Advances in New Catalytic Materials [1 ed.] 9783038133988, 9780878492404

Citation preview

Advances in New Catalytic Materials

Edited by Jin-An Wang Guozhong Cao José Manuel Domínguez

Advances in New Catalytic Materials Selected, peer reviewed papers from the Second International Symposium on New Catalytic Materials, Cancún, Mexico, 16-20 August, 2009

Edited by

Dr. Jin-An Wang National Polytechnic Institute, Mexico

Dr. Guozhong Cao University of Washington, USA

Dr. José Manuel Domínguez Mexican Petroleum Institute, Mexico

TRANS TECH PUBLICATIONS LTD Switzerland • UK • USA

Copyright  2010 Trans Tech Publications Ltd, Switzerland All rights reserved. No part of the contents of this publication may be reproduced or transmitted in any form or by any means without the written permission of the publisher. Trans Tech Publications Ltd Laubisrutistr. 24 CH-8712 Stafa-Zurich Switzerland http://www.ttp.net Volume 132 of Advanced Materials Research ISSN 1022-6680 Full text available online at http://www.scientific.net

Distributed worldwide by

and in the Americas by

Trans Tech Publications Ltd Laubisrutistr. 24 CH-8712 Stafa-Zurich Switzerland

Trans Tech Publications Inc. PO Box 699, May Street Enfield, NH 03748 USA

Fax: +41 (44) 922 10 33 e-mail: [email protected]

Phone: +1 (603) 632-7377 Fax: +1 (603) 632-5611 e-mail: [email protected]

Preface New catalytic materials have grown to a very important multidisciplinary research field including novel strategies for catalytic materials synthesis, control and manipulation of solid state chemistry and surface, innovative characterization techniques, and rapidly expanding catalysis applications, and thus have attracted a great attention of the scientists and engineers all over the world. To promote the development of new catalytic materials, the First International Symposium on New Catalytic Materials (NCM-1) was successfully organized in Cancun, Mexico, in August 16-21, 2008. Some selected manuscripts were published in a special volume of Catalysis Today in 2009 (Catalysis Today Vol. 148). The Second International Symposium on New Catalytic Materials (NCM-2) was held in the same location in August 16-20, 2009, as a part of the XVIII International Congress on Materials Research organized jointly by the Mexican Materials Society and the Materials Research Society (MRS) of the United States. This symposium attracted approximately 100 presentations from various countries and featured 1 plenary lecture, 7 keynote presentations, 30 oral talks and 70 posters. A total number of 83 full-length manuscripts were submitted for possible publication. After full peer review, some of these papers were published in a special volume of Advanced Materials Research and in a special issue of Catalysis Today, based on its scientific quality and novelty. This special volume of Advanced Materials Research consists of new contributions covering the aspects of catalysts preparation and characterization as well as various catalysis applications. Among these contributions, syntheses of novel catalytic materials involving surface functionalized mesoporous materials, hybrid inorganic-organic solids, nanotubes and nanorods of oxides, carbon nanotubes and activated carbon, carbide and nitride, zeolites, ion exchange resins, rare earth oxides doped solid superacids and superbases and double-layer hydrotalcites etc. were reported and applied for catalyzing a variety of reactions like hydrogen production and storage, NO reduction and photocatalytic degradation of VOCs, electrocatalytic synthesis of fine chemicals, Fischer-Tropsch synthesis, protroleum hydrodesulfurization, isomerization and alkylation of alkanes, transesterification of soybean oil to biodiesel etc.

Although the current volume does not cover all topics in the area of catalytic materials, it certainly reflects the recent advances and future trends in this field, which will undoubtedly benefit to the scientists and investigators from academic institutions and universities as well as engineers from industries. The editors gratefully acknowledge all the reviewers who contributed to the accomplishment of this work. We also wish to thank all the authors for their great efforts in the manuscripts preparation and revision. Special gratitude goes to the publishing editor, Mr. Thomas Wohlbier, who managed the complexity of information and communication aspects. Finally, we would like to acknowledge the financial support from the National Polytechnic Institute (IPN), the Mexican Petroleum Institute (IMP), National Consul of Sciences and Technology of Mexico (CONACyT), the Mexican Materials Society (MMS), and the Mexican Catalysis Society.

Editors Dr. Jin-An Wang National Polytechnic Institute, Mexico Dr. Guozhong Cao University of Washington, USA

Dr. José Manuel Domínguez Mexican Petroleum Institute, Mexico

The Second International Symposium on New Catalytic Materials Cancún, Mexico, August 16-20, 2009

Chairmen Dr. Jin-An Wang National Polytechnic Institute, Mexico Dr. José Manuel Domínguez E. Mexican Petroleum Institute, Mexico

Scientific Committee Dr. Jorge Ramírez Solís, Universidad Nacional Autónoma de México Dr. Gustavo Fuentes Zurita, Universidad Autónoma Metropolitana-I Dr. Ricardo Gómez, Universidad Autónoma Metropolitana-I Dr. Tatiana Klimova, Universidad Nacional Autónoma de México Dr. José Antonio De Los Reyes, Universidad Autónoma Metropolitana-I Dr. Jaime Sánchez Valente, Instituto Mexicano del Petróleo Dr. Griselda Corro, Universidad Autónoma de Puebla Dr. Luis Enrique Noreña, Universidad Autónoma Metropolitana-A Dr. Miguel Angel Valenzuela, Instituto Politécnico Nacional Dr. José Aarón Melo Banda, Instituto Tecnológico de Cd. Madero Dr. Xim Bokhimi, Universidad Nacional Autónoma de México

Table of Contents Preface and Committees

Plenary Lecture Nanostructured Materials for Hydrogen Storage S. Sepehri, Y.Y. Liu and G.Z. Cao

1

Catalysts Synthesis and Characterization Influences of Surface Chemistry on Dehydrogenation Kinetics of Ammonia Borane in Porous Carbon Scaffold S. Sepehri, B. Batalla García, Q.F. Zhang and G.Z. Cao Observation on the Structure of Ordered Mesoporous Materials at High Temperature via In Situ X-Ray Diffraction C.F. Zhou and J.H. Zhu Synthesis of Si-Based Mesoporous Materials with Different Structural Regularity L.F. Chen, J. López, J.A. Wang, L.E. Noreña-Franco, G.X. Yu, F.H. Cao, Y.Q. Song and X.L. Zhou Comparative Studies of the CoMo/MgO, CoMo/Al2O3 and CoMo/MgO-MgAl2O4 Catalysts Prepared by a Urea-Matrix Combustion Method L.B. Wu, D.M. Jiao, L.F. Chen, J.A. Wang and F.H. Cao ZnAlFe Mixed Oxides Obtained from LDH Type Materials as Basic Catalyst for the Gas Phase Acetone Condensation A. Mantilla, F. Tzompantzi, M. Manríquez, G. Mendoza, J.L. Fernández and R. Gómez Structure Sensitivity of Sol-Gel Alkali Tantalates, ATaO3 (A= Li, Na and K): Acetone Gas Phase Condensation L.M. Torres-Martínez, M.E. Meza-de la Rosa, L.L. Garza-Tovar, I. Juárez-Ramírez, F. Tzompantzi, G. Del Angel, J.M. Padilla and R. Gómez Phase Stabilization of Mesoporous Mn-Promoted ZrO2: Influence of the Precursor M.L. Hernández-Pichardo, J.A. Montoya, P. Del Angel and S.P. Paredes

19 29 38 45 55

61 68

Environmental Catalysis Promotional Effect of Gadolinia on CuO Catalyst for Reduction of NO by Activated Carbon Y.Y. Xue, G.Z. Lu, Y. Guo, Y.L. Guo, Y.Q. Wang and Z.G. Zhang Synthesis and Photocatalytic Performance of Hierarchical Porous Titanium Phosphonate Hybrid Materials T.Y. Ma, T.Z. Ren and Z.Y. Yuan Role of Nanocrystalline Titania Phases in the Photocatalytic Oxidation of NO at Room Temperature S. Castillo, R. Carrera, R. Camposeco, P. Del Angel, J.A. Montoya, A.L. Vázquez, M. MoránPineda and R. Gómez Chitin/TiO2 Composite for Photocatalytic Degradation of Phenol K. Wan, X.H. Peng and P.J. Du

76 87

96 105

Catalysts for Petroleum Hydrotreating Catalytic Properties of Ni-Mo Carbide and Nitride Phases Supported on SBA-15 and -16 in the Hydrodesulfurization of DBT E.C. Aguillón-Martínez, J.A. Melo-Banda, L.A. Guevara, T.A. Reyes, C.E. Ramos Galván, R.R. Silva and J.M. Domínguez Oxidative Removal of Dibenzothiophene by H2O2 over Activated Carbon-Supported Phosphotungstic Acid Catalysts G.X. Yu, R.X. Zhou, J.B. Li, X.L. Zhou, C.L. Li, L.F. Chen and J.A. Wang Adsorptive Removal of Dibenzothiophene in Diesel Fuel on an Adsorbent from Rice Hull Activated by Phosphoric Acid G.X. Yu, J. Sun, X.M. Hou, X.L. Zhou, C.L. Li, L.F. Chen and J.A. Wang

111 126 133

b

Advances in New Catalytic Materials

Adsorption of Dibenzothiophene on Transition Metals Loaded Activated Carbon G.X. Yu, J.B. Li, X.L. Zhou, C.L. Li, L.F. Chen and J.A. Wang

141

Catalysts for Petroleum Refining Studies on the Catalytic Activity of Sulfated Zirconia Promoted with Cerium Oxide F.E. Lugo del Ángel, R. Silva-Rodrigo, A. Vázquez Rodríguez, R. García Alamilla, J. Navarrete Bolaños, A. Castillo Mares, J.A. Melo-Banda, E. Térres-Rojas and J.L. Rivera Armenta Isomerization of Pinene with Al- and Ga-Modified MCM-41 Mesoporous Materials M. Díaz-García, J. Aguilar-Pliego, G. Herrera-Pérez, L. Guzmán, P. Schachat, L.E. NoreñaFranco, A. Aguilar-Elguezabal and M. Gutierrez-Arzaluz Yb2O3 Promoted Pt-SO42-/ZrO2-Al2O3 Catalyst in N-Hexane Hydroisomerization G.X. Yu, Y. Hu, D.N. Lin, X.L. Zhou, C.L. Li, L.F. Chen and J.A. Wang Effect of Hydrothermal Conditions on Isomerization Activity of Pt/SO42--ZrO2 J.J. Zhang, Y.Q. Song, X.L. Zhou, C.L. Li, J.A. Wang and L.Y. Xu Alkylation of Benzene with Propylene over H3PW12O40 Supported on MCM-41 and -48 Type Mesoporous Materials M. Gómez-Ruiz, J.A. Melo-Banda, C.E. Ramos Galván, S.E. López, R.R. Silva, R.I. Alamilla and J.M. Domínguez

149 162 174 183

192

Catalysts for Fuel Production Hydrogen Production by Steam Reforming of Methanol over a Ag/ZnO One Dimensional Catalyst R. Pérez-Hernández, A. Gutiérrez-Martínez, A. Mayoral, F. Leonard Deepak, M.E. FernándezGarcía, G. Mondragón-Galicia, M. Miki and M. José-Yacamán Basic Ion Exchange Resins as Heterogeneous Catalysts for Biodiesel Synthesis M.G. Falco, C.D. Córdoba, M.R. Capeletti and U. Sedran A Novel La2O3-ZnO/ZrO2 Solid Superbase and its Catalytic Performance for Transesterification of Soybean Oil to Biodiesel X. Li, G.Z. Lu, Y.L. Guo, Y. Guo and Y.Q. Wang Refinery Oil Fraction Fuels Obtained from Polyethylene Catalytic Cracking Employing Heteropolyacid-MCM-41 Materials A. Hernández, L.E. Noreña-Franco, L.F. Chen, J.A. Wang and J. Aguilar Synthesis Optimization of SAPO-34 in the Presence of Mixed Template for MTO Process L.P. Ye, F.H. Cao, W.Y. Ying, D.Y. Fang and Q.W. Sun Studies on Cobalt-Based Catalyst to Synthesize Heavy Hydrocarbons C. Li, P.L. Wang, W.Y. Ying and D.Y. Fang

205 220 228 236 246 257

Electrocatalysis A Novel Non-Metal Oxygen Reduction Electrocatalyst Based on Platelet Carbon Nanofiber J.S. Zheng, X.S. Zhang, S. Wen, P. Li, C.A. Ma and W.K. Yuan Enhanced Electrochemical Oxidation of BH4− on Pt Electrode in Alkaline Electrolyte with the Addition of Thiourea D.M. Yu, C.G. Chen, S. Lei, X.Y. Zhou and G.Z. Cao Pd Nanoparticles Supported on TiO2 Nanotubes for Ethanol Oxidation in Alkaline Media H.H. Yang, Y.H. Qin, S. Wen, X.S. Zhang, D.F. Niu, C.A. Ma and W.K. Yuan

264 271 279

Advanced Materials Research Vol. 132 (2010) pp 1-18 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.1

Nanostructured Materials for Hydrogen Storage Saghar Sepehri, Yanyi Liu and Guozhong Cao* Department of Materials Science and Engineering, University of Washington, Seattle, WA 98195, USA Email: [email protected], [email protected], *[email protected]

Key words: Hydrogen storage; Carbon cryogel; Hydride; Nanocomposite; Mesoporosity

Abstract: Hydrogen generated from clean and renewable energy sources has been considered as an alternate fuel to carbon based fossil fuels for several decades. Although many advances in hydrogen production and usage have been made, storing hydrogen remains a significant challenge. Many drawbacks including energy intensive processes, low volumetric densities, and safety concerns are associated with storing hydrogen as pressured or liquefied. Solid state hydrogen storage is considered to be the most promising method as a safe and effective storage option, but there is still no material or method that satisfies the requirements for a practical approach. A feasible hydrogen storage media should address several issues including targeted storage capacities, thermodynamics and hydrogen sorption kinetics, and safety. Nanostructured materials can provide tailor-made properties for storing and releasing hydrogen to fulfill, at least, the partial requirements. This short review, not a comprehensive review of all the materials or technologies in hydrogen storage, summarizes some of the recent developments in application of nanostructures for solid state hydrogen storage; particular attention has been devoted to the most recent development of nanocomposites with tuned dehydrogenation temperatures and kinetics through the control of pore size and surface chemistry. 1. Introduction During the recent years, significant progress has been made in the development of alternative energy technologies. Hydrogen has the potential to be a good energy carrier candidate in a carbonfree emission cycle. There are three components to the hydrogen economy: production, storage, and usage. Storing hydrogen is a challenging step in the hydrogen technology and considerable efforts have been made in synthesizing and investigating novel materials for hydrogen storage in the past decade [1]. Ideally, hydrogen should be stored in such a way to attain high storage capacity under near the ambient conditions to be safe and economical, and perform rapid and reversible hydrogenation and dehydrogenation process for practical applications. Although various techniques and materials have been used or studied to store hydrogen, there is neither method nor material that satisfies all the requirements for perceived hydrogen economy [2]. Hydrogen can be stored as gas, liquid, or solid. Storing hydrogen as compressed gas needs high pressure and heavy containers to support such pressure. Liquefaction of hydrogen needs energy and consumes more than 20% of the recoverable energy. Also cryogenic containers should be used to decrease the hydrogen boil-off. Storing hydrogen as solid may offer the best option to store hydrogen through two basic mechanisms: physisorption (or physical adsorption) and chemisorptions (or chemical adsorption). In physisorption, molecular hydrogen is adsorbed by intermolecular (van der Waals) forces. Examples of physisorption include storing of hydrogen in carbon structures and organic or inorganic frameworks. In chemisorption, hydrogen molecules and chemical bonding of the hydrogen atoms dissociate by integration in the lattice of a metal, an alloy, or by formation of a new chemical compound. Metal, chemical and complex hydrides are examples of chemisorption. Each principle has its own prospects and limitations. A given material can exhibit both chemisorption and

2

Advances in New Catalytic Materials

physisorption. Chemisorption may provide high volumetric and gravimetric storage capacities, but the chemical bonds need to split or recombine to release hydrogen. Storing hydrogen by physisorption is not subject to this constraint, because the hydrogen stays in its molecular form, but the challenge is to provide materials with a sufficient amount of bonding sites for the hydrogen per volume to achieve high storage capacity. One of the major differences between physisorption and chemisorption is their binding energies (Fig. 1). Physisorption bonding is usually too weak ( 50 kJ /mol) and demands high desorption temperatures [3].

Figure 1 Bond strengths for physisorption and chemisorption and the desirable range of binding energies that allow hydrogen release around room temperature. To achieve an ideal binding energy (in the range of 10 – 60 kJ /mol), we need to increase the physisorption binding energy or reduce the chemisorption binding energy. Nanostructured materials with their unique physical, chemical, thermodynamic and kinetics properties can provide effective and sufficient ways to address the challenges involve in hydrogen storage [4, 5]. Nanostructures can offer new opportunities for addressing these challenges. They have the potential for high surface areas and hybrid structures that allow multifunctional performance. This review gives a brief summary on the recent achievements in developing nanostructured materials and methods for hydrogen storage, particularly the most recent development in coherent porous carbon and hydride nanocomposites with reduced dehydrogenation temperature and improved kinetics with a reduced pore size and modified surface chemistry. Many topics and interesting research are out of the scope of this short review; for more comprehensive coverage of hydrogen storage materials, the readers should consult the excellent review articles published recently, for example, on complex hydrides by Schuth, et al.,[6] on metal-organic frameworks (MOFs) by Long et al.,[7] and on ammonia borane (AB) by Baker et al.. [8] 2. Hydrogen storage by physisorption in nanoporous materials Physisorption is a principle where the forces involved are weak intermolecular forces, therefore; in general it is associated with fast kinetics, excellent reversibility, and theoretically infinite lifetime. But the challenge with the physisorption of hydrogen also results from these weak forces. H2 is the smallest molecule and only has two electrons, hence it is hard to polarize and in the absence of relatively strong polarizing centers, interaction between the adsorbent and the non-polar hydrogen molecules relies on the weak dispersion forces which created by temporarily induced dipoles, and are typically of the order of 3–6 kJ/ mol [9]. Thus, significant hydrogen adsorption often takes place only at a cryogenic temperature. Nanostructured materials may offer advantages for molecular hydrogen storage by providing high surface areas, or by encapsulation or trapping hydrogen in microporous media. Using porous nanostructured materials, in general, can reduce the gravimetric and the volumetric storage densities. However, the increased surface area and porosity in

Advanced Materials Research Vol. 132

3

nanostructures will offer additional binding sites on the surface and in the pores that could increase storage mainly through physisorption. The possibility of storing a significant amount of hydrogen on high surface area density materials has been a key driver in the investigation of hydrogen sorption properties of nanotubes, graphite sheets, metal organic frameworks, template ordered porous carbons. Nanostructured carbons, zeolites, metal-organic frameworks, clathrates, and polymers with intrinsic microporosity are examples of the investigated physisorption materials. 2.1 Nanostructured carbon Carbon materials with high surface area, good chemical stability, and low density have received considerable attraction. Nanostructured carbon materials, such as graphitic nanofibers (GNF), multiwalled carbon nanotubes (MWNT), single walled carbon nanotubes (SWNT), carbon nanorods, and carbon aerogels, demonstrate novel but distinct properties that relates to many possible configurations of the electronic states of carbon atoms. Each carbon atom has six electrons which occupy 1s2, 2s2, and 2p2 atomic orbitals. The various bonding states are connected with certain structural arrangements, so that sp bonding gives rise to chain structure, sp2 bonding to planar structures, and sp3 bonding to tetrahedral structures [10]. The structural and practical properties of carbon critically depend on the ratio between the number of sp2 (graphite-like) and sp3 (diamondlike) bonds.

Figure 2 Adsorption isotherms at 77K for the carbon aerogels shows the linear dependency of hydrogen adsorption to the surface area Early reports [11, 12] on hydrogen storage in carbon nanotubes and graphitic nanofibers, proposed high storage capacities (to 67 wt.%) and started an extensive worldwide surge of research. Since then many succeeding experiments were carried out with different methods, but such high values have not yet been reproduced by other groups [13]. Furthermore, no hypothesis could support the unusually high storage capacities and the high storage capacity results were more related to the faults of experiment [14, 15]. Nevertheless, hydrogen adsorption on carbon materials is still an attractive and improving filed. The result of several investigations proposes that the amount of adsorbed hydrogen is proportional to the specific surface area of the carbon material [16, 17]. In case of activated carbons and activated carbon fibers, the hydrogen absorption of 5 wt.% is obtained at low temperature (77K) and high pressure (30 to 60 bar) [18]. For GNF,

4

Advances in New Catalytic Materials

SWNT, and MWNT, the reversible hydrogen uptake of 1.5 wt.% per 1000m2/g under ambient conditions is reported [19]. Hydrogen capacity of 7 wt.% is observed for ordered porous carbon with surface area of 3200m2/g, prepared by template, at 77K and 20 bar [20]. Recent studies on carbon aerogels (CAs), another class of amorphous porous carbon structures with high surface area, shows 5 wt.% of hydrogen adsorption for surface area of 3200m2/g at 77K and pressure 20-30 bar (Fig. 2) [21]. Recent research in hydrogen physisorption on carbon nanostructure involves efforts on increasing surface area of carbon to provide more binding sites and incorporating functional groups (dopants) in carbon to increase the binding energy between hydrogen and carbon surface [22]. 2.2 Zeolites Zeolite is an inorganic porous material consisted of hydrated aluminosilicate mineral with highly regular structure that exhibit reversible occlusion of gases. However, because of the high density of the aluminosilicate framework, which contains Si, Al, O, and heavy cations, a high gravimetric hydrogen storage density might not be achieved. On the other hand, zeolites can be ideal choice in studying the hydrogen binding because of their well known crystal structure and easy ion exchange and those studies may provide insight valuable for work on other hydrogen adsorbents. The working principle of hydrogen storage in zeolites is that the guest molecules under high temperature and pressure are forced into the cavities of the molecular sieve host. Upon cooling to room temperature or below, hydrogen is trapped inside the cavities and it can be released again by raising the temperature. The amount of encapsulated hydrogen in zeolite is related to the size of the exchanged cation and higher storage capacity is observed for zeolites with high number of small cavities in their structure [23]. It has been showed that the storage capacity of zeolite may be increased at low temperatures and a hydrogen storage capacity of 1.81 wt.% (at 15 bar and 77K) was obtained for NaY zeolite [24]. The calculated maximum possible hydrogen storage capacity of zeolites is less than 3 wt.% [25]. Large mass of the zeolite framework is a limiting factor in the storage capacity and improvement in the storage capacity of zeolites will therefore include using light elements in the framework and also enhancing the interaction energy between hydrogen and the framework. 2.3 Metal – organic frameworks Metal–organic frameworks (MOFs) are crystalline solid compounds consisting of organic ligands connecting metal ions or clusters that form a cage structure. Most MOFs have a three-dimensional interconnected porous framework with uniform pores that provides an ordered network of channels. MOFs can be synthesized using a self–organizational process that allows different combinations of organic linkers and provides a wide range of different functionality and pore size [26, 27]. MOFs can provide light porous framework with high surface areas and pore volumes. Surface area of MOFs is usually in the range of 500–3000 m2/g while values higher than 5000 m2/g are also attainable [28]. A large pore volume of 1.1 cm3/g is observed for some MOFs [29]. Similar to carbon nanostructures, hydrogen storage capacity of MOFs increases with the surface area and microporous volume [5]. At low temperature (77K) and high pressure (70-90 bar) MOFs with hydrogen adsorption of 7 wt.% are reported [30, 31] but at 298K and 90 bar the maximum observed hydrogen adsorption is only 1.4 wt.% [31]. Increasing the interaction of hydrogen with the organic ligands and metal centers in MOFs can improve the hydrogen adsorption at ambient temperature. Several approaches including obtaining MOFs with more polarized cations or optimized pore size are being investigated [32, 33, 34].

Advanced Materials Research Vol. 132

5

2.4 Clathrates Clathrates are crystalline structures consisting of a hydrogen-bonded water framework as the ‘host’ lattice providing cavities which hold ‘guest’ molecules. Several natural gases including methane and carbon dioxide are well-known to form water clathrates or clathrate hydrates [35] and since the first reports on hydrogen clathrate hydrates [36], possible hydrogen storage in different hydrogen clathrates has promoted an attractive research [37]. Hydrogen-bonded water molecules can produce polyhedral small and large cages around guest molecules to form solid clathrate hydrates. When these cages are empty, they are not stable and may collapse into ice crystal structure but inclusion of gas molecules can stabilize the cages. Formation and stability of H2/H2O clathrates need high pressure and low temperatures, therefore, research on facilitating the formation and stabilizing the structures is necessary. A binary hydrogen-water clathrate is reported to contain 5.3 wt.% hydrogen at 250K and very high pressure (2 kbar) [38]. To ease the fabrication process and reduce the synthesis pressure a second guest component (such as tetrahydrofuran) can be used to fill the cages of clathrate but this approach decreases the hydrogen storage capacity of the clathrate to less than 4 wt.% [39]. Nonetheless, clathrates have opened up an interesting field for further research. 2.5 Polymers with Intrinsic Microporosity Polymers with intrinsic microporosity (PIMs) are obtained by polymerization of large rigid molecules to form chains and networks that contain interconnected pores and large surface areas (500-1000 m2/g) [40]. PIMs can be considered as potential hydrogen storage materials because of the low density and large surface area [41]. For network- PIM and hyper-cross linked polymer a maximum hydrogen adsorption (at 77 K and 10-15 bar) of 3 wt.% is observed [42]. In order to improve the hydrogen adsorption on PIMS they must be optimized further to improve their porosity. However, hydrogen adsorption at near ambient temperature can be much lower than at 77 K due to the weak interaction between the hydrogen and polymer. For the hydrogen storage by physisorption, porous materials should possess a large surface area, the larger the better; however, the porosity should be kept low. Smaller pores may also be favor of surface physisorption of hydrogen, as the surface energy in smaller pores would be higher. Pore volume should be kept low as the hydrogen storage is due to the surface adsorption. Large pore volume or porosity would only reduce the volumetric storage density. 3. Hydrogen storage by chemisorption in hydrides Chemisorption is the adsorption of a particle with the formation of a chemical bond. Hydrogen can be stored in hydrides (hydrogen rich materials) by chemisorptions to offer high storage capacity at ambient conditions. Volumetric and gravimetric hydrogen densities of some selected hydrides are compared with other hydrogen storage methods in Fig. 3 [43]. However, various hydrides suffer from a range of drawbacks such as poor reversibility, poor thermal conductivity, and relatively high dehydrogenation temperature [44]. Developing of new hydrides has been a very active research topic. Nanostructures can be used to improve the hydrogen storage properties of hydrides. They can change the thermodynamic properties of hydride which define the theoretical working parameters, i.e. the pressure and temperature that hydrogen can be absorbed and desorbed. They can also change the kinetic properties that determine the rate of hydrogen release. This section discusses the recent finding and developments in this field.

6

Advances in New Catalytic Materials

Figure 3. The storage density of hydrogen in compressed gas, liquid, adsorbed monolayer (physisorbed), and selected chemical compounds, as a function of the hydrogen mass fraction. The straight lines indicate the total density of the storage medium including, hydrogen and host atoms. Pressurized gas storage is shown for steel (tensile strength 460MPa, density 6500kg/m3) and a hypothetical composite material (tensile strength 1500MPa, density 3000kg/m3) [43]. 3.1 Metal and complex hydrides Metal hydrides are solid alloys which are typically composed of metal atoms with a host lattice and hydrogen atoms that are trapped in the interstitial sites forming a single-phase compound between a metal host and hydrogen. Binary hydrides can essentially be classified into three categories depending on the nature of the bonding between hydrogen and the metal host. Ionic or saline hydrides (e.g. MgH2, NaH, and CaH2) are formed by alkali and alkaline earth atoms and exhibit ionic bonding between the hydrogen and metal atoms. Covalent hydrides are formed by nonmetal elements like S, Si, C or B. Metallic hydrides (e.g. LaNi5H6, PdH0.6, FeTiH2) originate from the metallic bonding between hydrogen and either a transition metal or a rare earth metal. In addition, group IA, IIA, and IIIA light metals form metal–hydrogen complexes [AlH4-], BH4-, which form covalent or ionic bonds with a cation, giving rise to highly stable complex hydrides (e.g. NaAlH4 , Mg(AlH4)2). Hydrides have higher hydrogen storage density than hydrogen in gas or liquid form. For example the hydrogen density of MgH2 is 6.5 H atoms/cm3, while those of gas and liquid hydrogen are 0.99 H atoms/cm3 and 4.2 H atoms/cm3, respectively [45]. Some metal hydrides absorb and desorb hydrogen at near ambient temperature and pressure, and demonstrate very high hydrogen density. However, all the reversible hydrides working around ambient temperature and atmospheric pressure consist of heavy transition metals; therefore, even in nanocrystalline form, the gravimetric hydrogen density of metal hydrides is limited to less than 3wt.% [46]. One way to improve the hydrogen capacity of metal hydrides is to use light weight materials such as magnesium [44]. Another challenge in hydrogen desorption from metal hydrides is their stability that demands elevated temperatures for release of hydrogen. Heat transfer is yet another challenge. In general, the

Advanced Materials Research Vol. 132

7

formation of metal hydrides is an exothermic reaction. Efficient heat removal (for absorption) and heat addition (for desorption) has proven extremely difficult to achieve in metal hydride based systems [47]. Synthesis of new hydrides, particularly complex hydrides, has potential to develop materials with superior hydrogen storage properties. Complex metal hydrides demonstrate higher gravimetric hydrogen capacities than simple metal hydrides. However, some of them show poor reversibility and once hydrogen is released they need high pressure to adsorb hydrogen again [48]. Moreover, due to the localization of the hydrogen and the slow diffusion rate of the metals in the solid, hydrogen sorption reactions are slow. Also, similar to metal hydrides, most complex hydrides suffer from high thermal stability [49].

Figure 4 Enthalpy diagram for destabilization of LiBH4 by MgX (X = H2, F2, S, Se) Destabilization approaches can be used to improve the hydrogen storage properties of stable hydrides materials [50]. Thermodynamic destabilization of light-metal hydrides is achieved by using additives that reacts with metals to form new compounds (an intermediate state) during dehydrogenation and lowers the enthalpy and hydrogen release temperature. Fig. 4 shows the decreased dehydrogenation enthalpies for LiBH4 by adding destabilization agents [51]. However, the dehydrogenation temperature for destabilized LiBH4 is still higher than 350˚C and the kinetics are slow. Also, it should be mentioned that adding any additive may increase the mass and reduce the hydrogen capacity, for example, the hydrogen capacity decreases from 13.6 wt.% in pure LiBH4 to 11.4 wt.% after adding MgH2, however, the enthalpy is lowered by 25 kJ/mol H2 [51]. Improving the reaction kinetics by decreasing the size of hydride may offer an interesting approach without increasing the mass. Increased surface area in nanosized hydrides can augment their surface energy and reduce the dehydrogenation enthalpy drastically. The increased surface area of hydrides facilitates the dissociation of hydrogen atoms by offering a larger number of dissociation sites and allowing fast gaseous diffusion [47]. Different methods (including laser ablation, vapor condensation, sputtering, and ball milling) can be used to reduce the size of metal hydride particles. Reduced particle and crystallite size is shown to enhance the hydrogen sorption kinetics in aluminum and magnesium based hydrides [52, 53]. Fig.5 demonstrates the effect of size reduction on dehydrogenation kinetics of nanosized sodium alanate (NaAlH4), deposited as

8

Advances in New Catalytic Materials

clustered on carbon nanofibers. The enhanced hydrogen desorption in nano- NaAlH4 can be attributed to the minimization of the solid-state diffusion path length during loading and desorption of hydrogen [54].

Figure 5. The improved hydrogen desorption for sodium alanate (NaAlH4) supported on carbon nanofibers. In conclusion, although the metal and complex hydrides are considered to be potential candidates for hydrogen storage, significant fundamental research should be performed to obtain hydrides with practical hydrogen storage properties. Dehydrogenation temperatures should be decreased and kinetics of reaction should be improved. Other issues, including thermal management, reversibility and durability, also need to be addressed [55]. 3.2 Chemical hydrides Chemical hydrides store hydrogen as M- H bonds where M is a light main group element such as C, B, N, or O [56]. They can release hydrogen through a chemical reaction which is typically not easily reversible. Sometimes, metal and complex hydrides are also categorized under chemical hydrides, but those classes often refer to reversible dehydrogenation. Common reactions to release hydrogen from chemical hydrides involve the reaction with water (hydrolysis) or alcohols (alcoholysis), and thermal decomposition (pyrolysis). In all these methods, several issues such as controllability of reaction and regeneration energy should be considered. A number of chemical hydrides, with both exothermic and endothermic dehydrogenation through different reaction are currently under investigation. Moreover, new chemical hydrides with high hydrogen densities can offer promising approaches for hydrogen storage. One of the early chemical hydrides studied, Ammonia (NH3) has been used in the fuel cells and power plants for more than 40 years [57, 58]. Anhydrous ammonia has high gravimetric hydrogen density of 17.5 wt.% and the byproduct of the hydrogen dissociation process is nitrogen that have no adverse environmental effects. However, decomposition (cracking) of ammonia is an endothermic reaction that happens efficiently at temperatures higher than 500°C, with an enthalpy of +46kJ/mol. Therefore, it takes energy to gain hydrogen from ammonia. There are also safety and toxicity issues such as propensity for reacting with water, reactivity with container materials, and high toxicity of the vapor if released into the air. These drawbacks should be addressed before using ammonia as a hydrogen storage material, however, because of high hydrogen density and wellestablished technology, ammonia is being considered as a means for delivering hydrogen. Several boron hydrides have high hydrogen content. Ammonia-borane (AB), also known as borazane or by formula NH3BH3, has been of great interest as a hydrogen storage material. At

Advanced Materials Research Vol. 132

9

ambient temperature and pressure, AB is a stable, white, crystalline solid (orthorhombic at lower temperatures and tetragonal more than -50°C), with low molecular weight (30.7 g /mol) with high gravimetric hydrogen capacity (19.6 wt.%) [59]. There have been several experimental studies on the multi-step thermal decomposition of AB [60, 61, 62]. It was found that AB releases one mole of hydrogen (per mole of AB) around 110°C and another mole of hydrogen at 150°C. Different methods, including milling and catalysts have been used to improve the kinetics of AB dehydrogenation reactions and lower its decomposition temperature [63]. These findings are encouraging, however, future research should address several issues including reducing the dehydrogenation temperature, minimizing the formation of volatile products and developing economically viable methods for regeneration of AB [64]. 4. Nanocomposites for hydrogen storage Nanocomposites refer to materials consisting of at least two phases with one with characteristic dimension in nanometer scale and uniformly dispersed in another that is called matrix and forms a three-dimensional network [65]. At the nano-scale, materials can show distinctly different properties than those of their bulk analogs. New fabrication techniques have offered new opportunities to design materials with specific structure to achieve desired properties. In hydrogen storage studies, nanocomposites have been observed to significantly improve the thermodynamics and kinetics of hydrogen sorption by providing high surface area and hybrid structures that offers multifunctional performance. Decreasing particle size increases surface/volume ratio, resulting in enhanced surface energies, and altering the hydrogen release mechanisms. 4.1 Mesoporous silica (SBA-15)–ammonia-borane (AB) nanocomposites Using nanoporous scaffold as structure-directing agents to host hydrides can facilitate the formation of nano-size hydrides within the scaffold while reduces the hydrogen diffusion distances. It has been shown that infusing ammonia-borane (AB) in nanoporous silica scaffold lowers the activation barrier for the hydrogen release, significantly improves the dehydrogenation kinetics, lowers the dehydrogenation temperature, and suppresses unwanted volatile products (Fig.6) [66].

