Time-Resolved Photoionisation Studies of Polyatomic Molecules: Exploring the Concept of Dynamophores [1st ed.] 9783030536282, 9783030536299

Citation preview

Springer Theses Recognizing Outstanding Ph.D. Research

Martin Alex Bjørnholst

Time-Resolved Photoionisation Studies of Polyatomic Molecules Exploring the Concept of Dynamophores

Springer Theses Recognizing Outstanding Ph.D. Research

Aims and Scope The series “Springer Theses” brings together a selection of the very best Ph.D. theses from around the world and across the physical sciences. Nominated and endorsed by two recognized specialists, each published volume has been selected for its scientific excellence and the high impact of its contents for the pertinent field of research. For greater accessibility to non-specialists, the published versions include an extended introduction, as well as a foreword by the student’s supervisor explaining the special relevance of the work for the field. As a whole, the series will provide a valuable resource both for newcomers to the research fields described, and for other scientists seeking detailed background information on special questions. Finally, it provides an accredited documentation of the valuable contributions made by today’s younger generation of scientists.

Theses are accepted into the series by invited nomination only and must fulfill all of the following criteria • They must be written in good English. • The topic should fall within the confines of Chemistry, Physics, Earth Sciences, Engineering and related interdisciplinary fields such as Materials, Nanoscience, Chemical Engineering, Complex Systems and Biophysics. • The work reported in the thesis must represent a significant scientific advance. • If the thesis includes previously published material, permission to reproduce this must be gained from the respective copyright holder. • They must have been examined and passed during the 12 months prior to nomination. • Each thesis should include a foreword by the supervisor outlining the significance of its content. • The theses should have a clearly defined structure including an introduction accessible to scientists not expert in that particular field.

More information about this series at http://www.springer.com/series/8790

Martin Alex Bjørnholst

Time-Resolved Photoionisation Studies of Polyatomic Molecules Exploring the Concept of Dynamophores Doctoral Thesis accepted by the University of Copenhagen, Denmark

123

Author Dr. Martin Alex Bjørnholst Department of Chemistry University of Copenhagen Copenhagen, Denmark

Supervisor Prof. Theis I. Sølling Department of Chemistry University of Copenhagen Copenhagen, Denmark

ISSN 2190-5053 ISSN 2190-5061 (electronic) Springer Theses ISBN 978-3-030-53628-2 ISBN 978-3-030-53629-9 (eBook) https://doi.org/10.1007/978-3-030-53629-9 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are solely and exclusively licensed 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

Supervisor’s Foreword

Femtochemistry brings together researchers from a range of different disciplines; however, those with a strong foundation in fundamental organic chemistry are scarce and this is where the thesis of Dr. Bjørnholst bridges a gap. Organic chemistry is rooted in the functional group, that is, the idea that reactivity can be predicted and utilised based on specific structural units in a molecule. Femtochemistry, on the other hand, is often approached on a case-to-case basis with a slightly more fundamental assessment starting from quantum chemical calculations. The overarching idea behind the dynamophore, a term coined by the Stolow group where Dr. Bjørnholst spent time during his studies, is that functional groups or combinations thereof will determine the initial nuclear motions (the dynamics) of molecules that are electronically excited. These initial motions are predetermining for the chain of events that follow the absorption of a photon, and therefore the aim is to ultimately be able to predict how the nuclei move and what photophysical or photochemical processes these motions trigger simply from a glimpse at the molecular structure—exactly as it is possible for an organic chemist to predict what a carbonyl group will do with an electrophile. The thesis of Dr. Bjørnholst is unique in the sense that it consistently follows this line of thought throughout. By attempting to assess a large range of the possible electronic transitions that can arise from exposing an organic functional group to a photon, the thesis provides a consistent overview of exactly how most functional groups in organic molecules induce specific nuclear motions that potentially lead to specific electronic transitions and eventually selective photoinduced bond breakage—something which has been considered the dream of most photochemists for decades. Copenhagen, Denmark May 2020

Prof. Theis I. Sølling

v

Abstract

Non-adiabatic transitions are central in the study of photoinduced dynamics of polyatomic molecules. Time-resolved photoelectron spectroscopy (TRPES) is a technique that is particularly sensitive to the changes in electronic structure, and thus non-adiabatic transitions between electronically excited states have been revealed in a vast number of molecules by TRPES studies. The non-radiative molecular response to excitation, e.g. internal conversion (IC), is often found to occur on the ultrafast timescale. The associated strong non-adiabatic couplings occur only along a small subset of nuclear degrees of freedom, and few structural changes are inherently responsible for ultrafast IC. The link between molecular structure and the excited state structural changes has previously been conceptualised as ‘dynamophores’. The VUV photoinduced dynamics of four cyclic ketones and one linear ketone is studied and exhibit qualitatively similar dynamics. The initially excited states have 3d Rydberg character but also display partial (p; p
) valence character. The observed excited state lifetimes are quantitatively similar, indicating that a common deactivation mechanism is associated with 3d Rydberg excitation in ketones and as such the dynamics are consistent with the dynamophore concept. The ring-opening and dissociative dynamics of cyclopropane are studied by a joint computational and experimental study. The computational results show that vertical excitation energies are inadequate to predict and assign the experimental absorption spectrum. The explicit inclusion of the electromagnetic field associated with a pump pulse is required to qualitatively reproduce the absorption spectrum. The ensuing dynamics are also simulated and show impressive quantitative agreement with the experimental results. Model systems for the disulfide bond and the peptide bond, which are both related to the structure of proteins, are additionally investigated. The dynamics are similar in the sense that a dense manifold of Rydberg states is present in both cases,

vii

viii

Abstract

and these efficiently couple with valence states, ultimately leading to dissociation. Finally, the potential ultrafast intersystem crossing (ISC) in three methylated benzene derivatives is investigated computationally, and the same structural change associated with IC is highlighted as being crucial to the potential ISC as well in all three compounds, consistent with a common dynamophore in the systems.

