摘要:

4369 4377 4387 4397 4407 4417 4427 4439 445 1 4457 447 1 4475 4487 4495 450 1 4509 Con tents A New Form of the High-temperature Isopiestic Technique and its Applica- tion to Mercury-Bismuth, Mercury-Cadmium, Mercury-Gallium, Mercury- Indium and Mercury-Tin Binary Amalgams Z-C. Wang, X-H. Zhang, Y-Z. He and Y-H. Bao The Derivation of Chemical-diffusion Coefficients of Oxygen in UO,,, over the range 180-300 "C. Spectroscopic Procedure and Preliminary Results T. R. Griffiths, H. V. St. Aubyn Hubbard, G. C. Allen and P. A. Tempest Pho tophysics at Solid Surfaces. Evidence of Dimer Formation and Polarization of Monomer and Excimer Fluorescences of Pyrene in the Adsorbed State on Silica-gel Surfaces T. Fujii, E. Shimizu and S. Suzuki Ordering in Monodispersed Polymer Latices induced by a Temperature Gradient K.Furusawa, N. Tobori and S. Hachisu X-Ray Diffraction Study of Molten Eutectic LiF-NaF-KF Mixture K. Igarashi, Y. Okamoto, J. Mochinaga and H. Ohno Viscosity Measurements of Some Tetra butylammonium, Copper( I), Silver( I) and Thallium( 1) Salts in Acetonitrile-Pyridine Mixtures at 15, 25 and 35 "C D. S. Gill and B. Singh The Ethane- 1,2-diol-Water Solvent System. The Dependence of the Dis- sociation Constant of Picric Acid on the Temperature and Composition of the Solvent Mixture Silver(1) Complexation with Tertiary Amines in Toluene M. Soledade Santos, E. F. G. Barbosa and M. Spiro Enhanced Oxygen Evolution through Electrochemical Water Oxidation mediated by Polynuclear Complexes embedded in a Polymer Film G. J. Yao, A. Kira and M. Kaneko Nature of Acid Sites in SAP05 Molecular Sieves.Part 1.-Effects of the Concentration of Incorporated Silicon C. Halik, J. A. Lercher and H. Mayer Hemimicelle Formation of Cationic Surfactants at the Silica Gel-Water Interface T. Gu, Y. Gao and L. He Nuclear Magnetic Resonance Relaxation in Micelles. Deuterium Relaxation at Three Field Strengths of Three Positions on the Alkyl Chain of Sodium Dodecyl Sulphate Studies of the Temperature Dependence of Retention in Supercritical Fluid Chromatography K. D. Bartle, A. A. Clifford, J. P. Kithinji and G. F. Shilstone Hydrogen and Muonium Atom Adducts of Trimethylsilyl Derivatives of Ethyne The Radical Cation of Formaldehyde in a Freon Matrix. An Electron Spin Resonance Study Phase Transition of the Water confined in Porous Glass studied by the Spin- probe Method H.Yoshioka G. C. Franchini, A. Marchetti, L. Tassi and G. Tosi 0. Soderman, G. Carlstrom, U. Olsson and T. C. Wong C. J. Rhodes and M. C. R. Symons C. J. Rhodes and M. C. R. Symons4369 4377 4387 4397 4407 4417 4427 4439 445 1 4457 447 1 4475 4487 4495 450 1 4509 Con tents A New Form of the High-temperature Isopiestic Technique and its Applica- tion to Mercury-Bismuth, Mercury-Cadmium, Mercury-Gallium, Mercury- Indium and Mercury-Tin Binary Amalgams Z-C. Wang, X-H. Zhang, Y-Z. He and Y-H. Bao The Derivation of Chemical-diffusion Coefficients of Oxygen in UO,,, over the range 180-300 "C. Spectroscopic Procedure and Preliminary Results T. R. Griffiths, H. V. St. Aubyn Hubbard, G. C. Allen and P. A. Tempest Pho tophysics at Solid Surfaces.Evidence of Dimer Formation and Polarization of Monomer and Excimer Fluorescences of Pyrene in the Adsorbed State on Silica-gel Surfaces T. Fujii, E. Shimizu and S. Suzuki Ordering in Monodispersed Polymer Latices induced by a Temperature Gradient K. Furusawa, N. Tobori and S. Hachisu X-Ray Diffraction Study of Molten Eutectic LiF-NaF-KF Mixture K. Igarashi, Y. Okamoto, J. Mochinaga and H. Ohno Viscosity Measurements of Some Tetra butylammonium, Copper( I), Silver( I) and Thallium( 1) Salts in Acetonitrile-Pyridine Mixtures at 15, 25 and 35 "C D. S. Gill and B. Singh The Ethane- 1,2-diol-Water Solvent System. The Dependence of the Dis- sociation Constant of Picric Acid on the Temperature and Composition of the Solvent Mixture Silver(1) Complexation with Tertiary Amines in Toluene M.Soledade Santos, E. F. G. Barbosa and M. Spiro Enhanced Oxygen Evolution through Electrochemical Water Oxidation mediated by Polynuclear Complexes embedded in a Polymer Film G. J. Yao, A. Kira and M. Kaneko Nature of Acid Sites in SAP05 Molecular Sieves. Part 1.-Effects of the Concentration of Incorporated Silicon C. Halik, J. A. Lercher and H. Mayer Hemimicelle Formation of Cationic Surfactants at the Silica Gel-Water Interface T. Gu, Y. Gao and L. He Nuclear Magnetic Resonance Relaxation in Micelles. Deuterium Relaxation at Three Field Strengths of Three Positions on the Alkyl Chain of Sodium Dodecyl Sulphate Studies of the Temperature Dependence of Retention in Supercritical Fluid Chromatography K. D. Bartle, A. A. Clifford, J. P. Kithinji and G. F. Shilstone Hydrogen and Muonium Atom Adducts of Trimethylsilyl Derivatives of Ethyne The Radical Cation of Formaldehyde in a Freon Matrix. An Electron Spin Resonance Study Phase Transition of the Water confined in Porous Glass studied by the Spin- probe Method H. Yoshioka G. C. Franchini, A. Marchetti, L. Tassi and G. Tosi 0. Soderman, G. Carlstrom, U. Olsson and T. C. Wong C. J. Rhodes and M. C. R. Symons C. J. Rhodes and M. C. R. Symons

ISSN:0300-9599    

DOI:10.1039/F198884FX005

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

NOMENCLATURE AND SYMBOLISM Units and Symbols. The Symbols Committee of The Royal Society, of which The Royal Society of Chemistry is a participating member, has produced a set of recommendations in a pamphlet 'Quantities, Units, and Symbols' (1 975) (copies of this pamphlet and further details can be obtained from the Manager, Journals, The Royal Society of Chemistry, Burlington House, London W1V OBN). These recommendations are applied by The Royal Society of Cemistry in all its publications. Their basis is the 'Systeme International d'Unit6s' (9). A more detailed treatment of units and symbols with specific application to chemistry is given in the IUPAC Manual of Symbols and Terminology for Physicochemical Quantities and Units (Pergamon, Oxford, 1979). Nomenclature. For many years the Society has actively encouraged the use of standard IUPAC nomenclature and symbolism in its publications as an aid to the accurate and unambiguous communication of chemical information between authors and readers.In order to encourage authors to use IUPAC nomenclature rules when drafting papers, attention is drawn to the following publications in which both the rules themselves and guidance on their use are given: Nomenclature of Organic Chemistry, Sections A, B, C, D, E, F, and H (Pergamon, Oxford, 1979 edn). Nomenclature of Inorganic Chemistry (Butterworths, London, 1971 , now published by Pergamon). Biochemical Nomenclature and Related Documents (The Biochemical Society, London, 1978). Compendium of Chemical Terminology: IUPAC Recommendations (Blackwells, Oxford, 1987).A complete listing of all IUPAC nomenclature publications appears in the January issues of J. Chem. SOC., Faraday Transactions. It is recommended that where there are no IUPAC rules for the naming of particular compounds or authors find difficulty in applying the existing rules, they should seek the advice of the Society's editorial staff. (xiv)NOMENCLATURE AND SYMBOLISM Units and Symbols. The Symbols Committee of The Royal Society, of which The Royal Society of Chemistry is a participating member, has produced a set of recommendations in a pamphlet 'Quantities, Units, and Symbols' (1 975) (copies of this pamphlet and further details can be obtained from the Manager, Journals, The Royal Society of Chemistry, Burlington House, London W1V OBN). These recommendations are applied by The Royal Society of Cemistry in all its publications.Their basis is the 'Systeme International d'Unit6s' (9). A more detailed treatment of units and symbols with specific application to chemistry is given in the IUPAC Manual of Symbols and Terminology for Physicochemical Quantities and Units (Pergamon, Oxford, 1979). Nomenclature. For many years the Society has actively encouraged the use of standard IUPAC nomenclature and symbolism in its publications as an aid to the accurate and unambiguous communication of chemical information between authors and readers. In order to encourage authors to use IUPAC nomenclature rules when drafting papers, attention is drawn to the following publications in which both the rules themselves and guidance on their use are given: Nomenclature of Organic Chemistry, Sections A, B, C, D, E, F, and H (Pergamon, Oxford, 1979 edn). Nomenclature of Inorganic Chemistry (Butterworths, London, 1971 , now published by Pergamon). Biochemical Nomenclature and Related Documents (The Biochemical Society, London, 1978). Compendium of Chemical Terminology: IUPAC Recommendations (Blackwells, Oxford, 1987). A complete listing of all IUPAC nomenclature publications appears in the January issues of J. Chem. SOC., Faraday Transactions. It is recommended that where there are no IUPAC rules for the naming of particular compounds or authors find difficulty in applying the existing rules, they should seek the advice of the Society's editorial staff. (xiv)

ISSN:0300-9599    

DOI:10.1039/F198884BX007

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

ISSN 0300-9238 JCFTAR 84(2) 367-674 (1 988) 367 379 39 1 397 41 3 43 3 44 1 459 467 6 473 483 49 1 50 1 51 I 52 1 529 539 55 I JOURNAL OF THE CHEMICAL SOCIETY Faraday Transactions I Physical Chemistry in Condensed Phases CONTENTS Mechanism of Electrohydrodimerization of 2-Cyclohexen- 1 -one on Mercury from Aqueous Solutions. Part 2.-Results obtained for Solutions containing Triton X-100 Characterisation of an RuO;xH,O Colloid and Evaluation of its Ability to Mediate the Oxidation of Water A. Mills and N. McMurray Modes of Enhancement of Physical Adsorption of Nitrogen and Water Vapour on Metal Oxides Solutions of Organic Solutes. Part 2.-Moderately Polar Compounds in Water ; Limiting Volumes and Compressibilities Paramagnetic Rhodium Species in Zeolites. Part 1 .-RhNa-X and RhNa-Y A.Sayari, J. R. Morton and K. F. Preston Raman Spectra of Aniline adsorbed on an Ag Electrode in Acidic Solutions H. Shindo and C. Nishihara Tin Dioxide Gas Sensors. Part 2.-The Role of Surface Additives J. F. McAleer, P. T. Moseley, J. 0. W. Norris, D. E. Williams and B. C. Tofield Electrochemical Studies of Hydrogen in Ordered and Disordered Pd,Mn Alloys Electron Spin Resonance and Electron Spin-echo Modulation Studies of 5- Doxylstearic Acid and N, N, N', N'-Tetramethylbenzidine Photoionization in Sodium Dodecylsulphate Micelles. Effects of 1 5-Crown-5 and 18-Crown-6 Ether Addition Rotating-disc Electrode Voltammetry. Waveshape Analysis for DISP2 and EC, Processes The Reduction of Fluorescein in Aqueous Solution (at pH 6). A New DISP2 Reaction Chemisorption of Propene on HZSM-5 by Ultraviolet and Infrared Spectro- scopy Reaction of the Aquacopper(1) Ion with Hydrogen Peroxide.Evidence for a CuTTT (Cupryl) Intermediate G. R. A. Johnson, N. B. Nazhat and R. A. Saadalla-Nazhat Infrared Studies of Adsorbed Species of H,, CO and CO, over ZrO, J. Kondo, H. Abe, Y. Sakata, K. Maruya, K. Domen and T. Onishi Modification of y-Alumina by Barium and Lanthanum, and the Consequential Effect on the Reducibility and Dispersibility of Nickel S. Narayanan and K. Uma Heat Capacities and Volumes of Mixtures of N,N-Dimethylformamide with Isobutanol, sec-Butanol and t-Pentanol. An Analysis of the Water-Non- electrolyte Enthalpic Pair Interaction Coefficients in N,N-Dimethylformamide Solution H. Piekarski and G. Somsen Thermodynamics of Fluorocarbon-Hydrocarbon Mixtures.The Systems formed by 2,2,4-Trimethylpentane with Hexafluorobenzene and with Hexa- fluorobenzene-Benzene J. Aracil, R. G. Rubio, M. Caceres, M. D. Peiia and J. A. R. Renuncio Thermodynamics of Micelle Formation of Alkali-metal Perfluorononanates in Water. Comparison with Hydrocarbon Analogues I. Johnson and G. Olofsson M. Y. Duarte, G. Pezzatini and R. Guidelli P. A. Sermon and R. R. Rajaram J. V. Leyendekkers K. Baba, Y. Sakamoto and T. B. Flanagan P. Baglioni and L. Kevan R. G. Compton, D. Mason and P. R. Unwin R. G. Compton, D. Mason and P. R. Unwin I. Kiricsi and H. Forster 13 FAR 156 1 575 58 1 59 1 60 1 609 617 63 1 647 657 665 Contents Solvent Properties of Polyaromatic Hydrocarbons G. Geblewicz and D. J.Schiffrin The Phase Response of the Explodator M. Eszterle, 2. Noszticzius and Z. A. Schelly Absorption and Diffusion of Sulphur Dioxide into Aqueous Sodium Chloride Solutions D. G. Leaist Enthalpic Pair Interaction Coefficients of NaI-Non-electrolyte in DMF Solution at 25 "C. A Comparison of Electrolyte-non-electrolyte Interactions in DMF and in Aqueous Solutions H. Piekarski Activation and Chain-carrying CH, Species in Terminal Alkene Metathesis on Molybdena-Titania Catalysts K. Tanaka and K-i. Tanaka Radical Cations of Nitroso Derivatives. A Radiation-chemical and Electron Spin Resonance Study H. Chandra, D. J. Keeble and M. C. R. Symons Electron Energy-loss Spectroscopy and the Crystal Chemistry of Rhodizite. Part 1 .-Instrumentation and Chemical Analysis W. Engel, H. Sauer, E. Zeitler, R. Brydson, B. G. Williams and J. M. Thomas Electron Energy-loss Spectroscopy and the Crystal Chemistry of Rhodizite. Part 2.-Near-edge Structure R. Brydson, B. G. Williams, W. Engel, Th. Lindner, M. Muhler, R. Schlogl, E. Zeitler and J. M. Thomas Differential Thermo-osmotic Permeability in Water-Cellophane Systems C. Fernandez-Pineda and M. I. Vhzquez-Gonzailez Spectrochemistry of Solutions. Part 20.l-The Infrared, Near Infrared and Visible Spectra of Liquid Ammonia J. C. Dougal, P. Gans and J. B. GiU Dynamic Studies of the Interaction between Diols and Water by Ultrasonic Methods. Part 4.-3-Methylbutane- 1,3-diol and 2,2-Dimethylpropane- 1,3-diol Solutions S. Nishikawa, N. Nakayama and N. Nakao

