaBEt3H as the hydride source [94]. Depending on the reaction time and work-up procedures, both iPrFeHBH4 and iPrFeHBr can be isolated as a mixture of syn and anti isomers or as a pure anti isomer, although it is expected to have no impact on the catalytic performance. This synthetic strategy has been extended to other ligand systems including EtPNHP [93, 96], CyPNHP [97, 98], the phospholane-based PNP ligand [58], and the Pstereogenic PNP ligands [39]. It is worth pointing out that (S,S)(tBuMePCH2CH2)2NH adopts the facial coordination mode upon formation of the dibromide complex, whose reaction with NaBEt3H must be carried out in CH2Cl2 instead of THF to avoid degradation. The three P-chiral precatalysts, MePhFeHBr, MeCy FeHBr, and MetBuFeHBr, decompose quickly in solution; therefore, they should be prepared right before use [39]. The more commonly used iron precatalyst iPr FeHBH4 can also be synthesized from the dichloride complex trans-(iPrPNHP) FeCl2(CO) and NaBH4 (10 equiv) in MeCN-EtOH, although applying this protocol to trans-(CyPNHP)FeCl2(CO) fails to generate CyFeHBH4 cleanly [99]. The five-coordinate complex iPrFeH can be obtained from dehydrohalogenation of iPrFeHBr [100] or iPrFeHCl (made from trans-(iPrPNHP)FeCl2(CO) and n Bu4NBH4) [101] with KOtBu (Scheme 17, Method B). The cyclohexyl analog Cy FeH is also available using this method [101]. Preparing the isocyanide derivatives iPrFeH(CNArMe2) and iPrFeH(CNArOMe) follows similar procedures (Method C) [102]. The key challenge here is in the synthesis of trans-(iPrPNHP) FeCl2(CNR). To avoid the undesired cationic bis(isocyanide) complexes, isocyanides must be diluted and added slowly to (iPrPNHP)FeCl2.

3.2.2

Hydrogenation of Esters, Ketones, and Their Derivatives

In 2014, the Beller group [24] and our group [95] independently reported that FeHBH4 was effective in catalyzing the hydrogenation of esters including lactones to alcohols (Eq. 25). Functional groups tolerated under the catalytic conditions include CF3, MeO, pyridyl, furyl, benzothiazolyl, and isolated C¼C bonds. In contrast, nitrile groups and conjugate C¼C bonds are hydrogenated along with the carbonyl groups, and phenol-type functionality shuts down the catalysis completely. For further applications (Fig. 4), iPrFeHBH4 has been utilized to catalyze the hydrogenation of a dodecapeptide, which is a precursor to the drug molecule Alisporivir [24], and an industrial sample CE-1270, which is derived from coconut oil and used in surfactant production [95]. As with the ruthenium system, iPr FeHBH4 has also been tested for direct catalytic hydrogenation of coconut oil (2.0 wt% catalyst loading, 135 C, 52.7 bar H2), although the fatty alcohol yield is iPr

288

D. A. Ekanayake and H. Guan

Fig. 4 A dodecapeptide and an industrial sample CE-1270 used in iron-catalyzed hydrogenation reactions

low (12%) due to the low thermal stability of the catalyst as well as its sensitivity toward impurities [66]. ð25Þ The complex iPrFeHBH4 is a precatalyst; under heating, it releases BH3 to generate the active species trans-(iPrPNHP)FeH2(CO) [22, 24]. This process can be facilitated by the addition of Et3N to trap BH3, resulting in a more efficient catalytic system [22]. However, using other bases such as KOtBu and Na2CO3 can reduce the alcohol yield [24]. In the presence of KOtBu (or NaOMe) and under H2, iPrFeHBr is also converted to trans-(iPrPNHP)FeH2(CO), thus catalyzing ester hydrogenation, although it can be complicated by base-promoted transesterification with the alcohol products [95]. To understand the substituent effects, Beller replaced the isopropyl groups in iPr FeHBH4 with ethyl or cyclohexyl groups and studied the catalytic performance of these new borohydride complexes [96]. Consistent with the steric argument, at a relatively low temperature of 60 C, EtFeHBH4 performs better than iPrFeHBH4, which is in turn more reactive than CyFeHBH4 for the hydrogenation of methyl benzoate. Notably, Me2NCH2CH2CO2Me, which is not a viable substrate for the Ru-MACHO system, can be smoothly hydrogenated to Me2N(CH2)3OH at 100 C under 30 bar H2 using EtFeHBH4 as the precatalyst (1 mol%). It should be emphasized here that temperature and H2 pressure play profound roles in controlling the activation, stability, and reactivity of the borohydride complexes and ultimately their catalytic efficiency. A closely related study by Langer showed a decreasing reactivity order of iPrFeHBH4 > CyFeHBH4 > EtFeHBH4 when the hydrogenation of methyl benzoate was conducted as 100 C under 10 bar H2 (Eq. 26) [93].

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

289

Scheme 18 Catalytic hydrogenation of alkyl levulinates

H3B R2 H CO H P Fe PR2 cat. N

O Ph

OMe

+ H2

pH2 = 30 bar, 60 oC o

pH2 = 10 bar, 100 C

H

THF, 6 h

Ph

OH

EtFeHBH 4

> iPrFeHBH4 > CyFeHBH4

iPrFeHBH 4

> CyFeHBH4 > EtFeHBH4

ð26Þ

Catalytic hydrogenation of levulinates, which contain two different types of carbonyl groups, has been explored with the iron PNP-type complexes. Very recently, Paixão and Nielsen showed that at 60 C under 10 bar H2, iPrFeHBH4 combined with KOMe catalyzed the hydrogenation of ethyl levulinate to γvalerolactone with 25: 1). Here, the presence of the NH moiety is critical to the success of the hydrogenation process. A control experiment using (iPrPNMeP)FeH (CO)(BH4) as the catalyst did not yield any hydrogenation product. In a follow-up study, Beller showed that CyFeHBH4 was similarly effective, whereas EtFeHBH4 became inactive when the catalyst loading was reduced from 1 mol% to 0.5 mol% [98]. According to that study, temperature is very critical for the outcome of the hydrogenation. Hydrogenation of PhCN performed below 70 C leads mainly to the secondary imine PhCH¼NCH2Ph.

ð28Þ N-heterocycles have been studied as potential organic hydrogen storage materials through reversible acceptorless dehydrogenation and hydrogenation reactions, both of which require a catalyst. In 2014, Jones reported that iPrFeHBr, when activated by KOtBu, was effective for the hydrogenation of quinoline derivatives to 1,2,3,4tetrahydroquinaldines (Eq. 29) [100]. Related N-heterocycles including 2-methylindole and 2,6-lutidine are also hydrogenated under similar conditions. As expected, iPrFeHBH4 also serves a precatalyst (without a base additive) for this process, although it is less active, resulting in 89% of quinoline being hydrogenated even at a higher temperature of 110 C. According to DFT calculations by Surawatanawong, the first hydrogenation event converts quinoline to 1,4-dihydroquinoline, which undergoes base-assisted isomerization to 3,4-dihydroquinoline [112]. Further hydrogenation of the C¼N bond furnishes the 1,2,3,4-tetrahydroquinaldine product.

ð29Þ

Typically, olefins are not considered viable substrates for hydrogenation systems that operate via metal-ligand cooperation. However, when the C¼C bonds are significantly polarized, they can accept H and H+ from H–M–N–H-type complexes in a similar way as carbonyl groups. In a recent study, Jones demonstrated this

292

D. A. Ekanayake and H. Guan

concept in studying iron-catalyzed hydrogenation of styrene and its derivatives [113]. At room temperature under an atmospheric H2 pressure, styrene is converted to ethylbenzene quantitatively in 24 h when iPrFeHBr (5 mol%) mixed with KOtBu (15 mol%) or iPrFeH (5 mol%) is employed as the catalyst. The borohydride complex iPrFeHBH4 is significantly less active due to the need to remove BH3, which is usually favored at elevated temperatures. Under the optimized conditions (Eq. 30), substituted styrenes, especially those containing electron-withdrawing groups, undergo C¼C bond hydrogenation smoothly. Because the reaction conditions are very mild, other reducible functional groups such as ester, pyridyl, and CN are tolerated, although hydrogenation of 4-cyanostyrene is sluggish due to catalyst inhibition by substrate coordination. Hydrogenation of trans-PhCH¼CHCOCH3 eventually gives the fully saturated product PhCH2CH2CH(OH)CH3. At the early stage of the reaction, C¼O hydrogenation is faster than C¼C hydrogenation. Consistent with a mechanism featuring metal-ligand cooperativity, weakly polarized C¼C bonds such as those in 1-hexene and tert-butylethylene resist hydrogenation, and the methylated complex (iPrPNMeP)FeH(CO)(BH4) shows no catalytic activity even at 100 C.

ð30Þ

3.2.4

Hydrogenation Reactions Related to CO2 or CO Reduction

Combining iron catalysis with CO2 reduction addresses many sustainability-related challenges [114]. Like the ruthenium-based systems described earlier, iron-based PNP-type complexes have also been explored in variety of transformations that are associated with CO2 reduction. Once again, the discussion here is organized based on how formal oxidation state of the carbon changes during hydrogenation (Fig. 3). For an example involving a change of +4 to +2 in carbon oxidation state, Hazari and Schneider showed in 2014 that hydrogenation of CO2 (1:1 mixture with a total pressure of 70 bar) could be catalyzed by CyFeH at 80 C in the presence of 300 equiv. DBU, which yielded formate with a TON of 186 in 12 h [101]. Adding 150 equiv. LiBF4 to the reaction mixture improves the TON to 289 in 4 h. Detailed mechanistic studies by Hazari and Bernskoetter suggest that the Lewis acid disrupts the intramolecular hydrogen bonding interaction between the NH moiety and the formato group and facilitates the release of HCO2 from iron [115]. Further screening of Lewis acids reveals that the hydrogenation reaction is best carried out in the presence of LiOTf with an optimal DBU to LiOTf ratio of 7.5 to 1. Under such conditions, hydrogenation of CO2 catalyzed by iPrFeH and CyFeH gives formate with TONs of 6,030 and 8,910, respectively (Scheme 19). The borohydride complex

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

293

Scheme 19 Hydrogenation of CO2 to formate catalyzed by various iron-based PNP pincer complexes

iPr

FeHBH4 displays a lower catalytic activity. Similar to the ruthenium-based catalytic systems, iron-catalyzed hydrogenation of CO2 to the formate stage does not require the presence of the NH moiety. As a matter of fact, (iPrPNMeP)FeH (CO)BH4 and (CyPNMeP)FeH(CO)BH4 are significantly more active with an about 30-fold increase in formate yield. For additional modification to the catalyst structure, Hazari and Bernskoetter incorporated different isocyanide ligands into the PNP pincer system. The five-coordinate complexes iPrFeH(CNArMe2) and iPrFeH (CNArOMe) prove to be less active than the CO analog iPrFeH [102]. The secondgeneration isocyanide-based catalysts supported by the methylated PNP ligand iPr PNMeP show some improvement over iPrFeH(CNArMe2) and iPrFeH(CNArOMe); however, they are still less effective than the corresponding CO derivatives [116]. Another formally two-electron reduction process with CO2 is N-formylation of amines, as mentioned in the ruthenium systems (Scheme 12). For iron-based catalysts, Bernskoetter compared the activity of iPrFeH, its adduct with HCONHPh, (iPrPNMeP)FeH(CO)BH4, and trans-(iPrPNMeP)FeH2(CO) for the N-formylation of morpholine [117]. Under the conditions outlined in Eq. 31, the reaction catalyzed by iPr FeH generates the formamide with a TON of 1930. The catalytic performance is slightly better than the HCONHPh adduct but worse than the methylated PNP complexes, again illustrating that the NH moiety is not needed for CO2 reduction to the formate stage.

294

D. A. Ekanayake and H. Guan

ð31Þ The iron-catalyzed amide hydrogenation has already been described in the previous section (see Eq. 27). Hydrogenation of formamides to methanol is singled out and discussed here due to its relevance to CO2 reduction (which changes the carbon oxidation state from +2 to 2). Under Sanford’s conditions (0.33 mol% Cy FeHBH4, 1.66 mol% K3PO4, 20 bar H2, 110 C, 3 h), N-formylmorpholine, HCONHAr, and HCONPh2 are hydrogenated to methanol with TONs of up to 300 [97]. Hydrogenation of HCONHMe and HCONH2 is problematic, providing methanol with only 1–12% yield. Bernskoetter’s system (0.018 or 0.07 mol% iPr FeH, 30.4 bar H2, 100 C, 4 h) hydrogenates N-formylmorpholine, HCONHAr, and HCONPh2 to methanol with TONs typically falling in the range of 1,190–4,430 [110]. Hydrogenation of HCONMePh under the same conditions is low yielding (TON ¼ 60) but can be improved by adding 20 equiv. of HCONHPh (TON ¼ 1,300). The overall hydrogenation process consumes 2 equiv. of H2 (for a formally fourelectron reduction process), first converting formamides to hemiaminals and then to methanol. This requires decomposition of hemiaminals to formaldehyde and amines, a process that can be catalyzed by iron or the formamide substrates, depending on the nitrogen substituents [118]. The process of CO to ethylene glycol via oxamide described in Scheme 13 has also been studied with iron-based PNP pincer complexes (i.e., iPrFeHBH4, Cy FeHBH4, EtFeHBH4, and trans-(EtPNHP)FeBr2(CO)), although the focus is on the second step that hydrogenates the oxamide to ethylene glycol [85]. With 0.2 mol % an iron catalyst and 1–1.5 mol% KOtBu, after 6 h, only 18–53% of the oxamide is hydrogenated. However, using 2 mol% EtFeHBH4 along with 5 mol% KOH and extending the reaction time to 24 h leads to a full conversion of the oxamide with 77% of the hydrogenation products attributed to ethylene glycol (Eq. 32).

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

295

Scheme 20 A two-step hydrogenation of CO2 to methanol catalyzed by an iron complex

ð32Þ Direct hydrogenation of CO2 to methanol assisted by amines, which changes the carbon oxidation state from +4 to 2, is more challenging with iron catalysts due to their relatively low thermal stability. The ruthenium systems described earlier operate most efficiently at 140–160 C. A thermal stability study of iPrFeHBH4 conducted by Jones showed that at 140 C this compound decomposed completely in 4 h [100]. Nevertheless, some of the iron-based PNP pincer complexes have been tested for this transformation. Olah and Prakash’s strategy of using PEHA to capture CO2 and iPrFeHBr to hydrogenate the captured CO2 (145 C, 70 bar H2, 72 h) failed to produce any methanol. Instead, formate and formamide were detected with NMR yields of 20% and 18%, respectively [89]. On the other hand, hydrogenation of the captured CO2 in a biphasic mixture (as illustrated in Scheme 15) was successful with iPr FeHBr at 55 C under 50 bar H2, which, after 10 h, gave formate in 96% yield. Like the Ru-MACHO-BH system, the iron catalyst can be reused at least four times without losing the catalytic activity [88]. Given the results in Eq. 31 and the fact that CyFeHBH4 and iPrFeH catalyze the hydrogenation of formamides to methanol [97, 110], one might expect some catalytic activity from these iron complexes for CO2 hydrogenation to methanol assisted by amines. A recent study by Bernskoetter suggests that these two steps are incompatible, thus preventing them being carried out in a single reactor [119]. In particular, CO2 poisons the catalyst during formamide hydrogenation. Furthermore, water (generated from CO2 hydrogenation) deactivates the catalytically active species. To solve these issues, N-formylation of morpholine was first catalyzed by iPr FeH in the presence of 3 Å sieves (Scheme 20). The resulting mixture was filtered to remove the sieves as well as ammonium carbamate salt of morpholine. A portion

296

D. A. Ekanayake and H. Guan

of the filtered mixture was then subjected to the second hydrogenation step catalyzed by iPrFeH. This two-step procedure provides methanol with a net TON of 590.

3.3

Osmium Catalysts

Osmium complexes have been rarely explored for hydrogenation reactions. DFT calculations on trans-(iPrPNHP)MH2(CO) (M ¼ Fe, Ru, Os) suggest that hydrogenation of MeCN is best catalyzed by iron and ruthenium and hydrogenation of methyl benzoate is best catalyzed by ruthenium [120]. Such predications have not been validated experimentally. The only known osmium system involving the PNP-type ligand is the one developed by Gusev, who treated the iPrPNHP ligand with OsHCl(CO)(PPh3)3 much like for the synthesis of iPrRuHCl (Scheme 21) [57]. The isolated product iPrOsHCl was identified as an isomeric mixture, which underwent dehydrochlorination with KOtBu followed by dehydrogenation of iPrOH to yield the dihydride complex iPrOsH2. What is remarkable about these osmium hydride complexes is that both iPrOsHCl and iPrOsH2 are air and moisture stable in solution. In terms of hydrogenation reactions, iPrOsHCl and iPrOsH2 have been evaluated to catalyze the hydrogenation of hexyl octanoate, cis-3-hexenyl hexanoate, and triglycerides [121]. As for the analogous ruthenium and iron complexes, iPrOsHCl needs to be activated by a strong base such as NaOtBu. The best conditions for hydrogenating hexyl octanoate (in toluene) involve 0.1 mol% iPrOsH2 (loaded in air) at 220 C under 55.2 bar H2 for 24 h, which results in 87% conversion of the ester with high selectivity for the alcohol products. The mixture of iPrOsHCl and NaOtBu shows slightly lower activity. The hydrogenation reaction is operative under neat conditions, and loading the catalysts under an inert atmosphere improves the yield by 6–15%. Hydrogenation of cis-3-hexenyl hexanoate with the osmium catalysts saturates the C¼C bond first, during which process the catalysts also degrade, showing no activity toward the ester functionality. To circumvent the issue, hydrogenation of cis-3-hexenyl hexanoate and seed oil (a mixture of canola and soybean oil) is first performed with Pd/C, a heterogeneous catalyst, to saturate the C¼C bonds (Scheme 22). After filtration to remove Pd/C, the resulting saturated esters are subjected to hydrogenation catalyzed by iPrOsH2, which reduces the esters with a 60–90% conversion.

Scheme 21 Synthesis of osmium-based hydrogenation (pre)catalysts

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

297

Scheme 22 A two-step process for the hydrogenation of seed oil

4 Group 9 Metal Systems 4.1

Rhodium Catalysts

As mentioned in Introduction, rhodium holds historical significance in hydrogenation catalysis, particularly for the early efforts to hydrogenate C¼C bonds. It is thus somewhat surprising that there is very little development of the PNP-ligated rhodium complexes as catalysts for the modern-day hydrogenation reactions. In 1984, Taqui Khan reported the synthesis of (PhPNHP)RhCl from the reaction of [RhCl(COE)2]2 (COE ¼ cyclooctene) with PhPNHP in benzene [122]. In a series of subsequent reports, this specific PNP complex was shown to catalyze the hydrogenation of cyclohexene [122], 1-heptene [123], and 1-pentene [124] at 10–50 C under 0.4–1 bar H2. The proposed mechanism is analogous to the one for Wilkinson’s RhCl(PPh3)3 catalyst, which involves H2 activation followed by olefin coordination [125]. Based on the NMR analysis, oxidative addition of H2 to (PhPNHP)RhCl produces thee dihydride complexes with the formula (PhPNHP)RhH2Cl. The major product (90%) is consistent with cis-(PhPNHP)RhH2Cl with the PhPNHP ligand adopting the meridional configuration [126]. A more recent study by Jagirdar showed that (PhPNHP)RhH2Cl was unable to catalyze the hydrogenation of aldehydes, ketones, imines, and CO2 at 50 C under 20 bar H2 [127]. These results do not rule out the possibility of using the rhodium-based PNP-type complexes for the hydrogenation of polar bonds, because the hydrogenation reactions were attempted under base-free conditions and the active species could be (PhPNHP)RhH3.

298

4.2 4.2.1

D. A. Ekanayake and H. Guan

Cobalt Catalysts Synthesis of (Pre)catalysts

In contrast to the limited examples for the PNP-type complexes of rhodium, many cobalt derivatives have been studied, including spectroscopic observation and crystallographic characterization of (iPrPNHP)CoH2Cl and (iPrPNHP)CoH3 [128]. While these Co(III) hydrides have yet to be employed for hydrogenation reactions, close to a dozen cobalt complexes supported by the RPNHP ligands have been prepared (Scheme 23) and evaluated as hydrogenation catalysts. As illustrated in Scheme 23, the first class of cobalt PNP-type complexes feature an alkyl or aryl donor, and the synthesis starts with (pyr)2Co(CH2SiMe3)2 (pyr ¼ pyridine) (Method A). Its ligand substitution reaction with CyPNHP gives Cy CoCH2TMS, which can be protonated on the nitrogen by Brookhart’s acid, [H (OEt2)2]BArF4, to form a cationic complex [CyCoCH2TMS]+ [16]. Treatment of [CyCoCH2TMS]+ with 1-phenylethanol results in a Co(III) hydride [CyCoHAr]+, which appears to dehydrogenate the alcohol and then activate the C–H bond of the dehydrogenation product, acetophenone [129]. The reaction of a RPNHP ligand with CoCl2 or CoBr2 provides another entry to cobalt-based PNP-type complexes (Method B). Exposure of iPrCoCl2 and iPrCoBr2 to CO produces iPrCoCl2(CO) and iPrCoBr2(CO) [48, 49, 52], and reduction of

Scheme 23 Synthetic routes to cobalt-based hydrogenation (pre)catalysts

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

299

iPr

CoCl2 with 1 equiv. of NaBH4 generates a Co(I) species iPrCoCl [48]. The latter compound can react with CO to yield a cationic bis(carbonyl) complex [iPrCo (CO)2]+ [48], which can alternatively be prepared from the reaction of iPrPNHP with CoCl(PPh3)3 under CO [130].

4.2.2

Applications for Catalytic Hydrogenation Reactions

The first hydrogenation system involving the cobalt-based PNP-type complexes appeared in a 2012 report by Hanson [16]. In that study, 1:1 mixture of Cy CoCH2TMS and [H(OEt2)2]BArF4, which essentially generated Cy + [ CoCH2TMS] in situ, was shown to catalyze the hydrogenation of terminal and disubstituted (1,1- or 1,2-) olefins at 25 C under an atmospheric H2 pressure (Eq. 33). Without the acid, CyCoCH2TMS alone is almost completely inactive. Aldehydes, ketones, and aldimines are also viable substrates under same conditions or a slightly higher temperature and/or H2 pressure. At 25 C, hydrogenation of C¼C bonds is unaffected by the presence of an ester, carboxylic acid, amine, or alcohol group in the olefin substrate and only slightly inhibited by water.

ð33Þ

In a follow-up study, Hanson used the methylated PNP complex [(CyPNMeP) CoCH2SiMe3]BArF4 to probe the role that NH moiety could play during the hydrogenation reactions [129]. Evidently, the NH functionality is not needed for olefin hydrogenation but absolutely required for ketone hydrogenation (performed at 25–60 C under 1 bar H2). The lack of metal-ligand cooperation in olefin hydrogenation has been supported by DFT calculations [131]. The proposed mechanism involves C¼C bond insertion into the Co–H bond of [(CyPNHP)CoH]BArF4 or [(CyPNMeP)CoH]BArF4 followed by hydrogenolysis of the cobalt alkyl species, in which NH or NMe does not directly participate. The Co(III) hydride [CyCoHAr]+ also shows good activity for hydrogenating styrene under ambient conditions but limited activity for hydrogenating acetophenone even at 60 C under 4.1 bar H2 [129]. Under more forcing conditions, carbonyl groups can be hydrogenated, not only by [CyCoCH2TMS]+ but also by the methylated derivative [(CyPNMeP) CoCH2SiMe3]BArF4. Jones reported in 2017 that both cationic complexes were effective catalysts for ester (or lactone) hydrogenation at 120 C under 55 bar H2

300

D. A. Ekanayake and H. Guan

(Eq. 34) [132]. As expected, hydrogenation of α,β-unsaturated esters with [CyCoCH2TMS]+ results in both C¼C and C¼O bonds being reduced, although C¼C bond hydrogenation appears to be faster. In contrast to olefin hydrogenation described earlier, carboxylic acid interferes with ester hydrogenation. No hydrogenation product was observed when adipic acid monoethyl ester was employed as the substrate. The uniqueness about this cobalt-based catalytic system is that methyl esters usually give lower alcohol yields when compared to the corresponding ethyl esters. Mechanistic investigation focusing on methyl benzoate revealed that [CyCoCH2TMS]+ lost its catalytic activity by forming [(CyPNHP)Co(κ 1-OCOPh) (κ2-OCOPh)]BArF4, presumably via methane elimination. Similar to the mechanism proposed for olefin hydrogenation, [(CyPNHP)CoH]BArF4 or [(CyPNMeP)CoH] BArF4 is thought to be the active species, although according to DFT calculations, some of intermediates during ester hydrogenation feature a significant distortion of the PNP ligand from the meridional geometry [133].

ð34Þ

Under similar conditions (100–140 C, 50 bar H2), the cobalt complexes listed in Scheme 23, Method B, when activated by NaOMe, all display some level of catalytic activity for the hydrogenation of methyl benzoate [48]. The best precatalyst is Ph CoCl2, which promotes the hydrogenation of various esters including lactones (Eq. 35). Unlike the catalytic system shown in Eq. 34, here C¼C bonds can be tolerated. Substrates that lead to low alcohol yields include PhCO2tBu (due to sterics) and chloro- or bromo-substituted methyl benzoate (due to dehalogenation). This particular catalytic system proves to operate via metal-ligand cooperation; control experiments using the methylated complex (PhPNMeP)CoCl2 did not yield any hydrogenation products.

ð35Þ

The cobalt-based PNP-type complexes can also be used to catalyze the hydrogenation of other multiple bonds including those in nitriles and N-heterocycles. In 2018, we reported that catalytic hydrogenation of PhCN could be affected by iPr CoCl2 or iPrCoBr2 in the presence of NaHBEt3, forming PhCH¼NCH2Ph exclusively as the hydrogenation product (Scheme 24) [134]. Adding 1 equiv. of CyNH2 to the reaction generated PhCH¼NCy selectively, which represents a hydrogenative coupling process. The selectivity of nitrile hydrogenation can be altered to favor primary amines, as demonstrated by Beller in a more recent study [135]. Among the

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

301

Scheme 24 Cobaltcatalyzed hydrogenation of nitriles leading to secondary imines or primary amines

PNP-ligated cobalt dihalide complexes shown in Scheme 23, PhCoCl2 is the most active precatalyst, converting various aromatic and aliphatic nitriles to primary amines (Scheme 24). Functional groups tolerated under the catalytic conditions include F, Cl, NH2, OMe, pyridyl, and pyrrolidyl groups; however, carbonyl groups in esters, ketones, and aldehydes are also hydrogenated along with the nitrile groups. The nature of the catalytically active species is ill-defined here, although all experiments suggest that the hydrogenation process is homogeneous. The lack of reactivity with the methylated complex (PhPNMeP)CoCl2 also supports a metal-ligand cooperative mechanism. As a further exploration of N-heterocycles as organic hydrogen storage materials, Jones studied the ability of [CyCoCH2TMS]+ to catalyze the hydrogenation of these molecules [136]. Under the conditions shown in Eq. 36, the hydrogenation process takes place very slowly, accepting 2 equiv. of H2 to saturate one nitrogen-containing ring. In contrast to the iron-based catalytic system (Eq. 29), 2,6-lutidine is not a viable substrate for the cobalt catalyst. Analogous to the olefin hydrogenation catalyzed by [CyCoCH2TMS]+, the NH moiety is not needed here.

ð36Þ

Catalytic hydrogenation of CO2 has not been explored extensively with the cobalt-based PNP-type complexes described above. The only known example is Bernskoetter’s study of [iPrCo(CO)2]+ as a potential catalyst [130]. Under the conditions outlined in Eq. 37, hydrogenation of CO2 gives the formate with 450 turnovers. Similar to the iron-based system (Scheme 19), the methylated complexes [(iPrPNMeP)Co(CO)2]Cl and [(CyPNMeP)Co(CO)2]Cl are more superior catalysts than [iPrCo(CO)2]+ for CO2 hydrogenation, increasing the TON by 64- or 53-fold

302

D. A. Ekanayake and H. Guan

[137]. Once again, for CO2 hydrogenation of to the formate stage, the metal-ligand bifunctional catalysts do not appear to have any advantage.

ð37Þ

4.3 4.3.1

Iridium Catalysts Synthesis of (Pre)catalysts

Iridium-based PNP-type complexes are also known in the literature. To develop a hydrogenation catalyst, Taqui Khan treated [Ir(COE)2Cl]2 with the hydrochloride salt of PhPNHP in refluxing benzene, which resulted in a compound with the formula (PhPNHP)IrCl, presumably PhIrCl as shown in Scheme 25 [122]. A more recent synthesis by Jagirdar employed [Ir(COD)Cl]2 and the neutral ligand PhPNHP, which formed [(PhPNHP)Ir(COD)]Cl with the COD ligand still bound to iridium [127]. Subsequent hydrogenation produced an air-stable Ir(III) dihydride PhIrH2Cl, which was further converted to PhIrH3 via dehydrochlorination under H2. It is interesting to

Scheme 25 Synthesis of iridium-based hydrogenation (pre)catalysts

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

303

note that PhIrH3 exists as a 1:1 isomeric mixture with the PNP ligand adopting either meridional or facial coordination mode. In contrast, the isopropyl analog iPrIrH3 displays the meridional mode only. This trihydride complex can be prepared from dehydrochlorination of the dihydride iPrIrH2Cl followed by dehydrogenation of i PrOH [138]. iPrIrH2Cl is commercially available but can be made from iPrPNHP and [Ir(COE)2Cl]2 in iPrOH at 80 C. It is also worth to point out that in the solid form iPr IrH2Cl is air stable and iPrIrH3 is moderately air stable.

4.3.2

Applications for Catalytic Hydrogenation Reactions

The use of iridium-based PNP-type complexes for catalytic hydrogenations reactions can be traced back to 1984, when Taqui Khan studied the hydrogenation of cyclohexene catalyzed by PhIrCl [122]. This reaction operates over the temperature range 20–50 C under 0.4–1 bar H2 and proceeds via an initial H2 activation to form Ph IrH2Cl [125]. The catalytic system that really takes advantage of metal-ligand cooperativity is the one developed by Abdur-Rashid in 2009 [139]. It was reported that aldehyde and ketone hydrogenation could be catalyzed by iPrIrH2Cl activated with KOtBu or by iPrIrH3 under base-free conditions. The catalysts are remarkably active at room temperature; the TONs for acetophenone hydrogenation are as high as 30,000 (Eq. 38). Hydrogenation of benzalacetone and β-ionone is chemoselective for the C¼O bonds; however, hydrogenation of 2-cyclohexen-1-one produces a 1:1 mixture of allyl alcohol and the fully saturated alcohol. In a related study, Jagirdar examined the catalytic activity of the phenyl derivatives (PhIrH3 and PhIrH2Cl/ KOtBu) in hydrogenation reactions, which were carried out at 50 C in methanol under 20 bar H2 with a catalyst loading of 0.1 mol% [127]. In addition to aldehydes and ketones, imines such as PhCH¼NPh and PhCH¼NBn are hydrogenated, albeit with moderate conversions (31–49% over 6 h). In contrast, methyl benzoate and styrene are completely unreactive.

ð38Þ Abdur-Rashid’s iridium complex iPrIrH2Cl has also been utilized to catalyze the hydrogenation of alkyl levulinates to γ-valerolactone (Eq. 39) [71]. The reaction is enhanced by added ethanol or methanol, and under the optimized conditions, γvalerolactone was obtained with TONs of up to 9,300. Compared to Ru-MACHO, iPr IrH2Cl is more active, although the ruthenium catalyst can be reused three times without noticeable catalyst decomposition.

304

D. A. Ekanayake and H. Guan

Scheme 26 Iridiumcatalyzed hydrogenation of CO2 or N-formylation of morpholine

ð39Þ Esters can be hydrogenated with the iridium-based PNP-type complexes, although the reaction must be conducted at higher temperatures and under higher H2 pressures. In 2014, Beller showed that in the presence of NaOMe and at 130 C under 50 bar H2, both iPrIrH2Cl and iPrIrH3 were efficient for catalytic hydrogenation of methyl benzoate [140]. Based on the proposed mechanism, iPrIrH3 is the active species transferring H+/H to the ester substrate, and therefore the base should not be needed for iPrIrH3. However, the addition of NaOMe does improve the conversion and yield, suggesting that the base plays multiple roles during the reaction. The catalytic system (Eq. 40) can tolerate functional groups including halogens, MeO, pyridyl, and furyl groups. Hydrogenation of p-NCC6H4CO2Me and PhCH¼CHCO2Me leads to saturation of CN, C¼O, and C¼C bonds. Hydrogenation of phthalic anhydride, on the other hand, can stop at the lactone stage to give phthalide in 71% yield.

ð40Þ Another important type of carbonyl substrates for the iridium-catalyzed hydrogenation reactions is CO2. In 2011, Hazari reported a very facile CO2 insertion process with iPrIrH3, resulting in an iridium formate complex iPrIrH2(OCHO) that is air stable and features a hydrogen bond between the NH group and the formato group (Eq. 41) [141]. iPrIrH2(OCHO) was then employed to catalyze the hydrogenation of CO2 in an aqueous solution of KOH (1 M), providing HCO2K with TONs of up to 348,000 (Scheme 26). The trihydride iPrIrH3 can also be used as the catalyst, although precaution needs to be taken to exclude oxygen from the reactor. Very recently, Jagirdar demonstrated that the phenyl derivative PhIrH3 (a 1:1

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

305

isomeric mixture) reacted with CO2 (1 bar at room temperature) to form an insertion product analogous to iPrIrH2(OCHO) [127]. Hydrogenation of CO2 in MOH (M ¼ Li, Na, K) with PhIrH3 or PhIrH2Cl produced HCO2M with much lower TONs of 65–144, although the hydrogenation reactions were tested under relatively low temperatures and pressures. In studying N-formylation of morpholine, Ding also examined the catalytic activity of iPrIrH2Cl (activated by KOtBu), which, under the conditions shown in Scheme 26, generated the formamide with a TON of 720 [82]. An attempt to use ethylene carbonate as CO2 surrogate had limited success with iPrIrH2Cl/KOtBu as the catalyst (0.1 mol%); at 140 C under 50.7 bar H2, ethylene glycol was obtained with only 10% yield [55]. The ruthenium system shown in Eq. 23 is far more reactive.

ð41Þ

5 Group 10 Metal Systems Group 10 metals bearing the PNP-type ligands have been rarely used as hydrogenation catalysts. The only known example of a nickel system is the one developed by Hanson in 2012 [142]. As summarized in Scheme 27, the reaction of Ni(diglyme)Br2 with CyPNHP produces a cationic PNP pincer nickel bromide complex, which can be converted to the hydride [CyNiH]+ using NaBH4 followed by anion exchange with NaBPh4. The neutral hydride CyNiH is available from [CyNiH]+ through deprotonation with KH. Complex [CyNiH]+ proves to be an active catalyst for the hydrogenation of styrene, α-methylstyrene, and tert-butylethylene at 80 C under 4.1 bar H2 (Eq. 42) [142]. Hydrogenation of 1-octene affords n-octane and internal octenes as a result of the competing olefin isomerization process. Under similar conditions, aldehydes are reduced to alcohols but in a non-catalytic manner. The neutral hydride CyNiH is also

Scheme 27 Synthesis of PNP-ligated nickel hydride complexes

306

D. A. Ekanayake and H. Guan

an active catalyst, although it is less reactive than [CyNiH]+. The methylated complex (CyPNMeP)NiH]BPh4 shows similar activity to [CyNiH]+, suggesting that here a metal-ligand cooperative mechanism is not involved.

ð42Þ

The analogous palladium and platinum hydrides have not been reported in the literature. The most relevant study is a 1988 report by Taqui Khan, who used [(PhPNHP)PdCl]Cl (made from Pd(COD)Cl2 and the hydrochloride salt of PhPNHP in benzene) to catalyze hydrogenation of cyclohexene [143]. The reaction was shown to operate at 10–40 C under 0.4–1 bar H2 and proceed via a palladium hydride intermediate.

6 Group 6 Metal Systems Mid-transition metal complexes supported by the PNP-type ligands have been studied. For group 6 metal systems, chromium complexes have never been utilized to catalyze hydrogenation reactions, although (RPNHP)CrCl2 [144] and (RPNHP) CrCl3 [145] have been known for many years. In contrast, PNP-ligated molybdenum and tungsten complexes have been developed specifically for various hydrogenation processes. They belong to two different types of complexes, each with a d6 electron configuration and isoelectronic to (RPNP)Fe(CO)H, which have already been established as active hydrogenation catalysts.

6.1

Nitrosyl Complexes

To synthesize the desired nitrosyl complexes (RPNP)M(NO)CO (M ¼ Mo, W), Berke used M(NO)(CO)4(AlCl4) as the metal precursors, which were shown to react with the iPrPNHP ligand to form (iPrPNHP)M(NO)(CO)Cl (Scheme 28) [146]. Upon further treatment with NaN(SiMe3)2, the five-coordinate complexes iPrMoNO and iPr WNO were isolated as highly air-sensitive materials. Activation of H2 by iPrMoNO and iPrWNO is feasible but slow at room temperature, forming a mixture of two isomers because of the availability of two sides (NO side vs. CO side) for H2 to approach (Eq. 43). As expected for other H–M–N–

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

N

( Pr)2 NO CO P NaN(SiMe3)2 M i P( Pr)2 THF, RT N H Cl i

i

H P( Pr)2 i

P( Pr)2

THF 90 oC

+ M(NO)(CO)4(AlCl4)

307

i

( Pr)2 NO P M CO i P( Pr)2 N iPrMoNO: iPrWNO:

M = Mo M=W

Scheme 28 Synthesis of Mo- and W-based hydrogenation catalysts bearing a nitrosyl ligand Scheme 29 Molybdenumand tungsten-catalyzed hydrogenation of imines and nitriles

H-type complexes, reduction of polar bonds is likely to occur with these hydrides. Indeed, Berke demonstrated that iPrMoNO and iPrWNO were efficient catalysts for the hydrogenation of aldimines bearing various aryl substituents (Scheme 29) [146]. Substrates that are unreactive under the catalytic conditions include pNO2C6H4CH¼NPh, PhCH¼NiBu, and surprisingly PhCHO. Acetophenone can be hydrogenated but with a low yield of 32%. Under slightly modified conditions, both aliphatic and aromatic nitriles are hydrogenated with selectivity favoring the secondary imines [147]. This implies that hydrogenation of the intermediate R0 CH¼NH to R0 CH2NH2 is less competitive than the reaction of R0 CH¼NH with R0 CH2NH2 to yield R0 CH¼NCH2R0 . In both catalytic processes, iPrMoNO displays high activity than iPrWNO, which is also confirmed by DFT calculations [148].

ð43Þ

Attempts have also been made to use iPrMoNO and iPrWNO to catalyze CO2 hydrogenation [149]. The in situ generated hydrides (see Eq. 43) were shown to undergo CO2 insertion to generate molybdenum and tungsten formate complexes, which could be converted back to iPrMoNO and iPrWNO through the addition of NaN(SiMe3)2. Unfortunately, catalytic hydrogenation of CO2 ( pH2 ¼ 70 bar, pCO2 ¼ 10 bar, 140 C) in the presence of a base and with iPrMoNO or iPrWNO (5 mol%) failed to produce HCO2 with a yield greater than 5%. This is likely due to the poisoning of the catalysts by CO2 to form carbamate species (Eq. 44), as separately studied.

308

D. A. Ekanayake and H. Guan

ð44Þ

6.2

Bis(Carbonyl) Complexes

The bis(carbonyl) system illustrated in Fig. 5 has recently been explored, though only focusing on molybdenum complexes. As shown in Scheme 30, treatment of iPr PNHP with Mo(CO)6 and Mo(PPh3)2(CO)2(MeCN)2 leads to ligand substitution, which generates iPrMo(CO)3 and iPrMoNCMe, respectively [150]. The latter reaction needs to be carried out in THF. Switching the solvent to CH2Cl2 can oxidize Mo (0) to Mo(I), giving the chloride complex iPrMoCl. The analogous bromide complex iPr MoBr is isolated as the minor product from the reaction of iPrPNHP with Mo(η3allyl)(PPh3)2(CO)2(MeCN)2Br (the major product is iPrMoNCMe). Both iPrMoCl and iPrMoNCMe, when activated by NaHBEt3, catalyze the hydrogenation of acetophenone, whereas iPrMo(CO)3 shows no activity [150]. Under the optimized conditions for iPrMoCl (Scheme 31), acetophenones substituted by F, MeO, MeS, and CF3 groups are all successfully hydrogenated to the corresponding alcohols in high yields. The reaction of (E)-PhCOCH¼CHPh Fig. 5 Molybdenum and tungsten complexes isoelectronic to (RPNP)Fe (CO)H

H N

OC CO CO

Mo

Mo(CO)6 toluene, reflux

AN = acetonitrile Mo(PPh3)2(CO)2(AN)2

( Pr)2NCMe CO H P Mo i P( Pr)2 N i

THF, RT P P( Pr)2 i i CO ( Pr)2 P( Pr) 2 H iPrMoNCMe iPr Mo(CO)3 i P( Pr)2 Br i N Cl i ( Pr)2 ( Pr)2 CO P CO H H P Mo Mo(PPh3)2(CO)2(AN)2 Mo(h 3-allyl)(CO)2(AN)2Br Mo i i P( Pr)2 N P( Pr)2 toluene, RT N CH2Cl2, RT CO CO iPrMoBr iPrMoCl i

Scheme 30 Synthesis of molybdenum-based precatalysts bearing at least two CO ligands

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

309

Scheme 31 Hydrogenation reactions catalyzed by iPrMoCl-NaHBEt3

results in both double bonds being reduced. Styrene derivatives are also viable substrates, although a higher temperature of 130 C is required. Hydrogenation reactions performed at 100 C allow the conversion of N-methylformanilides to Nmethylanilines with C¼C bond and ester functionality intact [151]. Other functional groups including PhCH2O, Me2N, CN, and NO2 are also compatible with the catalytic conditions; however, the product yields are low to moderate (6–52%). Amides of the type R0 CONPhR00 (R0 ¼ Me, CF3, Ph) are more challenging substrates, which typically give 11–28% yields for the hydrogenation products. The bromide complex iPrMoBr shows similar activity to iPrMoCl but outperforms iPr MoNCMe. The iPrMo(CO)3 is completely inactive. A detailed mechanistic study [151] focusing on iPrMoCl suggests that NaHBEt3 reduces the Mo (I) complex to several Mo(0) species including Na[(iPrPNHP)Mo(CO)2H] and Na [(iPrPNP)Mo(CO)2]. These two complexes represent the H–M–N–H and M–N molecules characteristic of metal-ligand bifunctional hydrogenation catalysts.