10

Advances in New Catalytic Materials

Figure 6 Mass spectrometry (at 1°C/min) of volatile products generated by heating neat AB (solid line) and AB:SBA-15 (dashed line); m/e=2 (H2) and m/e=80 (borazine, c-(NHBH)3) 4.2 Porous carbon scaffold nanocomposites Porous carbon – hydride nanocomposites have shown impressive results in enhancing the dehydrogenation kinetics by reducing the diffusion distance and increase the reaction surface area. Incorporation of hydrides into functional frameworks or matrices with desired chemical and physical properties must also be investigated more intensively. Increasing porosity of the nanoscaffold to attain large accessible pore volume will improve the storage capacity. The tunable structure of nanoporous carbon offers different ways to catalyze the dehydrogenation reaction. Thermal properties of the nanoscaffold play an important role in the reaction kinetics. Fabrication of a light and thermally conductive nanoporous material (such as porous carbon) provides a compelling approach. Modifying structure of carbon scaffold by adding other elements can increase the heat transfer while catalyzing the dehydrogenation furthermore. Using nanocomposites for hydrogen storage opens up the possibility of designing functional systems, where an external matrix could act as multi-functional destabilization system for hydride. It can decrease the hydride size, and increase surface energy, increase the heat transfer, and chemically catalyzing the dehydrogenation to achieve desirable thermodynamic and kinetic properties of the hydride. Using scaffolds with high pore volume and low weight can minimize the gravimetric and volumetric penalties associated with this method. Carbon aerogels (CAs) and carbon cryogels (CCs), the nanoporous carbons with high pore volume and surface area, and tunable densities and pore sizes, can serve this purpose. CAs and CCs can be prepared from hydrogels generated by sol-gel polycondensation of organic monomers such as resorcinol and formaldehyde in aqueous solution in the presence of a polymerization catalyst [67]. The precursor hydrogels can be dried by different methods including supercritical drying and freeze drying. Supercritically dried hydrogels are called aerogels, while freeze-dried gels are known as cryogels. CAs and CCs are produced by pyrolysis of aerogels and cryogels and are consisted of interconnected porous carbon skeleton with high porosity (above 90%) and surface area (above 1000 m2/g), and pore diameters ranging from less than 1nm to 100nm [68, 69]. Because of the low density and high porosity, CAs and CCs can accommodate a large fraction of hydrides with little addition of weight. Moreover, the extremely high surface area facilitates an intimate contact between hydrides and the carbon network.

Advanced Materials Research Vol. 132

11

4.2.1 Carbon aerogel (CA) - LiBH4 nanocomposites Incorporation of LiBH4 into CAs has shown to enhance the dehydrogenation kinetics and lower the dehydrogenating temperature of LiBH4. Fig.7 shows the thermo-gravimetric analysis for hydrogen release from LiBH4 confined in two aerogels with pore sizes of 13 and 26 nm, activated carbon with pore sizes of 99.5%, Sigma-Aldrich) and were used as received. Resorcinol (R) was mixed with formaldehyde (F) solubilized in distilled water (W), using sodium carbonate as a catalyst (C). The R/W, R/C, and R/F ratios were kept at 0.035 g/ml, 200, and 0.5, respectively. The clear solutions were poured into glass vials (inner diameter=10 mm) that were then sealed and cured at 90ºC for 7 days to complete the gelation process. The resulting RF hydrogels were acid washed in ten times their volume of trifluroacetic acid solution (pH: 1.9) at 45ºC for three days. Gels were put in in fresh t-butanol ten times their volume for solvent exchange (at room temperature for 24 hours), repeating twice more with fresh solution. For the modified samples the hydrogels were prepared as above then at the first solvent exchange step AB, decaborane, and aniline were individually added to hydrogeles (by dissorlving in t-butanol to make a 2 wt% solution) to obtain BNCC, BCC, and NCC, respectivley. The modified gels, initially dark red in color, changed to light red during this step. Samples were freeze dried for a week under vacuum (at -50ºC), and pyrolyzed for 4 hours at 1050ºC (heating rate 5ºC/min and, nitrogen flow 25 mL/min). Weight loss of the samples during pyrolysis was about 52% for CC, 46% for BNCC, 44% for BCC, and 48% for NCC. The bulk density of the samples were measured at 0.07 g/cc for CC, 0.08 g/cc for NCC, and 0.09g/cc for BNCC and BCC samples. CC-AB, BNCC-AB and BCC-AB nanocomposites were obtained by soaking monolithic CC, BNCC, NCC, and BCC samples in a solution of 10 wt% AB in tetrahydrofuran (THF) (99.9%, Sigma-Aldrich), under argon at room temperature, for 3-5 hours. The solvent was removed under vacuum. Sample weight gain (weight of AB loaded CC- weight of CC/ weight of CC) for all samples was about 45-50%. The CC-ABs were stored in liquid nitrogen before further analyses. The pore structure of carbon cryogels was analyzed by means of nitrogen sorption (BET technique) at -196 ºC using a Quantachrome NOVA 4200e instrument. Specific surface area, micropore and mesopore volumes were determined using multi point BET, t-method and BJH (desorption) analyses, respectively. Sample morphology was studied by scanning electron

Advanced Materials Research Vol. 132

21

microscopy (SEM) were done by JEOL JSM- 7000F. Elemental analysis was done by X-ray photoelectron spectroscopy (XPS) on a thin slice of the samples, using a Surface Science Instruments S-probe spectrometer (sampling depth about 50 Å, x-ray spot size 800µm). Dehydrogenation studies were performed using a Netzsch 200 differential scanning calorimeter (DSC). Samples were heated to 200 ºC at multiple heating rates under argon atmosphere (flow rate 15 ml/min). Mass spectrometry measurements were performed are analyzed by a JEOL HX-110 mass spectrometer at discrete temperatures under vacuum. Results and Discussion Table 1 exhibits the elemental and surface chemistry changes by introducing B and N in the CC matrix obtained by XPS measurements on the cross section of monolithic CC samples. Uniform distribution of boron and oxygen in BNCC and BCC samples was confirmed by observation of similar boron and oxygen concentrations at the center and the edge of the samples. The higher oxygen content in BNCC (~ 8.8 at %) and BCC (~8.7 at %) than that of CC (~2.9 at%, which is common for CC sample [23]), indicates that the boron in BNCC and BCC samples may be presented in the form of boron oxide or boron hydroxide. High resolution analysis revealed that majority (~70%) of detected surface oxygen are in form of hydroxyl group and that the B1s binding energy in BNCC and BCC is at 193 eV which is similar to the value commonly observed fro H3BO3 or B2O3 [24]. Also, in BNCC and NCC samples only a very small N1s peak (about 0.1 at %) was observed which may indicate that surface N (at least partially) volatizes and leaves the sample during pyrolysis. Table 1 Elemental distribution in the cross section of CC, BNCC, and BCC* Sample

C

O

B

CC (at %)

96.8

2.9

0

BNCC (at %)

88.7

8.8

2.5

BCC (at %)

88.9

8.7

2.4

NCC (at %)

97.3

2.7

0

* The small nitrogen content (~ 0.1 at %) in BNCC and NCC is not included in the table.

Figure 1. Pore size distribution of unmodified (CC) and modified (BNCC, BCC, and NCC) samples obtained by BJH method on desorption curve of nitrogen sorption isotherms.

22

Advances in New Catalytic Materials

The pore size distribution of the samples (obtained by applying BJH method on the desorption curve of the nitrogen sorption isotherms) is shown in Fig. 1. It can be seen that CC, NCC, and BNCC exhibit broad distribution of pores of 5-20 nm, while BCC shows a sharp pore size distribution with a peak of pores of ~ 9nm. It is known that pyrolysis at elevated temperatures can lead to a decrease in mesopore volume and pore size due to shrinkage resulted from partial sintering, which was commonly detected during the formation of CC [15]. The broad pore size distributions of CC, BNCC, and NCC can be attributed to the collapse of the mesopores during pyrolysis. Conversely, no such collapse is revealed in BCC which suggests that the presence of boron or boron oxide prevents the shrinkage or partial collapse of porous structure. On the other hand, the broader pore size distribution in the BNCC and NCC sample (as compared to that of CC) is more likely due to the escape of N during pyrolysis that can disturb the porous structure. Table 2 summarizes the detailed porous structure parameters of the samples. Boron- modified samples have both less macropore surface area that that of CC. BCC exhibit the largest mesopores volume and BET surface area among the samples. In BNCC 88% (i.e. 439/501), and in NCC and CC only 81% (i.e. 435/534) and 77% (i.e. 489/634) of the BET SA, respectively, is related to the mesopores, while 96% (i.e. 720/748) of BET SA in BCC is composed of mesopores. These results reveal more uniform and improved mesoporous structure of BCC as compared to CC, BNCC, and NCC. Moreover, presence of B in BNCC has improved its mesoporous structure as compared to CC and NCC. Table2 Porosimetry data for RF hydrogels and carbon cryogels BET SA (m2/g)

Mesopore SA* (m2/g)

CC

634

489

BNCC

501

BCC NCC

Sample

Macropore SA** (m2/g)

Mesopore Volume* (cc/g)

Pore Width* (nm)

145

1.01

7.0

466

35

1.06

10.0

748

720

28

1.48

8.5

534

435

99

1.03

8.3

*BJH desorption, **t- method, ** Macro SA= BET SA – (Meso SA + Micro SA (not detected))

(a)

(b)

(c)

(d)

Figure 2. SEM images of (a) CC, (b) BNCC, (c) NCC, and (d) BCC (scale bar = 100nm).

Advanced Materials Research Vol. 132

23

Scanning electron microscopy (SEM) images reveal that BCC (Fig.2 d) contains a more uniform mesoporous structure, while CC, BNCC, and NCC (Fig. 2a, b, and c) exhibits wide range of pores with clear macroporosity. The disturbed surface morphology that observed in NCC (Fig.2c) can be a result of releasing volatized N during pyrolysis. Calorimetric measurements of the dehydrogenation of neat AB, CC-AB, BNCC-AB, and BCCAB nanocomposites were carried out by differential scanning calorimetry (DSC) at multiple heating rates of 1, 2, 5, and 10 °C/min. At each rate, three runs were performed and the observed peak temperatures occurred in general at reproducible temperatures with less than 1˚C difference. Thermal decomposition of neat AB includes an endothermic dip (attributed to melting) and two exothermic maxima for two steps of hydrogen release. Fig. 3 shows the temperature dependency of the heat released and also the increase in the dehydrogenating peaks with rising heating rate which is in good agreement with literature.8

Figure 3. DSC exotherms for neat AB at heating rates of 1, 2, 5, and 10 ˚C/min. DSC exotherms for CC-AB, BNCC-AB, NCC-AB, and BCC-AB nanocomposites, at heating rates of 2, 5, and 10°C/min are depicted in Fig. 4 (exotherms for 1°C/min were almost flat at this scale, not shown). Similar to AB (Fig.3) dehydrogenation temperatures and heat flow for nanocomposites increases with increasing heating rate. Only in NCC-AB (Fig. 4c) the shift in dehydrogenation peak at different heating rate was very small but the reason is not known at this point. It can be seen that there is no endothermic peak for AB melting while confined in CCs, this observation can suggest the different start state of the AB inside the CC porous scaffold. Furthermore, the hydrogen release occurs in only one exothermic event and no other peak is observed at temperatures as high as 160°C which is indicative of different reaction pathways as compared to that of neat AB. Our previous studies using XRD, 11B NMR, and FTIR showed structural changes in AB and also different reaction pathways when AB confined in CCs to form alkyl borates in competition with conventional hydrogen loss polyborazylene products observed in neat AB decomposition [17, 25].

24

Advances in New Catalytic Materials

(a)

(b)

(c)

(d)

Figure 4. DSC exotherms for neat (a) CC-AB, (b) BNCC-AB, (c) NCC-AB, and (d) BCC-AB at heating rates of 2, 5, and 10 ˚C/min. (exotherm at heating rate of 1˚C/min was almost flat, not shown) Table 3 Peak temperatures for dehydrogenation of AB, CC-A, BNCC-AB AB

CC-AB

BNCC-AB

NCC-AB

BCC-AB

Heating rate (˚C/min)

First Tp (˚C)

Tp (˚C)

Tp (˚C)

Tp (˚C)

Tp (˚C)

10

124

109

107

104

112

5

118

102

99

103

104

2

112

94

92

102

93

1

107

88

85

101

84

Peak temperatures for dehydrogenations for nanocomposites along with the peak temperature for the first exothermic event in neat AB are shown in Table 3. Generally, in nanocomposites, dehydrogenation temperatures are lower as compared to those of neat AB as a result of large surface energy of nanosized AB in CCs. Different pore size distribution and peak pore size in CC, BNCC, NCC, and BCC affect the dehydrogenation peak, however, it can be seen that dehydrogenation happens at lower temperatures in all nanocomposites as compared to those of neat AB. Also, dehydrogenation temperatures for BNCC-AB is lower than those of CC- AB, while BNCC-AB shares similar broad pore size distribution but larger pore sizes as compared to CC-AB. This observation can show the effect of surface alteration and be attributed to further destabilization of AB due to the altered surface chemistry in BNCC-AB. Also, at lower heating rates the dehydrogenation temperatures in BCC-AB and BNCC-AB are similarly lower than that of CC-AB due to surface chemistry. The presence of B2O3 can provide surface interactions that disrupt the

Advanced Materials Research Vol. 132

25

dihydrogen bonding in AB and, therefore, lower the induction period for dehydrogenation and accelerate the dehydrogenation process. The different dehydrogenation peaks in BNCC-AB and BCC-AB at higher heating rate can be related to different pore size distribution that provides dissimilar nano-sized effects on AB and also results in different thermal conductivity in porous scaffold which can be a more dominant factor at higher heating rates. For NCC-AB, dehydrogenation peaks are lower than those of neat AB, but the peaks at different heating rates are very close. Activation energies are calculated for non-isothermal DSC runs using Kissinger equation given by [26]: ln (α / T 2) = – Ea/ RTp + constant (3) where Tp is the peak temperature, α = dT/dt is the heating rate and Ea is the activation energy. Plot of ln(α/ Tp2) versus 1/Tp is linear and the slope of the resulting line corresponds to the values of activation energies. Fig. 5 compares the Kissinger plots obtained for CC-AB, BCC-AB, and BNCCAB nanocomposites with that of neat AB. The plots for nanocomposites show the reduced slope as compared to that of neat AB, while boron-modified samples and especially BCC-AB shows the lowest slope. The Kissinger plot for NCC-AB (not shown in Fig. 5) was almost vertical which is not meaningful and it is purely due to the slight change in the dehydrogenation peak at different heating rates.

Figure 5. Kissinger plots for non-isothermal DSC runs for neat AB, CC-AB, BNCC-AB, and BCC-AB, at heating rates of 1, 2, 5, and 10 °C/min. Activation energies for the dehydrogenation were calculated from the slopes (Table 4). Activation energy for release of the first equivalent of hydrogen from neat AB is found to be ~ 160 kJ/mol which is comparable to the value reported in the literature [6]. The activation energies for nancomposites are smaller than that of neat AB, and BCC-AB reveals the lowest activation energy of 88 kJ/mol, which is about 55% of that of neat AB. The activation energies of nanocomposites can be attributed to the size dependent surface energy of AB confined inside the nanoscale pores of CC, and BNCC, and BCC. Carbon matrix can lower the hydrogen release barrier by reducing hydrogen diffusion distances, increasing the frequency of reaction which effectively accelerates the dehydrogenation process, and serving as efficient pathways for heat transfer. In addition, the lower activation energies in BNCC-AB and BCC-AB can be attributed to the surface interactions due to the presence of B2O3 that can destabilize AB furthermore and accelerate the dehydrogenation process. The lowest activation energy was observed in BCC-AB that can be attributed to both nanosized and catalytic effects resulting from smaller pores in a narrow pore size distribution in BCC (centered at ~9nm) and surface interactions.

26

Advances in New Catalytic Materials

Table 4 Activation energies for dehydrogenation of neat AB, CC-AB, BNCC-AB, and BCC-AB

Sample

Ea (kJ/mol)

AB

159

CC-AB

121

BNCC-AB

109

BCC-AB

88

The thermodynamics of dehydrogenation of AB, CC-AB, BNCC-AB, NCC-AB, and BCC-AB was compared by calculating the average enthalpy of dehydrogenation (Table 5) from the areas under the DSC exotherms (per mole AB) obtained at multiple heating rates. For neat AB, the area under the first peak is related to release of one mol hydrogen. The exothermic event was observed for nanocomposites is related to release of at least ~ 1.5 mol hydrogen (it is likely that some hydrogen may be lost during the fabrication of nanocomposites) [17]. Last column in Table 5 compares the corresponding enthalpies for the release of one mole hydrogen from AB, CC-AB, BNCC-AB, and BCC-AB. Higher enthalpy values in nanocomposites can be related to an alternative reaction pathway for the dehydrogenation of CC-AB, BNCC-AB, NCC-AB, and BCC-AB. In our previous study on dehydrogenation of CC-AB [25], new B-O peaks (from interaction of AB with surface oxygen) related to organic borates was observed in spectrum. The larger amount of oxygen and presence of boron on the surface of pores in BNCC and BCC can result in formation of new products and the increased exothermicity in BNCC-AB and BCC-AB as compared to that of CCAB. However, the exact mechanism is not known at this point. Table 5 Enthalpy of dehydrogenation calculated from DSC exotherms at multiple heating rates Sample

Enthalpy ∆H

Enthalpy ∆H

kJ/mol AB

kJ/mol H2

AB

-18.5 ±1.5

-18.5 ±1.5

CC-AB

-34 ± 3

-23± 2

BNCC-AB

-77 ± 4

-51 ± 3

NCC-AB

-62 ± 3

-41 ± 2

BCC-AB

-59 ± 3

-39 ± 2

Mass spectrometry was used to investigate the effect of the chemical modification on the purity of the released hydrogen in the thermal decomposition of neat AB, CC-AB, BNCC-AB, NCC-AB, and BCC-AB. Borazine is undesirable volatile product of AB decomposition since it can foul a fuel cell membrane and eventually render it unusable. Fig. 6 compares the borazine released in the decomposition of neat AB, CC-AB, BNCC-AB, and BCC-AB by using the intensities measured for hydrogen and borazine at 50°C (under vacuum) for each sample. Intensity of hydrogen is normalized to one for each sample. It can be seen that borazine is suppressed in all nanocomposites (but NCC-AB) as compared to neat AB. The Suppression of borazine provides a promising alternative for CC-ABs with fuel cell compatibility, while N modified CC does not seem to be a

Advanced Materials Research Vol. 132

27

proper choice for confinement of AB in hydrogen storage application due to increased borazine product. On the other hand, the decrease in the borazine release in CC-AB, BNCC-AB, and BCCAB, is possibly due to polymerization of borazine in cavities of carbon matrixes faster than it can escape to the gas phase [26].

Figure 6. Relative intensity of borazine to hydrogen for neat AB, CC-AB, BNCC-AB, and BCCAB nanocomposites. For easy comparison, intensity of hydrogen is normalized to one for each sample. The above results have clearly shown lowered dehydrogenation temperatures and improved reaction kinetics when AB confined in the B-modified CC scaffold as compared to that in CC or neat AB. N modification resulted in the increase of unwanted volatile product (borazine) during dehydrogenation of AB; however, the possible mechanism for such an increase in the formation of borazine requires further study. Conclusions This study has clearly demonstrated that the surface modification with B and/or N of CC scaffold not only reduced the dehydrogenation temperatures of AB confined inside the mesopores, but also noticeably changed the dehydrogenation kinetics of AB. Comparison of the hydrogen storage properties of CC-AB, BNCC-AB, NCC-AB, and BCC-AB nanocomposites suggested that Bmodification of CCs further promoted the destabilization of AB, and thus improved the kinetics of hydrogen release in BNCC-AB and BCC-AB as compared to that of CC-AB with the activation energy reduced from 159 kJ/mol for neat AB to 88 kJ/mol in BCC-AB nanocomposites. The study demonstrated that the dehydrogenation properties of hydrides can be readily tuned by altering the nanoporous structure and/or the surface chemistry of CC scaffolds. Further and better understanding of the mechanisms of the size confinement and surface catalytic properties is needed, which can be beneficial so as to predict and design CC- hydride nanocomposites with the ability of tuning the dehydrogenation temperatures and kinetics for specific applications. Acknowledgements. This work has been supported in part by National Science Foundation (DMI0455994 and DMR-0605159) and Air Force Office of Scientific Research (AFOSR-MURI, FA9550-06-1-0326). BBG acknowledges the University of Washington Bioenergy IGERT fellowship (DGE-0654252).

28

Advances in New Catalytic Materials

References [1] Fakioğlu, E.; Yurum, Y.; Veziroğlu, T.N. Int. J Hydrogen Energy 2004, 29, 1371. [2] Seayad, A. M.; Antonelli, D. M. Adv. Mater. 2004, 16, 765. [3] Baitalow, F.; Baumann, J.; Wolf, G.; Jaenicke-RoBler, K.; Leitner, G. Thermochm. Acta. 2002, 391, 159. [4] Baumann, J.; Baitalow, F.; Wolf, G. Thermochim. Acta. 2005, 430, 9. [5] Stowe, A.; Shaw, W.; Linehan, J.; Schmid, B.; Autrey, T. Phys. Chem. Chem. Phys. 2007, 9, 1831. [6] F. H. Stephens, V. Pons, R. T. Baker, Dalton Trans. 2007, 2613. [7] Hoon, C. F.; Reynhardt, E. C. J. Phys. C: Solid State Phys. 1983, 16, 6129. [8] Wolf, G.; Baumann, J.; Baitalow, F.; Hoffmann, F. P. Thermochim. Acta 2000, 343,19. [9] Hu, M. G.; Geanangel, R. A.; Wendlandt, W. W. Thermochim. Acta 1978, 23, 249. [10] Sit, V.; Geanangel, R. A.; Wendlandt, W. W. Thermochim. Acta 1987, 113, 379. [11] Benedetto, S. D.; Carewska, M.; Cento, C.; Gislon, P.; Pasquali, M.; Scaccia, S.; Prosini, P. P. Thermochim. Acta 2006, 441, 184. [12] Gutowska, A.; Li, L.; Shin, Y.; Wang, C.M.; Li, X.S..; Linehan, J.C.; Smith, R.S.; Kay, B. D.; Schmid, B.; Shaw, W.; Gutowski, M.; Autrey, T. Angew. Chem., Int. Ed. 2005, 44, 3578. [13] Vajo, J. J.; Olson, G. L. Scr. Mater. 2007, 56, 829. [14] Pekala, R. W.; Alviso, C. T.; Kong, F.M.; Hulsey, S.S. J. Non-Cryst. Solids 1992, 145, 90. [15] Al-Muhtaseb, S.A.; Ritter, J.A. Adv. Mater. 2003, 15, 101. [16] K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol and T. Siemieniewska, Pure Appl. Chem. 1985, 57, 603. [17] Feaver, A.; Sepehri, S.; Shamberger, P.; Stowe, A.; Autrey, T.; Cao, G.Z. J. Phys. Chem. B. 2007, 111, 7469. [18] S. Sepehri, B. B. García and G. Z. Cao, J. Mater. Chem. 2008, 18, 4034. [19]S. Sepehri, B. B. García and G. Z. Cao, Eur. J. Inorg. Chem. 2009, 5,599. [20] Tamon, H.; Ishizaka, H.; Yamamoto, T.; Suzuki, T. Carbon 1999, 37, 2049. [21] Pekala, R.W. J. Mater. Sci. 1989, 24, 3221. [22] Feaver, A.; Cao, G. Z. Carbon 2006, 44, 590. [23] P. V. Samant, F. Goncalves, M. A. Freitas, M. R. Pereira, J. L. Figueiredo, Carbon, 2004, 42, 1321. [24] E. A. Il’inchik, V. V. Volkov, L. N. Mazalov, J. Struct. Chem. 2005, 46, 523. [25] S. Sepehri, A. M. Feaver, W. J. Shaw, C. J. Howard, Q. Zhang, T. Autrey, G. Z. Cao, J. Phys. Chem. B, 2007, 111, 14285. [26] H. E. Kissinger, Anal. Chem. 1957, 11, 1702.

Advanced Materials Research Vol. 132 (2010) pp 29-37 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.29

Observation on the Structure of Ordered Mesoporous Materials at High Temperature via in situ X-ray Diffraction Chun Fang Zhou and Jian Hua Zhu* Key Laboratory of Mesoscopic Chemistry, School of Chemistry & Chemical Engineering, Nanjing University, Nanjing 210093, China E-mail:[email protected] Keywords: In situ XRD technique; Structure; Ordered mesoporous materials; Zeolites; High temperature

Abstract. This short paper reports the direct observation of the structure variation of mesoporous silica at temperatures higher than 600 oC by use of an in situ XRD technique. The mesostructure of SBA-15 or other mesoporous materials such as MCM-41 became almost invisible when the temperature rose to above 600 oC, but recovered or partially recovered once the temperature decreased. Contrarily, the characteristic XRD patterns of zeolites such as ZSM-5 kept unchangeable under the same conditions. On the basis of comparative experiments performed on various mesoporous samples, it is inferred that the reversible variation of XRD patterns probably originates from the thermal shock of the pore wall, not from the permanent collapse of the mesoscopic structure in these samples. This observation indicates the special features of SBA-15 at high temperature. Introduction Since the framework of ordered mesoporous materials provides a very robust, open and tunable periodic scaffold on the nanoscale [1], a great deal of effort has been directed toward their preparation, characteristics, morphological control, application and so on. In many applications, high hydrothermal and/or thermal stability of adsorbents or catalysts are required and consequently a lot of effort has been given to improve the structure stability of ordered mesoporous materials. However, the actual periodic structure of many ordered mesoporous materials at high temperature, which may differ from that under ambient conditions, is unclear up to now. Many scientists consider that the periodicity of the nanometer scale architecture in ordered mesoporous materials is held up to 800 oC [2]. However, they usually determine the ordering of ordered mesoporous materials at room temperature before and after thermal treatment, not examining it directly at high temperature. Recently the thermal stability of SBA-15 and Sn-SNBA-15 at high temperature is reported with an in situ HTXRD technique [3]. In this short paper we try to directly observe the mesoporous structure of many ordered mesoporous materials above 600oC by using in situ XRD technique. Experimental The details of sample MCM-41, MCM-48, HMS, SBA-15, KIT-6, SBA-16 and CMK-3 synthesis and the subsequent surfactant removal are described in references [2, 4-8]. The Al-containing SBA15 is prepared using HZSM-5 zeolites as part of the silica source and named as the SZ sample [9]. In a typical synthesis, the Al-containing SBA-15 was prepared as follows: 9.0 g KCl and 2 g

30

Advances in New Catalytic Materials

triblock copolymer P123 (EO20PO70EO20, Aldrich) were dissolved in 60 g 2 M HCl and 15 g H2O; 1.0 g finely-ground HZSM-5 zeolite (Si/Al = 12.5) was stirred vigorously in the clear solution at 30 oC for 0.2h and then well dispersed, followed by the addition of 4.25 g tetraethylorthosilicate (TEOS) while stirring at 30 oC. The mixture was stirred for 24h at 40 oC and heated at 100 oC for another 24h under static conditions; finally a white solid was filtered, washed and calcined at 550 oC to get about 2g samples. For in situ high temperature X-ray diffraction, finely ground powder samples were loaded into a HTK1200 high temperature chamber produced by Anton Paar GmbH in Austria. In situ low angle scattering was collected using Kα Cu irradiation from the Philips X’Pert X-ray Diffractometer. Diffraction patterns measured at constant temperature (20 min scanning time) follow how structure changes with temperature under a linear temperature ramp of 10oC/min in N2 gas (0.4 L/min) or vacuum (10-2 mba). TG/DTA curves were recorded on a Netzsch STA449C TG/DSC instrument. About 7.5 mg samples were heated in N2 gas with 25 ml/min from 25 oC to 800 oC at a heating rate 10 oC/min. Results and discussion A typical X-ray diffraction pattern of the SBA-15 samples in the HTK 1200 high temperature chamber at 25 oC (before increasing the temperature) is shown in Figure 1. The intense (100) diffraction peak and the existence of two higher order peaks, (110) and (200) indicated the p6mm periodicity of the pore structure in SBA-15 samples. Although SBA-15 is believed to have high thermal and hydrothermal stability above 800 oC [2, 10, 11], the actual periodicity of the mesoporous structure at a high temperature of 800 oC is seldomly observed. Actually, the intensity of the diffraction pattern of the sample gently decreased with the increasing temperature, and all diffraction peaks disappeared above 600 oC. More interesting, all diffraction peaks reappeared very well once the temperature was gently decreased as illustrated in Figure 1. This phenomenon implies that high thermal and hydrothermal stability of SBA-15 above 600 oC is controversial, and probably relates to the synthesis condition. In fact, the nanoscale periodicity of mesopores in SBA-15 seemed to disappear above 600 oC because of the unseen XRD profile that is based on the periodicity of

Figure 1: In situ XRD pattern for SBA-15 at different temperatures in nitrogen.

Advanced Materials Research Vol. 132

31

Figure 2: In situ XRD pattern of SZ sample in 2θ range of (upper) 0.5 – 2.5o and (below) 7-9° in nitrogen. mesostructure in the sample, though it could recover when the temperature decreased, similar to that reported by Shah and Ramasvamy [3]. At the first glance, this XRD result was absurd, but the same experiment was repeated several times and gave similar phenomena. In order to validate the above results, SZ samples with ordering in both mesopore and micropore ranges were selected [9]. Low-angle XRD pattern of SZ samples, showing unique excellence, exhibited not only typical p6mm symmetry in the 2θ range of 0.5~3 o, but also the diffraction peaks of HZSM-5 zeolite in the 2θ range of 7~9 o as shown in Figure 2, despite the fact that the diffraction intensity in 2θ = 7~9o was very weak. Similar to SBA-15, (100), (110) and (200) peaks for SZ’s mesostructure totally disappeared at 600 oC and above, but reappeared with the decreasing temperature. Meanwhile, the diffraction peaks for HZSM-5 kept well at 800oC while their intensities changed a little with the temperature as demonstrated in Figure 2. The basic structural building unit of zeolites is well known to be the TO4 tetrahedron that is quite rigid and with all corners connected via common oxygen to form a three-dimensional framework. Consequently, this periodicity of such framework results in the X-ray diffraction of zeolites [12]. The diffraction peaks

32

Advances in New Catalytic Materials

related to ZSM-5 units kept well at 800oC, probably due to the rather rigid framework composing of TO4 tetrahedron and the pore structure (0.54×0.56 nm). Contrarily, the periodicity of ordered mesoporous materials is from the ordered array of the mesoporous structure, and their pore wall is amorphous instead of crystalline [2]. XRD results of SZ sample revealed that the disappearance and reappearance of the ordering with the programmed temperature cycle was specific on the mesoscopic scale, not on the atomic scale or TO4 tetrahedron. One may argue these phenomena of structural variation come from the inherent drawback of the instrument, say, the thermal vibration of the sample holder itself causes the disappearence and re-appearance of XRD patterns in the low angle range, and such influece may be more serious on the lower angle range of 0.5 o – 3 o than that of 7~9 o. This argument, however, is not justfied by the further experiments.

Figure 3: In situ XRD pattern of HMS and CMK-3 samples in N2 gas.

Advanced Materials Research Vol. 132

33

Figure 4: The In situ XRD patterns of sample KIT-6, MCM-48 and SBA-16 measured in N2 gas flow.

34

Advances in New Catalytic Materials

This gradual disappearance and reappearance of mesoscopic ordering with the temperature changing seemed to be ubiquitous in the mesoscale, not only in the case of ordered mesoporous materials with p6mm symmetry but also in those with different symmetry or structure, either the silica host or carbon replica. In our measurements, several samples such as MCM-41, HMS with worm-like pores, MCM-48 and KIT-6 with Ia3d symmetry and SBA-16 with Im-3m symmetry all had similar behavior in the same programmed temperature cycle, though main diffraction peak intensity for these materials recovered to quite different extents, probably owing to their different symmetries and/or structure stabilities. The XRD profiles of SBA-15, MCM-41 and CMK-3 with 2D pore network disappeared at 600 oC (Figure 3). However, a different situation was observed on MCM-48, KIT-6 and SBA-16 samples with 3D pore system as shown in Figures 4, 5 and 6. The periodicity of mesoporous structure was still detected at 800 oC in the MCM-48 sample that was synthesized in alkaline conditions, whereas the periodicity remained at 700 oC in the samples of SBA-16 and KIT-6 that were prepared in acidic conditions. This difference may relate to their different synthesis conditions, since silica species are less condensed linear oligomers in acidic conditions, while in alkaline solution, the silica species are more cross-linked clusters [13]. Furthermore, these results proved the utility of both the X-ray diffraction instruments and the samples flake figure for the determination in the low-angle range under high temperatures because the ordered mesoporous structure was still detected in the MCM-48 sample at 800 oC. That is to say, the in situ XRD method was feasible under high temperature. Otherwise, all mesoporous silica samples should give the same disappearence and re-appearance of XRD patterns in the low angle range.

100

0.2 DSC

95

-0.2 -0.4

90

-0.6

TG

-0.8

DSC (mW/mg)

TG (%)

0.0

-1.0

85 100

200

300

400

500

600

700

-1.2

o

Temperature ( C) Figure 5: TG/DTA spectrum of SBA-15 sample If the programmed temperature cycle was operated in vacuum instead of in N2 gas, all diffraction peaks of SBA-15 disappeared at temperatures of 600 oC and above, but returned with the temperature decreasing, similar to the phenomenon in N2 gas. However, the recovery of (100) peak intensity in SBA-15 after the temperature cycle was only 58 %, much less than that in N2 gas. This alludes that both the vacuum and high temperature may lead to more permanent destruction of mesoscopic ordering in the ordered mesoporous materials. If the temperature was kept at 800 oC for 2h in N2 gas, the peak intensity in SBA-15 recovered 66 % after the temperature cycle. It is well

Advanced Materials Research Vol. 132

35

known that high temperature operation leads to the destruction of mesostructure [11]. However, mesoscopic ordering in the ordered mesoporous materials could disappear and partially or completely reappear with the temperature changing, which is a phenomenon that has been ignored for a long time, and may be beneficial for understanding the real state of ordered mesoporous materials at high temperature. To have a deeper inspection on the structural property of ordered mesoporous materials, a TG/DTA measurement of SBA-15 was completed in order to investigate possibility of the structure breaking, as shown in Figure 5. Physisorbed and hydrogen-bonded water was removed from the sample up to 200 °C (endothermic), while dehydroxylation of silanol groups (endothermic) occurred from 200 °C to 800 °C smoothly. No obvious endothermic or exothermic peaks occurred at 600 oC and above, only a gradual endothermic process, indicating no breaking of the mesostructure, but a gradual change at high temperature. TG/DTA results of MCM-48 and MCM-41 were similar to that of SBA-15 [14]. Based on these results it is safe to infer that the disappearance of periodicity in mesoporous silica at 600 oC and above is a gradual change process instead of a breaking one. It has been reported that diffraction intensity weakens with the temperature increasing owing to the increase of thermal shock vibration of the atoms [15]. As for zeolites, their ordered diffraction patterns result from the periodicity of the atomic arrangement. The tetrahedra are not free to move independently of each other, due to their covalent bonds [12]. Therefore, thermal shock of T atoms above 600 oC has little effect on diffraction intensity of zeolites as shown in Figure 2. For the ordered mesoporous materials, the situation is different: their ordering is in periodicity of pore arrangement, while the pore wall is amorphous. Two reasons should be taken into account for the disappearance of diffraction peaks in SBA-15 at high temperature. One is the collapse of ordered structure, and the other is thermal shock of the pore wall at high temperature. In fact, there exists partial collapse in SBA-15 at above 600 oC since the polymerization degree of Si-O-Si linkages is greatly enhanced [11]. However, polymerization of Si-O-Si is irreversible so that such collapse cannot recover even as the temperature lowers. Thus, the disappearance of diffraction peaks observed in our experiments should be attributed to thermal shock of the pore wall, not to permanent collapse of mesoscopic structure. At the atomic level the high temperature vibration of the lattice may shift very slightly the mutual distances between parallel channels in MCM-41 and SBA-15, which, however, is sufficient for disappearance of the XRD profile. This thermal shock of mesoporous structure induces the disappearance of the periodicity in ordered mesoporous materials owing to the anisotropic thermal effect [16]. Accordingly, thermal shock at high temperature existed at not only the atomic level, but also the mesoscopic level, and the influence in diffraction intensity becomes more obvious in mesoscopic ordering. Pore walls of MCM-48 or SBA-16 possess two branched 3D pores intercrossing and thus the thermal shock of pore walls is not free to move independently of each other, instead of cooperatively as the framework of zeolites does, so that the influence of thermal effect on MCM-48 is weaker than that on the ordered mesoporous materials with 2D pore system. If ordered mesoporous materials are used as catalysts supports or other applications at high temperature, their function certainly depends on their ordering of mesopores on which an active component is dispersed in the form of very small particles [17]. Thermal shock of pore walls under high temperature may affect the distribution and property-function of the guest more or less, altering their performance in catalysis or other processes. In addition, these results indicate that arrangement of pore structures also affects thermal stability of ordered mesoporous materials.