ISSN:0300-9599    

DOI:10.1039/F198884FP019

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

JOURNAL OF THE CHEMICAL SOCIETY Faraday Transactions II, Issue2,1988 Molecular and Chemical Physics 135 141 149 155 161 181 191 199 209 For the benefit of readers of Faraday Transactions I, the contents list of Faraday Transactions II, Issue 2, is reproduced below. The Gas-phase Reaction of CF, Radicals with HCN in the Temperature Range 333-523 K Selective Disruption of lH-lH Dipolar Connectivities within Two- dimensional Nuclear Overhauser Effect Spectra by Paramagnetic Centres E. Gaggelli, E. Tiezzi and G. Valensin Photosensitisation of Melanins covalently bound to Dyes J. Bielec, B. Pilas, T. Sarna, C. Knox and T. G. Truscott Transition Temperatures of Binary Nematic Mixtures predicted by the Humphries-James-Luckhurst Theory A. Kloczkowski and G. R. Luckhurst Painlev6 Solution of the Poisson-Boltzmann Equation for a Cylindrical Polyelectrolyte in Excess Salt Solution J.S. McCaskill and E. D. Fackerell Reactive Scattering of a Supersonic Fluorine Atom Beam: F+OCS N. Firth, N. W. Keane and R. Grice Emission Spectroscopy of the Plasma Decomposition of Silicon Tetra- chloride M. D. Rowe Determination of the Rate Constant for HO, + CH, -+ H,O, + CH, at 443 "C R. R. Baldwin, P. N. Jones and R. W. Walker Valence Ionization Energies of Ni(CN)t-, CO(C0):- and Fe(CN)t-, studied by X-Ray Emission Spectroscopy and Ab Initio Molecular Orbital Methods S. Smith, D. A. Taylor, I. H. Hillier, M. A. Vincent, M. F. Guest, A. A. MacDarell, W. von Niessen and D. S. Urch S. Lane, E. V. Oexler and E. H. Staricco The following papers were accepted for publication in Faraday Transactions I during November 1987.71650 71943 71 1009 71 1048 711 106 Solvatochromatic Indicator Study of Silicalite and Zeolite ZSM-5 G. P. Handreck and T. D. Smith Properties of Complexes with Cobalt-Carbon Bonds formed by Reactions of Aliphatic Free Radicals with Nitrilotriacetate-Cobalt(r1) in Aqueous Solution. A Pulse Radiolysis Study Ordered Distribution of Aluminium in Zeolite L T. Takaishi Volumetric Properties of Aqueous Solutions of Polyols between 0.5 and 25 "C S. Wurzburger, R. Sartorio, G. Guarino and M. Nisi A General Calculation of Molecular Solvation Energies R. J. Abraham, B. D. Hudson, M. W. Kermode and R. Mines D. Meyerstein and H. A. Schwarz711153 7/ 1209 711210 7/ 1282 7/ 1362 7/ 1370 711377 7/ 1379 7/ 1380 7/1381 7/ 1405 7/ 1425 711513 7/ 1537 7/ 1560 7/ 1584 7/ 1588 Volume and Compressibility Changes in Mixed Salt Solutions at 25 "C K.Patil and G. Mehta Adsorption Sites of Water and Methanol Molecules on Collodial Ferric Oxide Hydroxides T. Ishikawa, H. Sakaiya and S. Kondo The Effect of Pressure on the 'Crystal-like' Ordering of Monodisperse Poly- styrene Spheres in Deionized Aqueous Suspensions T. Okubo The Temperature Variation of the Hydrophobic Effect M. H. Abraham and E. Matteoli Infrared Study of Ammonia and Nitric Oxide Adsorption on Silica- supported Iron Catalysts Application of Kirkwood-Buff Theory to Enthalpies of Transfer and Expansibilities of Solutes in Binary Solvent Mixtures Rotating-disc Electrode Voltammetry. The Catalytic Mechanism (EC') and its Nuances R.G. Compton, M. J. Day, M. E. Laing, R. J. Northing, J. I. Penman and A. M. Waller Effect of Acidity on the Reactivity of the Triplet State of 2-Nitrothiophen L. J. A. Martins and T. J. Kemp Catalytic Properties of Aluminium Form of Zeolite Y Modified with Trifluoromethane S. Kowalak Thermodynamics of Water Sorption by Perfluorosulphonate (Nafion- 1 17) and Polystyrene-Divinylbenzene Sulphonate (Dowex 50W) Ion-exchange Resins at 298 Rotating-disc Electrodes. Single and Double-potential Step Chrono- amperometry and the ECE-DISP1 Problem R. G. Compton, D. Mason and P. Unwin Influence of Salts on Poly(vinylmethy1ether) at the Air/ Aqueous Solution Interface. Part 2.-Adsorption from Solution D. D. Eley, M. J. Hey and J. M. Speight Transient Discrimination of Carbon Monoxide Oxidation Kinetics with Two Reaction Paths Progressed on Zinc Oxide M.Kobayashi, T. Kanno and T. Kimura The Dehydrogenation of Ethanol in Dilute Aqueous Solution Photo- sensitised by Benzophenones P. Green, W. A. Green, A. Harriman, M. C. Richoux and P. Neta Application of Band-shape Analysis to Infrared Spectra of Adsorption of Nitrogen Monoxide on Iron Oxide-supported Silica, Alumina and Titania H. Miyata, Y. Nakagawa, S. Miyagawa and Y. Kubokawa The Water/Oil/Water Thermocouple and the Ionic Seebeck Effect H. H. Girault The Elimination of Internal and External Reference in Nuclear Magnetic Resonance Determinations of Fast Equilibria with Particular Reference to Electron-Donor-Acceptor (EDA) Complex Formation J. A. Chudek, R. Foster and R.L. MacKay C. Johnston, N. Jorgensen and C. H. Rochester K. E. Newman 1K K. K. Pushpa, D. Nandan and R. M. Iyer711613 7 1 1627 71 1628 71 1629 7/ 1663 7 / 1689 711714 71 1753 711812 71 1872 71 1872 711991 712075 712076 712077 712078 712079 Mass Transport in Channel Electrodes. The Application of the Backwards Implicit Method to Electrode Reactions (EC, ECE and DISP) involving Coupled Homogeneous Kinetics R. G. Compton, M. B. G. Pilkington and G. M. Stearn Carbon Monoxide Hydrogenation over Silica-supported Rhodium Catalysis : The Effect of the Rhodium Precursor S. D. Jackson, B. J. Brandreth and D. Winstanley Structure Dependence in the Hydrogenation of Diolefins over Ru Thin Films J. Tamaki, T. Miyanaga, T. Imanaka and T. Yamane Photoelectrochemical Electron Spin Resonance.Part 2.-The Reduction of Crystal Violet Solvation of Cyanoalkanes [CH,CN and (CH,),CCN] : An Infrared and Nuclear Magnetic Resonance Study G. Eaton, A. S. Pena-Nuiiez and M. C. R. Symons Spectroscopic Investigation of the Interaction of Co,(CO), with MgO and SiO, K. M. Rao, E. Guglieminotti and A. Zecchina The Theory of the Taylor Dispersion Technique for Three-component System Diffusion Measurements Linear Correlation between Entropies of Complexation of Cryptand 222 with Metal Ions in Non-aqueous Solvents and Entropies of Solvation of These Ions in These Solvents The Enthalpies of Interaction of some Amides with Urea in Water at 25 "C P. J. Cheek and T. H. Lilley Pulse Radiolysis Study of Salt Effects on Reactions of Aromatic Radical Cations with C1-.Part 2.-Spectral Shifts and Decay Kinetics of Diphenylpolyene Radical Cations in the Presence of Tetrabutylammonium Hexafluorophosphate The Dimer State of No Filled in Micropores of Cu(OH),-dispersed Activated Carbon Fibres K. Kaneko, A. Kobayashi, T. Suzuki, S. Ozeki, K. Kakei, N. Kosugi and H. Kuroda Excess Enthalpies and Excess Volumes of [xCO, + (1 - x) N,O] in the Liquid and Supercritical Regions Correlation between the Hydrophobic Nature of Monosaccharides and Chol- ates and their Hydrophobic Indexes K. Miyajima, K. Machida, T. Taga, H. Komatsu and M. Nakagaki A Nuclear Magnetic Resonance Study of the Isomeric Pentitols in Aqueous and Non-aqueous Solutions F. Franks, R. L. Kay and J. Dadok Influence of Water on Pure Sorbitol Polymorphism S. Quinquenet, C.Grabielle-Madelmont, M. Ollivon, M. Serpelloni and R. Freres Structure, Sweetness and Solution Properties of Small Carbohydrate Molecules Thermomechanical Properties of Small Carbohydrate-Water Glasses and ' Rubbers '. Kinetically Metastable Systems at Sub-zero Temperatures H. Levine and L. Slade R. G. Compton, B. A. Coles, G. M. Stearn and A. M. Waller W. E. Price A. F. Danil de Namor Y. Yamamoto, T. Aoyama and K. Hayashi C. J. Wormald and J. M. Eyears G. G. Birch and S. Shamil (iii)7/2080 Derivation of Parameters for Conformational Calculations on Carbohydrate Systems, including Bacterial Cell Wall Peptidoglycan A. Marsden, B. Robson and J. S. Thompson 7/208 1 Conformation of Polyols in Water, Molecular Dynamics Simulation of Man- nitol and Sorbitol 7/2082 Excess Enthalpy and Excess Volume in the Ternary Aqueous Solutions with Sucrose-Glucose, Sucrose-Glycerol and Glucose-Glycerol at 298.1 K N.Daldrup and H. Schonert 7/2084 Interactions between Cations and Sugars. Part 4.-Free Energy of Interaction of Calcium Ion with some Aldopentoses and Aldohexoses in Water at 298.15 K Jean-Pierre Morel, C. Lhermet and N. Morel-Desrosiers 7/2085 Solution Properties and the Sweet Taste of Small Carbohydrates M. Mathlouthi and A. M. Seuvre J. P. GrigeraCumulative Author Index 1988 Abe, H., 511 Abraham, M. H., 175 Allen, G. C., 165, 355 Anazawa, I., 275 Aracil, J., 539 Baba, K., 459 Baglioni, P., 467 Barna, T., 229 Bazsa, G., 215, 229 Berei, K., 367 Berroa de Ponce, H., 255 Blesa, M. A., 9 Borgarello, E., 261 Breen, J., 293 Brown, M.E., 57 Brydson, R., 617, 631 Busca, G., 237 Caceres, M., 539 Carbone, A. I., 207 Cavani, F., 237 Cavasino, F. P., 207 Centi, G., 237 Chandra, H., 609 Clarke, R. J., 365 Coates, J. H., 365 Compton, R. G., 473, 483 Danil de Namor, A. F., 255 Dash, A. C., 75 Dash, N., 75 Davydov, A., 37 Dawber, J. C., 41 Dawber, J. G., 41 de Bleijser, J., 293 Diaz Peiia, M., 539 Disdier, J., 261 Domen, K., 511 Dougal, J. C., 657 Duarte, M. Y., 97, 367 Egawa, C., 321 Engel, W., 617, 631 Eszterle, M., 575 Fernandez-Pineda, C., 647 Flanagan, T. B., 459 Foresti, E., 237 Foresti, M. L., 97 Forster, H., 491 Gabrail, S., 41 Galwey, A. K., 57 Gans, P., 657 Geblewicz, G., 561 Gill, J. B., 657 Gopalakrishnan, R., 365 Grampp, G., 366 Gratzel, M., 197 Green, S. I. E., 41 Guarini, G.G. T., 331 Guidelli, R., 97, 367 Hadjiivanov, K., 37 Harrer, W., 366 Hasebe, T., 187 Hashimoto, K., 87 Heatley, F., 343 Herrmann, J-M., 261 Hidalgo, M. del V., 9 Hill, A., 255 Huis, D., 293 Ige, J., 1 Ikeda, S., 151 Iwasawa, Y., 321 Jaenicke, W., 366 Johnson, G. R. A., 501 Johnson, I., 551 Johnston, C., 309 Jorgensen, N., 309 Kanno, T., 281 Kato, S., 151 Katz, N. E., 9 Keeble, D. J., 609 Kevan, L., 467 Kirby, C., 355 Kiricsi, I., 491 Kiss, I., 367 Klissurski, D., 37 Kobayashi, M., 281 Kondo, J., 511 Kondo, Y., 111 Konishi, Y., 281 Kusabayashi, S., 11 1 tajtar, L., 19 Lambi, J. N., 1 Lawrence, K. G., 175 Leaist, D. G., 581 Lengyel, I., 229 Leyendekkers, J. V:, 397 Leyte, J. C., 293 Lincoln, S. F., 365 Lindner, Th., 631 Malanga, C., 97 Marcus, Y., 175 Maroto, A.J. G., 9 Maruya, K., 511 Mason, D., 473, 483 Matsumura, Y., 87 McAleer, J. F., 441 McMurray, N., 379 Mensch, C. T. J., 65 Mills, A., 379 Mirti, P., 29 Mohamed, M. A. A., 57 Morton, J. R., 413 Moseley, P. T., 441 Muhler, M., 631 Nakamura, Y., 1 11 Nakao, N., 665 Nakayama, N., 665 Narayanan, S., 521 Nazhat, N. B., 501 Wishihara, C., 433 Nishikawa, S., 665 Nomura, H., 151 Norris, J. 0. W., 441 Noszticzius, Z., 575 Nucci, L., 97 Ohtani, S., 187 Olofsson, G., 551 Onishi, T., 51 1 Oosawa, Y., 197 Pelizzetti, E., 261 Peiia, M. D., 539 Pezzatini, G., 367 Piccini, S., 331 Pichat, P., 261 Piekarski, H., 529, 591 P6ta, G., 215 Preston, K. F., 413 Rajaram, R. R., 391 Renuncio, J. A. R., 539 Rochester, C. H., 309 Rubio, R. G., 539 Saadalla-Nazhat, R. A., 501 Saito, Y., 275 Sakamoto, Y., 459 Sakata, Y., 511 Sato, T., 275 Sauer, H., 617 Sawabe, K., 321 Sayari, A., 413 Sbriziolo, C., 207 Schelly, Z.A., 575 Schiffrin, D. J., 561 Schiller, R. L., 365 Schlogl, R., 631 Sellers, R. M., 355 Sermon, P. A., 391 Serpone, N., 261 Shindo, H., 433 Sokolowski, S., 19 Somsen, G., 529 Soriyan, 0. O., 1 Stevens, J. C. H., 165 Stone, W. E. E., 117 Symons, M. C. R., 609 Tanaka, K., 601 Tanaka, K-i., 601 Thomas, J. M., 617, 631 Tofield, B. C., 441 Torres-Sanchez, R-M., 117 Trifiro, F., 237AUTHOR INDEX Uematsu, R., 1 1 1 Uma, K., 521 Unwin, P. R., 