7 Group 7 Metal Systems 7.1

Manganese Catalysts

There has been an increasing interest in developing manganese-based hydrogenation catalysts [152]. This is in part motivated by the fact that manganese is the third most abundant transition metal (after iron and titanium) in the Earth’s crust. For PNP-type complexes, manganese species isoelectronic to (RPNP)Fe(CO)H would be (RPNP) Mn(CO)2. To date, strategies of using inexpensive sources of manganese such as MnCl2 to make these Mn(I) complexes have not had much success. For example,

310

D. A. Ekanayake and H. Guan

Scheme 32 Synthesis of manganese-based (pre)catalysts

(iPrPNHP)MnCl2 prepared from iPrPNHP and MnCl2 does not bind CO [153]. Successful routes to the carbonyl complexes have relied on the use of MnBr(CO)5 as the metal precursor. As illustrated in Scheme 32, the reaction of RPNHP with MnBr (CO)5 can lead to four different structures, which depend on the nature of phosphorus substituents. With medium-sized alkyl groups (R ¼ iPr, Cy, Et), a mixture of neutral dicarbonyl and cationic tricarbonyl complexes (e.g., iPrMnBr and [iPrMn (CO)3]+) is obtained [154]. The RPNHP ligands all adopt the meridional coordination mode, except that in [EtMn(CO)3]+, the EtPNHP ligand occupies three facial coordination sites [37]. The dicarbonyl and tricarbonyl complexes are separable due to solubility difference, although higher temperatures and longer reaction times usually facilitate the conversions to the neutral products. The phospholane-based chiral ligand, (S,S)-(CyMePCH2CH2)2NH, and (S,S)-(tBuMePCH2CH2)2NH can generate the cationic tricarbonyl complexes only [155, 156], whereas the more bulky ligand tBu PNHP leads to further extrusion of CO to yield [tBuMn(CO)2]+ [154]. The reaction of the phenyl-substituted ligand PhPNHP gives the neutral dicarbonyl complex PhMnBr [157]. Synthesis of the five-coordinate complexes iPrMn(CO)2 and EtMn(CO)2 has been accomplished via dehydrobromination of iPrMnBr [154, 158], EtMnBr [37], and [EtMn(CO)3]+ [37] with a strong base. The hydride iPr MnH is available from the reaction of iPrMnBr with NaHBEt3 [153]. The seminal work by Beller in 2016 showed that in the presence of NaOtBu, iPr MnBr were efficient in catalyzing the hydrogenation of nitriles, ketones, and aldehydes (Scheme 33) [153]. The relative difficulty for hydrogenating these substrates is reflected by the temperature and H2 pressure employed. As expected, aldehydes are the easiest ones to react. Various functional groups including halogens, CF3, NH2, pyridyl, furyl, and isolated C¼C bonds are tolerated under these conditions. The nitrile hydrogenation shows excellent selectivity for primary amines. Hydrogenation of PhCH¼CHCN produces the allylic and fully saturated

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

311

Scheme 33 Manganesecatalyzed hydrogenation of nitriles, ketones, and aldehydes

Scheme 34 Manganesecatalyzed asymmetric hydrogenation of ketones

amines. In contrast, hydrogenation of α,β-saturated aldehydes is highly selective for the C¼O bonds, forming the allylic alcohols exclusively. Under the conditions for ketone hydrogenation, other reducible functional groups such as ester, lactam, and cyclopropyl ring are unaffected. The proposed mechanism involves an outer-sphere hydrogen transfer from iPrMnH to the substrates followed by regeneration of the hydride through the reaction of iPrMn(CO)2 with H2. Consistent with this mechanism, iPrMnH catalyzes the hydrogenation of benzonitrile under similar but basefree conditions, although for some unknown reason, the addition of NaOtBu improves the catalytic efficiency. The cyclohexyl derivative CyMnBr is slightly less reactive than iPrMnBr in nitrile hydrogenation. Using iPrMnBr and CyMnBr as the precatalysts is more advantageous due to their high stability in air. Given the availability of chiral RPNHP ligands, it is possible to develop manganese-based PNP-type catalysts for asymmetric hydrogenation of ketones. Since high temperatures often erode enantioselectivity, the reaction conditions need to be further optimized. Using phospholane-based complex [R*Mn(CO)3]+ as the precatalyst and tert-amyl alcohol as the solvent, the Beller group was able to perform ketone hydrogenation at 40 C (Scheme 34) [58, 155]. The remarkable feature about this catalytic system is that aliphatic ketones are hydrogenated to alcohols with ee’s in the 51–83% range, which is typically difficult to achieve with other catalysts. However, bulky ketones including AdCOMe, tBuCOMe, and t BuCOnPr result in low alcohol yields, and ketones of the type PhCOR0 (R0 ¼ Me, Et, nPr, Cy) give low ee’s (11–19%). In a related study, Mezzetti explored Pstereogenic PNP-type manganese complexes as asymmetric hydrogenation catalysts [156]. Under the best conditions for (S,S)-[MetBuMn(CO)3]+ (Scheme 34),

312

D. A. Ekanayake and H. Guan

acetophenone derivatives are hydrogenated to the corresponding alcohols with ee’s up to 55%. Both experimental and computational studies support the involvement of (RPNP)Mn(CO)2 and (RPNHP)Mn(CO)2H in the catalytic cycle. It is interesting to know that the iron precatalyst (S,S)-MetBuFeHBr (shown in Scheme 17) is about 30 times more active than (S,S)-[MetBuMn(CO)3]+. This is due to the formation of a more stable alkoxide species with manganese, which is generated after hydrogen transfer from (RPNHP)Mn(CO)2H to the ketone substrate. Esters are comparatively more challenging substrates for hydrogenation. In a 2016 report, Beller showed that at 100 C under 30 bar H2, iPrMnBr and CyMnBr (activated with KOtBu) displayed very limited catalytic activity for the hydrogenation of methyl benzoate [37]. Lan and Liu recently also reported that at 120 C under 45 bar H2, iPrMnBr and PhMnBr (activated with KOtBu) catalyzed the hydrogenation of the ketone part of methyl 4-acetylbenzoate with only 3–9% of the ester functionality being reduced [159]. However, Beller demonstrated that both EtMnBr and [EtMn(CO)3]+ were effective hydrogenation catalysts for esters, converting methyl benzoate to benzyl alcohol in 97% yield. Under the conditions outlined in Eq. 45, various aromatic and aliphatic esters including lactones can be hydrogenated to alcohols. The catalytic system shows excellent functional group compatibility, similar to the iron system described earlier (Eq. 25).

ð45Þ Like the ruthenium- and iron-based catalytic systems, the manganese PNP-type complexes have been tested for the hydrogenation of amides and N-heterocycles. Hydrogenation of PhCONHPh proved to be unsuccessful with iPrMnBr (2 mol%, 110 C, 30 bar H2) [159] and [tBuMn(CO)2]+ (4 mol%, 130 C, 50 bar H2) [160] in the presence of KOtBu as the activator. Catalytic hydrogenation of quinoline was shown to be feasible with the neutral dicarbonyl complexes, although the conversions were low (Eq. 46) [159]. DFT calculations suggest that the lack of activity is in part due to the low hydricity of the manganese hydride intermediate. To improve the catalysts, one of the phosphorus donor groups was replaced by a pyridine or imidazole ring, which not only increases the hydricity but also creates a less crowded environment [159, 160].

ð46Þ

As suggested by DFT calculations [161], catalytic hydrogenation of CO2 to the formate stage should be possible with the manganese-based PNP-type complexes

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

313

(e.g., iPrMn(CO)2), although such process has not been validated experimentally. The closest work was done by Prakash, who used iPrMnBr and CyMnBr to catalyze the N-formylation of morpholine, benzylamine, and N-methylbenzylamine (Eq. 47) [162]. Here, the isopropyl derivative iPrMnBr is more efficient than CyMnBr. Other amines including amylamine and N,N0 -dimethylethylenediamine can also be converted to the formamides, but the yields are much lower (25–53%). The in situ generated N-formylmorpholine and HCONHBn can be further hydrogenated (at 150 C under 70–80 bar H2) to methanol with TONs of up to 36, again using iPr MnBr as the catalyst (0.5 mol% loading). Hydrogenation of pure Nformylmorpholine under the same conditions gives a substantially higher TON (128) for methanol. Unfortunately, direct hydrogenation of CO2 to methanol assisted by these amines including PEHA [89] has failed to produce any meaningful amount of methanol. This is likely due to the poisoning of the catalyst by CO2 during the formamide hydrogenation, as described in the iron system (Scheme 20).

ð47Þ

Two indirect methods of making methanol have been developed with the manganese-based PNP-type catalysts, one involving cyclic carbonates as CO2 surrogate and the other involving CO to methanol. In 2018, Leitner demonstrated the first approach by using iPrMnBr as the precatalyst and NaOtBu as the activator or the in situ generated iPrMn(CO)2 [158]. Under the best conditions for ethylene carbonate (0.1 mol% iPrMn(CO)2, 60 bar H2, 120 C), ethylene glycol and methanol were obtained with TONs of 620 and 400, respectively. As shown in Eq. 48, this process can be extended to other five- and six-membered cyclic carbonates.

ð48Þ Very recently, Checinski and Beller designed a CO-to-methanol process that utilized a nitrogen-containing promoter to capture CO in the form of formamide [163]. The subsequent manganese-catalyzed formamide hydrogenation was expected to produce methanol and regenerate the promoter. After an extensive

314

D. A. Ekanayake and H. Guan

computational screening and experimental validation, scatole and pyrrole were identified as the best promoters that balance the difficulty of amine carbonylation with that of hydrogenation to methanol. Under the optimized conditions shown in Eq. 49, methanol is obtained with high TONs. Other manganese-based PNP-type complexes have also been tested. Compared to iPrMnBr, CyMnBr, EtMnBr, and [EtMn(CO)3]+ are slightly less active, whereas PhMnBr and [tBuMn(CO)2]+ are almost inactive. The methylated complex (EtPNMeP)Mn(CO)2Br also displays limited reactivity, suggesting that in this case, the presence of the NH moiety is critical to the success of the hydrogenation process.

ð49Þ

7.2

Rhenium Catalysts

Rhenium-based PNP-type complexes have been prepared following the procedures established for the manganese system. The reaction of iPrPNHP with ReBr(CO)5 produces [iPrRe(CO)3]+ in which the PNP ligand adopts the facial coordination mode (Eq. 50) [164]. In contrast, the phospholane-based PNP ligand coordinates to the Re(CO)3+ fragment in a meridional fashion [58].

ð50Þ Both [iPrRe(CO)3]+ and [R*Re(CO)3]+ have been employed to catalyze the hydrogenation of ketones. Under the optimized conditions for [iPrRe(CO)3]+ (Scheme 35), most functional groups are tolerated with cyano, phenol, and boric acid being the exceptions [164]. While isolated C¼C bonds and internal C  C

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

315

Scheme 35 Rheniumcatalyzed hydrogenation of ketones

bonds are intact during the hydrogenation, α,β-unsaturated ketones typically give saturated alcohols as the major or sole products. Hydrogenation of R0 COCH2CH2CO2Me provides γ-butyrolactones, although a higher loading of [iPrRe(CO)3]+ (5 mol%) and KOtBu (10 mol%) is needed. Using the chiral precatalyst [R*Re(CO)3]+ results in low ee for hydrogenating acetophenone and moderate ee for hydrogenating cyclohexyl methyl ketone [58]. This level of enantioselectivity is lower than that achieved with the manganese analog but comparable to the ruthenium- and iron-based catalytic systems.

8 Summary and Outlook The use of the RPNHP ligands to design hydrogenation catalysts has been a fruitful patch in the field of homogeneous catalysis. As shown in this chapter, transition metal complexes supported by these ligands along with strong-field ligands such as CO, NO, and isocyanides have been so extensively studied that most of the mid- and late-transition metals have been involved. Some of the hydrogenation processes do not require the NH moiety. Examples include hydrogenation of weakly polarized C¼C bonds and hydrogenation of CO2 to formate. However, metal-ligand cooperativity enabled by the NH functionality does have advantage for the more challenging hydrogenation processes such as CO2 hydrogenation to methanol and amide hydrogenation. We envision that interests in using RPNHP ligated complexes for catalytic hydrogenation reactions will continue to grow in the future. In particular, group 5 and group 11 metals have not been explored to build PNP-type complexes specifically for hydrogenation reactions. A recent computation study focusing on (iPrPNHP)M (NO)2H (M ¼ V, Nb, Ta; see Fig. 6) suggests that they are promising catalysts [165]. Inspired by the structure of the active site of [Fe]-hydrogenase, Yang has computationally designed PNP-type complexes of iron [166] and cobalt [167] that contain acylmethylpyridinol as the ancillary ligand. These molecules present significant synthetic challenges but may provide a path for synthetic chemists to identify more robust and active hydrogenation catalysts.

316

D. A. Ekanayake and H. Guan

Fig. 6 Computationally designed hydrogenation catalysts

References 1. de Vries JG, Elsevier CJ (eds) (2007) The handbook of homogeneous hydrogenation. WileyVCH, Weinheim 2. Clarke ML (2012) Cat Sci Technol 2:2418 3. Werkmeister S, Junge K, Beller M (2014) Org Process Res Dev 18:289 4. Alig L, Fritz M, Schneider S (2019) Chem Rev 119:2681 5. Osborn JA, Jardine FH, Young JF, Wilkinson G (1966) J Chem Soc A:1711 6. Halpern J (1982) Science 217:401 7. Hallman PS, McGarvey BR, Wilkinson G (1968) J Chem Soc A:3143 8. Noyori R, Ohkuma T (2001) Angew Chem Int Ed 40:40 9. Shambayati S, Crowe WE, Schreiber SL (1990) Angew Chem Int Ed Engl 29:256 10. Zhang J, Leitus G, Ben-David Y, Milstein D (2006) Angew Chem Int Ed 45:1113 11. Saudan LA, Saudan CM, Debieux C, Wyss P (2007) Angew Chem Int Ed 46:7473 12. Ito M, Koo LW, Himizu A, Kobayashi C, Sakaguchi A, Ikariya T (2009) Angew Chem Int Ed 48:1324 13. Balaraman E, Gnanaprakasam B, Shimon LJW, Milstein D (2010) J Am Chem Soc 132:16756 14. Casey CP, Guan H (2007) J Am Chem Soc 129:5816 15. Sui-Seng C, Freutel F, Lough AJ, Morris RH (2008) Angew Chem Int Ed 47:940 16. Zhang G, Scott BL, Hanson SK (2012) Angew Chem Int Ed 51:12102 17. Shvo Y, Czarkie D, Rahamim Y, Chodosh DF (1986) J Am Chem Soc 108:7400 18. Ohkuma T, Ooka H, Hashiguchi S, Ikariya T, Noyori R (1995) J Am Chem Soc 117:2675 19. Doucet H, Ohkuma T, Murata K, Yokozawa T, Kozawa M, Katayama E, England AF, Ikariya T, Noyori R (1998) Angew Chem Int Ed 37:1703 20. Sui-Seng C, Haque FN, Hadzovic A, Pütz A-M, Reuss V, Meyer N, Lough AJ, Zimmer-De Iuliis M, Morris RH (2009) Inorg Chem 48:735 21. Otsuka T, Ishii A, Dub PA, Ikariya T (2013) J Am Chem Soc 135:9600 22. Qu S, Dai H, Dang Y, Song C, Wang Z-X, Guan H (2014) ACS Catal 4:4377 23. Sandoval CA, Ohkuma T, Muñiz K, Noyori R (2003) J Am Chem Soc 125:13490 24. Werkmeister S, Junge K, Wendt B, Alberico E, Jiao H, Baumann W, Junge H, Gallou F, Beller M (2014) Angew Chem Int Ed 53:8722 25. Dub PA, Gordon JC (2017) ACS Catal 7:6635 26. Smith NE, Bernskoetter WH, Hazari N (2019) J Am Chem Soc 141:17350 27. Dub PA, Gordon JC (2018) Nat Rev Chem 2:396 28. Dub PA, Scott BL, Gordon JC (2017) J Am Chem Soc 139:1245 29. Werkmeister S, Neumann J, Junge K, Beller M (2015) Chem Eur J 21:12226 30. Wilson ME, Nuzzo RG, Whitesides GM (1978) J Am Chem Soc 100:2269 31. Hermosilla P, López P, García-Orduña P, Lahoz FJ, Polo V, Casado MA (2018) Organometallics 37:2618 32. Kuriyama W, Matsumoto T, Ino Y, Ogata O (2011) PCT Int Appl WO/2011/048727 A1 33. Danopoulos AA, Wills AR, Edwards PG (1990) Polyhedron 9:2413

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

317

34. Ma W, Cui S, Sun H, Tang W, Xue D, Li C, Fan J, Xiao J, Wang C (2018) Chem Eur J 14:13118 35. Abdur-Rashid K, Graham T, Tsang C-W, Chen X, Guo R, Jia W, Amoroso D, Sui-Seng C (2008) PCT Int Appl WO/2008/141439 A1 36. Meiners J, Friedrich A, Herdtweck E, Schneider S (2009) Organometallics 28:6331 37. Elangovan S, Garbe M, Jiao H, Spannenberg A, Junge K, Beller M (2016) Angew Chem Int Ed 55:15364 38. Burk MJ, Feaster JE, Harlow RL (1991) Tetrahedron Asymmetry 2:569 39. Huber R, Passera A, Gubler E, Mezzetti A (2018) Adv Synth Catal 360:2900 40. Schneider S, Meiners J, Askevold B (2012) Eur J Inorg Chem:412 41. Coles SJ, Edwards PG, Hursthouse MB, Read PW (1994) J Chem Soc Chem Commun:1967 42. Coles SJ, Danopoulos AA, Edwards PG, Hursthouse MB, Read PW (1995) J Chem Soc Dalton Trans:3401 43. Sherwood R, Gonzàlez de Rivera F, Wan JH, Zhang Q, White AJP, Rossell O, Hogarth G, Wilton-Ely JDET (2015) Inorg Chem 54:4222 44. García-Seijo MI, Habtemariam A, Parsons S, Gould RO, García-Fernández ME (2002) New J Chem 26:636 45. Fillman KL, Bielinski EA, Schmeier TJ, Nesvet JC, Woodruff TM, Pan CJ, Takase MK, Hazari N, Neidig ML (2014) Inorg Chem 53:6066 46. Ma W, Cui S, Sun H, Tang W, Xue D, Li C, Fan J, Xiao J, Wang C (2018) Chem Eur J 24:13118 47. Nguyen DH, Morin Y, Zhang L, Trivelli X, Capet F, Paul S, Desset S, Dumeignil F, Gauvin RM (2017) ChemCatChem 9:2652 48. Junge K, Wendt B, Cingolani A, Spannenberg A, Wei Z, Jiao H, Beller M (2018) Chem Eur J 24:1046 49. Rozenel SS, Kerr JB, Arnold J (2011) Dalton Trans 40:10397 50. Shao Z, Fu S, Wei M, Zhou S, Liu Q (2016) Angew Chem Int Ed 55:14653 51. Lagaditis PO, Schluschaß B, Demeshko S, Würtele C, Schneider S (2016) Inorg Chem 55:4529 52. Fu S, Chen N-Y, Liu X, Shao Z, Luo S-P, Liu Q (2016) J Am Chem Soc 138:8588 53. Zhou W, Wei Z, Spannenberg A, Jiao H, Junge K, Jung H, Beller M (2019) Chem Eur J 25:8459 54. Kuriyama W, Matsumoto T, Ogata O, Ino Y, Aoki K, Tanaka S, Ishida K, Kobayashi T, Sayo N, Saito T (2012) Org Process Res Dev 16:166 55. Han Z, Rong L, Wu J, Zhang L, Wang Z, Ding K (2012) Angew Chem Int Ed 51:13041 56. Zhang L, Raffa G, Nguyen DH, Swesi Y, Corbel-Demailly L, Capet F, Trivelli X, Desset S, Paul S, Paul J-F, Fongarland P, Dumeignil F, Gauvin RM (2016) J Catal 340:331 57. Bertoli M, Choualeb A, Lough AJ, Moore B, Spasyuk D, Gusev DG (2011) Organometallics 30:3479 58. Garbe M, Wei Z, Tannert B, Spannenberg A, Jiao H, Bachmann S, Scalone M, Junge K, Beller M (2019) Adv Synth Catal 361:1913 59. Touge T, Aoki K, Nara H, Kuriyama W (2012) PCT Int Appl WO/2012/144650 A1 60. Nielsen M, Alberico E, Baumann W, Drexler H-J, Junge H, Gladiali S, Beller M (2013) Nature 495:85 61. Choi J-H, Schloerer NE, Berger J, Prechtl MHG (2014) Dalton Trans 43:290 62. Choi J-H, Heim LE, Ahrens M, Prechtl MHG (2014) Dalton Trans 43:17248 63. Ogata O, Nakayama Y, Nara H, Fujiwhara M, Kayaki Y (2016) Org Lett 18:3894 64. Lazzari D, Cassani MC, Bertola M, Casado Moreno F, Torrente D (2013) RSC Adv 3:15582 65. Deutsch J, Jiménez Pinto J, Doerfelt S, Martin A, Köckritz A (2015) Flavour Fragr J 30:101 66. Fairweather NT, Gibson MS, Guan H (2015) Organometallics 34:335 67. Nguyen DH, Raffa G, Morin Y, Desset S, Capet F, Nardello-Rataj V, Dumeignil F, Gauvin RM (2017) Green Chem 19:5665 68. Ziebart C, Jackstell R, Beller M (2013) ChemCatChem 5:3228

318

D. A. Ekanayake and H. Guan

69. Satapathy A, Gadge ST, Bhanage BM (2017) ChemSusChem 10:1356 70. Gao S, Tang W, Zhang M, Wang C, Xiao J (2016) Synlett:1748 71. Padilla R, JØrgensen MSB, Paixão MW, Nielsen M (2019) Green Chem 21:5195 72. Pittaway R, Fuentes JA, Clarke ML (2017) Org Lett 19:2845 73. Chen J, Wang J, Tu T (2018) Chem Asian J 13:2559 74. Thiyagarajan S, Gunanathan C (2019) Org Lett 21:9774 75. Neumann J, Bornschein C, Jiao H, Junge K, Beller M (2015) Eur J Org Chem:5944 76. Choi J-H, Prechtl MHG (2015) ChemCatChem 7:1023 77. Wei Q, Zhang F, Zhao X, Wang C, Xiao J, Tang W (2017) Org Biomol Chem 15:5468 78. Wang W-H, Himeda Y, Muckerman JT, Manbeck GF, Fujita E (2015) Chem Rev 115:12936 79. Liu Q, Wu L, Gülak S, Rockstroh N, Jackstell R, Beller M (2014) Angew Chem Int Ed 53:7085 80. Treigerman Z, Sasson Y (2018) ACS Omega 3:12797 81. Kothandaraman J, Czaun M, Goeppert A, Haiges R, Jones J-P, May RB, Prakash GKS, Olah GA (2015) ChemSusChem 8:1442 82. Zhang L, Han Z, Zhao X, Wang Z, Ding K (2015) Angew Chem Int Ed 54:6186 83. Koo J, Kim SH, Hong SH (2018) Chem Commun 54:4995 84. Kar S, Goeppert A, Prakash GKS (2019) J Am Chem Soc 141:12518 85. Dong K, Elangovan S, Sang R, Spannenberg A, Jackstell R, Junge K, Li Y, Beller M (2016) Nat Commun 7:12075 86. Rezayee NM, Huff CA, Sanford MS (2015) J Am Chem Soc 137:1028 87. Kothandaraman J, Goeppert A, Czaun M, Olah GA, Prakash GKS (2016) J Am Chem Soc 138:778 88. Kothandaraman J, Goeppert A, Czaun M, Olah GA, Prakash GKS (2016) Green Chem 18:5831 89. Kar S, Sen R, Goeppert A, Prakash GKS (2018) J Am Chem Soc 140:1580 90. Yoshimura A, Watari R, Kuwata S, Kayaki Y (2019) Eur J Inorg Chem:2375 91. Kothandaraman J, Kar S, Sen R, Goeppert A, Olah GA, Prakash GKS (2017) J Am Chem Soc 139:2549 92. Kar S, Sen R, Kothandaraman J, Goeppert A, Chowdhury R, Munoz SB, Haiges R, Prakash GKS (2019) J Am Chem Soc 141:3160 93. Schneck F, Assmann M, Balmer M, Harms K, Langer R (2016) Organometallics 35:1931 94. Alberico E, Sponholz P, Cordes C, Nielsen M, Drexler H-J, Baumann W, Junge H, Beller M (2013) Angew Chem Int Ed 52:14162 95. Chakraborty S, Dai H, Bhattacharya P, Fairweather NT, Gibson MS, Krause JA, Guan H (2014) J Am Chem Soc 136:7869 96. Elangovan S, Wendt B, Topf C, Bachmann S, Scalone M, Spannenberg A, Jiao H, Baumann W, Junge K, Beller M (2016) Adv Synth Catal 358:820 97. Rezayee NM, Samblanet DC, Sanford MS (2016) ACS Catal 6:6377 98. Lange S, Elangovan S, Cordes C, Spannenberg A, Jiao H, Junge H, Bachmann S, Scalone M, Topf C, Junge K, Beller M (2016) Cat Sci Technol 6:4768 99. Koehne I, Schmeier TJ, Bielinski EA, Pan CJ, Lagaditis PO, Bernskoetter WH, Takase MK, Würtele C, Hazari N, Schneider S (2014) Inorg Chem 53:2133 100. Chakraborty S, Brennessel WW, Jones WD (2014) J Am Chem Soc 136:8564 101. Bielinski EA, Lagaditis PO, Zhang Y, Mercado BQ, Würtele C, Bernskoetter WH, Hazari N, Schneider S (2014) J Am Chem Soc 136:10234 102. Smith NE, Bernskoetter WH, Hazari N, Mercado BQ (2017) Organometallics 36:3995 103. Yi Y, Liu H, Xiao L-P, Wang B, Song G (2018) ChemSusChem 11:1474 104. Bonitatibus Jr PJ, Chakraborty S, Doherty MD, Siclovan O, Jones WD, Soloveichik GL (2015) Proc Natl Acad Sci U S A 112:1687 105. Chakraborty S, Lagaditis PO, Förster M, Bielinski EA, Hazari N, Holthausen MC, Jones WD, Schneider S (2014) ACS Catal 4:3994

Hydrogenation Reactions Catalyzed by PNP-Type Complexes Featuring a. . .

319

106. Lagaditis PO, Sues PE, Sonnenberg JF, Wan KY, Lough AJ, Morris RH (2014) J Am Chem Soc 136:1367 107. Sonnenberg JF, Lagaditis PO, Lough AJ, Morris RH (2014) Organometallics 33:6452 108. Smith SAM, Lagaditis PO, Lüpke A, Lough AJ, Morris RH (2017) Chem Eur J 23:7212 109. Seo CSG, Tannoux T, Smith SAM, Lough AJ, Morris RH (2019) J Org Chem 84:12040 110. Jayarathne U, Zhang Y, Hazari N, Bernskoetter WH (2017) Organometallics 36:409 111. Bornschein C, Werkmeister S, Wendt B, Jiao H, Alberico E, Baumann W, Junge H, Junge K, Beller M (2014) Nat Commun 5:4111 112. Sawatlon B, Surawatanawong P (2016) Dalton Trans 45:14965 113. Xu R, Chakraborty S, Bellows SM, Yuan H, Cundari TR, Jones WD (2016) ACS Catal 6:2127 114. Bernskoetter WH, Hazari N (2017) Acc Chem Res 50:1049 115. Zhang Y, MacIntosh AD, Wong JL, Bielinski EA, Williard PG, Mercado BQ, Hazari N, Bernskoetter WH (2015) Chem Sci 6:4291 116. Curley JB, Smith NE, Bernskoetter WH, Hazari N, Mercado BQ (2018) Organometallics 37:3846 117. Jayarathne U, Hazari N, Bernskoetter WH (2018) ACS Catal 8:1338 118. Suàrez LA, Culakova Z, Balcells D, Bernskoetter WH, Eisenstein O, Goldberg KI, Hazari N, Tilset M, Nova A (2018) ACS Catal 8:8751 119. Lane EM, Zhang Y, Hazari N, Bernskoetter WH (2019) Organometallics 38:3084 120. Jiao H, Junge K, Alberico E, Beller M (2016) J Comput Chem 37:168 121. Acosta-Ramirez A, Bertoli M, Gusev DG, Schlaf M (2012) Green Chem 14:1178 122. Taqui Khan MM, Khan BT, Begum S, Nazeeruddin K (1984) J Mol Catal 26:207 123. Taqui Khan MM, Khan BT, Begum S, Ali SM (1986) J Mol Catal 34:283 124. Taqui Khan MM, Khan BT, Begum S (1987) React Kinet Catal Lett 34:105 125. Taqui Khan MM, Khan BT, Begum S (1986) J Mol Catal 34:9 126. Taqui Khan MM, Rao ER, Siddiqui MRH, Khan BT, Begum S, Ali SM, Reddy J (1988) J Mol Catal 45:35 127. Ramaraj A, Nethaji M, Jagirdar BR (2019) J Organomet Chem 883:25 128. Rozenel SS, Padilla R, Camp C, Arnold J (2014) Chem Commun 50:2612 129. Zhang G, Vasudevan KV, Scott BL, Hanson SK (2013) J Am Chem Soc 135:8668 130. Mills MR, Barnes CL, Bernskoetter WH (2018) Inorg Chem 57:1590 131. Jing Y, Chen X, Yang X (2015) Organometallics 34:5716 132. Yuwen J, Chakraborty S, Brennessel WW, Jones WD (2017) ACS Catal 7:3735 133. Chowdhury A, Biswas S, Pramanik A, Sakar P (2019) Dalton Trans 48:16083 134. Dai H, Guan H (2018) ACS Catal 8:9125 135. Schneekönig J, Tannert B, Hornke H, Beller M, Junge K (2019) Cat Sci Technol 9:1779 136. Xu S, Chakraborty S, Yuan H, Jones WD (2015) ACS Catal 5:6350 137. Spentzos AZ, Barnes CL, Bernskoetter WH (2016) Inorg Chem 55:8225 138. Clarke ZE, Maragh PT, Dasgupta TP, Gusev DG, Lough AJ, Abdur-Rashid K (2006) Organometallics 25:4113 139. Chen X, Jia W, Guo R, Graham TW, Gullons MA, Abdur-Rashid K (2009) Dalton Trans:1407 140. Junge K, Wendt B, Jiao H, Beller M (2014) ChemCatChem 6:2810 141. Schmeier TJ, Dobereiner GE, Crabtree RH, Hazari N (2011) J Am Chem Soc 133:9274 142. Vasudevan KV, Scott BL, Hanson SK (2012) Eur J Inorg Chem:4898 143. Taqui Khan MM, Khan BT, Begum S (1988) J Mol Catal 45:305 144. McGuinness DS, Brown DB, Tooze RP, Hess FM, Dixon JT, Slawin AMZ (2006) Organometallics 25:3605 145. McGuinness DS, Wassercheid P, Keim W, Hu C, Englert U, Dixon JT, Grove C (2003) Chem Commun:334 146. Chakraborty S, Blacque O, Fox T, Berke H (2014) Chem Asian J 9:328 147. Chakraborty S, Berke H (2014) ACS Catal 4:2191 148. Wei Z, Junge K, Beller M, Jiao H (2017) Cat Sci Technol 7:2298 149. Chakraborty S, Blacque O, Berke H (2015) Dalton Trans 44:6560

320

D. A. Ekanayake and H. Guan

150. Leischner T, Spannenberg A, Junge K, Beller M (2018) Organometallics 37:4402 151. Leischner T, Suarez LA, Spannenberg A, Junge K, Nova A, Beller M (2019) Chem Sci 10:10566 152. Zell T, Langer R (2018) ChemCatChem 10:1930 153. Elangovan S, Topf C, Fischer S, Jiao H, Spannenberg A, Baumann W, Ludwig R, Junge K, Beller M (2016) J Am Chem Soc 138:8809 154. Fu S, Shao Z, Wang Y, Liu Q (2017) J Am Chem Soc 139:11941 155. Garbe M, Junge K, Walker S, Wei Z, Jiao H, Spannenberg A, Bachmann S, Scalone M, Beller M (2017) Angew Chem Int Ed 56:11237 156. Passera A, Mezzetti A (2019) Adv Synth Catal 361:4691 157. Jang YK, Krückel T, Rueping M, El-Sepelgy O (2018) Org Lett 20:7779 158. Kaithal A, Hölscher M, Leitner W (2018) Angew Chem Int Ed 57:13449 159. Wang Y, Zhu L, Shao Z, Li G, Lan Y, Liu Q (2019) J Am Chem Soc 141:17337 160. Papa V, Carbero-Antonino JR, Alberico E, Spannenberg A, Junge K, Junge H, Beller M (2017) Chem Sci 8:3576 161. Rawat KS, Pathak B (2017) Cat Sci Technol 7:3234 162. Kar S, Goeppert A, Kothandaraman J, Prakash GKS (2017) ACS Catal 7:6347 163. Ryabchuk P, Stier K, Junge K, Checinski MP, Beller M (2019) J Am Chem Soc 141:16923 164. Wei D, Roisnel T, Darcel C, Clot E, Sortais J-B (2017) ChemCatChem 9:80 165. Wei Z, Junge K, Beller M, Jiao H (2018) CR Chim 21:303 166. Yang X (2015) Chem Commun 51:13098 167. Ge H, Jing Y, Yang X (2016) Inorg Chem 55:12179

Top Organomet Chem (2021) 68: 321–378 https://doi.org/10.1007/3418_2020_62 # Springer Nature Switzerland AG 2020 Published online: 1 December 2020

Catalytic Conversion of Nitriles by Metal Pincer Complexes Beibei Guo, Edwin Otten, and Johannes G. de Vries

Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Hydrogenation of Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Hydrogenation of Nitriles to Amines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Hydrogenation of Nitriles to Imines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 α-Functionalisation of Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 α-Alkylation with Alcohols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 α-Olefination of Benzyl Cyanide and Aliphatic Nitriles with Alcohols . . . . . . . . . . . . . 3.3 α-Alkylation of Aliphatic and Benzylic Nitriles via Michael Addition on Unsaturated Ketones or Esters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 α-Deuteration of Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 α-Acylation of Unsaturated Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Aldol and Mannich Reactions with Aldehydes or Protected Imines . . . . . . . . . . . . . . . . 3.7 Enantioselective α-Functionalisation of Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Hetero-Michael Addition to α,β-Unsaturated Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Lewis Acid Catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Catalysis via Metal-Ligand Cooperation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Hydration of Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Amination of Nitriles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Synthesis of Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

322 324 324 333 339 339 342 344 347 350 352 354 358 358 362 365 369 370 372 372

Abstract The nitrile is an extremely useful functional group in organic synthesis: it can be transformed into amides, carboxylic acids, amines and imines; yet it is relatively stable and can be easily carried through several synthetic steps before

B. Guo and E. Otten Stratingh Institute for Chemistry, University of Groningen, Groningen, The Netherlands e-mail: [email protected] J. G. de Vries (*) Leibniz Institut für Katalyse e.V., Rostock, Germany e-mail: [email protected]

322

B. Guo et al.

being converted. The conversions of nitriles under mild conditions are thus very important transformations. Great progress has been made in the last decade in the use of metal pincer complexes as catalysts for quite a number of reactions of nitriles and nitrile-containing molecules. The selective hydrogenation of nitriles either to the amines or to the imines usually follows a Noyori-type outer-sphere mechanism. Coordination of aliphatic nitriles to the metal centre renders the α-proton rather acidic allowing deprotonation followed by carbon-carbon coupling reactions. The pyridine-based metal pincer complexes introduced by Milstein allow for novel mechanisms based on metal-ligand cooperativity in which the pyridine undergoes dearomatisation induced by deprotonation of one of the side arms. The nitrile can undergo a cycloaddition to the complex in its dearomatised form, creating a new bond between the nitrogen atom and the metal, whereas the nitrile carbon atom forms a C-C bond with the carbon atom of one of the pincer side-arms. The resulting metalimide undergoes nucleophilic addition more easily than the nitrile. It can also easily rearrange to the enamide, which can undergo C-C bond forming reactions. Also, oxo- and aza-Michael reactions are facilitated on the unsaturated nitriles, such as acrylonitriles or pentenitriles. Most reactions proceed under mild conditions in excellent yields. Keywords Acylation · Aldol · Alkylation · Amide · Amidine · Amine · Deuteration · Hydration · Imine · Mannich · Mechanism · Michael addition · Nitrile · Olefination · Quinazoline · Quinoline

1 Introduction The nitrile functional group is very important from an organic synthetic perspective. It can be transformed into amides, carboxylic acids, amines and imines; yet it is relatively stable and can be easily carried through several synthetic steps before being converted. Thus, activation of the C
N bond under mild conditions can be seen as a very important research area. In this review, we will discuss the recent progress in reactions of nitriles catalysed by transition metal pincer complexes. In 1976, Moulton and Shaw were the first to report organometallic complexes based on pincer ligands [1]. Since then, pincer complexes have received increasing attention due to their unique stability, selectivity and often unusual reactivity [2– 12]. Taking advantage of these properties, efficient methodologies have been developed for the activation of nitriles that are often tolerant of many other functional groups. These pincer complexes may react in the same way as traditional metal complexes [13, 14]: the nitrogen atom of the nitrile may coordinate to the metal centre, which leads to increased acidity of the α-proton enabling the use of the resulting anion as a carbon nucleophile (Scheme 1, left side). In addition, they are

Catalytic Conversion of Nitriles by Metal Pincer Complexes

Traditional transition metal complexes R1

N

R2 R1 R2

323

Metal-ligand cooperation (MLC) PR2

M + base

X N

M

R1 R2

M

HH R

C N

R4

electrophile

R3 R4

(1,2- or 1,4-addition)

L L

X N M

PR2

M

N

Nu N M

L L

RCH2CN

• N

C-nucleophile

R3

M

N

R H

C N H PR2 M

N X

L L

Scheme 1 Activation of nitriles by conventional metal complexes (Lewis acids) and metal-ligand cooperative pincer complexes (X ¼ NR’2 or PR’2)

excellent hydrogenation and dehydrogenation catalysts that, depending on the ligand structure, may operate according to the classic Noyori-type (transfer) hydrogenation mechanism. However, in the PNP and PNN metal complexes based on pyridines, Milstein and co-workers also found novel mechanisms based on metal-ligand cooperativity [15] in which the pyridine undergoes dearomatisation induced by deprotonation of one of the side arms [16]. These mechanisms are also operative on nitriles, which were found to undergo cycloaddition onto these dearomatised complexes, forming a new bond between the nitrogen atom and the metal, whereas the carbon atom forms a C-C bond with the carbon atom of one of the side arms [17, 18]. The initially formed imide form can further tautomerise to the enamide form (Scheme 1, right side). The relative stabilities of these forms depend on the substituent R [18]. The addition as well as the tautomerisation is reversible. This change in bond order of the nitrile significantly reduces the activation barrier for its reaction with nucleophiles and other reactions. In addition, the metalated enamine form is a good nucleophile allowing substitution reactions and aldol condensations.