Preface

This thesis has been submitted to the Ph.D. School at the Faculty of Science at the University of Copenhagen as partial fulfilment of the requirements to obtain the Ph.D. degree in chemistry. The work has been carried out under the supervision of Prof. Theis I. Sølling at the Department of Chemistry at the University of Copenhagen during a 3-year period from September 2015 to 2018. The majority of this time was spent in the Femtolab at the University of Copenhagen, carrying out experiments and calculations. Part of the work was also carried out during a 7-month visit to the group of Prof. Albert Stolow at the National Research Council of Canada (NRC) and the University of Ottawa in 2016, with an additional 3-week re-visit in March 2018. A purely computational project was carried out during a 6-week visit to the group of Prof. Leticia González at the University of Vienna in late 2017. Copenhagen, Denmark

Martin Alex Bjørnholst

ix

Parts of this thesis have been published in the following journal articles: • Coherent Motion of Excited State Cyclic Ketones: The Have and The Have-Nots, M.A.B. Larsen, A.B. Stephansen, T.I. Sølling, Chem. Phys. Lett., (683), 495–499, 2017 • Vacuum Ultraviolet Excited State Dynamics of the Smallest Ring, Cyclopropane. II. Time-Resolved Photoelectron Spectroscopy and Ab Initio Dynamics, M.R. Coates, M.A.B. Larsen, R. Forbes, S.P. Neville, A.E. Boguslavskiy, I. Wilkinson, T.I. Sølling, R. Lausten, A. Stolow, M.S. Schuurman, J. Chem. Phys., (149), 144311, 2018 • Putting the Disulfide Bridge at Risk: How UV-C Radiation Leads to Ultrafast Rupture of the S-S Bond, M.A.B. Larsen, A.B. Skov, C.M. Clausen, J. Ruddock, B. Stankus, P.M. Weber, T.I. Sølling, ChemPhysChem, (19), 2829– 2834, 2018 • Vacuum Ultraviolet Excited State Dynamics of Small Amides, M.A.B. Larsena, T.I. Sølling, R. Forbesa, A.E. Boguslavskiy, V. Makhija, K. Veyrinas, R. Lausten, A. Stolow, M.M. Zawadskia, L. Saalbacha, N. Kotsina, M.J. Patersen, D. Townsend J. Chem. Phys., (150), 054301, 2019 (ajoint lead investigators)

xi

Acknowledgements

The work presented in thesis would not have been possible without the input, dedication and support from many great people. First and foremost I must thank my academic supervisor Prof. Theis I. Sølling for providing me with the opportunity to pursue a Ph.D. in the field of ultrafast science and my visits to distant places would not have happened without the freedom inherent to Theis’ supervision. Being in Theis’ group has also meant sharing offices with the outstanding people in Theis’ group and I am thankful for the many good times in the company of Alberto, Anders, Anne, Benjamin, Ditte, Kristin, Liv, and Max as well as all former group members that I have had the chance to meet. Especially Anne paved the way for me when I started in the group during my bachelor’s and made me feel most welcome in a crowd of talented people—and I am very thankful for our continued friendship. I am sincerely grateful to Prof. Albert Stolow for letting me visit his group and thereby giving me the possibility to learn from all the amazing people at the NRC. The many discussions on diverse subjects and field trips made my stay absolutely joyous thanks to excellent people like Andrey, Coates, Hockett, Kévin, Ruaridh, Rune, Ryan, Schuurman, Simon, Spanner, Varun and many more. I must especially thank Andrey for his ability to accommodate a solution to almost any technical request and Varun for his patient manners and his attempts to teach me some physics. I am also grateful to Spanner for breaking his own promise of never again spending time with temporary visitors to make friends and to Rune for providing opportunities to enjoy Danish liquorice, cross-country running, skiing and sailing. Last but not least, I probably spent the majority of my time in Ottawa in the company of Ruaridh and even late nights collecting data became tolerable (if not habit) in his dedicated and humorous company. I am also grateful to Prof. Leticia González and her friendly group for showing me the computational chemists’ perspective and I in particular owe thanks to Clemens, Maximilian, Nico, Patrick, Pedro and Sebastian for doing their best to teach me the basics of multi-reference quantum chemistry and their use in

xiii

xiv

Acknowledgements

simulations. Finally, I must thank my friends and family who have not been overly encumbered by my company during the last few years. I am most thankful to my parents, my brother and my sister for always encouraging me to travel and experience new things when opportunities arose, even when it meant leaving Denmark for extended periods of time. Lastly, I am eternally grateful for the never ending support from my dear Elisabeth.

Contents

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

1 2 3 6

2 Experimental Methods . . . . . . . . . . . . . . . . . . . 2.1 Time-Resolved Photoelectron Spectroscopy . 2.2 Velocity Map Imaging . . . . . . . . . . . . . . . . 2.2.1 Matrix Inversion . . . . . . . . . . . . . . . 2.3 Photoelectron Angular Distributions (PADs) 2.3.1 VMI Spectrometers and Laser Setups 2.3.2 Global Fitting Scheme . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

9 9 11 13 15 18 27 28

3 Theoretical Methods . . . . . . . . . . . . . . 3.1 Born-Oppenheimer Approximation 3.2 Potential Energy Curves . . . . . . . . 3.3 Electronic Structure Methods . . . . . 3.3.1 Density Functional Theory . 3.3.2 CASSCF . . . . . . . . . . . . . . 3.3.3 CASPT2 . . . . . . . . . . . . . . 3.3.4 Coupled-Cluster Methods . . 3.4 Surface Hopping . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

31 31 32 33 34 35 36 36 37 38

4 Results and Discussion . . . . . . . . . . . . 4.1 Cyclic Ketones . . . . . . . . . . . . . . . 4.1.1 Motivation . . . . . . . . . . . . 4.1.2 The Initially Excited States . 4.1.3 Photoelectron Spectra . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

41 41 41 42 43

1 Introduction . . . . . . . . . . . . . . . . . . 1.1 Rydberg States . . . . . . . . . . . . . 1.2 The Concept of a Dynamophore References . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