473, 483 van Veen, J. A. R., 65 van Wingerden, R., 65 Vasaros, L., 367 Vazquez-Gonzalez, M. I., 647 Viguria, E. C., 255 Vink, H., 133 Viswanathan, B., 365 Walker, R. A. C., 255 Williams, B. G., 617, 631 Williams, D. E., 441 Yoshida, S., 87 Zeitler, E., 617, 631 Zelano, V., 29 Zielinski, R., 151 THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION No.87 Catalysis by Well Characterised Materials University of Liverpool, 11-13 April 1989 Organising Committee: Professor R. W. Joyner (Chairman) Professor A. K. Cheetham Professor F. S. Stone The understanding of heterogeneous catalysis is an important academic activity, which compliments industry's continuing search for novel and more efficient catalytic processes. The emergence of relevant, in particular in situ techniques and new developments of well established experimental approaches to catalyst characterisation are making a very significant impact on our knowledge of catalyst composition, structure, morphology and their inter-relationships.Well characterised catalysts, which will be the subject of the Faraday Discussion, include single-crystal surfaces, whether of metals, oxides or sulphides; crystalline microporous solids, such as zeolites and clays, and appropriate industrial catalysts. The elucidation of structure/function relationships and catalytic mechanism will be important aspects of the scientific programme. Contributions describing novel methods for synthesising well characterised catalysts and also reporting important advances in characterisation techniques will also be welcome. Contributions for consideration by the Organising Committee are invited and abstracts of about 300 words should be sent by 31 May 1988 to: Professor R. W.Joyner, Leverhulme Centre for Innovative Catalysis, Department of Inorganic, Physical and Industrial Chemistry, University of Liverpool, Grove Street, P. 0. Box 147, Liverpool L69 3BX. Full papers for publication in the Discussion volume will be required by December 1988. Dr. K. C. Waugh Professor P. B. WellsTHE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION No. 8 5 Solvation University of Durham, 28-30 March 1988 Organising Committee: Professor M. C. R. Symons (Chairman) Professor J. S. Rowlinson Professor A. K. Covington Dr I. R. McDonald DrJ. Yarwood Dr A. D. Pethybridge Professor W. A. P. Luck Dr D. A. Young The purpose of the Discussion is to compare solvation of ionic and non-ionic species in the gas phase and in matrices with corresponding solvation in the bulk liquid phase.The aim will be to confront theory with experiment and to consider the application of these concepts to relaxation and solvolytic processes. Topics to be covered are: clusters, (c) Gas phase ionic clusters, (d) Liquid phase ionic solutions, (e) Dynamic processes including solvolysis. (a) Gas phase non-ionic clusters, (b) Liquid phase non-ionic Speakers include: H. L. Friedman, B. J. Howard, M. J. Henchman, S. Tomoda, 0. Kajimoto, M. H. Abraham, Yu Ya Efimov, J. L. Finney, P. Suppan, J. P. Devlin, D. W. James, G. W. Neilson, T. Clark, M. L. Klein, J. T. Hynes, G. A. Kenney-Wallace, G. R. Fleming, M. J. Blandamer and D. Chandler. The final programme and application form may be obtained from: Mr. Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN.THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION No. 8 6 Spectroscopy at Low Temperatures University of Exeter, 13-15 September 1988 Organising Committee: Professor A. C. Legon (Chairman) Dr P. B. Davies Dr B. J. Howard Dr P. R. R. Langridge-Smith Dr R. N. Perutz Dr M. Poliakoff The Discussion will focus on recent developments in spectroscopy of transient species (ions, radicals, clusters and complexes) in matrices or free jet expansions. The aim of the meeting is to bring together scientists interested in similar problems but viewed from the perspective of different environments. The Introductory Lecture will be given by G. C. Pimentel and speakers include: L. Andrews, K. H. Bowen, B.J. Howard, L. B. Knight Jr, E. Knozinger, D. H. Levy, J. P. Maier, J. Michl, M. Moskovits, A. J. Stace, M. Takami, J. J. Turner, M. Poliakoff, A. J. Barnes, J. M. Hollas, M. C. R. Symons and P. Suppan. The preliminary programme may be obtained from: Mrs Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN (vii)THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY WITH THE ASSOClAZlONE ITALIANA DI CHIMICA FISICA, DIVISION DE CHlMlE PHYSIQUE OF THE SOCICTf! FRANCAISE DE CHlMlE AND DEUTSCHE JOINT MEETING BUNSEN GESELLSCHAFT FUR PHYSIKALISCHE CHEMIE Structure and Reactivity of Surfaces Centro Congressi, Trieste, Italy, 13-16 September 1988 Organising Committee: M. Che V. Ponec F. S. Stone G. Ertl R. Rosei A. Zecchina The conference will cover surface reactivity and characterization by physical methods: (i) Metals (both in single crystal and dispersed form) (ii) Insulators and semiconductors (oxides, sulphides, halides, both in single crystal and dispersed forms) (iii) Mixed systems (with special emphasis on metal-support interaction) The meeting aims to stimulate the comparison between the surface properties of dispersed and supported solids and the properties of single crystals, as well as the comparison and the joint use of chemical and physical methods. Further information may be obtained from: Professor C.Motterra, lnstituto di Chimica Fisica, Corso Massimo D'Azeglio 48, 10125 Torino, Italy. THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SYM POSl U M Orientation and Polarization Effects in Reactive Collisions To be held at the Physikzentrum, Bad Honnef, West Germany, 12-14 December 1988 Organising Committee: Dr S.Stoke Professor J. P. Simons Dr K. Burnett Dr H. Loesch Professor R. N. Dixon Professor R. A. Levine The Symposium will focus on the study of vector properties in reaction dynamics and photodissociation rather than the more traditional scalar quantities such as energy disposal, integral cross-sections and branching ratios. Experimental and theoretical advances have now reached the stage where studies of Dynamical Stereochemistry can begin to map the anisotropy of chemical interactions. The Symposium will provide an impetus to the development of 3-D theories of reaction dynamics and assess the quality and scope of the experiments that are providing this impetus.Contributions for consideration by the Organising Committee are invited in the following areas: (A) Collisions of oriented or rotationally aligned molecular reagents (B) Collisions of orbitally aligned atomic reagents (C) Photoinitiated 'collisions' in van der Waals complexes (D) Polarisation of the products of full and half-collisional processes The preliminary programme may be obtained from: Mrs Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN. (Viii)FARADAY DIVISION INFORMAL AND GROUP MEETINGS Polymer Physics Group with the 3Ps Group Plastics, Packaging and Printing To be held at the Institute of Physics, 47 Belgrave Square, London on 18 February 1988 Further information from Dr M. Richardson, National Physical Laboratory, Teddington, Middlesex Tw11 OLW Theoretical Chemistry Group Postgraduate Students' Meeting To be held at University College, London on 2 March 1988 Further information from Dr G.Doggett, Department of Chemistry, University of York, York Molecular Beams Group Beam-Photon Interactions To be held at University of Durham on 24-25 March 1988 Further information from Dr J. C. Whitehead, Department of Chemistry, University of Manchester, Manchester M13 9PL Colloid and Interface Science Group with The Society of Chemical Industry and British Radio frequency Spectroscopy Group Spectroscopy in Colloid Science To be held at the University of Bristol on 5-7 April 1988 Further information from Dr R. Buscall, ICI Corporate Colloid Science Group, PO Box 11, The Heath, Runcorn WA7 4QE Polymer Physics Group with the Plastics and Rubber lnstitue Deformation, Yield and Fracture To be held in Cambridge on 11-14 April 1988 Further information from Dr M.J. Richardson, National Physical Laboratory, Teddington, Middlesex Tw11 OLW Annual Congress: Division with Electrochemistry Group Solid State Materials in Electrochemistry To be held at the University of Kent, Canterbury on 12-1 5 April 1988 Further information from Dr J. F. Gibson, Royal Society of Chemistry, Burlington House, London W1V OBN ~~~ ~ Electrochemistry Group with The Society of Chemical industry Electrolytic Bubbles To be held at Imperial College, London on 31 May 1988 Further information from Professor W. J. Albery, Department of Chemistry, Imperial College of Science and Technology, South Kensington, London SW7 2AY Electrochemistry Group with The Society of Chemical Industry Chlorine Symposium To be held at the Tara Hotel, London on 1-3 June 1988 Further information from Dr S.P. Tyfield, Central Electricity Generating Board, Berkeley Nuclear Laboratories, Berkeley, Gloucestershire GL13 9BP Gas Kinetics Group Xth International Symposium on Gas Kinetics To be held at University College, Swansea on 24-29 July 1988 Further information from Dr G. Hancock, Physical Chemistry Laboratory, South Parks Road, Oxford OX1 3QZ Electrochemistry Group with the Electroanalytical Group and the Society of Chemical industry Electrochemical Dynamics To be held at the University of Strathclyde on 5-10 September 1988 Further information from Dr S. P.Tyfield, CEGB, Berkeley Nuclear Laboratories, Berkeley, Gloucestershire GL13 9PB Statistical Mechanics and Thermodynamics Group Dense Fluids To be held at the University of Cambridge on 14-16 September 1988 Further information from Dr P. Francis, Department of Chemistry, University of Hull, Hull HU6 7RXCarbon Group with the Carbon and Graphite Group of the SC1 Carbon 88 To be held at the University of Newcastle upon Tyne on 18-23 September 1988 Further information from The Conference Secretariat, Carbon 88, Society of Chemical Industry, 14/15 Belgrave Square, London SWlX 8PS Division Autumn Meeting: Perspectives in Polymer Chemistry To be held at the University of Birmingham on 20-22 September 1988 Further information from Professor I. W. M. Smith, Department of Chemistry, University of Birmingham, PO Box 363, Birmingham 815 21T JOURNAL OF CHEMICAL RESEARCH Papers dealing with physical chemistry/chemical physics which appear currently in J. Chem. Research, The Royal Society of Chemistry's synopsis + microform journal, include the following : An E.s.r. Study of the Radiolysis of Acetylenic Acids and Esters in a Freon Matrix J. Rhodes and Martyn C. R. Symons (1988, Issue 1) Imbibition of Sodium Nitrate by Zeolite Na-Y at 25 "C and The Solubility of Carbon Dioxide in Mixtures of Water and Acetone Correlation Analysis of the Reactivity in the Oxidation of Aromatic Aldehydes by An E.s.r. Study of Azoalkane Radical Cations Electrochemical Studies of some Nickel(I1) Complexes of the Type [Ni(NNS)(Heterocycle)-1 Christopher Kevin R. Franklin, Barrie M. Lowe Gordon H. Walters (1988, Issue I ) Robert W. Cargill, Donald E. MacPhee and Kenneth Patrick (1988, Issue 1 ) N-Bromoacetamide Louwrier (1988, Issue 1 ) and [Ni2(NNS)~-(Heterocycle)-][C104] (1988, Issue 1 Anita Gupta, Sandhya Mathur and Kalyan K. Banerji (1 988, Issue 1 ) Christopher J. Rhodes and Pieter W. F. Sanat K. Mandal, Parimal Paul and Kamalaksha Nag Influence of the Acid-strength Distribution of the Zeolite Catalyst on the t-Butylation of Phenol Avelino Corma, Hermenegildo Garcia and Jaime Primo (1 988, Issue 1 )