324

B. Guo et al.

2 Hydrogenation of Nitriles 2.1

Hydrogenation of Nitriles to Amines

A range of different pincer complexes based on ruthenium, iron, cobalt and manganese have been used for the hydrogenation of nitriles to the primary amines (Scheme 2). cat

R C N

tBu P 2

HH

Ru P H Ph2

H

1 Leitner, 2011

Fe P H iPr 2

Fe P H Cy2

5a Beller, 2014

tBu P 2

N Co

H NHtBu

P H Et2

CO

P iPr CO 2

Cl

H

N

Co

8

Cl Cl

N

Liu, 2016

Cl N

9

N Mes

10 Fout, 2017

PR2

N

Co P R2

X

X

H N

N

Co X N X

Beller, 2019

11 R = iPr/Cy/Ph

PiPr2 CO

6a Beller, 2016

PtBu2

Cl

H Co N Mes Cl Py

Br

N

7 Milstein, 2015

N

N

Mn

5c Beller, 2016

N

Cl Cl

H

PEt2 Fe

PtBu2 Co

4 Milstein, 2017

HBH3 N

CO

PtBu2

H CO

3 Beller, 2016

5b Beller, 2016

N

N

H

N Ru

tBu P 2

N

PCy2

N

CO

H

CO

HBH3

H

PiPr2

N

PtBu2 HBH3 Ru CO

N

H

2 Beller, 2015

HBH3

H

PPh2

N

H Ru PtBu2

NH2

HBH3

H N

R

H2

12 X=Br/Cl

Scheme 2 Pincer complexes used as catalyst for the hydrogenation of nitriles to amines

Catalytic Conversion of Nitriles by Metal Pincer Complexes

325

Table 1 Hydrogenation of nitriles to amines by Ru pincer complex 1 1 mol% 1 5 mol% water

N R

N R

o

135 C, 75 bar H2

NH2 t Bu P 2

HH

toluene, 24 h

NH2

NH2

59 (65)%

62%

93 (75)% a

1

96 (88)%

NH2

NH2

80 (96)% NH2

H

NH2 O

95 (36)%

NH2

O

P t Bu2

O

Cl a

H

Ru

NH2 92 (82) %

95 (90)% N

NH2 90 (74) %

Numbers in bracket are yields without water additive

Leitner’s group reported the first pincer complex-catalysed hydrogenation of nitriles to amines [19]. With 0.4 mol% of the non-classical ruthenium hydride complex 1 as catalyst, they successfully reduced eight different nitriles including aliphatic and aromatic nitriles at 135 C (Table 1). In order to reach high selectivity and full conversion, a high H2 pressure (75 bar) and long reaction times (45 h) were required. Interestingly, they found that addition of 2 mol% of water increased the yields (especially for p-chlorobenzonitrile, from 35% to 95%) and significantly shortened the reaction time (from 45 h to 24 h). They proposed water played a key role in the prevention of secondary amine formation, a well-known side reaction in nitrile hydrogenation, via hydrolysis of the secondary imines: the water hydrolyses the secondary imine forming the primary amine and the aldehyde, which can react with the ammonia, released in the secondary imine-forming step (Scheme 3). Although this methodology required relatively harsh condition (135 C and 75 bar H2), it was the first highly selective and catalytic reduction of nitriles to amines by a pincer complex. The Beller group used the well-known ruthenium-MACHO-BH complex 2, which was developed earlier by Takasago chemists [20, 21], for the hydrogenation of nitriles under milder conditions (Scheme 4) [22]. Using 1 mol% of catalyst and 30 bar hydrogen at 100 C, aliphatic nitriles were hydrogenated with isolated yields

326

B. Guo et al. R C N Ru cat H2 R R

1 NH

NH2 N H

NH2

-RCHO H2O

-H2O

NH2

R

R

H2

-NH3 R

R

RCHO -H2O N

R

Scheme 3 Role of water in the Ru pincer complex 1 catalysed hydrogenation of nitriles HBH3

H R

C

N

1 mol% 4 100 o C, 30 bar H2

PPh2

N R

Ru

NH2

P H Ph2

i

PrOH, 3 h

CO 2

Scheme 4 Hydrogenation of nitriles using Ru-MACHO-BH P t Bu 2

R

C

N

0.5-1 mol% 3 o

50-150 C, 30 bar H2 i

R

NH 2

PrOH, 3 h

H

N

Ru H N

HBH 3 CO

N 3

Scheme 5 Hydrogenation of nitriles using Ru PNN complex 3

from 77 to 98%, whereas aromatic nitriles could be reduced with isolated yields between 68 and 99%. Interestingly, the usually problematic ester-substituted benzonitrile was selectively reduced (94% isolated yield) leaving the ester group intact. Ru NNP pincer complexes bearing an imidazole moiety were also developed by the same group [23]. Different substituents on the phosphorus atom were investigated, but complex 3 carrying tert-butyl substituents gave the best results. A large range of aromatic, heteroaromatic and aliphatic nitriles was hydrogenated at 30 bar hydrogen (Scheme 5). For most substrates, a temperature of 50 C was sufficient, but for more sluggish substrates, temperatures between 70 and 150 C were used. In all cases the nitrile was completely converted, and the primary amines were obtained in GC yields of 75–99%.

Catalytic Conversion of Nitriles by Metal Pincer Complexes

327

Milstein and co-workers developed the low-pressure hydrogenation of nitriles to primary amines catalysed by pyridine-based pincer Ru complex 4 in which the pyridine is dearomatised [24]. With lower catalyst loading (0.3 mol%), lower temperature (110 C) and no extra additives, aromatic nitriles could be reduced with high selectivity and conversions (Scheme 6). The authors assume a ligandassisted mechanism (Scheme 7), in which the catalyst 4 rapidly reacts with the nitrile to form nitrile complex I, which is hydrogenated to the dihydride complex II in which the aromaticity is restored and the side arm protonated. Next, in an outer-

0.3 mol% 4

Ar C N

NH2

Ar

o

100 C, 5 bar, 24-60 h, Benzene

t Bu

N Ru

2P

PtBu2

H CO 4

Scheme 6 Hydrogenation of nitriles to amines by Ru pincer complex 9

RCN

N t

Bu 2 P

Ru H

N

P t Bu 2

CO

t

Bu 2P

Ru H

4

N

CO

C

R

P t Bu 2 I

H2 -RCN N C R H

RCN

H t

N

Bu2 P

Ru H

CO

H

H H

P t Bu2

t

N

Bu2 P

Ru H

III

CO

H

P t Bu2 II

R

N t

Bu2 P

Ru H

CO

NH t

P Bu2

R

H C

NH H2

RCH2NH2

IV

Scheme 7 Proposed mechanism of nitrile hydrogenation using a dearomatised Ru PNP pincer complex

328

B. Guo et al.

sphere mechanism (intermediate III), the hydride is transferred to the carbon atom under simultaneous protonation of the nitrogen atom by the proton on the side arm under dearomatisation of the pyridine resulting in the imine complex IV, which releases the imine which is further hydrogenated to the amine by a similar mechanism. Beller and co-workers reported the first iron-based pincer complex-catalysed hydrogenation of nitriles to the amines [25]. The catalyst 5a they had developed before [26] showed high reactivity in the hydrogenation reaction and good selectivity (up to >99%) for the primary amines with isopropanol as solvent under 30 bar H2 atmosphere at 70–130 C. As to the substrate scope investigation, this reaction showed a very broad applicability: (hetero)aromatic nitriles, aliphatic nitriles and dinitriles were all efficiently converted (Table 2). Catalytic hydrogenation of benzonitriles with electron-donating substituents gave the amines good to excellent Table 2 Nitrile hydrogenation with iron-based pincer complex 5a R C N

1 mol% 5a 70 or 100 oC, 30 bar H2

NH3Cl

R

i

PrOH, 3h followed by acidification with 1M HCl MeOH NH3Cl

N

NH3Cl

NH3Cl

R' R' = H R' = Me R' = tBu R' = Ph R' = SMe R' = OMe R' = CF3 R' = F R' = Cl R' = Br R' = NH2

97% 96% 82% 99% 81% 88% 71% 81% 78% 88% 93%

99%

80%

NH3Cl

NH3Cl N H

S 58% NH3Cl MeOOC

40% NH3Cl

O N H

75%

70% NH3Cl

NH3Cl

NH3Cl

69% C16H33

63% NH3Cl

NH3Cl

63%

75%

88% ClH3N

NH3Cl 95%

Catalytic Conversion of Nitriles by Metal Pincer Complexes

329

yields (81–99%), whereas benzonitriles with electron-withdrawing substituents as well as heteroaromatic nitriles required slightly higher temperatures (100–130 C). Remarkably, not only common halogens (CF3, F, Cl, Br), amino and methoxy functional groups were tolerated as aromatic substituents (71–93% yields), but ester- and acetamide-substituted nitriles were also reduced in synthetically useful yields (75% and 70%). Primary, secondary as well as tertiary aliphatic nitriles were all converted with good yields (63–95%). Noteworthily, cinnamonitrile could be selectivity hydrogenated without reducing the C¼C bond (allylamine: saturated amine >25:1). Adiponitrile was reduced in excellent yield to hexamethylenediamine, a monomer for nylon-6,6, with good selectivity (95% isolated yield) and high rate (TOF of 250 h1). In addition, this reaction was scaled up to 25 mmol. Based on DFT studies, it was proposed that dissociation of BH3 in the form of B2H6 from 5a leads to the formation of the dihydride complex I (Scheme 5), which is the active catalyst. The calculations also allowed the authors to distinguish between the two possible mechanisms: in an inner-sphere mechanism, the CO needs to dissociate first in order to allow coordination of the nitrile. This is endergonic by 23.78 kcal mol1. In the outer-sphere mechanism, the iron-bound hydride and proton from the amine are transferred simultaneously, and the activation barrier for this is 15.35 kcal mol1. Based on this, the outer-sphere mechanism (Scheme 8) is clearly preferred. HBH3

H

PiPr2

N Fe

CO P H iPr 2 5a

-1/

2B

2H 6

H N

H Fe

R

N

P H iPr 2

PiPr2 CO

H R

NH

I

R

PiPr2

N

H NH

Fe P H iPr 2 II

CO

Scheme 8 Outer-sphere mechanism for nitrile hydrogenation with 5a

H R

H NH2

330

B. Guo et al.

To further improve the performance of iron pincer complex 5a, the Beller group studied the effect of the alkyl substituents at the phosphorus atoms. Thus, a new iron pincer complex was prepared with cyclohexyl substituents at the phosphorus atoms, and the performance of this complex was compared to the complexes 5a (R ¼ iPr) and 5c (R ¼ Et) in the hydrogenation of benzonitrile (Table 3) [27]. Although at 1 mol% catalyst loading the performance of the three catalysts was very similar, at catalyst loadings of 0.5 and 0.25 mol%, complex 5c was totally inactive, whereas 5a and 5b were able to convert the nitrile to the amine in good yields. Using the same ligand, the Beller group also prepared a manganese complex 6a which contains two additional CO ligands as well as bromide [28]. The presence of the halide necessitates the use of base (NaOtBu) to activate the catalyst; in this step the proton from the amine group is removed, together with the bromide. This activated catalyst is well set up to add hydrogen to form the hydride complex. This catalyst requires higher hydrogen pressure (50 bar) and catalyst loading (3 mol%) than the iron complexes, and yet reaction times of 24–60 h were still required. A range of aromatic and heteroaromatic nitriles were hydrogenated at 120 C during 24 h resulting in good yields (47–99%) of the benzylamines (Scheme 9). Aliphatic nitriles on average reacted more sluggishly and took 24–60 h. The aliphatic amines were obtained in isolated yields of 78–97% as the HCl salts. Cinnamonitrile was reduced to 3-phenyl-prop-2-enyl-1-amine in 54% isolated yield.

Table 3 Effect of the alkyl substituent of 5 on the hydrogenation of benzonitrile

C

N

cat

N

NH2

Entry 1 2 3

Catalyst loading (mol%) 1.0 0.5 0.25

C

N

Yield of benzylamine 5a 5b 92% 89% 90% 90% 88% 87%

H

3 mol% 6a t

10 mol% NaO Bu 120 oC, 50 bar H2 toluene, 24 h

PR2 Fe

CO P H R2 5a R = iPr 5b R = Cy 5c R = Et

70 oC, 30 bar, 3 h iPrOH

Ar

HBH3

H

N Ar

NH 2

5c 85% 0% 0%

Br

Mn

P iPr 2 CO

P CO i Pr2 6a

Scheme 9 Use of a manganese PNP pincer complex for the hydrogenation of nitriles

Catalytic Conversion of Nitriles by Metal Pincer Complexes

331

Table 4 Hydrogenation of nitriles to amines by cobalt pincer complex 7a

R C N

2 mol% 7 2mol% NaHBEt3 4.4 mol% NaOEt 135 o C, 30 bar, 36-60 h, Benzene

R

NH2

Bu2 P

N Co

NHtBu

Cl Cl

NH2

NH2

7

NH2 N

R' R' = H R' = Me R' = OMe R' = CF 3 R' = F R' = Cl R' = Br R' = NH2

t

85% 99% 99% 57% 83% 93% 6% 86%

98%

71% NH2

NH2 H 2N

85%

85%

NH2 67%

NH2 90%

a

Yields determined by 1H NMR spectroscopy with respect to toluene or dimethylformamide as an internal standard or by GC analysis

Milstein’s group reported the first cobalt pincer complex (7)-catalysed hydrogenation of nitriles to amines [29]. After activation of the catalyst by base (NaOEt) and a reducing agent (NaEt3BH, presumably to reduce Co(II) to the active Co(I)), aromatic nitriles could be hydrogenated to the primary amines in benzene at 135 C and under 30 bar H2, in up to >99% yield. A broad functional group tolerance including electron-donating groups and electron-withdrawing groups was found, with bromide being the exception (Table 4). In the case of pyridine-based substrates, the catalyst became sluggish (77% conversion, 71% yield after 36 h). Yields of aliphatic amines were somewhat lower, due to base-induced side reactions. In 2017, Fout and co-workers reported the hydrogenation of nitriles to primary amines by a bis(carbene) pincer cobalt complex 10 with broad scope and excellent yields (Scheme 10) [30]. Interestingly, they found that when 10 is reacted with NaHBEt3, not only does reduction occur to Co (I) complex 100 , but the Et3B that is formed in this reduction turned out to be an indispensable component in the nitrile reduction process. The authors presume that through coordination of Et3B with the nitrile, a side-on coordination of the nitrile to cobalt becomes possible, which effectively facilitates the insertion of the triple bond into the cobalt-hydride bond. The authors used para-hydrogen-induced polarisation transfer NMR studies to gain further mechanistic information. The intermediacy of the dihydride was proven, and the imine intermediate could be observed by NMR spectroscopy. By using 1.04 eq.

332

B. Guo et al. 2 mol% X 6 mol% NaOtBu 4 mol% additive

N R

115 oC,4 bar H2 toluene, 8h

R

NH2 22 samples up to >99% N

N H Co

N Mes Cl N

N Co N Mes Cl Py Additive

N

N

N Mes

Co

N Mes

Py

10

NaHBEt3

H R

N

NLA Mes

LA=lewis acid

N Mes 10'

BEt3

Scheme 10 Use of a CCC-pincer ligand in the Co-catalysed hydrogenation of nitriles to amines

C R

N

4 mol% 11a 16 mol% NaBHEt3 80 oC, 50 bar H2 dioxane, 6 h

H R

NH2

PPh2

N

Co

Cl

P Cl Ph2 11a

Scheme 11 Hydrogenation of nitriles to amines catalysed by Co(MACHO)Cl2 (11a)

of Et3B, it was possible to selectively reduce 4-acetyl-benzonitrile to 4-acetylbenzylamine. Using the MACHO ligand as well as analogues with different substituents on phosphorus (iPr, Cy), four pincer complexes were made based on cobalt by the Beller group [31]. These cobalt complexes contain halogen ligands; this necessitates the use of base and/or a reductant to activate the catalyst. Compared with iron analogue 5a, these complexes require higher H2 pressure (50 bar) and a higher catalyst loading (4 mol%) and still need a longer reaction time (6 h) to reach full conversion. The best catalyst was the one based on the MACHO ligand (11a). A range of aromatic and aliphatic nitriles were hydrogenated in good yields (Scheme 11). Thus far, there is only a single report on the transfer hydrogenation of nitriles catalysed by pincer complexes. Zhou, Liu and co-workers reported the Co-NNP pincer complex-catalysed transfer hydrogenation of nitriles to primary, secondary or tertiary amines using NH3BH3 as hydrogen source (Scheme 12) [32]. At 50 C in hexane, complex 8 bearing an imidazole moiety selectively converted aromatic and aliphatic nitriles to primary amines in good to excellent yields (26 examples). Transfer hydrogenation of aromatic nitriles with pyridine-based PNN cobalt complex 9 in hexafluoroisopropanol (HFIP) at room temperature surprisingly led to the formation of the symmetric secondary amines in good to excellent yields. A few aliphatic nitriles also gave the secondary amines, but in mediocre yields. If the same hydrogenation was performed in the presence of primary or secondary amines, the

Catalytic Conversion of Nitriles by Metal Pincer Complexes

333 1 mol% 8

Scheme 12 Transfer hydrogenation of nitriles catalysed by cobalt complexes 8 and 9

50oC, hexane NH 3-BH 3

R C N

R

NH2

R

N H

R

N

0.5 mol% 9 HFIP, rt

R

0.5 mol% 9 R 1 NHR 2 HFIP, rt

R1

H

H N

N

N

R2

N P t Bu2

Co Cl

Cl 8

N

P t Bu2

Co Cl

Cl 9

nonsymmetrical secondary and tertiary amines were formed, respectively. The analogous catalysts in which the ligand was N-methylated gave very similar results in the hydrogenation of benzonitrile as the non-methylated catalysts. Based on this observation, the authors assumed that these reactions proceed via an inner-sphere mechanism.

2.2

Hydrogenation of Nitriles to Imines

We have already seen the formation of secondary imines in the previous section, as an unwanted side reaction, which needed to be repressed, here we discuss the on-purpose hydrogenation of nitriles to the imines. A range of metal pincer complexes has been used for the hydrogenation of nitriles to the imines (Scheme 13). The first selective hydrogenation of nitriles to secondary imines using a pincer complex as catalyst was reported by the Milstein group [33]. In the absence of amines, benzonitriles with electron-donating substituents (H, methyl, methoxy) could be reduced to the corresponding symmetrical secondary imines at 70 C under 4 bar H2 using the bipyridine-based PNN Ru pincer complex 13 in excellent yields (Table 5). Using the same conditions, addition of (cyclo-)hexylamine led to hydrogenative cross-coupling forming unsymmetrical secondary imines. This reaction was shown to have a broad substrate scope with (hetero)aromatic and aliphatic nitriles forming the unsymmetrical imines in good to excellent yields. Electronwithdrawing substituents are not well-tolerated as para-fluoro-benzonitrile formed the imine in only 50% yield.

334

B. Guo et al.

R

cat

R C N

R'NH 2

H2

N N N

Ru

Pt Bu 2

R

13 Milstein, 2012

N

Pt Bu 2

H

X P X i Pr 2 M = Mo, W 15 Berke, 2014 H

Cl

Fe CO H

Pi Pr 2

Pi Pr2

N HN t

Bu 2 P

N

NH

Co

t

Co

P Bu 2

P Br i Pr2

Cl 16 Milstein, 2017

Pi Pr2 M

14 Prechtl, 2015

Br

R'

N

Ru CO

Pi Pr2 N

R

P t Bu 2

H CO

H

N

17 Huang, 2018

Br

11b Guan, 2018

Scheme 13 Hydrogenation of nitriles to imines by pincer complexes

In view of all the possible equilibria between the imines and amines during the hydrogenation (see also Scheme 3), it can be rather challenging to obtain high selectivity to a single product. Interestingly, Choi and Prechtl managed to selectively synthesise either imines or amines by simply tuning solvent and temperature in the hydrogenation of nitriles with complex 14 [34]. At 50 C in toluene under 4 bar H2, six nitriles could be reduced to the imines with or without addition of amines. However, at 90 C in isopropanol under 4 bar H2, it was possible to selectively hydrogenate the nitriles to the amines, and five primary amines were obtained in moderate to excellent yields (Scheme 14). Berke and Chakraborty used non-noble metal pincer complexes based on molybdenum and tungsten as catalyst for the hydrogenation of nitriles to secondary imines [35]. Using the molybdenum PNPiPr complex 15a, ten substrates, including aromatic, heteroaromatic and aliphatic nitriles, were reduced with up to >99% conversion and >96% selectivity (Table 6). Both electron-donating substituents (methyl and methoxy) and electron-withdrawing substituents (CF3, F, Br, Cl) are tolerated. The analogous tungsten complex 15b was much slower, although in some cases more selective towards formation of the desired secondary imine. Chakraborty, Milstein and co-workers synthesised iron PNP pincer complex 16 in 47% yield in three steps [36]; this catalyst needs to be activated by treatment with an equivalent of tBuOK. It was used for the hydrogenative cross-coupling of

Catalytic Conversion of Nitriles by Metal Pincer Complexes

335

Table 5 Hydrogenation of nitriles to imines by 13a 0.6 mol% 13 4 bar H2

R C N + R'NH2

o THF, 70 C -NH3

R

N

R'

N

N R1

N Ru

PtBu2

H CO

13

N R1

b 1= R

H R1 = Me R1 = OMe

R1

88% 81% 92%

R1=H R1=Me R1=OMe R1=F

N

85% 84% 90% 50%

N

R1

N

R1 = H R1 = OMe

81% c

91% 85% N

N 80%

74%

a Yields of the products and conversion of nitriles were determined by gas chromatography (GC) using toluene as an internal standard b no amines added; c0.8 mol% catalyst was used

R R

or N

N R'

R

Optionally:RNH2 1 mol% 14 o

90 C, 4 bar H2 toluene

N R

0.5 mol% 14 50 oC, 4 bar H2 iPrOH

N R

NH2

H

PtBu2

Ru CO

Pt Bu2 14

Scheme 14 Tunable hydrogenation of nitriles to imines or amines by 14

(hetero)aromatic nitriles with amines and anilines at 60 C and 20 bar H2 with benzene as the solvent, which resulted in good to excellent yields of the imines (Table 7). The more challenging aliphatic nitriles were still sluggishly reduced (5–58% yields) even at higher temperatures and/or higher catalyst/base loading. This catalyst was also used for the hydrogenation of nitriles to symmetrical secondary imines at higher temperature and pressure (90 C and 30 bar) (Scheme 15) [37]. With low catalyst loadings (1–2 mol%), benzonitriles with electron-donating (Me, MeO, NMe2) substituents and p-fluorobenzonitrile were hydrogenated to the imines in moderate to excellent yields. Heteroaromatic nitriles required 2 mol%

336

B. Guo et al.

Table 6 Hydrogenation of nitriles by 15a,ba

R C N

5 mol% 15a or 15b o

THF, 140 C, 60 bar H2 PiPr2

N Mo

P CO iPr 2 15a

a b

NO

NH2 + R

NH + R

R a

b PiPr2

N W

P CO iPr 2

NO

15b

N c

R

Catalytic Conversion of Nitriles by Metal Pincer Complexes

337

Table 7 Hydrogenation of nitriles to imines by 16a

R C N + R' NH2

N

R1

1 mol% 16 1mol% tBuOK o

C6H6, 60 C, 10-20 bar H2 -NH3

R

N

R'

PiPr2 Br Fe CO

H N H

PiPr2 16

C5H11

R1 = H R1 = o-Me R1 = p-OMe R1 = p-Cl R1 = p-Br N

N

N

96% 98% 96% 97% 66%

96%

N

N

N

64%

C5H11

N

10%b

88%

F

C5H11

89%

Ph

37%b

C5H11

N

C5H11

58%b

a

Yields and conversions determined by GC-MS and NMR analysis using m-xylene or b toluene as internal standards 8 mol% Catalyst used

R C

N

1-8 mol% cat and t BuOK o

C6 H6 , 90 C, 30 bar H2

H R

N

R

-NH3

Pi Pr 2 Br N

Fe CO H

Pi Pr 2 16

Scheme 15 Hydrogenation of nitriles to symmetrical imines by 16

catalyst loading to reach high conversions and yields. For the more challenging aliphatic nitriles, a decrease in conversion and yields (11–69%) was observed, even when using 8 mol% catalyst loading. The Milstein group proposed the same mechanism for this reaction as the one they had published earlier (Scheme 16) [36]. The bromide complex is deprotonated by base forming the amido complex I, which reacts with hydrogen to the cisdihydride complex II, which they assume will isomerise to the (unobserved) trans-complex III. The trans-dihydride complex, which they assume to be the active species, reduces the nitriles (or primary imines) to primary imines (or amines) regenerating the amido complex via hydride and proton transfer. Then the

338

B. Guo et al. H R

H N H

PiPr2 Br Fe CO

H

H

NH2

R

PiPr2 base

N

Fe

H2

CO H

PiPr2

PiPr

16

I

H

N

2

H

NH

PiPr2 CO Fe H PiPr

H

2

N H

II

PiPr2 H Fe CO PiPr2 III

H R'-NH2= added amines or generated in situ

R

N

R

NH

R' NH2 NH2 R

N H

H

R'

R

N

R'

NH3

Scheme 16 Hydrogenation of nitriles to imines by 16

R C N

1 mol% 17 4 mol% tBuOK toluene, 100oC, 62 bar H2 -NH3

Cl R

N

R

HN 2P

t Bu

N NH Co PtBu2 Cl 17

Scheme 17 Hydrogenation of nitriles to imines by 17

intermediary primary imines may react with the added or produced amines giving a gem-diamine. After liberation of ammonia, the desired imine is produced. Huang and co-workers reported the synthesis of Co-PN3P pincer complexes and their application in the hydrogenation of nitriles to secondary imines [38]. At 62 bar of H2, 100 C and using toluene as solvent, high selectivity and excellent yields of the imines could be achieved with only 1 mol% of catalyst 17 and 4 mol% base (Scheme 17). Under the same catalytic conditions, more challenging substrates (heterocyclic and aliphatic nitriles) were also converted to the corresponding imines with moderate to good yields.

Catalytic Conversion of Nitriles by Metal Pincer Complexes

Ph

C N

CyNH2 + Ph

2 mol% 11b 6 mol% NaHBEt 3 THF, 110oC, 20 bar H2

Ph

2 mol% 11b 4 mol% NaHBEt 3

C N

o

THF, 100 C, 20 bar H2

339

Ph

N 91%

H

Pi Pr2

N

Co Ph

N

Cy

91%

Br

P Br i Pr2

11b

Scheme 18 Hydrogenation of benzonitrile to imines by 11b

Guan and his co-worker used the Co-PNHP complex 11b to catalyse the selective hydrogenation of benzonitrile to secondary imines with or without an added amine (Scheme 18) [39]. Both reactions proceeded in excellent yields.

3 α-Functionalisation of Nitriles 3.1

α-Alkylation with Alcohols

The first ruthenium-catalysed α-alkylation of nitriles by alcohols was reported by Grigg almost 40 years ago [40]. This is a borrowing hydrogen-type reaction, in which the alcohol is dehydrogenated to the aldehyde, which undergoes a Knoevenagel reaction with the nitrile. After dehydration, the formed double bond is reduced with the hydrogen equivalents that were obtained in the first dehydrogenation step, leading to a net alkylation. Many transition metal complexes, mostly based on noble metals, were developed in this field. Esteruelas and co-workers successfully synthesised a series of Ru and Os pincer complexes based on xantphos [41]. They reported the alkylation of phenylacetonitrile with benzyl alcohol and 1-octanol, catalysed by Ru pincer complex 18 with 20 mol% base at 110 C using toluene as solvent (Scheme 19). The alkylation products were obtained in 79% yield with TOF50% ¼ 18 h1 (benzyl alcohol) and 68% yield with TOF50% ¼ 1.4 h1 (1-octanol).

CN + R

OH

1 mol% 18 20 mol% KOH toluene, 110oC

PiPr2

R CN + H O 2

R = Ph 79%; TOF50% = 18 h-1 R = (CH2)6CH3 68%; TOF50% = 1.4 h-1

Scheme 19 Alkylation of benzyl nitrile with alcohols by 18

H

O Ru H

H

PiPr2 18

B

H H

340

B. Guo et al.

Table 8 Alkylation of nitriles with alcohols by 19a

R1

CN + R2

OH

H

0.5 mol% 19 1 mol% KOtBu toluene 135oC, 4 h

CN

CN R1

H

N R 2 + H2O

CN

PPh2

Ru

CO P Cl Ph2 19 CN N

91%

95%

CN

61%

CN

CN

O O

93%

97%

CN

CN N

90%

Br

HN

96%

CN

CN

83%b

96% CN

CN

O

O

O

O

66%b a

61%c

83%c b

Isolated yield of products after column chromatography Catalyst (2 mol %), t

c

t

and KO Bu (4 mol %) Catalyst (2.5 mol %), and KO Bu (5 mol %), 40 h

Gunanathan investigated the same reaction using the ruthenium-MACHO catalyst 19 [42]. The reaction was performed at an oil-bath temperature of 135 C in toluene with 0.5–2.5 mol% catalyst and 2 eq. tBuOK (relative to catalyst). In this way, the α-alkylation of arylmethyl nitriles with primary alcohols was achieved with moderate to excellent yields and a broad substrate scope, including amino, halogen, methyl and methoxy substituents, on the aromatic nitrile (Table 8). 2-Pyridylsubstituted alcohols could also be used. Notably, the more challenging low-boiling alcohols (methanol and ethanol) could also be applied in this reaction, omitting the toluene as solvent at an oil-bath temperature of 135 C, resulting in synthetically useful yields (42–83%). The mechanism they proposed contains the following steps (Scheme 20): the catalyst is activated with base to form the amido complex I. This complex undergoes two separate reactions: on the one hand, it dehydrogenates the alcohol to the aldehyde via alkoxy species IV, which undergoes β-hydrogen elimination to form the aldehyde and the dihydride complex V. I also reacts with the alkyl nitrile to form

Catalytic Conversion of Nitriles by Metal Pincer Complexes

341

Scheme 20 Mechanism proposed by Gunanathan for the alkylation of nitriles with alcohols catalysed by 19

the imine species II via metal-ligand cooperation, which isomerises to the enamine form III. The complex III then reacts with the aldehyde and loses water to form species VI. This species releases the α,β-unsaturated nitrile, which is reduced by complex V to the α-alkylated nitrile. It should be stated, though, that four-membered unsaturated metalla-aminals such as II, III and VI are unprecedented and no proof for their existence was offered by the authors. An alternative mechanism via basecatalysed aldol condensation of the alkyl nitrile on the aldehyde seems more likely. More recently, calculations were performed by Zhang and co-workers, showing that the mechanism of Scheme 20 is unlikely in view of the high barriers involved [43]. Instead, their calculations favour the end-on mechanism as shown in Scheme 22 (bottom part). Zhu, Hao and co-workers synthesised a series of Ru NNN complexes based on bipyridyl imidazoline ligands and investigated the use of these complexes as

342

B. Guo et al.

Scheme 21 Alkylation of benzyl nitriles with alcohols by Ru NNN pincer complexes

catalysts for the α-alkylation of arylmethyl nitriles with primary alcohols (Scheme 21) [44]. With complex 21e, which showed the highest rates, a variety of alkylated nitriles could be successfully synthesised with 20–97% isolated yields at 140 C in toluene using 1.5 mol% catalyst loading and 0.15 eq. base additive. In most reactions benzyl alcohol or substituted benzyl alcohols were used as alkylating agents.

3.2

α-Olefination of Benzyl Cyanide and Aliphatic Nitriles with Alcohols

Milstein reported the α-olefination of nitriles with primary alcohols using Mn pincer complex 22 as catalyst [45]. This complex catalysed the dehydrogenative coupling of benzylic alcohols or purely aliphatic alcohols with (substituted) arylmethyl nitriles without any additives, resulting in moderate to excellent yields of the α,β-unsaturated nitriles. Notably, the α-olefination of benzyl cyanide with cinnamyl alcohol resulted in the diene in a very good yield (81%) (Table 9). Gunanathan and co-workers used the ruthenium-MACHO catalyst 19 for the olefination of (substituted) benzyl cyanide or aliphatic nitriles with secondary alcohols (Table 10) [46]. The reaction needs 2 eq. of base relative to 19; it not only serves to activate the catalyst but is also needed as catalyst for the Knoevenagel reaction. The reaction was performed in toluene at an oil-bath temperature of 135 C leading to the formation of the desired products in 20–93% yield. A broad substrate scope was achieved with methyl, methoxy, vinyl and halogen substituents on the aryl rings of the benzyl cyanide part. 2-Pyridin-2-ylacetonitrile was also used as substrate as were aliphatic nitriles and dinitriles. Most of the alkylations were performed with symmetrical secondary alcohols, such as cyclohexanol as this leads to a single product. Unsymmetrical secondary alcohols were also used in a number of cases, but their use leads to the formation of a mixture of E- and Zisomers. The proposed mechanism (Scheme 22, top) starts with the dehydrogenation of the alcohol by activated catalyst II via alkoxide complex III forming the corresponding

Catalytic Conversion of Nitriles by Metal Pincer Complexes

343

Table 9 Olefination of nitriles with primary alcohols catalysed by 22a

a

b

Isolated yields Yield by GC or NMR analysis using N,N-dimethylaniline internal standard

aldehydes or ketones by β-hydride elimination leading to the dihydride complex IV, which can lose hydrogen to reform II. Next, a Knoevenagel reaction between the nitriles and the in situ-formed aldehydes or ketones takes place catalysed by the extra base or by the Mn catalyst, which occurs via proton abstraction by the amide group in the activated complex. This latter mechanism, proposed by Milstein, was put in doubt by Sola, Poater and co-workers, who published an alternative version, based on DFT calculations, in which the activated complex II reacts with benzyl cyanide to form a (2-phenylvinylidene)amide manganese complex V, which is the nucleophile in the aldol condensation reaction leading to complex VI, which eliminates the nitrile (Scheme 22, bottom) [47]. Balaraman used manganese catalyst (6b) based on the MACHO ligand for the olefination of substituted and unsubstituted benzyl cyanides using mostly cyclic secondary alcohols such as cyclohexanol (Scheme 23) [48]. A total of 36 α, β-unsaturated nitriles were synthesised in this way with yields ranging from 20 to 88%.

344

B. Guo et al.

Table 10 Olefination of nitriles with alcohols by 19a

a d

b

c

Isolated yields Using 2 mol % catalyst and 4 mol % base Reaction time of 24 h Using 5 mol % catalyst 1 and 10 mol % base

3.3

α-Alkylation of Aliphatic and Benzylic Nitriles via Michael Addition on Unsaturated Ketones or Esters

Milstein and co-workers studied the reactivity of a Re complex based on a pyridinecontaining PNPtBu pincer with nitriles and found novel pathways for the activation of the C
N bond [17]. Based on the aromatisation and dearomatisation of the pyridine in the ligand backbone, the stoichiometric reaction of nitriles with the pincer complex 24 resulted in ketimido or enamino adducts, with remarkably low energy barriers. In addition, these processes are highly reversible (Scheme 24), opening the way to a lot of interesting reactivities. Taking advantage of these, they successfully demonstrated the Michael addition of benzyl nitriles to α,β-unsaturated esters and ketones with moderate to excellent yields (Table 11). Based on the stoichiometric studies, they proposed the following mechanism: reaction of 24 with benzyl cyanide leads to the enamido complex I. This complex reacts as a nucleophile in a Michael reaction on methyl acrylate to form the

Catalytic Conversion of Nitriles by Metal Pincer Complexes Scheme 22 Proposed mechanisms for olefination of nitriles with alcohols by pincer complexes

345

346

B. Guo et al.

Scheme 23 Manganese-catalysed olefination of benzyl cyanides with secondary alcohols

Scheme 24 Activation of nitriles via metal-ligand cooperation (MLC) by Re-pincer complex 24 (tBu groups on P atoms omitted for clarity)

zwitterionic intermediate II. Proton transfer leads to ketimido intermediate III, which undergoes elimination to form starting complex 24 and the alkylated benzyl cyanide (Scheme 25). The same group also reported the synthesis of the manganese pincer complex 25 based on the same ligand [49]. Interestingly, this catalyst was able to catalyse the same Michael reaction with lower catalyst loading and at a lower temperature. This paper mainly focused on fully aliphatic nitriles (Table 12). The effect of α- or β-substituents on the acrylate substrate was also investigated. Whereas the presence of an α-methyl group had no deleterious effect, the presence of a β-methyl group led to formation of the alkylated product in only 18% yield. More recently, Milstein and co-workers prepared pyridine-based PCP-ruthenium complexes in which the pyridine is bound via the carbon atom in the 4-position to

Catalytic Conversion of Nitriles by Metal Pincer Complexes

347

Table 11 Michael addition of benzyl nitriles to α,β-unsaturated esters and ketones by 24a

ruthenium [50]. This unusual structure is still capable of activating the C
N bond via metal-ligand cooperation. Under the influence of base, the methylene group in the side arm is deprotonated, and the ensuing negative charge is delocalised across the pyridine ring. This negatively charged species can react with nitriles as before leading to ketimido adduct 27 (Scheme 26). This adduct was used as catalyst for the double Michael reaction of benzyl cyanide on ethyl acrylate (2 eq.). If only one equivalent of ethyl acrylate was used, a mixture of mono- and dialkylated benzyl cyanide was obtained.

3.4

α-Deuteration of Nitriles

Gunanathan reported the selective deuterium labelling of aliphatic nitriles at the α-position using D2O as deuterium source [51]. With 2 eq. of base relative to catalyst and D2O as solvent, ruthenium-MACHO complex 19 catalysed the α-deuteration of nitriles resulting in 50–96.5% labelling (Scheme 27). Remarkably, a broad range of

348

B. Guo et al.

Scheme 25 Proposed mechanism for the Michael addition of benzyl nitriles to α,β-unsaturated esters and ketones catalysed by 24 (tBu groups omitted for clarity)

functional groups including amide, ester, thiol, amine, indole and heterocycles were tolerated. The authors propose a mechanism in which the nitrile adds across the ruthenium-nitrogen bond of the activated catalyst I to form a four-membered metallacycle II containing an imine group (Scheme 27). This complex can isomerise to the enamide form III in which the NH is exchanged for deuterium to form IV; isomerisation back to the imine form V and elimination of the nitrile complete the first deuteration step. The authors observe a species at m/z 669 which they propose as the protonated form of the fully deuterated adduct V0 . However, as discussed in Sect. 3.1 (Scheme 20), Zhang and co-workers performed DFT calculations on a similar mechanism involving structures II and III and concluded that the occurrence of such intermediates is highly unlikely in view of the high barriers that need to be overcome [52]. The species at m/z 669 is more easily explained by the end-on nitrile adduct VI or by the ketenimide adduct VIII (Scheme 28; see also structure V in Scheme 22). Deuterium exchange with D2O can either occur by deprotonation at the carbon centre α to the nitrile, the acidity

Catalytic Conversion of Nitriles by Metal Pincer Complexes

349

Table 12 Michael addition of nitriles to α,β-unsaturated esters and ketones catalysed by 25a

Scheme 26 Michael addition of benzonitrile to α,β-unsaturated ester catalysed by 26

of which is increased by coordination to the Lewis acidic metal centre. Alternatively, deprotonation/deuterium exchange can be facilitated by the basicity of the amido-N atom in the ligand backbone (Scheme 28).

350

B. Guo et al.

Scheme 27 Proposed mechanism for the α-deuteration of aliphatic nitriles catalysed by 19

3.5

α-Acylation of Unsaturated Nitriles

Szymczak and co-workers reported the hydrogenative acylation of α,β-unsaturated nitriles catalysed by a ruthenium pincer complex [53]. The reaction, which was performed with 4 eq. of the anhydride, complex 28 as catalyst at 100 C and 7 bar H2, using toluene as solvent and DBU as base, resulted in the desired products in good to excellent yields (Table 13). The reaction worked well on aliphatic as well as benzylic nitriles, and electron-donating as well as electron-withdrawing substituents on the

Catalytic Conversion of Nitriles by Metal Pincer Complexes

Scheme 28 Alternative mechanisms for the α-deuteration of aliphatic nitriles catalysed by 19 Table 13 Hydrogenative acylation of α,β-unsaturated nitriles by 28a

351

352

B. Guo et al.

Scheme 29 Synthesis of keteniminate adduct from 28 and 2,3-diphenylacrylonitrile and its reaction with Boc anhydride

aromatic rings were well-tolerated. Since acylation of saturated nitriles under the reaction conditions failed, the authors proposed that the role of DBU is solely to activate the anhydrides, but it does not promote the base-assisted acylation of the reduced nitriles. Stoichiometric reaction between 28 and 1,2-diphenylacrylonitrile allowed the successful isolation and characterisation of complex 29 (Scheme 29). Since 29 reacted stoichiometrically with (Boc)2O to form the acylated product, their hypothesis that the reaction proceeds via the intermediacy of keteniminates like 29 seems highly likely.

3.6

Aldol and Mannich Reactions with Aldehydes or Protected Imines

The combination of a transition metal complex with a base can function as an efficient catalyst for the α-alkylation of nitriles with aldehydes (aldol reaction) or imines (Mannich reaction). Ozerov and co-workers reported the coupling of acetonitrile with aldehydes using Ni(II)(PNP)OTf pincer complex 30a as catalyst [54]. At rt. or 45 C with an equivalent amount of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as base and excess acetonitrile as solvent (9.6–48 equivalents), aldehydes were converted, catalysed by 5 mol% of 30a, forming β-hydroxy nitriles in good to excellent yields (Scheme 30). The authors propose a mechanism in which the nitrile coordinates to the cationic nickel complex and is thus activated for deprotonation by DBU. The thus formed anion reacts with the aldehyde in an aldol condensation reaction. NMR investigation only showed the presence of the cationic complex formed from 30a and DBU. The reactions with the analogous palladium and platinum complexes led to the formation of the product in much lower yields under the same conditions. Szabó and co-workers reported the α-alkylation of nitriles with tosylimines catalysed by the Pd pincer complex 31a [55, 56]. This Mannich reaction, which required the addition of a weak base such as NaHCO3, allowed the high-yield synthesis of the tosylated β-aminonitriles from allyl or benzyl nitriles and tosylimines (Scheme 31, top). The authors assume that the nitrile binds to the

Catalytic Conversion of Nitriles by Metal Pincer Complexes

353

Scheme 30 Coupling of aldehydes with acetonitrile catalysed by 30a

Scheme 31 Mannich reaction of nitriles with tosylimines catalysed by 31a

cationic palladium complex, allowing facile deprotonation at the α-position even with a weak base. The anion then reacts with the tosylated imine to form the desired product. However, the authors propose a different mechanism for the Mannich reaction of allyl cyanides (Scheme 31, bottom) as this reaction does not proceed with other palladium salts or complexes, which seems to exclude a role as Lewis acid for the catalyst. They propose a mechanism that starts with formation of the cis-η1allylpalladium complex, rather than the trans-complex, the α-bound η1allylpalladium complex or the N-bound complex, all of which were found to be higher in energy by DFT calculations. This complex then reacts with the tosylimine in an SN2’-type reaction leading to selective alkylation in the α-position of the nitrile (Scheme 32). More recently, Butcher developed a series of bis-benzimidazole-based pincer complexes and applied 31b in the coupling of nitriles with sulfonimines with yields from 50 to 99% (Scheme 33) [57]. The catalyst was activated by reaction with AgOAc; in addition, 1.5 eq. of base (K2CO3) and molecular sieves (4 Å) were used as additives.

354

B. Guo et al.

Scheme 32 Stability of palladium-crotononitrile complexes calculated by DFT and proposed mechanism of the Mannich reaction between nitriles and sulfonimines by 31a

Scheme 33 Mannich reaction between benzyl cyanides and tosylimines catalysed by 31b

3.7

Enantioselective α-Functionalisation of Nitriles

Quite some effort has gone into developing enantioselective versions of the reactions described above. Richards and co-workers did pioneering work on the synthesis of cationic NCN Pd pincer complexes 32a–e bearing two chiral oxazoline ligands on the central benzene ring (Scheme 34) [58, 59]. They reported the asymmetric Michael addition of 2-methyl-cyanoacetate to α,β-unsaturated ketones/nitriles catalysed by these complexes. Unfortunately, the enantioselectivity of these reactions remained low and did not exceed 34%. Motoyoma and co-workers found that simply replacing the Pd metal by Rh (carrying additional chloride and Me3Sn ligands) but using similar ligands gave complexes 33a–f that led to a significant improvement of the enantioselectivity of this type of reaction, reaching values above 80% [60]. Uozumi [61] and co-workers used their unique Pd pincer complex 34a bearing two chiral hexahydro-1H-pyrrolo[1,2-c]imidazol-1-one moieties on the central benzene ring. In the Michael reaction on ethyl acrylate catalysed by this complex, an ee of 83% was obtained.