36

Advances in New Catalytic Materials

Therefore the intercrossed pore wall structure such as MCM-48 and SBA-16 with 3D pore structures may be firmer than the ordered mesoporous materials with simple pore walls at high temperature. 4. Conclusion The mesoporous structure of the materials such as SBA-15, MCM-41, MCM-48 and SBA-16 above 600 oC was directly examined by using an in situ XRD technique, and an unusual phenomenon was found. All samples showed declined XRD patterns in the 2-theta range of 0.5-3 o as the temperature rose to 600 oC or above, but recovered or partially recovered once the temperature decreased. Such phenomenon was absent in the 2-theta range of 7-9 o, so it is specific on the mesoscopic scale, not on the atomic scale or TO4 tetrahedron. This probably originates from the thermal shock of the pore wall, instead of the permanent collapse of the mesoscopic structure. Acknowledgment Financial support from 863 Program of MST of China (Grant 2008AA06Z327), NSF of China (20773061 and 20873059) and Analysis Center of Nanjing University is gratefully acknowledged. The authors are grateful to Professor D. Y. Zhao (Fudan University, China) for supplying the SBA-16 sample and beneficial discussion. References [1] Z. Kónya, V. F. Puntes, I. Kiricsi, J. Zhu, A. P. Alivisatos and G. A. Somorjai: Nano Lett., 2 (2002), p907. [2] D. Y. Zhao, J. L. Feng, Q. S. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka and G. D. Stuky: Science 279 (1998), p548. [3] P. Shah and V. Ramasvamy: Micropr. Mesopr. Mater. 114 (2008), p270. [4] F. Kleitz, S. H. Choi and R. Ryoo: Chem. Comm. (2003), p2136. [5] C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck: Nature 359 (1992), p710. [6] P. T. Tanev and T. J. Pinnavaia: Chem. Mater.8 (1996), p2068. [7] S. Jun, S. H. Joo, R. Ryoo, M. Kruk, M. Jaroniec, Z. Liu, T. Ohsuna and O. Terasaki: J. Am. Chem. Soc. 122 (2000), p10712 [8] D. Y. Zhao, Q. S. Huo, J. L. Feng, B. F. Chmelka and G. D. Stucky: J. Am. Chem. Soc. 120 (1998), p6024. [9] C. F. Zhou, Y. M. Wang, J.H. Xu, T. T. Zhuang, Y. Wang, Z.Y. Wu and J. H. Zhu: Stud. Surf. Sci. Catal. 156 (2005), p907. [10] Z. Zhang, Y. Han, L. Zhu, R. Wang, Y. Yu, S. Qiu, D. Zhao and F. S. Xiao: Angew. Chem., Int. Ed. 40 (2001), p1258. [11] F. Q. Zhang, Y. Yan, H. Yang, Y. Meng, C. Yu, B. Tu and D. Zhao: J. Phys. Chem. B, 109 (2005), p8723. [12] H. G. Karge and J. Weitkamp: Molecular Sieves – Science and Technology Vol.2: p141, Springer, 1999 [13] H. P. Lin and C. Y. Mou: Acc. Chem. Res. 35 (2002), p927. [14] H. Landmesser , H. Kosshck, W. Storek and R. Fricke: Solid State lonics (101) 1997, p271.

Advanced Materials Research Vol. 132

[15] R. Jenkms and R.L. Snyder: Introduction to X-ray Power Diffractometry, New York: John Wiley & Sons. Inc., 1996. [16] L. A. Villaescusa, P. Lightfoot, S. J. Teat and R. E. Morris: J. Am. Chem. Soc. 123 (2001), p5453. [17] A.T. Bell: Science 299 (2003), p1688.

37

Advanced Materials Research Vol. 132 (2010) pp 38-44 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.38

Synthesis of Si-based Mesoporous Materials with Different Structural Regularity Lifang Chen1, a,*, Jesus López1, Jin-An Wang1, Luis E.Noreña2 Guoxian Yu3,b, Fahai Cao4, Yueqing Song4, Xiaolong Zhou4 1

ESIQIE, Instituto Politécnico Nacional, Col. Zacatenco, 07738 México D. F., Mexico.

2

Departamento de Ciencias Básicas, Universidad Autónoma Metropolitana-A, Av. San Pablo 180, Col. Reynosa-Tamaulipas, 02200 México D.F., Mexico 3

School of Chemistry and Environment Engineering, Jianghan University, Wuhan, Hubei Province, 430056, P. R. China

4

School of Chemical Engineering, East China University of Science andTechnology, Shanghai 200237, P. R. China a

[email protected], [email protected]

Keywords: Si-MCM-41, TEOS, fumed silica, structure ordering, textural property, surface acidity.

Abstract. Two types of mesoporous Si-MCM-41 materials were synthesized via a cationic surfactant template method using different Si-precursors. The materials obtained were characterized by FTIR, XRD, BET, TEM and 29Si MAS-NMR techniques. When fumed silica was used as Si precursor, a Si-MCM-41-I solid with wormhole-like pore topologies was obtained. However, when tetraethylorthosilicate (TEOS) was used as Si precursor, a mesoporous Si-MCM-41-II solid with hexagonal arranges and a long-range ordered structure could be obtained. These two kinds of mesoporous materials had a uniform pore size distribution with an average pore diameter within 2.3-2.8 nm. Rather weak Lewis acid sites were formed on both the Si-MCM-41 samples prepared by the two methods.

1. Introduction In recent years, intensive research has been focused on the development of new synthesis approaches, aiming at obtaining new mesoporous materials type MCM-41 and understanding the role of synthesis parameters in the formation of the mesoporous structure. Due to their larger surface area, controlled big pore size and a very narrow pore size distribution, the ordered Si-based mesoporous materials have a great potential in fine chemistry, the pharmaceutical industry, as well as for the production of special polymer materials [1-5]. There are many factors in the synthesis procedure, for example, metal precursors, and surfactant concentration, pH of the solution or reaction time and temperature these may greatly affect the structural properties of the resulting materials. In the present work, we report the synthesis of mesoporous silica type MCM-41 by two different pathways, in order to investigate the effect of the Si precursors on the physicochemical properties and structure regularity of the obtained mesoporous materials. 2. Experimental 2.1 Synthesis of the Si-MCM-41 samples Si-MCM-41-I sample was synthesized by using fumed silica as Si precursor and cetyltrimethylammonium chloride (CTACl) as surfactant template. 0.6 g of fumed silica was added into 5.4 g tetrabutylammonium hydroxide (TBAOH, 40 wt. %) aqueous solution with vigorous

Advanced Materials Research Vol. 132

39

stirring for 5 min to form a transparent and uniform gel. Then 12 g of CTACl (25 wt. % solution in water) were added into the above gel with agitation. Afterwards, 1 g of fumed silica was immediately added into the above mixture which was vigorously agitated for approximately 15 min. The resultant gel was loaded into a stoppered Telfon bottle, kept without stirring at 100 ºC for 48 h. After cooling to room temperature, the resulting solid product was recovered by filtration, washed extensively with 500 ml deionized water for 4 times. The white solid was dried at 80 ºC for 24 h. Finally, the sample was calcined at 600 ºC for 6 h in air, with a flow rate of 60 ml/min. Si-MCM-41-II solid was prepared using TEOS as Si precursor and CTACl as synthesis template. 10.62 g of CTACl were dissolved into 110 ml of hot deionized water (around 40 ºC) to obtain the first solution. Then 110 ml of an ammonia aqueous solution (28 vol.%) were added into the surfactant solution with agitation. The ammonia solution can be added in two stages, 55 ml each time. The following step was to rapidly add 22.2 ml of TEOS (98 wt.%) into the above mixture with vigorous agitation for 30 min. The resultant gel was loaded into a stoppered Telfon bottle, kept without stirring at 100 ºC for 48 h. The other steps are the same as described in the synthesis of Si-MCM-41-I. 2.2 Characterization The low angle X-ray diffraction patterns of the samples were measured in a D-500 SIEMENS diffractometer with a graphite secondary beam monochromator to obtain a monochromatic CuKα1 radiation, and the evaluation of the diffractograms was made by DIFFRAC/AT software. The scanning was made from 1.5 to 10, 2θ step size of 0.02 and step time of 2 s. In order to avoid the problem of illuminated area at low 2 theta angle, all the samples were measured using the same sample holder. In this way the hexagonal reflection (100) planes as well as the intensities are directly comparative. Position correction was made using the NIST standard reference material 675. The specific surface area, pore volume and pore size distribution of the samples were measured in a Digisorb 2600 equipment by using low temperature N2 physisorption isotherms. Before the measurement, the sample was evacuated at 350 ºC under vacuum condition. The surface area was calculated using the BET method based on the adsorption data within the partial pressure P/Po range from 0.01 to 0.3. To evaluate and analyze the strength and types of acid sites, pyridine adsorption on the samples was performed on a 170-SX Fourier-Transform infrared (FTIR) spectrometer at different temperatures, ranging from 25 to 400 °C. Before pyridine adsorption, each sample was heated to 300 ºC for 60 min under vacuum, in order to eliminate the adsorbed water or impurities on the surface, and then cooled to room temperature. Afterwards, the solid wafer was exposed to pyridine, by breaking inside the spectrometer cell, a capillary containing 50 µl of liquid pyridine. The IR spectra were recorded at various conditions by increasing the cell temperature from 25 to 400 °C. The quantitative calculation of Lewis acid sites and Brönsted acid sites was made with respect to the area of the adsorption band at 1450 cm-1 and 1540 cm-1, respectively. The acid strength was determined with respect to the variation of the number of acid sites as a function of the temperature. Solid-state 29Si MAS-NMR spectra were recorded on a Bruker 400 MHz spectrometer at a frequency of 79.49 MHz, 7.5 kHz spinning, and using pulses at 90 s intervals and 4 mm zirconia rotors. For the 29Si analysis, tetramethylsilane was used as shift standard (0 ppm); Deconvolution of the spectra was performed with the Unity spectrometer software. Transmission electron microscopy (TEM) images of samples were carried out in a JEM-2200FS transmission electron microscope with accelerating voltage of 200 kV. The microscope is equipped with a Schottky-type field emission gun and an ultra high resolution (UHR) configuration (Cs = 0.5 mm; Cc 1.1 mm; point-to-point resolution, 0.19 nm) and in-column energy filter omegatype. The powder samples were grounded softly in an agate mortar and dispersed in isopropyl alcohol in an ultrasonic bath for several minutes. A few drops were then deposited on 200 mesh copper grids covered with a holey carbon film.

40

Advances in New Catalytic Materials

3. Results and Discussion 3.1. XRD analysis

100

The small angle X-ray diffraction pattern of the as-made Si-MCM-41-I sample is shown in Figure 1. Four Bragg peaks between 2 and 7° were observed, these can be indexed to a hexagonal lattice of (100), (110), (200) and (210) planes. Since the XRD peaks are very sharp, the hexagonal arrangement is, therefore, well ordered. After calcination, the four XRD peaks were replaced by a broad peak located at 2.9 °. This result clearly indicates that the structural order of calcined sample was strongly reduced. The channels derived from the ordered hexagonal arrangement, became rather disordered, forming a wormhole-like packing. The XRD patterns of the Si-MCM-41-II sample synthesized by the Method II are very sharp with good resolution, corresponding to the (100), (110), (200) and (210) planes of a hexagonal lattice. After calcination at 600 ºC, these reflections became much sharper, the intensities of all the peaks greatly enhanced, indicating that after the removal of the templating molecules from the pores, the mesopores may be rearranged into a more ordered form. In addition, the position of the peaks shifts in the direction to higher 2θ after calcination, therefore, the cell parameter a0 becomes smaller and thus the cell volume decreases or the lattice cell dimensions contract. It is evident that the Si-MCM-41-II prepared using TEOS as Si precursor has greater stability and structural regularity than the one obtained using fumed silica as Si precursor. These results show that the use of a different Si precursor strongly affects the structural regularity and thermal stability of the solids.

15000

12000

(100)

Si-MCM-41-I 80 ºC Si-MCM-41-I 600 C

o

600 C d100= 36.8 Å 6000

3000

1000 (110) (200)

210

2000

o

80 C d100 = 40.2 Å

9000

110

Intensity (a.u.)

3000

Intensity (a.u.)

Si-MCM-41-I

200

4000

Si-MCM-41-II

(210)

0

0

2

4

6

2-theta (degrees)

8

10

2

4

6

8

10

2-theta (degrees)

Figure 1. XRD patterns of the Si-MCM-41 samples. 3.2. Textural properties Figure 2 presents the profiles of the N2 adsorption-desorption isotherm of the Si-MCM-41-I sample. The inset is the pore size distribution profile. It is a type IV which exhibits four stages of adsorption: (1) The first stage (p/p0 < 0.2) is due to a monolayer adsorption of nitrogen on the walls of the mesopores. (2) The second stage (0.25 0.8, it is associated with capillary condensation in the interparticles of the mesoporosities. The adsorption and desorption lines in Figure 2 are almost overlapped, leading to no hysteresis loop formation. The absence of a hysteresis loop together with the sharp curvature along the second stage, suggests the existence of uniform and cylindrical channels throughout the material and similar behaviors of the adsorption and desoprtion procedures. Most of the mesopores have diameter around 2.7 nm. This solid has a surface area of approximately 935.8 m2/g and a pore volume of 0.68 cm3/g (Table 1). Table 1. Textural properties

S (m2/g)

Sample Si-MCM-41-I Si-MCM-41-II

Dp (nm)

935.8 1305.1

2.7 2.3

Vp (cc/g) 0.68 0.89

Similar N2 adsorption-desorption behaviors were observed in Si-MCM-41-II sample (Figure 2). It has surface area 1305 m2/g and the average pore diameter around 2.3 nm. Since all the inflections of N2 adsorption isotherm are very sharp, thus, Si-MCM-41-II synthesized using TEOS as Si precursor exhibits a very narrow pore size distribution. 600 DV/dlog (D) (cc/g)

4

Vol Adsorbed, (cc/g)

500

400

27.2 (Å) 3

Si-MCM-41-I

Si-MCM-41-II

2 1 0 10

100

1000

Pore Diameter, (Å)

300

ads des

200 0,0

0,2

0,4

0,6

0,8

1,0

Relative Pressure, (p/p0)

Figure 2. N2 adsorption-desorption isotherm of the Si-MCM-41 samples 3.3. Morphological features The TEM image of solid Si-MCM-41-I shows rather disordered channel arrays together with regions of regular ordering of the pores (Figure 3, left). TEM micrographs fully correlate to XRD analysis, which confirm wormhole-like pore topologies of this material. At high magnification, a “honeycomb” arrangement of the surface in some areas could be clearly observed, even in sample showing only a single diffraction peak in the XRD (Figure 1, left). TEM images for the calcined SiMCM-41-II solid shows a regular parallel ordering of the pores and their continuity through the entire length of the particles (Figure 3, right). It is clearly illustrated that the pore structure is well ordered with hexagonal arrays; the pore diameter is estimated to be approximately 2-2.5 nm. Obviously, the structural ordering of the Si-MCM-41-II is better than that of the Si-MCM-41-I.

42

Advances in New Catalytic Materials

Figure 3. TEM image of the Si-MCM-41 samples calcined at 600 ºC. left: Si-MCM-41-I and right: Si-MCM-41-II 29

3.4.

Si MAS-NMR spectroscopic analysis

Figure 4 shows the 29Si MAS-NMR spectra of the two samples. A broad peak between –90 and 125 ppm was observed, which can be deconvoluted to three main components with chemical shifts at ca. –92, -102 and –110 ppm, respectively. These signals resulted from Q2 (-92 ppm), Q3 (102 ppm) and Q4 (-110 ppm) silicon nuclei, where Qx corresponds to a silicon nuclei with x siloxane linkages in the framework, i.e., Q2 corresponds to disilanol Si–(O-Si)2(-O-X)2, where X is H or cationic defect; Q3 to silanol (X-O)-Si–(O-Si)3 and Q4 to Si–(O-Si)4 [6, 7]. The fraction of Q4 remains predominant, showing that silicon nuclei in the framework are mostly linked with O-Si ions. The ratio of (Q2+Q3)/Q4 in the Si-MCM-41-II solid is about 40 %, which is less than that shown on the sample Si-MCM-41-I where the (Q2+Q3)/Q4 ratio is approximately 49 % (Table 2). A larger (Q2+Q3)/Q4 value indicates more defects in the structure or a less ordered pore system. This result indicates that the Si-MCM-41-II solid has higher structural ordering in comparison with the Si-MCM-41-I solid, which is in good agreement with the XRD analysis and TEM observation.

4

Q

Q

Si-MCM-41-II

4

Si-MCM-41-I

3

Intensity (a.u.)

Intensity (a.u.)

Q

Q

2

Q

Q

0

-50

-100 δ (ppm)

-150

-200

0

-50

3

2

-100 δ

(ppm)

Figure 4. 29Si MAS-NMR spectra of the Si-MCM-41 samples

-150

-200

Advanced Materials Research Vol. 132

43

Table 2. Deconvolution of the 29Si MAS-NMR spectra.

Samples Si-MCM-41-I Si-MCM-41-II

Q2

Q3

Q4

(Q2+Q3)/Q4

3.6 1.6

29.4 26.9

67.0 71.5

49 40

3.5. Surface acidity The surface acidity of the Si-MCM-41-I sample was determined by the FTIR pyridine adsorption technique (Figure 5, left). When pyridine was adsorbed on the surface of the Si-MCM-41 solids at 25 °C, several peaks were observed at 1445, 1580 and 1595 cm-1. These are very typical characteristic of the vibrations of pyridine associated with Lewis acid sites [8, 9]. No band around 1540 cm-1 corresponding to Brönsted acids was observed, therefore, we conclude that on the surface of Si-MCM-41-I only Lewis acid sites existed. When the temperature of the IR cell increased to 100 °C, all the bands remained almost unchanged. However, at 200 °C, all the peaks entirely disappeared, indicating that the strength of the Lewis acidity is weak.

2,5

1,0

1445 L

1595 L

1578 L

0,6 25°C 0,4

0,2

0,0 1700

Si-MCM-41-II

2,0

Intensity (a.u.)

Absorbance (a.u.)

Si-MCM-41-I 0,8

100°C

1,0 25°C 0,5

200°C

1650

1,5

1600

1550

1500

Wavenumbers (cm-1)

1450

1400

0,0 1700

100°C 200°C 1650

1600

1550

1500

1450

1400

-1

Wavenumbers (cm )

Figure 5. A set of FTIR-in situ spectra of pyridine adsorption on the Si-MCM-41-I and Si-MCM-41-II at various temperatures. Similar to the Si-MCM-41-I sample, the Si-MCM-41-II shows sharp adsorption bands at 1445 cm and 1595 cm-1, corresponding to Lewis acid sites (Figure 5, right). No Brönsted acid sites exist on the solid because on band at 1545 cm-1 was observed. Compared to Si-MCM-41-I, the acidity of the Si-MCM-41-II is weaker. This result indicates that Si-MCM-41 materials, independent of the Si precursor fumed silica or TEOS, have a weak acidity. -1

5. Conclusions Si precursors exhibit a great influence on the thermal stability and structural regularity of the mesoporous Si-MCM-41materials. When fumed silica was used as Si precursor, the obtained SiMCM-41-I is thermally unstable, after calcination at 600 ºC, its ordered hexagonal arrangements became disordered in some degree, forming a wormhole-like packing. While the Si-MCM-41-II prepared by using TEOS as Si precursor shows high thermal stability with highly ordered hexagonal structure. After calcination at 600 °C, its hexagonal structure regularity could be further enhanced.

44

Advances in New Catalytic Materials

The mesopore diameter of these two solids is distributed within a narrow range between 2.2 and 2.8 nm and the surface area is 935 and 1305 m2/g, respectively. Only weak Lewis acid sites are present on both Si-MCM-41 materials. Acknowledgments The authors are thankful to the financial support from projects CONACyT-51007 and IPN-SIP20100733. References [1] A.Corma: Chemical Reviews Vol. 97 (1997), p. 2373. [2] B.M. Weckhuysen, R. Ramachandra Rao, J. Pelgrims, R. A. Schoonheydt, G. Bodart, P. Debras, O. Collart, P.Van Der Voort, E. F.Vansant: Chemistry: A European Journal Vol. 6 (2000), p. 2960. [3] J. H. Clark: Green Chemistry Vol. 1 (1999), p.1. [4] A. Karlsson, M. Stöcker, R. Schmidt, Micro. and Meso. Mater. Vol. 27 (1999), p.181. [5] A. Sayari, M. Jaroniec (editors): Nanoporous Materials IV, Stud. Surf. Catal. Sci., Vol. 141 (2002). [6] D. J. Rosenberg, F. Coloma, J. A. Anderson: J. Catal. Vol. 210 (2002), p. 218. [7] G. Engelhardt, D. Michel: High-Resolution Solid State NMR of Silicates and Zeolites (John Wiley & Sons, 1987). [8] T. López, J. Navarrete, R. Gómez, O. Novaro, F. Figueras, H. Armendáriz: Appl. Catal. A: General Vol. 125 (1995), p. 217. [9] J. Sánchez-Valente, X. Bokhimi, F. Hernández: Langmuir Vol. 19 (2003), p. 3583.

Advanced Materials Research Vol. 132 (2010) pp 45-54 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.45

Comparative Studies of the CoMo/MgO, CoMo/Al2O3 and CoMo/MgOMgAl2O4 Catalysts Prepared by a Urea-matrix Combustion Method Libao Wua, Dongmei Jiaoa, b, Lifang Chenc, Jin-An Wangc, Fahai Cao a,+ a

School of chemical engineering, East China University of Science and Technology, Shanghai 200237, China b

c

Luoyang Petrochemical Complex of SINOPEC, Luoyang 471000, China

ESIQIE, Instituto Politécnico Nacional, Col. Zacatenco, C. P. 07738 México D.F., Mexico + Email: [email protected]

Keywords: MgO–MgAl2O4; hydrodesulfurization; urea-matrix combustion method.

Abstract. Three CoMo supported catalysts with different supports, Al2O3, MgO and MgOMgAl2O4, were prepared by a urea matrix combustion method. The physicochemical properties of the catalysts were characterized by N2 isothermal adsorption–desorption, powder X-ray diffraction (XRD) and temperature programmed reduction (TPR) techniques. The activity of these catalysts was evaluated in a fixed-bed high-pressure reactor using hydrodesulfurization of dibenzothiophene as a model reaction. The urea matrix combustion preparation method greatly favored the formation of highly dispersed Co- and Mo-oxo species on the support, which had significant influence on the hydrodesulfurization (HDS) activity. XRD analysis showed that MgO was more sensitive to the deposition of Co-O or Mo-O species than Al2O3 and MgAl2O4; the former might be potentially used as an indicator of the Co- and Mo-oxo species formation. Among these catalysts, CoMo/MgOMgAl2O4 exhibited a high HDS activity. 1. Introduction Environmental regulations for sulfur specification in transportation fuels are becoming more and more stringent. The content of sulfur in diesel fuel will be limited to less than 10 ppm by 2011 throughout the world [1]. The catalytic activity of traditional catalysts is rather limited for sulfur deep removal in transportation fuels. To meet the aforementioned challenge, various attempts, for example, the creation of a preparation method by changing the composition of active metals, and the use of different supports and additives, have been made to improve the HDS activity [2]. González-Cortés et al. [3] reported an interesting synthetic method named urea matrix combustion for preparation of CoMo/Al2O3 catalyst, and experimental results showed that it improved the HDS activity. This method facilitates the formation of well-dispersed Co- and Mo-oxo species; it differs from the conventional impregnation techniques that usually lead to mixed-metal oxide formation on the Al2O3 support. Additionally, the urea-matrix combustion method may induce

46

Advances in New Catalytic Materials

a strong interaction between the promoters and Mo-oxo species on alumina, and illustrates a selective interaction between the supported metal oxide species [4]. In the design of the new HDS catalysts, selection of proper support materials is one of the key factors. Many single and combination oxides, such as zeolites [5], ZrO2 [6], MgO [7], Al2O3-TiO2 [8], TiO2-ZrO2 [9], MgO-Al2O3 [10], etc. have been used as catalysts support for HDS. Among these, MgO is a promoting support for HDS catalyst due to its basicity, which might increase the dispersion of acidic molybdenum species on it [11]. And, the promotional effect of Co and Ni on Mo supported catalysts can be enhanced in the case of a basic support compared to an acidic one [12]. However, such a strong interaction may make the sulfidation of the active component difficult, and its poor mechanical strength is a disadvantage for MgO to be a support. Fortunately, these shortcomings can be inhibited by introduction of a proper amount of Al2O3 into the lattice of MgO, which may not only increase the mechanical stability, but also creates a structural promoting role in the support-metal interaction and thus favors the HDS reaction. In the present work, three kinds of catalysts (CoMo/MgO, CoMo/Al2O3 and CoMo/MgOMgAl2O4) were prepared by the urea matrix combustion method. The HDS activities of the CoMosupported catalysts were tested using a fixed-bed high-pressure stainless reactor and the hydrodesulfurization of dibenzothiophene as a model reaction. It has been found that Co and Mo species were highly dispersed on CoMo/MgO-MgAl2O4 because of a strong interaction among the metal oxo species, which leads to a high HDS activity. 2. Experimental 2.1 Preparation of catalysts A MgO-MgAl2O4 (MgO:Al2O3=4:1, w/w) support was prepared by a homogenous precipitation method using aqueous solution of aluminum nitrate and magnesium nitrate (Sinopham Chemical, Shanghai), where a 10 wt.% water solution of NH3 (Ling Feng Chemical, Shanghai) was used as a precipitating agent to control the pH value of the solution at approximately 9.2. The mixture was stirred for 0.5 h. Then, the precipitate was filtered, followed by drying at 120 ºC for 12 h and calcination at 550 ºC for 4h. Pure Al2O3 and MgO supports were also prepared by the same method, by using aluminum nitrate and magnesium nitrate as Al2O3 and MgO precursors, respectively. The supported catalysts were prepared by a urea-matrix combustion method. Two solutions were prepared by dissolving appropriate amounts of ammonium heptamolybdate and cobalt nitrate (Sinopham Chemical, Shanghai) in water, keeping a constant molar ratio of [Co/(Co+Mo)] = 0.2. Urea was added into cobalt nitrate solution and kept a constant molar ratio of Urea/Co = 30. Then, ammonium heptamolybdate solution was added to the above mixture. Afterwards, a given amount of various supports, MgO, Al2O3 or MgO-MgAl2O4, was dipped into the above Co, Mo containing solution at 50 ºC for 3h. During the impregnation-deposition procedure, metal-oxo species were homogeneously dispersed throughout the inside and outside of the pores of the support. After that, the mixed slurry was dried at 120 ºC for 12 h. Finally, the solids were calcined at 450 ºC for 4 h under air at atmospheric conditions. The prepared catalysts were denominated as: CoMo/MgO, CoMo/Al2O3 and CoMo/MgO-MgAl2O4, respectively. 2.2 Catalyst characterization The surface properties of the synthesized supports and catalysts were measured by nitrogen adsorption-desorption isotherms at -196 ºC using an ASAP 2020 surface area and porosity analyzer.

Advanced Materials Research Vol. 132

47

The specific surface areas were calculated using the BET equation. The total pore volumes were evaluated from the nitrogen uptake at a relative N2 pressure around 0.99. The pore size distributions were obtained from the N2 desorption branch of isotherms applying the BJH pore model. Powder X-ray diffraction (XRD) technique was used to detect the phase composition and crystallite size of the calcined catalysts. The XRD patterns were obtained on a Rigaku D/max 2550 diffractometer, using Cu Kα radiation (λ=0.154056nm) over 2θ diffraction angles from 10° to 80° at a 0.02° step. Temperature programmed reduction (TPR) experiments were conducted on Micromeritics 2900 equipment. Prior to reduction, the catalysts (50 mg) were heated at a rate of 20 ºC/min to a final temperature of 400 ºC, and kept for 2 h at that temperature under a N2 flow to remove adsorbed water and other contaminants. Then the catalysts were cooled to ambient temperature in the same N2 flow, followed by reduction in a flowing H2/N2 mixed gas (10% vol. H2 in N2) at a total flow rate of 50 mL/min, and the temperature was raised at a rate of 15 ºC/min to a final temperature of 900 ºC. The TPR signals were recorded by using a thermal conductivity detector (TCD). 2.3 Catalyst test Catalytic tests of the catalysts were carried out on a fixed-bed reactor with a model compound of dibenzothiophene dissolved in xylene as feedstock. The catalyst in an oxidized state (1.0 g) was loaded in the bed with inert material and was pre-sulfided at 380 ºC and 3.0 MPa for 6 h, using a mixed solution of 3.0 % (w/w) CS2 in cyclohexane. Finally, the sample was collected periodically and analyzed by GC with a FPD detector. Specific reaction rate ( r ) was calculated according to the following expression: Vso ⋅ C0 ⋅ x w Vs0 is the flow velocity of the model compound, C0 is the initial concentration of dibenzothiophene, x is conversion of dibenzothiophene, which is stable after about 6 h of reaction, r=

and w is the loading amount of catalyst. 3. Results and Discussion 3.1 Characterization of the catalysts 3.1.1 N2 isothermal adsorption–desorption The loops of nitrogen adsorption-desorption isotherms of the oxidized catalysts are shown in Fig.1. The CoMo/Al2O3 exhibits a Type IV isotherm, which is indicative of a mesopore nature, while the CoMo/MgO catalyst exhibits a Type II isotherm, which suggests the appearance of macropores in this sample [13]. The CoMo/MgO-MgAl2O4 catalyst displays a distorted Type IV isotherm which indicates most macro-pores were changed to mesopores in the support when a little amount of Al2O3 was added. The specific surface areas (SBET) of the catalysts are in the order as follows: CoMo/Al2O3 (212 m2/g) > CoMo/MgO (197 m2/g) > CoMo/MgO-MgAl2O4 (163 m2/g). The low surface area of the CoMo/MgO-MgAl2O4 catalyst may indicate a different surface property and phase composition of the support.

48

Advances in New Catalytic Materials

360

CoMo/Al2O3 CoMo/MgO-MgAl2O4

320

CoMo/MgO

N2 Volume (cm3/g)

280 240 200 160 120 80 40 0 0.0

0.2

0.4

0.6

0.8

1.0

P/P0

Fig.1. N2 adsorption-desorption isotherm of oxidized catalysts

Pore Volume (cm3/g*nm)

0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 0

5

10 15 20 25 30

0

5

10 15 20 25 30 35

Pore diameter (nm) CoMo/Al2O3

CoMo/MgO-MgAl2O4

CoMo/MgO

Fig.2. Pore volume distributions of oxidized catalysts (left) and spent catalysts (right)

The pore diameter distributions of the synthesized oxidized catalysts are shown in Fig.2. The CoMo/MgO had a rather broad pore diameter distribution ranging between 10 and 30 nm and a rather narrow distribution at 2.5 nm and 3 nm. A great number of pores of the CoMo/Al2O3 catalyst are distributed within 3-6 nm. In the combined MgO-MgAl2O4 support, although the MgO/Al2O3 weight ratio is 4, the CoMo/MgO-MgAl2O4 exhibited a similar pore diameter distribution as that of CoMo/Al2O3 catalyst. Among the spent catalysts, the pore distribution of CoMo/Al2O3 did not change as compared with the one before reaction, indicating the high stability of the Al2O3 support. However, the pore diameters in CoMo/MgO-MgAl2O4 became slightly smaller after reaction. It is noted that the mesopores with a diameter of approximately 2.5 nm in CoMo/MgO almost

Advanced Materials Research Vol. 132

49

disappeared but the pores with diameters between 10 and 40 nm remained unchanged. This seems to indicate that Al2O3 has higher stability in comparison with MgO. 3.1.2 X-ray diffraction (XRD) The XRD patterns of the supports and catalysts are shown in Fig.3. For the MgO-Al2O3 mixed support, MgAl2O4 spinel phase was formed as indicated by the XRD peaks at 37.5º, 44.5 º and 64.7º. Because this sample was MgO-rich (weight ratio of MgO to Al2O3 is 4), MgO phase coexisted with the spinel phase (MgAl2O4) and all the alumina disappeared by reaction with MgO to form spinel. The pure MgO sample had very sharp XRD peaks, while pure alumina showed rather wide XRD peaks, indicating their different crystallite size.

∆

∆

∆

∆

Α

Intensity (a.u.)

Intensity (a.u.)

∆

∆ ∆

a

∆ ∆

B

∗

C

10

20

30

b

Ο

Ο

Ο

∗

40 50 2θ (°)

Ο

Ο

∆

Ο

Ο

c

∗

60



70

80

10

∗

20

30

∗

40 50 2θ (°)

∗

60

70

80

Fig.3.XRD patterns of the different supports (left) and catalysts (right) (*) Al2O3, (o) MgAl2O4, (∆) MgO, (□) MgMoO4 A: MgO; B: MgO-Al2O3; C: Al2O3 a: CoMo/MgO; b: CoMo/MgO-MgAl2O4; c: Co-Mo/Al2O3

No evident XRD signal corresponding to Co and Mo species was detected in the catalysts, indicating that these metal species are highly dispersed on the catalysts. There was a small peak of MgMoO4 appearing on the surface of the CoMo/MgO-MgAl2O4 catalyst, suggesting a strong interaction between the support and active component. An interesting fact observed, when comparing the diffractogram patterns of supports with the catalysts, is that the incorporation of molybdenum into the support induces a great loss in the crystallinity of the periclase phase, whose signals become broader and less intense. The intensity decrease of the diffraction lines of periclase phase may relate to molybdenum aqueous impregnation and to subsequent thermal treatment, as suggested by other work [14]. Moreover, we believe that the intense interaction between magnesia and active Mo or Co metals may be another important reason for this phenomenon. The diffraction intensity of spinel (MgAl2O4) does not change after impregnation. Therefore, it may act as a skeleton for the catalysts, which is in good agreement with the conclusion from the pore diameter distributions of the catalysts.

50

Advances in New Catalytic Materials

3.1.3 Temperature Programmed Reduction (TPR) Temperature programmed reduction profiles of the oxidized catalysts before catalytic test are shown in Fig. 4. In general, the reduction of MoO3 species occurs in two steps (MoO3→MoO2→Mo). Thus, the peaks around 470 ºC and 800 ºC observed in the TPR profile of the CoMo/Al2O3 catalyst can be assigned to the reduction of Mo6+ to Mo4+ and Mo4+ to Mo on Al2O3, respectively [12]. In the CoMo/MgO-MgAl2O4 catalyst, the peak around 470 ºC is absent, and instead, a broad peak between 500 and 700 ºC, partly overlapping with another peak at around 800 ºC, appeared. The appearance of the peak with a temperature maximum at 610 ºC should be related to the reduction of the Co- or Mo- oxo species formed on MgO phase in the MgO-MgAl2O4 support because this TPR peak is quite similar as the one in the CoMo/MgO catalyst. It was noteworthy that the TRP peak at 800 ºC in the CoMo/MgO-MgAl2O4 catalyst significantly differed from that shown on the CoMo/MgO catalyst where double peaks appeared or from that on the CoMo/Al2O3 catalyst where one very intense peak was located at 750 ºC.

CoMo/Al2O3

Intensity (a.u.)