ISSN:0300-9599    

DOI:10.1039/F198884BP021

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I , 1988, 84(2) 367-377 Mechanism of Electrohydrodimerization of 2-Cyclohexen- 1 -one on Mercury from Aqueous Solutions Part 2.-Results obtained for Solutions containing Triton X- 100 M. Yolanda Duarte Department of Chemistry, Nordeste University, Corrientes, Argentina Giovanni Pezzatini and Roland0 Guidelli" Department of Chemistry, Florence University, Florence, Italy The effect of Triton X-100 additions upon waves I and I1 for 2-cyclohexen- 1-one (R) electroreduction on mercury from aqueous solutions has been investigated. The presence of Triton X- 100, while allowing the electron transfer to remain Nernstian, affects the electrode reaction mechanism whenever the latter involves adsorbed reaction partners or else a protonation prior to or during the rate-determining step.In both cases the effect of Triton X-100 is mainly due to the fact that it displaces reaction partners from direct contact with the electrode surface. In the case of protonation, adsorbed water molecules are also regarded as reaction partners, in that they are much stronger proton donors than non-adsorbed ones. The effect of Triton X-100 is particularly remarkable on wave I1 for 6 < pH < 10. Here a gradual increase in Triton X-100 concentration shifts the rate control from the coupling step (2RH' + R2H2) to the preceding protonation step (R*- + HA $ RH' + A-) of the radical anion R*-. A systematic investigation of the mechanism of electrohydrodimerization (EHD) of diactivated alkenes in aqueous media carried out in this lab~ratoryl-~ has shown that adsorbed water molecules are much stronger proton donors than non-adsorbed ones and may protonate the anion radical R'- of any activated alkene R.With diactivated alkenes other than a, P-unsaturated carbonyls (e.g. fumarodinitrile, diethylfumarate and ethyl~innamate)l-~ the protonation of R'- with RH' formation is immediately followed by a further electronation yielding the saturated monomer RH,. Hence, in order to obtain the hydrodimer, the adsorbed water molecules must be completely displaced from direct contact with the electrode surface using an organic surfactant. The use of very strong surfactants, such as Triton X-100, is particularly convenient since they may displace adsorbed water molecules completely from the inner layer at bulk con- centrations sufficiently low (say lo-* mol dmP3) as not to alter appreciably the composition of the solution phase just outside the inner layer.a, P-Unsaturated carbonyls such as chalcone4v and 2-cyclohexen- 1 -ones show unusual behaviour in that their protonated radicals RH' (probably protonated at the oxygen atom) have a greater tendency to dimerize, yielding R2H,, thar, to be further electroreduced to the saturated monomer RH,. Hence, EHD of a,a-unsaturated carbonyls takes place even in the absence of strong surfactants. Even though the effect of strong surfactants upon the electroreduction of these compounds is not so striking as that observed with other diactivated alkenes, it is still significant and serves to evidence the important proton- donating role played by adsorbed water molecules.This paper aims to investigate the effect of Triton X- 100 additions upon the EHD of 2-cyclohexen- 1-one on mercury from aqueous solutions. 367 13-2368 Electrohydrodimerization of 2-Cyclohexen- I -one 1 I I I 2 3 4 5 PH Fig. 1. (a) Plot of < us. pH for mol dm-3 R reduction from citric (0, C,, = 0.2 mol dmP3), formic (m, C,, = 0.01 mol dm-3) and acetic (0, C,, = 0.01 mol dm-3) buffers containing 5 x mol dm-3 R reduction from phosphate buffers (C,, = 0.05 mol dm-3) in the absence of Triton X-100. Drop time td = 2 s; rnol dm-3 Triton X-100. (b) Plot of il us. pH for mercury flow rate rn = 7 x g s-l. - 1.6 w V m m 1 > -1.9 L;;" -1.2 I 1 I I I 1 4 6 8 10 12 us. pH for PH Fig. 2. (a) Plot of mol dmP3 R reduction from 0.01 mol formic (0) and acetic (D) buffers ; the dashed line has a slope of - 60 mV.(b) Plot of ,!Ti1 us. pH for mol dm-3 R reduction from 0.05 mol dmW3 phosphate (0) and 0.01 mol dm-3 6orate (a) buffers. (c) Plot of Eil us. pH plot for to lo-' mol dm-3 NaOH solutions. Triton mol dm-3 R reduction from X-100 concentration = 5 x mol dmP3.M. Y. Duarte, G . Pezzatini and R. Guidelli 369 1 -1.65 w > --- q“ -1.55 w -1.20 1 8 8 % 0 0 0 0 0 8 Q 0 0 . . . . 0 0 0 0 -4 -3 log(Ci/mol dm-3) Fig. 3. Curves (a), (b) and (c) are us. log C;ll plots as obtained from a pH 3.3 0.2 mol dm-3 formic buffer in tfie presence of lop4, 5 x rnol dm-3 Triton X-100. Curves (a’), (b’) and (c’) are I$ us. log C;T. plots as obtained from a pH 8.9 0.2 mol dmP3 borate buffer in the resence of lop4, 5 x lop4 and 2 x us.log C;t plots as obtained from a 0.01 mol dm-3 NaOH solution containing and 2 x mol dmP3 Triton X-100. Curves (a”), (b”) and (c”) are 5 x and 2 x mol dm-3 Triton X-100. Experimental The experimental conditions and the instrumentation employed have been described in Part 1.6 Results Even in the presence of Triton X-100,2-cyclohexen-l-one (henceforth denoted by R) is electroreduced, giving rise to waves I and 11. Here we focus our attention on the changes in the behaviour of these two waves produced by Triton X-100 additions. Wave I In the presence of Triton X-100 the limiting current 6 of wave I starts to decrease with respect to its diffusion-limiting value id at a pH value of ca. 3, becoming vanishingly small at pH 5.5. For comparison, fig.1 shows the limiting current of lop3 mol dm-3 R as a function of pH both in the absence and in the presence of 5 x rnol dm-3 Triton X-100. Note that the experimental curves in this figure depend to some extent on the nature and concentration of the buffer employed. The adsorption prewave which is observed at pH values < 2.5 in the absence of Triton370 1.5 1.0 - n I.-- I 1.2 v \ 1 . 2 u - ?? 0.5 Electrohydrodimerization of 2-Cyclohexen- 1 -oneM. Y. Duarte, G. Pezzatini and R. Guidelli 37 1 -1.60 f - w u m uj > \ =q"" -1.55 -4 -3 log([Triton X-1001/mol dm-3) Fig. 6. Plot of Eil us. the logarithm of Triton X-100 concentration for mol dm-3 R reduction from a pH 9.25 0.2 mol dm-3 borate buffer. the absence of Triton X-100. However, the E{ value is slightly more negative than in the absence of Triton X-100 under otherwise identical conditions.The positive shift of El with increasing CR at constant pH is characterized by a aEI/a log CR value of ca. 20 mV (see fig. 3), as distinct from the 30 mV value observed in the absence of Triton X-100. This implies that the negative shift in Ei following Triton X-100 addition, which is of a few mV at C;it = lop4 mol dm-3, increases progressively with an increase in C:, attaining a value of ca. - 20 mV at Cz = 5 x mol dm-3. The presence of Triton X- 100 also produces a change in the profile of wave I. Thus the difference (Ei-Ei) between the potentials at which the polarographic current is of the corresponding limiting value passes from ca. 41 to ca. 47 mV upon addition of Triton X-100.EI is independent of the buffer concentration at constant pH and Cg. As pH is increased above 3, the limiting current 6 starts to decrease with respect to its diffusion-limiting value id. A plot of log [</(Fd - <)I us. pH at a constant concentration C,, of the acid component of the buffer is approximately linear, with a slope of ca. - 0.9 (see fig. 4). For 6 < fd, an increase in C,, at constant pH and C;T. causes a moderate increase in 6 (see fig. 5), whereas Ei remains substantially unaltered. As in Part 1, 6 was measured at a potential E = Ei- 100 mV. and Wave I1 At pH values > 6, wave I is no longer present and the height of wave I1 attains its maximum diffusion-limiting value, id. Triton X- 100 additions exert a notable effect upon wave I1 in well buffered solutions of pH ranging from 6 to 10.Under these conditions the half-wave potential El1 and the shape of wave I1 are notably affected by a change in Triton X-100 concentration. Thus, an increase in Triton X-100 concentration from lo-* to 2 x mol dm-3 at pH 9.25 causes a negative shift of El1 by ca. 50 mV (see fig. 6), while the dependence of Ei' upon the reactant concentration Cg decreases progressively [see curves (a'), (b') and (c') in fig. 31. For a given Triton X-100 concentration the positive shift of ,?[I with increasing C;T. is more pronounced the higher the pH, as shown in fig. 7. This figure also shows that a decrease in the CR dependence is accompanied by a drawing out of the wave, namely by an increase in E;*--Ei1.An increase in buffer concentration from to lo-' mol dmP3 at constant pH and Cg has no detectable effect upon wave I1 under conditions in which aEi*/a log Cg is positive and greater than ca. 10 mV. On the other hand, under experimental conditions in which372 Electrohydrodimerization of 2-Cyclohexen- 1 -one (cl - O O O O (a l I I I -4 -3 -2 log(Ci /mol dm-3) Fig. 7. The circles on curves (a), (b), (c) and (a’), (b’), (c’) are Eir and (Er - E f ) values, respectively, plotted against log CE as obtained from pH 8, 9 and 10 borate buffers (CHA = 0.1 mol dm-3) containing mol dm-3 Triton X-100. The corresponding solid curves were calculated as described in the appendix. i3E[’/O? log C;E. is vanishingly small, such an increase in buffer concentration produces a small but detectable positive shift of E[’ by 10-15 mV.The above appreciable changes in kinetic behaviour with a gradual increase in Triton X-100 concentration are no longer observed in NaOH solutions, nor are they observed with wave I. This is shown in fig. 3, where the half-wave potentials of wave I [curves (a), (b) and (c)] and of wave I1 in 0.01 mol dm-3 NaOH [curves (a”), (b”) and (c”)] are only slightly affected by an increase in Triton X-100 concentration from to 2 x mol dmP3. In particular, wave I1 in NaOH solutions of concentrations from to 10-1 mol dm-3 is practically unaffected by Triton X-100 additions < 5 x lo-* mol dm-3. Thus its half-wave potential El1 and the rate of increase of this potential with an increase in the logarithm of the bulk reactant concentration, aE[*/O? log C;E.z 20 mV [see curves (a”) and (b”) in fig. 31, are practically identical to those observed in the absence of Triton X-100. Discussion The behaviour of wave I in the presence of Triton X-100 can be rationalized on the basis of the following mechanism : R + H A r RH++A- RH+ +e RH’ Eo k’ 2RH’ --+ R,H,M. Y. Duarte, G. Pezzatini and R. Guidelli EIVvs. SCE -1.55 -1.60 -1.65 I 373 h \ \ \ hl \ \ \ -e \ \ \ \ \ I I I I -1.15 -1.20 -1.25 -1.31 E/V vs. SCE Fig. 8. Plot of log [(I - i/<)/(z/<)i] us. E for 4 x lop3 mol dm-3 R reduction in the presence of mol dmP3 Triton X-100 from a pH 3.7 formic buffer (a) and from 0.01 mol dmP3 NaOH (b). which differs from the corresponding mechanism in the absence of Triton X- 100 only in that the coupling step (1 c ) is not homogeneous.The current potential characteristic corresponding to this mechanism is7,' E = E,, +f-' In [(il - F) $/(ig i,)] +f-' In ([H+]/Kd) + (3f)-' In (2d2k'C;/3D) ( 2 ) where 6 = ($)(77TDtd/ 12)$ and the limiting current 6 is given by: il/(id - 6) = (:)(7nt, k,[H+]/12Kd)~. (3) In the preceding equations Kd = C,[H+]/C,,+ is the dissociation constant of RH+, k' is the rate constant for the homogeneous coupling step and k, has the same meaning as in Part 1. From eqn (2) it follows that: (i3E[/i3 pH){,, c; = - 2.3/f z - 60 mV (i3E;/C? log CE)II,PH = 2 . 3 / 3 f z 20 mV (aE;/C? log &;,pH = - 2 . 3 ( 2 / 3 ) / f z -40 mV. (4) ( 5 ) ( 6 ) Curve ( a ) in fig. 2 and curves (a)-(c) in fig. 3 show that the predictions of eqn (4) and ( 5 ) are satisfied.Eqn (6) cannot be verified within the limits of sensitivity of the polarographic technique, since the changes in 6 produced by a change in buffer concentration at constant pH are so small (see fig. 5 ) as to produce a predicted change in Ei of only a few millivolts. A further confirmation of the validity of eqn (2) is provided374 Electrohydrodimerization of 2-Cyclohexen- 1 -one by the log [(I - f/<)/(f/<)i] vs. E plot in fig. 8, which is linear and has a (60 mV)-' slope. At values < i d , the experimental dependence of log [(/(La - ()] upon pH at constant concentration of the acid component HA of the buffer (see fig. 4) agrees with eqn (3) provided that the rate constant for protonation by hydroxonium ions, k,,,+ [H+], is much greater than that by HA, kv,HA CHA: k~ = 'v, H+LH+1 + 'v, HA 'HA kv, H+LH+]- (7) (8) and predicts a slope of -1 for the plot in fig.4, as actually verified to a first approximation. From the abscissa 4.46 of the intercept of this plot on the pH axis a value of 2.5 x lo4 dm3 rno1-l s-i for the (k,,H+/&)t ratio is immediately obtained. The validity of the approximation of eqn (7) is actually confirmed by the [&/(id - &)I2 vs. CHA plot at constant pH in fig. 5. In view of eqn (3) the slope of this plot equals (0.8 12)12 td kv, HA [H+]/ & and hence provides a (k,,HA/&)+ value of ca. 82 dm3 mo1-l s-5 which, once compared with the (kv, H+/&)+ value previously derived, yields a (k,, H+/kv, HA)' value of ca. 3 x lo2. Note that the (kV,,+/Kd)+ value of 2.5 x lo4 dm3m01-l s-i as obtained in the presence of Triton X-100 is ca.$ of that obtained in its absence (see Part 11). The decrease in (k,, H+/&)' following Triton x- 100 addition can only be partially ascribed to a decrease in the H,O+ concentration in the reaction layer, as produced by the decrease in the absolute value of the negative charge density cM on the metal. The maximum decrease in (k,, ,+/&)+ due to this #,-effect, as estimated on the basis of the Gouy-Chapman theory upon assuming that the reaction layer thickness is much smaller than the diffuse layer thickness, amounts to ca. 50 %. The discrepancy between the (k,, H+/&)' values obtained in the absence and in the presence of Triton X- 100 is likely to be ascribed, at least partly, to the slight potential dependence of 6 and to the resulting inaccuracy in its estimate.At any rate, the notable decrease in the rate of the protonation step (1 a) following Triton X-100 addition, as appears from fig. 1, is mainly due to the displacement of adsorbed water molecules, which practically eliminates the heterogeneous protonation pathway. The behaviour of wave I1 over the pH range from 6 to 10 can be explained by the same mechanism valid in the absence of Triton X-100 (see Part 1): Under these conditions eqn (3) can be written in the form log [il/(td-il)] = log [0.812(td k,, ~+/k$]-pH E; kl R'-+HAeRH'+A- R+e-eR'- kb k' 2RH' + R2H2. However, the gradual addition of Triton X-100 progressively decreases the rate of the protonation step (9 b), so that we pass from rate control by the coupling step (9c), to mixed control by the protonation and coupling steps, and ultimately to control by the protonation step alone.Since the exclusive control by a protonation step in which the main proton donor is water is characterized by an independence of Ei'from pH and reactant concentration C;, the shift of rate control from step (9c) to step (9b) involves a gradual decrease in the C; dependence of El* [see curves (a')-(c') in fig. 31 and in its pH dependence [cf. curve (b) in fig. 2 with the corresponding curve in fig. 1 of Part 13. Moreover, an increase in C;cl at constant Triton X-100 concentration accelerates the rate of the coupling step (9c), second-order in RHO, more than that of the protonation step (9 b), first-order in R*-, causing an increase in the rate control by the latter step and hence a decrease in the C;T.dependence of Ef (see fig. 7). On the other hand, an increase in pHM. Y. Duarte, G. Pezzatini and R. Guidelli 375 at constant Cc and Triton X-100 concentration enhances the bidirectional character of the protonation step (9b) by increasing its backward rate and hence shifts the rate control on the subsequent coupling step. This explains the increase in the C;l. dependence of E[' with increasing pH as shown in fig. 7 (aHc). Agreement between the experimental points in this figure and the corresponding solid curves calculated upon solving the boundary value problem for the mechanism of eqn (9) (see Appendix) is excellent. The gradual increase in the (EII-E!') value from ca.45 to ca. 57 mV with a decrease in pH and an increase in Cc is also satisfactorily accounted for on the basis of the increasing rate control by the protonation step over the coupling step [see curves (a')-(c') in fig. 71. From fig. 7 and curve (b) in fig. 2 it is apparent that the gradual increase in the rate control by the coupling step (9c) with an increase in pH at constant Triton X-100 concentration is accompanied by a progressive negative shift of EI'. Ultimately, at sufficiently high pH values, the pH-dependent rate of the reaction pathway of eqn (9) tends to become lower than the pH independent rate of the parallel reaction pathway (see Part 1): E; R+e+R'- (1W 2R'- + Ri- rds Ri- + 2H20 + R2H2 + 20H-. When this is the case the latter pathway predominates, as actually demonstrated by the attainment of a pH independent Ei' value [see curve (c) in fig.21 and by the 20 mV value for the quantity (aE;'/a log Cg) [see curves (a"Hc") in fig. 31. Moreover, the plot of log [(I - i/t,)/(i/id)'] vs. E is linear and has a (60 mV)-' slope [see curve (b) in fig. 81, in agreement with eqn (8) of Part 1. Note that at these high pH values (pH > 12) wave I1 is practically unaffected by Triton X- 100 additions. This indicates that Triton X-100, which allows the electron transfer to remain Nernstian, affects the electrode reaction mechanism only when the latter involves adsorbed reaction partners or else a protonation prior to or during the rate-determining step. In both cases the effect of Triton X-100 is mainly due to its displacing reaction partners from direct contact with the electrode surface.Naturally, in the case of protonation, adsorbed water molecules are also to be regarded as reaction partners, in that they turn out to be much stronger proton donors than non-adsorbed water molecules. This work was supported by the Consiglio Nazionale delle Ricerche (Progretto finalizzato 'Chimica Fine e Secondaria'). Thanks are due to the Italian Ministry of Foreign Affairs for a fellowship to M.Y.D. during the tenure of which the majority of the present results were obtained. Appendix Consider the reaction mechanism of eqn (9). In the framework of the diffusion-layer approximation, the diffusional problem relative to this mechanism is expressed by the equations : d2a/dy2 = 0 (A 1 4 (A 1b) (A 1 4 JJ = 0: a/b = 8 3 exp[F(E-Ek)/RT] (A 2 4 d2b/dy2 = Af b - Ab c d2C/dy2 = C- A, b + Ad C2 = O with the boundary conditions376 Electrohydrodimerization of 2-Cyclohexen- 1 -one (da/W = -(db/dY) y = l : a = l b=O (dbldy) = 0 where I , = k, d2/D I,, = kbd2/D Id = k'C:d2/D (A 4 4 In the preceding equations x is the distance from the electrode surface, S is the diffusion- layer thickness, D is the diffusion coefficient (assumed to be the same for all species) and C,, C,- and CRH are the space-dependent concentrations of R, R'- and RHO.Note that while in eqn (A 1 b) we have assumed stationary-state conditions (ab/at x 0), in eqn (A 1 c) we have also made the further assumption that d2c/dy2 = 0. The latter assumption is a good approximation for species which are both produced and consumed by very fast homogeneous reactions, whereas they are neither produced nor consumed by heterogeneous reactions. (A 5 ) The current i is given by where id is the diffusion-limiting current for a one-electron wave and the subscript zero refers to y = 0.Integration of eqn (A 1 a) with the boundary condition (A 3a) (A 6) yields where use has been made of eqn (A 5). If we express c as a function of b by means of the last two terms of eqn (A 1 c) and if we replace the expression so obtained into eqn (A 1 b), a differential equation is obtained which is readily integrated between y = 0 and y = 1, yielding ly = i/i, = (da/dy), (daldy) = (da/dy), = 1 -a, = v/ (dbldy); = (A:/&) b, + I , b: - (&/6I: If)[(A: + 41f Ad bo)f - A:]. (v//b,)2 = (I,/z2)(z2 + z - [( 1 + 42);- 1]/6j (A 7) In obtaining this equation use has been made of the boundary conditions (A 3b) and (A 3c).Upon combining eqn (A 7) with eqn (A 2b) and (A 6) we have (A 8) with r = I , I , / I : z = rb,. Using eqn (A 2a) and (A 6) b, is given by b, = t9-'(1- v/). (A 10) From eqn (A 8) and (A 10) it follows that E,-Ei-(2f)-' ln(k,d2/D) =f-l ln[(l- v / ) / v / ] + (2n-l In { 1 + 1 / z w - [( 1 + 4zJf - 1]/(6zE)) (A 1 1) where E, and z, are E and z values corresponding to a well defined i/id ratio. UponM. Y. Duarte, G. Pezzatini and R. Guidelli 377 substituting b, from eqn (A 10) into eqn (A9b) and eliminating (E,--E;) between the logarithmic form of the resulting equation and eqn (A 11) we obtain log Cg + log ( k , ki Da/kEd) = -log + f log { z t + z , - [( 1 + 4 ~ ~ ) ; - 11/61 (A 12) where use has been made of the definitions in eqn (A 4).Eqn (A 11) and (A 12) can be used to determine the dependence of (E, - E;) upon log Cg for any value ry of the ratio i/i,. For this purpose a series of arbitrary values is ascribed to z , on the right-hand side of eqn (A 11) and (A 12), and the corresponding pairs of values of [(E, -I?;) - (2fl-l In (k,d2/D)] = Y and of [log Cg + log(k, kf DB/kiG)] = X are calculated. In this way curves of El1 and of (I?kl-Ef) us. log Cg are readily calculated, apart from additive constants along both the Y and X axes. From general acid-base catalysis, the rate constants k, and k, in the mechanism of eqn (9) have the general form: kf = ' f , H 2 0 [ H 2 0 1 + ' f , H A C H A kb = kb, OH-[OH-] -k kb, *CA-. In fig. 7 C,, = [H,BO,] is kept constant and hence the same is true for k,. It follows that the additive constant along the Y axis is the same for all curves in this figure. The slight dependence of wave I1 upon buffer concentration indicates that we can set k , "N k,, OH- x [OH-] as a first approximation. This implies that a unitary increase of pH involves a positive shift by two units along the X axis. Fig. 7 was obtained by first fitting the calculated Y+ us. X curve to the experimental points relative to pH 9 [curve (b)]. The calculated Y; us. Xcurve was then shifted by two units in the positive (negative) direction in order to superimpose it on the experimental points at pH 10 (pH 8). Curves (a'), (b') and (c') were finally obtained by plotting calculated curves of (Yi- Y:) against the same X axis employed for the Y; vs. X calculated curve. References 1 R. Guidelli, G. Piccardi and M. R. Moncelli, J . Electroanal. Chem., 1981, 129, 373. 2 M. R. Moncelli, F. Pergola, G. Aloisi and R. Guidelli, J . Electroanal. Chem., 1983, 143, 233. 3 C. Amatore, R. Guidelli, M. R. Moncelli and J. M. Saviant, J . Electroanal. Chem., 1983, 148, 25. 4 M. R. Moncelli, L. Nucci, P. Mariani and R. Guidelli, J . Elecfroanal. Chem., 1984, 172, 83. 5 M. R. Moncelli, L. Nucci, P. Mariani and R. Guidelli, J . Electroanal. Chem., 1985, 183, 285. 6 M. Y. Duarte, C. Malanga, L. Nucci, M. L. Foresti and R. Guidelli, J . Chem. SOC., Faraday Trans. 1, 7 L. Camacho, J . Electroanal. Chem., 1984, 177, 59. 8 J. M. Rodriguez-Mellado, M. Blazquez, L. Camacho and J. J. Ruiz, J . Electroanal. Chem., 1985, 1988, 84, 14. 190.47. Paper 612500; Received 31st December, 1986