Catalytic Conversion of Nitriles by Metal Pincer Complexes

355

Scheme 34 Enantioselective Michael addition of 2-methyl-cyanoacetate catalysed by chiral pincer complexes

Scheme 35 Enantioselective Michael reaction of 2-methyl-cyanoacetate and MVK catalysed by a chiral palladium pincer complex

Mazet and Gade [62] developed an interesting ligand class based on the 2,5-bisoxazolinemethylpyrrole structure. These ligands were used in combination with palladium for the asymmetric Michael reaction of ethyl 2-methyl-cyanoacetate and methyl vinyl ketone (MVK) (Scheme 35). Unfortunately, these catalysts (34b,c) were very slow and conversions were as low as 14–29%; the product was obtained with ee values of 21 and 43%. Arai and co-workers developed a series of NCN pincer complexes bearing two imidazolidine moieties on the central benzene ring [63, 64]. The corresponding Pd complexes 35a,b were successfully applied in the conjugate addition of malononitrile to β-substituted nitroethylenes with good to excellent yields and ee’s (Scheme 36) [63]. Both aliphatic and (hetero)aromatic β-substituents were welltolerated. Using 3 mol% of the analogous Rh pincer complex 36a,b in toluene resulted in high yields and enantioselectivities in the asymmetric Mannich reaction of malononitrile with N-Boc imines [64]. Nakamura, Shibata and co-workers investigated the use in catalysis of bis (imidazoline) pincer ligands, which are easily synthesised and highly tunable. Using 1,3-bis(imidazoline)-benzene palladium pincer complexes as catalyst, they have successfully reacted many very challenging nitriles with imines (Scheme 37). Substrates included benzyl nitriles (up to 99% yield, 93:7 dr, 92% ee) [65],

356

B. Guo et al.

Scheme 36 Pincer complexes bearing two imidazolidine moieties as catalyst in enantioselective reactions: (a) Michael addition (top); (b) Mannich reaction (bottom)

Scheme 37 Pincer complexes based on bis(imidazoline) ligands as catalyst in enantioselective reactions

cyanoacetic acid (up to 82% yield, 90% ee) [66], α-phenylthioacetonitriles [up to 99% yield, 95:5 dr (anti/syn), 99% ee (anti)] [67], α-aminoacetonitriles [up to 95% yield, 97:3 dr (syn/anti), 99% ee (syn)] [68], dichloroacetonitrile (up to 99% yield, 94% ee) [69] and allenylnitriles (up to 89% yield, 99% ee) [70]. Whereas these

Catalytic Conversion of Nitriles by Metal Pincer Complexes

357

Scheme 38 Proposed mechanism for enantioselective Mannich reactions catalysed by 37

transformations are all Mannich-type reactions, the same catalyst has also been used for an aza-Morita-Baylis-Hillman reaction between acrylonitrile and tosylated imines (for the product, see Scheme 37, upper right). Here, the products were also isolated in excellent yields and ee’s (up to 99% yield, 98% ee) [71]. The authors proposed a general mechanism for the Mannich reactions (Scheme 38): first, the Pd-Br pincer complex is activated by the silver salt, leading to the formation of the cationic complex II in which the nitrile is end-on coordinated to the palladium. A loss of a proton, aided by the presence of basic acac anion, leads to the formation of the neutral keteneimido adduct III. This adduct reacts with the tosylated imine to form IV; the formed tosylated amine anion now binds to the palladium in preference to the neutral nitrile. Protonation of IV generates the product and regenerates I. When cyanoacetic acid is used in the Mannich reaction with sulfonylated imines, decarboxylation occurs leading to the formation of sulfonylated β-amino nitriles [66]. The authors claim that no decarboxylation was observed in the absence of

358

B. Guo et al.

Scheme 39 Asymmetric Mannich reaction of cyanoacetic acid catalysed by 37a followed by decarboxylation

imines or with α,α-disubstituted α-cyanoacetic acid. Using atmospheric pressure chemical ionisation (APCI) mass spectroscopic analysis, they established the presence of the sulfonylated α-cyano-β-aminopropionic acid intermediate in the reaction mixture at the end of the reaction (cationic mode, m/z calcd. for C15H14N3O4S, 332.1; found, 332.2), which suggests that the CO2 is extruded after the C-C bond formation event. They assume that the catalytic cycle starts by the binding of the acetoacetic acid via the carboxylate. They observe a complex in the APCI that has the correct mass for [M + H]+ (Scheme 39); however, the complex in which the cyanoacetate binds via the nitrile would give exactly the same MS and thus cannot be excluded. Recently, the same group used complex 37e as catalyst for the asymmetric conjugate addition of α,α-dithioacetonitriles to nitroalkenes. The reactions proceeded in excellent yield and the products were obtained with excellent ee’s [72].

4 Hetero-Michael Addition to α,β-Unsaturated Nitriles 4.1

Lewis Acid Catalysis

The hetero-Michael reaction on α,β-unsaturated nitriles gives access to a large number of interesting structures. Apart from the β-substituted hydroxy-, alkoxy-, amino-, phosphino- or (alkyl)thio-propionitriles, one can of course easily convert the nitrile group in these products by hydrogenation to the analogous amines, by hydration to the amides and by hydrolysis to the carboxylic acids. These reactions become even more interesting when they are enantioselective. The synthesis of chiral pharma intermediates is often quoted as a reason to develop these reactions. Trogler reported the first palladium pincer complex (38)-catalysed aza-Michael addition (Scheme 40) [73]. However, this reaction was limited to acrylonitrile as substrate. Nevertheless, up to 44 turnovers were achieved by this catalyst in the reaction between aniline and acrylonitrile at room temperature. Inspired by this work, Hartwig used in situ prepared PCP as well as PNP palladium pincer complexes and achieved good to excellent yields (76–99%) in the addition of primary and

Catalytic Conversion of Nitriles by Metal Pincer Complexes

359

Scheme 40 Aza-Michael addition on α,β-unsaturated nitriles catalysed by palladium pincer complexes

Scheme 41 Asymmetric aza-Michael addition catalysed by a Ni Pigiphos complex

secondary amines as well as aniline to methacrylonitrile and crotonitrile (Scheme 40) [74]. The Togni group developed a new nickel pincer complex 40 based on the Pigiphos ligand [75]. This complex showed good activity in the hydroamination of α,β-unsaturated nitriles (Scheme 41). Even the challenging substrate aniline did react with crotonitrile catalysed by 40 at rt. giving the aminated product in up to 91% yield with 22% ee. Aliphatic amines (morpholine and piperidine) were also evaluated in reactions with methacrylonitrile/crotonitrile leading to the products in excellent yields. The highest ee (69%) was obtained in the hydroamination of

360

B. Guo et al.

Scheme 42 Asymmetric aza-Michael addition of aliphatic cyclic amines to methacrylonitrile at 80 C catalysed by 40

methacrylonitrile with morpholine. To further increase the enantioselectivity of this reaction, ionic liquids were explored as solvent [76]. Unfortunately, no improvement w.r.t. the enantioselectivity resulted from this. Nevertheless, the catalytic activity was significantly enhanced by the ionic liquids: higher TON (up to 300, compared with 71 in THF), reusable after five recycles and air stable. Eventually, excellent enantioselectivity could be obtained in this reaction by lowering the temperature [77]. At 80 C, addition of aliphatic cyclic amines to methacrylonitrile still resulted in excellent yields of the products and >90% ee after 48 h using 5% [Ni(Pigiphos)(THF)](ClO4)2 as catalyst (Scheme 42). However, this high enantioselectivity could not be achieved in the conversion of other substrates, such as crotonitrile or cinnamonitrile. Based on DFT calculations, the authors proposed a mechanism involving a Lewis acid (metal) activation of the nitrile, which enables the 1,4-addition of the amine to the α,β-unsaturated nitrile forming the ammonium substituted alkyl nitrile (Scheme 43). An asymmetric proton transfer is followed by dissociation of the nitrile [78]. The same catalyst (40) was also investigated for the hydrophosphination of methacrylonitrile by the Togni group [79, 80]. One aromatic and five aliphatic secondary phosphines were reacted, and moderate to excellent yields of the tri-substituted phosphines were obtained at 20 C with acetone as solvent (Table 14). Enantioselectivities ranged from 32 to 94%. The authors proposed the same mechanism as for the amine addition. The Zargarian group extensively investigated the synthesis and use of nickel complexes based on PCP ligands using bisphosphines as well as bisphosphinites with aliphatic or aromatic backbones. These cationic nickel complexes were used as catalysts in the aza-Michael addition of amines such as morpholine and cyclohexylamine as well as aniline to acrylonitrile, crotonitrile and methacrylonitrile. The

Catalytic Conversion of Nitriles by Metal Pincer Complexes

361

Scheme 43 Proposed mechanism for the asymmetric aza-Michael addition catalysed by 40 Table 14 Asymmetric Michael addition of phosphines to methacrylonitrile by 40a

Entry 1 2 3 4 5 6 a

R2PH Cy2PH Ph2PH i Pr2PH t Bu2PH Ad2PH (EtMe2C)2PH

TON 10 15 45 100 100 116

Solvent Methacrylonitrile Methacrylonitrile Acetone Acetone Acetone Acetone

Yield (%) 71 10 Not isolated 95 97 86

ee (%) 70 32 78 94 89 90

Isolated yields, Ad ¼ 1-adamantyl

products were obtained in excellent yields and turnover numbers up to 2000 were achieved (Scheme 44) [81, 82]. The same group reported a dimeric nickel-PCN complex 44 which showed good activity in the oxa-Michael addition of aliphatic alcohols and phenols to acrylonitrile [83]. During the substrate scope investigation, the author found that the reaction rate is depending on the alcohol acidity (m-cresol > BnOH > aliphatic alcohols) and very sensitive to the steric hindrance (crotonitrile and methacrylonitrile EtOH > n-PrOH > iPrOH) (Table 15). In later research, the same group found that the reactivity of the Ni pincer complex 42 (X ¼ O) in the oxa/aza-Michael additions could be significantly increased by adding Et3N [84, 85]. This was particularly true for reactions with phenols and anilines. To some extent, water had the same effect. Liu, Imamoto, Zhang and co-workers prepared a series of chiral PXP Ni pincer complexes. Complex 45 was successfully applied in the hydroamination of methacrylonitrile, crotonitrile and other α- and β-substituted acrylonitriles

362

B. Guo et al.

Scheme 44 Ni PCP pincer complexes as catalysts for the aza-Michael addition Table 15 Oxa-Michael addition of alcohols to acrylonitrile by 44a

Entry 1 2 3 4 5 6 7 a

ROH m-Cresol MeOH EtOH CF3CH2OH n-PrOH i PrOH BnOH

X:nitrile:ROH 1:2000:4000 1:200:2000 1:200:2000 1:200:2000 1:200:2000 1:200:2000 1:200:2000

Time (h) 36 1.0 1.0 0.5 8 24 1.0

Yield (%) 99 100 87 90 64 23 100

TON ~2000 200 174 180 128 46 200

Reaction yields were determined by 1H NMR spectroscopy or GC/MS

[86]. However, the reactions were rather slow and the enantioselectivity did not surpass 46% (Scheme 45).

4.2

Catalysis via Metal-Ligand Cooperation

Thus far, in all hetero-Michael addition reactions, the catalyst has functioned as a Lewis acid activating the α,β-unsaturated nitrile for 1,4-addition and stabilising the developing negative charge on the nitrogen atom. An entirely different mode of

Catalytic Conversion of Nitriles by Metal Pincer Complexes

363

Scheme 45 Chiral nickel PCP pincer complex as catalyst for the aza-Michael addition

reactivity based on metal-ligand cooperativity (MLC) was reported by de Vries, Otten and co-workers [87]. Using the dearomatised Ru PNN pincer complex 46, which was developed by Milstein and co-workers, they examined the oxa-Michael reaction of α,β-unsaturated nitriles at room temperature in THF. A broad substrate range was investigated, and the product β-alkoxy-nitriles were obtained in moderate to excellent yields. Nitriles investigated included acrylonitrile, crotonitrile, 2- and 3-pentenenitrile and the ester containing 10-cyanodec-9-en-1-yl acetate. No transesterification occurred in the latter substrate. As nucleophiles, primary and secondary aliphatic alcohols, benzyl alcohols and thiols were used. In a later publication, the nitrile scope was further extended to more challenging structures containing β-substituents such as CF3, p-CF3C6H4 and cyclopentyl. Here the products were obtained in moderate to good yields [88]. Unfortunately, phenols deactivated the catalyst by strongly binding to the ruthenium. Aza-Michael reactions were extremely slow and yields below 50% of the products were obtained (Table 16). Based on the crystal structure of the dieneamido complex formed from 3-pentenenitrile and 46 (Scheme 46, complex II) as well as on DFT calculations, they proposed a mechanism (shown here for 2-pentenenitrile) involving (i) the activation of the C
N bond via metal-ligand cooperation resulting in addition of the nitrile across the ruthenium metal and the carbon atom of the deprotonated side arm leading to the formation of I in fast equilibrium with II; (b) hydrogen bondassisted 1,4-addition (III ! IV); (c) proton transfer (IV!V); and (d) release of the product and regeneration of the catalyst (it is the tautomeric form 46-taut that is the active catalyst). Milstein and co-workers used manganese PNN pincer complex 47 for the oxaand aza-Michael reaction for which they assumed the same mechanism as proposed by Otten and de Vries [89]. This catalyst is rather active, which allowed the use of a low catalyst loading of 0.1 mol% (compared to 0.5 mol% of the ruthenium catalyst 46) to achieve a similar rate in the oxa-Michael reaction. Interestingly, in contrast to the ruthenium catalyst, the aza-Michael addition also worked very well under the same conditions, with an even higher rate than the oxa-Michael reaction. Thus, in the presence of nBuOH (1 mL) and nBuNH2 (1 mL), catalyst 47 selectively promoted the aza-Michael addition (Scheme 47).

364 Table 16 Oxa-Michael addition of alcohols to nitriles catalysed by 46a

Scheme 46 Oxa-Michael addition via a MLC mechanism catalysed by 46

B. Guo et al.

Catalytic Conversion of Nitriles by Metal Pincer Complexes

365

Scheme 47 Aza-Michael addition of amines to nitriles catalysed by 47

4.3

Hydration of Nitriles

As amides are important structural motifs both in pharmaceuticals and bulk chemicals, their synthesis via hydration of nitriles has attracted considerable attention from the organic synthesis community. Several groups have used metal pincer complexes to catalyse this reaction (Scheme 48). In 2015, the Piers group reported the use of pincer complex 48 as catalyst for the hydration of nitriles to the analogous amides [90]. With low catalyst loadings (0.05–0.5 mol%) and without additives, excellent yields were achieved at 80 C using (hetero)aromatic and aliphatic nitriles as substrates. Substrates containing acidic groups such as phenols or acetone cyanohydrin did not react. Acrylonitrile did react, but the oxa-Michael addition with isopropanol (used as a solvent) was a major side reaction. The crystal structure of complex 48 showed an unusually long

Scheme 48 Hydration of nitriles catalysed by pincer complexes

366

B. Guo et al.

Scheme 49 Mechanism of the hydration of nitriles catalysed by 48

Ni-O bond (1.978 Å). In addition, they observed 17O exchange in a labelling experiment. This led the authors to propose that 48 reacts with water to form cationic aqua complex I (Scheme 49). Surprisingly, complex 48 did not react with benzonitrile. However, upon addition of water, benzonitrile was converted leading to the formation of benzamide. The authors found that several nickel species were present in solution during the reaction but were able to crystallise complex II from this mixture. It is assumed that benzonitrile displaces the water ligand in I after which attack of the hydroxide on the nitrile occurs leading to the formation of the benzamidyl species. Huang and co-workers developed PN3P pincer-based Ni-OH complex 51 (Scheme 48) [91]. Stoichiometric reactions between this complex and nitriles were successful, forming the amide adduct. The catalytic hydration of nitriles was achieved at 100 C, albeit with limited substrate scope (six examples); the amides were isolated in moderate to high yields. Boncella, Tondreau and co-worker described another approach for the hydration of nitriles using non-innocent ligand-based pincer complexes [92]. They synthesised and characterised 5 Ni PNP pincer complexes of which 50 was studied extensively together with the known Mn-OH pincer complex 49. They managed to obtain Ni and Mn carboxamide complexes from stoichiometric reactions between these complexes and acetonitrile or benzonitrile. However, catalytic conversions failed with the Ni complexes, and very low turnover numbers (1.7–3.9) were achieved with the Mn complex at 50 C using THF (10 wt% water) as solvent. During the mechanistic study, the authors found that nitriles coordinate to the Mn centre by replacing bromide; phenolate or benzyloxide could not be displaced (Scheme 50). Based on this, the authors propose that nucleophilic attack occurs from Mn-OH to the nitrile, which is activated by hydrogen bonding to the NH of the ligand (I in Scheme 50). However, the high stability of the metal amide products (III) inhibits the catalytic hydration of nitriles. Otten and co-workers successfully applied the concept of metal-ligand cooperation (MLC) to the hydration of nitriles [93]. Catalysed by the dearomatised Ru PNP and PNN pincer complexes with 5 equiv. water at room temperature, 33 substrates including (hetero)aromatic and aliphatic nitriles were converted in excellent yields to

Catalytic Conversion of Nitriles by Metal Pincer Complexes

367

Scheme 50 Proposed mechanism for hydration of nitriles by 49

the primary amides. Using either more or less water resulted in lower yields of the amides. Based on the stoichiometric reactions between on the one hand the catalyst and nitrile and on the other hand the catalyst with amide, they proposed the following mechanism (Scheme 51): (a) activation of the C
N bond via MLC leading to addition of the nitrile to form the ketimido complex I in which the nitrogen is bound to the ruthenium and the carbon atom to the carbon atom of one of the side arms; (b) nucleophilic attack at the carbon atom of the ketimido adduct by water, likely assisted by hydrogen bonding with a second molecule of water (TS);

368

B. Guo et al.

Scheme 51 MLC hydration of nitriles catalysed by 9

and (c) release of the initially formed iminol, which rapidly tautomerises to the amide product in solution, to regenerate the catalyst. Studies of catalyst speciation suggest that during turnover, the active catalyst 9 is in equilibrium with the aqua complex III and the carboxamide complex IV; thus, the product amide and water are competitive inhibitors. Nevertheless, the ligand exchange processes are rather fast enabling efficient catalysis, in spite of these inhibitors.

Catalytic Conversion of Nitriles by Metal Pincer Complexes

4.4

369

Amination of Nitriles

Zargarian and co-workers reported the amination of nitriles with primary or secondary amines to the amidines, catalysed by nickel pincer complexes. The reaction was initially discovered as an unwanted side reaction in the Michael addition of amines to cinnamonitrile [84]. More recently, together with French colleagues, they developed a new series of PCP ligands having an imidazolium group as part of the ligand backbone [94]. Bis-cationic complex 55 catalysed the addition of piperidine, morpholine or hexylamine to acetonitrile or benzonitrile with moderate to good yields (Table 17). Use of primary amines led to the formation of mixtures of the Nmono-substituted and the N,N0 -disubstituted amidines. Use of a 4:1 mixture of primary amine and nitrile led to the sole production of the disubstituted product. Arnold had previously shown that the Ni PNP complex 56 catalysed the addition of piperidine to acetonitrile with 5 mol% of catalyst at room temperature to give 68% yield of the amidine (Scheme 52) [95].

Table 17 PCP pincer complex-catalysed amination of nitriles

370

B. Guo et al.

Scheme 52 Nickel PNP pincer complex-catalysed amidine formation Table 18 Synthesis of 2-alkylaminoquinolines from o-aminobenzyl alcohol, benzyl nitriles and primary alcohols catalysed by Ru pincer complex 57a

a

4.5

Isolated yields

Synthesis of Heterocycles

Since nitrogen-containing heterocycles are omnipresent in natural products and bioactive molecules, their synthesis under mild conditions, using efficient and green methodologies, is an important topic. A novel route towards heterocycles from alcohols using an acceptorless dehydrogenative coupling (ADC) fits well with such sustainability demands [96]. Kundu reported the synthesis of 2-alkylaminoquinolines from o-amino-benzyl alcohol, benzyl nitriles and primary alcohols, catalysed by 3.5 mol% ruthenium pincer complex 57 [97]. This procedure is a one-pot two-step synthesis, which first step required Ru catalyst, nitrile, oaminobenzyl alcohol and base in dioxane for 0.5 h at 125 C for the formation of the 2-aminoquinoline, followed by the addition of 2 equivalents of primary alcohol for the borrowing hydrogen-type alkylation of the 2-amino group. The methodology was demonstrated on six examples, two of which were performed on gram scale; yields ranged from 45 to 78% (Table 18). Kundu also used a new cobalt pincer complex 58 bearing an NNN ligand as catalyst for this reaction (Scheme 53)

Catalytic Conversion of Nitriles by Metal Pincer Complexes

371

Scheme 53 Synthesis of 2-alkylaminoquinolines catalysed by Co pincer complex 58

Scheme 54 Tunable synthesis of heterocycles catalysed by Mn pincer complex 59

[98]. The substrates were similar to the ones used in the previous publication; however, a higher catalyst loading (5 mol% for each step), more base (1 equivalent for each step) and a higher temperature (150 C) were necessary. The reaction was demonstrated with eight examples which included substrates with heteroaromatic and aliphatic substituents, and the products were isolated in moderate to excellent yields. Srimani showed that the dehydrogenative coupling of o-aminobenzyl alcohol and nitriles catalysed by manganese NNS pincer complex 59 as catalyst can lead to two different products [99]. Either quinazoline or 2-aminoquinoline can be selectively obtained by simply tuning the solvent (xylene to toluene) and the base (KOtBu to KOH). Both pathways could be applied with high functional group tolerance. In addition, the one-pot synthesis of 2-alkylaminoquinolines was also achieved and demonstrated with nine examples in moderate yields (Scheme 54). The Togni group reported the enantioselective 1,3-dipolar cycloaddition of C, N-cyclic azomethine imines to acrylonitrile and crotonitrile catalysed by [Ni (PigiPhos)](ClO4) (40) [100]. Using acrylonitrile as substrate, a range of substituted C,N-cyclic azomethine imines based on (substituted) 3,4-dihydro-isoquinoline reacted at room temperature in CH2Cl2 resulting in excellent yields and ee’s of the products (Scheme 55). Using the same conditions, crotonitrile reacted much slower (4 days) and the product was obtained with lower ee (62%). More challenging nitriles reacted very sluggishly (methacrylonitrile, 52% conversion after 48 h at 40 C) or not at all (trans-cinnamonitrile and cis-2-pentenenitrile).

372

B. Guo et al.

Scheme 55 Enantioselective 1,3-dipolar cycloaddition catalysed by 40

5 Conclusions The use of pincer ligands in metal complexes allows a large number of different reactivities, many of which were previously unknown. This is particularly true for nitriles and α,β-unsaturated nitriles that can react in different ways with these complexes. Pincer complexes have been used extensively for hydrogenation reactions of nitriles to the amines, but also to the imines. Here, the classical Noyori-type metal-ligand cooperativity mechanism is often operative. Many catalysts, particularly cationic complexes, can activate the nitriles by functioning as Lewis acid, which makes the α-proton quite acidic leading to the formation of a metal ketenimide complex allowing C-C bond-forming reactions. Quite unusual reactivity was found in the dearomatised pyridine-containing PNN and PNP complexes that were developed by Milstein and co-workers. Here the nitrile can add to the dearomatised pincer complex via metal-ligand cooperation to form a metal nitrogen bond, whereas the carbon atom of the nitrile is bound to the carbon atom of one of the side arms in a fully reversible reaction. The initially formed metal ketimido complex can isomerise further to an enamide structure. In all cases the reactivity of the metal imide is much higher than that of the nitrile allowing a range of reactions such as its hydration. But also oxo- and aza-Michael reactions are facilitated on the unsaturated nitriles, such acrylonitriles or pentenitriles. It is clear that this interesting reversible reactivity will enable also other transformations, and we can expect more findings in this or similar fields in the near future. Particularly in the field of enantioselective transformations, more developments are expected.

References 1. Moulton CJ, Shaw BL (1976) Transition metal-carbon bonds. Part XLII. Complexes of nickel, palladium, platinum, rhodium and iridium with the tridentate ligand 2,6-Bis[(di-tbutylphosphino)methyl]phenyl. J Chem Soc Dalton Trans:1020–1024 2. Morales-Morales D, Jensen CM (2007) The chemistry of pincer compounds. Elsevier Science B.V, Amsterdam

Catalytic Conversion of Nitriles by Metal Pincer Complexes

373

3. Peris E, Crabtree RH (2018) Key factors in pincer ligand design. Chem Soc Rev 47:1959–1968 4. van Koten G, Milstein D (2013) Organometallic pincer chemistry. Topics in organometallic chemistry, vol 40. Springer, Cham 5. Szabo KJ, Wendt OF (2014) Pincer and pincer-type complexes: applications in organic synthesis and catalysis. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 6. van der Boom ME, Milstein D (2003) Cyclometalated phosphine-based pincer complexes: mechanistic insight in catalysis, coordination, and bond activation. Chem Rev 103:1759–1792 7. Selander N, Szabo KJ (2011) Catalysis by palladium pincer complexes. Chem Rev 111:2048–2076 8. Choi J, MacArthur AHR, Brookhart M, Goldman AS (2011) Dehydrogenation and related reactions catalyzed by iridium pincer complexes. Chem Rev 111:1761–1779 9. Gunanathan C, Milstein D (2014) Bond activation and catalysis by ruthenium pincer complexes. Chem Rev 114:12024–12087 10. Kumar A, Bhatti TM, Goldman AS (2017) Dehydrogenation of alkanes and aliphatic groups by pincer-ligated metal complexes. Chem Rev 117:12357–12384 11. Alig L, Fritz M, Schneider S (2019) First-row transition metal (De)hydrogenation catalysis based on functional pincer ligands. Chem Rev 119:2681–2751 12. Merz LS, Ballmann J, Gade LH (2020) Phosphines and N-heterocycles joining forces: an emerging structural motif in PNP-pincer chemistry. Eur J Inorg Chem 2023–2042 13. Michelin RA, Mozzon M, Bertani R (1996) Reactions of transition metal-coordinated nitriles. Coord Chem Rev 147:299–338 14. Yang X, Fleming FF (2017) C- and N-metalated nitriles: the relationship between structure and selectivity. Acc Chem Res 50:2556–2568 15. Khusnutdinova JR, Milstein D (2015) Metal-ligand cooperation. Angew Chem Int Ed 54:12236–12273 16. Gunanathan C, Milstein D (2011) Metal ligand cooperation by aromatization dearomatization: a new paradigm in bond activation and “green” catalysis. Acc Chem Res 44:588–602 17. Vogt M, Nerush A, Iron MA, Leitus G, Diskin-Posner Y, Shimon LJ, Ben-David Y, Milstein D (2013) Activation of nitriles by metal ligand cooperation. Reversible formation of ketimidoand enamido-rhenium PNP pincer complexes and relevance to catalytic design. J Am Chem Soc 135:17004–17018 18. Eijsink LE, Perdriau SC, de Vries JG, Otten E (2016) Metal-ligand cooperative activation of nitriles by a ruthenium complex with a de-aromatized PNN pincer ligand. Dalton Trans 45:16033–16039 19. Gunanathan C, Hölscher M, Leitner W (2011) Reduction of nitriles to amines with H2 catalyzed by nonclassical ruthenium hydrides - water-promoted selectivity for primary amines and mechanistic investigations. Eur J Inorg Chem 2011:3381–3386 20. Kuriyama W, Matsumoto T, Ogata O, Ino Y, Aoki K, Tanaka S, Ishida K, Kobayashi T, Sayo N, Saito T (2012) Catalytic hydrogenation of esters. Development of an efficient catalyst and processes for synthesising (R)-1,2-propanediol and 2-(l-menthoxy)ethanol. Org Process Res Dev 16:166–171 21. Touge T, Aoki K, Nara H, Kuriyama W (2012) Method for producing compound with carbonyl group from alcohols by using ruthenium carbonyl complex having tridentate ligand as dehydrogenation oxidation catalyst. WO2012144650A1 22. Neumann J, Bornschein C, Jiao H, Junge K, Beller M (2015) Hydrogenation of aliphatic and aromatic nitriles using a defined ruthenium PNP pincer catalyst. Eur J Org Chem 2015:5944–5948 23. Adam R, Alberico E, Baumann W, Drexler HJ, Jackstell R, Junge H, Beller M (2016) NNP-type pincer imidazolylphosphine ruthenium complexes: efficient base-free hydrogenation of aromatic and aliphatic nitriles under mild conditions. Chem Eur J 22:4991–5002

374

B. Guo et al.

24. Mukherjee A, Srimani D, Ben-David Y, Milstein D (2017) Low-pressure hydrogenation of nitriles to primary amines catalyzed by ruthenium pincer complexes. Scope and mechanism. ChemCatChem 9:559–563 25. Bornschein C, Werkmeister S, Wendt B, Jiao H, Alberico E, Baumann W, Junge H, Junge K, Beller M (2014) Mild and selective hydrogenation of aromatic and aliphatic (di)nitriles with a well-defined iron pincer complex. Nat Commun 5:4111 26. Alberico E, Sponholz P, Cordes C, Nielsen M, Drexler H-J, Baumann W, Junge H, Beller M (2013) Selective hydrogen production from methanol with a defined iron pincer catalyst under mild conditions. Angew Chem Int Ed 52:14162–14166 27. Lange S, Elangovan S, Cordes C, Spannenberg A, Jiao H, Junge H, Bachmann S, Scalone M, Topf C, Junge K, Beller M (2016) Selective catalytic hydrogenation of nitriles to primary amines using iron pincer complexes. Cat Sci Technol 6:4768–4772 28. Elangovan S, Topf C, Fischer S, Jiao H, Spannenberg A, Baumann W, Ludwig R, Junge K, Beller M (2016) Selective catalytic hydrogenations of nitriles, ketones, and aldehydes by welldefined manganese pincer complexes. J Am Chem Soc 138:8809–8814 29. Mukherjee A, Srimani D, Chakraborty S, Ben-David Y, Milstein D (2015) Selective hydrogenation of nitriles to primary amines catalyzed by a cobalt pincer complex. J Am Chem Soc 137:8888–8891 30. Tokmic K, Jackson BJ, Salazar A, Woods TJ, Fout AR (2017) Cobalt-catalyzed and Lewis acid-assisted nitrile hydrogenation to primary amines: a combined effort. J Am Chem Soc 139:13554–13561 31. Schneekönig J, Tannert B, Hornke H, Beller M, Junge K (2019) Cobalt pincer complexes for catalytic reduction of nitriles to primary amines. Cat Sci Technol 9:1779–1783 32. Shao Z, Fu S, Wei M, Zhou S, Liu Q (2016) Mild and selective cobalt-catalyzed chemodivergent transfer hydrogenation of nitriles. Angew Chem Int Ed 55:14653–14657 33. Srimani D, Feller M, Ben-David Y, Milstein D (2012) Catalytic coupling of nitriles with amines to selectively form imines under mild hydrogen pressure. Chem Commun 48:11853–11855 34. Choi J-H, Prechtl MHG (2015) Tuneable hydrogenation of nitriles into imines or amines with a ruthenium pincer complex under mild conditions. ChemCatChem 7:1023–1028 35. Chakraborty S, Berke H (2014) Homogeneous hydrogenation of nitriles catalyzed by molybdenum and tungsten amides. ACS Catal 4:2191–2194 36. Chakraborty S, Leitus G, Milstein D (2017) Iron-catalyzed mild and selective hydrogenative cross-coupling of nitriles and amines to form secondary aldimines. Angew Chem Int Ed 56:2074–2078 37. Chakraborty S, Milstein D (2017) Selective hydrogenation of nitriles to secondary imines catalyzed by an iron pincer complex. ACS Catal 7:3968–3972 38. Li H, Al-Dakhil A, Lupp D, Gholap SS, Lai Z, Liang LC, Huang KW (2018) Cobalt-catalyzed selective hydrogenation of nitriles to secondary imines. Org Lett 20:6430–6435 39. Dai H, Guan H (2018) Switching the selectivity of cobalt-catalyzed hydrogenation of nitriles. ACS Catal 8:9125–9130 40. Grigg R, Mitchell TRB, Sutthivaiyakit S, Tongpenyai N (1981) Oxidation of alcohols by transition metal complexes part V. selective catalytic monoalkylation of arylacetonitriles by alcohols. Tetrahedron Lett 22:4107–4110 41. Alos J, Bolano T, Esteruelas MA, Olivan M, Onate E, Valencia M (2014) POP-pincer ruthenium complexes: d(6) counterparts of osmium d(4) species. Inorg Chem 53:1195–1209 42. Thiyagarajan S, Gunanathan C (2017) Facile ruthenium(II)-catalyzed α-alkylation of arylmethyl nitriles using alcohols enabled by metal–ligand cooperation. ACS Catal 7:5483–5490 43. Huo R-P, Zhang X, Zhang C-F, Li X-X (2019) Computational insight into the mechanism of ruthenium(II)-catalyzed α-alkylation of arylmethyl nitriles using alcohols. J Phys Chem A 123:10263–10272

Catalytic Conversion of Nitriles by Metal Pincer Complexes

375

44. Zhu Z-H, Li Y, Wang Y-B, Lan Z-G, Zhu X, Hao X-Q, Song M-P (2019) α-Alkylation of nitriles with alcohols catalyzed by NNN0 pincer Ru(II) complexes bearing bipyridyl imidazoline ligands. Organometallics 38:2156–2166 45. Chakraborty S, Das UK, Ben-David Y, Milstein D (2017) Manganese catalyzed alphaolefination of nitriles by primary alcohols. J Am Chem Soc 139:11710–11713 46. Thiyagarajan S, Gunanathan C (2018) Ruthenium-catalyzed α-olefination of nitriles using secondary alcohols. ACS Catal 8:2473–2478 47. Luque-Urrutia JA, Sola M, Milstein D, Poater A (2019) Mechanism of the manganese-pincercatalyzed acceptorless dehydrogenative coupling of nitriles and alcohols. J Am Chem Soc 141:2398–2403 48. Yadav V, Landge VG, Subaramanian M, Balaraman E (2020) Manganese-catalyzed alphaolefination of nitriles with secondary alcohols. ACS Catal 10:947–954 49. Nerush A, Vogt M, Gellrich U, Leitus G, Ben-David Y, Milstein D (2016) Template catalysis by metal-ligand cooperation. C-C bond formation via conjugate addition of non-activated nitriles under mild, base-free conditions catalyzed by a manganese pincer complex. J Am Chem Soc 138:6985–6997 50. Tang S, von Wolff N, Diskin-Posner Y, Leitus G, Ben-David Y, Milstein D (2019) Pyridinebased PCP-ruthenium complexes: unusual structures and metal-ligand cooperation. J Am Chem Soc 141:7554–7561 51. Krishnakumar V, Gunanathan C (2018) Ruthenium-catalyzed selective alpha-deuteration of aliphatic nitriles using D2O. Chem Commun 54:8705–8708 52. Huo RP, Zhang X, Zhang CF, Li XX (2019) Computational insight into the mechanism of ruthenium(II)catalyzed alpha-alkylation of arylmethyl nitriles using alcohols. J Phys Chem A 123:10263–10272 53. Hale LVA, Sikes NM, Szymczak NK (2019) Reductive C-C coupling from alpha,betaunsaturated nitriles by intercepting keteniminates. Angew Chem Int Ed 58:8531–8535 54. Fan L, Ozerov OV (2005) Efficient nickel catalyst for coupling of acetonitrile with aldehydes. Chem Commun:4450–4452 55. Aydin J, Conrad CS, Szabó KJ (2008) Stereoselective pincer-complex catalyzed C-H functionalization of benzyl nitriles under mild conditions. An efficient route to β-aminonitriles. Org Lett 10:5175–5178 56. Aydin J, Szabó KJ (2008) Palladium-pincer complex catalyzed C-C coupling of allyl nitriles with tosyl imines via regioselective allylic C-H bond functionalization. Org Lett 10:2881–2884 57. Rani V, Singh HB, Butcher RJ (2018) Protic and substituted NCN palladium(II) pincer complexes with 1,3-bis(benzimidazol-20 -yl)-2-bromobenzenes: structure and catalysis. J Organomet Chem 859:33–43 58. Stark MA, Richards CJ (1997) Synthesis and application of cationic 2,6-Bis(2-oxazolinyl) phenylpalladium(ll) complexes. Tetrahedron Lett 38:5881–5884 59. Stark MA, Jones G, Richards CJ (2000) Cationic [2,6-Bis(20 -oxazolinyl)phenyl]palladium (II) complexes: catalysts for the asymmetric michael reaction. Organometallics 19:1282–1291 60. Motoyama Y, Koga Y, Kobayashi K, Aoki K, Nishiyama H (2002) Novel asymmetric michael addition of cyanopropionates to acrolein by the use of a bis(oxazolinyl)phenylstannanederived rhodium(III) complex as a chiral Lewis acid catalyst. Chem Eur J 8:2968–2975 61. Takenaka K, Minakawa M, Uozumi Y (2005) NCN pincer palladium complexes: their preparation via a ligand introduction route and their catalytic properties. J Am Chem Soc 127:12273–12281 62. Mazet C, Gade LH (2003) Double bonds in motion: bis(oxazolinylmethyl)pyrroles and their metal-induced planarization to a new class of rigid chiral C2-symmetric complexes. Chem Eur J 9:1759–1767 63. Arai T, Oka I, Morihata T, Awata A, Masu H (2013) A neutral, chiral, bis(imidazolidine)derived NCN-type palladium pincer complex with catalytic activity. Chem Eur J 19:1554–1557

376

B. Guo et al.

64. Arai T, Moribatake T, Masu H (2015) Chiral bis(imidazolidine)-derived NCN pincer Rh complex for catalytic asymmetric mannich reaction of malononitrile with N-Boc imines. Chem Eur J 21:10671–10675 65. Hyodo K, Nakamura S, Tsuji K, Ogawa T, Funahashi Y, Shibata N (2011) Enantioselective reaction of imines and benzyl nitriles using palladium pincer complexes with C2-symmetric chiral bis(imidazoline)s. Adv Synth Catal 353:3385–3390 66. Hyodo K, Kondo M, Funahashi Y, Nakamura S (2013) Catalytic enantioselective decarboxylative cyanoalkylation of imines by using palladium pincer complexes with C2-symmetric chiral bis(imidazoline)s. Chem Eur J 19:4128–4134 67. Kondo M, Kobayashi N, Hatanaka T, Funahashi Y, Nakamura S (2015) Catalytic enantioselective reaction of alpha-phenylthioacetonitriles with imines using chiral bis (imidazoline)-palladium catalysts. Chem Eur J 21:9066–9070 68. Kondo M, Nishi T, Hatanaka T, Funahashi Y, Nakamura S (2015) Catalytic enantioselective reaction of alpha-aminoacetonitriles using chiral bis(imidazoline) palladium catalysts. Angew Chem Int Ed 54:8198–8202 69. Kondo M, Sugimoto M, Nakamura S (2016) Direct catalytic enantioselective Mannich-type reaction of dichloroacetonitrile using bis(imidazoline)-Pd catalysts. Chem Commun 52:13604–13607 70. Kondo M, Omori M, Hatanaka T, Funahashi Y, Nakamura S (2017) Catalytic enantioselective reaction of allenylnitriles with imines using chiral bis(imidazoline)s palladium(II) pincer complexes. Angew Chem Int Ed 56:8677–8680 71. Hyodo K, Nakamura S, Shibata N (2012) Enantioselective aza-Morita-Baylis-Hillman reactions of acrylonitrile catalyzed by palladium(II) pincer complexes having C2-symmetric chiral bis(imidazoline) ligands. Angew Chem Int Ed 51:10337–10341 72. Nakamura S, Tokunaga A, Saito H, Kondo M (2019) Enantioselective conjugate addition of an alpha,alpha-dithioacetonitrile with nitroalkenes using chiral bis(imidazoline)-Pd complexes. Chem Commun 55:5391–5394 73. Seligson AL, Trogler WC (1993) Protonolysis approach to the catalytic amination of olefins with bis(phosphine)palladium(II) dialkyls. Organometallics 12:744–751 74. Kawatsura M, Hartwig JF (2001) Transition metal-catalyzed addition of amines to acrylic acid derivatives. A high-throughput method for evaluating hydroamination of primary and secondary alkylamines. Organometallics 20:1960–1964 75. Fadini L, Togni A (2003) Ni(II) complexes containing chiral tridentate phosphines as new catalysts for the hydroamination of activated olefins. Chem Commun:30–31 76. Fadini L, Togni A (2007) Asymmetric catalytic hydroamination of activated olefins in ionic liquids. HeIv Chim Acta 90:411–424 77. Fadini L, Togni A (2008) Asymmetric hydroamination of acrylonitrile derivatives catalyzed by Ni(II)-complexes. Tetrahedron Asymmetry 19:2555–2562 78. Senn HM, Blöchl PE, Togni A (2000) Toward an alkene hydroamination catalyst: static and dynamic ab initio DFT studies. J Am Chem Soc 122:4098–4107 79. Sadow AD, Haller I, Fadini L, Togni A (2004) Nickel(II)-catalyzed highly enantioselective hydrophosphination of methacrylonitrile. J Am Chem Soc 126:14704–14705 80. Sadow AD, Togni A (2005) Enantioselective addition of secondary phosphines to methacrylonitrile: catalysis and mechanism. J Am Chem Soc 127:17012–17024 81. Pandarus V, Zargarian D (2007) New pincer-type diphosphinito (POCOP) complexes of nickel. Organometallics 26:4321–4334 82. Castonguay A, Spasyuk DM, Madern N, Beauchamp AL, Zargarian D (2009) Regioselective hydroamination of acrylonitrile catalyzed by cationic pincer complexes of nickel(II). Organometallics 28:2134–2141 83. Spasyuk DM, Zargarian D (2010) Monomeric and dimeric nickel complexes derived from a pincer ligand featuring a secondary amine donor moiety. Inorg Chem 49:6203–6213 84. Lefèvre X, Durieux G, Lesturgez S, Zargarian D (2011) Addition of amines and phenols to acrylonitrile derivatives catalyzed by the POCOP-type pincer complex [{κP,κC,κP-2,6(i-Pr2PO)2C6H3}Ni(NCMe)][OSO2CF3]. J Mol Catal A Chem 335:1–7