xv

xvi

Contents

4.1.4 4.1.5 4.2 VUV 4.2.1 4.2.2 4.2.3 4.2.4 4.2.5 4.2.6 4.2.7

Linear Interpolations . . . . . . . . . . . . . . . . . . Summarising Discussion . . . . . . . . . . . . . . . Excited Ketones . . . . . . . . . . . . . . . . . . . . . . Motivation . . . . . . . . . . . . . . . . . . . . . . . . . The Initially Excited States . . . . . . . . . . . . . . Ionic State Calculations . . . . . . . . . . . . . . . . 3-Pentanone . . . . . . . . . . . . . . . . . . . . . . . . Cyclopentanone . . . . . . . . . . . . . . . . . . . . . . 2-Methylcyclopentanone . . . . . . . . . . . . . . . . Photoelectron Angular Distributions for VUV Ketones . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.8 Summarising the Ketone Dynamics . . . . . . . 4.3 Cyclopropane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2 The Initially Excited States . . . . . . . . . . . . . . 4.3.3 Experimental Photoelectron Spectrum . . . . . . 4.3.4 Summarising Discussion . . . . . . . . . . . . . . . 4.4 1,2-Dithiane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.2 The Initially Excited State . . . . . . . . . . . . . . 4.4.3 Time-Resolved Photoelectron Spectrum . . . . 4.4.4 SHARC Dynamics . . . . . . . . . . . . . . . . . . . . 4.4.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Amides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . 4.5.2 Calculations . . . . . . . . . . . . . . . . . . . . . . . . 4.5.3 Formamide HC(¼O)NH2 . . . . . . . . . . . . . . . 4.5.4 N,N-Dimethylformamide HC(¼O)N(CH3 )2 . . 4.5.5 N,N-Dimethylacetamide CH3 C(¼O)N(CH3 )2 . 4.5.6 Photoelectron Angular Distributions for VUV Amides . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5.7 Summarising the Amide Dynamics . . . . . . . . 4.6 Benzenes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . 4.6.2 MS-CASPT2 Potential Energy Curves . . . . . 4.6.3 Spin-Orbit Couplings . . . . . . . . . . . . . . . . . . 4.6.4 SHARC Dynamics . . . . . . . . . . . . . . . . . . . . 4.6.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

46 48 49 50 51 52 53 56 58

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

59 62 63 64 64 68 72 73 73 75 77 82 84 86 86 88 93 96 98

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

100 103 105 105 107 109 113 114 116

Excited . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

Excited . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

5 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 5.1 Rydberg-Valence Interactions in Molecules . . . . . . . . . . . . . . . . . 123 Appendix: Cartesian Coordinates of Optimised and Interpolated Molecular Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

Abbreviations

B.O. BBO CCD CH CI CP DEDS DMA DMF E-L FC fs/ps/ns/ls FOR FWHM IC IE ISC MCP MECP PAD PEL PEC REMPI SOC TRPES TRMS VMI VUV

Born-Oppenheimer bbarium borate Charge-coupled device Cyclohexanone Conical intersection Cyclopentanone Diethyldisulfide N,N-Dimethylacetamide N,N-Dimethylformamide Even-Lavie Franck-Condon Femto-/pico-/nano-/microsecond Formamide Full-width at half-maximum Internal conversion Ionisation energy Intersystem crossing Multi-channel plate Minimum energy crossing point Photoelectron angular distribution Potential energy landscape Potential energy curve Resonance-enhanced multi-photon ionisation Spin-orbit coupling Time-resolved photoelectron spectroscopy Time-resolved mass spectrometry Velocity map imaging Vacuum ultraviolet

xvii

xviii

1,2-DT 2-MCP 2-MCP 3P

Abbreviations

1,2-dithiane 2-Methylcyclopentanone 3-Methylcyclopentanone 3-Pentanone

List of Figures

Fig. 1.1

Fig. 2.1

Fig. 2.2 Fig. 2.3

Fig. 2.4

Fig. 2.5

Fig. 2.6 Fig. 2.7

Illustration of the structural changes in excited states of nitrobenzene and thus the nitro dynamophore. (Figure adapted from [24] with permission. Copyright (2011) American Chemical Society) . . . . . . . . . . . . . . . . . . . . Illustration of Koopmans’ correlation and its effect on preferential ionisation to either D0 or D1 depending on the electronic configuration of the molecular core. Only the electron in the excited state (red) is affected, and the core is left unchanged . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Time-, energy- and angle-resolved yields can be combined in TRPES, allowing for more differential information . . . . . . . Sketch of a standard three-electrode VMI spectrometer. The molecular beam (MB) and pump-probe laser pulses overlap in the interaction region (indicated by a blue sphere), producing photoelectrons with distinct kinetic energies (here two different kinetic energies are illustrated) which are focused by onto a position-sensitive detector with a certain radius from the centre . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Raw VMI image (left) and Abel inverted image (right) of cyclopropane at Dt ¼ 0 for a 160 nm pump and 267 nm probe. The images have been background subtracted for onecolour signals and fourfold symmetrised. The polarisation axis of the pump and probe pulses were along the vertical direction in the images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The partitioning of the detector plane used for the Abel inversion technique. Equidistant lines (j) and lines of radius r (along i) segment the plane into areas denoted Aij . . . . . . . . Schematic of the Copenhagen laser setup . . . . . . . . . . . . . . . . Schematic of the 4x box with optics and polarisations indicated in the figure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

..

5

..

10

..

11

..

12

..

14

.. ..

15 18

..

19

xix

xx

Fig. 2.8

Fig. 2.9

Fig. 2.10

Fig. 2.11 Fig. 2.12

Fig. 2.13

Fig. 2.14

Fig. 2.15

Fig. 2.16

List of Figures

Sketch of the VMI setup at the University of Copenhagen. The molecular beam (MB) produced by the Even-Lavie (E-L) valve is skimmed before entering the interaction (main) chamber. The photoelectron produced by ionisation of the pump-probe laser pulses travel inside a Mu metal tube to shield the photoelectrons from interference due to magnetic background fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total ion count transient of the 200 nm + 400 nm (1 þ 20 ) nonresonant ionisation signal of xenon. The Gaussian fit indicates the time-zero position at 0.14 ps, and the FWHM of the fit is 144 fs. The corresponding (1 þ 10 ) FWHM with the same pffiffi pulses is 166 fs, when scaled by the appropriate factor of 23 . . . Energy (2.10a) and phase-matching (2.10b) diagram for the non-collinear four-wave mixing process employed to generate the fifth harmonic, i.e. the 160nm fs VUV pulse . . . . . . . . . . Schematic of the rare gas-filled box used for fifth-harmonic generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Spectra of the 160 nm VUV pulse measured on three separate occasions and Gaussian fits to the data. Note that the wavelength axis of the spectrometer has not been calibrated and the central wavelengths cannot be trusted to be exact . . . Sketch of the VMI setup in Ottawa at the National Research of Canada (NRC) illustrating the baffles that limit the background from scatter immensely. The Einzel lens assembly allowing compression and extension of the produced photoelectrons or ions. The short flight path, optimised for photoelectrons, to reach the detector and the achromatic relay imaging of the phosphor screen to the top of the time-of-flight (TOF) tube. Adapted with permission from [54] . . . . . . . . . . . Raw image illustrating the three contributions to the signal: 1 þ 10 , 3267 nm, and 2160 nm non-resonant ionisation of Xe (2.14a) and the raw image with time-independent 160 nm only contributions subtracted (2.14b) . . . . . . . . . . . . . . . . . . . Energy calibration from 267 nm only ionisation of xenon (2.10a) and determination of VUV central wavelength based on the 1 þ 10 ionisation of xenon (2.10b). The latter displays time-independent contributions from pump alone and probe alone signals which are subtracted in all other data presented throughout but are included here . . . . . . . . . . . . . . . . . . . . . . Cross-correlation from the 1 þ 10 non-resonant ionisation of xenon with 160 nm + 267 nm. The Gaussian fit indicates the time-zero position at 0.03 ps, and the FWHM of the fit is 108 fs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

..