ISSN:0300-9599    

DOI:10.1039/F19888400367

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1988, 84(2), 379-390 Characterisation of an RuO, xH,O Colloid and Evaluation of its Ability to Mediate the Oxidation of Water Andrew Mills* and Neil McMurray University College of Swansea, Department of Chemistry, Singleton Park, Swansea SA2 8PP A colloid of Ru0,-xH,O supported by polybrene was prepared, cha- racterised and assessed as an 0, catalyst. Thermal analysis (t.g.a. and d.t.g.a.) of RuO, .xH,O precipitated from the colloid indicated the presence of both weakly and tightly bound water. Dynamic light scattering indicated that the coagulated colloidal particles were large (d = 825 nm) and positively charged. Transmission electron microscopy demonstrated that the colloidal particles were themselves aggregates of crystallites too small (< 10 nm) to be clearly resolved.The colloid proved unstable towards coagulation under conditions of high electrolyte concentration (2 mol dm-3) even when the electrolyte was H,SO,. In the presence of CeIV ions the colloid did show some activity as an 0, catalyst (0, yield = 73 %) but also underwent some anodic corrosion to RuO, (27%). At low concentrations of CeIV ions (4.5 x lo5 mol dm-3) and colloid ([RuO,] = 0.02-0.001 25 g dm-3) the col- loid appeared to mediate the oxidation of polybrene over that of water by the Ce" ions. Kinetic studies performed under these conditions and in the presence of a high constant background concentration of polybrene (0.015 g dmW3) showed the kinetics to be biphasic with an initial fast step (associated with charging of the catalyst) followed by a second step which was proportional to the concentrations of both Ce'" ions and colloid.Under conditions where no extra polybrene was added to dilutions of the colloid some 0, evolution was observed (ca. 20%) and the kinetics of Ce" disappearance once again appeared biphasic, although more complicated and difficult to interpret. A major approach1 towards the conversion and storage of solar energy into chemical energy is the development of a photochemical system capable of cleaving water into hydrogen and oxygen efficiently, i.e. sunlight 2H20 )2H, + 0,. (1) photochemical system It is generally recognized' that an essential step towards achieving this objective is the discovery and development of catalysts which are stable, specific and fast-acting towards water reduction and oxidation, respectively.A great deal of progress has been made in the area of water reduction catalysts (H, catalysts), and there are now materials such as colloidal platinum which are stable, specific and able to act in the ps time domain., In contrast, progress in the area of water oxidation catalysts (0, catalysts) has been slow and, so far, most of the materials tested appear t o be either inactive as 0, catalysts or themselves undergo anodic corrosion when subjected to the strong oxidising conditions [E"(O,/H,O) = 1.23 V us. NHE) necessary to oxidise water to oxygen. For example, although ruthenium dioxide hydrate (RuO, - xH,O) has, for several years, found frequent use as a catalyst for the oxidation of ~ a t e r ~ - ~ by a strong oxidant such as CeIV ions (2) RuO,.zH ,O 2H20 + 4Ce4+----+4Ce3+ + 4H+ + 0, 379380 RuO, * xH,O Colloids and the Oxidation of Water we have7-' now established that for all commercial samples in the presence of CeIV ions some degree of anodic corrosion to ruthenium tetroxide (RuO,) occurs.Indeed, with the majority of samples tested, anodic corrosion, and not water oxidation, was the only process ob~erved.~ In addition we have shown that Ru0,-xH,O powder may be ' activated ' as a catalyst in reaction (2) and passivated towards anodic corrosion simply by heating the sample at ca. 145 "C in air for 5 h.' The 0, catalyst produced using this procedure is not, however, very fast-acting. For example, using 10mg of catalyst dispersed in a solution (100 cm3) of sulphuric acid (0.5 mol dm-3) containing CeTV ions (3.6 x In the search for faster-acting 0, catalysts several groups10-12 have produced colloids of RuO;xH,O; however, little effort has been made to characterise these colloids so as to help evaluate their potential as 0, catalysts.An ideal 0, catalyst colloid should be stable towards anodic corrosion, active towards water oxidation and stable towards coagulation at high ionic strength. In this paper we report the findings of a detailed characterisation of a real colloid (a colloid of RuO;xH,O supported by polybrene) similar to one previously reported as an 0, catalyst," and yet, from our work, apparently far from ideal as such a catalyst. mol dm-3), ti = 104 s. Experimental Materials Polybrene ( 1,5-dimethyl- 1,5-diazaundecamethylene polymethobromide, hexadimethrine bromide) and cerium(1v) sulphate tetrahydrate ( > 99 %) were obtained from Aldrich.Solutions containing Ce(SO4);4H,O were made up from the solid using 0.5 mol dm-3 H,SO, (AR; BDH) and standardised spectrophotometrically [e(Ce4+) = 5580 dm3 mol-1 cm-l at Amax = 320 nm].13 The solid ruthenium tetroxide (RuO,) used in the colloid preparation was made using a modified version of a procedure described by Connick and H~r1ey.l~ Thus anhydrous RuO, was prepared by distillation, in a stream of nitrogen, from a solution containing a dispersion of RuO, - xH,O (Johnson Matthey), potassium permanganate (AR; BDH) and dilute sulphuric acid. The volatile RuO, (m. pt 24.4-24.8 "C; literature value15 25.4 "C) was passed through a trap of anhydrous magnesium perchlorate and collected in a U-tube immersed in liquid nitrogen.In this work all the sdutions were made up using doubly distilled deionised water. Methods All u.v.-visible absorption spectra were recorded using a Perkin-Elmer Lambda 3 spectrophotometer. The gravimetric analyses (t.g.a. and d.t.g.a.) were performed using a Stanton Redcroft TG-750 instrument and a Linseis chart recorder (type LS4). Using this apparatus a sample (ca. 7 mg) was usually heated from ambient temperature (typically ca. 22 "C) to 900 "C at a rate of 20 "C min-' and the percentage weight loss was recorded as a function of time with an x / t chart recorder. In all cases N, was flowing at a rate of 25 cm3 min-l. Transmission electron micrographs of the colloid were recorded using an electron microscope (1 20 C TEM-SCAN) manufactured by JEOL.Copper grids, covered with a carbon support film, were loaded with sample as follows : a grid was placed on a droplet of colloid lying on a surface of dental wax; after ca. 30 min the grid was lifted off using a pair of tweezers, the excess liquid was drawn off with a piece of tissue paper and the grid was then left to dry (30 min) before being used in the TEM-SCAN. The dynamic light scattering experiments were performed by Dr J. R. Danvent and Ms Anne Lepre (Birkbeck College, London) using a Malvern Instruments Zeta-Nanosizer. A ' test system ' was devised to assess the ability of any sample of RuO, - xH,O, powder or colloid, to act as a catalyst or undergo corrosion when exposed to a strong oxidisingA .Mills and N. McMurray 38 1 agent such as CeIV ions. The 'test system' comprised a N, cylinder which provided a continuous flow (fz 180 cm3 min-l) of gas through two Dreschel bottles (each 125 cm3), and an oxygen membrane polarographic detector (0,-MPD) coupled in series. Of the Dreschel bottles, the first contained 100 cm3 of a CeIV solution (3.6 x rnol dm-3 in 0. I mol dm-3 H,SO,) and also had the additional feature of a rubber septum through which the stock colloidal RuO,.xH,O suspension (20 cm3; containing ca. 0.4 g dm-3 RuO, and 0.2 g dm-3 polybrene) were injected, and the second contained 100 cm3 0.1 mol dm-3 NaOCl in 1 mol dm-3 NaOH [used to trap, in the form of per-ruthenate (RuOJ any RuO, produced]. The percentage corrosion was calculated using the equation where N(Ru0;) is the number of moles of RuO, trapped in the hypochlorite solution (measured spectrophotometrically)16 and N(Ru0, * xH,O) is the number of moles of RuO, - xH,O injected.The percentage corrosion calculated using eqn (3) represented a minimum value, since RuO, also attacked the rubber septum and the glassware (as evidenced by blackening of both) before reacting in the hypochlorite trap. Owing to the high solubility of RuO, (ca. 20.3 g dm-3, at 20 OC),15 a long time (6-8 h) was required to flush out most of the RuO, produced (> 98 YO) following colloid injection, since the time taken to sweep out half of the RuO, produced was found to be ca. 70 min. The 0,-MPD has been described el~ewhere'~ and allowed the determination of the number of moles of 0, produced, following catalyst injection.The percentage 0, yield corrosion = [N(RuO,)/N(RuO, -xH,O)] x 100 YO (3) was calculated using the equation moles 0, produced x 400 moles CerV consumed 0, yield = YO (4) where the number of moles of CeIV consumed was determined spectrophotometrically from the drop in absorbance of the CeIV solution following catalyst injection. Catalyst Preparation The colloid of RuO, .xH,O supported by polybrene was prepared as follows : RuO, (200 mg) was dissolved in 200 cm3 water contained in a stoppered flask in an ice bath. This solution was stirred under these conditions for 30 min, during which time no change in the u.v.-visible absorption spectrum of the dissolved RuO, was observed. At the end of this period, with the solution still briskly stirred, an equal volume of polybrene solution (0.4 g dmP3) at 0 "C was added quickly.A reaction commenced immediately, with the solution darkening rapidly and taking on a greenish hue. The reaction appeared to be the reduction of RuO, (yellow/green) to colloidal Ru0,-xH,O (black) by the polybrene which also acted as a colloid support, preventing coagulation. This process was best observed by u.v.-visible spectrophotometry, and fig. 1 shows the observed change in absorption spectrum of an aliquot of a freshly mixed reaction solution placed directly into a sealed quartz spectrophotometer cell and thermostatted at 30 "C, as a function of time. From fig. 1, the reduction of RuO, (A,,, = 309 and 385 nm)" appeared to be complete within ca.7 h. In the catalyst preparation the reaction mixture was allowed to reach ambient temperature (ca. 20 "C) and a series of spectral changes identical to those illustrated in fig. 1 were observed but over a longer timescale, 48 h being required before the reaction went to completion. Using a similar procedure, Minero et al.ll reported that the resulting colloid is green with a maximum absorbance at ca. 400 nm. In our work, however, the maximum absorbance observed in this wavelength region (A,,, = 385 nm) was due to unreacted RuO, (see fig. 1) and the spectral profile of the colloid was always of the form illustrated in fig. 1 (e) after all the RuO, had reacted. This u.v.-visible spectrum remained unchanged over a period of382 RuO, * xH,O Colloids and the Oxidation of Water I I I I I I I 200 300 400 500 600 700 900 wavelength/ nm Fig.1. U.v.-visible spectra of a reaction solution thermostatted at 30 "C, containing RuO, (0.5 g drn-$) and polybrene (0.2 g dm-3) and recorded at 100 min intervals following mixing [for curve (a) t = 01. several months and there was no evidence of precipitation over this period. The appearance of the RuO;xH,O colloid was that of a black solution, transparent when diluted, with no trace of turbidity, but with a slight green coloration in transmitted light. Results and Discussion Catalyst Characterisation Determination of Ruthenium Metal Content The ruthenium metal content of the colloid was determined spectrophotometrically as the perruthenate ion (RuO,; A,,, = 385 nm, E~~~ = 2162 dm3 mol-1 cm-l) using the procedure developed by Larson and Thus, a 5 cm3 aliquot of colloid was made up to 50 cm3 in a volumetric flask with a solution containing sodium hypochlorite (NaOCl, 0.1 mol dm-3) plus NaOH (1 mol dm-3) and left ovenight to ensure complete oxidation of the RuO;xH,O to RuO; by the NaOCl.The final absorbance of the RuOh solution at 385 nm was measured relative to a blank consisting of 5 cm3 of 0.2 g dm-3 polybrene plus 45 cm3 of 0.1 mol dm-3 NaOCl in 1 mol dm-3 NaOH. This procedure was repeated several times, and from the appropriate Beer's law calculations the colloid was found to contain (2.8 +, 0.3) x mol dm-3 ruthenium metal, a figure which compared favourably with the value calculated (3 x mol dmW3) from the amount of RuO, added in the catalyst preparation.The concentration of RuO, in the colloid was calculated, from the Ru content, to be 0.37k0.04 g dm-3. Thermal Analysis of RuO, * xH,O Precipitated from the Colloid A 50 cm3 aliquot of the colloid was coagulated by the addition of an equal volume of 0.1 mol dm-3 H,SO,, precipitation was completed by centrifugation (126 g for 10 min) and the excess liquid was then decanted. The precipitate was then washed by dispersing it in 150 cm3 of 0.1 mol dm-3 H,SO, using stirring and ultrasound (4 min), followed by centrifugation and decantation of the resulting supernatant. The washing procedure, aimed at removal of the polybrene support, was repeated four times with 0.1 mol dme3 H,SO, and four times with 1 x mol dm-3 H,SO,.(Water could not be employed in this latter washing process since peptisation was found to occur at low ionic strengths.) The washed material (ca. 20 mg) was then allowed to dry in air at ambient temperature (ca. 20 "C) for a period of 10 days. Elemental analysis of the product revealed <4% C and <0.6 YO N by weight.J. Chem. SOC., Faraday Trans. I , vol. 84, part 2 Plate 1 (c 1 (dl Plate 1. Transmission electron micrographs of the colloid [(a) x 33 000, (b) x 2500001, the colloid coagulated by Na,SO, [(c) x 1000001 and a commercial sample (Johnson Matthey) of RuO;xH,O heated at 110 "C in air for 5 h [(d) x 1000001. A. Mills and N. McMurray (Facing p . 383)A. Mills and N. McMurray 100 90 383 - - B Fig. 2. T.g.a. (a) and d.t.g.a. (b) curves recorded for samples of RuO;xH,O obtained: (A) by precipitation of the colloid and (B) from a commercial source (Johnson Matthey).Fig. 2 shows the t.g.a. and d.t.g.a. profiles recorded for the precipitated Ru0,-xH,O [fig. 2(A)] and a commercial sample of RuO;xH,O (Johnson Matthey batch no. 061 174 [fig. 2(B)], and both are broadly similar. From fig. 2 both samples appear to be highly hydrated, and from their total percentage weight losses the values for x in the simple formula RuO;xH,O may be calculated as (A) x = 4.5 (colloidal) and (B) x = 3.2 (Johnson Matthey). The d.t.g.a. profiles for both samples are very similar and indicate the presence of two very distinct types of water in the samples, namely (i) loosely bound, probably physically absorbed water (d.t.g.a.Tmax = 80 "C) and (ii) tightly bound, probably chemically bonded water (d.t.g.a. Tmax x 220 "C). Transmission Electron Microscopy, Electron Diffraction and Energy Dispersive Analysis Transmission electron micrographs of the colloid (a) and (b), coagulated colloid (c) and a commercial sample of RuO, -xH,O (heated to 110 "C for 5 h, in air) ( d ) are illustrated in plate 1. The last transmission electron micrograph has been included for comparison with that of the coagulated colloid since it appeared to possess a similar stability towards anodic corrosion and activity as an 0, catalyst. From plate 1 and additional micrographs it appeared that all the samples of RuO;xH,O used were not composed of regular or well defined particles, but rather contained aggregates of crystallites too small ( c 10 nm) to be resolved readily. Electron diffraction experiments on both dispersed and coagulated specimens produced no discernible diffraction pattern, corroborating the t.e.m.evidence of their amorphous natures. Qualitative elemental analysis was performed by energy dispersive analysis on all the samples illustrated in plate 1, and confirmed that the subject material contained ruthenium and was not foreign matter or an artifact.384 RuO, - xH,O Colloids and the Oxidation of Water Dynamic Light Scattering It was found, using a Malvern Instruments Zetasizer 2c, that the stock colloid did not scatter light sufficiently to permit an estimation of the mean particle size or potential. Minero et al., however, estimated the hydrodynamic radii of their polybrene-stabilised RuO, colloid to be ca. 30 nm, using a dynamic light scattering technique.ll Interestingly the coagulated colloid, dispersed in water at pH 7, did scatter enough light to allow its mean particle size and potential to be estimated; these were determined to be 825 nm and 40 mV, respectively.The point of zero charge (P.z.c.) for RuO;xH,O powder is usually at pH < 3, therefore at pH 7, the Ru0,-xH,O colloid might be expected to exhibit a negative surface charge and thus a negative zeta potential. The observed positive zeta potential can be reconciled with the above if an overall superequivalent of cationic polybrene is adsorbed on the surfaces of the aggregate particles. Coagulation Studies In any chemical or photochemical study involving an 0, colloidal catalyst it may be necessary to subject the catalyst to conditions of appreciable electrolytic concentration (say > mol dm-3).For example, in order to evaluate the catalytic activity of an RuO, - xH,O colloid for water oxidation, Minero et a1.l' used a test system of CeIV ions in 0.1 mol dm-3 H,SO,, a high-ionic-strength medium. Ideally therefore, before any colloid is used as a catalyst, some assessment should be made of its stability towards coagulation in the presence of the intended high-ionic-strength medium. In our work two basic techniques were employed in order to determine (a) the maximum electrolyte concentration (for a variety of electrolytes) at which the colloid still possesses long-term (220 h) stability and (b) the timescale over which the coagulation occurs in a high-ionic-strength medium such as 0.1 mol dmd3 H,SO,.The results of this work are as follows. 