Catalytic Conversion of Nitriles by Metal Pincer Complexes

377

85. Salah AB, Offenstein C, Zargarian D (2011) Hydroamination and alcoholysis of acrylonitrile promoted by the pincer complex {κP,κC,κP-2,6-(Ph2PO)2C6H3}Ni(OSO2CF3). Organometallics 30:5352–5364 86. Yang Z, Liu D, Liu Y, Sugiya M, Imamoto T, Zhang W (2015) Synthesis and structural characterization of nickel complexes possessing P-stereogenic pincer scaffolds and their application in asymmetric Aza-Michael reactions. Organometallics 34:1228–1237 87. Perdriau S, Zijlstra DS, Heeres HJ, de Vries JG, Otten E (2015) A metal-ligand cooperative pathway for intermolecular oxa-Michael additions to unsaturated nitriles. Angew Chem Int Ed 54:4236–4240 88. Guo B, Zijlstra DS, de Vries JG, Otten E (2018) Oxa-Michael addition to alpha,betaunsaturated nitriles: an expedient route to gamma-amino alcohols and derivatives. ChemCatChem 10:2868–2872 89. Tang S, Milstein D (2019) Template catalysis by manganese pincer complexes: oxa- and aza-Michael additions to unsaturated nitriles. Chem Sci 10:8990–8994 90. Borau-Garcia J, Gutsulyak DV, Burford RJ, Piers WE (2015) Selective hydration of nitriles to amides catalysed by PCP pincer supported nickel(II) complexes. Dalton Trans 44:12082–12085 91. Yao C, Chakraborty P, Aresu E, Li H, Guan C, Zhou C, Liang LC, Huang KW (2018) Monomeric nickel hydroxide stabilized by a sterically demanding phosphorus-nitrogen PN(3) P-pincer ligand: synthesis, reactivity and catalysis. Dalton Trans 47:16057–16065 92. Anderson NH, Boncella JM, Tondreau AM (2018) Investigation of nitrile hydration chemistry by two transition metal hydroxide complexes: Mn–OH and Ni–OH nitrile insertion chemistry. Organometallics 37:4675–4684 93. Guo B, de Vries JG, Otten E (2019) Hydration of nitriles using a metal–ligand cooperative ruthenium pincer catalyst. Chem Sci 10:10647–10652 94. Vabre B, Canac Y, Lepetit C, Duhayon C, Chauvin R, Zargarian D (2015) Charge effects in PCP pincer complexes of Ni(II) bearing phosphinite and imidazol(i)ophosphine coordinating jaws: from synthesis to catalysis through bonding analysis. Chem Eur J 21:17403–17414 95. Rozenel SS, Kerr JB, Arnold J (2011) Metal complexes of Co, Ni and Cu with the pincer ligand HN(CH2CH2PiPr2)2: preparation, characterization and electrochemistry. Dalton Trans 40:10397–10405 96. Gunanathan C, Milstein D (2013) Applications of acceptorless dehydrogenation and related transformations in chemical synthesis. Science 341:1229712 97. Maji M, Chakrabarti K, Paul B, Roy BC, Kundu S (2018) Ruthenium(II)-NNN-pincercomplex-catalyzed reactions between various alcohols and amines for sustainable CN and CC bond formation. Adv Synth Catal 360:722–729 98. Shee S, Ganguli K, Jana K, Kundu S (2018) Cobalt complex catalyzed atom-economical synthesis of quinoxaline, quinoline and 2-alkylaminoquinoline derivatives. Chem Commun 54:6883–6886 99. Das K, Mondal A, Pal D, Srimani D (2019) Sustainable synthesis of quinazoline and 2-aminoquinoline via dehydrogenative coupling of 2-aminobenzyl alcohol and nitrile catalyzed by phosphine-free manganese pincer complex. Org Lett 21:3223–3227 100. Milosevic S, Togni A (2013) Enantioselective 1,3-dipolar cycloaddition of C,N-cyclic azomethine imines to unsaturated nitriles catalyzed by NiII-pigiphos. J Org Chem 78:9638–9646

Top Organomet Chem (2021) 68: 379–450 https://doi.org/10.1007/3418_2020_71 # Springer Nature Switzerland AG 2020 Published online: 24 January 2021

The Application of Pincer Ligand in Catalytic Water Splitting Hong-Tao Zhang and Ming-Tian Zhang

Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Artificial Photosynthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Water Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Water Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 Types of Pincer Ligand Used in HER/OER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Molecular Catalysts with Pincer Ligand for HER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Non-Hydrogenase Type HECs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Models of Hydrogenase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Assembled Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Molecular Catalysts with Pincer Ligand for OER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Noble Metals-Based Water Oxidation Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Base Metals-Based Water Oxidation Catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion and Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

380 380 382 384 387 390 390 405 412 413 413 429 436 436

Abstract The study of catalytic water splitting is one of the most active areas of research across many sub-disciplines of chemistry. To understand the mechanistic details and design artificial molecular catalysts for both water reduction (Hydrogen Evolution Reaction, HER) and water oxidation (Oxygen Evolution Reaction, OER) continue to be a challenge for the development of artificial photosynthetic system. This chapter will focus on the summarization of recent development in the rapidly growing field of artificial molecular catalysts with pincer ligand for both HER and OER. Keywords Artificial photosynthesis · Hydrogen evolution · Pincer ligand · Water oxidation

H.-T. Zhang and M.-T. Zhang (*) Center of Basic Molecular Science (CBMS), Department of Chemistry, Tsinghua University, Beijing, China e-mail: [email protected]

380

H.-T. Zhang and M.-T. Zhang

Fig. 1 Photosynthesis and the photosynthetic machinery. Reprinted with permission from Ref. [15]

1 Introduction The study of catalytic water splitting is one of the most active areas of research across many sub-disciplines of chemistry [1–10]. To understand the mechanistic details and design artificial molecular catalysts for both water reduction (Hydrogen Evolution Reaction, HER) and water oxidation (Oxygen Evolution Reaction, OER) continue to be a challenge for the development of artificial photosynthetic system. The search for novel and improved water splitting has led to the development of a number of homogeneous and heterogeneous catalysts for HER and OER. In the following sections, this chapter will focus on the summarization of recent development in the rapidly growing field of artificial molecular catalysts with pincer ligand for both HER and OER.

1.1

Artificial Photosynthesis

In nature, green plants use photosynthesis to convert CO2 and H2O into glucose for life, accompanied by the release of oxygen (Eq. 1), which ingeniously realizes the conversion of light energy to chemical energy (Fig. 1). In the photosystem II (PSII), there is a central pair of chlorophylls, P680, which can be excited by photon. After irradiation, the excited state P680* undergoes electron transfer to generate P680•+, which has strong oxidation ability. The radical cation can oxidize manganese

The Application of Pincer Ligand in Catalytic Water Splitting

381

Fig. 2 Schematic picture of a molecular assembly for overall H2O splitting

calcium (Mn4Ca) cluster which can be used to catalyze water oxidation to promote oxygen release [11–14]. 6CO2 þ 12H2 O ! C6 H12 O6 þ 6O2 þ 6H2 O

ð1Þ

Inspired by this natural system, great efforts have been made on artificial photosynthesis systems for water splitting to achieve the conversion of solar energy to clean and renewable hydrogen energy (Eq. 2). The artificial device needs several basic but challenging photophysical steps and complex catalytic conversion to achieve the combination of two half reactions [16–18]: (1) Oxidation of H2O to release O2 and generate multiple protons and electrons (Eq. 3); (2) Reduction of H+ to form molecular hydrogen (Eq. 4) [1–3, 6–10]. In artificial photosynthesis system, it is essential to design an effectively continuous process. First, a chromophore (photosensitizer) is used to absorb a photon and form a charge separation state by transferring electrons to a reduction catalyst which can continuously accept and accumulate two electrons for the reduction of two protons to produce a molecule of hydrogen; The charge-separated chromophores can return to the initial state by accepting the electrons given by the oxidation catalyst which has to reach a high oxidation state by losing four electrons in succession, then grabs another four electrons from two water molecules and back to the initial state and produces one molecule of oxygen and four protons [19, 20]. A molecular system is assembled by photosensitizer, water oxidation catalyst (WOC), and hydrogen evolving catalyst (HEC) to realize aforementioned water splitting process (Fig. 2). Despite this cartoon discussed above seem simple, it is a challenge to combine all the process well to achieve efficient photosensitization, rapid electron transfer, and ultimately efficient catalysis. 2H2 O ! O2 þ 2H2 þ

2H2 O ! O2 þ 4H þ 4e 4Hþ þ 4e ! 2H2

ð2Þ 

ð3Þ ð4Þ

382

1.2

H.-T. Zhang and M.-T. Zhang

Water Reduction

Effective process to achieve a sustainable large-scale hydrogen economy is still in challenge [21–24]. The reduction of protons to produce hydrogen by photochemistry and electrochemistry is a green and renewable method, while the reduction of protons to hydrogen has a large activation barrier and thus leads to a high overpotential and a slow reaction rate. Therefore, an efficient catalyst is necessary to lower the barrier and improve the performance. (Eq. 4) [25–28]. In nature, a class of metal enzymes, hydrogenase, can catalyze hydrogen to convert protons and electrons, and can also perform its reversible reaction to give hydrogen [29]. Hydrogenase exists widely in bacteria, archaea, and some eukarya in nature [30–34], and can be divided into [FeFe]-, [NiFe]-, and [Fe]-hydrogenases according to the metal ions in their active centers (Fig. 3) [30, 31, 35, 36]. In the active center of [FeFe]-hydrogenase, the diiron center coordinated with carbonyl and cyanide ligands is bridged by a dithiolate ligand, and one of the two Fe ions is connected to the [4Fe4S] cluster via a cysteinyl residue. One iron center has an octahedral coordination environment, while the other iron has a five-coordinated tetragonal pyramid geometry. It is precisely because of the unsaturated iron central coordination that it can bind protons in the reduced state or hydrogen molecules in the oxidized state, thus realizing the reversibility between protons and hydrogen. In this process, the proton relay (-NH-) in the molecule plays an important role. The TOF of hydrogen production catalyzed by [FeFe]-hydrogenase can reach as high as 9,000 s1 [37, 38]. However, the [NiFe]-hydrogenase with heteronuclear bimetals maintains a certain difference with [FeFe]-hydrogenase. The active sites are mostly coordinated by cysteine residues, carbonyl and cyanide ligands. The full-coordinated octahedral structure is maintained around the iron atom, while the penta-coordinated triangular bipyramidal geometry is adopted for the nickel atom. It was found that the active center of [NiFe]-hydrogenase was on the nickel atom, while the iron atom existed as a redox auxiliary metal ion, which may lead to its catalytic activity different from that of [FeFe]-hydrogenase. Nickel-iron hydrogenase has excellent catalytic properties in hydrogen-activated, but its TOF for reducing proton to hydrogen evolution is about 700 s1 [38]. Inspired by hydrogenase, researchers have developed a large number of hydrogenase models to explore their ability of hydrogen production from proton reduction [29, 39, 40]. Platinum-based compounds are known as excellent proton reduction catalysts [41, 42], and are still used in the industrial field of hydrogen production from water electrolysis. Even though its catalytic performance has tended to be perfect, platinum is a rare precious metal with limited abundance, which limits its application in largescale industrial hydrogen production. Therefore, it is of great significance to develop a cheap catalyst for water reduction and hydrogen evolution based on the first transition metal. In addition to the exploration of proton reduction by mimicking

Fig. 3 Structures of the [NiFe]-hydrogenase from DvMF [43], of the [FeFe]-hydrogenase from Dd [44] and [Fe]-hydrogenase (C176A) with the substrate, methylene-H4MPT in the open form (PDB 3H65) [45]. Reprinted with permission from Ref. [29]

The Application of Pincer Ligand in Catalytic Water Splitting 383

384

H.-T. Zhang and M.-T. Zhang

hydrogenase, hydrogen evolution catalysts based on other first-row transition metals, including macrocyclic cobalt-based [46–52] and nickel-based [53–58] compounds, have been studied for decades [59].

1.3

Water Oxidation

Up to now, water oxidation has become a bottleneck restricting the development of artificial photosynthesis, which is manifested that the catalytic performance of water oxidation catalyst cannot match that of reduction catalyst. The main challenges in water oxidation research are as follows: 1. Involving multi-proton and electron transfer – for a redox process, generally only one or two electrons are involved, while the process of water oxidation to give oxygen requires four consecutive four electrons oxidation and four protons release; 2. Involving multiple bonds breakage and formation – the breakage of OH bonds and the formation of OO bond; 3. Ambiguous mechanism of OO bond formation – as a recognized key step in catalysis, the formation of OO bond is still unclear. In the photosynthetic system (PSII), the site of water oxidation is located in the Mn4Ca cluster (Fig. 4). It consists of four manganese atoms and one calcium atom, and five metal atoms are grasped together by surrounding μ-oxo and μ-hydroxo ligands to form a cubic-like structure, in which three manganese atoms in the manganese center form a cubane structure with calcium atom, and the fourth manganese atom, like a pendant, is connected to the cube by two oxo groups [60–70]. The mechanism of water oxidation catalyzed by the oxygen evolving complex has also been extensively studied [71–73]. The general view of the OEC cycle is that five oxidation states are involved in the catalytic process, designated as S0-S4 (Kok cycle) [73]. S0 represents the most reduced state, reaching the most oxidized state of S4 through four consecutive photo-induced proton-coupled electron transfer (PCET) processes, and then the high-energy intermediate releases oxygen back to the initial state of S0 (Scheme 1). In many biological and chemical processes, proton-coupled electron transfer (PCET) is an effective way to generate highly oxidized species without changing the total charge of molecules. It is also an important feature of the four electrons oxidation of water molecule in artificial photosynthesis. Since 1982, a large number of novel molecular catalysts for water oxidation have been developed. In order to evaluate the catalytic activity of these artificial catalysts, it is often necessary to utilize oxidants with high oxidation potential to drive water oxidation catalyst (WOC) to react with water to produce oxygen. One of the

The Application of Pincer Ligand in Catalytic Water Splitting

385

Fig. 4 Structure of the manganese calcium cubane, Mn4Ca, located in the OEC. Reprinted with permission from Ref. [60]

prerequisites of these sacrificial oxidants is that they have higher oxidation potential than WOC, which makes the oxidation of H2O thermodynamically advantageous. At present, there are three methods to drive catalysts to catalyze water oxidation: chemical oxidation, photocatalytic oxidation, and electrocatalytic water oxidation. In chemical oxidation, CeIV with 1.74 V vs NHE redox potential is commonly used as a sacrificial oxidant, but it usually needs acidic conditions [74, 75]. In the lightdriven catalytic process, the corresponding Ru3+ species produced by commonly [Ru (bpy)3]2+-based photosensitizers excited by photo are employed as oxidants [76]. The advantages of this system are that this series photosensitizers have orderly different oxidation potentials. Choosing the right oxidizer as needed can effectively avoid over-oxidation of catalysts. The method of electrocatalytic water oxidation provides reliable information support for the study of catalytic kinetics and mechanism. For catalytic reaction, the precondition of designing more efficient and perfect catalysts is the mechanism research. Recognition of every elementary reaction and key step in catalytic process facilitates the transformation of better catalysts. The formation of OO bond as a key step in catalytic water oxidation needs further understanding. Although the verification of the mechanism is difficult to fully discuss with the experimental results, after decades of development, two widely

386

H.-T. Zhang and M.-T. Zhang

Scheme 1 Kok cycle in which the OEC cycles through five redox states, S0S4, via consecutive photo-oxidation. Reprinted with permission from Ref. [61]

Scheme 2 Schematic representation of the two mechanistic pathways for OO bond formation. Reprinted with permission from Ref. [61]

accepted mechanisms of OO bond formation in water oxidation have been proposed (Scheme 2): (1) water nucleophilic attack (WNA) and (2) the interaction of two metal-oxo (I2M) [77–86]. In WNA mechanism, water molecules play the role of nucleophilic reagent, attacking the formed high-energy metal-oxo species to form

The Application of Pincer Ligand in Catalytic Water Splitting

387

metal peroxide species (M-OO-H), which can be further oxidized to release oxygen. In I2M mechanism, two metal-oxo species which hold significant radical character generate a (M-O-O-M) species by radical coupling, and then M-O-O-M species is further oxidized to release oxygen. Even though these two mechanisms have been generally recognized, it is still necessary to study in depth which mechanism is preferred in different catalysts and catalytic systems.

1.4

Types of Pincer Ligand Used in HER/OER

Pincer ligands are a kind of tridentate ligand, which can form a planar coordination mode around the metal center, resulting in the unique stability and reactivity of pincer-metal complexes [87]. The design and synthesis of pincer ligands become easily in a few steps, so that their electronic effect and steric nature can be easily adjusted. Based on the special coordination mode and function of pincer ligands, transition metal complexes containing pincer ligands exhibit excellent performance in the fields of synthesis, bond activation, and catalysis [88–90]. Pincer ligands have been applied for proton reduction catalysts and water oxidation catalysts because of catalyst structure and activity could be easily tuned based on pincer platform. For hydrogen evolution, the catalysts are required to increase the density of electron cloud in the metal center with the enhancement of reduction ability. Based on this consideration, some phosphine-based, sulfur-based, and anionic pincer ligands with strong electron-donating ability will have priority in the synthesis of proton reduction catalysts. For water oxidation, the metal centers of catalysts often need to reach a high oxidation state to drive the conversion of water to oxygen, which requires that the ligands have strong antioxidant capacity and stability. And thus tpy and its derivatives are extensively used in water oxidation catalysis. In the HER/OER, various pincer-type ligands have been employed in designing catalysts and regulating the catalytic activity. According to the difference of skeletal structure, the ligands will be divided into three classes – tpy and its derivatives L1–34 (Fig. 5), non-completely rigid tridentate pincer ligands L35–71 (Fig. 6), and unconventional pincer ligands L72–74 (Fig. 7). As mentioned earlier, tpy-type ligands are widely used for both synthetic HECs (hydrogen evolution catalysts) and WOCs (water oxidation catalysts). Besides that, other NNN-, NNP-, PNP-, PCP-, ONO- ligands based on pincer platform have also been employed for their strong plasticity in redox, electron-donating ability and degree of deformation. Interestingly, some multi-dentate ligands (Hthiop, dpktsc, and Cl-TMPA) can also be regarded as pincer-like ligands when they are used as tridentate ligand.

388

Fig. 5 Tpy-tpye pincer ligands used in HER/OER

H.-T. Zhang and M.-T. Zhang

The Application of Pincer Ligand in Catalytic Water Splitting

Fig. 6 Non-completely rigid tridentate pincer ligands used in HER/OER

389

390

H.-T. Zhang and M.-T. Zhang

Fig. 7 Unconventional tridentate ligands used in HER/OER

Fig. 8 Structure of Co(tpp)2Cl2 (R1)

2 Molecular Catalysts with Pincer Ligand for HER 2.1

Non-Hydrogenase Type HECs

As early as 1930, it has been known that a large amount of energy is stored in hydrogen molecule, which can provide the necessary energy for the life activities of some microorganisms [91]. After the energy crisis broke out in 1970, the research of using solar energy to produce clean and renewable hydrogen has become a popular field [92]. However, hydrogen production from water reduction by homogeneous catalysts based on pincer-type ligands has rarely been reported. Until 1997, Bauer’s group [93] reported the first novel cobalt-based catalyst [Co (tpp)2]Cl2 (R1, Fig. 8; tpp ¼ 2,3,5,6-tetra(pyridin-2-yl)pyrazine) containing pincerlike tpp ligand. It was found that catalyst R1 was successfully used as homogeneous catalysts for photo-induced hydrogen production with [Ru(bpy)3]2+ as photosensitizer and ascorbic acid as the sacrificial electron donor. The drawback was that quantum yields for H2 formation were below 1%. As the catalyst was reduced to a low-valent intermediate on the surface of the electrode, electrochemical signals of free tpp ligands are not captured, which is different from the more efficient catalyst [Co(pdt)3]2+ (pdt ¼ 5,6-diphenyl-3-(pyridin-2-yl)-1,2,4-triazine), which contains three dentate ligands. Obviously, the large steric hindrance and the strong chelation of tpp ligand make it difficult to be replaced by water molecules and dissociate into [Co(tpp)]2+ species with mono-ligand. So, it is not easy for low-valent intermediates

The Application of Pincer Ligand in Catalytic Water Splitting

391

Fig. 9 Structure of [Co (tpy)2]l2 (R2)

to react with proton. This might be the reason that catalytic efficiency of catalyst R1 is much lower than that of [Co(pdt)3]2+. According to both electrochemical- and light-driven hydrogen evolution, the author proposed the following mechanisms: Firstly, [Co(tpp)2]2+ was reduced to [Co(tpp)2]+ by 3*Ru; Then a water molecule coordinates with [Co(tpp)2]+ and dissociates one of the ligands to form a [CoH (tpp)]+ species; Finally, hydrogen is released from the interaction between the two [CoH(tpp)]+ species, and then combined with the free tpp to return to the initial state of the catalyst. In 2001, Abe group [94] studied the reduction of proton to producing hydrogen by [Co(terpy)2]I2 (R2, Fig. 9, tpy ¼ 2,20 :60 ,200 -terpyridine) containing tpy ligand under electrocatalytic conditions. In DMA solution containing 0.1 M NaClO4, [Co (tpy)]2+ showed three redox couples at +0.36 V (CoIII/II), 0.67 V (CoII/I), 1.51 V (CoI/ 0 ) in the range of 2.0 to 0.5 V vs AgCl/Ag. For the third reduction process, the formation of [Co(tpy)]2•+ by one-electron reduction occurring on the ligand is formally recorded as Co(0) [95]. When 20% H2O was introduced into the system, CoIII/II and CoII/I processes were still clearly visible in cyclic voltammetry, while CoI/0 redox pairs disappeared, accompanied by a sharp increase in potential below 1.5 V. Experiments showed that Co(0) species formed by electroreduction can effectively catalyze proton reduction, and H2 evolution was detected by separation. TOF of hydrogen evolution at 1.7 V is estimated to be 250 h1. In this work, it is considered that the very electron-rich Co(0) species will coordinate with H+ to transform into Co(0)-H intermediates and be adsorbed on the electrode surface. Because Co(0)-H exhibited hydrophobic interaction, it is not easy for water molecules to react further, so a bimolecular catalytic dehydrogenation process was proposed. In addition, Abe also used [Co(tpy)]2+ to synthesize molecular aggregates (Nf[Co(terpy)22+]) [94]. It was found that in electrocatalysis at pH < 3.8, the heterogeneous catalyst will form Co(I)-H species active species to further reduce proton and give out hydrogen; while at pH > 3.8, the Co(I) species could not react with protons because of its weak alkalinity, and thus it was further reduced to form an effective active Co(0)-H species. Similarly, the heterogeneous (Nf[Co(terpy)22+] catalyst has also been proposed as a catalytic process for hydrogen evolution from the interaction between two Co-H species. So far, noble metal platinum is still the most effective catalyst for hydrogen production in industrial aquatic electrolysis, although Pt-based catalyst is one of the earliest hydrogen evolution catalysts. Despite the high price of platinum metal,

392

H.-T. Zhang and M.-T. Zhang

Fig. 10 Structures of [Pt (ttpy)Cl]+ (R3) and [Pt (tpy)Cl]+ (R4)

molecular Pt-based catalysts for proton reduction are designed for significantly improving the catalytic efficiency. In early 2000, many platinum group metal complexes as molecular catalysts [96, 97] were reported to enhance catalytic efficiency and cost-effectiveness instead of the traditional heterogeneous colloidal Pt. Eisenberg group [98] had different opinions on whether these Pt(II) complexes are H2 generating catalysts. Two Pt(II) compounds were selected for generating H2, one of which was a mononuclear catalyst [Pt(ttpy)Cl]+ (R3, Fig. 10, ttpy ¼ 4-tolyl2,20 :60 ,200 -terpyridine) composed of pincer-type ttpy ligand. It is well known that platinized titanium dioxide (TiO2) can reduce proton in aqueous solutions to produce large amounts of hydrogen under ultraviolet light irradiation in the presence of electronic sacrificial agents such as Na2S, Na2SO3 [99, 100]. Hence, using compound R3 instead of colloidal Pt can also produce the hydrogen gas under similar conditions. After irradiation, free Pt(II) complexes adsorbed on the surface of TiO2 particles were removed after treatment. ESM and EDAX measurements showed that the surface of TiO2 particles changed significantly before and after irradiation. After irradiation, there were many fine particles adsorbed on TiO2 surface, and EDAX showed that there were Pt elements on TiO2. These TiO2 particles were placed in a fresh reaction solution without Pt(II) and a large amount of hydrogen was also produced after irradiation. Because mercury has a strong toxic effect on colloidal Pt, no side effects on molecular Pt-based complexes have been proved by experiments. When mercury was introduced into the catalytic system, the system could not release hydrogen with excitation. A large number of controlled experiments have proved that Pt(II) “molecular” catalysts can be photo-decomposed to form colloidal Pt in the process of hydrogen evolution under illumination. The former only exists as a catalytic precursor, while the latter is the real catalyst. Through the experiments of some photostable Pt (II) compounds [101], they found that these complexes could not catalyze hydrogen production under the same conditions. And thus “Noble metal complexes as H2 generating catalysts that in fact photodecomposition of these complexes may be occurring and that the question of catalysis by the resultant colloids needs to be addressed rigorously.” [98]. Nevertheless, it is important to investigate the catalytic activity of molecular species in solution. In order to get direct and convincing evidence for H2 generating activity of Pt(II)-based molecular catalysts, Sakai group [102] investigated the activity and stability of a series of Pt(II)-based compounds for thermal reduction

The Application of Pincer Ligand in Catalytic Water Splitting

393

Scheme 3 The process of catalyzing water reduction to H2 by Milstein catalyst R5. Reprinted with permission from Ref. [104]

of water to hydrogen under dark conditions. Complex [Pt(tpy)Cl]Cl (R4, Fig. 10) containing tpy ligand was utilized as a molecular catalyst. In the atmosphere of inert gas, the reaction process between MV•+ produced by in-situ controlled potential electrolysis and complex R4 was studied by stopped-flow technique. And by monitoring the decay process of MV•+, it is confirmed that MV•+ can reduce Pt (II) by one electron [103]. Stopped-flow experiments show that there is no solid particle sedimentation during the mixing process, so the formation of colloidal platinum can be excluded. Under dark conditions, in AcOH/AcONa aqueous solution system, MV•+ prepared by pre-electrolysis reacted with catalyst R4, and the gas produced in the system was carried out by Ar gas stream. It was confirmed that hydrogen was emitted, but the catalytic efficiency was relatively low. The development of efficient photo-driven water splitting catalysts is the main goal of sustainable hydrogen energy research. In 2009, Milstein group [104] developed a Ru-based catalyst R5 [RuIIH(tBuPNEtN)(CO), tBuPNEtN ¼ (2-(di-tertbutylphosphinomethyl)-6-diethylaminomethyl)pyridine] for water splitting, which can catalyze thermal hydrogen evolution and photochemical oxygen evolution. In the absence of sacrificial oxidants and reductants, the two half-cycles of water oxidation and water reduction were successfully combined to realize the splitting of water. This work demonstrates the mechanism of hydrogen formation and OO bonding by stepwise stoichiometric manner. Here, the catalytic proton reduction is in detail, while catalytic water oxidation will be discussed in later chapters. First, de-aromatized [Ru(II)] pincer complex R5 can react with water to get aromatic Ru(II) hydrido-hydroxo complex R6 which was characterized by X-ray and NMR. It is confirmed that after water molecules were activated, OH coordinated to Ru atom and H+ transferred to the side arm of P-terminal by isotope labeling. When compound R6 was heated and refluxed in water for a long time, it will produce green cis-dihydroxo complex R7 accompanied by evolution of H2 (Scheme 3). The generation of hydrogen was quantified by GC, and the yield was 37%. The author suggested that the hydrogen is formed by the electrophilic attack of water molecules on the hydride ligand, in which the aromatization and de-aromatization of pincer ligand played an important role in the catalytic process. Subsequently, DFT calculations [105–109] were carried out to study the mechanism of the catalytic reaction. Yoshizawa group [105] carried out a detailed theoretical study on metal-ligand cooperation for the aromatization and de-aromatization of catalyst R5 in the catalytic

394

H.-T. Zhang and M.-T. Zhang

Fig. 11 Structures of NiIIH (tBuPCP) (R8) and W-H species

process. However, they used the models containing PMe2 and NMe2 to replace PtBu2 and NEt2 groups, respectively. It is pointed out that the release of hydrogen from the interaction between protons and hydride ligand is the rate-limiting step in the whole reaction, and protonation takes place preferentially with hydride rather than OH. Hall group [106] calculated the position of pincer ligand de-aromatization after H2 formation using the same model as R5. Deprotonation on methylene with PtBu2 group is more favorable than on the side with NEt2. In 2011, Suresh group [107] proposed a new catalytic process, which did not involve the aromatization–dearomatization of pincer ligands. In 2012, Fabris group [108] used “an explicit description of the solvent and employ metadynamics coupled with the Car-Parrinello method” to propose a catalytic mechanism with lower activation energy. The aromatization and de-aromatization of ligand are also not involved, and it shows that water molecules play an active role in the catalytic process. In 2015, 56 models with different substituents at 4-py or N-arm or P-arm were calculated and compared by Suresh [109]. When only substituents on para-position of the pyridyl were changed, there was only 1–1.4 kcal/mol energy effect between them. When the P atom was substituted by Et or Me, the water molecule coordination would be promoted and the catalytic efficiency would be greatly improved. Some transition-metal hydrides also have good reactivity as proton sources [110]. Because hydrides of different metal hydrides show “acidity” or “alkalinity,” they react with each other and form hydrogen such as [OsH2(PMePh2)4]/[CpM(H) (CO)3] system (M ¼ Mo, W) [111]. In 2011, Peruzzini and coworkers reported that the Ni-H species NiIIH(tBuPCP) (R8, Fig. 11, tBuPCP ¼ 2,6-C6H3(CH2PtBu2)2) employing pincer ligands interacted with W-H species to produce hydrogen through acid–base interactions [112]. By combining lots of crystallographic, spectroscopic, and DFT analysis, this is the first example of hydrogen release from a dihydrogen bond adduct formed by two transition-metal hydrides. The study of the reactivity of acidic and basic transition-metal hydrides provides an important reference for the study of proton reduction and hydrogen evolution by bimetallic synergistic catalysis. In 2011, Fujita and coworkers [113] developed two polypyridine Ru-based catalysts p-[Ru(tpy)(pynap)(OH2)]2+ (R9) and d-[Ru(tpy)(pynap)(OH2)]2+ (R10) (Fig. 12, p ¼ proximal, d ¼ distal), which differ in the orientation of the asymmetric pynap ligand (pynap ¼ 2-(pyrid-20 -yl)-1,8-naphthyridine). Although the two isomers are so similar, they exhibit completely different properties in the electrocatalytic process. Catalyst R9 showed the catalytic activity of proton reduction and unobvious catalysis on water oxidation, but its isomer R10 showed the opposite activity. When pH < 8 in aqueous solution, R9 is reduced by an electron

The Application of Pincer Ligand in Catalytic Water Splitting

395

Fig. 12 Structures of p-[Ru (tpy)(pynap)(OH2)]2+(R9) and its isomer d-[Ru(tpy) (pynap)(OH2)]2+(R10)

Fig. 13 Structures of [(tpy) (bpy)RuII(NCCH3)]2+ (R11) and [(tpy)(bpy)RuII(H)]+ (R12)

and followed proton coupling to form intermediate [Ru(tpy)(pynapH)(OH2)]2+. In acetonitrile, R9 displays enhanced catalytic current for reduction proton with the addition of acetic acid. However, this phenomenon has not been observed for R10. These preliminary results clearly show that the proton relay should be in appropriate position. The catalytic mechanism has not been studied in detail in this work. Meyer [114] studied in detail the electrocatalytic reduction of water to hydrogen with its analogues [(tpy)(bpy)RuII(NCCH3)]2+ (R11) and [(tpy)(bpy)RuII(H)]+ (R12) (Fig. 13). Based on the experimental evidence, the mechanism of catalytic water reduction of R11 in acetonitrile was proposed (Scheme 4(1)). Firstly, tpy ligand in R11 was reduced to [(tpy)(bpy)RuII(NCCH3)]+ by one electron, and then bpy ligand was reduced to intermediate [(tpy)(bpy)RuII(NCCH3)]0 by one electron. This is followed by a chemical process in which the acetonitrile molecule leaves and is bound to a proton in the water to obtain R12; The active species [(tpy) (bpy)RuII(H)]0 is obtained by electroreduction of tpy ligand, and then combined with a proton in water to form (tpy)(bpy)RuII(H2)]+ with H2 ligand; It then releases hydrogen back to [(tpy)(bpy)RuII(NCCH3)]+ or further accepts one electron to get [(tpy)(bpy)RuII(H2)]0 and then releases H2 back to[(tpy)(bpy)RuII(NCCH3)]0 to complete a cycle. The kinetic isotope effect analysis shows that kH2O/kD2O is 3.7, which clearly demonstrates that the attack of water molecules on [(tpy)(bpy)RuII(H)]0 is the rate-limiting step. When weak acid H2PO4 is added to the system as proton source, the hydrogen production efficiency will increase by four orders of magnitude, because H2PO4 is 8 pKa units smaller than water, which is easier to give protons. In this regard, the mechanism of proton reduction to hydrogen by R11 in the presence of H2PO4 was proposed (Scheme 4(2)); R11 is electronically reduced to [(tpy)(bpy)RuII(NCCH3)]+, then an electron is obtained and [(tpy)(bpy)

396

H.-T. Zhang and M.-T. Zhang

Scheme 4 (1) Proposed mechanism for electrocatalytic reduction of water by [(tpy)(bpy)RuII(NCCH3)]2+ (R11) (S ¼ CH3CN); (2) Proposed mechanism for electrocatalytic reduction of proton from H2PO4 by R11 (S ¼ CH3CN); Reprinted with permission from Ref. [114]

The Application of Pincer Ligand in Catalytic Water Splitting

397

Fig. 14 Structures of [NiIICl(C6H3-2,6(OPPh2)2)] (R13) and [NiIICl(tBuPCP)] (R14)

RuII(NCCH3)]0 is formed; The rate-limiting step is to capture a proton to form R12 by interaction with H2PO4; Further, a proton of another H2PO4 is captured and H2 is released back to R11. Non-noble metal nickel-based catalysts of hydrogen evolution have been widely developed, such as the DuBois catalysts [54, 58, 115]. In 2012, Crabtree group [116] investigated the electrocatalytic properties of two mononuclear nickel compounds R13–14 (R13 ¼ [NiIICl(C6H3–2,6-(OPPh2)2)]; R14 ¼ [NiIICl(tBuPCP)]) containing PCP pincer ligands as proton reduction catalysts (Fig. 14). In 0.1 M NBu4BF4 acetonitrile solution, two catalysts have been proved that both of them can be used as effective proton reduction catalysts by electrochemical studies. Catalyst R14 has a TOF about 209 s1, which is faster than R13 (TOF ¼ 54.6 s1) under the same reaction condition. The Faraday efficiency of the catalysts for hydrogen production was estimated to be about 90% by the means of bulk electrolysis and quantitative detection of hydrogen. For catalyst R14, two possible intermediates R8 and R15 with catalytic activity coordinated with CH3CN and H were separated and detected, and the reaction mechanism of hydrogen production catalyzed by R15 was proposed by DFT calculation (Scheme 5). DFT calculation showed that R14 loses Cl to form R15 in acetonitrile, and then acetonitrile molecule leaves after one-electron reduction and combines with proton to form NiIII-H, then further reduces to R8, and then the hydride reacts with proton to release hydrogen back to the initial state. They believe that the initial active species is the solvated complex R15; Then the acetonitrile molecule is lost with the combination of proton to form NiIII-H species by one-electron reduction; After further reduction, the hydride species R8 reacts with one proton to release hydrogen back to its initial state [116]. Concurrently, a mononuclear nickel catalyst R16 ([NiII{2,6-(Ar1NCMe)2C5H3N} Br2] (Ar1 ¼ 2,6-dimethyl phenyl), Fig. 15) based on redox-active pincer ligand was developed [117]. The catalyst R16 can effectively reduce proton to hydrogen by electrochemical method, whether in acetonitrile or aqueous solution. In CH3CN, R16 shows a lower overpotential of 140 mV at current density of 1 mA/cm2, comparing thermodynamic potential of 84 mV vs NHE under the same conditions. The TOF is calculated to be 65 h1 and the Faraday efficiency of hydrogen production is about 95% by bulk electrolysis in 0.1 M KCl/HCl solution (pH ¼ 1). DFT calculation shows that the starting state of catalyst is NiII(NNN)(OH2)2+, which has a water molecule coordination rather than the original two Br. Then the pincer ligand is reduced by one electron to form NiII(NNN)(OH2)+, followed by once proton-coupled electron transfer (PCET) reduction and removal of water molecule to obtain NiII(NNN)(H)+. Finally, the intermediate binds with a proton and releases hydrogen

398

H.-T. Zhang and M.-T. Zhang

Scheme 5 Proposed catalytic cycle of proton reduction supported by DFT calculations using R15 formed by R14 loss of Cl in MeCN; Reprinted with permission from Ref. [116]

back to NiII(NNN)(OH2)2+ (Scheme 6). The activation energy of the catalytic process is greatly reduced by PCET, which makes an important contribution to the low overpotential. A few years later, Groysman group [118] synthesized a series of Ni-based catalysts similar to the R16 framework by ligand modification. R17–18 (R17 ¼ 2,6-(Ar2N¼CH)2C5H3N, Ar2 ¼ 2,6-diisopropyl phenyl; R18 ¼ 2,6(Ar3N¼CH)2C5H3N, Ar3 ¼ 2,4,6-trimethyl phenyl, Fig. 15) were found as CO2 catalysts in the catalytic process. The catalyst has high selectivity for reducing proton hydrogen production. Through a continuous one-electron reduction reaction, they were reduced to form an active intermediate which catalyzes the evolution of hydrogen. For catalytic proton reduction, the catalysts have not been subjected to subsequent studies. Effective strategy for realizing multi-electron reactions of metal compounds by incorporating redox active parts into ligand skeleton. In 2014, Crabtree group [119] developed two novel redox-active ligands (N5 ¼ 2,6-di(1,8-naphthyridin-2-yl)

The Application of Pincer Ligand in Catalytic Water Splitting

399

Fig. 15 Chemical structures of [NiII(NNN)Br2] type catalysts R16–18

Scheme 6 Catalytic cycle based on formation of a NiII hydride intermediate by PCET proposed based on calculations using NiII(NNN)(OH2)2+ came from conversion of R16; Reprinted with permission from Ref. [117]

pyridine, N7 ¼ 2,6-di(1,5,8-triazanaphthalene-2-yl)pyridine) which were tpy derivatives. Based on these ligands, nickel-based catalysts R19–20 (R19 ¼ NiII(N5)Br2, R20 ¼ NiII(N7)Br2, Fig. 16) were synthesized for water reduction. By electrochemical comparison with Zn-based complexes containing same ligands, the potential of Ni (II) to Ni (I) in R19–20 were 1.05 V and 0.51 V (vs NHE), respectively,

400

H.-T. Zhang and M.-T. Zhang

Fig. 16 Chemical structures of catalysts R19–20 with N5, N7 ligands

Fig. 17 Structures of catalysts R21–22 studied by He group

though the evidence is insufficient. The catalytic activity of the two catalysts for water reduction was tested, and the dramatic difference between the two catalysts was found under the same conditions. In KCl/HCl (pH ¼ 1), the turnover numbers of controlled potential electrolysis R19–20 were 55 and 43 at 0.5 V (vs NHE), respectively; In phosphate buffer (pH ¼ 4), the TON of R19 was three times than R20 at 0.8 V (vs NHE); Interestingly, in neutral phosphate buffer, R20 exhibited excellent catalytic performance 30 times higher than R19. No explanation was given for this interesting experimental result. In terms of stability, only blank experiments were carried out on electrolytic electrodes, while no detailed study was proceeded on the real catalytic process. In 2015, He group [120] synthesized Ni-based R21 and Co-based R22 catalysts (Fig. 17) using pincer-like HthioP (2-(2-diphenylphosphino) benzylidenehydrazinecarbothioamide) ligand. The HthioP ligand with redox activity can stabilize the intermediates in the catalytic process. Luminescence titrations showed that R21 could be used as an effective fluorescent quencher when fluorescein was used as a photosensitizer. There is an opportunity for R21 to improve its reducibility under photocatalytic conditions, which can effectively reduce proton to hydrogen production. Hence, under the conditions of fluorescein as photosensitizer and triethylamine used as sacrificial agent, R21 has the capability of efficient photocatalytic hydrogen generation. The catalyst exhibited excellent catalytic activity, TON reaching 8,000 after 24 h, and the initial TOF was greater than 500 h1. It can be seen from single crystal that the geometric structure of Ni atom is square planar and Ni atom interacts with hydrogen atom in the ligand to featuring Ni•••H interactions. During the catalytic process, the proton is transferred to the empty site to form hydride, which is also confirmed by DFT calculations. For catalyst R22, the cobalt atom is surrounded by an octahedral geometry and coordinated by two HthioP ligands. R22 is considered to be the most active hydrogen evolution catalyst under Fl-cobalt systems, with initial TOF of 200 h1 and TON of 2000.