20

..

21

..

23

..

23

..

24

..

25

..

26

..

26

..

27

List of Figures

Fig. 3.1

Fig. 3.2

Fig. 4.1 Fig. 4.2 Fig. 4.3 Fig. 4.4

Fig. 4.5 Fig. Fig. Fig. Fig.

4.6 4.7 4.8 4.9

Fig. 4.10 Fig. 4.11

Fig. 4.12 Fig. 4.13 Fig. 4.14 Fig. 4.15

Fig. 4.16 Fig. 4.17

Adiabatic Born-Oppenheimer potential energy curves (left) illustrating that transitions between states cannot occur unless their non-adiabatic coupling is taken into account. Conversely, the diabatic potential energy curves (right) are associated with a specific orbital character and may cross each other . . . . . . . Sketch of the active space principle in CASSCF. Most electrons in doubly occupied orbitals are inactive (blue), while the electrons in the active space (red) sample every possible configuration within the active space—here a (2,3) active space with two electrons in three orbitals. No virtual orbitals outside of the active space are sampled . . . . . . . . . . . Molecular structures of the four cyclic ketones investigated and their abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The TRPES for the cyclic ketones . . . . . . . . . . . . . . . . . . . . . Fits, integrated traces and Fourier transform of the residuals (shown in insets) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Linear interpolations from the FC region to the S2 minima and the resulting D0 S2 energy differences summarised in Fig. 4.4f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Molecular structures of the three ketones investigated and their abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The experimental data and associated fit for 3-pentanone . . . . The experimental data and fit for cyclopentanone . . . . . . . . . . The experimental data and fit for 2-methylcyclopentanone . . . PADs for 3-pentanone, cyclopentanone and 2methylcyclopentanone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Molecular structure of cyclopropane and definitions of the two most important modes . . . . . . . . . . . . . . . . . . . . . . Experimental and simulated absorption spectra of cyclopropane, showing excellent agreement when nonadiabatic couplings between the bright and dark Rydberg states are included. Adapted from [78], with permission of AIP Publishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results from AIMS simulations . . . . . . . . . . . . . . . . . . . . . . . DAS and histogram for AIMS simulations of cyclopropane . . The experimental data and fit for cyclopropane excited at 160 nm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The molecular structures of the cyclic disulphide 1,2-dithiane and the related open-chain analogue diethyldisulphide, along with their abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparison of EOM-CCSD and TD-CAM-B3LYP orbitals for 1,2-dithiane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The simulated and experimental absorption spectrum for 1,2-dithiane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xxi

..

33

..

35

.. ..

42 44

..

45

..

47

. . . .

. . . .

49 54 56 59

..

61

..

63

.. .. ..

66 69 70

..

71

..

73

..

76

..

77

xxii

List of Figures

Fig. 4.18 Fig. 4.19

Fig. 4.20 Fig. 4.21 Fig. 4.22 Fig. 4.23 Fig. 4.24 Fig. Fig. Fig. Fig. Fig. Fig.

4.25 4.26 4.27 4.28 4.29 4.30

Fig. 4.31 Fig. 4.32 Fig. 4.33 Fig. 4.34

The experimental data and models for 1,2-dithiane . . . . . . . . . The Brown dataset on 1,2-dithiane excited at 206 nm displays the same features and qualitative behaviour observed in the CPH dataset with excitation at 201 nm . . . . . EOM-CCSD potential energy curves for 1,2-dithiane . . . . . . . SHARC populations and the presumed coupling mode in 1,2-dithiane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Molecular structures of the three investigated amides and their abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TD-CAM-B3LYP/aug-cc-pVDZ adiabatic potential energy curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diabatic PECs, EOM-CCSD and TD-CAM-B3LYP for FOR, DMF, and DMA . . . . . . . . . . . . . . . . . . . . . . . . . . . The experimental data and fit for formamide . . . . . . . . . . . . . The experimental data and fit for DMF . . . . . . . . . . . . . . . . . The experimental data and fit for DMA . . . . . . . . . . . . . . . . . PADs for formamide, DMF and DMA . . . . . . . . . . . . . . . . . . Molecular structures of the three benzenes investigated . . . . . MS-(5S,6T)-CASPT2(6,6)/cc-pVDZ adiabatic potential energy curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The orbitals associated with the (6,6) active space of toluene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MS-(5S,6T)-CASPT2(X,Y) spin-orbit coupling strengths for toluene as a function of active space . . . . . . . . . . . . . . . . . The orbitals involved in the larger (14,10) active space for toluene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results from SHARC simulations on benzene including the possibility of single-triplet transitions . . . . . . . . . . . . . . . .

..

79

.. ..

81 82

..

83

..

86

..

91

. . . . . .

. 93 . 94 . 97 . 100 . 102 . 106

. . 108 . . 109 . . 112 . . 112 . . 114

List of Tables

Table 4.1

Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 4.7

Table 4.8

Table 4.9

Table 4.10

Comparison of experimental and calculated values for vertical excited state energies of the S2 state and vertical ionisation energies for the investigated cyclic ketones . . . . . Timescales for the cyclic ketones excited to the (n,3s) states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vertical excitation energies for 3-pentanone . . . . . . . . . . . . D0 and D1 ionisation energies and characters . . . . . . . . . . . Fitted timescales for the VUV excited ketone IC dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cyclopropane vertical excitation calculations. All energies are given in eV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vertical excitation energies for the first 10 excited states in 1,2-dithiane comparing EOM-CCSD, TD-DFT and experimental values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Computational and experimental vertical excitation energies of FOR (with calculated oscillator strengths in parentheses). The states have significant mixed character and especially the two states near 160 nm (7.75 eV) are very heavily mixed. Note that A and B are characters based on an NTO analysis whereas C is based on the contributions without any postprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Computational vertical excitation energies of DMF and DMA (with calculated oscillator strengths in parentheses). States in the VUV region are heavily mixed, so the reported character is the main contributor at a given energy. Note that A and B are characters based on an NTO analysis whereas C is based on the contributions without any post-processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Spin-orbit couplings calculated for the four different toluene MECPs (given in cm1 ) . . . . . . . . . . . . . . . . . . . . . . . . . . .