20 h Coagulation Experiments. The method employed was based on that developed by Furlong et al.,' The colloid studied contained 0.02 g dm-3 RuO,, 0.015 g dm-3 polybrene, in order to simulate the catalyst and polymer concentrations reported by Minero et a1.l' in their CeIV kinetic experiments. The extent of coagulation induced over a 20 h period by the addition of H,SO,, Na,SO, and MgSO, was measured as a function of electrolyte concentration. The experiments with the last two electrolytes were conducted at pH 7. Solutions containing colloid and electrolyte at various concentrations were prepared in 20 cm3 glass sample bottles and allowed to stand at 20 "C for 20 h.Each solution was then transferred to a 15 cm3 centrifuge tube and spun at 126 g for 20 min before the top 3 cm of liquid was drawn off andits absorbance (A,) measured at A = 300 nm. A Stokes law calculation showed that aggregates with a radius > 120 nm should have been swept out of the liquid sample taken following centrifugation." A measure of the stability of the colloid towards coagulation was calculated using the equation stability = (AJA,) x 100% ( 5 ) where A , = absorbance of colloid with electrolyte present and A , = absorbance of colloid with no electrolyte present. Fig. 3 shows the variation of the observed colloid stability us. the negative logarithm of the cation concentration, from which the following is seen.(a) The colloid was completely coagulated even by very low concentrations of electrolyte (< 1 x mol dmP3). This suggests that the polybrene present was rather ineffectual as a steric stabiliser and that a large part of the colloid's stability was electrostatic in origin. (b) Coagulation385 A. Mills and N. McMurray 120, .tL 0 2.6 -0-- 100 x .- 60 n 0 20 0 2.8 3.0 3.2 3.4 3.6 3.8 4.0 - 4 .O -log ([cation]/mol dmd3) Fig. 3. Plot of the colloid stability (YO) us. the negative logarithm of the cation concentration of the coagulating electrolyte. The cations used were H' (O), Na' (a) and Mg2+ (0). of the colloid by the neutral salts (Na,SO, and MgSO,) occurred in almost the same region of ionic strength. Given that the colloid was most likely stabilised electrostatically, this result implies that, in accordance with the Schultz-Hardy rule, counter-ions surrounding the colloid were predominately anionic and that the colloid particles were, therefore, cationic.The measured positive zero potential of the colloidal material at pH 7 using dynamic light scattering supports this conclusion. (c) Coagulation of the colloid by H,SO, occurred at a slightly higher (ca. 1.5 times) ionic strength than that found for the neutral salts Na,SO, and MgSO, at pH 7. This enhancement in colloid stability towards coagulation was most likely due to the potential-determining action21 of the protons present on the colloid particles. At a pH below the P.Z.C. of the RuO, .xH,O, as in this work, the surface of the particles would be highly protonated and, therefore, electrostatically more stable towards coagulation.Rate of Colloid Coagulation. Several groups, including Minero et al.," employed CerV ions in H,SO, (30.1 mol dm-3) to test [uia reaction (2)] the activity of their RuO;xH,O colloids. It was decided, therefore, to study the variation of the rate of coagulation of our colloid in 0.1 mol dm-3 H,SO, as a function of its initial concentration. The RuO;xH,O colloid stock solution was used to prepare a range of colloid concentrations (0.02-0.001 25 g dm-3 RuO, with 0.01 5 g dm-3 polybrene). A fixed aliquot (2.5 cm3) of each solution was stirred, at a constant rate, in a 1 cm quartz spectrophotometric cell thermostatted at 30 "C. The process of coagulation was initiated by an injection of 5 x lo-, cm3 of 5 mol dm-3 H2S0, solution which gave (within 10 s) a concentration of H,SO, in the cell of 0.098 mol dm-3.Observation of the progress of coagulation was achieved by following the change in absorbance (at 1 = 320 nm) of the colloid as a function of time and a typical trace is shown in fig. 4. This method of monitoring the progress of coagulation exploits the fact that the absorption/scattering spectrum of the colloid decayed as flocculation proceeded such that dA/dA, over the range 250-400 nm, tended to zero. It was assumed that when the absorbance of the suspension had fallen by half the value it was to fall over the complete coagulation process (AA), a constant degree of coagulation had occurred. The time taken, in s, for this constant degree of coagulation [t(AA/2)] was plotted as a function of the negative logarithm of the colloid concentration and the result is shown in fig.5. From the value of the gradient (=0.99 kO.01) of the386 RuO, - xH,O Colloids and the Oxidation of Water I 1 I I 100 200 300 40 0 ti s Fig. 4. Plot of the absorbance (A = 320 nm) of the colloid (0.01 g dm-3 [RuO,], 0.015 g dm-3 [polybrene]) us. time following injection (at t = 0) of H,SO, (final [H,SO,] = 0.098 mol dm-3). '3 2 1 -log([R~O~]/gdm-~) Fig. 5. Plot of the logarithm of the time taken for the absorbance of the colloid (at 320 nm) to diminish by half, i.e. log [t(AA/2)] following addition of H,SO, (0.098 rnol dm-3) us. the negative logarithm of the colloid concentration, expressed in g dm-3. straight line shown in fig.5, it would appear that coagulation of the colloid is a second- order process. This result is in good agreement with the theory of diffusion-controlled flocculation developed by Smoluchowski,2' which predicts that the coagulation of primary colloidal particles will be second order provided sufficient electrolyte is present to remove the repulsive energy barrier. Interestingly, although it is apparent from fig. 4 and 5 that coagulation in 0.1 mol dm-3 H,SO, can be rapid at high concentrations of colloid, the time for coagulation was always found to be at least 100 times that for the colloid to mediate the reduction of any CeIV ions present. Thus, although not ideal, it does appear that for this colloid a system comprised of CeLV ions in 0.1 mol dm-3 H,SO, can be used to help evaluate, via reaction (2), its ability to mediate the oxidation of water.Corrosion of the Colloid by CeIV ions and Stoichiometry of Oxygen Release Using the 'test system ' described in the Experimental section, 20 cm3 of the stock colloid (0.37 g dm-3 RuO,, 0.2 g dm-3 polybrene) were purged with N, to remove 0, and then injected into the first Dreschel bottle of the test system containing 3.6 x mol dm-3A . Mills and N. McMurray 387 Celv ions in 0.1 mol dm-3 H,SO,. Any 0, (due to catalysis) or RuO, (due to corrosion) produced following catalyst addition was swept out of the reaction vessel by a continuous stream of N, and then analysed (see Experimental section for details). The percentage corrosion and percentage 0, yield exhibited by the colloid were determined, using eqn (3) and (4), to be 27 YO and 73 %, respectively.The resistance of the catalyst towards corrosion was greater than might have been expected in the light of its apparent high degree of hydration. This is because it has been established that the percentage corrosion of RuO, - xH,O increases with increasing degree of hydrati~n.~ Although the colloid exhibited a high degree of catalytic activity towards 0, evolution, the appreciable degree of corrosion which occurred concomitantly makes it far from ideal as an 0, catalyst. Kinetics of eelv Reduction A study of the kinetics of CeIV reduction mediated by the colloid was undertaken using a stopped-flow technique. The stock colloid solution (0.4 g dm-3 RuO,, 0.2 g dm-3 polybrene) was diluted with water to produce a variety of dilute colloid concentrations (0.02-0.001 25 g dm-3 RuO,) which allowed the reduction of CeIV ions to be monitored spectrophotometrically.[Note that with these dilute colloids the RuOJpolybrene ratio (w/w) was constant (= 2).] Previous workers5~ l1 have suggested that a high [Ce"]/[RuO,] ratio favours corrosion. In the previous section we established with the test system that 0, yields of ca. 73 O h are achieved using a [Ce4+]/[Ru0,] ratio = (3.6 x lop3 mol dm-3)/(0.08 g dm-3). Thus, in order to favour 0, evolution, rather than catalyst corrosion, in our kinetic studies we employed a sufficiently small CeIV concentration (4.5 x mol dm-3) to ensure that for any of the dilute colloids used the [Ce4+]/[Ru0,] ratio was always less than or equal to that used in the test system.The observed kinetics of the reduction of the CeIV ions appeared to be biphasic. The initial phase was a fast, high-order (> 1) process, possibly corresponding to the oxidation or charging of the colloidal particles. The second phase was a slower process which appeared to obey first-order kinetics. A plot of the logarithm of the first-order rate constant for this second phase (log k,) us. -log [RuO,] produced a good straight line with a gradient of 1.67. This result implied that for the second phase the rate of reaction ( - d[Ce4+]/dt) was proportional to ([RuO,)"[polybrene]"), where n + m = 1.67. How- ever, if polybrene was not involved in the second phase of the reaction (i.e. rn = 0), then it is likely that a major reaction product would have been 0,.In order to check this, 10 cm3 of a Ce'" solution (9 x mol dm-3) in 0.1 mol dmP3 H,SO, were placed in a cylindrical quartz reaction cell, incorporating in its base an 0,-MPD for dissolved 0, mesurernent.,, This solution was purged with N, for 20 min, sealed and, after a few minutes, 10 cm3 of stock colloid (0.0025 g dmP3 RuO,) also purged with N,, were added to produce a final colloid concentration of 0.001 25 g dm-3 RuO,. Fig. 6 shows the observed changes in current from the 0,-MPD [curve (b)] as a function of time as well as the decay of the CeIV absorbance [fig. 6(a)] observed with the stopped-flow apparatus under identical reaction conditions. From the 0,-MPD, it was found that the maximum amount of 0, observed under these reaction conditions corresponded to an 0, yield of only 20 %.This value for the 0, yield is in marked contrast to that of 73 O/O determined above using the test system (in which the CeIV, RuO, and polybrene concentrations were ca. 70 times greater). The reasons for these two very different 0, yields remain, as yet, unclear. However, it is clear that in our stopped-flow study of the reduction of CeIV ions, the major reaction is not water oxidation but, most probably, polybrene oxidation (i.e. rn # 0). Colloidal RuO, appears to be able to mediate both these oxidation reactions since, in the absence of RuO,, the oxidation of either polybrene or water by CeIV ions was found to be very slow. Since the oxidation of polybrene appears in the rate law we388 RuO; xH20 Colloids and the Oxidation of Water Fig.6. Reduction of CeTV ions (4.5 x mol dm-3) in 0.1 mol dm-3 H,SO,, mediated by the colloid ([RuO,] = 0.001 25 g dmP3, [polybrene] = 0.00063 g dm-3). The decay of the CeIV absorbance (a) was monitored using a stopped-flow technique. The concomitant evolution of 0, (b) was monitored using an 0,-MPD. The maximum change in current [Ai(O,-MPD)] observed corresponded to a 20% 0, yield. I I I I I t l s Fig. 7. Reduction of CerV ions (4.5 x mol dmP3 H,SO,, mediated by the colloid ([RuO,] = 0.01 g dm-3) with added polybrene ([polybrene] = 0.01 54 g dm-3. Plot of the negative logarithm of the observed change in absorbance of the Ce'" ions us. time. 0.5 1.0 1.5 2 .o must assume that either its coverage of the surface of the colloidal particles is low or that the polybrene counterions (Br-) are involved in the oxidation process.Although it is unclear what happens to the polybrene during oxidation, we did not find any evidence for bromine formation due to the oxidation of the Br- counterions. Minero et u1.l' reported the details of a kinetic study of Ce'" reduction mediated by an RuO, - xH,O/polybrene colloid, under reaction conditions in which ' dioxygenA . Mills and N . McMurray 389 -log ( [ RuO, 1 /g dm - 3 ) Fig. 8. Plot of the logarithm of the first-order rate constant for the second phase of the CeIv reduction kinetics us. -log [RuO,] ; [polybrene] was fixed at 0.01 5 g dm-3. evolution should be the predominant pathway '. These reaction conditions are similar to those employed in our initial kinetic study reported above (i.e.[Ce'"] = 4.5 x loe5 mol dm-3 in 0.1 mol dm-3 H,SO,, [RuO,] = 0.02-0.0005 g dmP3) with the exception that Minero et al. employed a fixed polybrene concentration (0.015 g dm-3) for all values of We repeated the kinetic study of Minero et a/." using our colloid and found, once again, that the kinetics of the Cel" reduction were biphasic (see fig. 7). The initial phase was fast, with no simple order. In addition it appeared very distinct from the second phase, and this allowed us to determine readily the quantity of Ce'" reduced in the initial phase (A[Ce4+]). This quantity was found to be directly proportional to the amount of RuO, present. The second phase was a slower process, which obeyed first-order kinetics almost perfectly (correlation coefficients were usually 2 0.9999).The kinetics reported by Minero et al. appear to be concerned solely with this second phase and their findings agree very well with our own." Fig. 8 shows a plot of log k, us. -log [RuO,]. A similar plot has been reported by Minero et d.," and in both plots it appeared that at high [polybrene]/[RuO,] ratios (i.e. % 2), the rate of Ce'" reduction was proportional to Identical reaction conditions as those employed in the above kinetic study were used to investigate the variation of the 0, yield as a function of [RuO,]. An 0,-MPD was used, as before, to monitor the amount of 0, produced; however, in no case was there found to be any evidence for 0, evolution during the reduction of the CeI" ions, mediated by the variety of colloid concentrations employed in the presence of the fixed high polybrene concentration (0.01 5 g dmP3).From this work it would appear, therefore, that the kinetics for CeIV reduction reported by Minero et al." and repeated by ourselves (see above and fig. 7 and 8) are not associated with the oxidation of water, but, more likely, with the oxidation of polybrene. [RuO,l* [RuO,I. Conclusion The results of the catalyst characterisation demonstrated that a true colloidal dispersion of RuO;xH,O supported by polybrene was prepared and that it possessed some catalytic activity for the oxidation of water. However, the stability of the colloid towards: (i) coagulation (including H,SO,), and (ii) anodic corrosion by CeI" ions390 RuO, - xH,O Colloids and the Oxidation of Water appeared poor.In addition, at the low concentrations of CerV ions and colloid necessary to carry out a kinetic study using a stopped-flow technique the colloid appeared to mediate the oxidation of polybrene rather than water. Indeed, at ratios of [polybrene]/ [RuO,] $ 2 no 0, evolution was observed during the reduction of the CerV ions. As a result of this work it is apparent that the colloid is unsuitable for use as an 0, catalyst or incorporation as such in a photochemical system capable of splitting water. Ideally, such a catalyst should be fast-acting, stable towards corrosion and coagulation and specific to the oxidation of water, i.e. not able to mediate more readily the oxidation of the support. Further work is now in progress to produce colloids of RuO;xH,O which utilise different supports and which are generated via reaction pathways other than that reported here.The colloids produced will be characterised using the techniques described above and through such research it is hoped that an ‘ideal’ 0, catalyst may be found. We thank the S.E.R.C. and the Royal Society for financial support of this work. References 1 Energy Sources through Photochemistry and Catalysis, ed. M. Gratzel (Academic Press, New York, 2 A. Demortier, M. De Backer and G. Lepoute, Nouv. J. Chim., 1983, 7, 421. 3 J. Kiwi and M. Gratzel, Chemia, 1979, 33, 289. 4 J. Kiwi, M. Gratzel and G. Blondeel, J. Chem. SOC., Dalton Trans., 1983, 2215. 5 G. Blondeel, A. Harriman, G. Porter, D. Urwin and J. Kiwi, J. Phys. Chem., 1983, 87, 2629. 6 A. Harriman, G. Porter and P. Walters, J. Chem. SOC., Faraday Trans. 2, 1981, 77, 2373. 7 A. Mills and M. L. Zeeman, J. Chem. SOC., Chem. Commun., 1981, 948. 8 A. Mills, J. Chem. SOC., Dalton Trans., 1982, 1213. 9 A. Mills, C. Lawrence and R. Enos, J. Chem. SOC., Chem. Commun., 1984, 1436. 1983). 10 J. Kiwi, J. Chem. SOC., Faraday Trans. 2, 1982, 78, 339. 11 C. Minero, E. Lorenzi, E. Pramauro and E. Pelizzetti, Inorg. Chim. Acta, 1984, 91, 30 L. 12 P. A. Christensen, A. Harriman, G. Porter and P. Neta, J. Chem. SOC., Faraday Trans. 2, 1984, 80, 13 T. J. Sworski, J. Am. Chem. SOC., 1958, 79, 3655. 14 R. E. Connick and C. R. Hurley, J. Am. Chem. SOC., 1952, 74, 5012. 15 W. P. Griffith, in The Chemistry of the Rarer Platinum Metals (Interscience, London, 1967). 16 J. L. Woodhead and J. M. Fletcher, J. Chem. SOC., 1961, 5039. 17 A. Mills and C. Lawrence, Analyst (London), 1984, 109, 1549. 18 P. Wehner and J. C. Hindman, J. Am. Chem. SOC., 1950, 72, 3911. 19 R. P. Larsen and L. E. Ross, Anal. Chem., 1959, 31, 176. 20 D. N. Furlong, A. Launikonis, W. H. F. Sasse and J. V. Sanders, J. Chem. SOC., Faraday Trans. I , 21 D. J. Shaw, in Introduction to Colloidal and Surface Chemistry (Butterworths, London, 3rd edn, 1980), 22 A. Mills, A. Harriman and G. Porter, Anal. Chem., 1981, 53, 1254. 145 1. 1984, 80, 57 1. p. 150. Paper 710’73; Received 14th January, 1987