CoMo/MgO-MgAl2O4 CoMo/MgO

300

400

500

600

700

800

900

o

Temperature ( C)

Fig.4 TPR patterns of CoMo supported catalysts Kumar et al. once observed that the corresponding reduction peaks of Mo-O species on MgO-Al2O3 are located at 600 ºC and 1000 ºC, respectively [15]. Lian and coworkers also found the formation of similar species with the Mo-O-Co bond on the CoMo/MgO catalyst, and these species would weaken the interaction between the active component and support, thus increasing the sulfidation degree during activation of the catalyst [16]. In our work, the first reduction peak (between 500 and 700 ºC) of the CoMo/MgO-MgAl2O4 catalyst is higher than that of CoMo/Al2O3 and similar with that of Kumar. This may be due to the formation of new Co-O-Mo species resulting from strong interaction between Co and Mo species [12]. But the second TPR peak showed that the reduction temperature shifts to a lower region than that reported by Kumar. There are two possibilities leading to TPR performance modification: the first is the formation of the Mo-oxo species on the MgAl2O4 spinel. As shown above, in this catalyst, spinel phase was separated from the MgO phase. After metal precursors were loaded on MgAl2O4 by impregnation-deposition reaction, new species like Mo- or Co- oxo- MgAl2O4 may be formed; its reduction behaviors should be different from that on MgO or Al2O3. The second possibility is the formation of the Co-O-Mo species due to interaction between Co and Mo species

Advanced Materials Research Vol. 132

51

[12]. Because the peak around at 600 ºC and 800 ºC is very broad, it is suggested that this reduction peak of Mo6+ and species with Mo-Co interaction may overlap together. The new support MgOMgAl2O4 strengthens the interaction between Mo and Co species by forming new Co-O-Mo species but weakens the interaction between the support and active metal as evidenced the TPR temperature shift. Therefore the reduction temperature of Mo4+ to Mo becomes lower than the 1000 ºC reported by Kumar et al. This new species with Mo-Co interaction may be related to the coordination of MgAl2O4 support in which the electronegativity of Mg ions is relatively lower with respect to Al3+ in Al2O3, which leads to a weak Mg-O bond and hence the oxygen coordination ability of the host support to Mo or Co guest metal is higher [15]. It is expected that these species could be sulfided to the Co-Mo-S active phase more easily in comparison with the CoMo/Al2O3 or CoMo/MgO catalyst. 3.2 HDS activities 3.2.1 Effect of catalyst support 10

DBT

-1

6

4

r×10 , mol*h *g

-1

8

4

2

0 CoMo/Al2O3

CoMo/MgO-MgAl2O4 CoMo/MgO

Catalysts

Fig.5 HDS activities of the prepared catalysts Operating condition: P=3.0MPa, T=300 °C, H2/HC=400, WHSV=10.0 h-1 The HDS reaction rates of different catalysts calculated from the DBT conversions that were obtained after 6 h of reaction are shown in Fig.5. The catalytic activity varied with the following order: CoMo/MgO-MgAl2O4 > CoMo/MgO > CoMo/Al2O3 It is well known that high surface area of the support can facilitate the dispersion of the active metals. In our work, the CoMo/MgO-MgAl2O4 catalyst exhibited the lowest surface area among the three catalysts. However, it showed the highest HDS activity. It is speculated that there are more active sites on the surface of the catalyst with basic MgO support and the surface area herein is not the key factor determining the catalytic activity.

52

Advances in New Catalytic Materials

3.2.2 Effect of reaction conditions DBT

10

r×104, mol*h-1*g-1

8

6

4

2

0 280

300

320 o

Temperature, C

A: Reaction condition: P=3.0 MPa, H2/HC=400, WHSV=10.0 h-1

10

DBT

r×104, mol*h-1*g-1

8

6

4

2

0 100

200

300

400

H2/HC

B: Reaction condition: P=3.0 MPa, T=300 ºC, WHSV=10.0 h-1

Fig.6. Effect of reaction conditions on the HDS activity 10

DBT

-1

6

4

r×10 ,mol*h *g

-1

8

4

2

0 2

3

4

P, MPa

C: Reaction condition: T=300 ºC, H2/HC=400, WHSV=10.0

Fig.6. Effect of reaction conditions on the HDS activity (continue)

Advanced Materials Research Vol. 132

53

Effects of the reaction temperature, pressure and the H2/HC ratio on the catalytic performance of the CoMo/MgO-MgAl2O4 catalyst in the HDS reaction of DBT were studied and the results are shown in Fig. 6. The activity of the catalyst monotonously increased with the increase of the reaction temperature (Fig.6 A). The same trend was observed for the reaction rate changing with the H2/HC ratio (Fig. 6 B). However, the pressure effect shows that there is an optimum reaction pressure around 3.0MPa for the HDS reaction on CoMo/MgO-MgAl2O4 (Fig.6 C). 4. Conclusions The CoMo/MgO-MgAl2O4 catalyst prepared by the urea-matrix combustion method exhibited an improved HDS activity at a moderate operating condition. The active metals on the supports were highly dispersed. Although the specific surface area of CoMo/MgO-MgAl2O4 was smaller in comparison to that of CoMo/MgO and CoMo/Al2O3, it showed the highest HDS activity probably due to an interaction between Mo and Co species that appeared in the former that may facilitate the formation of Co-Mo-S active phase. It was found that periclase phase is very sensitive to the formation of Co- or Mo-oxo species on it which might be used as a potential indicator for studying such kind active species in the HDS catalysts. Acknowledgement The authors would like to express their gratitude to the financial support from Luoyang Petrochemical Company of SINOPEC, China. We also thank Dr. Liping Ye from ECUST for her enthusiastic help in the experiment of N2 adsorption–desorption isotherm. Referentes [1] B. Pawelec, T. Halachev, A. Olivas, T.A. Zepeda: Appl. Catal., A. Vol. 348 (2008), p. 30. [2] M.S. Rana, E.M.R. Capitaine, C. Leyva, J. Ancheyta: Fuel. Vol. 86 (2007), p. 1254. [3] S.L. Gonzalez-Cortes, T.C. Xiao, P.M.F.J. Costa, B. Fontal, M.L.H. Green: Appl. Catal., A. Vol. 270 (2004), p. 209. [4] S.L. Gonzalez-Cortes, T.C. Xiao, T.W. Lin, M.L.H. Green: Appl. Catal. A. Vol. 302 (2006), p.264. [5] M.S. Rana, J. Ancheyta, S.K. Maity, P. Rayo: Catal. Today Vol. 130 (2008), p. 411. [6] S.K. Maity, M.S. Rana, B.N. Srinivas, S.K. Bej, G.M. Dhar, T.S.R.P. Rao: J. Mol. Catal. A: Chem. Vol. 153 (2000), p. 121. [7] M. Zdrazil: Catal. Today Vol. 86 (2003), p. 151. [8] W.Q. Huang, A.J. Duan, Z. Zhao, G.F. Wan, G.Y. Jiang, T. Dou, K.H. Chung, J. Liu: Catal. Today Vol. 131 (2008), p. 314. [9] S.K. Maity, M.S. Rana, S.K. Bej, J. Ancheyta, G.M. Dhar, T.S.R.P. Rao: Catal. Lett. Vol. 72 (2001), p. 115. [10] T. Klimova, D.S.Â. Casados, J.R. Ârez: Catal. Today Vol. 43 (1998), p. 135. [11] T. Klicpera, M. Zdrazil: J. Catal. Vol. 206 (2002), p. 314. [12] M.S. Rana, M.L. Huidobro, J. Ancheyta, M.T. Gomez: Catal. Today Vol. 107-108 (2005), p. 346. [13] D. Gulkova, O. Solcova, M. Zdrazil, Micropor. Mesopor. Mat. Vol. 76 (2004), p.137. [14] D. Solis, T. Klimova, J. Ramirez, T. Cortez: Catal. Today Vol. 98 (2004), p. 99.

54

Advances in New Catalytic Materials

[15] M. Kumar, F. Aberuagba, J.K. Gupta, K.S. Rawat, L.D. Sharma, G.M. Dhar, J. Mol. Catal. A: Chem. Vol. 213 (2004), p. 217. [16] Y.X. Lian, H.F. Wang, Q.X. Zheng, W.P. Fang, Y.Q. Yang, J. Nat. Gas Chem. Vol. 18 (2009), p. 161.

Advanced Materials Research Vol. 132 (2010) pp 55-60 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.55

ZnAlFe Mixed Oxides Obtained from LDH Type Materials as Basic Catalyst for the Gas Phase Acetone Condensation Angeles Mantilla1,a, Francisco Tzompantzi2,b, María Manríquez2,c, Guadalupe Mendoza2,d Jose L. Fernández1,e and Ricardo Gómez1,f* 1

CICATA-IPN, Av. Legaria 694, 11500, D.F. México

2

Universidad Autónoma Metropolitana-Iztapalapa, San Rafael Atlixco 186, 09340 D.F. México a

email: [email protected]; bemail: [email protected];

c

email: [email protected]; demail; [email protected]; e

email; [email protected]; femail:[email protected]

Keywords: LDH materials; solid basic catalyst; aldol condensation; acetone self condensation.

Abstract: ZnAlFe mixed oxides with high surface area were obtained by the calcination (723 K) of ZnAlFe layered double hydroxides (LDHs). The calcined materials proved as basic catalysts in the gas phase acetone condensation exhibited high activity and high selectivity towards the formation of mesityl oxide. 1. Introduction Nowadays, the interest in aldol condensation reactions and their products has been increased in the fine chemical industry. Aldol condensation reactions are currently carried out with homogeneous basic catalysts and their replacement by solid basic is a goal to reach. The use of solid basic catalysts offers the advantage of decreasing corrosion and environmental problems, allowing an easier separation and recovery of the catalysts. In this way, aldol condensation reactions have been successfully carried out using MgAl hydrotalcites [1-3], zeolites [4] and alkaline doped MgO and TiO2 catalysts [5,6]. However, the catalysts obtained by the calcination of layered double hydroxides LDHs (with general formula [M1-x2+Mx3+(OH)2]x+ . Ax/nn- mH2O]) have become an important alternative to obtain basic catalysts, since the composition of the mixed oxides and hence their basic properties can be tailored from the LDHs formulation [7-9]. With the purpose to develop new solid basic catalysts, in the present work, ZnAlFe LDHs materials were prepared and calcined to obtain ZnAlFe mixed oxides. In order to study their basic properties, the materials were evaluated in the acetone gas-phase self condensation. X-ray diffraction, nitrogen adsorption and Transmission Electron Microscopy (TEM) were used for the characterization of the mixed oxides. 2. Experimental 2.1 Catalysts preparation ZnAlFe layered double materials were prepared by the co-precipitation method by using aqueous solutions of Zn(NO3)2.6H2O:Al(NO3)3.9H2O and Fe(NO3)3.9H2O (J.K. Baker AR) in the appropriate amounts to obtain different ZnAlFe molar ratios. The solutions containing the precursors were added drop wise under vigorous stirring in a glass reactor vessel containing bidistilled water. Afterwards, the pH of the solutions was adjusted to 9 by adding NH2CONH2 as the precipitant agent. The resulting suspensions were vigorously stirred for 4 h at 363 K and maintained under reflux for 36 h. Then, the obtained materials were filtered and washed with deionized water. Before characterization, the solids were dried at 393K (dried samples) for 12 h and finally annealed at 723 K for 12 h in N2 flux (calcined samples). A reference ZnAl sample was prepared in the same way as described but without using iron precursor.

56

Advances in New Catalytic Materials

2.2 Characterization X-ray analyses were obtained with a Siemens D500 X-ray diffractometer using a graphite crystal as monochromator to select Cu-Kα radiation (1.5406 Ǻ) with a step of 0.02◦ s−1. Specific surface areas were determined by the BET method from the N2 adsorption isotherms obtained with a Quantachrome Autosorb-3B equipment. TEM micrographs were obtained with electron microscope Carl Zeiss-EM910 with 120kV and 0.6 nm of resolution. Before observation the samples were prepared using EPON 912 EMS. Cuts were made with an ultra-microtome RMC Mod MT 7000. 2.3 Activity test The aldol condensation reaction was carried out at 523K in a glass tubular reactor (3 ml) through acetone was feed using a saturator at 0°C and nitrogen as gas carrier; the mass of the catalyst used was of 50 mg and the nitrogen flow of 1.8 L/h. The products of the reaction (aldol, mesityl oxide (MO), isomesityl oxide (ISMO) and isophorone) were analyzed on line by gas chromatography. 3. Results and Discussion 3.1 Specific surface areas. The BET values obtained on the different samples are reported in Table 1. A diminution on the specific surface area was observed in the oxides with high Fe3+ content, it varies from138 m2/g to 70 m2/g for the ZAF-5.0 and ZAF-1.0 mixed oxides respectively. Table 1. Identification, chemical composition, molar ratio and BET areas for the LDHs. Composition

Name

Mol 2+

3+

Zn /Al +Fe

Mol 3+

3+

Al /Fe

BET 3+

2

m /g

Zn1.7Al1.0

ZA-3

1.7

-

44

Zn1.99Al2.04Fe 0.38

ZAF- 5.0

0.824

5.36

138

Zn1.27Al0.90Fe 0.66

ZAF- 1.5

0.814

1.36

97

Zn1.99Al0.802Fe 0.74

ZAF- 1.0

1.3

1.08

70

3.2 X-ray diffraction. The X-ray diffraction patterns of the fresh samples showed a layered double hydroxide type structure in all of them (Fig. 1). The peak at 11.75 in 2 θ angle, which corresponds to the 003 reflection of the interlayer distance in the carbonated solid, can be clearly noted.

Advanced Materials Research Vol. 132

57

* Zincite ** Hematite

1200 1100

ZA-3 ZA-3

1000 900 800 700 600 500

ZAF – 5.0

400 300 30 200

*

**

100

**

* ZAF – 5.0 (723 K)

**

*

*

*

0

ZAF – 1.5 ZAF – 1.5 (723 K) ZAF – 1.0 ZAF – 1.0 (723 K)

10

20

30

40

50

60

70

2 Theta ( degrees )

Figure1. XRD patterns for calcined (723K) and fresh (393K) LDHs The 003 reflection is typical of layered double hydroxide materials and its intensity is related to the crystallinity degree of the material. The cell parameters were calculated assuming an hexagonal packing and using the 003 and 110 reflection values in the equations c = 3d003 and a = 2d110, where c corresponds to three times the interlayer distance (003) and a is the average metal-metal distance in the interlayer structure (110). In Table 2, it can be seen that the cell parameters for the interlayer distance (d003) increases from 7.5600 to 7.5864 when the Fe content varies from 0.38 to 0.74 respectively. The variation in the cell parameter could be due to the different MII/MIII ratio in the materials; thus, the increasing in the positive charge produces a higher repulsion between the material layers. The variation on the parameter c can also be taken as a proof that the substitution of Al3+ by Fe3+ in the LDHs structure occurred during the synthesis. Thus, it can be assumed that the intercalation of Zn+2, Al3+ and Fe3+ was almost complete. In the calcined LDHs, the mixed oxides showed XRD patterns where the presence of hematite and zincite can be observed Fig.1 (723K samples) showing that the hydrotalcite type structure was destroyed. Table. 2. ZnAlFe LDH materials cell parameters. Hydrotalcites

d003

Parameter

Parameter

ZA-3

8.6345

c (Å) 25.903

a (Å) 3.549

ZAF-5.0

7.5600

22.685

3.548

ZAF-1.5

7.5673

22.702

3.549

ZAF-1.0

7.5864

22.759

3.549

58

Advances in New Catalytic Materials

3.3 Transmission electron microscopy On the selected samples ZA-3, ZAF-5.0 ZAF-1.0 the corresponding transmission electron micrographs are presented in Fig. 2, where the layered structure characteristic of LDH can be observed. LDHs with a particle size lower than 100 nm are obtained.

Figure 2. TEM micrographs form left to right for; ZA-3, ZAF-5.0 and ZAF-1.0 3.4 Acetone self condensation. The variation in the catalytic activity as function of time for the acetone gas-phase condensation using mixed oxides is presented in Fig. 3. Some deactivation, usually present in gas phase reactions, can be observed [5, 6]; hence, the activity and selectivity for the various catalysts were calculated after 180 min on stream (steady state) and they are reported in Table 3. -5

1.2x10

Rate (mol/g s)

ZAF-5.0 -6

8.0x10

ZA-3 ZAF-1.5

-6

4.0x10

ZAF-1.0

0.0

0

50

100

150

200

250

Time (min)

Figure. 3. Acetone aldol condensation as function of time for ZnAlFe mixed oxides The higher activity corresponds to the solid with the low Fe content (Al/Fe ratio 5.42). In Fig. 1, for the ZAF-5.0 calcined samples, broad peaks identifying the presence of zincite suggest the presence of ZnO nanosized particles. For the ZAF-1.5 and ZAF-1.0 samples the peaks assigned to zincite are not well defined and hence a low crystallinity of ZnO can be expected. The high activity showed by the ZAF-5.0 sample can related to the formation of high dispersed ZnO nanoparticles coexisting with Al2O3 and Fe2O3 oxides. It must be noted that the ZAF-5.0 catalyst showed the

Advanced Materials Research Vol. 132

59

highest specific surface area 138 m2/g (Table 1) and then in this solid a better dispersion of zincite, which can be considered the basic catalyst for the acetone aldolization reaction was inferred. In Table 3, it can be noted that the selectivity patterns are very similar for the three mixed oxides tested, with selectivity to mesityl oxide (MO) comprised between 75 and 77 mol %. The activity for the condensation reaction could be related to the number of basic sites exposed which will depend on the specific surface area of the mixed oxides, while the selectivity can be related to the electronic properties of ZnO. Therefore, any modification on the electronic properties of ZnO was provoked by the Al2O3 or Fe2O3 accompanying oxides. The reaction pathway for the acetone self condensation is represented in Fig. 4. The aldol molecule is stable at low temperature reaction and it is an important product. However, at high temperature (gas-phase reaction) the aldol is quickly dehydrated to form isomesityl oxide (ISMO) and mesityl oxide (MO); the addition of a third molecule of acetone to these molecules produces phorone and mesitylene, respectively. According to the selectivity patterns, we can see that mesityl oxide is the preferred product, i.e. the catalysts are high selective to the formation of the dimers, avoiding the successive addition of acetone and hence the formation of the corresponding trimers. Table 3. Activity (noted as reaction rate) and selectivity (Se%) for the acetone self condensation on basic ZnAlFe mixed oxides after 180 min on stream Sample

Se %

Se %

Se %

Se %

Rate

Aldol

Mesityl oxide

Isophorone

mol/g s 10

ZA-3

14

71

Isomesityl oxide 5

10

2.2

ZAF-5.0

14

77

7

2

4.7

ZAF-1.5

15

75

2

8

2.1

ZAF-1.0

16

76

5

3

2.0

Figure 4. Pathway for the acetone aldol condensation.

6

60

Advances in New Catalytic Materials

4 Conclusions It has been shown that basic ZnAlFe mixed oxides with high specific surface areas can be prepared by the thermal decomposition of the corresponding LDHs materials. It is proposed that for the gasphase acetone self condensation reaction, the catalytic activity depends on the specific surface area and then on the dispersion of ZnO. Acknowledgements The authors knowledge to SEP-CONACYT grant (2007-2010) and A.M. the support given by CONACYT for by the Scholarship. References 1. E. Susuki and Y. Ono: J. Mol. Catal. Vol. 61 (1988), p 283. 2. J.C.A.A. Roelofs, D.J. Lensveld, A.J. Van Dillen and K.P. de Jong: J. Catal. Vol. 203 (2001), p 184. 3. S. Abelló, T. Medina, D. Tichit, J. Pérez-Ramírez, X. Rodríguez, J.E. Sueiras, P. Slagre and Y. Cesteros: Appl.Catal. A: Gen Vol. 281 (2005), p. 191. 4. M.J. Climnet, A. Corma, S. Oborra and J. Primo: J. Catal. Vol.151(1995), p. 60. 5. J.I. Di Cosimo and C.R. Apesteguía: J. Mol. Catal: A Chem. Vol .130 (1998), p. 177. 6. M. Zamora, T. Lopez, M. Asomoza, R. Melendrez and R. Gómez: Catal. Today Vol. 116 (2006), p. 234. 7. F. Tzompantzi, J. Valente, M.Cantú and R. Gómez: in S.R. Schmidt (Ed) Catalysis in Organic Reactions (2006), Taylor & Francis Press, Vol. 55. 8. P.C. Noda, C.A. Pérez, C.A. Henriques and J.L.F. Monteiro: Appl. Catal. A: General Vol. 272 (2004), p. 229. 9. P. Kustrowski, D. Sulkowska, L. Chmielarz and J. Dziembaj R Appl. Catal. A. General: Vol. 302 (2006), p. 317.

Advanced Materials Research Vol. 132 (2010) pp 61-67 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.61

Structure Sensitivity of Sol-Gel Alkali Tantalates, ATaO3 (A= Li, Na and K): Acetone Gas Phase Condensation Leticia M. Torres-Martínez1,a, M. Elena Meza-de la Rosa1,b, Lorena L. Garza-Tovar1,c, Isaías Juárez-Ramírez1,d, Francisco Tzompantzi2,e, Gloria Del Angel2,f, Juan. M. Padilla2,g and Ricardo Gómez*2,h 1

Universidad Autónoma de Nuevo Leon, Instituto de Ingeniería Civil, Ecomateriales y Energía, Av Universidad y Av. Fidel Velásquez S/N, Cd. Universitaria, San Nicolás d e los Garza, N.L. 66540 México. 2

Universidad Autónoma Metropolitana-Iztapalapa, Depto. de Química, Av. San Rafael Atlixco No 186, México 09340, D.F. México

Emails: [email protected]; [email protected]; [email protected]; [email protected]; e [email protected]; [email protected]; [email protected]; [email protected]

Keywords: sol-gel alkali tantalates; aldol condensation; acetone condensation, base catalyst.

Abstract. Perovskite-type compounds such as alkali tantalates, ATaO3 (A = Li, Na and K), prepared by the sol-gel method are reported as heterogeneous basic catalysts for the acetone aldol condensation. It has been proposed that the activity order and the selectivity patterns (LiTaO3 > > KTaO3 > NaTaO3) depend on the octahedral arrangements of the TaO6 tantalates. Introduction The increasing demand of fine chemical intermediates has prompted a renewed interest in the study of novel catalysts for the acetone aldol condensation [1], and the development of alternative basic catalysts still is an interesting research subject. Aldol condensation products such as diacetone alcohol (DAA), mesityl oxide (MO) and isophorone (IP) are widely used as starting materials for the production of a large number of fine chemical compounds. The acetone condensation has usually been performed in liquid phase using ion exchangers [2, 3], hydrotalcites [4-6] and rare earth doped oxides as catalysts [7], however, for the reaction in gas phase the reports are considerably limited. For instance alkali doped magnesium [8, 9] and titanium oxides [10,11] have been used to perform such a reaction. In this work, perovskite-type compounds are proposed as basic catalytic materials for the condensation of acetone in gas phase. Perovskite-type compounds such as alkali niobates and tantalates, AMO3 (A = Li, Na and K; M = Nb and Ta) have been extensively studied as ferroelectric materials [12], and more recently, as photocatalytic materials to generate hydrogen by water splitting [13] and for the photodegradation of organic compounds [14]. In their structure, basic sites can be developed by the proper distortion of the TaO6 connections, which depend on the considered alkali metal [15, 16]. Tantalates have usually been prepared by the solid state reaction; however, this method requires high temperatures (~ 1200°C). In the present work, alkali tantalates, ATaO3 (A = Li, Na and K), were prepared by the sol-gel method, which allows the formation of their crystalline structure at lower temperatures. Experimental Lithium acetylacetonate (Aldrich Chemicals, 97.0%), sodium acetylacetonate monohydrate (Aldrich Chemicals, 97.0%), and potassium acetate (Aldrich chemicals 99.0%) were mixed in the appropriate amounts with tantalum (V) ethoxide (Aldrich Chemicals 99.98 %) in an ethanol-water

62

Advances in New Catalytic Materials

solution (4:1). Afterwards, the pH of the solution was adjusted to 3 with acetic acid, and the solution was maintained under stirring at 70°C for 5 days. Afterwards, the pH was adjusted to 9 using ammonium hydroxide. Then, the solution was kept under reflux until the gel was formed. The obtained gel was dried at 70°C and annealed at 600°C for 6 h. The specific surface area of the solids was determined from the nitrogen adsorption isotherms using Quantachrom equipment. The CO2-TPD was carried out after saturating the solids with an ArCO2 flow, with a Chembet 3000 and a programme rate of 10°C/min. The X-ray patterns were obtained with a diffractometer (Siemens, D-5000) with CuKα radiation. SEM analysis was carried out using a Scanning Electron Microscope LEICA model S440 coupled with an Energy Dispersion Spectroscopy system (EDS). Powder samples were placed in a stainless steel sample holder and then covered with a thin gold layer. Images were recorded using a magnification of 10,000X. The catalytic activity for the acetone condensation was evaluated by using a fixed bed reactor at comparable conversions (1.9 – 6.0 %) by varying the mass of the catalysts (20-100 mg). Prior to the catalytic test, the catalysts were reactivated at 400 °C for 1 h in nitrogen flow. After, the temperature was lowered to 250 °C, and using nitrogen flow the acetone (66 Torr) was fed into the reactor through a saturator to the reactant system (1.8 L/h). The reaction rate was followed as a function of time by analyzing the products with a FID-gas chromatograph coupled with the reactant system. Diacetone alcohol (DAA), mesityl oxide (MO) and isophorone (IP) were the only detected products (Fig. 1). O O

OH

O

-H 2O

O

C2H O 6

2 DAA

MO

IP

Fig. 1. Acetone aldol condensation pathway.

Results and Discussion The crystalline structure of the powders as a function of the annealing temperature was analyzed by XRD, and it is illustrated for the NaTaO3 sample in Fig. 2. Below 400°C, the sample was almost amorphous; as for the sample annealed at 600°C, the corresponding NaTaO3 orthorhombic crystalline structure (01-089-8061 ICDD) can be seen. For the KTaO3 and LiTaO3 samples, the XRD patterns show the same behaviour, i.e. well defined orthorhombic crystalline structures were obtained at 800 °C (Fig 3). As expected, the crystalline size of the solids increased with increasing the calcinations temperature. According to the XRD peak sharpness, at the same annealing temperature 800 ºC, the crystalline size of the different solids showed the following order: KTaO3 > NaTaO3 > LiTaO3.

Advanced Materials Research Vol. 132

63

Fig. 2. XRD patterns for NaTaO3 as function of the annealing temperature.

INTENSITY

T = 800°C, t = 6 h

KTaO3

T = 800°C, t = 6 h

NaTaO3

T = 800°C , t = 6 h

LiTaO3 10

20

30

40

50

60

70

80

90

2θ Fig. 3. XRD patterns for the alkali tantalates annealed at 800°C. SEM micrographs of the alkali tantalates annealed at 400°C show solids with a particle size lower than 1 micron in (Fig. 4). The particles observed by SEM are joined together forming clusters with a type-sponge microstructure as it is observed in Fig.4. However the size of the clusters is different for each one of the alkali tantalates. According to the SEM micrographs, the clusters size of the samples annealed at 400 ºC increases in the following order: KTaO3 > LiTaO3 > NaTaO3.

64

Advances in New Catalytic Materials

LiTaO3

KTaO3

NaTaO3

2 µm

2 µm

2 µm

Fig . 4. SEM micrographs for the alkali tantalates annealed at 400°C.

Rate(mol/g s)

The evolution of the acetone condensation as a function of time with the alkali tantalates is illustrated in Fig. 5. As for the gas phase, acetone aldol condensation deactivation phenomena have been frequently observed as an effect of oligomerization reactions [8, 9]. 1.0x10

-5

8.0x10

-6

6.0x10

-6

4.0x10

-6

2.0x10

-6

LiTaO3

KTaO3 NaTaO3

0.0 0

50

100 150 Time(min)

200

Fig. 5. Catalytic evolution as a function of time for the acetone aldol condensation on alkali tantalates. In some cases, the efficiency of the catalysts strongly depends on a moderate deactivation. In our case, the loss of activity for the Li, Na and K tantalates was around 30 %, and after 150 min in stream, the steady state was reached. Then, they showed a moderate evolution as a function of time. By taking into account the deactivation of the catalysts, the conversion, rate and selectivities reported in Tables 1 and 2 were calculated from the values obtained after 180 min in stream (Fig. 5). The activity and selectivity order to mesityl oxide was the following: LiTaO3 > > KTaO3 > NaTaO3. Table 1. Characterization and activity in the acetone aldol condensation for the alkali tantalates. Perovskite-type LiTaO3 NaTaO3 KTaO3 a

BET Areaa CO2b 16 26 16 15 5 41

C (%)c 6.0 1.9 2.4

Rated 7.3 0.67 0.80

Rate/CO2e 280 44 19

m2/g; bµmol/g; cconversion; dmol/g sx10-6 (180 min); e s-1 x10-3

Advanced Materials Research Vol. 132

65

Table 2. Selectivity (% mol) in the acetone aldol condensation. Perovskite-type LiTaO3 NaTaO3 KTaO3

DAA Mesityl Oxide 17 80 31 61 26 70

Isophorone 3 8 4

It has been reported that both the activity and selectivity in the acetone aldol condensation depend on the strength of the basic sites [1]. In Fig. 6, the CO2-TPD desorption peaks appear around 115-125 °C for the three studied tantalates, and then, we can consider similar basic site strengths of these samples. However, the activity per adsorbed CO2 molecule, reported in Table 1, shows that LiTaO3 is 6 or 14 times more active than both NaTaO3 and KTaO3, respectively.

500 KTaO3 LiTaO3 NaTaO3

Intensity (a.u.)

400 300 200 100 0 0

25

50

75

100 125 150 175 200

Temperature (°C) Fig. 6. CO2-TPD for the perovskite-type tantalates. The marked differences in activity per basic site determined by the number of desorbed CO2 molecules by the TPD-CO2 of catalysts cannot be justified by the differences in either the specific surface area or CO2 desorption temperature; then, both the activity and selectivity of the basic sites could reside in the structure of the tantalates. The basic sites in the tantalates can be developed upon the proper distortion of the TaO6 angles. The LiTaO3 structure shows the presence of TaO6 octahedral arrangements forming angles of 143°, whereas in the Na and K tantalates, the octahedral arrangements form angles of 163 and 180°, respectively [15, 16], (Fig. 7).

66

Advances in New Catalytic Materials

Fig. 7. Perovskite-type structures for: a) LiTaO3; b) NaTaO3 and c) KTaO3 with octahedral arrangement angles of 143, 163 and 180°, respectively The most distorted structure corresponds to LiTaO3, which is the tantalate with the highest activity and selectivity. As the structure becomes less distorted, as it can be seen in the Na and K tantalates, the activity in the acetone aldol condensation strongly diminishes.

Conclusions According to the aforementioned results, it can be proposed that the acetone condensation, when performed on sol-gel prepared perovskite-type alkali tantalates, is a reaction depending on the TaO6 tantalate octahedral arrangements. As far as we are concerned, it is the first time that perovskite-type tantalates were reported as structure sensitivity basic catalysts. These results open important future applications for the tantalates when used as basic catalysts. References [1] H. Hattori: Appl.Catal. A: General, Vol 222 (2001) p. 247. [2] N. Sisul, N. Cokovic, J. Jelencic and N. Wolf: React. Kinet. Catal. Lett., Vol 40 (1989) p. 227. [3] N. Sisul, N. Cokovic, N. Wolf and J. Jelencic: React. Kinet. Catal. Lett., Vol 47 (1992) p. 65. [4] P. Kustrwski, D. Sulkowska, L. Chmielarz, P. Olszewski, A. Rfalska-Lasocha and R. Dziembaj: React. Kinet Catal. Lett., Vol 85 (2005) p. 383. [5] P. Kustrowski, D. Sulkowska, L. Chmielarz, A. Rafalska-Lasocha, B. Dudek and R. Dziembaj: Micropor. Mesopor. Mat., Vol. 78 (2005) p. 11. [6] S. Abelloacute, D. Vijaya-Shankar and J.Peacuterez-Ramiacuterez: Appl. Catal. A: General, Vol. 342 (2008) p. 119. [7] A. P. Maciel, N. L. V. Carreño, P. R. de Lucena, E. R. Leite, E. Longo, H. V. Fajardo, A. Valentini and L.F.D. Porbst: React. Kinet. Catal. Lett. , Vol. 81 (2004) p. 211. [8] J.I. Di Cosimo, V.K. Diez and C.R. Apesteguia: Appl. Catal. A: General, Vol 137 (1997) p. 149. [9] J. I. Di Cosimo and C. R. Apesteguia: J. Mol. Catal. A: Chemical, Vol. 130 (1998) p. 177. [10] M. Zamora, T. Lopez, M. Asomoza, R. Melendrez and R. Gomez: Catal. Today, Vol 116 (2006) p. 234.

Advanced Materials Research Vol. 132

67

[11] M. Zamora, T. Lopez, R. Gomez, M. Asomoza and R. Melendrez: Catal. Today Vol. 107-108 (2005) p. 289. [12] H. Hungria, M. Alguero and A. Castro: Chem. Mater. Vol. 18 (2006) p. 5370. [13] C.C. Hu and H. Teng: Appl. Catal. A General, Vol 331 (2007) p. 44. [14] L. M. Torres- Martínez, L. L. Garza and E. M. López: J. Ceramic Transactions Series, Vol. 193 (2006) p. 197. [15] H. Kato and A. Kudo: J. Phys. Chem B Vol. 105 (2001) p. 4285. [16] M. Wiegel, M.H.J. Emond, E.R. Stobbe and G. Blasse: J. Phys. Solids, Vol. 55 (1994) p. 773.

Advanced Materials Research Vol. 132 (2010) pp 68-75 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.68

Phase Stabilization of Mesoporous Mn-Promoted ZrO2: Influence of the Precursor M. L. Hernandez-Pichardo1, *, J. A. Montoya2, P. Del Angel2, S. P. Paredes1 1

Instituto Politécnico Nacional, ESIQIE, Laboratorio de Catálisis y Materiales, Av. IPN S/N Zacatenco, 07738 México, México

2

Instituto Mexicano del Petróleo, DIyP, Eje Central Lázaro Cárdenas Norte 152, 07730 México D.F., México *

Keywords: Mn-Promoted Transformations.

[email protected]

Zirconia;

Mesoporous

Zirconia;

Precursor;

Surfactant;

Phase

Abstract. Mesoporous zirconia-Mn oxides were prepared by surfactant-assisted precipitation using different zirconia precursors and cetyl-trimethyl-ammonium bromide (CTAB) as a synthetic template. The objective of this work was to find out the influence of the zirconia precursors over the structural and textural characteristics of Mn-doped mesoporous zirconia solids. A series of syntheses were carried out by two methods using different zirconia precursors, modifying the Zr:surfactant ratio and the hydrolysis rate of the precipitate. After calcination at 500 °C, the samples were characterized by XRD, DTA, TEM and nitrogen adsorption-desorption isotherm. The use of the zirconium nitrate leads to materials having higher surface areas and narrow pore size distributions in the range of mesoporous materials; however, the preferential formation of the zirconia in the metaestable tetragonal phase was identified as the effect of the particle size allowed by the preparation method rather than the effect of the precursor. It was also found that the Mn and surfactant addition enhances the stabilization of the tetragonal crystalline phase and porosity. Introduction Zirconium oxide has found multiple applications in chemical and environmental chemistry, including its appliance as catalyst and support due to both, acidic and basic surface sites [1]. Moreover, the use of promoted zirconia is also of great interest for many applications in the field of new materials [2, 3]. In catalysis, it has been found that the addition of promoters to the zirconia improves the thermal stability and enhances the catalytic activity at low temperatures in the reactions of hydrocarbons conversions [4, 5]. For example, manganese has been used as a promoter of zirconia and the Mn doped materials have been employed in combustion of methane [6] and alkanes isomerization when it is sulfated [7] or tungstated [8]. These chemical processes require a high density of surface active sites on materials with high surface areas and controllable pore size distributions. Mesoporous zirconium oxides have been studied since these materials allow the improvement of mass transport, reactants adsorption and dispersion of active sites. The synthesis of mesoporous zirconia was first reported by Hudson and Knowles [9] who suggested that the formation of these materials is occurring via scaffolding rather than a templating mechanism. The main difference between these routes is that templated materials present ordered pore systems while scaffolding solids show a disordered pore arrangement [10]. The loose of the porosity seems to correlate to the monoclinic transformation, since sample calcination at high temperatures leads to the structural transformation to the monoclinic phase and the particles sintering, which results in the collapse of the mesopores. On the other hand, it has also been observed that the stability of the phase observed depends on the synthesis method and the size of the resulting particles. It has been reported that there exists a critical particle size in which zirconia can

Advanced Materials Research Vol. 132

69

no longer be stabilized in the tetragonal phase, since the small particles has a big contribution to the total free energy [2]. Despite zirconia are intensely studied, there are few works dedicated to the study of the influence of the precursor over the structure and textural characteristic of ZrO2 solids. In addition, it has not yet investigated the effect of the addition of a promoter such as manganese to a mesoporous system. In this work, the influence of some parameters, for instance, the precursor and the promoter, over the texture of zirconia-manganese oxides using surfactant-assisted synthesis was investigated, focusing on structural and textural characterization. It has been found that the structure and porosity of these ZrO2-Mn oxides can be tailored by means of the control of some parameters such as the kind of precursor and the zirconium hydrolysis rate. Experimental Mesoporous oxides were prepared from two different zirconium precursors, zirconium nitrate ZrO(NO3)2 (Aldrich, 99.9 %) and zirconium chloride, ZrOCl2⋅8H2O (Aldrich, 98 %). In all cases the surfactant was cetyl-trimethyl-ammonium bromide (CTAB, Aldrich, 99.9 %) in aqueous solution (25 wt %). The Mn precursor was manganese nitrate (Aldrich, 99.9 %) and the Mn loading was 1 wt.% The various preparations differed in the Zr:surfactant molar of 5 or 10. The precipitates were prepared by two different methods: Method A. An aqueous solution of CTAB was added to an aqueous solution of zirconium and manganese, this solution was then stirred for 1h. Then NH4OH was added until the pH reached 11 (rapid precipitation). Method B. An aqueous solution of CTAB was added to an aqueous solution of zirconium and manganese, in this method urea was used which releases the precipitating agent slowly, and this solution was stirred under reflux at 90 °C for 20 h in order to modify the hydrolysis rate of the precipitate. After that time NH4OH was dropped into the solution until the pH reached 11 in order to complete the precipitation (controlled precipitation). All the precipitates were aged for 24 h. After precipitation, the precipitates were washed repeatedly and then dried at 80 °C for 24 h. The material calcination was performed in static air at 500 °C for 4 h. The samples were labeled as ZMx-yz, with x = precursor (N:nitrate, Cl:chloriode), y = the Zr:surfactant molar ratio (5 or 10) and z = the synthesis method (A or B). Pure zirconium oxides, surfactant- and Mn-free samples, were also prepared by method A, these samples were labeled as ZN and ZCl. After calcination, the samples were characterized by differential thermal analysis (DTA), X-ray powder diffraction (XRD), nitrogen adsorption-desorption isotherm and transmission electron microscopy (TEM). The differential thermal analyses (DTA) were performed by using a DTA-7 Perkin Elmer. Samples were heating at a rate of 10 °C/min from 0 to 1000 °C. The zirconium oxides were studied by XRD by means of a Siemens D-500 system, using Ni-filtered Cu-Kα radiation. The nitrogen adsorption isotherm and BET surface areas were measured at 77 K with the ASAP 2405 system. Prior to the measurements the samples were out-gassed at 350 °C for 15 h in vacuum. The pore size distribution was calculated from the N2 desorption isotherm by the BJH method. A JEOL 100CX TEM was used to examine the morphology of the solids using a voltage of 100 kV. Results and Discussion Nitrogen Adsorption-Desorption Isotherm The characterization of the texture was done by nitrogen adsorption-desorption isotherm. Table 1 shows the influence of the two precursors and synthesis methods on the textural parameters of these materials such as surface area, pore volume and pore diameter. The surfactant addition allows the improving of the surface area and the pore volume as it is compared with the data for surfactant free samples (ZN and ZCl). The data show that these oxides present a direct relation between the surface

70

Advances in New Catalytic Materials

area and the Zr:surfactant ratio, this dependence shows that the surface area decreases with a lower surfactant ratio. As far as the effect of the method B is concerned, the using of slow precipitation, yields solids with high average pore diameter and volumes of about 0.21 cm3g-1, but the surface areas obtained were lower than those obtained by the method A. From the data of the textural properties of samples synthesized by method B we can see that when reflux is used, the textural characteristics of both samples are very similar, independently of the precursor used. Table 1. Properties of Mn-doped ZrO2 oxides synthesized by different precursors.