..

42

.. .. ..

48 52 53

..

62

..

65

..

75

..

89

..

98

. . 110

xxiii

xxiv

Table 4.11 Table 4.12 Table A.1 Table A.2 Table A.3 Table A.4 Table A.5 Table A.6 Table A.7 Table A.8 Table A.9 Table A.10 Table A.11 Table A.12 Table A.13 Table A.14 Table A.15 Table A.16 Table A.17 Table A.18 Table A.19

List of Tables

Spin-orbit couplings calculated for the two para-xylene MECPs with two different basis sets (given in cm1 ) . . . . . Comparison of spin-orbit couplings for the three different molecules (given in cm1 ). . . . . . . . . . . . . . . . . . . . . . . . . . Cyclopentanone ground state minimum energy structure (CAM-B3LYP/6-311+g(d,p)). . . . . . . . . . . . . . . . . . . . . . . . Cyclopentanone S2 minimum energy structure (CAM-B3LYP/6-311+g(d,p)). . . . . . . . . . . . . . . . . . . . . . . . 2-methylcyclopentanone ground state minimum energy structure (CAM-B3LYP/6-311+g(d,p)) . . . . . . . . . . . . . . . . 2-methylcyclopentanone S2 minimum energy structure (CAM-B3LYP/6-311+g(d,p)). . . . . . . . . . . . . . . . . . . . . . . . 3-methylcyclopentanone ground state minimum energy structure (CAM-B3LYP/6-311+g(d,p)) . . . . . . . . . . . . . . . . 3-methylcyclopentanone S2 minimum energy structure (CAM-B3LYP/6-311+g(d,p)). . . . . . . . . . . . . . . . . . . . . . . . Cyclohexanone ground state minimum energy structure (CAM-B3LYP/6-311+g(d,p)). . . . . . . . . . . . . . . . . . . . . . . . Cyclohexanone S2 minimum energy structure (CAM-B3LYP/6-311+g(d,p)). . . . . . . . . . . . . . . . . . . . . . . . 3-pentanone ground state minimum energy structure (PBE0/aug-cc-pVDZ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benzene ground state minimum energy structure (MP2/aug-cc-pVDZ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benzene S2 -S1 minimum energy crossing point (MS-5-CASPT2(6,6)/cc-pVDZ) . . . . . . . . . . . . . . . . . . . . . . Toluene ground state minimum energy structure (MP2/aug-cc-pVDZ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Toluene S2 -S1 minimum energy crossing point for the ipso carbon (MS-5-CASPT2(6,6)/cc-pVDZ) . . . . . . . . . . . . . . . . Toluene S2 -S1 minimum energy crossing point for the ortho carbon (MS-5-CASPT2(6,6)/cc-pVDZ) . . . . . . . . . . . . . . . . Toluene S2 -S1 minimum energy crossing point for the meta carbon (MS-5-CASPT2(6,6)/cc-pVDZ) . . . . . . . . . . . . . . . . Toluene S2 -S1 minimum energy crossing point for the para carbon (MS-5-CASPT2(6,6)/cc-pVDZ) . . . . . . . . . . . . . . . . Para-xylene ground state minimum energy structure (MP2/aug-cc-pVDZ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Para-xylene S2 -S1 minimum energy crossing point for the ipso carbon (MS-5-CASPT2(6,6)/cc-pVDZ) . . . . . . Para-xylene S2 -S1 minimum energy crossing point for the a-carbon (MS-5-CASPT2(6,6)/cc-pVDZ) . . . . . . . . .

. . 111 . . 113 . . 128 . . 128 . . 129 . . 129 . . 130 . . 130 . . 131 . . 131 . . 132 . . 132 . . 133 . . 133 . . 134 . . 134 . . 135 . . 135 . . 136 . . 137 . . 138

Chapter 1

Introduction

The molecular response to photoexcitation is studied herein by ultrafast spectroscopy. The vibrational movement of the atomic nuclei in molecules occurs on a femtosecond (fs) timescale as exemplified by typical values for C−H stretching (∼10 fs/3000 cm−1 ) and C=O stretching (∼20 fs/1700 cm−1 ). Investigations on the fs timescales thus open the possibility of determining which molecular modes dictate the ensuing photochemistry and photophysics after excitation [1], and the late Prof. Ahmed Zewail was awarded the 1999 Nobel prize in Chemistry in recognition of his pioneering studies into ultrafast spectroscopy. The ultrashort timescale enables experiments that study not only reactants and products but also the extremely shortlived structures in between known as transition states and certain ultrafast spectroscopic techniques have therefore also been referred to as transition state spectroscopy [2–5]. Firstly, two concepts at the core of this thesis are introduced in the following sections: Rydberg states and the dynamophore concept. In Chap. 2, the experimental technique of choice, time-resolved photoelectron spectroscopy, is described along with the technical details associated with performing, analysing and modelling the outcome of such experiments. Next, in Chap. 3, the theoretical methods underlying the calculations of electronic structures are introduced along with the Born-Oppenheimer approximation which makes such calculations on polyatomic molecules possible in the first place. The significance of terms such as diabatic, adiabatic and non-adiabatic is introduced in Chap. 3 as well along with the potential energy curves and the concepts involved in simulations of the molecular response to photoexcitation by surface hopping methods. The experimental and theoretical results are gathered in Chap. 4, starting with the investigation of the lowest Rydberg state in cyclic ketones in Sect. 4.1, followed by investigations of higher lying Rydberg state in ketones (cyclic as well as linear) in Sect. 4.2. The ring-opening and mixed Rydberg-valence character in the first bright absorption band of cyclopropane is investigated in Sect. 4.3 before the Rydberg and σ ∗ valence mixing in the disulfide 1,2-dithiane is investigated in Sect. 4.4. The heavily mixed Rydberg-valence excited states of three amides (including both (π, π ∗ ) and σ ∗ valence states) are © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 M. A. Bjørnholst, Time-Resolved Photoionisation Studies of Polyatomic Molecules, Springer Theses, https://doi.org/10.1007/978-3-030-53629-9_1

1

2

1 Introduction

investigated in Sect. 4.5. Finally, a purely theoretical investigation into the potential ultrafast singlet-triplet intersystem crossing (ISC) in benzene, toluene and p-xylene are investigated in Sect. 4.6. Chapter 5 contains a summarising conclusion across investigations, and how the studies have elucidated the dynamophore concept.