The Application of Pincer Ligand in Catalytic Water Splitting

401

Fig. 18 Catalysts R23–24 with NS2-type ligand studied by Castillo and coworkers

Castillo and coworkers [121] also reported a novel aminobis(thiophenolate) pincer-type ligand NS2, in which bis(trifluoromethyl)phenyl substituent, a steric bulky and electron-withdrawing substituent were introduced. The ligand provides electronic plasticity to metal center. Interestingly, they used the ligand to synthesize a dimeric nickel complex R23 ([NiII(NS2)]2, Fig. 18). Despite the coordination of reductive S2, the complex maintains the corresponding skeleton structure in the presence of O2 or I2, due to the introduction of high steric hindrance bis (trifluoromethyl)phenyl, which provides steric protection for the metal center from the formation of metal oxides. UVvis spectroscopy shows that the dimer also keeps the corresponding stability under acidic conditions and is not degraded by protonation. By cyclic voltammetry, it is found that there is a reversible redox process at 0.47 V and 1.66 V (vs Fc+/0), respectively. It is considered that [NiIIINiIII] and [NiINiI] species are formed by two-electron oxidation and reduction, respectively. This conclusion is obtained by comparing corresponding monomer R24 (NiII(NS2) (2,6-Me2C6H3NC), Fig. 18) and dimeric Zn complex with the same ligand. On the reduction, an enhanced current appeared at 1.66 V (vs Fc+/0) after adding HBF4 to the system, and the reversible redox peak became irreversible, which indicates that R23 could effectively catalyze proton reduction. In addition, R23 can also electrocatalyze the oxidation of hydrides such as NaBH(OAc)3, which gives the catalyst a hydrogenase-like redox activity. In 2016, Wang group developed an atypical pincer-based copper-based catalyst R25 (CuII(Cl-TMPA)Cl2) using four-teeth Cl-TMPA ligand [122]. It is the first copper-based catalyst for water reduction. Uncoordinated pendant pyridine may play the triple roles: a monodentate ligand dissociation trigger, proton relay, and electronic relay. The catalyst exhibits significant photo-catalyzed water reduction and hydrogen evolution. With [Ir(ppy)2(dtbpy)]Cl as photosensitizer and triethylamine as sacrificial reductant in CH3CN/H2O (9/1,v/v) solution, the turnover number can reach 10,014 within 6 h, and the photo-quantum yield is estimated to be 5.6%. In 2016, Chidsey and coworkers [123] developed R26–29 (R26 ¼ NiII(POCOP) (H); R27 ¼ NiII(POOMeCOP)(H); R28 ¼ NiII(POMeCOP)(H); II ditBu R29 ¼ Ni (PO COP)(H), Fig. 19), a nickel-hydrogen compound containing different pincer ligands, which was synthesized by reacting the corresponding NiCl species with LiAlH4. Although metal hydrides show high activity, these three compounds were relatively stable and their crystal structures were obtained by X-ray. By cyclic voltammetry and DFT calculation, the hydrides did not release electrons and protons during electrochemical oxidation, but evolved hydrogen with

402

H.-T. Zhang and M.-T. Zhang

Fig. 19 Catalysts R26–29 with POCOP pincer ligands and H studied by Chidsey and coworkers

Scheme 7 Mechanisms for hydrogen production of R26–28; Reprinted with permission from Ref. [123]

energy release. Two electrocatalytic mechanisms, ErErCi (Electrochemical-Electrochemical-Chemical process) and ErCiEr’ (Electrochemical-Chemical-Electrochemical process), have been proposed (Scheme 7). ErErCi can be summarized as follows: NiIIIH+ species was obtained from NiIIH species through one-electron oxidation process, and then two NiIIIH+ molecules were chemically reacted with the participation of solvent molecules to form two NiII-NCCH3+ and release hydrogen. However, the ErCiEr’ process is that the NiIIH species were oxidized to NiIIIH+ species by one electron, and then NiIIIH+ reacted with another molecule NiIIH to obtain NiI-NCCH3 and NiII-NCCH3+ via a chemical process combining with solvent molecule and releasing hydrogen; The low-valent NiI-NCCH3 from the last step was then oxidized to NiII-NCCH3+ by one electron. Experiments showed that these metal hydrides R26–29 oxidative H2 evolution was through the bimolecular mechanism. If these metal hydrides are distributed and anchored on solid support in fuel cells, the bimolecular oxidation mechanism will be inhibited, resulting in the release of protons and electrons. How to use metal hydrides as oxidative H2 evolution catalysts was still a great challenge.

The Application of Pincer Ligand in Catalytic Water Splitting

403

Fig. 20 Pd-based catalysts R30–32 studied by Lawrence and coworkers

In 2017, Lawrence and coworkers [124] reported a catalyst R30 (PdII2Cl3(κ5-Npy, Nim,S,Npy,Nam-dpktsc-H), Fig. 20) containing two Pd atoms. It contained an unconventional pincer ligand, in which one Pd atom was in a pincer coordination environment of NNS, while the other one coordinated with two nitrogen atoms in the ligand. The crystal of R30 containing a water molecule was obtained by standing in DMF. The electrocatalytic performance of the catalyst for proton reduction was tested by adding acid in organic phase. The reduction peak at 0.8 V (vs Ag+/0) was designated as the catalytic peak for proton reduction. After analysis, with p-TsOH as proton source, the TOF of proton reduction catalyzed by R30 was 65 min1 within 20 min, and the corresponding TON was about 1,300. The highlight of this catalyst was its low overpotential, only 180 mV. In the following year, the group developed mononuclear “k3-SNS” R31 (PdIICl(pdcta), Fig. 20) and “k3-SNN” R32 (PdIICl (pbcta), Fig. 20) catalysts by using two known pincer ligands [125], pdcta and pbcta, respectively. The two catalysts continued the good electrocatalytic hydrogen evolution effect of R30. Within 20 min of DMF solution, TONs of R31 and R32 catalyzed H2 evolution were 8,000 and 5,300, and TOFs were 400 min1 and 265 min1, respectively. Their brilliant features were that they have lower overpotential of 97 mV and 156 mV, respectively. Cyclic voltammetry showed that Pd(II) was reduced to Pd(0) by two electrons in one step in the process of proton reduction catalyzed by R32, and then hydrogen was generated by Pd(0) in the process of proton reduction. This is different from the catalytic process of R31. In the catalytic process of R31, Pd(I) was formed firstly and then converted into Pd(0) by twice one-electron reduction, and Pd(I) also can be used as an active species to catalyze proton reduction. At present, these Pt-, Pd-based and other noble metal catalysts still have advantages in catalytic water reduction to hydrogen. In 2018, two binuclear nickel-based water reduction catalysts R33–34 (R33 ¼ [NiII2(PyC2S)2]Br2; R34 ¼ [NiII2(PyC3S)2]Br2, Fig. 21) containing different thiolate-functionalized N-heterocyclic carbene (NHC) ligands were synthesized by Bouwman group [126]. The sulfur group on the ligand acts as a bridge between the two nickel atoms. Fortunately, the structures of the two catalysts had been characterized by X-ray, and the nickel centers were in a distorted square-planar geometry. CV and controlled-potential coulometry experiments were carried out to investigate the intrinsic redox properties of R33–34 and their electrocatalytic proton reduction activity. The results showed that the performance of two catalysts was unsatisfactory, because the TOFs of both catalysts were not more than 0.1 min1 within

404

H.-T. Zhang and M.-T. Zhang

Fig. 21 Dimer Ni-based catalysts R33–34 studied by Bouwman and coworkers

Fig. 22 Ni-based catalysts R35–37 with tBuPNEtN ligand studied by Fan group

10 min. In addition, electron-rich sulfur ligands is unfavorable for the stability of catalysts in the presence of oxygen, so these factors limited their potential application as hydrogen production catalysts. The “Milstein catalysts” [104] mentioned above provide a valuable example for the design and development of catalysts for water splitting, in which PNN pincer ligand played an important role in the catalytic process. Based on this ligand, Fan group [127] had developed three Ni-based catalysts R35–37 in 2018 (Fig. 22). All of three catalysts contained the same PNN ligands, but the difference lied in the monodentate ligands (ONO2 R35, Cl R36, SC6H4Me R37) coordinated on nickel atoms. All three catalysts exhibited good stability in the presence of oxygen and had the ability to catalyze proton reduction of acetic acid and trifluoroacetic acid in acetonitrile. However, the crossover waveforms of CVs for catalytic proton reduction by R35 and R36 may be due to the deposition and nucleation of the catalysts on the surface of the electrode. When the electrode was placed in fresh solution, it also showed a certain catalytic current. These phenomena indicated that in the process of electrocatalytic proton reduction, some protons were reduced by heterogeneous reaction on the electrode surface. However, catalyst R37 showed completely homogeneous catalytic behaviors in the same condition. With acetic acid as proton source, the peak of catalytic reduction potential was observed at 0.2 V in acetonitrile solution, and the corresponding overpotential was 0.56 V. However, when trifluoroacetic acid was used as proton source, the catalytic reduction potential was 1.43 V and the corresponding overpotential was 0.54 V. Based on the experiments of variable scan rates, the following mechanism of proton reduction catalyzed by R37 was proposed (Scheme 8): Firstly, the reversible dissociation and coordination of catalyst R37 with monodentate sulfur ligand in solution phase resulted in the release of [(PNN)Ni]2+; Then [(PNN)Ni]2+ was converted to [(PNN)Ni]+ by an electron reduction; After electron-rich [(PNN)Ni]+ bound a proton to form NiIIH

The Application of Pincer Ligand in Catalytic Water Splitting

405

Scheme 8 Proposed mechanism for the electrocatalytic proton reduction process using complex R37 as the example. Reprinted with permission from Ref. [127]

Fig. 23 Structure of catalyst R38 with PNP-C2Me ligand

species, NiIIH species was further reduced by one electron to form NiIH species; Finally, one proton was obtained to form and release hydrogen back to [(PNN)Ni]2+. Recently, Brueggeller group [128] developed a Pd-based catalyst R38 (PdIICl (PNP-C2-Me), Fig. 23) based on PNP ligand, which anchors Pd like pincers and forms a square-planar geometry through the coordination of Cl with Pd. Firstly, the photo-driven proton reduction of R38 was studied. The catalyst was irradiated for 360 h to hydrogen evolution. After the poor catalytic performance of the first 100 h, the rate of hydrogen evolution suddenly increased dramatically, and finally TON could reach nearly 400. When photosensitizer was added continuously in 1,245 h of catalysis, TON could reach 1986. However, the stability of the catalyst under irradiation had not been concerned. From the experimental phenomena, catalyst R38 may undergo an activation process to produce a real active intermediate in the photocatalytic process, which does not exclude the possibility of heterogeneous catalysis. In addition, the catalysts were electrochemically experimented in anhydrous acetonitrile. The compound showed two weak reversible reduction waves, Pd2 +/+ and Pd+/0. When 2 M H2O was gradually added into acetonitrile, the current at +/0 Pd reduction peak (2.4 V vs Fc+/0) increased gradually. It showed that the catalyst can catalyze proton reduction from water to produce hydrogen without some strong acid as proton source, although the overpotential was high.

2.2

Models of Hydrogenase

Through the introduction of Sect. 1.2, we know that [NiFe]-hydrogenase is one of the hydrogenases. It can catalyze proton reduction to produce hydrogen, and also can activate hydrogen molecules by heterolysis. The study of [NiFe]-hydrogenase is of

406

H.-T. Zhang and M.-T. Zhang

Fig. 24 Structures of [NiFe]-hydrogenase and model R39

great significance for hydrogen production and storage. It was not until 1996 that the specific structure of [NiFe]-hydrogenase activity center was confirmed by X-ray crystallography (Fig. 24) [129–132]. The structure showed that the nickel atom was coordinated by four CysS, iron by two CysS, two CNs, and one CO. Through many years of research, it has been found that only the valence of Ni atom has changed in the catalytic process of [NiFe]-hydrogenase. Therefore, it is believed that nickel center is the real catalytic active center [133, 134]. The simulation of the structure and function of natural [NiFe]-hydrogenase by synthetic metal complexes has become the research interest of many researchers. Lots of sulfur-rich nickel-based and palladium-based complexes containing ‘S3’ type pincer ligands were synthesized by Sellmann group as early as 1999 [135, 136]. D2 O þ H2 Ð HD þ HDO

ð5Þ

In 2000, the group [137] reported the first nickel thiolate complex [Ni(NHPnPr3) (‘S3’)] R39 (Fig. 24), which successfully simulated the structure and function of nickel site in [NiFe] hydrogenase. With pincer-type ‘S3’ [‘S3’2 ¼ bis (2-sulfanylphenyl)sulfide(2-)] as the framework, Ni atom is coordinated with NHPnPr3 to form a four-coordinated NiII center surrounded by one N atom, one thioether, and two thiolate donors. The [NiNS3] geometry of R39 is neither tetrahedral nor planar, but strongly distorted, and is similar with [NiS4] in natural [NiFe] hydrogenase. When reacted with D2 at slightly elevated pressure (10 bar), [Ni (NHPnPr3) (‘S3’)] is converted into [Ni(NDPnPr3) detected by H/D NMR with the release of HD. Further, when R39 reacts with D2O, [Ni(NDPnPr3) is also detected accompanied by the formation of HDO. Combination of the two reaction sums up to successfully establish the H2/D+ exchange experiment (hydrogenase activity test reaction (Eq. 5) [138]). The feasible mechanism of the H2/D+ exchange experiment is proposed as shown in Scheme 9, in which the NHPnPr3 ligand of R39 becomes an important probe for the reaction mechanism. The experimental results clearly show that the heterolysis of D2 is catalyzed by R39, which further shows that it can effectively split and release hydrogen. The author believes confidently that “The H2 heterolysis catalyzed by [NiFe] hydrogenases can take place at a NiII center coordinated by cysteinate ligands” [137]. More than 10 years later, Reisner group [139] reported the first series of mononuclear nickel models R40–42 containing Se substituted sulfur ligand for the structural and functional simulation of [NiFeSe] hydrogenase active center

The Application of Pincer Ligand in Catalytic Water Splitting

407

Scheme 9 Mechanism of the R39 catalyzed D2/H+ exchange according to Eq. (5). Reprinted with permission from Ref. [137]

Fig. 25 Structures of models R40–43

(Fig. 25). At the same time, model R43 was synthesized without seleniumsubstituted. The electrocatalytic hydrogen production of four models was studied, but it was disappointing that although there was an obvious enhanced current in the catalytic process, it was not homogeneous catalysis. Detection by SEM and EDX, it was determined that solid particles were continuously deposited on the surface of the

408

Scheme 10 Syntheses of new model R44 from (C6F5)4]. Reprinted with permission from Ref. [141]

H.-T. Zhang and M.-T. Zhang

(tBuPCP)NiIICl

with[(Et3Si)2H][B

electrode during the catalytic process, and these deposits were the real active substances. These molecules existed only as catalytic precursors. Activation of hydrogen is an important function of natural [NiFe] hydrogenase. It is a common method to simulate this function by artificial models. Although many models [140] can activate hydrogen as well as [NiFe] hydrogenase, the study of activation mechanism is still incomplete and the capture of key intermediates is still lacking. Heinekey and coworkers [141] reported a crystal structure of nickel dihydrogen complex R44, which filled the gap in the key intermediate of Ni-H2 (Scheme 10). (tBuPCP)NiCl containing PCP pincer ligand reacts with the chlorideabstracting reagent [(Et3Si)2H][B(C6F5)4] to form an intermediate containing a open site. The intermediate reacts with 1 atm H2 to obtain R44 which was characterized by X-ray and NMR. The distance between the two hydrogen atoms in the single crystal is 0.55 Å, which may be due to thermal rotation of the H2 ligand [142]. Adding triethylamine to the solution of R44 can promote the heterolysis of hydrogen ligand and obtain Ni-H species (tBuPCP)NiH, which is characterized by NMR. The determination of Ni-H2 structure provides important evidence for understanding how [NiFe] hydrogenase binds hydrogen molecules. [Fe]-hydrogenase is also a kind of natural enzymes that can activate hydrogen [143–145]. After activating hydrogen, [Fe]-hydrogenase can act as a hydride reagent to transfer hydride to methenyl-H4MPT+, which is the intermediate step of reducing CO2 to CH4 in nature. Since the structure of [Fe]-hydrogenase has been determined [146], various mononuclear iron models [147–155] have been synthesized. In 2012, Song group [156] reported a series of single iron model R45–47 containing acylmethyl pincer ligand (Scheme 11). R45 and R46 were synthesized by the reaction of Fe (0) species with Br2 or I2, respectively. And both models could react with KSCOMe to obtain R47. A common feature of these three models is that they all contain a hydroxyl ligand, which is very similar to the hydroxyl

The Application of Pincer Ligand in Catalytic Water Splitting

409

Scheme 11 Syntheses of new models R45–47. Reprinted with permission from Ref. [156]

structure of [Fe]-hydrogenase. The uncoordinated hydroxyl group of [Fe]-hydrogenase exists as an intramolecular lewis base to accept proton generated from hydrogen heterolysis. The synthesis of these three models is great significance for simulating and understanding the role of hydroxyl groups in hydrogen activation. Kirchner group reported that Fe(II) hydride can be generated by activating hydrogen in hydrogen atmosphere by bis-carbonyl Fe(II) complex with Zn as reductant [157]. Strictly speaking, this complex does not belong to the model of hydrogenase, nor has it been studied as a proton reduction catalyst, so it is not described in detail. A key intermediate in the catalytic cycle of hydrogen activation by [Fe]-hydrogenase is considered to be Fe(II) hydride, although it has not been found in any enzymatic studies. Therefore, it is of great significance to identify and study this active intermediate. In 2016, Rose group [158] synthesized an iron hydrogenase model R48 with FeH moiety using CNS pincer ligand (Scheme 12). The model is obtained by the reaction of R49 with NaHBEt3 containing Br coordination. It is extremely unstable and can be detected by cryogenic nuclear magnetic resonance only when the temperature is below 40  C. When the temperature is higher than 40  C, CH3S is released by breaking the CS bond and intramolecular hydride transfer, and the model becomes a highly active fivecoordination intermediate R50 containing CNC pincer ligand. By adding PPh3 to capture the intermediate R50, complex R51 containing two PPh3 was obtained. These reactions provide much key information for understanding and imitating the activation of hydrogen by [Fe]-hydrogenase, and provide some ideas for designing more importantly active models. In 2017, the group [159] carried out detailed research on reactivity of R49, and synthesized a series of models by substitution reaction (Scheme 13). When large bulk NaBArF4 exists, ligand Br in R49 will be replaced by PPh3, PMe3, Py, P(OEt)3, or tBuNC to form R52–56. When AgBF4 is used, the ligand Br can be replaced by CH3CN to produce R57. Br can also be substituted by stronger ArS-ligand to produce R58. The models R49,52,54 containing two CO ligands can be oxidized by Me3NO, removed one CO and combined with PPh3 or Py to form R59–61 containing one CO ligand, respectively. These experiments provide a synthetic route for us to synthesize models with more different substituents, and also make these models more widely used.

410

H.-T. Zhang and M.-T. Zhang

Scheme 12 R49-led reaction to synthesize R48, R50–51. Reprinted with permission from Ref. [158]

Scheme 13 A series of [Fe]-models R52–61 were synthesized starting from R49. Reprinted with permission from Ref. [159]

The Application of Pincer Ligand in Catalytic Water Splitting

411

Fig. 26 Analogous iron hydrogenase models R62–76 with SCN-type pincer ligand

[FeFe]-hydrogenase is a natural enzyme with a catalytic efficiency of 9,000 s1 about hydrogen evolution [38], and its active center is a binuclear iron structure with polycarbonyl coordination. In 2012, Hirotsu group [160] synthesized a series of [FeFe]-models R62–66 (Fig. 26) containing SCN-type pincer ligand by photoinduced C-S bond breaking of dibenzothiophene derivatives [161]. Here, S atom is connected to two iron atoms as a bridge group, and the models were characterized by X-ray and NMR. The electrochemical behavior of the five models was studied. There are two reversible one-electron reduction processes at 1.25 V and 1.90 V (vs Fc+/0), which were assigned to be the processes of [Fe2]0/ and [Fe2]/2respectively. Acetic acid as the proton source, R62, R64 and R66 show apparent catalytic current at 2.0 V (vs Fc+/0). With the increase of acetic acid, the catalytic current increases linearly, which indicates that they can effectively reduce proton. Apparently, the overpotential of catalysis is too high and needs further optimization. Hence, in 2015, the group fine-tuned the apyDBT ligand to synthesize diiron models R67–69 (Fig. 26) whose pyridine rings were substituted by amino group at different position [162]. It is hoped that the catalytic differences of proton reduction can be explored by introducing pendant amino groups in different positions. Under the same conditions, electrochemical studies showed that the catalytic potential of R67 with o-NH2 is more positive than the other two. It is suggested that the amino group introduced in the models may act as a proton acceptor. When the amino group is proximal to the metal iron center, it is beneficial to the formation of HH bond

412

H.-T. Zhang and M.-T. Zhang

between the proton and the hydride on the iron center, thus improving the efficiency of hydrogen evolution. These models can also react with PMe2Ph to form the corresponding mononuclear iron models R70–73 with phosphine ligand [162, 163]. In addition, based on dibenzothiophene derivatives, the team has developed some other similar models R74–76 [163], which will not be discussed here (Fig. 26).

2.3

Assembled Catalysts

In order to effectively transfer the electrons of photo-induced charge separation to the active sites of catalysts, a method of covalent bonding between photosensitizers and catalysts was proposed to promote the transfer of electrons in the catalytic process. In 2003, Sun group [164] developed an assembled catalyst R77 (Fig. 27) by covalently bonding a model of [FeFe]-hydrogenase with a photosensitizer containing [Ru(tpy)2]2+ to release hydrogen through photochemical catalysis of proton reduction. The assembly of R77 was synthesized by the reaction of terpyridine-coupled dinuclear iron complex (which was characterized by X-ray) with [RuCl2(terpy)]DMSO under dark and low temperature conditions. The reason why [Ru(tpy)2]2+ type photosensitizer was chosen is that the tridentate tpy has stronger chelating effect on metals, which makes [Ru(tpy)2]2+ have better geometric structure than other ruthenium compounds containing bidentate ligands [165]. The acetylenic linker, which links the photosensitizer with azapropylene-bridged diiron species which is the first model of N-atom linked to an aromatic system, not only

Fig. 27 Structures of assemblies R77–79 with RuII-based photosensitive parts

The Application of Pincer Ligand in Catalytic Water Splitting

413

accurately controls the spatial distance, but also increases the lifetime of the excited Ru state. In the following year, the photophysical properties of the assembled R77 were studied [166]. Comparing with the non-substituted [Ru(tpy)2]2+, it is found that the lifetime of the excited states of R77 or [Ru(tpy)2]2+ containing phenylacetylene tether substituted by different alkyl groups is greatly enhanced by at least one order of magnitude. The driving force required for the transfer of electrons from the excited ruthenium to the dirron portion can be calculated by using the energy of the excited state and electrochemical data, which is uphill by 0.59 eV. The reason is that the protonation of N atom on ADT bridge (ADT ¼ azadithiolate (SCH2NRCH2S)) will make the reduction potential of the assembly R77 shift more positively, so that the electron transfer is more thermodynamically advantageous. The disadvantage of this series of research is that the exact single crystal structure information of R77 has not been obtained, although R77 has been indirectly demonstrated by other spectroscopic methods such as NMR, IR and MS. More regrettably, there is a lack of in-depth study on the catalytic properties of the assembly. Cobalt complexes are a kind of cheap and efficient molecular catalysts with low overpotential, which can effectively catalyze proton reduction under light-driven conditions [28, 46, 47, 167]. In 2010, Tiede and coworkers [168] hoped to explore the relationship between structure of different supramolecular R78–79 (Fig. 27)and photo-induced electron transfer in catalytic proton reduction. The assemblies are composed of different Ru-based photosensitizers and cobaloxime catalyst. In this study, the differences between assemblies and fragments were compared by solution-phase X-ray scattering and ultrafast transient optical spectroscopy analyses; further, the self-assembly process of assemblies was analyzed. Notably, the photosensitizer with push–pull effect in the assembly makes the excited state Ru center quenched rapidly by the Co-based part after illumination, thus greatly increasing the chances of solar energy converting into hydrogen energy.

3 Molecular Catalysts with Pincer Ligand for OER 3.1

Noble Metals-Based Water Oxidation Catalysts

Hundreds of molecular ruthenium-based water oxidation catalysts [40, 61] had emerged since 1982 when Meyer group reported the first water oxidation catalyst containing binuclear ruthenium named “blue dimer” [169–171]. Under the condition of pH ¼ 7, the thermodynamic driving potential for water oxidation is as high as 1.23 V (vs NHE). At such a high potential, many ligands containing P and S are difficult to avoid being oxidized, so most of the catalysts for water oxidation use N-based ligands such as polypyridine. As a result, pincer-type tpy and its derivatives ligands have been widely developed and applied to water oxidation catalysts. In 2000, Tanaka and coworkers reported the first Ru-based water oxidation catalyst O1 [172] containing tpy units. When the catalyst was anchored on the

414

H.-T. Zhang and M.-T. Zhang

Fig. 28 Structure of the [Ru2(OH)2(3,6-tBu2qui)2(btpyan)]2+ complex O1 containing the redoxactive quinone ligand, the analogous O2-O4

surface of ITO electrode, and strong catalytic current can be observed at 1.7 V by electrochemical measurement. TON of catalytic water oxidation to oxygen was as high as 33,500 [173]. It was because of such excellent catalytic ability that researchers had great interest in its catalytic mechanism. Catalyst O1 contained redox active quinone ligand, which enabled the catalyst to catalyze water oxidation at a lower potential. Catalyst O2 obtained by substituting bipyridine for quinone was electrochemically studied, and its catalytic ability decreased sharply under the same conditions [173]. It was concluded that the presence of quinone ligands was an important part of the efficient catalytic oxidation of water. In order to explore the mechanism of OO bond formation in catalyst O1, the single-site analogue O3 [174, 175] and binuclear analogue O4 [176] with chlorine bridge were synthesized and studied. It was pointed out that compound O2 was basically inactivated under similar conditions, which indicated that two closely Ru centers were essential characteristic structures in the catalytic process. Another structural analogue O4 exhibited a certain catalytic activity, but it is far less than O1, which indicated that the face-to-face double RuOH provided an advantageous configuration for OO bond formation (Fig. 28). Since these binuclear ruthenium catalysts may form OO bond by the mechanism of I2M, whether the OO bond forming process could be regulated by fine-tuning ligands to explore the relationship between distance of Ru•••Ru and catalytic activity. Thus, Berlinguette and Nocera developed Ru-based catalysts O5–7 (Fig. 29), which orderly adjust the distance between the two Ru centers by adjusting the dihedral angles of the two tpy planes. Although all three compounds can catalyze water oxidation by chemical oxidation method under acidic condition, O5 with 5 Å distance between the two Ru centers had a larger kobs than the other two [177, 178]. In 2004, Llobet group reported a binuclear Ru catalyst O8 containing rigid pyrazole ligand framework (Fig. 30) [179]. Later, the catalyst was extensively mechanistically studied and well-recognized as a water oxidation catalyst [179– 187]. Two ruthenium centers were deliberately designed to be placed in close proximity, and the two Ru-OH2 moieties were in a cis fashion. This also resulted in the formation of OO bond between double Ru(IV)-oxos through the I2M way, which has been proved by extensive research. By chemical oxidation, compound O8 still suffers from a lower TON of 17.5, and exhibits a 70% efficiency relative to the oxidant CeIV. When catalyst O8 was anchored on the solid support surface, such as

The Application of Pincer Ligand in Catalytic Water Splitting

415

Fig. 29 Representation of the related dinuclear Ru complexes with different distance of Ru•••Ru

Fig. 30 Structures of the RuHbpp catalyst, [Ru2(OH2)2(bpp)-(tpy)2]2+ O8 and its derivatives O9–12, developed by Llobet and coworkers

FTO and SiO2, the catalytic performance was significantly improved. Under the optimum conditions, the TON of the catalyst for catalytic water oxidation to oxygen can reach 250. When the tpy ligand in the molecule is replaced by monodentate substituted-pyridine ligand to form analogous catalyst, the isotope labeling experiments demonstrate that the catalytic water oxidation is through WNA mechanism [187]. This result indicates that pincer-type tpy ligands provide tighter bimetallic cooperation for this family of catalysts. A few years later, Llobet and coworkers prepared a tetranuclear ruthenium-based catalyst O9–11 (Fig. 30) by linking double molecular O8 with the linkers, which can combine bpp-ligands by an ortho-, meta-, or para-substituted xylyl moieties. When excessive CeIV is employed as oxidant, tetranuclear catalysts O9–11 and their binuclear monomer O12 (Fig. 30) can catalyze water oxidation to oxygen with a similar efficiency, accompanying by the release of CO2. The ratio of [O2]/[CO2] proves the evidence that the carbon source of CO2 comes from linker’s methylene, and according to the amount of CO2

416

H.-T. Zhang and M.-T. Zhang

Fig. 31 Molecular structures of the single-site ruthenium catalysts O13–15

produced by four ruthenium-based catalysts, the difficulty extent of oxidizing methylenic moieties can be reasonably speculated [ 188]. Initially inspired by Mn4Ca-cluster, researchers designed a large number of binuclear and multinuclear catalysts, because people mistakenly believed that single-site catalysts could not catalyze water oxidation independently, and that catalytic water oxidation could be achieved at least by bimetallic cooperation. In 2005, Thummel group provided reliable evidence for the first time that mononuclear ruthenium catalyst O13–15 (Fig. 31) catalyzed the oxidation of water to oxygen [189], and then mononuclear ruthenium catalysts have been widely developed as mushrooms after rain. The group of catalysts O13–15 contains a tridentate polypyridine pincer ligand, accompanied by two naphthyridine nitrogens. In the structure of the catalyst, the nitrogen atom of the naphthyridine forms an intramolecular hydrogen bond with the coordinated H2O, which can stabilize the water molecule. The catalysts axially coordinate pyridine ligands substituted by different electronic substituents. Among them, catalyst O13 with the most electron-donating -NMe2 substitute did not show the best catalytic performance for water oxidation under the same conditions, but catalyst O14 with methyl substitution carried the highest catalytic capacity through four consecutive electrons losing. This might be due to the fact that dimethylamines are easily protonated, which greatly weakens their electron-donating ability in acidic conditions. Subsequently, Tummel’s team developed other groups of mononuclear ruthenium catalysts to study the relationship between stereoelectronic effects of ligands and catalytic activity (Fig. 32) [190, 191]. In 2008, a group of ruthenium catalysts O16–20 containing tpy and substituted bpy ligands were developed to explore the correlation between structure and activity by changing the electronic effect of substituents on bpy. It was found that the electron donor group of Me, OMe in catalyst O17, O18 had a lower RuIII/II potential, indicating that the introduction of electron donor group could stabilize the RuIII center. On the contrary, the O19, O20 with NO2 and CO2Et had a higher redox potential. Similarly, oxygen evolution experiments also revealed that O17–18 containing electron-rich groups had a higher catalytic rate than O19–20. It should have been inferred that compounds O17–18 containing electron-rich substituents have higher TONs, but the experimental results are dramatically opposite. Actually, catalyst O19 exhibits about three times TON than O17–18, which may be related to the stability of the catalyst in high oxidation

The Application of Pincer Ligand in Catalytic Water Splitting

417

Fig. 32 Molecular structures of the [Ru(tpy)(NN)Cl]+ complexes O16–26 and other complexes O27–29 without Cl ligand

state. Another group ruthenium catalyst O21–26 was then developed by replacing bpy ligand, which was replaced by a series of its derivatives or analogues. Bidentate nitrogen ligands with different structures have different stereoelectronic effects, which may give different catalytic effects to the catalysts. This is true, but only catalyst O21 with phenanthroline ligand exhibited catalytic activity in this group of catalysts. Catalysts O22–24 were considered to have a benzo-derivative ligand with high steric hindrance, which prevented RuCl from being exchanged by H2O, resulting in its inability to catalyze water oxidation. Compounds O25–26 were not easily oxidized by CeIV due to the protonation of bpm and bpz ligands under acidic conditions. In addition, ruthenium catalysts O27–29 containing six pyridinium nitrogen coordination were synthesized and its catalytic activity for water oxidation was investigated. Interestingly, compounds O27–28 with methylpyridine coordination have the ability to produce oxygen, while no oxygen was monitored for compound O29 with two tpy ligands.

418

H.-T. Zhang and M.-T. Zhang

Fig. 33 Single-site RuIIaqua complexes [Ru (tpy)(bpm)-(OH2)]2+ O30 and [Ru(tpy)(bpz)(OH2)]2+ O31 studied by Meyer and coworkers

Fig. 34 Structures of the [Ru(tpy)(bpy)(OH2)]2+-type complexes O32–41

Although many mononuclear ruthenium catalysts were initially developed, the mechanism was proposed by Meyer and his coworkers until 2008 [192]. When chloride in inactivated compounds O25–26 was replaced by water molecules, catalysts O30–31 were formed which can catalyze water oxidation (Fig. 33) [192]. The mechanism named WNA is proposed for the first time. Firstly, [RuIIOH2]2+ was oxidized by three eq. of CeIV continuously to lose three electrons and two protons to form [RuV¼O]3+. It was the electron-deficient RuV¼O that make it easy to convert into [RuIII-OOH]2+ by nucleophilic attack of water molecules, and then [RuIII-OOH]2+species was oxidized by one eq. CeIV and released oxygen back to the original [RuII-OH2]2+. Then the mechanism of catalysts O13–15 to oxidize water was studied again, which was proved that they also catalyzed the oxidation of water through single-site Ru center. The difference is that the onset potential of electrocatalytic water oxidation by complexes O13–15 was lower than the corresponding potential of [RuV/IV¼O], which indicated that the formation of OO bond is caused by directly nucleophilic attacking to [RuIV¼O]2+ by water molecules, thus avoiding the formation of higher valent [RuV¼O]3+ intermediate [193, 194]. In 2010, Berlinguette and coworkers developed a family of [Ru(tpy)(bpy)(OH2)]2 + -type catalysts O32–41 with different substituents (Fig. 34), which can provide insight into the effect of the electronic properties on the catalytic activity of the single-site ruthenium catalysts [195]. Cl/OH2 exchange in compounds was monitored by methods of NMR and UVvis spectroscopy, resulting in conversion of catalyst O16 to O32. In addition, it was found that the electronic properties of bpy ligands could affect the exchange rate of Cl / OH2. When there are electron-

The Application of Pincer Ligand in Catalytic Water Splitting

419

withdrawing groups in bpy, the rate of Cl replacement will slow down. Through the electrochemical study of this group of compounds, it was found that the influence of electronic effect on the redox potential of ruthenium centers by substituted tpy ligands was no more important than the modification of bpy. More interestingly, when the bpy ligand was introduced into the electron-withdrawing group, the catalytic rate of water oxidation decreased, but when tpy ligand was introduced into the electron-withdrawing group, the catalytic rate increased. In view of this phenomenon, it is difficult for researchers to give a regular summary [195, 196]. Next, through extensive mechanism studies of catalysts O32, O34, and O40, it was found that modification of bpy ligands would lead to greater catalytic difference of water oxidation, which was due to the axial para-position of RuO bond occupied by a nitrogen atom of bpy. The electronic density and reactivity of RuO bond will be directly affected by the change of electrical properties on bpy ligand. The mechanism of OO bond formation by high-valent [RuV¼O] species was also proposed for this group of catalysts [197, 198]. Isotope labeling experiments showed that only one O atom was labeled by 18O in generated O2 molecule and unlabeled O atom came from NO3- while WO catalysis was studied in H218O and [Ce(NO3)6]2+ as oxidant. So, this OAT (oxygen atom transfer) pathway of CeIV has also been suggested by many researchers [199, 200], thus this method provides us with an effective way to prove the catalytic mechanism. In 2010, Meyer and coworkers synthesized a series of mononuclear Ru-OH2 compounds O42–52 (Fig. 35) with pincer ligands as catalysts for water oxidation, and explored how the diversity of ligand environments affected the reaction rate and thermodynamics in water oxidation. Based on a large number of experimental data, a platform for the design of catalysts with controllable reactivity was provided [201]. In the subsequent paper, the electrocatalytic water oxidation by catalyst O42 with carbene ligand was studied in detail [202]. The experimental results of catalyst O42 supported the previously proposed WNA catalytic mechanism. It was believed that the multi-step PCET process resulted in the formation of highly active [RuV¼O] intermediate, which were attacked by H2O to form OO bond. Next, the [RuIII-OOH] species was oxidized and oxygen was released. By comparing catalyst O45, it was found that carbene ligands with strong electron donation ability can effectively increase the catalytic rate of water oxidation. The electrochemical performance also showed that the rate-limiting step of the whole catalytic reaction should be before the electrons are transferred to the electrodes. In other words, it may be a chemical process or diffusion process that becomes the limiting factor of the catalytic rate. In order to eliminate the influence of diffusion rate of catalyst molecule on the catalytic rate, a series of assemblies have been carried out to attach catalyst to semiconductor surface through covalent bond [203–211]. In 2017, Chen group employed a tridentate pincer ligand with carbene structure to synthesize two ruthenium-based catalysts O53–54 (Fig. 35) for water oxidation [212]. The tridentate ligand contains two strong electron donors, carbene and carboxyl, which greatly reduces the overpotential of catalytic water oxidation by O53–54. Therefore,

Fig. 35 Mononuclear RuII-OH2 complexes O42–54 studied by groups of Meyer and Chen

420 H.-T. Zhang and M.-T. Zhang

The Application of Pincer Ligand in Catalytic Water Splitting

421

in the three-component photo-induced water oxidation system, O54 has comparable catalytic activity with state-of-the-art catalysts, and TON can reach 273 [212]. In 2011, Llobet group synthesized single-site ruthenium compounds O55–58 (Fig. 36) with a structure similar to that of binuclear Ru-Hbpp catalyst O8 [213]. All four Ru-aqua compounds were equipped with a tpy ligand and a Hbpp ligand or its analogues. In organic solution, the electrochemical measurements of Ru-Cl species which were the precursors of catalysts O55–58 showed that only one RuIII/II redox couple existed, while RuIII/II and RuIV/III couples appeared in active complexes O55–58, which indicated that the existence of protons could make the metal centers reach higher oxidation state through PCET pathway. More importantly, the investigation indicated that the out isomer had lower redox potential than the corresponding in isomer. In the process of catalytic oxygen evolution, not only the formation of O2 was detected, but also the CO2 was also discovered in chemicaldriven water oxidation process. It was found that CO2 may originate from the oxidative decomposition of phenyl in catalysts O55 and O56. Because no CO2 was detected in the process of complexes O57 and O58 catalyzed water oxidation under the same conditions, this may be due to the increase of oxidation potential of pyridine rings after protonation under acidic conditions, which makes pyridine rings not easy to be oxidized and decomposed. In 2016, Llobet group first developed tri-heteronuclear Ru2M (M ¼ Mn, Co) precursors O59–O62 (Fig. 36) containing tpy ligands for water oxidation catalysis based on corresponding Ru-Cl species of catalyst O58 [214]. They can be converted into Ru2M-(H2O)4 (M ¼ Mn, Co) O63–O64 (Fig. 36) as real catalytic active substances in the catalytic environment. Compared with catalyst O58, the catalytic activity of Ru2M-(H2O)4 species O63– O64 in the process of catalytic oxygen evolution was greatly improved whether in photocatalysis or chemical oxidation. The results showed that the active center was still on metal Ru, and the O-O bond was formed by WNA mechanism. In 2011, Yagi and Fujita studied the catalytic differences of two geometric isomers O65–66 (Fig. 37) as catalysts for water oxidation. Isomer O66 was obtained by photoisomerization of O65 [113, 215]. Compared with Ru(tpy)(bpy) catalyst O32, compounds O65 and O66 containing pynap ligand have lower RuIII/II redox potential. However, the two geometric isomers O65 and O66 with such similar structures differ greatly in their electrochemical properties. It was found that the uncoordinated pyridinium nitrogen atom in isomer O66 was very close to the coordinating H2O molecule and formed intramolecular hydrogen bond, which increased the difficulty of producing high-valent Ru ¼ O and eventually led to its use as a poor catalyst for water oxidation. But it is the presence of intramolecular hydrogen bond that makes O66 a more active proton reduction catalyst than O65 [113]. In 2012, Thummel group synthesized a large number of [Ru(NNN)(NN)(X)] and [Ru(NNN)(pic)2(X)]-type ruthenium compounds O67–90 (Fig. 38) to study the relationship between catalyst structure and activity [216, 217]. When CeIV was employed as oxidant, most of the catalysts in this group exhibited corresponding catalytic activity for water oxidation, except compounds O75, O77, and O80. According to the data analysis, the catalysts containing picoline ligands had higher

Fig. 36 Structures of complexes O55–64 with Hbpp or its analogues ligand studied by Llobet group

422 H.-T. Zhang and M.-T. Zhang

The Application of Pincer Ligand in Catalytic Water Splitting

423

Fig. 37 Molecular structures of the two isomers O65 and O66

catalytic activity than those containing bpy-type ligands. Most of these catalysts containing halogen ligands had a catalytic induction period, except for I ligand. This indicated that these catalysts generally needed to form real catalytic active species through halogen-water exchange, so that the formation of OO bonds can be easily catalyzed by a higher valent state which comes from multi-step PCET on the metal centers. However, the ruthenium compounds with iodine coordination can achieve an unexpectedly high initial rate of oxygen evolution in the initial catalysis without the induction period. This indicates that the exchange of water molecules with halogens was not an essential process. Prior to this, Sun group had used photocatalytic methods to drive the catalysts O69–70 to catalyze water oxidation [217]. The study found that compared with chemical oxidation, there is no induction period for oxygen generation under photocatalytic conditions. The possible reason is that there is an ultrafast ligand exchange under photo-induced conditions. In order to distinguish the effects of electronic and hydrogen bonds, Llobet and coworkers developed ruthenium complexes O91–92 (Fig. 39) containing fluorinesubstituted bpy ligands [218], in which a hydrogen bond formed between fluorine atom and hydrogen from H2O existed in the single crystal structure of compound O92. By cyclic voltammetry, the redox potential of RuIII/II in complex O92 becomes higher compared with catalyst O32 which has not been replaced by fluorine, whereas the potential of RuIV/III decreases. So that the two redox potentials are coupled together to show a two-electron RuIV/II process. In the oxygen evolution experiment, catalyst O92 containing intramolecular hydrogen bond had a lower TON than catalyst O91, which may be due to the presence of intramolecular hydrogen bond inhibiting the key PECT process in the catalytic process. Then Grotjahn and coworkers studied the catalytic performance of ruthenium compounds O93–95 (Fig. 39) containing bpy ligand modified by electronic pendant as a catalyst for water oxidation [219]. Using CeIV as oxidant, compound O95 exhibited higher TON and TOF than compounds O93–94. This may be due to the formation of intramolecular hydrogen bonds between OMe and water ligands, which can promote the rotation of catalytic cycles by filling the empty site left by oxygen release. Besides pincer ligands of tpy and its derivatives have been widely used in the design of catalysts for water oxidation, pincer-type ligands containing carboxyl group and their analogues have also been widely used. Coordination between ionic ligands with negative charges and metal centers can greatly reduce the redox potential of metal center, thus using photosensitizers to oxidize metals to reach a high-valent, further realizing the conversion of solar energy to chemical energy. This