Sample

ZrO2 precursor Zr:Surfactant Sg-BET (m2g-1)

Vpa (ccg-1)

Dpa (A)

ZN ZMN-5A

Nitrate Nitrate

-1:5

64 104

0.09 0.20

39 50

ZMN-10A

Nitrate

1:10

118

0.18

36

ZMN-10B

Nitrate

1:10

80

0.21

71

ZCl

Chloride

--

60

0.11

51

ZMCl-5A

Chloride

1:5

52

0.12

57

ZMCl-10A

Chloride

1:10

72

0.20

71

ZMCl-10B

Chloride

1:10

82

0.22

72

a

BJH desorption, average value.

Figure 1 shows the nitrogen adsorption-desorption isotherms at 77 K for samples synthesized from zirconium nitrate (Fig. 1a) and oxychloride (Fig. 1b) by method A. These materials present hysteresis loop characteristic of the type IV corresponding to mesoporous solids; however, the loop shape is different for the samples using different precursor. It is clear that in the case of the samples synthesized from chloride, the loop is sharp, indicating a wide range of mesopores; when nitrate is used, the loop presents a flat plateau at about p/p0 = 0.9, attributed to N2 filling in the pores, suggesting a narrow pore size distribution (PSD). This result is confirmed by the pore size distribution plots, the use of nitrate leads to materials with narrow monomodal pore size distributions, whereas, the samples made from chloride show wider dispersions of the pore size, but still in the mesoporous range (20-100 A). On the other hand, Figure 2 shows, as an example, the N2 adsorption-desorption isotherm of the ZMCl-10B sample; from the analysis of the textural properties and the isotherms of samples prepared by method B, compared with samples prepared by method A, it is possible to observe that in samples prepared with chloride by the method B present a slight increase in surface area, however these samples show a wider pore size distribution. Moreover, the use of method B represents a decrease in the surface area value and also a wide PSD is obtained. The nitrogen adsorptiondesorption results show that the characteristics of the final oxide were heavily dependent on the synthesis conditions.

Advanced Materials Research Vol. 132

71

120 1.8 1.6

PORE VOLUME (cc/g)

1.4

VOL ADSORBED , (cc/g)

100

1.2 1.0 0.8 0.6 0.4

80

0.2 0.0 10

100 PORE DIAMETER (A)

60

40

a) 20 0.0

0.2

0.4

0.6

0.8

1.0

RELATIVE PRESSURE , (P/Po)

140 0.8

0.6

VOL ADSORBED , (cc/g)

PORE VO LUME

(cc/g)

120

100

0.4

0.2

0.0 10

100 PORE DIAMETER (A)

80

60

40

20 0.0

0.2

0.4

0.6

0.8

1.0

RELATIVE PRESSURE , (P/Po)

Fig. 1 Pore size distributions and adsorption-desorption N2 isotherms at 77K of ZrO2-Mn samples prepared from: a) zirconium nitrate (ZMN-10A) and b) zirconium oxychloride (ZMCl-10A) prepared by method A. Differential Thermal Analysis (DTA) Figure 3 shows the typical DTA curve obtained from the uncalcined materials. It is possible to identify several decomposition stages of the zirconia-Mn solids; the first one is the removal of water as indicated by the endothermic peak at 93 °C. The degradation of the precursors begins at about 220 °C and there are two regions between 260-400 °C which correspond to the dehydroxylation and the oxidation of the surfactant. Finally a small exothermic peak can be observed at about 490 °C attributed to the crystallization of the zirconia [11].

72

Advances in New Catalytic Materials

160 0.8 0.7 0.6 PORE VOLUME (cc/ g )

VOL ADSORBED , (cc/g)

140 120

0.5 0.4 0.3 0.2

100

0.1 0.0 10

1 00 PORE DIAMETER(A)

80 60 40 20 0.0

0.2

0.4

0.6

0.8

1.0

RELATIVE PRESSURE , ( P/Po)

Fig. 2 Pore size distributions and N2 adsorption-desorption isotherm of ZMCl-10B sample.

320

ZMCl

∆Τ °C

250 380

490

90 0

200

400

600

800

1000

Temperature (°C)

Fig. 3 Typical DTA curve obtained from zirconia-Mn uncalcined samples.

X-Ray Powder Diffraction Figures 4 and 5 show the XRD patterns of the samples synthesized from zirconium nitrate (Fig. 4) and zirconium oxychloride (Fig. 5). The patterns of pure zirconia (ZN and ZCl) show that for the ZN sample the peaks correspond to the preferential formation of the tetragonal crystalline phase with only traces of the monoclinic form, whereas the sample ZCl reveals the formation of a mixture of tetragonal and monoclinic phases. For the Mn-doped samples synthesized by method A, the XRD patterns show that the manganese is well dispersed in the structure of the zirconia and the combination with the surfactant produces wide peaks associated with small particles. In the formation of the crystalline zirconia, these samples show a behavior similar to the pure samples: the tetragonal phase is preferentially formed for the samples synthesized by using zirconium nitrate. This effect is more obvious than that resulted from the surfactant content augments.

Advanced Materials Research Vol. 132

73

T

Intensity (a.u.)

T

T M ZMN -10 B ZM N-10A ZM N-5A ZN

25

30

35

40

45

50

55

60

2θ

Fig. 4 XRD patterns of ZrO2-Mn calcined at 500 °C synthesized from zirconium nitrate (T:tetragonal, M:monoclinic zirconia) with different methods and surfactant contents. T

Intensity (a.u.)

T

M

T

M ZMCl-10B ZMCl-10A ZMCl-5A ZCl

25

30

35

40

45

50

55

60

2θ Fig. 5 XRD patterns of ZrO2-Mn calcined at 500 °C synthesized from zirconium oxychloride (T:tetragonal, M:monoclinic zirconia) with different methods and surfactant contents.

Figures 4 and 5 also show the diffractograms corresponding to samples prepared by the method B, these patterns indicate that when a controlled hydrolysis is used, the tetragonal phase is the main form in both samples, independently of the precursor used. It has been observed that the use of surfactant leads to the zirconia formation through a scaffolding mechanism [9], giving a partially ordered materials; however, in our case the transmission electron micrograph (Fig. 6) shows that the three-dimensional pores are disordered. No any peaks are observed in the low angle XRD patterns (not shown). These results confirmed that the mesoporous zirconia-Mn materials have short-range order, suggesting that the porosity arises from the spaces among particles.

74

Advances in New Catalytic Materials

Fig. 6 Typical transmission electron micrograph of mesoporous zirconia-Mn samples. The structural analysis results indicate that there is dependence between the precursor used and the crystalline phase in these Mn-ZrO2 materials. It was observed that when the nitrate precursor was used, zirconium oxides with preferential formation of the tetragonal structure are formed, even when the sample is free of manganese and surfactant, whereas the use of zirconium chloride does not avoid the transformation from the metaestable tetragonal phase to the monoclinic form, and it leads to the formation of a mixture of both phases. However, when a slow precipitation is used (method B), the tetragonal phase is stabilized in all samples, independently of the precursor used. These results suggest that under the conditions used in the method A, the initial hydrolysis of the zirconium oxychloride is rapid, allowing a sintering process which permits the particles to reach a critical diameter in which the zirconia can no longer be stabilized in the tetragonal phase. Thus the transformation from tetragonal to monoclinic phase occurs at the same temperature, differing from that in the samples prepared from the zirconium nitrate. On the other hand, if the hydrolysis process is carried out slowly, more time is necessary to reach the equilibrium and the interaction between the hydrous zirconia-Mn and the surfactant avoids the particles growth, retarding the transformation of the tetragonal to the monoclinic phase. This observation is consistent with the wider peaks associated with small particles obtained by the method B (Fig. 5 and 6). So, it seems that the stabilization of the tetragonal zirconia depends on an effect of the precursor allowed by the preparation method, rather on the formation of small particles. Finally, it was also observed that the use of the surfactant in the preparation leads to the formation of materials with higher superficial areas, but the extent of increment depends on the preparation method and the precursor used (Table 1). In the case of the materials synthesized from zirconium nitrate, the formation of higher superficial areas seems to be favored by the use of the method A. We suggest that the synthesis conditions used in method A permit the surfactant to envelop the hydrous zirconia, preventing the particles growth before the formation of the oxides. Conclusions This work shows that the textural and structural features of the Mn-doped zirconia oxides depend largely on the synthesis method. The control of the synthesis parameters, such as the precursor, hydrolysis rate and the Zr:surfactant ratio, allows the formation of mesoporous materials with monomodal pore size distributions in the range of mesoporous solids. The stabilization in the tetragonal phase was favored by the addition of Mn to the synthetic system. The preferential

Advanced Materials Research Vol. 132

75

formation of the metaestable tetragonal form instead of the monoclinic phase was considered as a particle size effect rather than a precursor consequence. These synthesis methods allow the formation of mesoporous zirconia owning disordered structures as confirmed by TEM and the low angle XRD analyses, suggesting that the porosity arises from the spaces among particles. Acknowledgements Authors would like to acknowledge the financial support from the Instituto Politécnico Nacional for the financial support for this study. References [1] T. Yamaguchi: Catal.Today Vol. 20 (1994), p. 199. [2] B. L. Kirsch, S. H. Tolbert: Adv. Funct. Mater. Vol. 13 (2003), p. 281. [3] M. Garcia-Hipolito, O. Alvarez-Fregoso, E. Martínez, C. Falcony, M.A. Aguilar-Frutis: Opt. Mater. Vol. 20 (2002), p. 113. [4] H. Ohtsuka: Catal. Lett. Vol. 90 (2003), p. 213. [5] M. Valigi, D. Gazzoli, R. Dragone, A. Marucci, G. Mattei: J. Mater. Chem. Vol. 6 (1996), p. 403. [6] V. R. Choudhary, B. Uphade, S. G. Pataskar: Appl. Catal. A Vol. 227 (2002), p. 29. [7] X. Song, K. R. Reddy, A. Sayari: J. Catal. Vol. 161 (1996), p. 206. [8] M. L. Hernández-Pichardo, J. A. Montoya, P. del Angel, A. Vargas, J. Navarrete: Appl. Catal. A Vol. 345 (2008), p. 233. [9] M. J. Hudson, J. A. Knowles: J. Mater. Chem. Vol. 6 (1996), p. 89. [10] U. Ciesla, M. Fröba, G. Stucky, F. Schüth: Chem. Mater. Vol. 11 (1999), p. 227. [11] M. Valigi, D. Gazzoli, R. Dragone, A. Marucci, G. Mattei : J. Mater. Chem. Vol. 6 (1996), p. 403.

Advanced Materials Research Vol. 132 (2010) pp 76-86 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.76

Promotional Effect of Gadolinia on CuO Catalyst for Reduction of NO by Activated Carbon Yuye Xue, Guanzhong Lu*, Yun Guo, Yanglong Guo, Yanqin Wang, Zhigang Zhang Key Laboratory for Advanced Materials and Research Institute of Industrial Catalysis, East China University of Science and Technology, Shanghai 200237, P. R. China *Corresponding author. Tel: +86-21-64252923; Fax: +86-21-64253703; *E-mail address: [email protected] (G.Z. LU) Keywords: Gadolinia promoter; CuO catalyst; Catalytic reduction; Nitric oxides; Activated carbon.

Abstract: The Gd2O3 (gadolinia) modified CuO/AC catalysts for NO reduction by activated carbon were prepared and characterized by XRD, TPD-MS, EPR, XPS techniques. The results show that adding a small amount of Gd2O3 in the CuO catalyst can improve effectively its catalytic performance for NO reduction by activated carbon, and the appropriate molar ratio of Gd2O3/CuO is 0.03:1. The promotional effect of Gd2O3 stems from the cooperative effects between CuO and Gd2O3. The presence of Gd2O3 in the catalyst can alter the chemical state and environment of the CuO active sites and improve the catalytic activation of carbon by CuO to form more carbon reactive sites, resulting in the quicker transfer and release of oxygen decomposed from NO. The carboxylic groups on the surface of activated carbon play an important role in the catalytic reduction of NO by carbon at temperature below 300 °C. 1. Introduction The selective catalytic reduction (SCR) of nitrogen oxide is one of the very important methods of eliminating NOx pollution, in which a limited amount of reducing agent selectively react with NOx over a catalyst. Based on different reducing agents, the SCR technologies include urea- or ammonia-SCR, hydrocarbon-SCR, and plasma–assisted SCR and so on [1, 2]. However, the ammonia/urea-SCR requires the development of an ammonia/urea distribution network, and may result in additional sociability (NH3 slip and odor). Hydrocarbon-SCR must be operated at high temperature, and usually the catalysts in this process have a low stability especially in the presence of water, and they are easily poisoned by sulphur dioxide. Plasma-assisted SCR is just studied in laboratory and it is very difficult to make further application. If activated carbon is used as the reducing agent, instead of ammonia (and/or urea), to reduce NOx, the re-pollution of unreacted ammonia can be avoided and the cost of SCR of nitric oxide would be decreased obviously, which is the most ideal method to remove NOx, but the high effective catalysts must be designed and used. Activated carbon is a very common and effective catalyst support as well as a reducing agent. Carbon-based catalysts can be used in the reduction of NOx at relatively low temperatures [3-10]. Among the carbon-based catalysts, activated carbon impregnated with Cu(NO3)2 has been proved to be very active for the catalytic reduction of NO with CO [11] or NH3 [12] as the reducing agent. The Cu–C complexes were suggested to be responsible for the high activity in the reduction of NO with NH3 [12]. Besides, numerous studies have shown that the reactive activity of carbon can be

Advanced Materials Research Vol. 132

77

enhanced through the introduction of acidic surface oxygen functional groups that would serve as the promoter for the dispersion of active sites. Although some catalysts with a high activity could be achieved in the purification of NOx, there is no single-phase catalyst capable of satisfying the practical demand for the NOX removal [13]. Recently, some publications have reported that the catalytic performance of catalyst for the NO reduction can be enhanced by combination of two (or more than two) catalytic species in catalysts. For example, Long and Yang [14] reported that, adding a small amount of CeO2 and/or PrO1.83 to Fe-TiO2-PILC catalyst could increase its catalytic activity for the SCR of NO with NH3 (by 35%) in the absence and presence of H2O + SO2. In2O3 or SnO2-doped Ga2O3-Al2O3 catalysts show higher activities than the Ga2O3-Al2O3 catalyst for the selective reduction of NO with propylene in the presence of H2O; in the absence of H2O, CoO, CuO or Ag show good additive effects for the Ga2O3-Al2O3 catalyst [15]. The presence of Zn can improve effectively the selectivity of methane towards NOx reduction over the Co/HZSM-5 catalyst, which was stem from the cooperative effect between Zn, Co and zeolite [16]. The promotional effects of the rare earth metal oxides on the catalysts were studied for the reduction of NOx by methane or CO [17-22]. Modification of catalysts with the rare earth elements can increase their activity and selectivity for SCR of NOx and the catalyst durability. We have developed the catalytic process of the NO reduction by AC over CuO [23], and the influence of pretreatment method of activated carbon on the catalytic reduction of NO by AC was studied. As the rare earth elements have some special redox properties and activated carbon has numerous surface oxygen groups, it is meaningful to study the influence of the rare earth elements on the activity of carbon-based CuO catalyst for the reduction of NO by activated carbon. In this paper, the promotional effect of Gd2O3 on the carbon-supported CuO catalyst for the reduction of NO by activated carbon was studied, in order to design and prepare the high effective catalyst for the NO reduction by activated carbon at lower temperature. 2. Experimental 2.1. Preparation of catalyst Activated carbon (AC) made from a coconut shell was used as the reducing agent and the support of CuO. The textural properties of AC are, total pore volume 0.336 cm3/g, BET surface area 928 m2/g, bulk density 399 g/dm3, ash content 1.79 % (w/w). Before being used, AC was grinded into 20~40 mesh. The Gd2O3-CuO/AC sample was prepared at room temperature by immersing AC in the aqueous solution of Cu(NO3)2•3H2O and Gd(NO3)3•6H2O, in which CuO loading was controlled to 20 wt % and the molar ratio of Gd2O3/CuO was 0.025~0.5/1. After being impregnated for 4h, the sample was dried at 60 °C for 2h and then at 110°C overnight, and finally calcined in air at 250°C for 4h. The loadings of metal oxides were determined by burning off the carbonaceous support. In mGd2O3-CuO/AC, where m is the molar ratio of Gd2O3/CuO, such as, in the 0.1Gd2O3-CuO/AC catalyst, Gd2O3/CuO = 0.1 (mol). 2.2. Characterization of catalyst The N2 adsorption-desorption isotherms and BET surface areas of samples were measured at -196 °C on a Micromeritics ASAP 2020 Sorptometer using static adsorption procedures. The X-ray diffraction (XRD) patterns of samples were obtained on a Rigaku D/max-II/2550/VC X-ray powder diffractometer with CuKα radiation. The electron paramagnetic resonance (EPR) spectrum of sample was recorded on a Bruker ER 200D spectrometer operating in the X-band frequency (E9.205 GHz) with a field modulation frequency of 100 kHz. 25 mg sample was placed inside a quartz probe cell for the EPR measurement. Polycrystalline DPPH with g=2.0036 was used as a standard field marker.

78

Advances in New Catalytic Materials

The XPS measurements were performed on a thermo ESCALAB 250 spectrometer with Al Kα radiation (1486.6 eV) at 1×10-9 to 2×10-10 torr. The binding energies were determined with respect to the C1s line (284.6 eV) originating from adventitious carbon. Temperature-programmed desorption (TPD-MS) of sample was carried out with a quartz reactor (Φ10 mm × 300 mm) coupled to a quadrupole mass spectrometer (IPC400, INFICON Co. Ltd.). 200 mg sample was used, the flow rate of Ar carrier gas was 100 ml/min, and the heating rate was 20 °C/min from 100~900 °C. 2.3. Catalytic reduction of NO by carbon The catalytic reduction of NO by AC was carried out at atmospheric pressure in a fixed bed microreactor (Φ10 mm × 300 mm). The outlet gases were analyzed by two gas chromatographs (Fuli 9790) with two thermal conductivity detectors two separation columns, a column of Poropak Q for the separation of CO2, N2O, H2O, and a column of 5A zeolite for the separation of O2, N2 and CO. 500 mg catalyst was used. The space velocity (GHSV) was 20,000 h-1, and the concentration of NO in Ar was 2000 ppm. The catalytic reduction of NO was studied by a temperature programmed procedure at a heating rate of 3 °C/min from 100 to 400 °C. The conversion of NO was calculated by the amount of N2 produced. 3. Results and Discussion 3.1. Effect of Gd2O3 additive on the catalytic performance of CuO/AC The feed gas consisted of 2000 ppm NO and Ar as a balance gas. AC was used as the reducing agent of NO and the support of Gd2O3-CuO catalyst. In the NO reduction by AC, N2 is the aim and main product, and N2O has also been formed but its concentration was is too low to be calculated. Therefore, the conversion of NO was calculated according to the amount of nitrogen produced.

100

0.02Gd2O3-CuO/AC

NO Conversion /%

0.03Gd2O3-CuO/AC 80

0.05Gd2O3-CuO/AC 0.1Gd2O3-CuO/AC

60

0.5Gd2O3-CuO/AC CuO/AC

40

20

0 100

150

200

250

300

350

o

Temperature / C

Fig. 1. Effect of Gd2O3 loading on the catalytic activity of Gd2O3-CuO/AC for NO reduction by AC.

Fig.1 shows the influence of Gd2O3 loading on the catalytic activity of CuO/AC for the reduction of NO by AC at different temperatures. The results show that, adding Gd2O3 in CuO/AC can vary its catalytic performance. When the molar ratio of Gd2O3/CuO is more than 0.05 (such as 0.1 or

Advanced Materials Research Vol. 132

79

0.5), its activities decrease in some sort, especially at reaction temperatures below 280 °C. When the molar ratio of Gd2O3/CuO is < 0.05, with an increase in the molar ratio of Gd2O3/CuO, the performance of catalyst for the reduction of NO increases, such as, the temperature of the complete reduction of NO decreases to 285 °C, which is about 50 °C lower than that over CuO/AC. We have observed that single Gd2O3 supported on AC exhibits very low activity for the NO reduction by AC, that is to say, Gd2O3 is inactive for the reduction of NO by AC, and Gd2O3 acts only as a promoter or positive additive of CuO/AC. The phenomenon that a decrease in the activity of CuO/AC catalyst with an increase in Gd2O3 loading may be due to the coverage of Gd species on Cu sites, resulting in the decreases of the surface area of CuO and the active CuO sites. Therefore, the loading of Gd2O3 should be controlled to an appropriate amount. For the CuO/AC catalyst used in the reduction of NO by AC, the appropriate molar ratio of Gd2O3/CuO should be 0.03. 1600 1400

CuO/AC

1200

Concentration /ppm

1000 800 600 400 200 0 1600 1400

0.03Gd2O3-CuO/AC

1200 1000 800 600 400 200 0

100

150

200

250

300

350

o

Temperature / C Fig. 2. Products concentrations vs. the reaction temperature in the reduction of NO by AC over 0.03Gd2O3-CuO/AC and CuO/AC. (● — N2; ○ — CO2)

Fig. 2 shows the concentration variations of the products (N2 and CO2) in the reduction of NO by AC over 0.03Gd2O3-CuO/AC. No CO or N2O was detected in the range of reaction temperature. For the reaction of 2 NO + C → N2 + CO2, the mole number of N2 produced should be equal to that of CO2 formed in the products if no oxygen is added or adsorbed on the catalyst surface. The results in Fig.2 show that, increasing the reaction temperature, the evolutions of N2 and CO2 increase obviously, however the concentration of CO2 is always higher than that of N2. This means that except for NO, the surface oxygen-containing groups also react with AC, which can be seen from the TPD experiments. The results in Fig.2 also show that, the CO2 concentration on 0.03Gd2O3-CuO/AC is higher than that of CO2 on CuO/AC, which shows that the presence of Gd2O3 can improve the gasification of carbon and the catalytic performance of CuO for the oxidation of AC, resulting in an improvement of the catalytic reduction of NO by AC. 3.2. Effect of Gd2O3 additive on the physicochemical properties of CuO/AC The BET surface areas of the Gd2O3-CuO/AC catalysts are summarized in Table 1. It can be seen that CuO/AC has the largest surface area among all samples. The addition of Gd2O3 leads to an evident decrease of the catalyst surface area: the more the Gd2O3 loading, the more the decrease of

80

Advances in New Catalytic Materials

the surface area is, because a high loading of Gd2O3 is able to block more pores of AC, which results in a decrease of the catalyst performance. Table 1. BET surface areas of the Gd2O3-CuO/AC catalysts. Gd2O3/CuO (molar ratio)

0.5

0.1

0.05

0.03

0.02

0

BET Surface area /m2g−1

485

591

707

730

778

836

The XRD patterns of CuO/AC and Gd2O3 modified CuO/AC catalysts are shown in Fig. 3. The crystallite sizes of CuO, Gd2O3, and CuGd2O4 on the catalysts determined by Scherrer equation are summarized in Table 2.

Intensity /a.u.

f e d c b a 10

20

30

40

50

60

70

80

2-Theta /degree Fig. 3. XRD patterns of Gd2O3-CuO/AC catalysts with the Gd2O3/CuO ratio of (a) 0.5, (b) 0.1, (c) 0.05, (d) 0.03, (e) 0.02 and (f) 0. (● — CuO; ■— Gd2O3; ▲— CuGd2O4). Table 2. Crystallite sizes of CuO, Gd2O3 and CuGd2O4 determined by Scherrer equation. Crystallite size /nm Catalyst CuO

Gd2O3

CuGd2O4

0.5Gd2O3-CuO/AC

12.4

20.6

18.5

0.1Gd2O3-CuO/AC

12.8

8.6

12.9

0.05Gd2O3-CuO/AC

13.0

-

10.7

0.03Gd2O3-CuO/AC

13.0

-

-

0.02Gd2O3-CuO/AC

12.8

-

-

CuO/AC

12.5

-

-

The results show that all the catalysts exhibit the similar diffraction peaks of AC, and the diffraction peaks of CuO decreases gradually with an increase in the Gd2O3 loading. The diffraction peaks of Gd2O3 are not observed when the molar ratio of Gd2O3/CuO is less than 0.05, because the

Advanced Materials Research Vol. 132

81

amount of Gd2O3 is lower and the Gd species are highly dispersed in the catalyst. When Gd2O3/CuO = 0.05, the diffraction peaks corresponding to CuGd2O4 spinel appear and no Gd2O3 can be observed, which shows that the Gd species reacts easily with Cu species to form spinel. In the low Gd2O3 loadings, the Gd species are dispersed among Cu species, which favors the interaction of Gd and Cu species. When the Gd2O3 loading increases further to Gd2O3/CuO = 0.1, the diffraction peaks of Gd2O3 crystallite are observed, and the peaks of CuGd2O4 spinel are enhanced obviously. It means that Gd species gather quickly with the increase of Gd2O3 loading. The studies of Lee et al. [24] showed that the smaller particle sizes of CuO in CuO(7)-Cr2O3/DM22 may be caused by the consumption of CuO to form the Cu-Cr oxides complex. For the Gd2O3-CuO/AC with Gd2O3/CuO > 0.05, the decrease in the size of CuO crystallites may be attributed to the consumption of CuO to form the CuGd2O4 spinel.

2.24 2.00

0.1Gd2O3-CuO/AC

Intensity /a.u.

0.05Gd2O3-CuO/AC 0.03Gd2O3-CuO/AC

CuO/AC 2.33

2.06

1000

2000

3000

4000

5000

6000

4

Magnetic field /10 T

Fig. 4. EPR spectra of CuO/AC and Gd2O3-CuO/AC at room temperature. X-ray photoelectron spectroscopy has been used to characterize the catalyst samples and the results obtained are shown in Table 3 and Fig.5. The binding energies (BEs) of carbon (C 1s) are around 284.6 and 286.6 eV, the former can be attributed to reduced-state carbon and the latter is to oxidized-state carbon [29]. It can be seen that carbon exists mainly in reduced-state (about 78%), and oxidized-state carbon is about 22% that can be attributed to the presence of oxygen species such as lactones, quinones, carboxylic acids, etc. The BEs of Cu 2p3/2 are around 932.9 and 933.9 eV, the former (932.9 eV) indicates the presence of Cu0 and the latter (933.9 eV) is characteristic of Cu2+ species in CuO [29, 30]. The results in Table 3 show that, Cu2+(~90%) and Cu0(~10%) coexist in the CuO catalyst. When the loadings of Gd2O3 is low (such as Gd2O3/CuO < 0.03) the presence of Gd2O3 varies hardly the content of Cu0; when the loading of Gd2O3 is higher, the presence of Gd2O3 would lead to a decrease of the Cu0 content, such as in the 0.05Gd2O3-CuO/AC sample the Cu0 content is almost to null. The BE of Gd 3d5/2 in 0.02Gd2O3-CuO/AC is about 1186.6 eV which may be attributed to CuGd2O4. In 0.03Gd2O3-CuO/AC and 0.05Gd2O3-CuO /AC, the EBs of Gd 3d5/2 are 1187.35 and 1188.00 eV respectively, which show the presence of Gd2O3 [31]. These data of XPS above are in agreement with the results of XRD, that is, with the increase of loading of Gd2O3 in the CuO catalysts the Gd2O3 and CuGd2O4 phases strengthen. Auger kinetic energy of Cu LMM is around 917.7 eV, which shows that CuO mainly exists in the catalyst. However the distinguish

82

Advances in New Catalytic Materials

between Cu0 and Cu1+ based on the XPS data is difficult, because the BE of Cu0 (932.7 eV) is very close the BE of CuI (932.6 eV) [32].

Intensity /a.u.

Cu 2p3/2

a

b c d 930

940

950

960

970

Binding Energy /eV Fig. 5. Cu 2p3/2 XPS spectra of the Gd2O3-CuO/AC sample with the Gd2O3/CuO ratio of (a) 0/1, (b) 0.02/1, (c) 0.03/1, and (d) 0.05/1.

Fig. 5 shows the Cu 2p3/2 XPS spectra of Gd2O3-CuO/AC samples, and the relative intensities of peaks are different, because of the different loadings of Gd2O3. 0.02Gd2O3-CuO/AC, 0.03Gd2O3-CuO/AC and CuO/AC exhibit the peaks of Cu2+ (933.9 eV) and Cu0 (932.9 eV), while 0.05Gd2O3-CuO/AC only has the peak of Cu2+ (934.07 eV). With an increase of Gd2O3 loading, the areas of Cu 2p3/2 peaks in Gd2O3-CuO/AC reduce, that is, the concentration of Cu cations declines because of dilution of Gd on the surface. However, among three Gd-doping samples, 0.03Gd2O3-CuO/AC has the highest concentration of reduced species Cu0(11%). Comparing with the activities of the catalyst samples, it may be concluded that the coexistence Cu2+ and Cu0 can improve the reduction of NO by AC, in which the redox recycle of Cu2+/Cu0 has been established. Illan-Gomez et al. [33-35] thought that alkaline, alkaline-earth and transition metals catalyzing NO reduction by carbon is based on its role in the redox cycle, which involves active oxygen transferring from the catalyst to the carbon surface and the reaction between active oxygen and AC to form CO and CO2. Their studies show that the catalyst that is active for carbon gasification will also be active for the reduction of NO by carbon. With the increase of the reaction temperature, the surface oxygen groups on AC would be activated to decompose to CO and CO2, so the TPD technique can be used to characterize the activation of AC catalyzed by CuO [23]. In the TPD processes of carbon or the carbon-supported catalysts, CO2 can release by the decomposition of surface carboxylic acid at low temperatures and from the decomposition of anhydrides, lactones or lactols at higher temperatures. Carboxylic anhydrides can decompose to both CO and CO2, while phenols, ethers and carbonyls (and quinines) can only produce CO [12, 36]. Fig. 6 shows the TPD profiles of CO (m/z = 28) and CO2 (m/z = 44) on the Gd2O3-CuO/AC samples. The results show that Gd2O3/AC is slightly active for the decomposition of surface oxygen groups at T< 800 °C, and CuO/AC and Gd2O3-CuO/AC are active for the decomposition of surface oxygen groups. For all the samples, there are three major desorption peaks in the CO2 desorption profiles (m/z=44) at ~ 260 °C, ~ 410 °C, and 520~700 °C, respectively. The first desorption peak of

Advanced Materials Research Vol. 132

83

CO2 at ~ 260 °C can be attributed to the decomposition of surface carboxylic acid groups, the second at ~ 410 °C is resulted from lactones or lactols and the third at T > 520°C may come from the carboxylic anhydrides [12].

140 0.03Gd2O3-CuO/AC

120

0.05Gd2O3-CuO/AC

13

CO2(m/z =44) Intensity X10 /torr

0.02Gd2O3-CuO/AC

0.1Gd2O3-CuO/AC

100

0.5Gd2O3-CuO/AC

80

CuO/AC Gd2O3/AC

60 40 20 0

200

400

600

800

1000

o

Temperature / C

CO(m/z =28) Intensity X1013/torr

140

0.02Gd2O3-CuO/AC 0.03Gd2O3-CuO/AC

120

0.05Gd2O3-CuO/AC 0.1Gd2O3-CuO/AC

100

0.5Gd2O3-CuO/AC CuO/AC Gd2O3/AC

80 60 40 20 0

200

400

600

800

o

Temperature / C

Fig. 6. TPD profiles of CO(m/z=28) and CO2(m/z = 44) on the Gd2O3-CuO/AC samples. After CuO/AC is doped with Gd2O3, desorption peaks of CO2 would vary remarkably with different loadings of Gd2O3. As Gd2O3/CuO < 0.03, three desorption peaks of CO2 are larger than that of CuO/AC and augment with an increase in the loading of Gd2O3. At Gd2O3/CuO = 0.03, the areas of desorption peaks of CO2 reach the maximum value. As Gd2O3/CuO > 0.03 (such as 0.05), desorption peaks of CO2 decrease sharply. That Gd2O3-CuO/AC (Gd2O3/CuO>0.05) samples having a less CO2 evolution than that of CuO/AC can be attributed to an extensive coverage of AC by Gd compounds (such as Gd2O3) and a reduction of CuO dispersion because of the formation of CuGd2O4 spinel and Gd2O3, as well as the decrease of the active sites of CuO. Comparing with the TPD spectra of CO2, the desorption peaks of CO present mainly at T > 700 °C, the peaks at ~260 °C and ~410 °C are very small. CO comes from the CO-desorbing functional

84

Advances in New Catalytic Materials

groups, such as phenols, carbonyl, quinine groups and so on. It is well known that the CO-desorbing functional groups are more thermally stable than the CO2-desorbing ones. The top temperatures of peaks at ~260 °C and ~410 °C are consistent with that of first two desorption peaks of CO2, which show that these CO results from the incomplete decomposition of the CO2-desorbing oxygen groups, such as carboxyl and lactones or lactols. The presence of Gd2O3 in CuO/AC makes the CO desorption peak at ~700 °C shift to ~750 °C. Adding a small quantity of Gd2O3 (Gd2O3/CuO=0.03) may increase the areas of the CO desorption peaks at low temperature, that is to say, modification of Gd2O3 on CuO/AC is in favor of the activation of AC, especially at low temperature. It can be concluded from Fig. 6 that the adding of Gd2O3 can improve the catalytic ability of CuO for the activation of AC. When the results in Fig. 1 are correlative with that in Fig. 6, it can be found that the catalytic activity of Gd2O3-CuO/AC for the activation of AC is consistent with its catalytic performance for the NO reduction by AC, especially at low temperature (such as 260 °C). The catalytic reduction of NO by carbon should accord with the oxidation-reduction (redox) mechanism, which can be described as following kinetic steps: (1) The transfer of oxygen from NO to catalytically active sites by the dissociative chemisorption of NO on the catalyst surface; (2) The transfer of adsorbed oxygen on the catalytically active sites to the carbon reactive sites; (3) The reaction between adsorbed oxygen on the carbon reactive sites and carbon to form CO2 and/or CO; (4) Desorption of CO2 and/or CO from carbon. The carbon reactive sites can be produced through the oxidation of carbon by the carbon–oxygen surface complexes and these “nascent” carbon active sites can accelerate greatly the transfer and release of oxygen decomposed from NO, thus promoting the reduction of NO by carbon [23]. The presence of Gd2O3 in CuO catalyst alters the chemical state and environment of Cu active sites, resulting in the production of much more carbon reactive sites and accelerating the transfer and release of oxygen atoms from the decomposition of NO, which improve the catalytic activity of the catalyst for the NO reduction by carbon. In 0.03Gd2O3-CuO/AC, CuO and Gd2O3 have a good dispersion on AC and the cooperation of CuO and Gd2O3 promotes effectively the activation of AC by the decomposition of surface oxygen groups (especially carboxylic groups) to form more carbon reactive sites, resulting in the improvement of the catalytic reduction of NO by carbon. As the reduction of NO by AC occurs at T < 300 °C, the carboxylic groups play an important role in the activation of AC. 4. Conclusions The Gd2O3-doped CuO/AC catalyst shows high catalytic performance for the reduction of NO by activated carbon. The complete reduction of NO can be obtained at 285 °C on Gd2O3 modified CuO/AC, which is about 50 °C lower than that of CuO/AC. The appropriate molar ratio of Gd2O3/CuO is 0.03:1. In the Gd2O3-CuO/AC catalyst, CuO and Gd2O3 are highly dispersed on activated carbon, and the cooperation between CuO and Gd2O3 alters the chemical state and environment of Cu active sites and improves the activation of activated carbon to form carbon reactive sites, resulting in the improvement of the catalytic reduction of NO by carbon over CuO. The coexistence of Cu2+ and Cu0 (or Cu1+) is useful for the reduction of NO by carbon. In the catalytic reduction of NO by carbon at T< 300 °C, the carboxylic groups play an important role in the activation of AC. Acknowledgements This project was supported financially by the National Basic Research Program of China (2010CB732300), the National Key Technologies R & D Program of China (2007BAJ03B01), and the National Natural Science Foundation of China (20601008).