1.1 Rydberg States The majority of the excited states investigated in this thesis have Rydberg character, and a brief introduction to these states is presented here. Historically, the atomic Rydberg formula was derived to describe the regularity of absorption lines in atomic spectra and thus closely connected to the birth of quantum mechanics and the idea of discreet energy levels [6]. Rydberg states are thus often considered like the hydrogen atom, an extended electronic orbital, orbiting a positive ion core. For hydrogen, the core is a single proton, whereas for molecules the core still contains many electrons but is reminiscent of a cationic species. It may appear surprising that Rydberg states are even present in molecules at all if small saturated molecules without any hetero atoms are considered, e.g. methane and ethane. From a molecular orbital perspective, the outer atomic orbitals of carbon and/or hydrogen are combined to form bonding and anti-bonding σ-orbitals. Excitation of any valence electron, whether it is associated with the C−C bond or the C−H bond, should lead to immediate dissociation as only anti-bonding σ ∗ -orbitals would be expected to be available. Nevertheless, Rydberg states are present and interact with the dissociative valence states in both molecules [7, 8]. The ubiquitous presence of Rydberg states in molecules has lead to several reviews on the properties of molecular Rydberg states [6, 9–11]. In general, the energy of a given molecular Rydberg state E n can be expressed as En = I E −

R∞ , (n − δ)2

(1.1)

where R∞ is the universal Rydberg constant (∼13.6 eV), I E is the molecular ionisation energy, n is the principal quantum number and δ is the quantum defect. A large quantum defect may indicate a less atomic-like molecular Rydberg state and the picture of an extended electronic Rydberg orbital, orbiting a positive ion core, may not hold. In photoelectron spectroscopy studies of molecular Rydberg states, the photoelectron energies have previously been reported in binding energies (i.e. the fraction in Equation 1.1, see, e.g. [12, 13]) rather than the kinetic energies of the photoelectrons. This may facilitate easy comparison between different experiments using different wavelengths for excitation (pump wavelength) and subsequent ionisation (probe wavelength). However, reporting binding energies requires well-established ionisation energies and for some of the molecular species considered herein, the exact ionisation energies are simply not known. On the other hand, photoelectron

1.1 Rydberg States

3

kinetic energies allow estimation of the ionisation energy when pump and probe wavelengths are reported. Part of the studies reported herein utilises an fs VUV source, and thus higher members of the Rydberg manifold are excited than with typical UV-Vis fs sources. The difference in behaviour of high- and low-lying Rydberg states may therefore be considered. Although the distinction between high-lying and low-lying Rydberg states is somewhat diffuse, the range of low-lying Rydberg states covers excitation of states with principal quantum numbers, n, in the range from three to ten. These lowlying Rydberg states generally penetrate the ion core significantly and consequently interact with the molecular valence states. High-lying Rydberg states have n >> 100 and are accessed, for instance, in zero kinetic energy (ZEKE) photoelectron spectroscopy [14]. These states have lifetimes on the order of μs, while low-lying Rydberg states have typical lifetimes on the fs-ps timescale. All Rydberg states investigated herein are thus comfortably within the low-lying range (n = 3, 4). As already mentioned, Rydberg states are ubiquitous in molecules and their interaction with valence states continues to be a very active field of study. Recent reviews highlight the interaction between Rydberg and dissociative σ ∗ states which may have relevance for the dynamics in complex systems such as DNA/RNA and amino acids [15, 16].

1.2 The Concept of a Dynamophore Photophysics and photochemistry cover many complex processes that have attained their own conceptualisation. Examples include phenomena such as internal conversion (IC) (a non-adiabatic process involving the nuclear motions coupling electronic and vibrational degrees of freedom), intersystem crossing (ISC) (additionally involving spin-orbit coupling between the electronic spin and the orbital angular momentum leading to a flip in the electronic spin), luminescence (spontaneous emission of light from excited states such as fluorescence and phosphorescence) and intramolecular vibrational energy redistribution (IVR) (coupling of vibrational energy from a subset of nuclear degrees of freedom to a larger and possibly complete set of vibrational degrees of freedom). These concepts are taught in introductory photochemical courses and really should be considered common knowledge within the field of photophysics and photochemistry. An interesting question is now whether the new concept ‘dynamophore’, as defined below, belongs among these well-known concepts and should be added to the introductory curriculum in photophysics. It is not the purpose of this thesis to answer this question. However, the question is undoubtedly tied to the relevance and use of the concept. This thesis undeniably contributes to the latter and tries to explore the former, and so an indirect answer to the question will be contained within this thesis. Generally, conceptualisations must rely on commonly accepted knowledge or phenomena in order to effectively communicate information. Thus the ‘dynamophore’