ISSN:0300-9599    

DOI:10.1039/F19888400379

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. I, 1988, 84(2), 391-395 Modes of Enhancement of Physical Adsorption of Nitrogen and Water Vapour on Metal Oxides Paul A. Sermon* and Raj R. Rajaram Department of Chemistry, Brunel University, Uxbridge UB8 3PH A comparison of hydrogen chemisorption and physical adsorption on Co,Fe,-,O, gives information of the mode of specific interactions which enhance physical adsorption. It is notable that 1-2 N, and 3-5 H,O molecules are adsorbed per exposed cation-anion surface pair, enhancing their adsorption to an extent which varied with x. Physical adsorption is widely used to characterise the total surface areas and porosities of powders. It can be enhanced 1-4 when the radii of the adsorbent pores and the adsorbate approach one another and when there is some specific interaction between the adsorbate and some adsorbent functional groups.Thus the total extent of physical adsorption would be given by a summation of non-specific and specific contributions; these have always been difficult to separate. Here an attempt to overcome some of this uncertainty by applying both physical adsorption and chemisorption to probe well defined oxide surfaces is described. Experimental Co,Fe,-,O, (0 < x < 3) oxides were selected because of their high surface area, variable porosity and cation state. These were prepared5 with only marginal surface enrichment and surface areas stable to 600 K5 As x increased larger pores became more significant; at x = 0 53 % of the detected pore volume is within pores 4-25 nm in radius, whereas at x = 3 70% of the detected pore volume is within pores of 25-250 nm,5 although the results of mercury porosimetry are only in a limited size range. n-Hexane, N, and H,O were physically adsorbed upon these spinels as described previously:596 with N, (White Spot, BOC) at 77 K a semi-automatic Carlo-Erba 1800 Sorptomatic was used; with the vapour of digilled water at 299.65 & 0.02 K a quartz- spring balance was used after outgassing to < 13.3 mPa; with n-hexane (AnalaR, BDH) adsorption was measured at 298 K using a flow-chromatographic method.The cross-sectional areas of these adsorbates were assumed to be 0.162, 0.106 and 0.541 nm2, respectively. Results Type I1 isotherms5g for n-hexane, N, and H,O adsorption were exhibited and analysed to give the surface areas and C values.' The surface areas estimated by X-ray diffraction line broadening (XRDLB) and n-hexane adsorption were in moderate agreement.5 If XRDLB is accurate in its assessment of average crystallite sizes and hence surface area, then hexane (which is the largest adsorbate) presumably adsorbs over the entire surface (irrespective of which pore size this happens to be predominantly in; however, there is no independent evidence that this is indeed the case). This could be taken to mean that adsorbate molecular sizes do not define the fraction of surface on which physical adsorption is observed here.The increase in the extent of adsorption from n-hexane to nitrogen and then water vapour could result if the most polar molecules interact more 39 1392 Adsorption of N, and Water Vapour on Metal Oxides I 1 I I I I 1 2 3 X Fig.1. Effect of composition on the numbers of hydrogen (A), carbon monoxide (0) and oxygen (0) molecules chemisorbed on samples of Co,Fe,-,O, at 373, 295 and 473 K, respectively. Total surface areas were deduced by B.E.T. analysis of N, adsorption at 77 K. specifically with functional groups on the spinel surface. Values of r in the Frenkel- Hill-Halsey equation [i.e. In (p,/p) = -kq-']' from plots at 0.4 c p/po < 0.9 for adsorption of water vapour decreased from 3.28 to 2.36 as x increased,6 suggesting a decreasing specificity of adsorption as x decreases, although there is some uncertainty over this interpretation. Interestingly, the B.E.T. C values increase as x decreases, also suggesting greater specificity of physical adsorption at low x, in apparent contradiction of degrees of enhancement of the extent of adsorption (which increase as x increases).It is also entirely unexpected that C values are greater at low x and for n-hexane. The chemisorption of hydrogen* at 373K was essentially irreversible but was well represented by a Langmuir isotherm with a plateau. Such adsorption probes the concentrations of incompletely coordinated or partially exposed adjacent surface cations and anions :g Ha- - Ha+ H H and at 373 K the numbers of hydrogen molecules chemisorbed per nm2 on these spinels increases steadily with x (see fig. 1). The repeatability of such isotherms indicates that neither adsorbent reduction nor reconstruction is significant, although reduction certainly occurs at and above 423K.The extent of CO chemisorption, measured volumetrically at 273, 295 and 348 K on the oxide samples outgassed at.573 K for 6 h'O did not reach a monolayer CO capacity at any experimental pressure and was represented by the Freundlich equation. Although CO adsorption on lattice cations could involve the formation of carbonyls, it is moreP. A . Sermon and R. R. Rajaram 393 likely to interact only with lattice anions producing a surface carbonate : CO(g) + 20,- = C03,-+2e-. Fig. I shows that the numbers of CO molecules chemisorbed per nm' of the spinels at 295 K also increased simply with x; however, repeated isotherms showed a decreased CO uptake. This irreversibility of chemisorption (possibly associated with surface oxygen depletion and reduction) mitigated against its use here.The slow rate of oxygen chemisorption at 373473 KIO and its chemisorption upon only surface divalent cations in octahedral sites : Co2+ + 0, = Co3+-O,-, and its absorption at high temperature and x also mitigated against its use here. Because monolayer capacity was reached rapidly with hydrogen and the adsorption sites are closely related to those cation-anion pairs c o n c I ~ d e d ~ ~ - ~ ~ to be involved in nitrogen and water vapour adsorption on dehydroxylated rutile surfaces, hydrogen chemisorption was selected as being most suitable to compare with the results of physical adsorption. One way then to resolve these ambiguities in the evidence provided by physical adsorption a10nell-l~ and to disentangle specific and non-specific contributions is to have such a method of probing the functional groups on existing adsorbents independently and to compare its results with those of physical adsorption.This has been attempted here. The assumption then which is crucial to the comparison below is that hydrogen chemisorption' and enhancement of N, and H,O physical adsorption5 on these spinels both involve incompletely coordinated cation-anion pair sites at the oxide surfaces. Past evidence for rutile etc.,11'12 albeit sparse, suggests that this is so; conversely, the agreement between the two sets of adsorptive data here (see description below) underlines the correctness of this assumption. Since heterolytic adsorption of hydrogen must involve exposed cation sites, alternative adsorption models are unlikely to be as satisfactory.Here we use the term enhancement of physical adsorption to indicate an effect of specific interactions with such surface sites. Discussion Assume that physical adsorption of n-hexane, nitrogen and water vapour occurs upon the same accessible surface areas on each spinel here, but that the latter adsorbates are also held upon specific cation-anion sites in addition to a physically held coverage (i.e. either on top of the physically held monolayer or more likely that initial N, and H,O molecules are strongly held on cation-anion sites and essentially incorporated into the spinel surface : OH H N2 OH H N = N + _I - M(3-A) i- - 0 ( 2 - A ) -- M(3-A)+- -M(2-A)+ - o(I-A)-- M(3-A)+, at incompletely coordinated sites generated the surface and on top of which the normal physically held monolayer subsequently formed).Then the numbers of such N, and H,O molecules per nm2 of the spinel surface (estimated by N, B.E.T. measurements) which are so adsorbed by specific interations in addition to the normal physically held capacity can be calculated. Fig. 2 shows the ratios of the surface concentrations of specifically held N, (or H,O) to chemisorbed H,, which may be taken to be the ratio of specifically held N, (or H,O) per exposed surface cation-anion pair, as a function of x. First, it is clear that both adsorption stoichiometries are greatest at low x. This may be because at x = 0 Fe,"' and Fe?' will accept electrons and moderately adsorb with a modest change in CFSE;15 two thirds of surface cations might interact thus with an enthalpy, C value and r value greater than in mere physical adsorption. At394 Adsorption of N, and Water Vapour on Metal Oxides 0 f X 2 Fig.2. Stoichiometry of specific adsorption of water molecules (0) and N, molecules (0) per surface cation-anion site (as estimated by hydrogen chemisorption; see fig. 1) in addition to a physically held monolayer adsorption capacity. x = 3, Co," can only interact without electron transfer and hence less specifically and more weakly, but it appears from the present results that such sites are in a higher surface concentration than those in Fe,O,. Thus greater extents of enhanced physical adsorption are seen at x = 3, but the highest enthalpies of physical adsorption are seen at x = 0.As yet it is uncertain why C and r values are greater for n-hexane than for water vapour. Secondly, 1-2 N, molecules appear to be adsorbed per surface cation-anion pair site, but it is premature to ponder whether this adsorbate is stretched between two cation sites. Thirdly, 5 H,O molecules appear specifically adsorbed per exposed cation site at x = 0, dropping to 2-3 at x = 3. Either the former cation sites are very incompletely coordinated or there is great specificity over inner shell and outer shell water adsorbate layers. Such extended interactions involving water are known to involve water strongly bound to the surface as well as hydrogen bonding to relatively immobile water in the next molecular layer and so on; such interactions clearly decrease in strength as one proceeds away from the specific adsorption centres.However, further details cannot yet be resolved. Nevertheless, these results suggest why B.E.T. C and FHH r values increase as x decreases and explain why enhancement in extents of adsorption are greatest for water vapour. Conclusions The idea of specific interactions in physical adsorption is not new;16 however, here a relationship between sites chemisorbing hydrogen (but not those chemisorbing CO or 0,) and the enthalpy and extent of specific physical adsorption of N, and H,O has been shown; doubtless a similar approach would be of value in characterising other solid surfaces. Just as weak chemisorption overlaps with formal physical adsorption, specific interactions draw physical adsorption of N, and H,O on oxides towards chemisorption.Their net effect is to raise the extents of physical adsorption and hence lead to the B.E.T.P. A . Sermon and R. R. Rajaram 395 analysis over-estimating the surface areas. The present approach may allow a correction of the B.E.T.-derived areas. All too frequently the application and measurement of physical adsorption and chemisorption data are artifically separated. The present results hint at the value of inter-relating such data for some systems. Previously, physical adsorption on unsupported metal powders has enabled the stoichiometry of chemisorption to be ascertained : conversely here chemisorption enables a better understanding of physical adsorption. Since specific and non-specific interactions will show a different temperature and pressure dependence, the precise correlations found under the present adsorption conditions may not be valid at other temperatures and pressures.Nevertheless, the results here hint at the potential of the combined chemisorption-physical adsorption approach. Further work would be needed to justify comparisons under a wide range of adsorption conditions on a variety of surfaces. References 1 A. A. Isinkyan and A. V. Kiselev, J. Phys. Chem., 1961, 65, 601. 2 E. M. McCafferty, and A. C. Zettlemoyer, J. Colloid Interface Sci., 1970, 34, 452; Discuss. Faraday 3 R. E. Day, G. D. Parfitt and J. Peacock, Discuss. Faraday Soc., 1971, 52, 215. 4 S. J. Gregg and K. S . W. Sing, Adsorption, Surface Area and Porosity (Academic Press, London, 5 R.R. Rajaram and P. A. Sermon, J. Chem. SOC., Faraday Trans. I , 1985, 81, 2577. 6 R. R. Rajaram and P. A. Sermon, Fundamentals of Adsorption, ed. A. I. Liapis (Engineering Founda- 7 A. C. Zettlemoyer, J. Colloid Interface Sci., 1968, 28, 343. 8 R. R. Rajaram and P. A. Sermon, J. Chem. SOC., Faraday Trans. I , 1985, 81, 2593. 9 T. Ito, T. Murakami and T. Tokuda, J. Chem. Soc., Faraday Trans. I , 1983, 79, 913. SOC., 1971, 52, 239. 1982). tion, 1986), p. 493. 10 R. R. Rajaram, Ph.D. Thesis (Brunel University, 1983). 1 1 D. N. Furlong, F. Rouquerol, J. Rouquerol and K. S . W. Sing, J. Chem. SOC., Faraday Trans. 1, 1980, 76, 774. ; J. Colloid Interface Sci., 1980, 75, 68. 12 D. N. Furlong, K. S. W. Sing, and G. D. Parfitt, Ads. Sci. Technol., 1986, 3, 25; J. Rouquerol, F. Rouquerol, Y. Grillet and M. J. Torralvo, in Fundamentals of Adsorption, ed. A. L. Myers and G. Belfort (Engineering Foundation 501, 1984). 13 P. T. Dawson, J. Phys. Chem., 1967, 71, 838. 14 P. J. M. Carrott, A. I. McLeod and K. S. W. Sing, in Ahorption at the Gas-Solid and Liquid-Solid 15 J. B. Goodenough, Proc. R. SOC. London, Ser. A, 1984, 393, 215. 16 K. C. Campbell and D. T. Duthie, Trans. Faraday Soc., 1965, 61, 558; J. D. Carruthers, D. A. Payne, Inferface, ed. J. Rouquerol and K. S. W. Sing (Elsevier, Amsterdam, 1982). K. S. W. Sing and L. J. Stryker, J. Colloid Interface Sci., 1971, 36, 205. Paper 71083; Received 16th January, 1987