Advanced Materials Research Vol. 132

85

References [1]

W.S. Epling, L.E. Campbell, A. Yezerets, N.W. Currier, J.E. Parks II: Catal. Rev. 46 (2004), p. 163.

[2]

V.I. Parvulescu, P. Grange, B. Delmon : Catal. Today 46 (1998), p. 233.

[3]

G. Marban, R. Antuna, A.B. Fuertes : Appl. Catal. B. 41 (2003), p. 323.

[4]

G.S. Szymanski, T. Grzybek, H. Papp, Catal. Today 90 (2004), p. 51.

[5]

L.Y. Hsu, H. Teng: Appl. Catal. B 35 (2001), p.21.

[6]

N. Shirahama, I. Mochida, Y. Korai, K.H. Choi, T. Enjoji, T. Shimohara, A. Yasutake: Appl. Catal. B 52 (2004), p. 173.

[7]

N. Shirahama, I. Mochida, Y. Korai, K.H. Choi, T. Enjoji, T. Shimohara, A. Yasutake: Appl. Catal. B 57 (2005), p. 237.

[8]

E. Bekyarova, M. Khristova, D. Mehandjiev: J. Colloid Interf. Sci. 213 (1999), p. 400.

[9]

H.H. Tseng, M.Y. Wey, Y.S. Liang, K.H. Chen: Carbon 41 (2003), p. 1079.

[10] Z.H. Zhu, L.R. Radovic, G.Q. Lu: Carbon 38 (2000), p. 451. [11] N.B. Stankova, M.S. Khristova, D.R. Mehandjiev: J. Colloid Interf. Sci. 241 (2001), p. 439. [12] L.Y. Hsu, H.S. Teng: Appl. Catal. B 42 (2003), p. 69. [13] A. Obuchi, I. Kaneko, J. Oi, A. Ohi, A. Ogata, G..R. Bamwenda, S. Kushiyama: Appl. Catal. B

15 (1998), p. 37. [14] R.Q. Long, R.T. Yang: Appl. Catal. B 27 (2000), p. 87. [15] M. Haneda, Y. Kintaichi, H. Hamada: Catal. Today 54 (1999), p. 391. [16] L.L. Ren, T.Zhang, D.B. Liang, C.H. Xu, J.W. Tang, L.W. Lin: Appl. Catal. B 35 (2002),

p. 317. [17] M.D. Fokema, J.Y. Ying: Catal. Rev. 43 (2001), p. 1. [18] J.A.Z. Pieterse, H. Top, F. Vollink, K. Hoving, R.W. van den Brink: Chem. Eng. J. 120 (2006),

p. 17. [19] X. Zhang, A.B. Walters, M.A. Vannice: Catal. Today 27 (1996), p. 41. [20] M.D. Fokema, J.Y. Ying: J. Catal. 192 (2000), p.54. [21] J.H. Holles, M.A. Switzer, R.J. Davis: J. Catal. 190 (2000), p. 247. [22] S. Wisniewski, J. Belkouch, L. Monceaux: C.R. Acad. Sci. Series IIC- Chem. 3 (2000), p. 443. [23] Y.Y. Xue, G.Z. Lu, Y. Guo, Y.L. Guo, Y.Q. Wang, Z.G. Zhang: Appl. Catal. B 79 (2008),

p. 262. [24] C.Y. Lee, T.H. Jung, B.H. Ha: Appl. Catal. B 9 (1996), p. 77. [25] A. Martinez-Arias, R. Cataluna, J.C. Conesa, J. Soria: J. Phys. Chem. B 102 (1998), p. 809. [26] A.B. Hungria, A. Iglesias-Juez, A. Martinez-Arias, M. Fernandez-Garcia, J.A. Anderson,

J.C. Conesa, J. Soria: J. Catal. 206 (2002), p. 281. [27] A. Murali, R.P. S. Chakradhar, J. L. Rao: Physica B 364 (2005), p. 142. [28] E. Malchukova, B. Boizot, D. Ghaleb, G. Petite: J. Non-cryst. Solids 352 (2006), p. 297. [29] N. Barrabes, J. Just, A. Dafinov, F. Medina, J.L.G. Fierro, J.E. Sueiras, P. Salagre, Y. Cesteros:

Appl. Catal. B 62 (2006), p. 77.

86

Advances in New Catalytic Materials

[30] R. Nickolov, T. Tsoncheva, D. Mehandjiev: Fuel 81 (2002), p. 203. [31] B. Glorieux, R. Berjoan, M. Matecki, A. Kammouni, D. Perarnau: Appl. Surf. Sci. 253 (2007),

p. 3349. [32] C. D. Wagner, W. M. Riggs, L. E. Moulder, G. E Muilenberg: Handbook of X-ray

Photoelectron spectroscopy; Perkin-Elemer Corporation Physical Electronics Division: U.S.A., 1979 [33] M.J. Illan-Gomez, A. Linares-Solano, C. Salinas-Martinez de Lecea, Energ. Fuel. 9 (l995),

p. 976. [34] M.J. Illan-Gomez, A. Linares-Solano, L.R. Radovic, C. Salinas-Martınez de Lecea: Energ.

Fuel. 10 (1996), p.158. [35] M.J. Illan-Gomez, C. Salinas-Martınez de Lecea, A. Linares-Solano, L. R. Radovic: Energ.

Fuel. 12 (1998), p. 1256. [36] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão: Carbon 37 (1999), p. 1379.

Advanced Materials Research Vol. 132 (2010) pp 87-95 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.87

Synthesis and Photocatalytic Performance of Hierarchical Porous Titanium Phosphonate Hybrid Materials Tian-Yi Ma1,a, Tie-Zhen Ren2,b and Zhong-Yong Yuan1,c,* 1

Institute of New Catalytic Materials Science, Engineering Research Center of Energy Storage and Conversion (Ministry of Education), College of Chemistry, Nankai University, Tianjin 300071, China. 2

School of Chemical Engineering, Hebei University of Technology, Tianjin 300130, China. a

[email protected], [email protected], [email protected]

Keywords: Hierarchical porosity; Titanium phosphonate; Titanium phosphate; Organic-inorganic hybrid; Photocatalyst.

Abstract. A hierarchical meso-/macroporous titanium phosphonate (TPPH) hybrid material was prepared via a simple surfactant-assisted process with the use of the precursor tetrabutyl titanate and 1-hydroxy ethylidene-1,1-diphosphonic acid. The prepared hybrid TPPH presented amorphous phase, exhibiting a hierarchical macroporous structure composed of mesopores with a pore size of 2.0 nm. The BET surface area is 256 m2/g. The hydroxyethylidene-bridged organophosphonate groups were homogeneously incorporated in the network of the hierarchical porous solid, as revealed by FT-IR, MAS NMR, XPS, and TGA measurements. The optical properties and photocatalytic activity of the hierarchical TPPH material were investigated in comparison with those of hierarchical porous titanium phosphate and pure mesoporous titania materials, showing superiority of the inorganic-organic hybrid framework, suggesting promising photocatalysts for wastewater cleanup. Introduction In recent years, designing advanced materials with hierarchical meso-/macroporous structure has attracted tremendous interest [1-6], since the occurrence of macropores incorporated in mesoporous materials is important and useful for catalysis and adsorption/separation, and for engineering of pore systems. The enhanced performance of the hierarchical meso-/macroporous structures could be attributed to the increased mass transfer and reduced diffusion resistance [7]. The superior photocatalytic activity has been observed in the obtained hierarchical macro-/mesoporous titanias, due to the light-harvesting macroporous channels [8]. Although the combination of the surfactant templating techniques and the micromold methods of emulsion droplets [9], colloid crystals [10], or bacterial threads [11] could allow for the construction of meso-/macroporous materials, it has always been a challenge to design the architecture and tailor the porous hierarchy. Many metal phosphonate and metal oxide/organophosphonate hybrid porous materials have been synthesized recently [12-18]. Some of the alkyl-monophosphonic acids and their derivatives (salts, esters) were used as organophosphorus coupling molecules to modify metal oxide surfaces by grafting [19,20]. Alternatively a two-step nonhydrolytic/hydrolytic sol-gel process was performed to prepare metal oxide/organophosphonate hybrids, in which the organophosphorus coupling molecules were incorporated in the inorganic network [21]. Several mesoporous aluminum organophosphonate materials were also reported recently by a surfactant-templating method [17,18,22,23]. Song et al. synthesized metal phenylphosphonate nanoparticles and nanorods by surfactant-assisted methods [24,25]. Vasylyev et al. reported titanium and oxide-phosphonate porous nanospherical particles by non-hydrolytic condensation of water-insoluble arylphosphoric acid (tetrakis-1,3,5,7-(4-phosphonatophenyl) adamantane) and titanium isopropoxide [26]. Threedimensional ordered macroporous titanium phosphonate hybrids using polystyrene spheres as hard templates [27] and mesoporous titania–phosphonate materials with organically bridged tetra- or

88

Advances in New Catalytic Materials

penta-phosphonates [28] were prepared in our previous work, in which higher photocatalytic activity could be observed for the hybrid materials than for a pure titania photocatalyst. In this work, we report the preparation of a hierarchical titanium phosphonate material with wormhole-like mesostructure in the macroporous nanoarchitecture, in which hydroxyethylidene groups were anchored in the titanium phosphonate network. The superior photocatalytic performance of the obtained hybrid titanium phosphonates to that of meso-/macroporous titanium phosphate and titania materials with pure inorganic framework is highlighted, which could benefit from the elaborate nanostructure and the large scaled micro-architecture and the homogeneous incorporation of phosphorus and carbon by the organic-inorganic network. It also sets an advanced example of the assembly of individual nanoparticles of metal organophosphonate into hierarchical nanoarchitectures with complex shapes for multifunctionalization and potential applications. Experimental Section Material Synthesis. All chemicals were used as received without further purification. In a typical synthesis procedure, 0.05 mol of 1-hydroxy ethylidene-1,1-diphosphonic acid (HEDP) was dissolved in a mixed solution of ethanol (10 ml) and distilled water (30 ml) in the presence of cetyltrimethylammonium bromide (CTAB) with stirring, followed by the dropwise addition of 0.05 mol of tetrabutyl titanate solution. After further stirring for 24 h, the mixture was sealed in a Teflonlined autoclave and heated statically at 80 ºC for 24 h. The product was filtered, washed with water, dried at 60 ºC, and then transferred to a Soxhlet reactor to remove the attached surfactant by ethanol for 24 hours, which was denoted as TPPH. Hierarchical porous titanium phosphate materials were also prepared by a similar procedure with the use of phosphoric acid instead of phosphonic acid, denoted as TPO. Mesoporous pure titania was also prepared in the absence of phosphoric acid/phosphonic acid. Characterization. X-ray diffraction (XRD) patterns were recorded on a Rigaku D/max-2500 diffractometer with Cu-Kα radiation operated at 40 kV and 100 mA. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were carried out on a Shimadzu SS-550 microscope at 15 keV and a Philips Tecnai G20 at 200 kV, respectively. N2 adsorption and desorption isotherms were obtained on a Quantachrome NOVA 2000e sorption analyzer at liquid nitrogen temperature (77 K). The samples were degassed at 80 ºC overnight before measurements were made. Fourier transform infrared (FT-IR) spectra were recorded on a Bruker VECTOR 22 spectrometer with the KBr pellet technique. Diffuse reflectance UV-vis absorption spectroscopy was employed on a JASCO V-570 UV-V-NIR spectrophotometer using BaSO4 as a reference. Xray photoelectron spectroscopy (XPS) measurements were performed on a Kratos Axis Ultra DLD (delay line detector) spectrometer equipped with a monochromatic Al-Kα X-ray source (1486.6 eV). All XPS spectra were recorded using an aperture slot of 300 × 700 microns, survey spectra were recorded with a pass energy of 160 eV, and high resolution spectra with a pass energy of 40 eV. Solid-state 31P and 13C magic angle spinning (MAS) nuclear magnetic resonance (NMR) spectra were recorded on a Varian Infinity plus-400 spectrometer at spinning rates of 12 and 6 kHz and resonance frequencies of 161.9 and 100.5 MHz with recycle time of 5 and 3 s, respectively. Thermogravimetry (TG) and differential scanning calorimetry (DSC) were performed using a TA SDT Q600 instrument at a heating rate of 5 º/min using α-Al2O3 as the reference. The chemical compositions of Ti and P were analyzed by inductively coupled plasma (ICP) emission spectroscopy on a Thermo Jarrell-Ash ICP-9000 (N+M) spectrometer, and C, N and H were analyzed on a Vario-EL elemental analyzer. Photocatalytic Activity Test. The photocatalytic activity experiments were performed by the degradation of rhodamine B (RhB) dye under either UV or visible-light irradiation in the air at room temperature. In UV-photocatalysis experiment, 5.5 mg of the synthesized catalyst was placed into a tubular quartz reactor of 100 ml of RhB aqueous solution (1×10-5 mol/L). A 125W UV lamp with maximum emission at 365 nm was located at 10 cm higher than the solution surrounded by a circulating water tube. The reaction mixture was stirred under UV-light irradiation. The mixture

Advanced Materials Research Vol. 132

89

sampled at different times was centrifuged for 5 min to discard any sediment. The absorbance of reaction solutions was measured by a SP-722 spectrometer at λmax=554 nm. The visible-light photodecomposition of RhB was carried out with a household desktop lamp with a 40-watt tungsten bulb as the visible light source, of which the wavelength range is usually considered as 4002500 nm, and the concentration of RhB solution and the amount of catalyst used are 1×10-5 mol/L and 20 mg respectively. Results Material Synthesis and Characterization. The synthesis of hierarchical titanium phosphonate (TPPH) was accomplished by a simple CTAB-assisted method. Tetrabutyl titanate was added into a CTAB solution of ethanol and water in the presence of HEDP, followed by autoclaving at 80 ºC for 24 h. In order to investigate the superiority of the organophosphonic coupled hybrid material, titanium phosphate (TPO) and pure titania (TiO2) materials were also prepared in a similar process for comparison. By doping organophosphonic acid or phosphoric acid, typical hierarchically meso/macroporous structures were obtained, which are shown in Fig. 1. Macropores with diameter of 80200 nm can be seen in Fig. 1a of the TPO sample and the macroporous walls were composed of disorderd nano-sized mesopores that were assembled by the titanium phosphate nanoparticles (Fig. 1b). For the hybrid TPPH material, a large number of alveolate macropores with openings ranging from 90 to 400 nm spread over the entire structure (Fig. 1c). The macroporous frameworks are composed of wormhole-like mesopores [29] (Fig. 1d). a

b

e

d

Intensity (a. u.)

c

TiO 2

TPPH

TiPO

10 nm

10

20

30

40

50

60

70

80

2 θ (degree)

Fig. 1. SEM (a) and TEM (b) images of sample TPO; SEM (c) and TEM (d) images of sample TPPH; (e) XRD patterns of the synthesized TPPH, TiPO and TiO2.. The XRD patterns of the synthesized TPPH and TPO samples show one wide diffraction peak in the 2θ range of 15 to 40° (Fig. 1e), indicating amorphous frameworks of titanium phosphonates and titanium phosphates, while pure titania obtained in the absence of phosphonic acid shows the bicrystalline phases of anatase and brookite (Fig. 1e). This indicates that the hydrolysis of titanium alkoxide in the phosphonic acid solution resulted in the incorporation of organophosphonate into the titania network, leading to amorphous titanium phosphonate nanoclusters.

90

Advances in New Catalytic Materials

Fig. 2 shows the N2 adsorption–desorption isotherms and the corresponding pore size distribution curves of the hybrid TPPH, TPO and pure titania, and their textural properties are listed in Table 1. The isotherm of TPPH is of type II, showing a gradual increase of nitrogen adsorbed volume with an increase in the relative pressure, which has been observed previously in several assynthesized surfactant-containing mesoporous silica materials, and some macroporous materials [30]. The pore size distribution curve derived from the adsorption branch of the isotherm using the BJH method exhibits an asymmetric peak maximized at 2.0 nm, which corresponds to the wormhole-like mesostructure observed in titanium phosphonate hybrids (Fig. 1). The multi-point BET surface area is 256 m2/g with a total pore volume of 0.26 cm3/g. The isotherms of TiO2 and TPO are of type IV with H2 hysteresis loop and type II with H3 hysteresis loop, respectively. The surface area of synthesized TPO material is 232 m2/g with a total pore volume of 0.28 cm3/g, while the surface area of pure TiO2 is 245 m2/g with a total pore volume of 0.27 cm3/g. This suggests that the synthesized TPO and TiO2 have similar textural properties to that of the TPPH hybrid. 0.12

0.09

3

320

TPPH TiPO TiO2

240 0.06

3

dV/dD (cm /g/nm)

Volume adsorbed (cm /g)

400

160 0.03 80

0.0

0.2

0.4

0.6

0

0.8

Relative pressure (P/P0)

10

20

30

40

0.00 50

Pore diameter (nm)

Fig. 2. N2 adsorption-desorption isotherms (left) and the corresponding BJH pore size distribution curves (right) of the TPPH, TiPO and TiO2 samples. The volume adsorbed was shifted by 180 and 90 for TPPH and TiPO, and dV/dD value was shifted by 0.05 for TPPH. Table 1. Summary of the physicochemical properties, photocatalytic rate constants of the synthesized materials. SBET a DBJH-ads b Dave c Vpore d photodegradation rate constant, k (min-1) 2 3 Sample [m /g] [nm] [nm] [cm /g] UV Visible-light TPPH 256.7 2.0 4.4 0.263 0.0168 0.00526 TiPO

232.7

3.9

1.1

0.283

0.0141

0.00502

TiO2

244.5

2.2

4.3

0.265

0.0134

0.00366

a

BET surface area calculated from the linear part of the 10-point BET plot. b Estimated using the adsorption branch of the isotherm by the BJH method. c BJH average pore size (4V/A). d Single point total pore volume of pores at P/P0 = 0.97. The FT-IR spectrum of TPPH is shown in Fig. 3, and is compared with the spectrum of HEDP. The broad band at 3400 cm-1 and the sharp band at 1638 cm-1 correspond to the surface-adsorbed water and hydroxyl groups. The strong band at 1046-1150 cm-1 is from the phosphonate P-O···Ti stretching vibrations. An obvious absorption peak at 928 cm-1 assigned to P-O···H is observed in the infrared spectrum of HEDP but absent for TPPH [21,26], suggesting the extensive condensation

Advanced Materials Research Vol. 132

91

Absorbance

between Ti-OBu and P-OH groups to form Ti-O-P bridges. The bands at 1380 and 1450 cm-1, attributed to C-O and P-C stretching vibrations, respectively [20], still remain in TPPH. This implies that the organophosphonate groups retain their integrity in TPPH. The 31P MAS NMR spectrum of the resultant TPPH sample shows the resonance at δ = 16.25 ppm (Fig. 4), which can be attributed to diphosphonate groups [≡P-C-(OH)(CH3)-P≡] linked to the Ti atoms [17,18,27]. The sharp 31P NMR resonance signal for layered titanium phosphonate was not observed at δ = -4 ppm [21]. A 13C MAS NMR spectrum of the sample exhibits resonances at δ = 19.7 and 69.0 ppm (Fig. 4), which correspond to the C atoms of the terminal CH3 group and the quaternary carbon atom connected with the P=O group of the phosphonate, respectively. It is thus deduced from the FT-IR and NMR spectroscopic results that organophosphonate groups are homogeneously anchored in the hierarchical nanostructured/porous solid.

(a)

(b)

4000

3500

3000

2500

2000

1500

1000

500

-1

Wavenumber (cm )

Fig. 3. FT-IR spectra of (a) HEDP and (b) TPPH. 19.7

16.25 13 31

C

P

69.0

*

* 150

100

50

0

-50

Chemical shift (ppm)

-100

-150

200

150

100

50

0

-50

-100

Chemical shift (ppm)

Fig. 4. 31P and 13C MAS NMR spectra of TPPH material. High-resolution XPS spectra were also taken on the surface of the TPPH sample for an investigation of the chemical state and surface stoichiometry (Fig. 5). The Ti 2p line of the TPPH sample is composed of two single peaks situated at 459.4 eV for Ti 2p3/2 and 465.2 eV for Ti 2p1/2, which are characteristic of Ti4+. The P 2p binding energy of TPPH is observed around 133.1 eV, characteristic of P5+ in phosphonate groups. No peaks of Ti–P bonds appear at 128.6 eV. The broad O 1s signal might be fitted by four components, situated at 530.7, 531.4, 532.1, and 533.0 eV, ascribed to the oxygen contribution from the Ti-O, P-O, O-H, and C-O bonds, respectively [3]. The surface atomic composition was calculated as 4.31% for Ti, 10.58% for P, 38.76% for C, and

92

Advances in New Catalytic Materials

44.58% for O. The Ti/P ratio is almost 1:2, and a molecular unit of [Ti{O3P-C(CH3)(OH)PO3}]·xH2O can be formulated for TPPH. Alternative formulation can be expressed as [Ti(HEDP)]·xH2O, which is mostly consistent with the data from the elemental microanalysis (experimental 15.55% Ti, 12.31% P, 6.03% C, and 5.29% H by mass), suggesting compositional homogeneity throughout the hybrid material. Ti 2p

P 2p

133.1

Intensity (a.u.)

Intensity (a.u.)

459.4

465.2

470 468 466 464 462 460 458 456 454 452

140

138

136

O 1s

134

132

130

128

Binding energy (eV)

Binding energy (eV) 531.2

Intensity (a.u.)

Intensity (a.u.)

O1s

542

540

538

536

534

532

530

528

536

Binding energy (eV)

534

532

530

528

Binding energy (eV)

Fig. 5. High-resolution XPS spectra of the Ti 2p, P 2p, and O 1s regions of TPPH material.

Absorbance (a.u.)

The amount of water and the thermal stability of TPPH were determined by thermal gravimetric analysis and differential scanning calorimetry. The TGA curve demonstrates an initial weight loss of 10.77% from room temperature to 217 °C, accompanied by an endothermal peak around 108 °C in the DSC curve, which may be assigned to the desorption of the adsorbed and intercalated water. The weight loss of 7.24% from 217 to 620 °C, accompanied by two exothermic peaks at 308 and 556°C, can be attributed to the decomposition of the organic species and the coke combustion. Thus, 1.7 molecules of H2O per formula unit were calculated.

423nm TPPH

408nm

TiPO

400nm

200

300

400

TiO2 500

600

700

800

Wavelength (nm)

Fig. 6. UV-vis. diffuse reflectance spectra of TPPH, TiPO and TiO2.

Advanced Materials Research Vol. 132

93

Photocatalytic Activity. Titania requires the near-UV light to be used to activate its attractive photocatalytic ability. Unfortunately, in solar energy applications only ca. 3% of the solar light is absorbed. Many technologies have been utilized to promote photoabsorption efficiency of titania, such as surface chelation, selective metal ion doping and platinization. Herein, the structure modification of a porous hierarchy and the inorganic-organic hybrid network contribute cooperatively to promote the photocatalytic activity of the TPPH material. UV-vis diffuse reflectance spectroscopy was performed to assess the optical properties and electronic structure of the hierarchical TPPH. Fig. 6 presents the diffuse reflectance spectra of the TPPH, TPO and pure TiO2 samples, where a low reflectance means a high absorption at the corresponding wavelength. The onset wavelength of absorption (λonset) for TPPH is about 423 nm, which is larger than that of TiPO (about 408 nm) and pure TiO2 (about 400 nm). The bandgap value (Eg) of the TPPH and TPO are estimated to be 2.93 eV and 3.03 eV from absorption spectra by a linear fit of the square root of the absorption coefficient (α1/2) as a function of the photon energy (hν) near the bandgap. Compared with the Eg of pure TiO2 (3.10 eV), the bandgap narrowing observed in TPPH and TPO should be the result of the homogeneous doping of phosphorus into the framework of the porous TPPH and TPO materials. 10

8 1.2

-6

Concentration (x10 mol/L)

1.6

ln(C0/C)

6

4

Self-degradation TiO2

0.4

TiPO TPPH

2 0

0.8

0.0

20

40

60

80

0

100

20

40

60

80

100

UV-light Irradiation Time (min)

UV-light Irradiation Time (min)

9

0.6

8

ln(C0/C)

-6

Concentration (x10 mol/L)

10

7

0.2

6

Self-degradat TiO2

5 4

0.4

TiPO TPPH 0

20

0.0

40

60

80

100

Visible-light Irradiation Time (min)

120

0

20

40

60

80

100

Visible-light Irradiation Time (min)

Fig. 7. The residual concentration of RhB solution after UV- and visible-light photodecomposition by TPPH, TiPO, TiO2 and the corresponding plots of ln(C0/C) versus the irradiation time, showing the fitting results using the pseudo-first-order reaction. The photocatalytic activities of the synthesized hierarchical meso-/macroporous titanium phosphonate materials were evaluated by photodegradation of Rhodamine B (RhB) under UV and visible-light irradiation (Fig. 7), and compared with that of the TPO and TiO2. A blank experiment (self-photosensitized process) was also performed in the absence of any catalysts for comparison. As shown in Fig. 7, the TPPH sample exhibits superior photocatalytic activity to the others, whether under UV or visible-light irradiation, giving the following sequence: TPPH > TPO > TiO2 > Self-

94

Advances in New Catalytic Materials

degradation. The photocatalytic degradation rate constants (k) of TPPH, TPO and TiO2, calculated by a pseudo-first-order expression, are 0.0168, 0.0141 and 0.0134 when under UV irradiation, while 0.00526, 0.00502 and 0.00366 under visible-light irradiation, listed in Table 1. Discussion Titania photocatalysts doped with C [31], P [32] N [33], etc. were reported to show their absorption edge red-shifted to lower energies (longer wavelengths), enhancing photocatalytic efficiencies in the visible light range. This gives part of the explanation for the higher photocatalytic activity of TPPH and TPO samples, coupled with either organo phosphonic acid HEDP or H3PO4, than that of pure titania. On the other hand, the micro/nanocomposite architecture could also increase the efficiency of photoabsorption and improve mass transfer [8]. It is well known that the photoabsorption efficiency is one of the main influencing factors for the over-all photocatalytic activity, which is strongly influenced by the pore-wall structure of photocatalysts. In the meso/macroporous TPPH or TPO photocatalyst, the macropores acted as light-transfer paths for the distribution of scattering and absorption taking place in such a porous structure. This could enlarge the effective light-activated surface area, leading to the increasing of photoabsorption efficiency [8], and making them better photocatalysts than pure mesoporous TiO2. In the case of the higher photocatalytic activity of TPPH than TPO, both with amorphous framework, the possible carbon-doping in the TPPH hybrid material could lead to the final distinct performances. Because of the existence of hydroxyethylidene groups in the 1-hydroxy ethylidene1,1-diphosphonic acid, the possible carbon-doping was accessed in the final product, which could be confirmed by the elemental analysis and TG-DSC measurement. A red-shift of the adsorption edge of TPPH (423 nm) was observed compared with TPO (408 nm, Fig. 6), similar with the previously reported carbon-doping titania [31], which had stronger absorption in the UV–visible range and a remarkable red shift of the spectrum onset in the absorption spectrum. This made the TPPH hybrid better utilize the visible light, leading to the enhancement of photocatalytic activity. Therefore, because of the similar surface areas of the synthesized TPPH (256 m2/g), TPO (232 m2/g) and pure titania (245 m2/g), the influence of the surface area on the photocatalytic ability can be ignored. The efficient photocatalytic ability of the hierarchical titanium phosphonate material should be a result of well-structured meso-/macroporous architecture and the adding of organophosphonic acid HEDP during the preparation, which could offer an inorganic-organic hybrid framework with homogeneous incorporation of phosphorus and carbon. Conclusions In conclusion, a nanostructured titanium phosphonate hybrid material with a hierarchical macro/mesoporous structure has been prepared by using HEDP as organophosphorus coupling molecules. A higher photocatalytic activity than titanium phosphate and pure TiO2 materials was demonstrated for the synthesized hybrid material, which could be attributed to the hierarchical structure and the carbon-/phosphor-doping into the inorganic-organic framework, suggesting that they have potential in practical applications. The fabrication of other titanium/titania phosphonates with different organic bridging groups and hierarchical porous structures can also be expected. Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 20803017 and 20673060), the National Basic Research Program of China (No. 2009CB623502), the Specialized Research Fund for the Doctoral Program of Higher Education (20070055014), the Natural Science Foundation of Tianjin (08JCZDJC21500), the MOE Supporting Program for New Century Excellent Talents (NCET-06-0215), and Nankai University.

Advanced Materials Research Vol. 132

95

References [1] K. Suzuki, K. Ikari and H. Imai: J. Mater. Chem. Vol. 13 (2003), p. 1812 [2] A. Collins, D. Carriazo, S.A. Davis and S. Mann: Chem. Chommun. (2004), p. 568 [3] T.Z. Ren, Z.Y. Yuan, A. Azioune, J. Pireaux and B.L. Su: Langmuir Vol. 22 (2006), p. 3886 [4] T.Z. Ren, Z.Y. Yuan and B.L. Su: Chem. Commun. (2004), p. 2730 [5] Z.Y. Yuan, T.Z. Ren, A. Azioune, J.J. Pireaux and B.L. Su: Chem. Mater. Vol. 18 (2006), p. 1753 [6] H.L. Fei, X.Q. Zhou, H.J. Zhou, Z.R. Shen, P.C. Sun, Z.Y. Yuan and T.H. Chen: Micropor. Mesopor. Mater. Vol. 100 (2007), p. 139 [7] Z.Y. Yuan and B.L. Sun: J. Mater. Chem. Vol. 16 (2006), p. 663 [8] X. Wang, J.C. Yu, C. Ho, Y. Hou and X. Fu: Langmuir Vol. 21 (2005), p. 2552 [9] D.M. Antonelli: Micropor. Mesopor. Mater. Vol. 33 (1999), p. 209 [10] B.T. Holland, C.F. Blanford, T. Do and A. Stein: Chem. Mater. Vol. 11 (1999), p. 795 [11] B. Lebeau, C.E. Fowler, S. Mann, C. Farcet, B. Charleux and C. Sanchez: J. Mater. Chem. Vol. 10 (2000), p. 2105 [12] A. Clearfield and Z. Wang: J. Chem. Soc. Dalton Trans. (2002), p. 2937 [13] A. Clearfield: Curr. Opin. Solid State Mater. Sci. Vol. 6 (2000), p. 495 [14] P.H. Mutin, G. Guerrero and A. Vioux: J. Mater. Chem. Vol. 15 (2005), p. 3761 [15] A. Vioux, J.Le Bideau, P.H. Mutin and D. Leclercq: Top. Curr. Chem. Vol. 232 (2004), p. 145 [16] K. Maeda: Micropor. Mesopor. Mater. Vol. 73 (2004), p. 47 [17] T. Kimura: Chem. Mater. Vol. 15 (2003), p. 3742 [18] T. Kimura: Chem. Mater. Vol. 17 (2005), p. 337 [19] S. Marcinko and A.Y. Fadeev: Langmuir Vol. 20 (2004), p. 2270 [20] G. Guerrero, P.H. Mutin and A. Vioux: Chem. Mater. Vol. 13 (2001), p. 4367 [21] G. Guerrero, P.H. Mutin and A. Vioux: Chem. Mater. Vol. 12 (2000), p. 1268 [22] J.El. Haskouri, C. Guillem, J. Latorre, A. Beltran, D. Beltran and P. Amoros: Eur. J. Inorg. Chem. (2004), p. 1804 [23] X. Shi, J. Yang and Q. Yang: Eur. J. Inorg. Chem. (2006), p. 1936 [24] S.Y. Song, J.F. Ma, J. Yang, M.H. Cao and K.C. Li: Inorg. Chem. Vol. 44 (2005), p. 2140 [25] S.Y. Song, J.F. Ma, J. Yang, M.H. Cao, H.J. Zhang, H.S. Wang and K.Y. Yang: Inorg. Chem. Vol. 45 (2006), p. 1201 [26] M. Vasylyev, E.J. Wachtel, R. Popovitz-Biro and R. Neumann: Chem. Eur. J. Vol. 12 (2006), p. 3507 [27] T.Y. Ma, X.J. Zhang, G.S. Shao, J.L. Cao and Z.Y. Yuan: J. Phys. Chem. C Vol. 112 (2008), p. 3090 [28] X.J. Zhang, T.Y. Ma and Z.Y. Yuan: J. Mater. Chem. Vol. 18 (2008), p. 2003 [29] Z.Y. Yuan, T.Z. Ren and B.L. Su: Adv. Mater. Vol. 15 (2003), p. 1462 [30] M. Kruk and M. Jaroniec: Chem. Mater. Vol. 13 (2001), p. 3169 [31] M. Shen, Z. Wu, H. Huang, Y. Du and P. Yang: Mater. Lett. Vol. 60 (2006), p. 693 [32] L. Lin, W. Lin, J.L. Xie, Y.X. Zhu, B.Y. Zhao and Y.C. Xie: Appl. Catal. B: Environ. Vol. 75 (2007), p. 52 [33] S. Sato: Chem. Phys. Lett. Vol. 123 (1986), p. 126 [34] F. Lu, W. Cai and Y. Zhang: Adv. Funct. Mater. Vol. 8 (2008), p. 1

Advanced Materials Research Vol. 132 (2010) pp 96-104 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.96

Role of Nanocrystalline Titania Phases in the Photocatalytic Oxidation of NO at Room Temperature S. Castillo1, 2 a, R. Carrera1, 2 b, R. Camposeco 1 c, P. Del Ángel1 d, J. A. Montoya1 e, A. L. Vázquez1, 2 f, M. Morán-Pineda2 g, R. Gómez3 h. 1

Programa de Ingeniería Molecular, Instituto Mexicano del Petróleo, D.F., México,

2

Departamento de Ingeniería Metalúrgica, ESIQIE-IPN, AP. 75-876, D.F. México

3

Dept. de Química, Universidad Autónoma Metropolitana -Izatapalapa, A.P. 55-534, México, D.F.,09340, Mexico *Corresponding author: a [email protected]. b e

[email protected]; c [email protected]; d [email protected]; [email protected]; f [email protected]; g [email protected]; h

[email protected]

Keywords: TiO2 nanocrystalline; anatase-brookite phases; crystallite size; NO photo-oxidation.