4

1 Introduction

concept also relies on well-established knowledge. It is, for instance, well known that the symmetry of small molecules dictates which subset of the available vibrational modes (those of the right symmetry) can couple electronic states of a given symmetry. This has furthermore led to the related conclusion that only a subset of vibrational modes are important when describing transitions between electronic states on an ultrafast timescale. Due to this selectivity non-radiative transitions between states are non-ergodic, that is, non-statistical in terms of the distribution of energy in the modes. It has been pointed out that if the energy was distributed ergodically/statistically between the vibrational modes, the transition rates could not be as short as the experimental evidence suggests in a number of cases [17, 18]. A particularly relevant analogue of the term ‘dynamophore’ is the chromophore— a localised part of the molecule (an atom or a subset of bonded atoms) which is responsible for the absorption of light at a particular wavelength. The remainder of the atoms in the molecule are considered spectators. The IUPAC definition of a chromophore is ‘the part of a molecular entity in which the electronic transition responsible for a given spectral band is approximately localised’ [19]. This in principle allows the possibility of tailoring different chromophores together, provided that the synthesis of a compound containing multiple chromophores does not significantly alter the original chromophores (i.e. induce significant changes in the electronic structure). Similarly, the term dynamophore refers to ‘the part of the molecule where the (excited state) dynamics are approximately localised’ as ‘the non-adiabatic transitions between electronic states often arise from a small subset of distinct molecular displacements’ [20]. The active dynamophore may thus change depending on excitation wavelength or multiple dynamophores may be in play at a given excitation wavelength. Both of these features are analogous to the chromophore. Additionally, the chromophore allows scientists to predict which types of transitions will be present simply by inspection of the molecular structure and, with a little experience, predict the relative ordering of these electronic transitions. The dynamophore holds the same simplifying potential of predicting excited state dynamics simply from inspection of the molecular structure although knowledge of the relative ordering of electronic states also appears to be crucial in cases of higher excited states as explored in later chapters. The dynamophore concept was introduced in 2011 by the Stolow group [20], but the term has not gained widespread use in the field of ultrafast spectroscopy. Nonetheless, it is used as a conceptual framework in which to analyse the molecular photoinduced dynamics in this thesis. The dynamophores introduced by the original paper include the C=C double bond in ethylene and cyclohexene (spatially similar to the (π, π ∗ ) chromophore), one of the C=C double bonds in 1,3-butadiene and 1,4-cyclohexadiene (spatially smaller than the (π, π ∗ ) chromophore) and the whole carbon skeleton of 1,3-cyclohexadiene (spatially larger than the (π, π ∗ ) chromophore). These chemical subunits are identified as dynamophores because they are associated with specific nuclear motions, e.g. a twist pyramidalisation and a [1,2] hydrogen bridge/migration associated with the C=C double bond in ethylene and (one of the C=C double bonds) in 1,3-butadiene and 1,4-cyclohexadiene, a [1,3] hydrogen migration associated with the allyl group

1.2 The Concept of a Dynamophore

5

Fig. 1.1 Illustration of the structural changes in excited states of nitrobenzene and thus the nitro dynamophore. (Figure adapted from [24] with permission. Copyright (2011) American Chemical Society)

(CH=CH−CH2 ) in cyclohexene and 1,4-cyclohexadiene and finally ring-opening associated with the whole carbon skeleton in cyclohexene and the cyclohexadienes. These examples were all presented in the introductory paper, and since further examples have not been highlighted as dynamophores by other groups, it is appropriate to briefly describe an additional example not associated with ethylene-like systems and double bonds: The nitro group. Investigations into nitrobenzene and larger aromatic systems containing nitro groups all point to the same cardinal motions of the nitro group as key in excited state IC and ISC. In nitrobenzene, the chromophore associated with the S1 (n,π ∗ ) transition is delocalised across the entire molecule— covering the nitro group as well as the π-system of the benzene ring—but the structural changes leading to changes in the electronic state are almost exclusively associated with changes in the nitro group [21–25]. The three important structural changes in the nitro group can be described as ONO scissoring angle (α), out-of-plane rotation (φ) and pyramidalisation (τ )—see Fig. 1.1. The motion leading to IC and ISC is complex and consists of a combination of the three modes occurring simultaneously as shown in Fig. 1.1. Here, the left-most structure is the ground state equilibrium geometry of nitrobenzene, the middle structure is the S1 excited state minimum and the right-most structure is the S1 -S0 conical intersection (CI). The nitro group can thus be identified as another example of a dynamophore. The same combination of spatially localised modes is crucial for the ultrafast ISC observed in larger aromatic systems containing nitro groups [26–31]. These systems are also called nitrated polycyclic aromatic hydrocarbons (NPAHs). The extended π-systems of the larger NPAHs means that the (π, π ∗ ) transition is lower in energy than the (n,π ∗ ) transition, and thus an additional structural change associated with the planar π-system is introduced. The primary change is an in-plane deformation resembling the structure of a quinone [26–31]. Such quinoid changes are well established in benzene analogues [32, 33] (potentially being a dynamophore itself) and can thus be considered separate from the nitro dynamophore because the latter occurs in cases of both (n,π ∗ ) and (π, π ∗ ) transitions, whereas the quinoid deformation is clearly associated with (π, π ∗ ) transitions. The change in chromophore when going from nitrobenzene to larger NPAHs highlights that the active dynamophore

6

1 Introduction

is tightly connected with the chromophore and thereby dependent on the excitation wavelength. Finally, a few comments on what will not be considered a dynamophore in this thesis. Generally, ring-opening processes involve the entire carbon skeleton and do not appear as localised changes on a subset of atoms. This is exemplified by the sub-100 fs ring-opening of 1,3-cyclohexadiene (with two double bonds) leading to 1,3,5-hexatriene (with three double bonds) and thus a change in at least one of the crucial structural parameters (bond lengths, angles and dihedral angles) for all C−C bonds [34]. Furthermore, ring-opening is already a useful conceptualisation in itself. The same is true for fragmentation/dissociation and labelling such reactivity as a dynamophore relegates the term from being a novel concept to being an umbrella term covering dynamics that is already under the broader umbrella term of photochemistry (also including electrocyclic reactions, bleaching and isomerisations). These types of dynamics could be appended at a later stage to the full list of dynamophores if more examples of novel non-conceptualised localised dynamics are identified but for the remainder of this thesis ring-opening, fragmentation and dissociation are not included as examples of dynamophores.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Zewail AH (2000) Angew Chem Int Ed 39:2586–2631 Dantus M, Rosker MJ, Zewail AH (1987) J Chem Phys 87:2395–2397 Bowman R, Dantus M, Zewail A (1989) Chem Phys Lett 161:297–302 Zewail AH (1991) Faraday Discuss Chem Soc 91:207–237 Neumark DM (1992) Annu Rev Phys Chem 43:153–176 Sandorfy C (1999) Understanding chemical reactivity. Springer Gordon MS (1977) Chem Phys Lett 52:161–167 Buenker RJ, Peyerimhoff SD (1975) Chem Phys 8:56–67 Stebbings R, Dunning F (1983) Rydberg states in atoms and molecules. Cambridge University Press Pratt ST (1995) Rep Prog Phys 58:821–883 Chupka WA (1987) J Chem Phys 87:1488–1498 Klein LB, Morsing TJ, Livingstone RA, Townsend D, Sølling TI (2016) Phys Chem Chem Phys 18:9715–9723 Klein LB, Thompson JOF, Crane SW, Saalbach L, Sølling TI, Paterson MJ, Townsend D (2016) Phys Chem Chem Phys 18:25070–25079 Müller-Dethlefs K, Schlag EW (1991) Annu Rev Phys Chem 42:109–136 Roberts GM, Stavros VG (2014) Chem Sci 5:1698–1722 Ashfold MNR, King GA, Murdock D, Nix MGD, Oliver TAA, Sage AG (2010) Phys Chem Chem Phys 12:1218–1238 Sølling TI, Kuhlman TS, Stephansen AB, Klein LB, Møller KB (2013) Chem Phys Chem 15:249–259 Kuhlman TS (2013) The non-ergodic nature of internal conversion, PhD thesis. Springer Braslavsky SE (2006) Pure Appl Chem 79:360–361 Schalk O, Boguslavskiy AE, Stolow A, Schuurman MS (2011) J Am Chem Soc 133:16451– 16458 Takezaki M, Hirota N, Terazima M (1997) J Phys Chem A 101:3443–3448