424

H.-T. Zhang and M.-T. Zhang

Fig. 38 Molecular structures of Ru-based complexes O67–90 studied by Thummel and coworkers

The Application of Pincer Ligand in Catalytic Water Splitting

425

Fig. 39 Structures of Ru complexes O91–95

Fig. 40 Representation of the two Ru-based complexes O96 and O97, housing the negatively charged tridentate pdc ligand

function is beyond the reach of many ordinary neutral ligand catalysts, which require stronger chemical oxidants to assist catalysts in catalytic water oxidation. In 2009, Ru(bda)(pic)2, a mononuclear ruthenium catalyst reported by Sun group as water oxidation catalyst, confirmed the rationality of the analysis and showed significantly lower redox potential [220]. Then in 2010, Sun group synthesized and characterized mononuclear ruthenium compounds O96–97 (Fig. 40) containing double negative charge tridentate pdc ligands [221]. The electron cloud density of Ru center was increased dramatically by pdc ligand, which reduced the initial potential of catalytic water oxidation and matched RuIII/II potential of common [Ru(bpy)3]2+ photosensitizer. Although the

426

H.-T. Zhang and M.-T. Zhang

Fig. 41 Structurally related Ru-(pdc) complexes O98–102

structures of compounds O96 and O97 differ little, complex O96 exhibited stronger catalytic activity compared with modest O97, using both photocatalytic and chemical oxidation methods. The initial catalytic rate of the catalysts was linearly correlated, which indicated that the OO bond formation was most likely to adopt WNA mechanism based on the high-valent Ru ¼ O species. Because of the need to generate high-valent Ru ¼ O species, the process of H2O-picoline exchange becomes a necessary step. Compound O96 contains three picoline ligands which can easily generate [Ru(pdc)(pic)2(OH2)]+ active species in the catalytic system for further catalytic water oxidation. On the contrary, O97 contains only one picoline ligand, and its para-position is occupied by a nitrogen of bpy, which makes H2O -picoline exchange not easy to occur, thus reducing the catalytic efficiency of O97 for water oxidation. Next, Ru(pdc) type catalysts O98–102 (Fig. 41) containing different axial ligands were synthesized and their structure-activity relationships were explored [222]. The electrochemical measurements showed that catalysts O96–98 can produce enhanced catalytic current in catalytic water oxidation, while compound O99–100 exhibited weak catalytic current. A linear correlation was found by the ratio of icat to id, suggesting that the electrochemical activity was directly related to the electrondonating ability of the monodentate ligand. In this group of catalysts, the stronger the electron-donating ability is showed by ligands, the higher the catalytic activity is found by catalysts. The experiment of oxygen evolution by chemical oxidation also showed that the most electron-rich catalyst O96 had the strongest oxygen evolution ability. In a subsequent work, two compounds O103–104 (Fig. 42) with tridentate bpc ligand (Hbpc ¼ 2,20 -bipyridine-6-carboxylic acid) including single anion were synthesized because ruthenium-based complexes with double anion ligands always had poor water solubility [223]. As expected, the ionicity of O103 containing only single carboxyl bpc ligand was stronger than that of catalyst O96, and solubility of O103 in water was much better than catalyst O96. The properties of compound O103–104 were studied in detail by various techniques such as stopped-flow, mass spectrometry, UVvis spectroscopy, and electrochemistry. Finally, five advantages of introducing carboxyl ligands were summarized as follows: (i) reducing the redox

The Application of Pincer Ligand in Catalytic Water Splitting

427

Fig. 42 Structures of the two Ru complexes [Ru(bpc) (pic)3]+ (O103) and [Ru (bpc)(bpy)(OH2)]+ (O104)

Fig. 43 Structures of ruthenium complexes O105 and O106

potential of accessing high-valent (RuIV and RuV, for example) states of ruthenium WOCs; (ii) reducing the overpotential of catalytic water oxidation; (iii) enhancing the rate of ligand exchange (H2O/pic in this case); (iv) facilitating release of dioxygen from the ruthenium center, and (v) possibly drawing the water molecule close to the reactive Ru site via a hydrogen bond.” [223] Of course, these ligands also have some shortcomings. For example, the existence of multiple oxygen atoms can easily be protonated, which hinders proton transfer. There are also potential degradation possibilities of decarboxylation and ligand dissociation in high oxidation state. In 2012, two novel RuIII-based catalysts O105–106 (Fig. 43) for water oxidation were synthesized by Åkermark group using double anionic tridentate ligands (hpbc and hpb) containing phenol, carboxyl, and imidazole [224]. Because the negative charged tridentate ligands make both compounds have low redox potential, compounds O105–106 have also been successfully used to catalyze water oxidation when [Ru(bpy)3]3+-type compounds are used as oxidants. The experimental results showed that catalyst O105 had more effective catalytic ability than compound O106. Using [Ru(bpy)3]3+-type as oxidant, catalyst O105 has a large TON of up to 4,000 and an initial TOF of 7 s1, which are tens of times higher than catalyst O106. When a photosensitizer with stronger oxidation ability is used, a higher turnover number can be obtained. In this paper, the catalytic mechanism of the two catalysts was studied in detail, and the catalytic process was suggested as follows: Firstly, catalysts with three picoline ligands were transformed into a species containing RuIII-OH2 through picoline-H2O ligand exchange reaction [221]; The equilibrium reaction between RuIII-OH2 species and RuIII-OH species was constructed by intramolecular

428

H.-T. Zhang and M.-T. Zhang

Fig. 44 Structures of ruthenium complexes O107 with hqc ligand

proton transfer; RuV¼O species was formed in a multi-step PCET process, and water molecules attack RuV¼O nucleophilically to form OO bond, eventually releasing oxygen back to RuIII-OH2 state. Because the uncoordinated nitrogen in imidazole can be used as an effective proton transfer station in the whole catalytic cycle, the important step of PCET can occur smoothly. When 10 times the amount of oxidant [Ru(bpy)3]3+ was added to the catalytic system, the existence of corresponding highenergy [RuV¼O + H+]+ intermediates can be detected by high-resolution mass spectrometry. This indicated that the bi-anionic tridentate ligand skeleton can greatly improve the stability of high-energy intermediates [224]. Catalysts with low redox potential, such as O105 and O106, have a great opportunity to design molecular assemblies for the conversion of solar energy to chemical energy. In 2012, Mononuclear Ruthenium Compound O107 ([Ru(hqc)(pic)3], Fig. 44) with a tridentate hqc ligand with better π-donor ability was developed by Sun group [225]. The redox potential of the catalyst O107 for RuIII/II is only 0.23 V vs NHE, which is much lower than that of the pdc-based catalyst O96 because the hqc ligand contains the negative part of phenol with stronger electron-donating ability. It is because of the excellent electron-donating ability of hqc ligand that the coordination ability of picoline ligands at equatorial position becomes weaker and the picolineH2O exchange process is more likely to occur, which provides an opportunity to improve the activity of catalysts. Catalyst O107 displays much better catalytic activity than ruthenium catalysts containing neutral ligands through chemical oxidation or photo-driven. As mentioned earlier, for single-site ruthenium catalysts, the rate-limiting step in the whole catalytic water oxidation process is mostly the step of oxygen leaving from the metal center [192, 197]. For catalyst O107, the strong electron donation ability of hqc ligand also weakens the Ru-OO bond, just as it weakens the Ru-picoline bond. Ultimately, what it shows is the improvement of TON and TOF compared with that of its analogue O96. The above-mentioned ruthenium-based catalysts for water oxidation mostly adopt the recognized O-O bond formation mechanisms, called “I2M” and “WNA.” In 2009, Milstein reported that ruthenium catalyst O108 containing NNP pincer ligand used a unique fashion of OO bond formation in catalytic water oxidation [104]. It was also mentioned in the above section of water reduction.

The Application of Pincer Ligand in Catalytic Water Splitting

429

Scheme 14 Proposed mechanism for the formation of H2 and O2 from H2O by Milstein catalyst O108. Reprinted with permission from Ref. [104]

The five-coordinated catalyst O108 first binds to water molecule and splits OH bond to form aromatized Ru-OH species O109 (Scheme 14); Then, under the condition of heating reflux, species O109 reacts further with water and releases hydrogen to obtain species O110 containing cis dihydroxyl coordination. Next, when illumination is introduced, the formation of hydrogen peroxide can be found from the decomposition of O110. Isotope labeling experiments confirmed that hydrogen peroxide originated from the coupling of two hydroxyl groups in compound O110. This case also shows that catalyzing the formation of OO bond does not necessarily require the coordination of multiple metals, nor does it necessarily require the formation of high-valent metal oxo, which also provides a new idea for the design of more reasonable water oxidation catalysts.

3.2

Base Metals-Based Water Oxidation Catalysts

Although many catalysts based on precious metals have been considered to possess high catalytic capacity for water oxidation [226–229], the catalyst composed of inexpensive and earth-abundant metals is the key to realize the economic conversion

430

H.-T. Zhang and M.-T. Zhang

of solar to chemical energy in the future. Therefore, the development of earthabundant first-row transition metal catalysts has become an interesting research field for many researchers. In the field of water oxidation catalysts, after decades of development, a large number of Mn, Fe, Co, Ni, Cubased water oxidation catalysts have been developed [61, 230–234], then their catalytic activity, catalytic mechanism, and other scientific issues have been extensively explored. Although Ru-based catalysts for water oxidation based on pincer ligands such as tpy and pdc have been widely developed, excellent catalytic performance has been demonstrated. However, pincer ligand-based catalysts using the first-row transition metals for water oxidation have not been widely developed and explored. In 1999, Crabtree and his coworkers reported a Mn-based catalyst O111 containing tpy-ligand for water oxidation (Scheme 15) [235]. The discovery of this catalyst means a major breakthrough, due to the first Mn-based catalyst with the ability to catalyze water oxidation by chemical oxidation. It is found that the catalytic center of catalyst O111 is bis(μ-O)MnIIIMnIV composed of two different valent Mn atoms. In fact, the two Mn atoms are identical due to the average valence. In addition, each Mn center contains a water molecule coordination, which is the basic structure of the catalyst for water oxidation. In this work, when unconventional NaCIO was used as oxidant, catalyst O111 could effectively catalyze water oxidation to oxygen, accompanied by general TOF of 12 h1 and TON of 4 in 6 h. According to the analysis of UVvis spectra, the ligands on catalyst O111 were oxidized and decomposed to release manganese ions in the process of catalytic water oxidation driven by chemical oxidation, and manganese ions were further oxidized to form permanganate, which deactivates the catalyst, thus resulting in modest catalytic capacity. Permanganate and sodium hypochlorite are both oxygen-producing reagents, so the oxygen source of O2 evolution by catalytic water oxidation needs to be determined. The isotope labeling experiments with H218O confirmed that 75% of the oxygen released came from water, while the remaining 25% came from hypochlorite. Based on the experimental results, a possible catalytic mechanism was proposed: Firstly, compound O111 was oxidized by one electron to form MnIVMnIV species O112; Then it underwent ligand exchange reaction of H218O-Cl16O to form the intermediate O113 containing Cl16O coordination; Then chlorine atom coupled two electrons to leave in the form of Cl ion, and species O113 containing 16O ¼ MnVMnV was obtained. Species O114 could convert itself into species O114 containing MnVMnV ¼ 18O through equilibrium reaction; Whether O114 or O115, the final reaction of H2O molecules with electron-deficient MnV¼O produced MnIIIMnIII species O116 and released oxygen. Under this catalytic mechanism, it can be calculated by statistics that 75% of the oxygen atoms in released O2 are from water molecules and 25% are from ClO, which is consistent with the experimental results of isotope labeling experiment. Catalyst O111 has been further extensively studied with different oxidants. When Oxone (KSO5) replaces ClO as oxidant, the catalytic activity of catalyst O111 for oxygen evolution becomes enhanced [236–238]. The improvement of catalytic performance is attributed to the higher reactivity between Oxone and catalyst O111, and production of bis(μ-O)MnIVMnV¼O active species becomes easy,

The Application of Pincer Ligand in Catalytic Water Splitting

431

Scheme 15 Proposed mechanism for O2 evolution mediated by the [(H2O)(tpy)MnIII(μ-O)2MnIV(tpy)(OH2)]3+ (O111). Reprinted with permission from Ref. [235]

432

H.-T. Zhang and M.-T. Zhang

which further reacts with Oxone to release oxygen. However, the oxygen evolution capacity in this system is correlated with the ratio between catalyst and oxidant. When the oxidant content is 500 times higher than that of the catalyst, most of the catalysts are oxidized to permanganate, and only 10% of the oxygen released comes from water; When the content of oxidant is less than 100 times, it is no longer easy for the catalyst to be deactivated by oxidation, and 50% of the oxygen atoms in the released oxygen comes from water. From these experimental results, it can be seen that the oxygen transfer reagent must be used to drive the oxidation of water that becomes the main limiting factor of catalyst O111. The utilization of CeIV as an oxidant to drive O111 catalytic water oxidation at the molecular catalysis has been reported, but different views have been maintained among researchers [239– 242]. Subsequent studies have found that catalyst O111 and its analogues can dimerize into tetranuclear manganese compounds, known as “dimer of dimers” [243–245]. It was found that the dimers also had the ability of electrocatalytic water oxidation with poor TON of 2.8. However, DFT calculations showed that the dimers should first be oxidized and decomposed into binuclear Mn species, and then react with water molecules to produce oxygen [245]. In 2005, Crabtree and his coworkers synthesized a series of catalysts O117–128 (Fig. 45) with Mn(μ-O)2Mn core employed various planar tridentate ligands, and extensively studied the structure-activity relationship of catalysts [246]. In 2012, Yagi group studied the substituent effect of catalysts O111, O117, O122, O129–133 (Fig. 45) containing 40 -substituted terpy (R-terpy) ligands as O111 derivatives. Those derivatives were attached to the surface of mica for catalytic water oxidation, and obvious substituent effect was observed [247]. In 2016, mononuclear manganese compounds O134–135 (Fig. 46) containing bipy-alkH ligand were reported by Crabtree group [248]. Oxone was used as oxidant to investigate the oxygen evolution of complexes O134–135. When 25 μm compound O134 was directly added into the solution containing low concentration Oxone, the initial oxygen release rate maintained 2.15 h1; When catalyst O134 was pre-oxidized and then added into the reaction system, the initial oxygen evolution rate increased to 5.27 h1. However, when the concentration of O134 was increased to 125 μm, untreated and pre-oxidized compounds maintained basically the same rate of oxygen evolution. This indicated that mononuclear Mn compound O134 only existed as a catalytic precursor, and the real active intermediate was formed by interacting with oxygen transfer reagent. Complex O135 with saturated coordination has no empty coordination site to accommodate a water molecule, which indicates that it needs to dissociate the ligand before it can obtain a truly catalytic intermediate. These hypotheses were confirmed in 2019 [249]. Compound O134 was treated with oxidant Oxone, and the binuclear catalyst O136 containing Mn2O2 core was obtained by separation and characterization to catalyze water oxidation. With Oxone as sacrificial oxidant, catalyst O136 exhibited an initial TOF of 0.0055 s1, which is only half of that of compound O134 with the same Mn concentration in catalytic system (TOF ¼ 0.0099 s1). This indicated that compound O134, as a catalytic precursor, undergoes a rather intricate process to form truly active species. Based on isotope labeling experiments, a possible mechanism of catalytic oxidation of water by catalyst O136 was proposed. Firstly, catalyst O136

The Application of Pincer Ligand in Catalytic Water Splitting

433

Fig. 45 Structures of [(H2O)(NNN)MnIII(μ-O)2MnIV(NNN)(OH2)]3+-type catalysts O117–133

reacts with oxygen transfer reagent (Oxone) to form an active intermediate of MnVMnV¼O; Then water molecules attack the intermediate nucleophilically and O-O bond is formed, which are further oxidized to release oxygen; Or MnVMnV¼O species reacts with another HSO5 (Oxone) molecule to produce oxygen. In 2012, Llobet and Stahl synthesized Co-based compounds O137–138 (Fig. 47) containing peroxide bridge [250], mimicking the previously studied binuclear Ru-Hbpp catalyst O8. Co-Pi has been found to be a stable and efficient heterogeneous catalyst for water oxidation [251–255]. Calculation supported that the key

434

H.-T. Zhang and M.-T. Zhang

Fig. 46 Structures of Mn-based complexes O134–136, housing bipy-alkH ligand

Fig. 47 Structures of dinuclear CoIII,III peroxo complexes O137 and O138

intermediates of CoIII-OO-CoIII can be produced in the catalytic water oxidation process of Co-Pi [256]. Inspired by this, the role of peroxide bond in the catalytic cycle of homogeneous catalyst O137–138 was studied. It is well known that Co-based compounds are easily oxidized to cobalt oxides under neutral or alkaline conditions, because of cobalt oxides having strong catalytic capacity for water oxidation, which always obscures the effects of molecular catalyst themselves. Thus, the electrocatalytic water oxidation of complexes O137–138 was measured at pH ¼ 2.1, because the possibility of Co-oxides formation was reduced under acidic conditions. In fact, the catalyst O137–138 did maintain a stable catalytic state in a few hours without the formation of Co-oxide nanoparticles. By cyclic voltammetry, compound O137 showed a quasi-reversible redox peak at 1.56 V vs NHE, which was referred to as CoIVCoIII/CoIIICoIII process; When the voltage rose to 1.91 V, the second oxidation peak belonged to CoIVCoIV/CoIVCoIII was observed, accompanied by an enhanced current; This phenomenon was interpreted as the reaction of CoIVCoIV species with water to catalyze water oxidation. For O138 containing Me2bimpy ligand, two redox peaks, CoIVCoIII/CoIIICoIII and CoIVCoIV/ CoIVCoIII, moved to low potential due to the stronger electron-donating ability of the ligand, which were 1.35 V and 1.84 V vs NHE, respectively. By controlling the potential electrolysis at 2 V for a long time, the stable electrolytic current showed that the solution kept a certain stability, thus eliminating the possibility of Co-oxide formation. Finally, the possible mechanism of catalytic oxidation of water by catalyst O137 was proposed by combining isotope labeling experiment and DFT calculation. The initiator of CoIII-OO-CoIII was first converted into CoIV-OO-CoIII

The Application of Pincer Ligand in Catalytic Water Splitting

435

Scheme 16 Synthesis of Cu-based complex O139 by compounds O140–141. Reprinted with permission from Ref. [260]

through one-electron oxidation, then CoIV-OO-CoIV was synthesized by one-electron oxidation of CoIV-OO-CoIII; And oxygen was released immediately after further oxidation; With the participation of water molecules and through multistep proton-coupled electron transfer reaction, it returned to the initial state of CoIIIOO-CoIII. In 2016, Llobet discovered an important intermediate [H2OCoIIICoIIIOO•]4+ species in the process of water oxidation catalyzed by catalyst O137, which was characterized resonance Raman, electron paramagnetic resonance and X-ray absorption spectroscopy [257]. The capture of this species provided a strong support for uncovering the veil of oxygen release process. Recently, copper-based catalysts for water oxidation have not been discovered and studied [232, 258, 259]. In 2017, the only copper-based water oxidation catalyst O139 containing pincer-type ligand was reported by Cao group (Scheme 16) [260]. Originally, in order to study the catalytic properties of copper complexes with N,N0 -2,6-dimethylphenyl-2,6-pyridinedicarboxamidate ligand as a catalyst for water oxidation, copper-based compounds O140–141 containing auxiliary DMF and AcO ligands were synthesized. When O140–141 was placed in the buffer of carbonate for electrochemical study, the real catalyst O139 containing the coordination of HCO3 was rapidly formed by ligand exchange reaction, which was characterized by the X-ray. Through cyclic voltammetry and DFT calculation, it was proposed that the Cu(II) catalyst O139 was first oxidized by one electron to form [L-CuIII-OCO2H] species in the process of electrocatalytic water oxidation; Then the geometric structure of five-coordinated tetragonal Cu(III) species was formed by the coordination of water and [L-CuIII-OCO2H]; Further the bicarbonate as an intramolecular base accepted protons and left in the form of H2CO3 to obtain [L-CuIII-OH] species coordinated by hydroxyl group; After the formation of [L-CuIII-O•-] oxo radical species by one step of PCET, OO bond was formed by water nucleophilic attack, and finally O2 was released through multi-step oxidation. The common shortcoming of copper-based water oxidation catalysts is that they have much high overpotential, and catalyst O139 is no exception in this case. Its overpotential

436

H.-T. Zhang and M.-T. Zhang

reaches 650 mV on the condition of pH ¼ 10. However, the calculation showed that the presence of intramolecular base could provide effective assistance for proton transfer in the catalytic process, which provided some new ideas for the design of better catalysts and catalytic systems.

4 Conclusion and Perspective The exploration of artificial photosynthesis is of great significance for the development of green hydrogen energy. It can provide a large amount of scientific theoretical support for the future conversion of light energy to hydrogen energy. In artificial photosynthesis, hydrogen evolution catalysts (HECs) and water oxidation catalysts (WOCs) play an essential role, liking the engine of a car. In the last few decades, researchers have developed a variety of HECs and WOCs and accumulated some knowledge on the catalytic mechanism. However, these developed catalysts do still not meet industrial applications, and there is still a long way to go. Pincer ligands have been extensively applied in molecular catalysts for artificial photosynthesis, they are generally used as a ligand to construct the molecular skeleton to provide enhanced chemical and thermal stability which serve to minimize the leaching of the metal during the catalytic cycle, but less directly participating in the regulation and transportation of electrons or protons in the catalytic process. To develop a new type of molecular catalyst is the bottleneck for water splitting and required easily functionalizable and tunable ligand systems for controlling the electron-proton transfer and bond cleavage/formation. Pincer ligands provide such a platform for molecular design and modification and thus good candidates.

References 1. Sun L, Hammarström L, Åkermark B, Styring S (2001) Towards artificial photosynthesis: ruthenium–manganese chemistry for energy production. Chem Soc Rev 30(1):36–49 2. Chow J, Kopp RJ, Portney PR (2003) Energy resources and global development. Science 302 (5650):1528–1531 3. Lewis NS, Nocera DG (2006) Powering the planet: chemical challenges in solar energy utilization. Proc Natl Acad Sci U S A 103(43):15729–15735 4. Morton O (2006) Solar energy: a new day dawning? Silicon Valley sunrise. Nature 443 (7107):19–22 5. Allakhverdiev SI, Casal JJ, Nagata T (2009) Photosynthesis from molecular perspectives: towards future energy production. Photochem Photobiol Sci 8(2):137–138 6. Cook TR, Dogutan DK, Reece SY, Surendranath Y, Teets TS, Nocera DG (2010) Solar energy supply and storage for the legacy and nonlegacy worlds. Chem Rev 110(11):6474–6502 7. Tollefson J (2011) How green is my future? Nature 473(7346):134–135 8. Service, R. F (2011) Turning over a new leaf. Science 334(6058):925–927 9. Tachibana Y, Vayssieres L, Durrant JR (2012) Artificial photosynthesis for solar watersplitting. Nat Photonics 6(8):511–518

The Application of Pincer Ligand in Catalytic Water Splitting

437

10. Karkas MD, Johnston EV, Verho O, Akermark B (2014) Artificial photosynthesis: from nanosecond electron transfer to catalytic water oxidation. Acc Chem Res 47(1):100–111 11. Nelson N, Ben-Shem A (2004) The complex architecture of oxygenic photosynthesis. Nat Rev Mol Cell Biol 5(12):971–982 12. McEvoy JP, Brudvig GW (2006) Water-splitting chemistry of photosystem II. Chem Rev 106 (11):4455–4483 13. Barber J (2006) Photosystem II: an enzyme of global significance. Biochem Soc Trans 34 (Pt 5):619–631 14. Barber J (2009) Photosynthetic energy conversion: natural and artificial. Chem Soc Rev 38 (1):185–196 15. Vengadesh P (2012) Nanocrystals – synthesis, chacterization and applications. InTech, Rijeka, pp 41–60 16. Amouyal E (1995) Photochemical production of hydrogen and oxygen from water: a review and state of the art. Sol Energy Mater Sol Cells 38(1–4):249–276 17. Bard AJ, Fox MA (2002) Artificial photosynthesis: solar splitting of water to hydrogen and oxygen. Acc Chem Res 28(3):141–145 18. Frischmann PD, Mahata K, Wurthner F (2013) Powering the future of molecular artificial photosynthesis with light-harvesting metallosupramolecular dye assemblies. Chem Soc Rev 42(4):1847–1870 19. Gust D, Moore TA, Moore AL (2009) Solar fuels via artificial photosynthesis. Acc Chem Res 42(12):1890–1898 20. Concepcion JJ, House RL, Papanikolas JM, Meyer TJ (2012) Chemical approaches to artificial photosynthesis. Proc Natl Acad Sci U S A 109(39):15560–15564 21. Lubitz W, Tumas W (2007) Hydrogen: an overview. Chem Rev 107(10):3900–3903 22. Palo DR, Dagle RA, Holladay JD (2007) Methanol steam reforming for hydrogen production. Chem Rev 107(10):3992–4021 23. Marbán G, Valdés-Solís T (2007) Towards the hydrogen economy? Int J Hydrog Energy 32 (12):1625–1637 24. Nocera DG (2009) Chemistry of personalized solar energy. Inorg Chem 48(21):10001–10017 25. Esswein AJ, Nocera DG (2007) Hydrogen production by molecular photocatalysis. Chem Rev 107(10):4022–4047 26. Thoi VS, Sun Y, Long JR, Chang CJ (2013) Complexes of earth-abundant metals for catalytic electrochemical hydrogen generation under aqueous conditions. Chem Soc Rev 42 (6):2388–2400 27. Morris AJ, Meyer GJ, Fujita E (2009) Molecular approaches to the photocatalytic reduction of carbon dioxide for solar fuels. Acc Chem Res 42(12):1983–1994 28. Du P, Eisenberg R (2012) Catalysts made of earth-abundant elements (Co, Ni, Fe) for water splitting: recent progress and future challenges. Energy Environ Sci 5(3) 29. Lubitz W, Ogata H, Rudiger O, Reijerse E (2014) Hydrogenases. Chem Rev 114 (8):4081–4148 30. Vignais PM, Billoud B, Meyer J (2001) Classification and phylogeny of hydrogenases. FEMS Microbiol Rev 25(4):455–501 31. Vignais PM, Billoud B (2007) Occurrence, classification, and biological function of hydrogenases: an overview. Chem Rev 107(10):4206–4272 32. Tamagnini P, Axelsson R, Lindberg P, Oxelfelt F, Wunschiers R, Lindblad P (2002) Hydrogenases and hydrogen metabolism of cyanobacteria. Microbiol Mol Biol Rev 66(1):1–20 33. Tamagnini P, Leitao E, Oliveira P, Ferreira D, Pinto F, Harris DJ, Heidorn T, Lindblad P (2007) Cyanobacterial hydrogenases: diversity, regulation and applications. FEMS Microbiol Rev 31(6):692–720 34. Thauer RK, Kaster AK, Goenrich M, Schick M, Hiromoto T, Shima S (2010) Hydrogenases from methanogenic archaea, nickel, a novel cofactor, and H2 storage. Annu Rev Biochem 79:507–536

438

H.-T. Zhang and M.-T. Zhang

35. Fontecilla-Camps JC, Volbeda A, Cavazza C, Nicolet Y (2007) Structure/function relationships of [NiFe]- and [FeFe]-hydrogenases. Chem Rev 107(10):4273–4303 36. Fontecilla-Camps JC, Amara P, Cavazza C, Nicolet Y, Volbeda A (2009) Structure-function relationships of anaerobic gas-processing metalloenzymes. Nature 460(7257):814–822 37. Adams MWW (1990) The structure and mechanism of iron-hydrogenases. Biochim Biophys Acta (BBA) Bioenergetics 1020(2):115–145 38. Frey M (2002) Hydrogenases: hydrogen-activating enzymes. Chembiochem 3(2–3):153–160 39. Schilter D, Camara JM, Huynh MT, Hammes-Schiffer S, Rauchfuss TB (2016) Hydrogenase enzymes and their synthetic models: the role of metal hydrides. Chem Rev 116(15):8693–8749 40. Zhang B, Sun L (2019) Artificial photosynthesis: opportunities and challenges of molecular catalysts. Chem Soc Rev 48(7):2216–2264 41. Sakai K, Ozawa H (2007) Homogeneous catalysis of platinum(II) complexes in photochemical hydrogen production from water. Coord Chem Rev 251(21–24):2753–2766 42. Fukuzumi S, Yamada Y, Suenobu T, Ohkubo K, Kotani H (2011) Catalytic mechanisms of hydrogen evolution with homogeneous and heterogeneous catalysts. Energy Environ Sci 4(8) 43. Ogata H, Hirota S, Nakahara A, Komori H, Shibata N, Kato T, Kano K, Higuchi Y (2005) Activation process of [NiFe] hydrogenase elucidated by high-resolution X-ray analyses: conversion of the ready to the unready state. Structure 13(11):1635–1642 44. Nicolet Y, Piras C, Legrand P, Hatchikian CE, Fontecilla-Camps JC (1999) Desulfovibrio desulfuricans iron hydrogenase: the structure shows unusual coordination to an active site Fe binuclear center. Structure 7(1):13–23 45. Hiromoto T, Warkentin E, Moll J, Ermler U, Shima S (2009) The crystal structure of an [Fe]hydrogenase-substrate complex reveals the framework for H2 activation. Angew Chem Int Ed Engl 48(35):6457–6460 46. Dempsey JL, Brunschwig BS, Winkler JR, Gray HB (2009) Hydrogen evolution catalyzed by cobaloximes. Acc Chem Res 42(12):1995–2004 47. Artero V, Chavarot-Kerlidou M, Fontecave M (2011) Splitting water with cobalt. Angew Chem Int Ed Engl 50(32):7238–7266 48. Losse S, Vos JG, Rau S (2010) Catalytic hydrogen production at cobalt centres. Coord Chem Rev 254(21–22):2492–2504 49. Zhang P, Wang M, Gloaguen F, Chen L, Quentel F, Sun L (2013) Electrocatalytic hydrogen evolution from neutral water by molecular cobalt tripyridine-diamine complexes. Chem Commun (Camb) 49(82):9455–9457 50. Stubbert BD, Peters JC, Gray HB (2011) Rapid water reduction to H2 catalyzed by a cobalt bis (iminopyridine) complex. J Am Chem Soc 133(45):18070–18073 51. Marinescu SC, Winkler JR, Gray HB (2012) Molecular mechanisms of cobalt-catalyzed hydrogen evolution. Proc Natl Acad Sci U S A 109(38):15127–15131 52. Du P, Schneider J, Luo G, Brennessel WW, Eisenberg R (2009) Visible light-driven hydrogen production from aqueous protons catalyzed by molecular cobaloxime catalysts. Inorg Chem 48 (11):4952–4962 53. Rakowski DuBois M, DuBois DL (2009) The roles of the first and second coordination spheres in the design of molecular catalysts for H2 production and oxidation. Chem Soc Rev 38(1):62–72 54. Rakowski DuBois M, DuBois DL (2009) Development of molecular electrocatalysts for CO2 reduction and H2 production/oxidation. Acc Chem Res 42(12):1974–1982 55. Wang M, Chen L, Sun L (2012) Recent progress in electrochemical hydrogen production with earth-abundant metal complexes as catalysts. Energy Environ Sci 5(5) 56. Jacques PA, Artero V, Pecaut J, Fontecave M (2009) Cobalt and nickel diimine-dioxime complexes as molecular electrocatalysts for hydrogen evolution with low overvoltages. Proc Natl Acad Sci U S A 106(49):20627–20632 57. Le Goff A, Artero V, Jousselme B, Tran PD, Guillet N, Metaye R, Fihri A, Palacin S, Fontecave M (2009) From hydrogenases to noble metal-free catalytic nanomaterials for H2 production and uptake. Science 326(5958):1384–1387

The Application of Pincer Ligand in Catalytic Water Splitting

439

58. Helm ML, Stewart MP, Bullock RM, DuBois MR, DuBois DL (2011) A synthetic nickel electrocatalyst with a turnover frequency above 100,000 s(-)(1) for H(2) production. Science 333(6044):863–866 59. Fisher BJ, Eisenberg R (1980) Electrocatalytic reduction of carbon dioxide by using macrocycles of nickel and cobalt. J Am Chem Soc 102(24):7361–7363 60. Umena Y, Kawakami K, Shen JR, Kamiya N (2011) Crystal structure of oxygen-evolving photosystem II at a resolution of 1.9 A. Nature 473(7345):55–60 61. Karkas MD, Verho O, Johnston EV, Akermark B (2014) Artificial photosynthesis: molecular systems for catalytic water oxidation. Chem Rev 114(24):11863–12001 62. Ferreira KN, Iverson TM, Maghlaoui K, Barber J, Iwata S (2004) Architecture of the photosynthetic oxygen-evolving center. Science 303(5665):1831–1838 63. Loll B, Kern J, Saenger W, Zouni A, Biesiadka J (2005) Towards complete cofactor arrangement in the 3.0 a resolution structure of photosystem II. Nature 438(7070):1040–1044 64. Yano J, Kern J, Irrgang KD, Latimer MJ, Bergmann U, Glatzel P, Pushkar Y, Biesiadka J, Loll B, Sauer K, Messinger J, Zouni A, Yachandra VK (2005) X-ray damage to the Mn4Ca complex in single crystals of photosystem II: a case study for metalloprotein crystallography. Proc Natl Acad Sci U S A 102(34):12047–12052 65. Yano J, Kern J, Sauer K, Latimer MJ, Pushkar Y, Biesiadka J, Loll B, Saenger W, Messinger J, Zouni A, Yachandra VK (2006) Where water is oxidized to dioxygen: structure of the photosynthetic Mn4Ca cluster. Science 314(5800):821–825 66. Kern J, Alonso-Mori R, Tran R, Hattne J, Gildea RJ, Echols N, Glockner C, Hellmich J, Laksmono H, Sierra RG, Lassalle-Kaiser B, Koroidov S, Lampe A, Han G, Gul S, Difiore D, Milathianaki D, Fry AR, Miahnahri A, Schafer DW, Messerschmidt M, Seibert MM, Koglin JE, Sokaras D, Weng TC, Sellberg J, Latimer MJ, Grosse-Kunstleve RW, Zwart PH, White WE, Glatzel P, Adams PD, Bogan MJ, Williams GJ, Boutet S, Messinger J, Zouni A, Sauter NK, Yachandra VK, Bergmann U, Yano J (2013) Simultaneous femtosecond X-ray spectroscopy and diffraction of photosystem II at room temperature. Science 340(6131):491–495 67. Cox N, Messinger J (2013) Reflections on substrate water and dioxygen formation. Biochim Biophys Acta 1827(8–9):1020–1030 68. Glatzel P, Schroeder H, Pushkar Y, Boron 3rd T, Mukherjee S, Christou G, Pecoraro VL, Messinger J, Yachandra VK, Bergmann U, Yano J (2013) Electronic structural changes of Mn in the oxygen-evolving complex of photosystem II during the catalytic cycle. Inorg Chem 52 (10):5642–5644 69. Rapatskiy L, Cox N, Savitsky A, Ames WM, Sander J, Nowaczyk MM, Rogner M, Boussac A, Neese F, Messinger J, Lubitz W (2012) Detection of the water-binding sites of the oxygen-evolving complex of photosystem II using W-band 17O electron-electron double resonance-detected NMR spectroscopy. J Am Chem Soc 134(40):16619–16634 70. Kulik LV, Epel B, Lubitz W, Messinger J (2007) Electronic structure of the Mn4OxCa cluster in the S0 and S2 states of the oxygen-evolving complex of photosystem II based on pulse 55Mn-ENDOR and EPR spectroscopy. J Am Chem Soc 129(44):13421–13435 71. Haumann M, Liebisch P, Muller C, Barra M, Grabolle M, Dau H (2005) Photosynthetic O2 formation tracked by time-resolved x-ray experiments. Science 310(5750):1019–1021 72. Renger G, Renger T (2008) Photosystem II: the machinery of photosynthetic water splitting. Photosynth Res 98(1–3):53–80 73. Kok B, Forbush B, McGloin M (1970) Cooperation of charges in photosynthetic O2evolution–I. A linear four step mechanism. Photochem Photobiol 11(6):457–475 74. Parent AR, Crabtree RH, Brudvig GW (2013) Comparison of primary oxidants for wateroxidation catalysis. Chem Soc Rev 42(6):2247–2252 75. Wadsworth E, Duke FR, Goetz CA (2002) Present status of cerium(IV)-cerium(III) potentials. Anal Chem 29(12):1824–1825 76. Bolletta F, Juris A, Maestri M, Sandrini D (1980) Quantum yield of formation of the lowest excited state of Ru(bpy)2+3 and Ru(phen)2+3. Inorg Chim Acta 44:L175–L176

440

H.-T. Zhang and M.-T. Zhang

77. Privalov T, Akermark B, Sun L (2011) The O-O bonding in water oxidation: the electronic structure portrayal of a concerted oxygen atom-proton transfer pathway. Chemistry 17 (30):8313–8317 78. Wang LP, Wu Q, Van Voorhis T (2010) Acid-base mechanism for ruthenium water oxidation catalysts. Inorg Chem 49(10):4543–4553 79. Romain S, Vigara L, Llobet A (2009) Oxygen-oxygen bond formation pathways promoted by ruthenium complexes. Acc Chem Res 42(12):1944–1953 80. Concepcion JJ, Tsai MK, Muckerman JT, Meyer TJ (2010) Mechanism of water oxidation by single-site ruthenium complex catalysts. J Am Chem Soc 132(5):1545–1557 81. Lin X, Hu X, Concepcion JJ, Chen Z, Liu S, Meyer TJ, Yang W (2012) Theoretical study of catalytic mechanism for single-site water oxidation process. Proc Natl Acad Sci U S A 109 (39):15669–15672 82. Vigara L, Ertem MZ, Planas N, Bozoglian F, Leidel N, Dau H, Haumann M, Gagliardi L, Cramer CJ, Llobet A (2012) Experimental and quantum chemical characterization of the water oxidation cycle catalysed by [RuII(damp)(bpy)(H2O)]2+. Chem Sci 3(8) 83. Romero I, Rodriguez M, Sens C, Mola J, Rao Kollipara M, Francas L, Mas-Marza E, Escriche L, Llobet A (2008) Ru complexes that can catalytically oxidize water to molecular dioxygen. Inorg Chem 47(6):1824–1834 84. Kikuchi T, Tanaka K (2014) Mechanistic approaches to molecular catalysts for water oxidation. Eur J Inorg Chem 2014(4):607–618 85. Savini A, Bucci A, Bellachioma G, Rocchigiani L, Zuccaccia C, Llobet A, Macchioni A (2014) Mechanistic aspects of water oxidation catalyzed by organometallic iridium complexes. Eur J Inorg Chem 2014(4):690–697 86. Sala X, Romero I, Rodriguez M, Escriche L, Llobet A (2009) Molecular catalysts that oxidize water to dioxygen. Angew Chem Int Ed Engl 48(16):2842–2852 87. van der Boom ME, Milstein D (2003) Cyclometalated phosphine-based pincer complexes: mechanistic insight in catalysis, coordination, and bond activation. Chem Rev 103 (5):1759–1792 88. Choi J, MacArthur AH, Brookhart M, Goldman AS (2011) Dehydrogenation and related reactions catalyzed by iridium pincer complexes. Chem Rev 111(3):1761–1779 89. Selander N, Szabó KJ (2011) Catalysis by palladium pincer complexes. Chem Rev 111 (3):2048–2076 90. Schneider S, Meiners J, Askevold B (2012) Cooperative aliphatic PNP amido pincer ligands – versatile building blocks for coordination chemistry and catalysis. Eur J Inorg Chem 2012 (3):412–429 91. Tausz J, Donath P (1930) Über die oxydation des Wasserstoffs und der Kohlenwasserstoffe mittels Bakterien. Hoppe Seylers Z Physiol Chem 190(3–6):141–168 92. Getoff N (1990) Photoelectrochemical and photocatalytic methods of hydrogen production: a short review. Int J Hydrog Energy 15(6):407–417 93. Konigstein CBR (1997) Photoinduced hydrogen production from aqueous solution?Frontiers and new trends*1. Int J Hydrog Energy 22(5):471–474 94. Abe T, Kaneko M (2001) pH-dependent electrocatalysis for proton reduction by bis(2,20 :60 ,200 -terpyridine) cobalt(II) complex embedded in Nafion® membrane. J Mol Catal A Chem 169 (1–2):177–183 95. Guadalupe AR, Usifer DA, Potts KT, Hurrell HC, Mogstad AE, Abruna HD (1988) Novel chemical pathways and charge-transport dynamics of electrodes modified with electropolymerized layers of [Co(v-terpy)2]2+. J Am Chem Soc 110(11):3462–3466 96. Elvington M, Brown J, Arachchige SM, Brewer KJ (2007) Photocatalytic hydrogen production from water employing a Ru, Rh, Ru molecular device for photoinitiated electron collection. J Am Chem Soc 129(35):10644–10645 97. Rau S, Schafer B, Gleich D, Anders E, Rudolph M, Friedrich M, Gorls H, Henry W, Vos JG (2006) A supramolecular photocatalyst for the production of hydrogen and the selective hydrogenation of tolane. Angew Chem Int Ed Engl 45(37):6215–6218