Abstract. Nanocrystalline TiO2 powders were prepared by the sol-gel method and evaluated in the NO photocatalytic oxidation. Samples annealed at 200 and 500°C (TiO2-P-200, TiO2-P-500) were characterized by nitrogen adsorption, XRD-Rietveld refinements, TEM, FTIR and UV-vis spectroscopies. The photocatalytic test of the sol-gel TiO2 samples was carried out in an insulated chamber with 10 ppm of NO, using a 365-nm UV light lamp; the test results were compared with those obtained with a commercial catalyst (P25). Improved photoactivity (89 % of NO oxidized in 60 min) was obtained with the TiO2-P-200 solid which showed high surface area, small crystallite size, higher amount of OH and highly abundant brookite phase (37.2 %) coexisting with the anatase phase (62.8 %). The photo-oxidation activity of the sol-gel catalyst annealed at 500 °C (TiO2-P500) showed changes in its textural and morphologic properties and therefore, less photoactivity. Sol-gel photocatalysts could be a good option for abating pollution in both indoor and outdoor environments at room temperature. 1 Introduction Nowadays, the nanostructured TiO2 photocatalytic materials offer promising opportunities for their application in the environmental pollution catalytic abatement [1, 2]. Photocatalytic processes using UV-illuminated TiO2 have been successfully used for abating pollution in indoor and outdoor environments. For instance, organic volatile compounds such as acetaldehyde and formaldehyde [3-6], carcinogenic hydrocarbons [7], chloride compounds [8], and toluene [9] have been effectively removed from atmospheric environments using TiO2 as a photocatalyst and UV radiation as a light source, both in indoor and outdoor environments. However, the elimination of inorganic gaseous compounds, which are present in considerable amounts in the atmosphere, has been scarcely studied by photocatalytic methods. In this way, NOx decomposition has attracted great attention [10-14]. These pollutants are formed during the gasoline combustion process in automobiles, and hence important amounts of them can be found in the atmosphere. The NOx elimination has been carried out by oxidation and reduction photoreactions using TiO2 films and coatings in the anatase or antase/rutile crystalline forms [14, 15]. Three crystalline forms (brookite, anatase and rutile) can be present in the titanium oxide with phase transitions around 200, 400 and 600°C, respectively. Anatase and rutile are two of the titania crystalline phases that have been extensively studied, and their photocatalytic properties are well documented. Nevertheless, the photocatalytic properties of the brookite phase, as far as we know, have not been deeply studied. Thus, it has been reported that the proportion of brookite coexisting with the anatase and rutile phases depends on both the

Advanced Materials Research Vol. 132

97

preparation method and thermal stabilization conditions [16]. For instance, by using either thermolysis or hydrothermal synthesis, it is possible to obtain brookite at high temperature; likewise, in recent works, the role of brookite in the TiO2 crystal size has been analyzed [17, 18]. Additionally, it seems that there are special effects in TiO2 photoactivity related to the crystalline phase composition, nanocrystallite size, high surface area, hydroxyl surface density and band gap energy [19]. In the present work, in order to study the variables that affect the titania photoactivity, wet chemistry routes (sol-gel) were used to prepare photoactive TiO2. The catalysts were annealed at 200 and 500 °C, and characterized by XRD-Rietveld refinement, nitrogen adsorption, high resolution transmission electron microscopy (HRTEM), FTIR and UV-vis spectroscopy. The photoactivity tests were carried out in the NO photo-oxidation, which led to the formation of HNO3 that can be easily removed by using water traps. The aim of this work is to show that this kind of photocatalyst could be used for abating pollution in both indoor and outdoor environments at room temperature. 2 Experimental 2.1 Preparation of sol-gel catalysts The sol-gel TiO2 catalysts were prepared as follows: 36.6 mL of titanium (IV) isopropoxide (Aldrich, 99.9%) were dissolved in 60 mL of 2-propanol (Baker 99.9%). The solution was put under constant stirring, and then hydrochloric acid (Baker 36.5 vol. % in water) was added to adjust the reaction medium at pH 3. The hydrolysis was accomplished by adding 18 mL of bidistilled water (water/alkoxide molar ratio 1/0.125). The solution was then maintained under stirring and reflux until the gel was formed. Afterwards, the gel was dried at 70 °C for 12 h, and then annealed at 200 and 500 °C for 4 h. 2.2 Characterization The specific surface areas were calculated by the BET method from the nitrogen adsorption isotherms obtained with Micrometrics ASAP-2000 equipment. Pore volume and pore size distributions were calculated from the desorption isotherms by applying the BJH method. XRD diffraction patterns were obtained at room temperature with a Siemens D-5005 diffractometer with Cu-Kα radiation and s diffracted-beam graphite focusing monochromator, and the intensities were determined in the 2θ range between 20 and 80° with a 0.02° step and a measuring time of 5 s per point. For the analysis of the diffraction patterns by Rietveld refinement, the Full Proof software was used [20]. The crystallite size was determined by the Rietveld refinement and Scherrer equation [21, 22]. High resolution transmission electron microscopy (HRTEM) was performed in a JEOL JEM2200FS microscope with a Schottky type field gun, working at 200 kV. The point resolution was of 0.19 nm; and the information limit was better than 0.10 nm. From the obtained micrographs, the average particle size was calculated by the surface/volume equation [23]. UV-vis spectra were obtained in self-supported samples with a Cary III spectrophotometer equipped with a diffuse reflectance accessory. Once the diffuse reflectance spectra were obtained Eg was calculated by applying a method reported elsewhere [24-26]. The FTIR spectra were obtained with the Nicolet-710 equipment. The samples were prepared in potassium bromide tablets with an approximate ratio of 10 KBr parts to 1 part of sample. They were placed in a glass cell that allows “in situ” thermal treatments. 2.3 Measurement of photocatalytic activity The photocatalytic oxidation of NO was carried out at room temperature in a quartz cell photoreactor using 200 mg of catalyst and a 365-UV light lamp (UVP-Light-Sources) with an intensity of 100 µW/cm2. The gas composition was 10 ppm of NO, 2.0 % of O2 and a humidity of 15,500 ppm in helium balance. The NO photo-oxidation behavior consisted of a series of oxidation

98

Advances in New Catalytic Materials

steps by OH· radical and O2- as follows: NO → HNO2 → NO2 → HNO3 [27]. The photocatalytic oxidation of NO was followed by means of infrared spectroscopy using an IFS66V/s equipment with a high sensibility MCT detector (Mercury-Tellurium-Cadmium), resolution of 0.5 cm-1 and an optic step length of 25 cm. 3 Results and Discussion 3.1 BET and XRD analysis A very important effect of the annealing temperature on the specific surface area and pore volume on the sol-gel catalysts can be observed in Table 1. The specific surface area of TiO2-P-200 was 189 m2/g (pore volume of 0.17 cc/g), meanwhile the specific surface area of the TiO2-P-500 sample was strongly diminished to 60 m2/g (pore volume of 0.11 cc/g). XRD Rietveld refinements with the phases formed in the sol-gel TiO2 catalysts can be observed in Figure 1. In the TiO2-P-200 sample, anatase (tetragonal) and brookite (orthorhombic) phases were obtained, while in the TiO2-P-500 sample, anatase, brookite and rutile were observed. In both samples, anatase was found in high proportion (Table 1). Banfield et al. [28, 29] reported the importance of the crystallite size in the titania phase and its correlation with the polymorphic phase transformation of the TiO2 nanoparticles. Although the anatase and rutile phases have been widely studied, the properties and applications of the brookite phase is comparatively new and especially when they are synthesized as nanoparticles [30]. The formation of the anatase phase usually occurs in titania samples thermally treated at around 450 ˚C, and then the co-existence of anatase and brookite phases can be obtained at annealing temperatures below 450 °C [31, 32]. Table 1 Textural and morphological properties of the sol-gel TiO2 catalysts. Catalysts Cat

TiO2- P-200 TiO2- P-500 TiO2-P25

BET (m2 g-1)

189 60 55

BJH (cc g-1)

0.17 0.11 -

% Phases XDR*

Particle size (nm)

A

B

R

63 83 70

37 15 -

2 30

XRD*

TEM

7.0 22.0 -

7.0 17 30

Eg (eV)

NO Conversion (%)

3.95 3.15 3.2

89 80 50

* XRD-Rietveld refinements A= anatase; B= brookite; R= rutile

The TiO2-P-200 sample showed an anatase crystallite size of 7 nm, which indicated that the nanocrystalline phase structure could be obtained at low temperature (200 ºC). The anatase/brookite and crystallite size/phase composition ratios have been specifically reported for sol-gel iron-doped TiO2 solids [33]. The broad peaks and the unstable base line in Figure 1 suggest the presence of amorphous TiO2, which has not been calculated. On the other hand, for the TiO2-P-500 sample, the anatase, rutile and brookite phases coexist. The anatase phase is relatively abundant (83 wt.%) with a crystallite size of 17 nm. The simultaneous formation of the three phases was certainly an effect of the synthesis reaction conditions, which controlled the titania grain size as it has been reported by several authors [34-36].

Advanced Materials Research Vol. 132

99

Figure 1 Rietveld refinement plots for the TiO2-P-200 sample. The upper marks correspond to anatase (A), and the lower tick marks correspond to brookite (B). As for the TiO2-P-500 sample, the upper tick marks correspond to anatase (A), the middle tick marks correspond to brookite (B) and the lower tick marks correspond to rutile (R). 3.2 TEM and HRTEM analysis

45 40

Average crystal size = 7.52 nm

35 30 25 20 15 10 5 0 4

5

6

7

8

9

10

11

12

Crystal size (nm)

25 Average crystal size = 22.32 nm

20

15

10

5

0 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Crystal size (nm)

Figure 2. TEM micrographs and its average crystal size distribution for: A) TiO2-P-200 and B) TiO2-P-500 sol-gel photo catalysts. The HRTEM micrograph shows that the sol-gel-TiO2-P-200 catalyst exhibits a small particle size around 7 nm, meanwhile the sol-gel-TiO2-P-500 sample showed an increased crystallite size

100

Advances in New Catalytic Materials

(22 nm), (Figure 2). These results are in good agreement with those obtained by using XRDRietveld Refinement (7 and 17 nm), (Table1). On the other hand, by means of HRTEM, the anatase and brookite phases in the TiO2-P-200 sample were identified; as for the TiO2-P-500 sample, the coexistence of the anatase, rutile and brookite phases was confirmed (Figure 3).

Figure 3 Micrograph phases details and the corresponding diffraction patterns for: I) TiO2-P-200 and II) TiO2-P-500 samples.

3.3 Band gap and FTIR analyses The band gap energy (Eg) of the solids was calculated from the UV absorption spectra, and the Eg values for the TiO2-P-200 and TiO2-P-500 photocatalysts were 3.95 and 3.15 eV, respectively (Figure 4); according to the aforementioned, the samples should present different photocatalytic properties [37]. The fit-plot-yield-band-gap values could be also related to quantum-size effects through the TiO2 crystalline phases (anatase-brookite) and particle size, since it has been reported that the band gap of a crystalline semiconductor depends on the particle size [38-41]. The study of the dehydroxylation of the samples as a function of the annealing temperature was followed by FTIR, where a shoulder at 3648 cm-1 due to OH bending vibrations can be seen for the TiO2-P-200 catalyst; likewise, the broad band centered at around 3205 cm-1 showed that the catalyst was highly

Advanced Materials Research Vol. 132

101

hydroxylated at this temperature (Ti-OH). The absorption bands observed at around 2930 cm-1 can be related to shallow electron-trapping states, which is of great interest for photocatalytic purposes [41]. The bands at 1620 and 1610 cm-1 were assigned to the molecular water bending mode [42]. These absorption bands could not evidently be observed for the TiO2-P-500 samples since at 400 °C, total dehydroxylation occurs (Figure 5).

0.8 0.7

2 2 (hv) α

0.6 0.5

TiO2-P-500 3.15 eV

0.4 0.3 0.2

TiO2-P-200

3.95 eV

0.1 0.0 3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

Time (min) Figure 4 Plot of (hv)2 α2 versus photon energy for the TiO2-P-200 and TiO2-P-500 sol-gel catalysts. 3.4 Photocatalytic activity The photocatalytic oxidation of NO at room temperature showed that the TiO2-P-200 sample was the most active catalyst. It oxidized 89 % of NO in 60 minutes, meanwhile the TiO2-P-500 catalyst oxidized 80 %, and as for the commercial catalyst, the conversion was of 50 % (Figure 6). Important differences in textural and morphologic properties were identified in the sol-gel samples annealed at 200 and 500°C. However, the high photocatalytic activity showed by TiO2-P-200 is related to the high specific surface area, nanocrystallite size, higher amount of OH (hydroxylation), and its special properties on crystalline titania phases, and Eg band gap, since in this catalyst, the anatase-brookite phases coexist and it contains the highest amount of the brookite phase (37.2 %) and showed a strong change in Eg (3.95 eV) [43, 44]. On the other hand, the TiO2-P-500 photocatalyst showed a diminution in the photocatalytic oxidation of NO; this behavior could be related to its lower surface area, increased nanocrystallite size and lower hydroxylation, likewise, with its different crystalline phases.

102

Advances in New Catalytic Materials

Figure 5 FTIR spectra for the TiO2-P-200 catalyst annealed at different temperatures.

Figure 6. NO photo-oxidation as a function of irradiation time for the sol-gel TiO2 photo catalysts.

Advanced Materials Research Vol. 132

103

4 Conclusions By varying the sol-gel parameters and controlling the annealing temperature, it is possible to obtain nanocrystalline TiO2 with high surface area and higher amount of OH (hydroxylation), likewise, the coexistence of the anatase–brookite phases can be obtained. Likewise, the phsicochemical properties improve the photocatalytic efficiency in the NOx photo-oxidation at room temperature, and as a result, they could be used to abate pollution in both indoor and outdoor environments at room temperature. Acknowledgments We are indebted to the IMP-Molecular Engineering Program for its financial support. The authors want to thank Technician Rufino Velázquez Lara for his assistance and technical support in this work. References [1] N Stock, J Séller, K Vinodgal, P Kamat: Environ. Sci. Techmol. Vol. 34 (2000), p,1747. [2] J Lee, Y Yang: Mater. Chem. Phys. Vol 93 (2005), p. 237. [3] M Takeuchi, T Kimura, M Hidaka, M Rakhmawaty, M Anpo: J. Catal. Vol. 246 (2007), p. 235. [4] K Nishijima, B Ohtani, X Yan, T Kamai, T Chiyoya, T Tsubota, N Murakami, T Ohno: Chem, Phys. Vol. 339 (2007), p.64. [5] F Shiraishi, S Yamaguchi, Y Ohbuchi: Chem. Eng. Sci. Vol. 58 (2003), p.929. [6] F Shiraishi, D Ohkubo, K Toyoda, S Yamaguchi: Chem. Eng. Sci. Vol. 114 (2005), p.159. [7] U Diebold: Surf. Sci. Reports Vol. 48 (2001), p.53. [8] I Arabatzis, S Antonaraki, T Stergiopoulos, A Hiskia, E Papaconstantinou, M Bernard, P Falaras: J. Photochem. Photobiol. A Vol. 149 (2002), p.237. [9] J Zhao, X Yang: Building and Environment Vol. 38 (2003), p.645. [10] C Ao, S Lee, C Mak, L Chan: Appl. Catal. B Vol. 42 (2003), p.119. [11] S Devahasdin, C Fan, K Li, D Chen: J. Photochem. Photobiol. A Vol.156 (2003), p.161. [12] T Maggos, J Bartzis, M Liakou, C Gobin: J. Hazardous Materials Vol. 146 (2007), p.668. [13] T Lim, S Jeon, S Kim, J Gyenis: J. Photochem. Photobiol. A Vol. 134 (2000), p. 209. [14] H Ibrahim, H Lasa: Appl. Catal. B Vol. 38 (2002), p.201. [15] D Uhlman, G Teowe: J. Sol Gel Sci. Thecnol. Vol.13 (1998), p.153. [16] Y Hu, H Tsai, C Huang: J. Euro. Ceram. Soc. Vol. 23 (2003), p.691. [17] I Kuznetsova, V Blaskov, L Znaidi: Mater. Sci. Eng. B Vol.137 (2007), p.31. [18] R Bhave, B Lee: Mater. Sci. Eng. A Vol. 467 (2007), p.146. [19] D Bahnemann: Res. Chem. Intermed. Vol. 26 (2000), p.207. [20] Rodríguez-Carbajal: J. Phys. B Vol. 192 (1993), p.55. [21] X Orlhac, C Fillet, P Deniard, A Dulac, R Brec: Appl. Cryst. Vol. 34 (2001), p.114. [22] L Fuentes (1998) Análisis de minerales y el método de Rietveld, Sociedad Mexicana de Cristalografía, A.C., Mexico. [23] S Castillo, M Morán-Pineda, V Molina, R Gómez, T López: Appl. Catal. B Vol. 15 (1998), p. 203. [24] E Sánchez, T López: Mat. Lett. Vol. 25 (1995), p.271. [25] K Reddy, S Manorama, A Reddy: Mat. Chem. Phys. Vol. 78 (2002), p.239. [26] C Sanchez, J Livage, M Henry, M Babonneau: J. Non-Cryst- Solids Vol.100 (1988), p. 65. [27] H Q Wang, Z B Wu, W R Zhao, B H Guan: Chemosfere Vol. 66 (2007), p.185. [28] H Zhang, J Banfield: J. Mater. Chem. Vol. 8 (1998), p.2073. [29] H Zhang, J Banfield: J. Phys. Chem. B Vol. 104 (2000), p.3481. [30] K Zhu, M Zhang, J Hong, Z Yin: Mater. Sci. Eng. A Vol. 403 (2005), p.87. [31] J Ovenstone: J. Mater Sci. Vol. 36 (2001), p.1325. [32] C Radhica, B Bhave, I Lee: Mater. Sci. Eng. A Vol. 467 (2007), p.146.

104

Advances in New Catalytic Materials

[33] J A Wang, R Lima-Ballesteros, T López, A Moreno, R Gómez, O Novaro, X Bokhimi: J Phys. Chem. B Vol. 105 (2001), p. 9692. [34] K Okada, N Yamamoto, Y Kameshima, A Yasumori: J. Am. Ceram. Soc. Vol. 84 (2001), p. 1591. [35] X Bokhimi, A Morales, O Novaro, T López, E Sánchez, R Gómez: J. Mater. Res. Vol. 10 (1995), p. 2788. [36] Y Sun, T Egawa, L Zhang, X Yao: Jpn. J. Appl. Phys. 41 (2002), p. L945. [37] K M Reddy, C V Gopal Reddy, S V Manorama: J. Solid State Chem.Vol. 158 (2001), p.180. [38] T Tora, K Hiroshi, S Ping, K Akihik, O Masahiro: Jpn. J. Appl. Phys. 5B 39 (2000), p.3160. [39] L Bras: J. Chem. 90 (1986), p.2555. [40] J Rino, N Studart: Phys. Rev. B Vol. 59 (1999), p.6643. [41] K Kanaca, J White: J. Phys. Chem. 86 (1982), p.4708. [42] M Ampo, Y Ichihashi, M Takeuchi, H Yamashita: Res. Chem. Intermed. Vol. 24 (1998), p. 143. [43] U Diebold: Surf. Sci. Reports Vol. 48 (2003), p. 53. [44] I Kuznetsova, V Blaskov, L Znaidi: Mater. Sci. Eng. B Vol. 137 (2007), p.31.

Advanced Materials Research Vol. 132 (2010) pp 105-110 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.105

Chitin/TiO2 Composite for Photocatalytic Degradation of Phenol Kun Wan a, Xianghong Peng b*, Pingjing Duc School of Chemistry and Environmental Engineering, Jianghan University, Wuhan 430056, China a

E-mail: [email protected],bE-mail: [email protected], ﹡correspondence author c

E-mail: [email protected]

Keyword: TiO2, Chitin, Photocatalysis, Phenol, Sol-gel

Abstract: Chitin/TiO2 composite was prepared through colloid TiO2 deposited on the chitin by controlling the pH value of the system, while colloid TiO2 was synthesized by the sol–gel method using tetrabutyl titanate as a precursor. The structures and morphologies of the chitin/TiO2 composite were characterized by FT-IR, X-ray diffraction (XRD) and scanning electron microscopy (SEM). The photocatalytic degradation of phenol was investigated by HPLC method. The results revealed that the chitin/ TiO2 composite was an efficient photocatalyst for the degradation of phenol, and 99.2% of the phenol was degraded after 6h under UV light. The TiO2 was adsorbed on the chitin by hydrogen and titanoxane bonds between them. Colloid TiO2 was gradually deposited to form the anatase crystallographic structures, showing 2θ = 25.3°, 37.8°, 47.8° and 54.6°. Such biocompatible photocatalyst might be applied in the field of various phenol pollutants abatement.

1. Introduction Recently, TiO2 photocatalyst has attracted great attention because of high efficiency, photochemical stability, non-toxic, nature and low cost. Nanometer TiO2 has proved to be an excellent catalyst in the photo degradation of organic pollutants [1-3]. However, with TiO2 powder some limitations exist in recycling and reuse at industrial scale. Immobilized TiO2 on solid media can dissolve the problem mentioned above [4, 5]. Some techniques have been developed for immobilizing TiO2 onto a solid substrate, such as dip coating from suspension, spray coating, sputtering, sol-gel-related methods, and electrophoretic deposition. The substrates include glass beads and tubes, fiberglass, quartz, stainless steel, activated carbon, silica and perlite [6-11]. In the 21st century, science and technology have moved toward renewable raw materials and more environmentally friendly and sustainable resources and processes. The natural biopolymer chitin is the second most abundant polysaccharide in the world after cellulose. Chitosan is the derivative of chitin. Chitin and chitosan have been found wide applications in the field of wastewater treatment, for example, the removal of dye and heavy metals [12-14]. TiO2 have been immobilized on chitosan to form the photocatalyst for degradation of various organic pollutants [15, 16]. However, most of the chitosan/TiO2 composite is prepared by TiO2 powder dispersed on chitosan. There exists some disadvantage of TiO2 aggregation. In present, phenol has become the most abundant pollutant in industrial wastewater, because it is widely used in the synthesis of resins, dyes, pharmaceuticals, perfumes, pesticides, tanning agents, solvents and lubricating oils [17]. The phenol wastewater needs to be degraded to an acceptable level before it is discharged to

106

Advances in New Catalytic Materials

the environment. As reported, nanometer TiO2 can effectively degrade phenol to safe end products, such as CO2, H2O, and mineral acids [18, 19]. Some researchers are devoted to enhancing the photocatalytic activity of TiO2 by doping with Pt, Au, Cu metal ions [20-22]. In addition, colloid TiO2 synthesized by the sol-gel method has high photocatalytic activity because of its large surface area, high surface hydroxyl content, and lower band gap [23]. In the present paper, the colloid TiO2 was synthesized by the sol-gel method using tetrabutyl titanate as a precursor, and then immobilized on the chitin by controlling the pH value in the system. Using the phenol solution as a model, the photocatalysis of chitin/ TiO2 composite was investigated. This biocompatible and low-cost material might be used as a photocatalysts for the environmental applications. 2. Experimental Section 2.1. Materials Chitin was purchased from Yuhuan Ocean Biochemistry Co. Ltd., in Zhejiang, China. Tetrabutyl titanate (TBT) was chemical grade and obtained from Shanghai Chemical Reagent Co. in Shanghai. The other chemical reagents were commercially obtained from China and of analytical grade. 2.2 Sample preparation The colloid TiO2 was prepared by the sol-gel method. 4.3 mL of tetrabutyl titanate was mixed with 25 mL of alcohol, and then dropping into 500 mL 1% (V/V) HNO3 aqueous solution with 0.5 drops per second at stirring condition to obtain the TiO2 colloid solution. 50.0 g amount chitin powder and 18.0 g amount urea was added into the above TiO2 solution, and stirred for 20h. The temperature was changed from 60 to 80 oC, and the pH value was raised from 1.0 to 7.0 in the process. The mixture product was treated by centrifugation, and then the white precipitate was collected and washed with distilled water and dried at 80 oC. 2.3 Photocatalytic activity studies The photocatalytic activity of the chitin/TiO2 composite was evaluated according to the photodegradation of phenol aqueous solutions under different irradiation time. 100 mL of 100 mg/L phenol aqueous solution and composite containing 0.1 g TiO2 were loaded in a glass container, and stirred with a magnetic stirrer under the irradiation of 400 W high pressure mercury UV lamps (Philips Yaming Co. Ltd., Shanghai, China). The main emission of the lamp was 365 nm. The distance between the UV lamp and the surface of the solute was set at 10 cm. At the desired irradiation time interval, 1.5 mL portions of the suspension were taken out and the concentration of phenol was further analyzed by high performance liquid chromatography (HPLC) (DIONEX) equipped with a diode array detector (UVD 170U). The reaction products were separated on a reverse-phase C18 (Diamonsil), 3 µm, 4.6×250 mm, using a mixture H2O/CH3OH/CH3CN = 40/30/30 (v/v) as eluent at a flow rate of 1 mL/min. The UV absorbance detector was set at 270 nm for phenol. The suspensions were filtered through 0.45 µm hydrophilic membranes. Then second distilled water was added to maintain a constant volume during the process. The total irradiation time was 6 h in the experiment. 2.4 Characterization To clarify the structure of chitin, chitin/TiO2 composite, Fourier transform infrared spectroscopy (FTIR) was recorded on a Nicolet 5700 spectrometer (Nicolet, Minnesota) in a range from 4000 to 400 cm-1 using a KBr-pellet method. The chitin/TiO2 composite was calcined at 500 oC for 2 h, and carried out on an X-ray diffractometer (D/MAX, RIGAKU, Japan) with a Cu Ka radiation source at

Advanced Materials Research Vol. 132

107

40 kV and 30 mA. The diffraction angles ranged from 10o to 90o. The morphologies of the microspheres were observed by a scanning electron microscope (SEM, S-570, Hitachi, Japan). The samples were coated with gold for SEM observation. 3. Results and Discussion 3.1. Structure of chitin/ TiO2 composite The FT-IR spectra of the chitin and chitin /TiO2 composite are shown in Figure 1. The peaks at 1627 cm-1, 699 cm-1 and 572 cm-1 of the composite belong to the Ti–O structures [1, 15]. The peaks at 1666 cm−1 and 1071 cm−1 are assigned the bonds of the C=O imide group and Ti-OH in the composite, respectively [24]. The characteristic peak of -OH at 3438 cm−1 is shifted to a lower wave number at 3433 cm−1 in the composite because of the hydrogen bonds between chitin and TiO2. The bands from 2877 to 2927 cm−1 are attributed to the symmetric stretching of -CH2- of the composite [25]. The above results indicate that Ti–O–Ti bonds have been formed in the composite, suggesting that the TiO2 can be adsorbed on the chitin by the hydrogen and titanoxane bonds between them.

chitin/TiO2

2972

699

2877

572 1666 1627

chitin 3433

1071

2927 1655

3438

4000

3500

3000

2500

2000

1500

1000

500

-1

Wavenumbers/cm

Fig. 1, FT-IR spectra of chitin and chitin/TiO2 composite Figure 2 shows the X-ray diffraction pattern of chitin/TiO2 composite after calcination at 500 °C. The peaks at 2θ = 25.3°, 37.8°, 47.8° and 54.6° can be indexed to the (101), (103), (004), (200), (105) and (211) crystal faces of anatase TiO2 [26]. As reported, TiO2 was amorphous in the chitosan/TiO2 hybrid film, in which TiO2 was synthesized by the sol-gel method [22]. It is noted that anatase phase has been formed by deposited colloid TiO2 gradually in this work.

3.2. Morphology of chitin/ TiO2 composite The SEM images of the chitin/ TiO2 composite are shown in Figure 3. Small TiO2 particles are adsorbed on the high porosity chitin surface, as shown in Figure 3a. The microporous chitin is a good support for TiO2. Figure 3b exhibits that the TiO2 crystal is multilayer and has porous structure, suggesting that the crystal structure TiO2 is gradually formed. As reported, the progressive decomposition of urea in solution at temperature above 60 oC releases OH – ions, which gradually increase the medium pH [27]. Urea is used as a hydrolysis agent to control the pH value of the system to induce TiO2 slow precipitation onto the chitin support in this work. The formation of

108

Advances in New Catalytic Materials

TiO2 crystal is described as follows. The colloid TiO2 is adsorbed on the surface of chitin to form the small TiO2 particles, and then other colloid TiO2 in the solution is gradually deposited on the surface of small TiO2 particles as the pH value of solution increases. The progress continues to form the anatase crystallization. Such TiO2 owing large surface area and pore volume is beneficial to the degradation of the phenol. As reported, the porous TiO2-chitosan materials are used to remove the active dyes [28]. Therefore, the chitin/ TiO2 composite could be used to treat the complex wastewater because of its photocatalysis and adsorption properties of chitin.

20

40

60

80

2θ/(degree)

Fig. 2, X-ray diffraction pattern of chitin/TiO2 composite calcined at 500 oC for 2 h

Fig. 3, SEM images of chitin/ TiO2 composite 3.3 Photocatalytic Degradation of Phenol The result of the photocatalytic degradation of phenol of chitin/ TiO2 composite is shown in Figure 4. After 6 h of UV light irradiation, the concentration of phenol is decreased from100 mg/L to 0.08 mg/L, and 99.2% of phenol has been degraded. It is well known that most of the photocatalysts TiO2 prepared by the sol-gel method need to be calcined to form the anatase to enhance the photocatalytic activity. It is worth noting that the TiO2 synthesized in this paper have high

Advanced Materials Research Vol. 132

109

photocatalytic activity. For the first time, the colloid TiO2 has been immobilized on the biocompatible chitin substrate to obtain the high effective photocatalysis. Such composite might be a good candidate catalyst for industrial-scale treatment the complex phenol wastewater because of its photocatalytic properties and low price.

phenol concentration(mg/L)

100 80 60 40 20 0 1

2

3

4

5

6

time / h

Fig. 4, Degradation of phenol under UV light. 4. Conclusion Chitin/TiO2 composite was successfully prepared by colloid TiO2 deposited gradually on the chitin through controlling the pH value of solution. The chitin/TiO2 composite exhibited efficient photocatalytic degradation for phenol: 99.2 % of phenol was degraded after 6h of UV light. The TiO2 on the chitin shows a multilayer structure having crystallized anatase phase. Such photocatalyst having biocompatibility and low-cost could find broad application in the field of phenol pollution treatment.

Acknowledgment. This work was supported by the Foundation of Science and Technology Bureau of Wuhan (200751699478-06), the Education Bureau of Wuhan (2008K040) and Education Bureau of Hubei Province (B20093404).

References [1] H. Li, J. Li and Y. Huo: J. Phys. Chem. B Vol. 110 (2006), p. 1559 [2] B. Iurascu, I. Siminiceanu, D. Vione, M.A. Vicente and A. Gil: Water Res. Vol. 43 (2009), p.1313 [3] H. Yu, X. Li, X. Quan, S. Chen and Y. Zhang: Environ. Sci. Technol. Vol. 43(2009), p. 7849 [4] H. Al-Ekabi, N. Serpone, E. Pelizzetti, C. Minero, M. A. Fox and R. B. Draper: Langmuir Vol. 5(1989), p. 250

110

Advances in New Catalytic Materials

[5] J. Sabate, M. A. Anderson, H. Kikkawa, M. Edwards and C. G. Hill: J. Catal. Vol. 127(1991), p. 167 [6] J.C. Lee, M.S. Kim and B.W. Kim, Water Res. Vol. 36 (2002), p. 1776 [7] I.N. Martyanov and K.J. Klabunde: J. Catal. Vol. 225 (2004), p. 408 [8] J. Shang, W. Li and Y. Zhu: J. Mol. Catal. A Vol. 202 (2003), p. 187 [10] M. S. Vohra and K. Tanaka, Water Res. Vol. 37 (2003), p3992 [11] S. N. Hosseini, S. M. Borghei, M. Vossoughi and N. Taghavinia: Applied Catalysis B: Environmental Vol.74 (2007), p. 53 [12] G. Crini and P.M. Badot: Prog. Polym. Sci. Vol.33 (2008), p. 399 [13] K.D.Trimukhe and A.J. Varma: Carbohydr. Polym. Vol.71 (2008), p. 66 [14]N. Sankararamakrishnan, A. K. Sharma and R. Sanghi: J. Hazardous Materials Vol.148 (2007), p. 353 [15] Q. Li, H. Su and T. Tan: Biochemical Engineering Journal Vol. 38 (2008), p. 212 [16] Z. Zainal, L.K. Hui, M. Z. Hussein, A. H. Abdullah and I. R. Hamadneh: Journal of Hazardous Materials 164 (2009), p. 138 [17] S. Budavari, M.J. O’Neil, A. Smith, P.E. Heckelman and J.F. Kinneary (Eds.), The Merck Index, 12th ed. Merck and Co., New Jersey, 1996 [18] R.W. Matthews: J. Phys. Chem. Vol. 91 (1987), p. 3328 [19] L. Liu, H. Liu, Y. Zhao, Y. Wang, Y. Duan, G. Gao, M. Ge and W. Chen: Environ. Sci. Technol. 42(2008), p. 2342 [20] R. Vinu and G. Madras: Environ. Sci. Technol. Vol. 42(2008), p. 913 [21] E. Kowalska, H. Remita, C. Colbeau-Justin, J. Hupka and J. Belloni: J. Phys. Chem. C Vol.112 (2008), p. 1124 [22] R.S. Sonawane and M.K. Dongare: Journal of Molecular Catalysis A: Chemical Vol.243 (2006), p68 [24] K. Nagaveni, M. S. Hegde, N. Ravishankar, G. N. Subanna and G. Madras: Langmuir Vol. 20 (2004), p. 2900 [24] Y. Tao, J. Pan, S. Yan, B. Tang and L. Zhu: Materials Science and Engineering B Vol.138 (2007), p. 84 [25] X. Peng and L. Zhang: Langmuir Vol. 21(2005), p. 1091 [26] Z. Liu, X. Zhang, S. Nishimoto, M. Jin, D. A. Tryk, T. Murakami and A. Fujishima: J. Phys. Chem. C Vol.112 (2008), p. 253 [27] L.H. Kao, T.C. Hsu , H.Y. Lu: Journal of Colloid and Interface Science Vol. 316 (2007), p.160 [28] C. E. Zubieta, P. V. Messina, C. Luengo, M. Dennehy, O. Pieroni and P. C. Schulz: J. Hazardous Materials Vol. 152 (2008), p. 765

Advanced Materials Research Vol. 132 (2010) pp 111-125 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.111

Catalytic Properties of Ni-Mo Carbide and Nitride Phases Supported on SBA-15 and -16 in the Hydrodesulfurization of DBT Aguillón-Martínez Edgar Caleb1, Melo-Banda José Aarón1,a, Guevara L. A2,b, Reyes T. A. 1, Ramos Galván. C. E.1, Silva R. R.1, Domínguez J. M.3,c 1

Instituto Tecnológico de Ciudad Madero, División de Estudios de Posgrado e Investigación, Juventino Rosas y Jesús Urueta S/N, Col. Los Mangos C.P. 89440, Cd. Madero, Tamps., México. 2

Universidad Autónoma del Estado de Hidalgo, Centro de Investigaciones Químicas, Carretera Pachuca-Tulancingo km. 4.5, C.P. 42184, Pachuca Hgo., México 3

Instituto Mexicano del Petróleo, Programa de Ingeniería Molecular, Eje central L. Cárdenas No. 152 México D.F., México

a

[email protected], [email protected], [email protected]

Keywords: Carbides, Nitrides, SBA-15, SBA-16, Hydrodesulphurization, Dibenzothiophene, Mesoporous.