References

7

22. Takezaki M, Hirota N, Terazima M, Sato H, Nakajima T, Kato S (1997) J Phys Chem A 101:5190–5195 23. Takezaki M, Hirota N, Terazima M (1998) J Chem Phys 108:4685 24. Quenneville J, Greenfield M, Moore DS, McGrane SD, Scharff RJ (2011) J Phys Chem A 115:12286–12297 25. Mewes J-M, Jovanovic V, Marian CM, Dreuw A (2014) Phys Chem Chem Phys 16:12393– 12406 26. Morales-Cueto R, Esquivelzeta-Rabell M, Saucedo-Zugazagoitia J, Peon J (2007) J Phys Chem A 111:552–557 27. Reichardt C, Vogt RA, Crespo-Hernández CE (2009) J Chem Phys 131:224518 28. Plaza-Medina EF, Rodríguez-Córdoba W, Peon J (2011) J Phys Chem A 115(35):9782–9789 29. Orozco-Gonzalez Y, Coutinho K, Peon J, Canuto S (2012) J Chem Phys 137:054307 30. Vogt RA, Reichardt C, Crespo-Hernández CE (2013) J Phys Chem A 117:6580–6588 31. Vogt RA, Crespo-Hernández CE (2013) J Phys Chem A 117(51):14100–14108 32. Buma WJ, van der Waals JH, van Hemert MC (1990) J Chem Phys 93:3746–3751 33. Palmer IJ, Ragazos IN, Bernardi F, Olivucci M, Robb MA (1993) J Am Chem Soc 115:673–682 34. Deb S, Weber PM (2011) Annu Rev Phys Chem 62:19–39

Chapter 2

Experimental Methods

A range of different experimental methods have previously been applied in the field of ultrafast spectroscopy; these include but are not limited to transient absorption, laser-induced fluorescence, time-resolved mass spectrometry (TRMS), vibrational fs spectroscopy and resonance-enhanced multi-photon ionisation (REMPI). These methods share the first step of any ultrafast experiment, the excitation of a molecule by a pump pulse. The methods differ in how the resulting dynamics are monitored, i.e. the probe pulse. The experimental technique that has been employed throughout this thesis for interrogation of the ultrafast response to photoexcitation is time-resolved photoelectron spectroscopy (TRPES) in which the excited state is projected onto the ionisation continuum. The experiments are limited to the weak-field regime in which the molecular species can be assumed to absorb only a single photon at a time, as opposed to the strong-field regime in which very intense laser fields lead to direct ionisation [1, 2].

2.1 Time-Resolved Photoelectron Spectroscopy A thorough review of the technique can be found in [3], and many of its aptitudes have previously been highlighted [4]. One of the major advantages of TRPES is that the angular momentum and symmetry selection rules are relaxed compared to transient absorption techniques as the outgoing photoelectron can attain a range of symmetries. This means that ionisation of any electronic state is allowed and as such there are no ‘dark states’ in photoelectron spectroscopy. Still, different ionisation propensities for electronic states with different characters manifest themselves [3–5]. Ionisation to different ionic states from neutral states with different characters can be understood by considering Koopmans’ theorem [6]. The correlation to different ionic states (here either D0 or D1 ) for two different neutral states is illustrated in Fig. 2.1. The essential approximation in Koopmans’ theorem is that only the ionised electron is affected © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 M. A. Bjørnholst, Time-Resolved Photoionisation Studies of Polyatomic Molecules, Springer Theses, https://doi.org/10.1007/978-3-030-53629-9_2

9

10

2 Experimental Methods

Fig. 2.1 Illustration of Koopmans’ correlation and its effect on preferential ionisation to either D0 or D1 depending on the electronic configuration of the molecular core. Only the electron in the excited state (red) is affected, and the core is left unchanged

and the remaining electronic structure does not change. The lowest singly occupied orbital associated with the neutral excited state thus remains singly occupied and can be directly correlated with an ionic state. In Fig. 2.1, this is illustrated as either excitation from a lone-pair (n) or a π orbital (π). As D0 is associated with removal of a lone-pair electron, the excited state on the left will ionise preferentially to D0 , whereas the excited state on the right will ionise preferentially to D1 . This propensity can result in the observation of closely spaced neutral excited states as energetically separated bands in TRPES since the ionisation energies of D0 and D1 will not be identical. As such the propensities may become beneficial as they can result in less congested spectra which is often favourable when investigating dynamics involving multiple electronic states. Additionally, this may result in some neutral states not being observed at all despite attaining significant population if the probe energy is insufficient to ionise the associated ionic state. One example of this is the (π, π ∗ ) valence state excited by VUV pulses in acetone, which correlates to D1 (>2.5 eV higher in energy than D0 [7]) but had not been observed until sufficiently energetic probe pulses were used [8–11]. This is a general limitation in TRPES that dynamics can only be followed as long as the probe pulse is sufficiently energetic to ionise the populated state. In cases of ground state population, the probe energy must therefore be larger than the ionisation energy although this will also result in a time-independent single-photon ionisation background signal. The information content of an experiment is partially determined by the applied detection technique. The 2D detection method applied herein, velocity map imaging (VMI), offers highly differential information as illustrated in Fig. 2.2. Simply monitoring the photoelectron yield (Fig. 2.2a) offers information on the timescales involved in a time-resolved experiment. The energy-resolved yield (Fig. 2.2b) offers potential information on which states are associated with a given timescale and the possibility of deconvolving the timescales obtained from the integrated yield. Finally, the angle-resolved yields offer detailed information on the photoionisation process and thereby potentially direct information on the character of the excited state [12].

2.1 Time-Resolved Photoelectron Spectroscopy

11

Fig. 2.2 Time-, energy- and angle-resolved yields can be combined in TRPES, allowing for more differential information

The presence of a CI has, for instance, been argued from a rapid (