The Application of Pincer Ligand in Catalytic Water Splitting

441

98. Du P, Schneider J, Li F, Zhao W, Patel U, Castellano FN, Eisenberg R (2008) Bi- and terpyridyl platinum(II) chloro complexes: molecular catalysts for the photogeneration of hydrogen from water or simply precursors for colloidal platinum? J Am Chem Soc 130 (15):5056–5058 99. Galińska A, Walendziewski J (2005) Photocatalytic water splitting over PtTiO2in the presence of sacrificial reagents. Energy Fuel 19(3):1143–1147 100. Liu H, Yuan J, Shangguan W (2006) Photochemical reduction and oxidation of water including sacrificial reagents and Pt/TiO2Catalyst. Energy Fuel 20(6):2289–2292 101. Zhang J, Du P, Schneider J, Jarosz P, Eisenberg R (2007) Photogeneration of hydrogen from water using an integrated system based on TiO2 and platinum(II) diimine dithiolate sensitizers. J Am Chem Soc 129(25):7726–7727 102. Yamauchi K, Masaoka S, Sakai K (2009) Evidence for Pt(II)-based molecular catalysis in the thermal reduction of water into molecular hydrogen. J Am Chem Soc 131(24):8404–8406 103. Hill MG, Bailey JA, Miskowski VM, Gray HB (1996) Spectroelectrochemistry and dimerization equilibria of chloro(terpyridine)platinum(II). Nature of the reduced complexes. Inorg Chem 35(16):4585–4590 104. Kohl SW, Weiner L, Schwartsburd L, Konstantinovski L, Shimon LJ, Ben-David Y, Iron MA, Milstein D (2009) Consecutive thermal H2 and light-induced O2 evolution from water promoted by a metal complex. Science 324(5923):74–77 105. Li J, Shiota Y, Yoshizawa K (2009) Metal-ligand cooperation in H2 production and H2O decomposition on a Ru(II) PNN complex: the role of ligand dearomatization-aromatization. J Am Chem Soc 131(38):13584–13585 106. Yang X, Hall MB (2010) Mechanism of water splitting and oxygen-oxygen bond formation by a mononuclear ruthenium complex. J Am Chem Soc 132(1):120–130 107. Sandhya KS, Suresh CH (2011) Water splitting promoted by a ruthenium(II) PNN complex: an alternate pathway through a dihydrogen complex for hydrogen production. Organometallics 30(14):3888–3891 108. Ma C, Piccinin S, Fabris S (2012) Reaction mechanisms of water splitting and H2 evolution by a Ru(II)-pincer complex identified with Ab initio metadynamics simulations. ACS Catal 2 (7):1500–1506 109. Sandhya KS, Remya GS, Suresh CH (2015) Pincer ligand modifications to tune the activation barrier for H2 elimination in water splitting milstein catalyst. Inorg Chem 54 (23):11150–11156 110. Kristjansdottir SS, Norton JR (1991) Transition metal hydrides. VCH, New York, pp 309–357 111. Abdur-Rashid K, Fong TP, Greaves B, Gusev DG, Hinman JG, Landau SE, Lough AJ, Morris RH (2000) An acidity scale for phosphorus-containing compounds including metal hydrides and dihydrogen complexes in THF: toward the unification of acidity scales. J Am Chem Soc 122(38):9155–9171 112. Levina VA, Rossin A, Belkova NV, Chierotti MR, Epstein LM, Filippov OA, Gobetto R, Gonsalvi L, Lledós A, Shubina ES, Zanobini F, Peruzzini M (2011) Acid-base interaction between transition-metal hydrides: dihydrogen bonding and dihydrogen evolution. Angew Chem 123(6):1403–1406 113. Boyer JL, Polyansky DE, Szalda DJ, Zong R, Thummel RP, Fujita E (2011) Effects of a proximal base on water oxidation and proton reduction catalyzed by geometric isomers of [Ru (tpy)(pynap)(OH2)]2+. Angew Chem Int Ed Engl 50(52):12600–12604 114. Chen Z, Glasson CR, Holland PL, Meyer TJ (2013) Electrogenerated polypyridyl ruthenium hydride and ligand activation for water reduction to hydrogen and acetone to iso-propanol. Phys Chem Chem Phys 15(24):9503–9507 115. Kilgore UJ, Roberts JA, Pool DH, Appel AM, Stewart MP, DuBois MR, Dougherty WG, Kassel WS, Bullock RM, DuBois DL (2011) [Ni(P(Ph)2N(C6H4X)2)2]2+ complexes as electrocatalysts for H2 production: effect of substituents, acids, and water on catalytic rates. J Am Chem Soc 133(15):5861–5872

442

H.-T. Zhang and M.-T. Zhang

116. Luca OR, Blakemore JD, Konezny SJ, Praetorius JM, Schmeier TJ, Hunsinger GB, Batista VS, Brudvig GW, Hazari N, Crabtree RH (2012) Organometallic Ni pincer complexes for the electrocatalytic production of hydrogen. Inorg Chem 51(16):8704–8709 117. Luca OR, Konezny SJ, Blakemore JD, Colosi DM, Saha S, Brudvig GW, Batista VS, Crabtree RH (2012) A tridentate Ni pincer for aqueous electrocatalytic hydrogen production. New J Chem 36(5) 118. Reed BR, Yousif M, Lord RL, McKinnon M, Rochford J, Groysman S (2017) Coordination chemistry and reactivity of bis(aldimino)pyridine nickel complexes in four different oxidation states. Organometallics 36(3):582–593 119. Luca OR, Konezny SJ, Hunsinger GB, Müller P, Takase MK, Crabtree RH (2014) Ni complexes of redox-active pincers with pendant H-bonding sites as precursors for hydrogen production electrocatalysis. Polyhedron 82:2–6 120. Jing X, Wu P, Liu X, Yang L, He C, Duan C (2015) Light-driven hydrogen evolution with a nickel thiosemicarbazone redox catalyst featuring Ni⋯H interactions under basic conditions. New J Chem 39(2):1051–1059 121. Mondragon A, Flores-Alamo M, Martinez-Alanis PR, Aullon G, Ugalde-Saldivar VM, Castillo I (2015) Electrocatalytic proton reduction by dimeric nickel complex of a sterically demanding pincer-type NS2 aminobis(thiophenolate) ligand. Inorg Chem 54(2):619–627 122. Wang J, Li C, Zhou Q, Wang W, Hou Y, Zhang B, Wang X (2016) Photocatalytic hydrogen evolution by Cu(II) complexes. Dalton Trans 45(13):5439–5443 123. Ramakrishnan S, Chakraborty S, Brennessel WW, Chidsey CED, Jones WD (2016) Rapid oxidative hydrogen evolution from a family of square-planar nickel hydride complexes. Chem Sci 7(1):117–127 124. Bakir M, Lawrence MAW, Ferhat M, Conry RR (2017) Spectroscopic, electrochemical and X-ray crystallographic properties of a novel palladium(II) complex of thioamide deprotonated di-2-pyridyl ketone thiosemicarbazone (dpktsc-H). J Coord Chem 70(17):3048–3064 125. Lawrence MAW, Mulder WH (2018) SNS versus SNN pincer ligands: electrochemical studies and their palladium(II) complexes as electro-catalyst for proton reduction. Chem Select 3 (28):8387–8394 126. Luo S, Siegler MA, Bouwman E (2018) Dinuclear nickel complexes of thiolate-functionalized carbene ligands and their electrochemical properties. Organometallics 37(5):740–747 127. Yap CP, Chong YY, Chwee TS, Fan WY (2018) Electrocatalytic proton reduction by an air-stable nickel(ii)-thiolato PNN pincer complex. Dalton Trans 47(25):8483–8488 128. Viertl W, Pann J, Pehn R, Roithmeyer H, Bendig M, Rodriguez-Villalon A, Bereiter R, Heiderscheid M, Muller T, Zhao X, Hofer TS, Thompson ME, Shi S, Brueggeller P (2019) Performance of enhanced DuBois type water reduction catalysts (WRC) in artificial photosynthesis – effects of various proton relays during catalysis. Faraday Discuss 129. Volbeda A, Charon MH, Piras C, Hatchikian EC, Frey M, Fontecilla-Camps JC (1995) Crystal structure of the nickel-iron hydrogenase from Desulfovibrio gigas. Nature 373(6515):580–587 130. Halcrow MA (1995) Die Struktur der [NiFe]-Hydrogenase ausD. gigas und die art ihres nickelkomplexes. Angew Chem 107(11):1307–1310 131. Halcrow MA (1995) The structure of theD. gigas[NiFe] hydrogenase and the nature of the hydrogenase nickel complex. Angew Chem Int Ed Engl 34(11):1193–1195 132. Volbeda A, Garcin E, Piras C, de Lacey AL, Fernandez VM, Hatchikian EC, Frey M, Fontecilla-Camps JC (1996) Structure of the [NiFe] hydrogenase active site: evidence for biologically uncommon Fe ligands. J Am Chem Soc 118(51):12989–12996 133. Graf E-G, Thauer RK (1981) Hydrogenase from methanobacterium thermoautotrophicum, a nickel-containing enzyme. FEBS Lett 136(1):165–169 134. Maroney MJ, Pressler MA, Mirza SA, Whitehead JP, Gurbiel RJ, Hoffmann BM (1995) Insights into the role of nickel in hydrogenase. Adv Chem Ser 246:21–60 135. Sellmann D, Häußinger D, Heinemann FW (1999) Synthesis, structures, and reactions of sulfur-rich nickel and platinum complexes with [MS3] and [MNS2] cores. Eur J Inorg Chem 1999(10):1715–1725

The Application of Pincer Ligand in Catalytic Water Splitting

443

136. Sellmann D, Geipel F, Heinemann FW (2000) Synthesis and exchange reactions of sulfur-rich nickel and palladium [M(L)(‘S3’)] complexes [‘S3’2– ¼ bis(2-mercaptophenyl)sulfide(2–)]. Eur J Inorg Chem 2000(2):271–279 137. Sellmann D, Geipel F, Moll M (2000) [Ni(NHPnPr3)(`S3')], the first nickel thiolate complex modeling the nickel cysteinate site and reactivity of [NiFe] hydrogenase. Angew Chem Int Ed 39(3):561–563 138. Cammack R (1993) Bioinorganic catalysis. Marcel Dekker, New York, pp 189–226 139. Wombwell C, Reisner E (2014) Synthesis, structure and reactivity of Ni site models of [NiFeSe] hydrogenases. Dalton Trans 43(11):4483–4493 140. Yang JY, Bullock RM, Shaw WJ, Twamley B, Fraze K, DuBois MR, DuBois DL (2009) Mechanistic insights into catalytic h(2) oxidation by ni complexes containing a diphosphine ligand with a positioned amine base. J Am Chem Soc 131(16):5935–5945 141. Connelly SJ, Zimmerman AC, Kaminsky W, Heinekey DM (2012) Synthesis, structure, and reactivity of a nickel dihydrogen complex. Chem Eur J 18(50):15932–15934 142. Kubas GJ, Nelson JE, Bryan JC, Eckert J, Wisniewski L, Zilm K (1994) Isolation of an extremely labile dihydrogen complex, Cr(CO)3(PPri3)2(H2), containing the shortest ligated H-H bond. Inorg Chem 33(13):2954–2960 143. Shima S, Thauer RK (2007) A third type of hydrogenase catalyzing H2 activation. Chem Rec 7(1):37–46 144. Wright JA, Turrell PJ, Pickett CJ (2010) The third hydrogenase: more natural organometallics. Organometallics 29(23):6146–6156 145. Corr MJ, Murphy JA (2011) Evolution in the understanding of [Fe]-hydrogenase. Chem Soc Rev 40(5):2279–2292 146. Shima S, Pilak O, Vogt S, Schick M, Stagni MS, Meyer-Klaucke W, Warkentin E, Thauer RK, Ermler U (2008) The crystal structure of [Fe]-hydrogenase reveals the geometry of the active site. Science 321(5888):572–575 147. Guo Y, Wang H, Xiao Y, Vogt S, Thauer RK, Shima S, Volkers PI, Rauchfuss TB, Pelmenschikov V, Case DA, Alp EE, Sturhahn W, Yoda Y, Cramer SP (2008) Characterization of the Fe site in iron-sulfur cluster-free hydrogenase (Hmd) and of a model compound via nuclear resonance vibrational spectroscopy (NRVS). Inorg Chem 47(10):3969–3977 148. Wang X, Li Z, Zeng X, Luo Q, Evans DJ, Pickett CJ, Liu X (2008) The iron Centre of the cluster-free hydrogenase (Hmd): low-spin Fe(II) or low-spin Fe(0)? Chem Commun (Camb) 30:3555–3557 149. Liu T, Li B, Popescu CV, Bilko A, Perez LM, Hall MB, Darensbourg MY (2010) Analysis of a pentacoordinate iron dicarbonyl as synthetic analogue of the Hmd or mono-iron hydrogenase active site. Chem Eur J 16(10):3083–3089 150. Royer AM, Rauchfuss TB, Gray DL (2009) Oxidative addition of thioesters to iron(0): activesite models for Hmd, nature’s third hydrogenase. Organometallics 28(13):3618–3620 151. Chen D, Ahrens-Botzong A, Schunemann V, Scopelliti R, Hu X (2011) Synthesis and characterization of a series of model complexes of the active site of [Fe]-hydrogenase (Hmd). Inorg Chem 50(11):5249–5257 152. Royer AM, Salomone-Stagni M, Rauchfuss TB, Meyer-Klaucke W (2010) Iron acyl thiolato carbonyls: structural models for the active site of the [Fe]-hydrogenase (Hmd). J Am Chem Soc 132(47):16997–17003 153. Turrell PJ, Wright JA, Peck JN, Oganesyan VS, Pickett CJ (2010) The third hydrogenase: a ferracyclic carbamoyl with close structural analogy to the active site of Hmd. Angew Chem Int Ed Engl 49(41):7508–7511 154. Chen D, Scopelliti R, Hu X (2010) [Fe]-hydrogenase models featuring acylmethylpyridinyl ligands. Angew Chem Int Ed Engl 49(41):7512–7515 155. Chen D, Scopelliti R, Hu X (2011) A five-coordinate iron center in the active site of [Fe]-hydrogenase: hints from a model study. Angew Chem Int Ed Engl 50(25):5671–5673

444

H.-T. Zhang and M.-T. Zhang

156. Song LC, Xie ZJ, Wang MM, Zhao GY, Song HB (2012) Biomimetic models for the active site of [Fe]hydrogenase featuring an acylmethyl(hydroxymethyl)pyridine ligand. Inorg Chem 51(14):7466–7468 157. Bichler B, Holzhacker C, Stoger B, Puchberger M, Veiros LF, Kirchner K (2013) Heterolytic cleavage of dihydrogen by an iron(II) PNP pincer complex via metal-ligand cooperation. Organometallics 32(15):4114–4121 158. Durgaprasad G, Xie ZL, Rose MJ (2016) Iron hydride detection and intramolecular hydride transfer in a synthetic model of mono-iron hydrogenase with a CNS chelate. Inorg Chem 55 (2):386–389 159. Xie ZL, Durgaprasad G, Ali AK, Rose MJ (2017) Substitution reactions of iron(ii) carbamoylthioether complexes related to mono-iron hydrogenase. Dalton Trans 46(33):10814–10829 160. Hirotsu M, Santo K, Hashimoto H, Kinoshita I (2012) Carbon- and sulfur-bridged diiron carbonyl complexes containing N,C,S-tridentate ligands derived from functionalized dibenzothiophenes: mimics of the [FeFe]-hydrogenase active site. Organometallics 31 (21):7548–7557 161. Hirotsu M, Tsuboi C, Nishioka T, Kinoshita I (2011) Carbon-sulfur bond cleavage reactions of dibenzothiophene derivatives mediated by iron and ruthenium carbonyls. Dalton Trans 40 (4):785–787 162. Nakae T, Hirotsu M, Kinoshita I (2015) Di- and mononuclear iron complexes of N,C,Stridentate ligands containing an Aminopyridyl group: effect of the pendant amine site on catalytic properties for proton reduction. Organometallics 34(16):3988–3997 163. Hirotsu M, Santo K, Tsuboi C, Kinoshita I (2014) Diiron carbonyl complexes bearing an N,C, S-pincer ligand: reactivity toward phosphines, heterolytic Fe–Fe cleavage, and electrocatalytic proton reduction. Organometallics 33(16):4260–4268 164. Ott S, Kritikos M, Akermark B, Sun L (2003) Synthesis and structure of a biomimetic model of the iron hydrogenase active site covalently linked to a ruthenium photosensitizer. Angew Chem Int Ed Engl 42(28):3285–3288 165. Hammarström L, Barigelletti F, Flamigni L, Indelli MT, Armaroli N, Calogero G, Guardigli M, Sour A, Collin J-P, Sauvage J-P (1997) A study on delocalization of MLCT excited states by rigid bridging ligands in homometallic dinuclear complexes of ruthenium(II). Chem A Eur J 101(48):9061–9069 166. Ott S, Borgstrom M, Kritikos M, Lomoth R, Bergquist J, Akermark B, Hammarstrom L, Sun L (2004) Model of the iron hydrogenase active site covalently linked to a ruthenium photosensitizer: synthesis and photophysical properties. Inorg Chem 43(15):4683–4692 167. Hawecker JL, Lehn JM, Ziessel R (1983) Efficient homogeneous photochemical hydrogen generation and water reduction mediated by cobaloxime or macrocyclic cobalt complexes. New J Chem 7:271–277 168. Mulfort KL, Tiede DM (2010) Supramolecular cobaloxime assemblies for H2 photocatalysis: an initial solution state structure-function analysis. J Phys Chem B 114(45):14572–14581 169. Gersten SW, Samuels GJ, Meyer TJ (1982) Catalytic oxidation of water by an oxo-bridged ruthenium dimer. J Am Chem Soc 104(14):4029–4030 170. Gilbert JA, Eggleston DS, Murphy WR, Geselowitz DA, Gersten SW, Hodgson DJ, Meyer TJ (1985) Structure and redox properties of the water-oxidation catalyst [(bpy)2(OH2)RuORu (OH2)(bpy)2]4+. J Am Chem Soc 107(13):3855–3864 171. Collin JP, Sauvage JP (1986) Synthesis and study of mononuclear ruthenium(II) complexes of sterically hindering diimine chelates. Implications for the catalytic oxidation of water to molecular oxygen. Inorg Chem 25(2):135–141 172. Wada T, Tsuge K, Tanaka K (2000) Electrochemical oxidation of water to dioxygen catalyzed by the oxidized form of the bis(ruthenium – hydroxo) complex in H2O. Angew Chem Int Ed 39(8):1479–1482 173. Wada T, Tsuge K, Tanaka K (2001) Syntheses and redox properties of bis(hydroxoruthenium) complexes with quinone and bipyridine ligands. Water-oxidation catalysis. Inorg Chem 40 (2):329–337

The Application of Pincer Ligand in Catalytic Water Splitting

445

174. Kobayashi K, Ohtsu H, Wada T, Kato T, Tanaka K (2003) Characterization of a stable ruthenium complex with an oxyl radical. J Am Chem Soc 125(22):6729–6739 175. Tsai MK, Rochford J, Polyansky DE, Wada T, Tanaka K, Fujita E, Muckerman JT (2009) Characterization of redox states of Ru(OH(2))(Q)(tpy)(2+) (Q ¼ 3,5-di-tert-butyl-1,2-benzoquinone, tpy ¼ 2,20 :60 ,200 -terpyridine) and related species through experimental and theoretical studies. Inorg Chem 48(10):4372–4383 176. Wada T, Ohtsu H, Tanaka K (2012) Catalytic four-electron oxidation of water by intramolecular coupling of the oxo ligands of a bis(ruthenium-bipyridine) complex. Chem Eur J 18 (8):2374–2381 177. Wasylenko DJ, Ganesamoorthy C, Koivisto BD, Berlinguette CP (2010) Examination of water oxidation by catalysts containing cofacial metal sites. Eur J Inorg Chem 2010(20):3135–3142 178. Betley TA, Wu Q, Van Voorhis T, Nocera DG (2008) Electronic design criteria for O-O bond formation via metal-oxo complexes. Inorg Chem 47(6):1849–1861 179. Sens C, Romero I, Rodriguez M, Llobet A, Parella T, Benet-Buchholz J (2004) A new Ru complex capable of catalytically oxidizing water to molecular dioxygen. J Am Chem Soc 126 (25):7798–7799 180. Roeser S, Ertem MZ, Cady C, Lomoth R, Benet-Buchholz J, Hammarstrom L, Sarkar B, Kaim W, Cramer CJ, Llobet A (2012) Synthesis, structure, and electronic properties of RuN6 dinuclear Ru-Hbpp complexes. Inorg Chem 51(1):320–327 181. Romain S, Bozoglian F, Sala X, Llobet A (2009) Oxygen-oxygen bond formation by the Ru-Hbpp water oxidation catalyst occurs solely via an intramolecular reaction pathway. J Am Chem Soc 131(8):2768–2769 182. Bozoglian F, Romain S, Ertem MZ, Todorova TK, Sens C, Mola J, Rodriguez M, Romero I, Benet-Buchholz J, Fontrodona X, Cramer CJ, Gagliardi L, Llobet A (2009) The Ru-Hbpp water oxidation catalyst. J Am Chem Soc 131(42):15176–15187 183. Ellis CD, Gilbert JA, Murphy WR, Meyer TJ (1983) Electrocatalytic oxidation of chloride to chlorine based on polypyridine complexes of ruthenium. J Am Chem Soc 105(14):4842–4843 184. Francas L, Sala X, Benet-Buchholz J, Escriche L, Llobet A (2009) A Ru-Hbpp-based wateroxidation catalyst anchored on rutile TiO2. ChemSusChem 2(4):321–329 185. Mola J, Mas-Marza E, Sala X, Romero I, Rodriguez M, Vinas C, Parella T, Llobet A (2008) Ru-hbpp-based water-oxidation catalysts anchored on conducting solid supports. Angew Chem Int Ed Engl 47(31):5830–5832 186. Aguiló J, Francàs L, Liu HJ, Bofill R, García-Antón J, Benet-Buchholz J, Llobet A, Escriche L, Sala X (2014) Characterization and performance of electrostatically adsorbed Ru–Hbpp water oxidation catalysts. Cat Sci Technol 4(1):190–199 187. Neudeck S, Maji S, Lopez I, Meyer S, Meyer F, Llobet A (2014) New powerful and oxidatively rugged dinuclear Ru water oxidation catalyst: control of mechanistic pathways by tailored ligand design. J Am Chem Soc 136(1):24–27 188. Francas L, Sala X, Escudero-Adan E, Benet-Buchholz J, Escriche L, Llobet A (2011) Synthesis, structure, and reactivity of new tetranuclear Ru-Hbpp-based water-oxidation catalysts. Inorg Chem 50(7):2771–2781 189. Zong R, Thummel RP (2005) A new family of Ru complexes for water oxidation. J Am Chem Soc 127(37):12802–12803 190. Tseng HW, Zong R, Muckerman JT, Thummel R (2008) Mononuclear ruthenium (II) complexes that catalyze water oxidation. Inorg Chem 47(24):11763–11773 191. Kaveevivitchai N, Kohler L, Zong R, El Ojaimi M, Mehta N, Thummel RP (2013) A Ru (II) bis-terpyridine-like complex that catalyzes water oxidation: the influence of steric strain. Inorg Chem 52(18):10615–10622 192. Concepcion JJ, Jurss JW, Templeton JL, Meyer TJ (2008) One site is enough. Catalytic water oxidation by [Ru(tpy)(bpm)(OH2)]2+and [Ru(tpy)(bpz)(OH2)]2+. J Am Chem Soc 130 (49):16462–16463

446

H.-T. Zhang and M.-T. Zhang

193. Polyansky DE, Muckerman JT, Rochford J, Zong R, Thummel RP, Fujita E (2011) Water oxidation by a mononuclear ruthenium catalyst: characterization of the intermediates. J Am Chem Soc 133(37):14649–14665 194. Badiei YM, Polyansky DE, Muckerman JT, Szalda DJ, Haberdar R, Zong R, Thummel RP, Fujita E (2013) Water oxidation with mononuclear ruthenium(II) polypyridine complexes involving a direct Ru(IV) horizontal lineO pathway in neutral and alkaline media. Inorg Chem 52(15):8845–8850 195. Wasylenko DJ, Ganesamoorthy C, Koivisto BD, Henderson MA, Berlinguette CP (2010) Insight into water oxidation by mononuclear polypyridyl Ru catalysts. Inorg Chem 49 (5):2202–2209 196. Masaoka S, Sakai K (2009) Clear evidence showing the robustness of a highly active oxygenevolving mononuclear ruthenium complex with an aqua ligand. Chem Lett 38(2):182–183 197. Wasylenko DJ, Ganesamoorthy C, Henderson MA, Koivisto BD, Osthoff HD, Berlinguette CP (2010) Electronic modification of the [Ru(II)(tpy)(bpy)(OH(2))](2+) scaffold: effects on catalytic water oxidation. J Am Chem Soc 132(45):16094–16106 198. Wasylenko DJ, Ganesamoorthy C, Henderson MA, Berlinguette CP (2011) Unraveling the roles of the acid medium, experimental probes, and terminal oxidant, (NH4)2[Ce(NO3)6], in the study of a homogeneous water oxidation catalyst. Inorg Chem 50(8):3662–3672 199. Yoshida M, Masaoka S, Abe J, Sakai K (2010) Catalysis of mononuclear aquaruthenium complexes in oxygen evolution from water: a new radical coupling path using hydroxocerium (IV) species. Chem Asian J 5(11):2369–2378 200. Ikeda-Ohno A, Tsushima S, Hennig C, Yaita T, Bernhard G (2012) Dinuclear complexes of tetravalent cerium in an aqueous perchloric acid solution. Dalton Trans 41(24):7190–7192 201. Concepcion JJ, Jurss JW, Norris MR, Chen Z, Templeton JL, Meyer TJ (2010) Catalytic water oxidation by single-site ruthenium catalysts. Inorg Chem 49(4):1277–1279 202. Chen Z, Concepcion JJ, Meyer TJ (2011) Rapid catalytic water oxidation by a single site, Ru carbene catalyst. Dalton Trans 40(15):3789–3792 203. Ashford DL, Stewart DJ, Glasson CR, Binstead RA, Harrison DP, Norris MR, Concepcion JJ, Fang Z, Templeton JL, Meyer TJ (2012) An amide-linked chromophore-catalyst assembly for water oxidation. Inorg Chem 51(12):6428–6430 204. Concepcion JJ, Jurss JW, Hoertz PG, Meyer TJ (2009) Catalytic and surface-electrocatalytic water oxidation by redox mediator-catalyst assemblies. Angew Chem Int Ed Engl 48 (50):9473–9476 205. Ashford DL, Song W, Concepcion JJ, Glasson CR, Brennaman MK, Norris MR, Fang Z, Templeton JL, Meyer TJ (2012) Photoinduced electron transfer in a chromophore-catalyst assembly anchored to TiO2. J Am Chem Soc 134(46):19189–19198 206. Ashford DL, Lapides AM, Vannucci AK, Hanson K, Torelli DA, Harrison DP, Templeton JL, Meyer TJ (2014) Water oxidation by an electropolymerized catalyst on derivatized mesoporous metal oxide electrodes. J Am Chem Soc 136(18):6578–6581 207. Chen Z, Concepcion JJ, Jurss JW, Meyer TJ (2009) Single-site, catalytic water oxidation on oxide surfaces. J Am Chem Soc 131(43):15580–15581 208. Chen Z, Concepcion JJ, Hull JF, Hoertz PG, Meyer TJ (2010) Catalytic water oxidation on derivatized nanoITO. Dalton Trans 39(30):6950–6952 209. Chen Z, Concepcion JJ, Luo H, Hull JF, Paul A, Meyer TJ (2010) Nonaqueous catalytic water oxidation. J Am Chem Soc 132(50):17670–17673 210. Hoertz PG, Chen Z, Kent CA, Meyer TJ (2010) Application of high surface area tin-doped indium oxide nanoparticle films as transparent conducting electrodes. Inorg Chem 49 (18):8179–8181 211. Chen Z, Concepcion JJ, Hu X, Yang W, Hoertz PG, Meyer TJ (2010) Concerted O atomproton transfer in the O-O bond forming step in water oxidation. Proc Natl Acad Sci U S A 107 (16):7225–7229

The Application of Pincer Ligand in Catalytic Water Splitting

447

212. Su W, Younus HA, Zhou K, Khattak ZAK, Chaemcheun S, Chen C, Verpoort F (2017) Chemical and photochemical water oxidation catalyzed by novel ruthenium complexes comprising a negatively charged NCNHCO ligand. Cat Sci Technol 7(2):387–395 213. Roeser S, Farras P, Bozoglian F, Martinez-Belmonte M, Benet-Buchholz J, Llobet A (2011) Chemical, electrochemical, and photochemical catalytic oxidation of water to dioxygen with mononuclear ruthenium complexes. ChemSusChem 4(2):197–207 214. Mognon L, Mandal S, Castillo CE, Fortage J, Molton F, Aromi G, Benet-Buchhlolz J, Collomb MN, Llobet A (2016) Synthesis, structure, spectroscopy and reactivity of new heterotrinuclear water oxidation catalysts. Chem Sci 7(5):3304–3312 215. Yamazaki H, Hakamata T, Komi M, Yagi M (2011) Stoichiometric photoisomerization of mononuclear ruthenium(II) monoaquo complexes controlling redox properties and water oxidation catalysis. J Am Chem Soc 133(23):8846–8849 216. Kaveevivitchai N, Zong R, Tseng HW, Chitta R, Thummel RP (2012) Further observations on water oxidation catalyzed by mononuclear Ru(II) complexes. Inorg Chem 51(5):2930–2939 217. Duan L, Xu Y, Tong L, Sun L (2011) Ce(IV)- and light-driven water oxidation by [Ru(terpy) (pic)3]2+ analogues: catalytic and mechanistic studies. ChemSusChem 4(2):238–244 218. Maji S, Lopez I, Bozoglian F, Benet-Buchholz J, Llobet A (2013) Mononuclear rutheniumwater oxidation catalysts: discerning between electronic and hydrogen-bonding effects. Inorg Chem 52(7):3591–3593 219. Marelius DC, Bhagan S, Charboneau DJ, Schroeder KM, Kamdar JM, McGettigan AR, Freeman BJ, Moore CE, Rheingold AL, Cooksy AL, Smith DK, Paul JJ, Papish ET, Grotjahn DB (2014) How do proximal hydroxy or methoxy groups on the bidentate ligand affect [(2,20 ;60 ,200 -Terpyridine)Ru(N,N)X] water-oxidation catalysts? Synthesis, characterization, and reactivity at acidic and near-neutral pH. Eur J Inorg Chem 2014(4):676–689 220. Duan L, Fischer A, Xu Y, Sun L (2009) Isolated seven-coordinate Ru(IV) dimer complex with [HOHOH]() bridging ligand as an intermediate for catalytic water oxidation. J Am Chem Soc 131(30):10397–10399 221. Duan L, Xu Y, Gorlov M, Tong L, Andersson S, Sun L (2010) Chemical and photochemical water oxidation catalyzed by mononuclear ruthenium complexes with a negatively charged tridentate ligand. Chem Eur J 16(15):4659–4668 222. An J, Duan L, Sun L (2012) Ru complexes containing pyridine dicarboxylate ligands: electronic effects on their catalytic activity toward wateroxidation. Faraday Discuss 155:267–275 223. Tong L, Inge AK, Duan L, Wang L, Zou X, Sun L (2013) Catalytic water oxidation by mononuclear Ru complexes with an anionic ancillary ligand. Inorg Chem 52(5):2505–2518 224. Karkas MD, Akermark T, Johnston EV, Karim SR, Laine TM, Lee BL, Akermark T, Privalov T, Akermark B (2012) Water oxidation by single-site ruthenium complexes: using ligands as redox and proton transfer mediators. Angew Chem Int Ed Engl 51 (46):11589–11593 225. Tong L, Wang Y, Duan L, Xu Y, Cheng X, Fischer A, Ahlquist MS, Sun L (2012) Water oxidation catalysis: influence of anionic ligands upon the redox properties and catalytic performance of mononuclear ruthenium complexes. Inorg Chem 51(6):3388–3398 226. Duan L, Bozoglian F, Mandal S, Stewart B, Privalov T, Llobet A, Sun L (2012) A molecular ruthenium catalyst with water-oxidation activity comparable to that of photosystem II. Nat Chem 4(5):418–423 227. Duan L, Araujo CM, Ahlquist MS, Sun L (2012) Highly efficient and robust molecular ruthenium catalysts for water oxidation. Proc Natl Acad Sci U S A 109(39):15584–15588 228. Jiang Y, Li F, Zhang B, Li X, Wang X, Huang F, Sun L (2013) Promoting the activity of catalysts for the oxidation of water with bridged dinuclear ruthenium complexes. Angew Chem Int Ed Engl 52(12):3398–3401 229. Matheu R, Ertem MZ, Benet-Buchholz J, Coronado E, Batista VS, Sala X, Llobet A (2015) Intramolecular proton transfer boosts water oxidation catalyzed by a Ru complex. J Am Chem Soc 137(33):10786–10795

448

H.-T. Zhang and M.-T. Zhang

230. Singh A, Spiccia L (2013) Water oxidation catalysts based on abundant 1st row transition metals. Coord Chem Rev 257(17–18):2607–2622 231. Su XJ, Zheng C, Hu QQ, Du HY, Liao RZ, Zhang MT (2018) Bimetallic cooperative effect on O-O bond formation: copper polypyridyl complexes as water oxidation catalyst. Dalton Trans 47(26):8670–8675 232. Su X-J, Gao M, Jiao L, Liao R-Z, Siegbahn PEM, Cheng J-P, Zhang M-T (2015) Electrocatalytic water oxidation by a dinuclear copper complex in a neutral aqueous solution. Angew Chem 127(16):4991–4996 233. Du HY, Chen SC, Su XJ, Jiao L, Zhang MT (2018) Redox-active ligand assisted multielectron catalysis: a case of Co(III) complex as water oxidation catalyst. J Am Chem Soc 140 (4):1557–1565 234. Karkas MD, Akermark B (2016) Water oxidation using earth-abundant transition metal catalysts: opportunities and challenges. Dalton Trans 45(37):14421–14461 235. Limburg J, Vrettos JS, Liable-Sands LM, Rheingold AL, Crabtree RH, Brudvig GW (1999) A functional model for O-O bond formation by the O2-evolving complex in photosystem II. Science 283(5407):1524–1527 236. Limburg J, Vrettos JS, Chen H, de Paula JC, Crabtree RH, Brudvig GW (2001) Characterization of the O2-evolving reaction catalyzed by [(terpy)(H2O)MnIII(O)2MnIV(OH2)(terpy)] (NO3)3(terpy ¼ 2,20 :6,20 ‘-Terpyridine). J Am Chem Soc 123(3):423–430 237. Chen H, Tagore R, Olack G, Vrettos JS, Weng TC, Penner-Hahn J, Crabtree RH, Brudvig GW (2007) Speciation of the catalytic oxygen evolution system: [MnIII/IV2(mu-O)2(terpy)2 (H2O)2](NO3)3 + HSO5. Inorg Chem 46(1):34–43 238. Tagore R, Crabtree RH, Brudvig GW (2008) Oxygen evolution catalysis by a dimanganese complex and its relation to photosynthetic water oxidation. Inorg Chem 47(6):1815–1823 239. Tagore R, Chen H, Zhang H, Crabtree RH, Brudvig GW (2007) Homogenous water oxidation by a Di-mu-Oxo dimanganese complex in the presence of Ce. Inorg Chim Acta 360 (9):2983–2989 240. Kurz P, Berggren G, Anderlund MF, Styring S (2007) Oxygen evolving reactions catalysed by synthetic manganese complexes: a systematic screening. Dalton Trans 38:4258–4261 241. Baffert C, Romain S, Richardot A, Lepretre JC, Lefebvre B, Deronzier A, Collomb MN (2005) Electrochemical and chemical formation of [Mn4(IV)O5(terpy)4(H2O)2]6+, in relation with the photosystem II oxygen-evolving center model [Mn2(III,IV)O2(terpy)2(H2O)2]3+. J Am Chem Soc 127(39):13694–13704 242. Yagi M, Narita K (2004) Catalytic O2 evolution from water induced by adsorption of [(OH2) (terpy)Mn(mu-O)2Mn(terpy)(OH2)]3+ complex onto clay compounds. J Am Chem Soc 126 (26):8084–8085 243. Chen H, Faller JW, Crabtree RH, Brudvig GW (2004) Dimer-of-dimers model for the oxygenevolving complex of photosystem II. Synthesis and properties of [MnIV4O5(terpy)4(H2O)2] (ClO4)6. J Am Chem Soc 126(23):7345–7349 244. Chen H, Collomb MN, Duboc C, Blondin G, Riviere E, Faller JW, Crabtree RH, Brudvig GW (2005) New linear high-valent tetranuclear manganese-oxo cluster relevant to the oxygenevolving complex of photosystem II with oxo, hydroxo, and aqua coordinated to a single Mn (IV). Inorg Chem 44(25):9567–9573 245. Gao Y, Crabtree RH, Brudvig GW (2012) Water oxidation catalyzed by the tetranuclear Mn complex [Mn(IV)4O5(terpy)4(H2O)2](ClO4)6. Inorg Chem 51(7):4043–4050 246. Chen H, Tagore R, Das S, Incarvito C, Faller JW, Crabtree RH, Brudvig GW (2005) General synthesis of di-mu-oxo dimanganese complexes as functional models for the oxygen evolving complex of photosystem II. Inorg Chem 44(21):7661–7670 247. Yamazaki H, Igarashi S, Nagata T, Yagi M (2012) Substituent effects on core structures and heterogeneous catalytic activities of Mn(III)(mu-O)2Mn(IV) dimers with 2,20 :60 ,200 -terpyridine derivative ligands for water oxidation. Inorg Chem 51(3):1530–1539

The Application of Pincer Ligand in Catalytic Water Splitting

449

248. Michaelos TK, Lant HMC, Sharninghausen LS, Craig SM, Menges FS, Mercado BQ, Brudvig GW, Crabtree RH (2016) Catalytic oxygen evolution from manganese complexes with an oxidation-resistant N,N, O-donor ligand. ChemPlusChem 81(10):1129–1132 249. Lant HMC, Michaelos TK, Sharninghausen LS, Mercado BQ, Crabtree RH, Brudvig GW (2019) N,N,O pincer ligand with a deprotonatable site that promotes redox-leveling, high Mn oxidation states, and a Mn2 O2 dimer competent for catalytic oxygen evolution. Eur J Inorg Chem 2019(15):2115–2123 250. Rigsby ML, Mandal S, Nam W, Spencer LC, Llobet A, Stahl SS (2012) Cobalt analogs of Ru-based water oxidation catalysts: overcoming thermodynamic instability and kinetic lability to achieve electrocatalytic O2 evolution. Chem Sci 3(10) 251. El Wakkad SES, Hickling A (1950) The anodic behaviour of metals. Part VI.—cobalt. Trans Faraday Soc 46:820 252. Benson P, Briggs GWD, Wynne-Jones WFK (1964) The cobalt hydroxide electrode—I. Structure and phase transitions of the hydroxides. Electrochim Acta 9(3):275–280 253. Benson P, Briggs GWD, Wynne-Jones WFK (1964) The cobalt hydroxide electrode—II. Electrochemical behaviour. Electrochim Acta 9(3):281–288 254. Kanan MW, Nocera DG (2008) In situ formation of an oxygen-evolving catalyst in neutral water containing phosphate and Co2+. Science 321(5892):1072–1075 255. Surendranath Y, Dinca M, Nocera DG (2009) Electrolyte-dependent electrosynthesis and activity of cobalt-based water oxidation catalysts. J Am Chem Soc 131(7):2615–2620 256. Wang L-P, Van Voorhis T (2011) Direct-coupling O2 bond forming a pathway in cobalt oxide water oxidation catalysts. J Phys Chem Lett 2(17):2200–2204 257. Gimbert-Surinach C, Moonshiram D, Francas L, Planas N, Bernales V, Bozoglian F, Guda A, Mognon L, Lopez I, Hoque MA, Gagliardi L, Cramer CJ, Llobet A (2016) Structural and spectroscopic characterization of reaction intermediates involved in a dinuclear co-Hbpp water oxidation catalyst. J Am Chem Soc 138(47):15291–15294 258. Barnett SM, Goldberg KI, Mayer JM (2012) A soluble copper-bipyridine water-oxidation electrocatalyst. Nat Chem 4(6):498–502 259. Zhang MT, Chen Z, Kang P, Meyer TJ (2013) Electrocatalytic water oxidation with a copper (II) polypeptide complex. J Am Chem Soc 135(6):2048–2051 260. Chen F, Wang N, Lei H, Guo D, Liu H, Zhang Z, Zhang W, Lai W, Cao R (2017) Electrocatalytic water oxidation by a water-soluble copper(II) complex with a copper-bound carbonate group acting as a potential proton shuttle. Inorg Chem 56(21):13368–13375

Top Organomet Chem (2021) 68: 451–452 https://doi.org/10.1007/3418_2020_74 # Springer Nature Switzerland AG 2020 Published online: 28 January 2021

Correction to: Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems Seji Kim, Yeong-Eun Kim, and Yunho Lee

Correction to: Chapter “Metal-Ligand Cooperativity of Phosphorus-Containing Pincer Systems” in: Seji Kim et al., Top Organomet Chem, https://doi.org/10.1007/3418_2020_69 The original version of this chapter was revised with an addition of the below acknowledgement text for funding at the end of the Conclusion Section: This work was supported by C1 Gas Refinery Program (NRF2015M3D3 A1A01064880). The original chapter was corrected.

The updated online version of this chapter can be found at https://doi.org/10.1007/3418_2020_69