Abstract. In this paper, the incorporation of carbide and nitride phases supported on mesoporous materials like SBA type is described in order to obtain a better material than commercial catalysts (NiMoS/Al2O3, Sg=269 m2.gr-1) specifically in the HDS of dibenzothiophene. The XRD patterns exhibit the presence of carbide and nitride phases in each series, respectively. In the nitride materials, the presence of oxide phases was more evident than in the carbide catalysts. The principal product was biphenyl (BF) for all the analyzed series. This behavior suggests that the DBT desulphurization pathway for the carbide materials was similar to that of sulfide catalysts. Bicyclohexyl (BCH) was analyzed as a product, and cyclohexylbenzene traces (CHB) were determined in a single catalyst (NiMoC-2%P/SBA-16). This was attributed to the hydrogenation character of carbide catalyst reported previously. 1. Introduction. The present problems of environmental pollution have increased the toxic gases in the air, partly generated by the increment in the concentration of polluting agents, and also by the combustion of heavy oil fractions, which have caused great challenges in the crude oil industry. At the same time, strict environmental regulations for decreasing sulfur, nitrogen and aromatic compound contents in fuels have been described. In this way, the oil industry has been focused on the hydrotreatment process (HDT) as a way to remove these heterocompunds, specifically, the hydrodesulfurization process (HDS). This has motivated the necessity of developing new catalytic materials with better catalytic properties and capacity to withstand harsh conditions in the process. Mesoporous materials with higher pore sizes and better stability such as M 41S materials were synthesized by Zhao et al. [1]. These mesoporous silica materials called SBA-n show pore sizes close to 30 nm with high wall thicknesses, which makes them resistant to certain hydrothermal conditions. These materials also present a narrow pore size distribution in the mesoporous range (2-30 nm), high ordered structure, surface areas near to 1000 m2·g-1 and a highly active surface [2-3]. Moreover, ceramic compound such as transition metal carbides and nitrides with high surface areas have been tested in hydrotreatment processes due to their mechanical resistance and potential catalytic activity [4]. The Mo carbides have been tested in HDN, HDS and HDO. These materials exhibit higher activity than a commercial sulfide catalyst of Co-Mo/Al2O3 [5]. P. Da Costa et al. found that the unsupported Ni-Mo carbides promoted with P were more active than an industrial catalyst of Co-Mo/Al2O3 [6]. They have also demonstrated the potential of the Mo and W carbides as great hydrogenation materials in the presence of sulfur compound mixtures [7]. Nagai et al. researched on the HDS of dibenzothiophene (DBT) with Mo2N supported on alumina; they found that the nitride phase was

112

Advances in New Catalytic Materials

1.1-1.2 times more active than the Mo/Al2O3 sulfide catalyst with less hydrogen consumption [8]. The nitride catalysts after HDS reaction were characterized by many studies [9-12]; and Gong et al. showed that the γ-Mo2N crystalline phase was present after the HDS reaction; nevertheless, an inevitable partial sulfurization happened during the test [12]. Melo-Banda et al. analyzed the catalytic properties of NiMo carbide and nitride catalysts employed in the HDS reaction of heavy vacuum gas oil; the final activity was attributed to the presence of superficial MoS or MoS/MoC mixtures [13]; whereas, the catalytic activity of the carbide and nitride phases of Ni, Co and Mo seem to be intimately related to their structure and the crystal size. In this way, Santillán et al. studied the effect of phosphorus content on Ni-Mo carbide and nitride phases supported on alumina. Their results indicate that phosphorus allowed the metal dispersion (i.e. Ni, Co, Mo), increasing the formation of carbide and nitride phases as a consequence of the original oxide species [9]. In 1980 Houalla M et al.[14] reported a pseudo first order kinetics and the HDS reaction mechanism for DBT. Figure 1 shows the reaction mechanism of the DBT hydrodesulfurization. Currently, this mechanism is the most accepted and it has been corroborated by further research works; other proposed mechanisms are only simplifications of this reaction scheme. This mechanism presents two reaction routes, sequential and parallel; in addition, the mechanism indicates that the conversion takes place through the less hydrogen consuming pattern.

Fig. 1. Reaction pathways proposed by Houalla for the hydrodesulfurization of DBT [14]. The first one implies the rupture of the C-S bonds, keeping the aromatic rings and producing biphenyl (BF); this reaction route is frequently known as direct desulfurization (DDS) or hydrogenolysis. Once the BF is produced, one of the aromatic rings is hydrogenated to produce cyclohexylbenzene (CHB); this last step is very slow. The second route is the direct hydrogenation (DHY), where the heteroatom is kept, but one of the aromatic rings is hydrogenated, promoting the formation of partially hydrogenated compounds such as the tetrahydrobibenzothiophene (THDBT) and hexahydrodibenzothiophene (HHDBT); because of the instability of these compounds, they react quickly through the rupture of the C-S bond and a second hydrogenation to produce CHB takes place, where the two reaction routes converge; finally, the second aromatic ring is hydrogenated to produce bicyclohexyl (BCH). In this work, supported Ni-Mo carbides and nitrides were synthesized in mesoporous structures of SBA-15 and 16 types to compare the catalytic activity between two different mesoporous

Advanced Materials Research Vol. 132

113

arrangements and a commercial alumina support; furthermore, the phosphorous content (from 1 to 2 wt % P) was analyzed in both mesoporous materials and in the two different phases in order to observe the synergism between the phases. The catalytic properties of the materials were tested in the hydrodesulfurization of DBT used as a model molecule at 573 K and P= 30 bar for 8 hours. 2. Experimental 2. 1 Preparation of supports and catalysts SBA -15 and -16-type materials were synthesized by the hydrothermal method according to Zhao et al. [1]; other detailed procedures are also reported in the literature [15-17]. Hexagonal SBA-15 silica structure p6mm was synthesized using a polymer in triblock copolymer Pluronic P123 (EO20PO70-EO20, BASF) as the structure-directing agent, and tetraethyl orthosilicate (TEOS, Chemical Aldrich Co., 99.9 %) as silica source. Pluronic P123 was dissolved in 105 ml of deionized water and 20 ml of HCl (Aldrich Chemical, 37 % vol.) at 318 K. Then, TEOS (9.1 g) was added dropwise to the solution. The mixture was stirred at 318 K for 24 h and later, it was placed in polyethylene boats for 24 hours at 363 K. The final solid was filtered, washed (with deionized water) and finally, it was dried at 363 K overnight. The cubic SBA-16 silica structure Im3m was synthesized by means of a similar procedure. Pluronic F127 (EO106-PO70-EO106, BASF) and NaCl (7 g) were dissolved in 80 ml of HCl (0,5 M). Then, TEOS (8.4 g) was added dropwise to the final solution. The final mixture was stirred for 20 h at 313 K, and later, it was aged in polyethylene boats at 373 K for 24 h. The solid was filtered, washed with deionized water and dried at 363 K overnight. Both solids were calcined under dynamic air atmosphere at 823 K for 6 h. The metallic oxide precursors were obtained by means of the incipient wetness technique, using Ni(NO3)2 (Chemical Aldrich Co.,99.99 %) and (NH4)6Mo7O24 4H2O (Chemical Aldrich Co.,99.5 %) with an atomic ratio Ni/(Ni+Mo)= 0.3, (2.8 Mo atoms/nm2, i. e., Mo~15-18 wt.% and Ni ~ 711 wt % loads, where the surface area of each initial support, before reduction treatments, was used to calculate the metal contents (according to a specific atomic ratio); both materials were successively incorporated using ethanol as solvent; afterward the materials were dried at 393 K overnight. Phosphorus was incorporated using the same route with (NH4)2HPO4 (Chemical Aldrich Co., 99 %) in order to obtain 1, 1.5 and 2 wt % P loads. Later, the solids were dried at 393 K; afterward, the final material was calcined at 723 K. These materials were treated with reducing agents such as an anhydrous combination of 1 part of hydrogen (H2) diluted in 2 parts of methane (CH4), i.e., CH4/H2 = 2/1, which is a reduction procedure that has been reported elsewhere [9]. After reduction, all the solids loaded with metals were passivated under a diluted flow of oxygen (O2:1% vol.) combined with helium (He) in order to avoid spontaneous combustion in air. The synthesis of Ni-Mo supported nitrides was carried out according to the procedure reported by Afanasiev and Wang [18-19]. The nitride catalysts were prepared in the same way using the incipient wetness technique; nickel and molybdenum were successively impregnated; after, hexamethylenetetramine ((CH2)6N4, Aldrich Chemical, 99.7%) was added by the same route. All the samples were dried at room temperature; and later, the temperature was increased at 393 K for 6 h. A similar procedure was used for the nitride phase, using dynamic argon flow as inert atmosphere to promote the reduction effect on the mixed metallic oxide with stoichiometric hydrogen, nitrogen and carbon contents in the samples. Finally, all the samples were cooled at room temperature and passivated with O2/He flow ( CuCl2 > NiCl2 was seen for that from chlorides and Ni(NO3)2 > Cu(NO3)2 > Co(NO3)2 for that from nitrates. Yang et al. [8] reported that transition metal performed an adsorption by π-complexation with thiophene sulfur-containing compounds. It was also proved in this work that calcination step has powerful effect on the adsorption behaviors of the modified carbons. Fig. 6 gave the DBT adsorption capacities of the calcined precursor-impregnated adsorbents. These six adsorbents were prepared by a combination of impregnation, drying and calcination steps. It was seen the adsorption capacity was reduced in an order of CuCl2 > CoCl2 > NiCl2, when chloride salts were the metal precursors; meanwhile this order was Ni(NO3)2 > Co(NO3)2 > Cu(NO3)2 when transition metal nitrates were the metal precursors. Moreover, it was also seen from Fig. 6, that the DBT adsorption capacities of the calcined copper or cobalt chloride-impregnated carbon-based adsorbents were apparently higher than that of metal nitrates, whilst that of the calcined nickel chloride-impregnated carbon-based adsorbents was lower than that of the absorbents from the respective nitrates. All of the above results elucidated that the calcination step significantly affected on the adsorption behaviors of the adsorbents; it even altered the order of the adsorption capacity in comparison with the dried carbon-based samples shown in Fig. 5. These results were probably correlated to the species of the adsorbents on the carbon substrate. It probably showed that the change of metallic salts caused an effect on DBT adsorption capacity of these materials provoking a change in order of activity, and the calcinations process caused a

Advanced Materials Research Vol. 132

147

change of the geometry of the metallic centre or even particle size, which probably has effects on DBT adsorption. It was observed that on the CuCl2-loaded carbon, calcined at 350 ºC, the copper was dispersed as CuCl and CuCl2 crystals, wherein CuCl content was higher, and for the Cu(NO3)2-loaded carbon, it was mainly dispersed as Cu2O. For the nickel-modified carbon, the nickel state on the carbon substrate was either NiCl2 or NiO as the precursor was NiCl2 or Ni(NO3)2. As for that from cobalt, the metal states were either CoCl2 or Co3O4 and CoO orderly. From our tests for adsorption behaviors, it was clear that CuCl > Cu2O, CoCl2>Co3O4, CoO and NiO>NiCl2. Compared the results in Fig. 5 and Fig. 6, it was observed that for the metal chloride, the adsorption capacities of the calcined adsorbent sample from NiCl2 was higher than that of only being dried, meanwhile the opposite order was true for the CoCl2-modified sample. As for the CuCl2 –modified one, both dried and calcined were similar each other. On the other hand, the calcined samples presented a higher performance for the adsorbents by all of these metal nitrates than that only being dried. These results revealed that a good preparation method to obtain an adsorbent with high adsorption capacity depended greatly on the kinds of the precursor anions. From FTIR characterization shown in Fig. 4, the oxygen-containing groups on the carbon surface after calcination under nitrogen stream increased by virtue of decomposition of nitrate salts that favored to DBT adsorption on the carbon-based adsorbent [23]. Therefore, the adsorption capacity of the adsorbents modified by nitrate salts was enhanced after calcination. As for the sample modified by the chloride precursors, the calcination step must strengthen the synergistic effect between metal ions and the carbon support, resulting in a decrease in the adsorption capacity for Co2+ ions, whilst an increase for Ni2+ ions.

4. Conclusions Among these dried metal-modified carbon-based adsorbents, each of the samples from metal chlorides presented a higher adsorption capacity than that from respective nitrates, wherein CoCl2 gave the highest performance, reaching 36.25 mg S/g. For these carbon-based adsorbents modified with the metal nitrates, the number of oxygen-containing groups on the carbon support surface were increased by virtue of decomposition of nitrate ions after the calcination step under nitrogen stream at 350 ºC, boosting their DBT adsorption capacities compared to the respective dried adsorbents. As for the absorbent from chlorides, the interaction might play a role in reducing the DBT adsorption capacity of cobalt ions or increasing this capacity for nickel ions. For the adsorption behaviors for DBT, an order was observed in varying metal species CuCl>Cu2O, CoCl2>Co3O4 and CoO, whilst NiO>NiCl2. Therefore, our experimental results show that both the transition metal precursors and the carbon surface metal species and kinds have significant effects on DBT adsorption capacity of the transition metal loaded activated carbon adsorbents.

Acknowledgements L F. Chen thanks the financial support from Instituto Politecnioco Nacional, Mexico (Grant No. SIP-20100733).

148

Advances in New Catalytic Materials

References [1] C. Song: Catal. Today Vol. 86 (2003), p. 211 [2] A.J. Hernandez-Maldonado, S.D. Stamatis, R.T. Yang, A.Z. He, W. Cannella: Ind. Eng. Chem. Res. Vol. 43 (2004), p. 769 [3] A.J. Hernandez-Maldonado, R.T. Yang: Ind. Eng. Chem. Res. Vol. 42 (2003), p. 3103 [4] A.J. Hernandez-Maldonado, R.T. Yang: AIChE J. Vol. 50 (2004), p. 791 [5] A.J. Hernandez-Maldonado, R.T. Yang, Ind. Eng. Chem. Res. Vol. 42 (2003), p. 123 [6] S.G. McKinley, R.J. Angelici, Chem. Commun. Vol. (2003), p. 2620 [7] S. Reut, A. Prakash: Fuel Proc. Tech. Vol. 87 (2006), p. 217 [8] R.T. Yang, A.J. Hernandez-Maldonado, F.H. Yang: Science Vol. 301 (2003), p. 79 [9] A.J. Hernandez-Maldonado, S F.H. Yang, G.S. Qi, R.T. Yang: Appl. Catal. B: Environ. Vol. 56 (2005), p. 111 [10] A. Takahashi, F.H. Yang, R.T. Yang: Ind. Eng. Chem. Res. Vol. 41(2002), p. 2487 [11] A.J. Hernandez-Maldonado, S.D. Stamatis, R.T. Yang, A.Z. He, W. Cannella: Ind. Eng. Chem. Res. Vol. 43 (2004), p. 769 [12] X. Ma, M. Sprague, C. Song: Ind. Eng. Chem. Res. Vol. 44 (2005), p. 5768 [13] X. Ma, L. Sun, C. Song: Catal. Today Vol. 77 (2002), p. 107 [14] J. Kim, X. Ma, A. Zhou, C. Song: Catal. Today Vol. 111 (2006), p. 74 [15] Y. Sano, K. Choi, Y. Korai, I. Mochida: Appl. Catal. B Environ. Vol. 49 (2004), p. 219 [16] Y. Sano, K. Choi, Y. Korai, I. Mochida: Energy & Fuels Vol. 18 (2004), p. 644 [17] Y. Sano, K. Sugahara, K. Choi, Y. Korai, I. Mochida: Fuel Vol. 84 (2005), p. 903 [18] M. Yu, Z. Li, Q. Xia, H. Xi, S. Wang: Chem. Eng. J. Vol. 132 (2007), p. 233 [19] V. Selvavathi, V. Chidambaram, A. Meenakshisundaram, B. Sairam, B. Sivasankar: Catalysis Today Vol. 141 (2009), p. 99 [20] M. Seredych, T.J. Bandosz: Energy Fuels, Vol. 23(2009), p, 3737 [21] P.E. Fanning, M.A. Vannice: Carbon Vol. 31(1993), p. 721 [22] J.L. Figueiredo, M.F.R. Pereira, M.M.A Freitas: Carbon Vol. 37 (1999), p. 1379 [23] A. Zhou, X. Ma, C. Song: Appl. Catal. B: Environ. Vol. 87 (2009), p. 190

Advanced Materials Research Vol. 132 (2010) pp 149-161 © (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.132.149

Studies on the Catalytic Activity of Sulfated Zirconia Promoted with Cerium Oxide F. E. Lugo del Ángel a, *, R. Silva-Rodrigo a, 1, A. Vázquez Rodríguez b, R. García Alamilla a, J. Navarrete Bolaños b, A. Castillo Mares a, J. A. Melo Banda a, E. Terres Rojas b, J. L. Rivera Armenta a a

Instituto Tecnológico de Cd. Madero, División de Estudios de Posgrado e Investigación; Juventino Rosas y Jesús Urueta S/N; Col. Los Mangos, 89440, Cd. Madero, Tam., MÉXICO. b

Instituto Mexicano del Petróleo; Eje Central Lázaro Cárdenas No. 152, México D.F., MÉXICO. 1

Email: [email protected]

Keywords: Isomerization; Sol-gel ZrO2; Pt/ ZrO2SO4-CeO2; BET; FTIR pyridine; XRD.

Abstract. Pure zirconia, sulfated zirconia and sulfated zirconia modified with 2, 3, 5 and 10 wt. % of cerium oxide were synthesized by sol-gel method. Pt phase was impregnated on the supports using the incipient wet technique. Sulfated zirconium oxide showed tetragonal phase only. Addition of cerium to sulfated zirconia did not modify the tetragonal phase but produced a marked effect on the surface area. Low cerium content may greatly increase the surface area; however, too high cerium content (10 wt.%) may decrease the surface area. Pore size had influence on the catalytic activity and ZrO2 acidity was favored by the sulfate ion incorporation. All catalysts having Brönsted and Lewis acid sites were active in the n-hexane isomerization. The highest n-hexane conversion (40%) and selectivity towards DMB (26%), 2-methyl pentane (61%) and 3-methyl pentane (13%) were reached over the catalyst with 10 wt. % cerium oxide. In addition, sulfated zirconium oxide presented high selectivity of light products (< C6), which indicated that the addition of this doping agent (CeO2) made the catalysts more selective toward the desired reaction products. 1. Introduction The technological and industrial advance generated by man has been translated into multiple benefits for the population worldwide. Nevertheless, these facts have grown along with a phenomenon, for example, environmental pollution, whose impact is evidently negative for the subsistence of life on the whole planet. To find a solution is not an easy task due to the demographic increase and the technological developments. Responsibility of the problem is not for a single sector because many industries as well as homes and automobiles participate in the generation of polluting agents. Therefore, an alternative would be to invest in new and better technologies to guarantee gasoline free from polluting octane compounds such as additives, lead, tetraethyl and tert-butyl methyl ether through the n-hexane isomerization, which is an important reaction that allows a remarkable increase in the octane number (index that expresses the antidetonation power of a fuel) [1-2]. In the last years, new catalysts for the n-hexane isomerization have appeared in the international bibliography, the so-called super acid catalysts [3-6], where zirconia promoted with sulfate ions is an example. This one presents high activity, but has the disadvantage of losing sulfur during the reaction, which originates structural modifications [7]. These limitations are the points of interest of this work. The main objective is to study the effect of adding cerium oxide to the ZrO2-SO4 system by means of different characterization techniques. One of the interesting aspects of using oxides mixed with rare earths in the isomerization reaction is to increase the high selectivity toward the interest products.

150

Advances in New Catalytic Materials

2. Experimental 2.1. Synthesis of the catalysts The synthesis of sulfated zirconia modified with cerium oxide was performed by the sol-gel method, varying the CeO2 content (2, 3, 5 and 10 wt. %) [8-9]. This route consisted of a hydrolysis stage carried out in two steps so that the less reactive precursor, cerium nitrate, was partially hydrolyzed when added to zirconium butoxide (zirconia precursor). The zirconium butoxide solution was mixed with terbutylic alcohol and remained in agitation until reaching 343 K. On the other hand, cerium nitrate [Ce(NO3)3⋅6H2O] was dissolved in a mixture of tert-butyl alcohol [(CH3)3COH] and deionized water by a dropping system. During this time the gel was formed and then aged for 24 hours. The sulfation of the materials was carried out in situ using concentrated H2SO4 as sulfating agent, maintaining the amount of the sulfate ion constant in 20 wt. % [10]. All the materials were dried at 393 K, and calcined in a dynamic air atmosphere at 873 K. Pure sulfated materials were synthesized by using the same procedure. The support was impregnated with a necessary metallic amount, for example, 0.3 wt % of platinum [11]. The impregnation of the supports was performed in a 50-ml ball flask covered with a heating mantle with agitation. Diaminedinitro platinum [II] (Pt(NH3)2(NO2)2) was added with water, and later, slowly added to the solid (ZrO2-SO4-2-CeO2) at a temperature of 333 K with continuous agitation and heating for 2 hours. All the impregnated materials were dried at 383 K for 24h and then were calcined for 3h at 773 K in an air flow, using a heating rate of 3 K/min. Table 1 presents the compositions used for preparing the supports and catalysts. In this table, zirconia is represented by Z, sulfated zirconia by ZS and the mixed supports by ZSnCe with n as cerium oxide concentration (n = 2, 3, 5 and 10 wt. %). The same notation is used for the PtZSnCe catalysts. Table 2 shows the actual platinum content of the synthesized catalysts obtained by atomic absorption technique as well as the actual contents of sulfur and CeO2 determined by ASTM D1552 and EDS (SEM), respectively. The sulfur content in the catalyst is lower than the sample before calcination.

Table 1.Compositions used for preparation of supports and catalysts. Composition (wt. %) Supports Z ZS ZS2Ce ZS3Ce ZS5Ce ZS10Ce Catalysts PtZ PtZS PtZS2Ce PtZS3Ce PtZS5Ce PtZS10Ce

CeO2 --2 3 5 10

ZrO2 100 80 78 77 75 70

--2 3 5 10

100 80 78 77 75 70

SO4-2 20 20 20 20 20

Pt -------

20 20 20 20 20

0.3 0.3 0.3 0.3 0.3 0.3

2.2. Characterization Different characterization techniques were necessary to evaluate and identify the physicochemical properties of the catalysts.

Advanced Materials Research Vol. 132

151

Standard test method for sulfur. The samples were evaluated on the basis of the method ASTM D-1552. Energy dispersive X-ray spectroscopy (EDS). To perform this analysis we used the scanning electron microscope Philips XL30 ESEM-EDAX associated with an EDX microprobe for elemental microanalysis using a semiquantitative point. Table 2. Metallic contents of the synthesized catalysts. Composition wt.%

Catalysts CeO2 PtZ PtZS PtZS2Ce PtZS3Ce PtZS5Ce PtZS10Ce

2 3 5 10 a

EDS;

b

a

CeO2

1.98 2.94 4.8 10.28

b

S

--3.99 2.78 3.71 2.85 3.51

c

Pt

0.37 0.45 0.43 0.53 0.50 0.45

ASTM D-1552 (S Theoretic wt. % = 3.32) and c Atomic Absorption

N2 adsorption-desorption isotherms (BET). The textural properties of the supports were determined by N2 adsorption at -77 K using a Quantachrome AUTOSORB-1 sorptmeter. The surface area was calculated from the Brunauer-Emmett-Teller (BET) equation, while pore and average pore diameters were obtained from the desorption isotherm using the Barret, Joyner and Halenda (BJH) method. Before performing the BET measurements, the samples were pretreated at 623 K for 2 h [12-13]. Fourier-Transform Infrared (FTIR) spectroscopy. The study of functional groups of the supports was made by means of FTIR technique (conventional transmittance) with a Perkin-Elmer apparatus (Paragon 2000) in which wafers were prepared with KBr. X-ray diffraction. The crystalline phases were determined by using a Bruker AXS diffractometer model 8000 Advance equipped with an X-ray Cu Kα (λ =1.5405 Å) cathode operating at 35 kV and 25 mA. The sample was crushed until a fine and homogenous powder was obtained. A 1 gram sample was used for this analysis. The spectra were obtained in the 2θ interval = 10-70º. The software of the equipment has a data base of reference files from the Joint Committee of Power Diffraction Standards (JCPDS). The diffractograms were compared with standard files for the determination of the present phases. The crystallite size of the solid phase were evaluated on the basis of the Debye-Scherrer formula [9, 14], (τ = 0.9λ/βcos θ) where τ is the grain size of crystal, λ is the wavelength of the incident beam, β is the full width at half of the maximum of the diffraction band intensity and θ is half the Bragg angle. The isomerization reaction needs acid support: two techniques were carried out to evaluate the type and total acidity of the materials. (a) FTIR adsorbed pyridine. The type and quantity of acid sites (Brönsted and/or Lewis) were determined with a Fourier Transform Infrared (FTIR) Nicolet 170 SX spectrometer by means of pyridine adsorption. The annealed materials were pressed into thin self-supported wafers and placed in a Pyrex glass cell with a CaF2 window coupled to a vacuum line, degassing the sample at 673K for 2h to eliminate humidity. The adsorption was carried out on the IR cell at room temperature. The pyridine was desorbed under vacuum, from room temperature to 323, 373, 473, 573 and 673 K. The quantities of adsorbed pyridine were obtained from the integrated absorbance of the respective bands in the 1400 - 1700 cm-1 range in which the vibrations of the pyridine ring are located.

152

Advances in New Catalytic Materials

(b) 2-propanol dehydration. By analyzing the selectivity of products in the decomposition of 2propanol, fundamental information on the acid-base properties of the catalytic sites in the materials was obtained. The catalysts were classified according to their selectivity towards the formation of propylene (dehydration) or acetone (dehydrogenation) [15-16]. The tests were performed in a microplant that operates under atmospheric pressure and continuous flow. The reaction conditions and analysis of products were constant for each sample. The reaction temperature was 353 K. The reaction was monitored every 80 min [17]. Catalytic evaluation of n-hexane isomerization. The catalytic evaluation of the materials was carried out in a continuous flow micro plant by means of the n-hexane isomerization reaction. Prior to the reaction, each catalyst was reduced in the presence of hydrogen (30 ml/min) for 2 hours at 573 K. Later, the temperature was reduced to 523 K to start the isomerization reaction. Other reaction conditions were: pressure = 1.013 bars, catalytic mass = 250 mg, molar ratio H2/C6 = 6, volumetric flow C6 = 5 ml min-1, H2 volume = 30 ml min-1, weight hourly space velocity (WHSV) = 3 h-1. 3. Results and discussion 3.1 Textural Properties

Table3. Textural properties of synthesized solids.

Supports Z ZS ZS2Ce ZS3Ce ZS5Ce ZS10Ce Catalysts PtZS2Ce PtZS3Ce PtZS5Ce PtZS10Ce

2

Textural properties PV (cc/g) DPSa(Å)

26 85 143 130 125 66

0.068 0.170 0.284 0.361 0.360 0.252

105 78 88 101 115 151

Dhklb ( Å) 118 88 127 108 100 94

118 93 66 51

0.273 0.210 0.526 0.410

92 90 318 326

---------

SA(m /g)

Surface area (SA), pore diameter (DPS), pore volume (PV) and crystallite size (Dhkl); aPore diameter determined from the desorption isotherm by the BJH method. bCrystallite size determined from Scherrer’s equation

The BET surface area was obtained by the N2 adsorption isotherms, as shown in Table 3. As can be observed in Table 3, the addition of cerium oxide to sulfated zirconia strongly increases the surface area; for example, for the support with the lowest cerium oxide content (ZS2Ce), the surface area is increased from 85 m2 g-1 to 143 m2 g-1. At higher cerium concentrations, the surface area diminishes, and for 10 wt % cerium oxide, it reaches 66 m2 g-1. At the same time, the pore diameter increases from 88 Å, at 2 wt % of cerium, to 151 Å, at 10 wt % of cerium oxide. It is possible that cerium at high contents forms CeO2 conglomerates that block the support pores, thus promoting a diminution of the surface area. The addition of platinum into supports led to an obvious decrease by 50 % in the surface area, especially for ZS5Ce. Moreover, the addition of platinum to the support increased the pore diameter more than 200 % for the sample with 5 and 10 wt % of cerium oxide. Nitrogen adsorption isothermals at 77K of these materials are shown in Fig. 1(a). The isotherms were of type III

Advanced Materials Research Vol. 132

153

(IUPAC classification), which occur when the adsorbate-adsorbent interaction is low, with the hysteresis loop characteristic of mesoporous materials with open pore systems. (a) ZS10Ce ZS5Ce ZS3Ce ZS2Ce ZS Z

Volume, [cc/g]

100

50

0 0.0

0.2

0.4

0.6

0.8

1.0

Relative pressure, (P/Po)

(b)

PtZS2Ce PtZS3Ce PtZS5Ce PtZS10Ce

350

Volume, [cc/g]

300 250 200 150 100 50 0 0.0

0.2

0.4

0.6

0.8

1.0

Relative pressure [P/Po]

Fig.1. N2 adsorption-desorption isotherms at 77K of (a) supports and (b) catalysts. All the solids presented a unimodal pore size distribution. The addition of cerium oxide led to a distinct increase in pore size of ZS (Table 3). The presence of cerium did not cause any significant effect on the type of isotherm with the exception of the ZS2Ce support that presented the type II isotherm corresponding to a solid crossed by almost cylindrical channels or made up of aggregates (consolidated) or agglomerates (not consolidated) of spheroidal particles with irregular forms and sizes [18]. The isotherms and hysteresis curves and pore size distributions of the catalysts appear in Fig.1 (b), which presented type III isotherms with a unimodal pore size distribution.

154

Advances in New Catalytic Materials

3.2 FTIR characterization.

4000

(b)

3500

2500

2000

1500

1000

500

Z10Ce

Z5Ce

OH (H2O)

SO4

M-O

OH

Z3Ce

Z2Ce

1630

T r a n s m i t a n c e, [%]

3000

3500

3000

2500

2000

1500

1000

500

-1

W a v e n u m b e r, [cm ]

Fig. 2. Spectra for FTIR: (a) ZrO2, ZrO2 - SO4-2 and (b) ZrO2-SO4-CeO2 Fig. 2 (a) shows the IR spectra of zirconium and sulfated zirconium oxides. Zirconium oxide presents strong absorption bands between 3650-2500 cm-1, which are assigned to stretching vibrations of O-H groups [19]. These groups belong to OH connected with the zirconium network, OH occluded water and terbutylic alcohol. This absorption is correlated to the 1600 cm-1 absorption band due to flexion vibrations of these groups. The bands that appear in the 400 - 800 cm-1 range of the spectrum are generated by the stretching vibrations of Zr-O and Ce-O bonds. Sulfated zirconium oxide exhibits the same bands as pure zirconia with the presence of SO42- groups characterized by absorptions at 1242, 1148 and 1055 cm-1 [10-11]. Figure 2 (b) presents the spectra of sulfated

Advanced Materials Research Vol. 132

155

zirconia samples modified with cerium oxide. It is important to emphasize that this characterization was also done to the catalysts obtained after of the reaction, which did not suffer any significant change in the functional groups. 3.3 X-Ray diffraction (a )

∗

[1 1 1 ]

M o n o c lin ic

[2 2 0 ] [0 0 2 ] [2 0 0 ]

∗

∗

Z

[2 2 2 ]

[1 1 2 ]

∗

I n t e n s i t y, [u. a]

[3 1 1 ]

∗∗

Z S

2 0

2 5

3 0

3 5

4 0

4 5

5 0

5 5

X -r a y d i f f r a c t i o n ,

6 0

6 5

7 0

2 θ

(b)

[1 1 1] [2 2 0]

[2 0 0]

ZS2Ce

[3 1 1]

I n t e n s i t y, [a.u.]

ZS3Ce

ZS5Ce

ZS10Ce

20

25

30

35

40

45

50

55

60

65

X-r a y d i f f r a c t i o n, 2 θ

Fig. 3. XRD patterns (a) ZrO2 (Z), ZrO2SO4 (ZS) and (b) mixed oxides (ZrO2-SO4-CeO2). Fig. 3 (a) shows the diffractograms of zirconium oxide and sulfated zirconium oxide. In the case of zirconium oxide, we observe a mixture of tetragonal and monoclinic phases, according to the J.C.P.D.S. database. The tetragonal phase corresponds to the most intense planes [1 1 1], [2 0 0], [2 2 0], [3 1 1] observed at 30, 35, 50 and 60º. The monoclinic phase of Baddeleyite (ZrO2) shows diffraction lines of weaker intensity at 2θ = 24, 28 and 31º. This result is in good agreement with the work reported by Huang et al. [20]. The diffractogram of sulfated zirconium oxide only presents the tetragonal phase, which is in agreement with K. Fottinger [21].

156

Advances in New Catalytic Materials

The diffractograms of modified mixed oxides are presented in Fig. 3(b). The addition of cerium to sulfated zirconia did not modify the tetragonal phase. In Table 3, the values of the crystallite size evaluated by means of the Scherrer equation, using the [1 1 1] diffraction of the tetragonal phase of zirconium oxide are shown. In sulfated zirconia, the crystallite size is reduced with respect to zirconia. When 2 wt % CeO2 is added to sulfated zirconia, there is an increase in the crystallite size, which passes from 88 to 127 Å; nevertheless, when the concentration of the doping agent (CeO2) is increased, a diminution of the crystal size occurs. 3.4 FTIR pyridine The pyridine IR spectra were analyzed in the region assigned to the vibrations of the pyridine aromatic ring from 1700 to 1400 cm-1. All the samples were scanned from 298 to 673 K. The acidity results obtained by means of FTIR-pyridine adsorption are presented in Table 4. All the samples exhibited the presence of Brönsted and Lewis sites, considering important behavior between 473 – 573 K, since the isomerization reaction of n-hexane was carried out at 523 K. Nascimento et al. [22] observed that the ratio of Brönsted /Lewis sites was correlated with the catalytic activity in n-butane isomerization, where the maximum activity occurred at a ratio of 1. In the spectrum of pure zirconium oxide (see Figure 4) taken after adsorption of pyridine, strong adsorption bands at 1610 and 1444 cm-1 are observed for the bonds of coordinated pyridine to a strong Lewis acid site, which remained well defined up to 673 K. At 1490 cm-1 there is a well defined band attributed to the vibrations of the pyridine ring involving two types of Lewis and Brönsted acid sites. This band is characteristic of the total acidity of the material [23]. The band at 1575 cm-1 is representative of Lewis type acidity with medium intensity corresponding to the Hpyridine bonds. Finally, at 1590 cm-1 there is a small band corresponding to Brönsted sites of medium intensity which disappeared with the increasing desorption temperature [24-26]. Table 4. Brönsted and Lewis Acidity for ZrO2-SO4-CeO2. Pyridine µmoles/g Supports Z ZS ZS2Ce ZS3Ce ZS5Ce ZS10Ce

Brönsted sites 5 43 23 117 15 18

Lewis sites 65 29 60 72 72 13

Total sites 70 72 83 189 87 31

Relation B/L 0.08 1.48 0.38 1.63 0.21 1.38

Sulfated zirconia analyzed by infrared spectroscopy of adsorbed pyridine can be seen in Figure 4, which shows the appearance of new bands at 1540 cm-1 and 1640 cm-1 attributed to vibrations of the pyridinium ion chemisorbed on Brönsted acid sites with medium and strong intensity, respectively. Also, there are bands characteristic of Lewis sites (1610 and 1444 cm-1) and Lewis and Brönsted acid sites (1488-1490 cm-1). At 673 K all bands are observed but with less intensity.

Advanced Materials Research Vol. 132

157

Fig. 4. IR spectra pyridine of ZrO2 (Z), ZrO2-SO4-2(ZS) and mixed oxides (ZrO2-SO4-CeO2). The IR spectra of modified oxides with cerium oxide are also presented in Figure 4. In general all samples have the same bands of sulfated zirconium oxide but as the concentration of cerium increases the intensity of the Lewis sites remains while the Brönsted sites disappear at 673 K with the exception the ZS10Ce support, which keeps all the bands but with less intensity, showing strong sites.

158

Advances in New Catalytic Materials

3.5. 2-Propanol dehydration Table 5 compared the activities of 2-propanol dehydration over three catalysts: zirconium oxide, sulfated zirconium oxide and sulfated zirconium oxide modified with rare earth. The conversion of isopropanol over the ceria modified oxides is between 12 to 24 %, obviously, the catalytic activity increases when the concentration of CeO2 is increased. However, it is emphasized that as the concentration of cerium increases in the ceria doped materials, their selectivity to propylene largely grows; and it has been observed that with 10 wt. % cerium, a higher selectivity to diisopropyl ether (DIPE) can be obtained. Table 5. Dehydration of 2-propanol with materials based on ZrO2-SO4-CeO2 T= 353 K, P= 1.023bars, WHSV= 1.5h-1, θrxn=80min. Supports

Z ZS ZS2Ce ZS3Ce ZS5Ce ZS10Ce

Catalytic activity in 2-propanol dehydration wt. % % % % CeO2 Conversion SPropylene SEterdiisopropilic ----2 3 5 10

0.16 14 12 18 24 18

100 97 78 90 89 24

0 3 22 10 11 76

3.6 Catalytic activity All the catalysts showed activity in the isomerization of n-hexane. The Pt catalyst on pure zirconia (Pt/Z) (see Table 6 and Figure 5) had a 2 % of catalytic activity at this condition, confirming that this type of reaction demands Brönsted type acidity. The highest conversion (67 %) was obtained over the catalyst Pt/ZS. In addition to the products of interest, the catalytic cracking products (