ISSN:0300-9599    

DOI:10.1039/F19888400391

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

J . Chem. Soc., Faraday Trans. I , 1988, 84(2), 397411 Solutions of Organic Solutes Part 2.-Moderately Polar Compounds in Water ; Limiting Volumes and Compressibilities Jean V. Leyendekkers Department of Physical Chemistry, University of Sydney, 2006 New South Wales, Australia The chemical thermodynamic (CT) model used to develop equations for the partial molal volumes and compressibilities of amino acids (J. V. Leyendekkers, J . Phys. Chem., 1986, 90, 5449) has been applied to moderately polar solutes; eight solids and 22 liquids. Radii and packing distances of the solutes in water have been estimated from the scaled particle theory, for 25 and 5 "C. The space requirements of the solute due to its polar character could then be deduced from the CT model. These results plus an analysis of the compressibilities indicated that the polar interactions are linked to dynamic correlations between the solute and water and have only a small effect on the volume changes.The latter are dominated by the geometric packing effects. Clathrate-like local bonding is implied, with solute to water ratios of 1 : 17 and 1 :28 (for the bulkier solutes). The overall results are in accord with deductions from computer simulations and a variety of experimental data (e.g. Raman, X-ray and 1 7 0 n.m.r. techniques). In part 1 of this series' a chemical thermodynamic (CT) model was combined with the dielectric theories of Kirkwood, Fuoss and Buckingham in order to derive equations for the partial molal volumes and compressibilities of organic compounds in solution. These equations were then used to predict and analyse the volume properties of amino acids in water.As the dipole moments, p, of the amino acids are large the effects related to p may have obscured other effects which would be more important for less polar solutes; for example, the change of the polarizability of water with solute concentration. Consequently, in parts 2 and 3 of this series the equations are used for a number of apolar to moderately polar solutes in water, in order to test the generality of the previous results. In the present paper the limiting volume properties of the solute (i.e. in very dilute solutions) will be considered. These are analysed on the basis of the CT model combined with the scaled particle theory (SPT) and results from a statistical-mechanical theory of aqueous solutions of hydrocarbons, together with dielectric and dynamic properties of the solutes.In part 3, the effect of changes in concentration will be considered. Equations The following equations were derived in part 1 . l For moderate concentrations, the partial molal volume v of the solute is given by V = V0+2S,m (1) where m is the concentration of solute in mol (kg H,O)-l and V" is the limiting partial molal volume (at infinite dilution), given by 14 V" = vl; + v,,,, = v:n + (B, + 1) KZomp. 397 (2) FAR 1398 Limiting Volumes and Compressibilities Komp is the compressional effect on the water due to the solute's volume and charge,, Vi", is the volume of the solute in solution (the intrinsic volume), BT is the Tait Parameter' and K:omp [or -(aV:omp/ap)T] is a component of the limiting partial molal compressibility (isothermal), K& i.e.where Ki", is (-3 Vin/i3p)T. - - Kg = Ki",+K:omp (3) Values of the intrinsic volume are obtained from1 (1 +k1)Vi"2- V"Vo,+k; = 0 ( 4 4 where k; = - 2.52 x 1024N@~f(D,J] (4b) withf(D) = (1/D2)(aD/i3p)T, D is the static dielectric constant of the solution, the zero subscript indicates the pure solvent, ps is the dipole moment of the solute and N is Avogadro's number. For water' at 25 "C ki = -0.684p:, with ps in D (1 D = 3.335 64 x C m). Values of K&,p and hence Kio, may be obtained from eqn (2H4) and V", K ; and ,us data. At 25 "C the parameter k, = -0.0638 (derived previously).' The Limiting Partial Molal Volume Estimation of Vg and KO,,,,, The value of Vi", at 25 "C was estimated for 30 organic solutes, using eqn (4) and experimental values of V".Eomp could then be estimated from eqn (2). Values are listed in table 1, together with the experimental V" values and h, the molar volume of the pure solute. The effect of p is only small, and errors of +2D give negligible errors to Via, so that uncertainties in ViOn are the same as the experimental errors for V" (ca. kO.1 cm3 mol-1 j. Calculations of Vin and Eomp were made as well for the alcohols and diols at 5 "C (table 1). A value for k, was deduced by assuming that the relative values of Kio, for the alcohols were as for 25 "C (estimates of K; are discussed below). This gave k, = - 0.1267. Components of Vg The intrinsic volume of the solute can be represented by an equivalent sphere of radius (r + A) where r is the radius of the solute and A covers packing effects (including shape and polar influences).Values of (r + A) estimated on this basis for aqueous electrolyte solutions were found to be closely related to the ion-water geometry elucidated from X- ray and neutron diffraction work,2 which indicates that such a basis is useful. Estimating r Pure Liquid. The scaled particle originally developed for hard-sphere or simple, non-polar fluids, gives reasonable estimates of molecular diameters even for such complex liquids as water. Various equations, enabling an estimate of the hard-sphere diameter (HSD), have been developed from the theory. For example, with y = nNd;/ 6 5 , the molar volume, 5, is given by5 so that dl (A), the HSD of the liquid, can be estimated, using experimental compressibility (p bar-l) and volume data, R is the gas constant (cm3 bar mol-1 K-l) and Tis the temperature (K). A slightly different equation is6 Yo = R7','.oy(7+~2+~)/(l -.Y)~ ( 5 ) 5 = RTP,[( 1 + 2 ~ ) ~ - 4z3 + z4]/( 1 - z ) ~ with z = IINd:/65.J .V . Leyendekkers 399 Table 1. Limiting volume properties no. solute -~ - 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 10 11 12 13 14 15 16 17 18 19 - k i ,uU eqn (46) V o b vp,c -(V"-V,P,) v,. dextrose D-ribose sucrose urea 1,3-dimethyl urea thiourea ace tamide thioacetamide 1,4-dioxane methanol ethanol propan- 1-01 butan-1-01 pentan- 1-01 ethane- 1,2-diol propane- 1,2-diol butane- 1,2-diol pentane- 1,2-diol hexane- 1,2-diol diethylene glycol triethylene glycol te trae thylene glycol 1,2-dimethyoxyethane bis(2-methoxyethy1)ether triethylene glycol dimethyl tetraethylene glycol dimethyl 2-methoxyethanol 2-ethoxyethanol 2-propox yethano 2- bu toxye t hanol ether ether methanol ethanol propan- 1-01 butan- 1-01 pentan- 1-01 ethane- 1,2-diol propane- 1,2-diol butane- 1,2-diol pentane- 1,2-diol hexane- 1,2-diol 2 2 2 4.6 6.5 4.9 3.8 4.6 0 3 3 3 3 3 4 4 4 4 4 5.5 5.6 5.8 4 4 4 4 5 5 5 5 T = 25°C 3 3 3 14 29 16 10 15 0 6 6 6 6 6 11 11 11 11 11 21 21 23 11 11 11 11 17 17 17 17 T = 5 " C 112.04 95.21 21 1.32 44.23 80.03 54.79 55.60 66.42 80.94 38.17 55.08 70.74 86.63 102.55 54.65 71.89 88.36 104.43 120.39 92.25 129.28 166.39 95.85 132.72 169.83 206.88 75.20 91.29 107.13 122.98 38.10 55.12 70.48 86.00 101.43 53.77 71.40 87.8 1 103.54 119.16 119.71 101.73 225.75 47.57 85.85 58.83 59.57 71.17 86.46 40.94 58.95 75.66 92.62 109.62 58.50 76.96 94.53 11 1.68 128.71 98.79 138.28 177.91 102.52 141.88 181.51 22 1.08 80.57 97.72 114.61 131.53 43.63 63.12 80.70 98.48 f 1 16.14f 61.57f 81.76f 100.55f 1 18.56 136.45 * 7.7 6.5 14.4 3.3 5.8 4.0 4.0 4.8 5.5 2.8 3.9 4.9 6.0 7.1 3.9 5.1 6.2 7.3 8.3 6.5 9 .O 11.5 6.7 9.2 11.7 14.2 5.4 6.4 7.5 8.5 5.5 8.0 10.2 12.5 14.7 7.8 10.4 12.7 15.0 17.3 - - (216) (46) - - - - 85.71 40.70 58.73 75.22 91.94 108.75 55.92 73.68 91.4" 109.2e 127.0e 95.38 134.13 173.4e 104.54 142.14 179.7e 217.3e 79.25 97.41 1 14.5" 131.84 39.74 57.48 73.71 90.33 106.84 55.22 72.67 90.1 " 107.6" 125.0" a Experimental data for pure compound at 25 "C, from: A.L. McClelfan, Tables of Experimenfal Dipole Moments (W. H. Freeman, San Francisco, 1963) and ref. (18). Experimental data: solutes 10-19, T. Nakajima, T. Komatsu and T. Nakagawa, Bull. Chem. SOC. Jpn, 1975, 48, 783; solutes 1-9, A. Lo Surdo, C. Shin and F. J. Millero, J. Chem. Eng. Data, 1978, 23, 197; solutes 20-30, ref. (30). Experimental data: ref. (15) and (27); bracketted values from M. R. J. Dack, Aust. J. Chem., 1976, 29, 771. From eqn (4a), neglecting k i , k , = -0.1267. Units: volumes in cm3 mol-'; p in D. Calculated from eqn (4a) with ki = - 0.684,~~ and k, = - 0.064. Interpolated or extrapolated from data of other members of homologous series. 14-2400 Limiting Volumes and Compressibilities Table 2. Components of VP, no.a %Wb rOe 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 20 21 23 27 28 29 30 10 11 12 13 14 - - - - - - - - - 2.049 2.33 1 2.557 2.749 2.917 2.438 - - - 2.647 2.827 2.987 3.131 - - - - - - - - - - - - - - 1.746 1.769 2.079 2.08 1 2.334 2.329 2.561 2.551 2.769 2.754 2.239 2.227 2.76 2.74 3.16 3.14 2.83 2.8 1 2.48 2.47 2.7 1 2.70 - - 1.767 1.775 2.097 2.095 2.347 2.340 2.578 2.566 2.769 2.754 3.206 3.003 4.030 2.180 2.759 2.386 2.338 2.505 2.699 1.934 2.3 13 2.589 2.833 3.018 2.302 - - 2.906 2.653 2.880 3.072 3.245 1.995 2.398 2.705 2.941 3.150 T=25'C - - - - - - - - - 0.780 0.768 0.777 0.775 0.767 0.773 0.754 0.766 0.738 0.753 0.570 0.582 0.59 0.6 I 0.60 0.62 0.63 0.65 0.68 0.69 0.67 0.68 - - T=5'C 0.740 0.732 0.738 0.740 0.733 0.740 0.718 0.730 0.717 0.732 0.414 0.426 0.443 0.482 0.48 1 0.47 1 0.53 1 0.539 0.549 0.598 0.546 0.518 0.502 0.498 0.550 - - 0.532 0.520 0.503 0.496 0.49 1 0.592 0.527 0.470 0.45 1 0.434 -0.140 -0.128 -0.1 11 - 0.072 - 0.073 - 0.083 - 0.023 -0.015 - 0.005 + 0.045 - 0.007 - 0.034 - 0.05 1 - 0.055 - 0.004 - - -0.012 - 0.034 -0.049 - 0.058 - 0.063 + 0.034 - 0.03 1 -0.088 -0.106 -0.124 - 13 -9 - 14 0.9 1.2 0.1 8 10 13.1 12.8 10.0 7.4 5.2 4.7 10.4 - - 10.9 7.6 5.3 4.0 2.9 7.28 - 1.09 - 11.15 - 19.41 -27.13 a See table 1.Hard-sphere radius (A) of pure solute : first value from eqn (6), second value from eqn (5). Compressibilities listed in table 7 (IateQ, V, for water: 18.070 cm3 mol-: (25 'C) and 18.016 cm3 mol-' (5 "C). Hard-sphere radius (4) in water, eqn (f). [( V,/2.523)5- rJ/A in same order as for footnote c.[( Vf,/2.523)5- rJA. (Aw -0.554)/A. Partial molal isothermal compressibility of solute, cm3 mo1-1 bar-l; expen- mental data from: solutes 1-9, estimates based on & (expansivity) data for sucrose from J. L.. Neal and D. A. I. Goring, J. Phys. Chem., 1970, 74, 658; solutes 1&14, ref. (31); solute 15, esti- mated from alkoxyethanol series; solute 23, ref. (30) and (31); solutes 27-30, ref. (30). Van der Waals radii: ref. (12) and (13), in A.J . V . Leyendekkers 40 1 Both eqn ( 5 ) aFd (6) can be simply extended to solutions.DEqn ( 5 ) gives the HSD for water as 2.740 A, whilst eqn (6) gives a value of 2.733 A. Another method4** for estimating HSD values is from an extrapolation technique using experimental solubilities of race gases in the liquid. However, an estimate of errors from the extrapolation gave 0.06 A for the HSD4 and this method is not easily applied to solutions.Values of the HSD were estimated for a number of pure liquids from eqn (5) and (9) (table 2). Errors in P of 5 x bar-' give a corresponding uncertainty in ro of 0.01 A. Experimental compressibilities appear to be only available for the alcohols and ethylene glycol. However, for 25 "C, values of the Tait parameter, BT, for these compounds (and other liquids such as benzene, benzene derivatives, aniline and CCl,) fit the equation (to f 100 bar) : where t, is the boiling point ("C) and (dtldp) is the pressure coefficient of temperature (KTorr-'). Some estimates of the compressibilities (ca. 3 70 uncertainty) could therefore be made from the Tait equation,' i.e./3 = (C/ln lO)/(B,+ 1) with C taken as 0.22.' B,(bar) = - 272.8 + 0.593tb(dp/dt) (7) Solutions. For aqueous electrolyte solutions the crystal radius of the ions has been shown2* lo, '' to be appropriate for r, but for solutes such as organic liquids and solids the selection is not so easy. However, it has been shown' that the HSD values of ions estimated from the SPT are close to the crystallographic diameters and it is therefore appropriate to use this theory to estimate r for organic solutes in solution, the aim being to develop general equations to cover all solutes and solvents. The approach used to develop eqn (6) yields the following relationship for dilute solutions6 where K = 10-3d0PiRT/(l - z ) ~ , with z as for eqn (6) and (8) - lO-'(R; -Po V") = K(Bo + B, d2 + B, d2, + B, d i ) Bo = F,-(V0/V,)(F,+61;,+3F,+2FJ; B, = 6F,/d1; B, = 3F,/dt and B, = 2F,/df; with 4 = (1 --2)3; 4 = z(i -212; 4 = z(i -zz)(2+7z-4z2+z3) and 4 = ~ ( l +7z+8z2-5z3+z4).do and Do are the density and compressibility of the solvent, respectively, & and dl are the molar volume apd HSD of the solvent; d, is the HSD of the solute. The value of dl is taken as 2.733 A [eqn (6)] for consistency. Values of d, (table 2) were estimated from eqn (8) using the experimental values of V" and K;. The latter data are few, but values may be calculated from K: using other thermodynamic data'? (table 2). Solvent data are given in the footnotes of table 2. Eqn (8) can also be used for those solutes which are solids at 25 "C; using this equation an error of 1 x units in K; gives a corresponding change in the hard-sphere radius of 0.01 A, whilst an error of 0.1 in V" gives a corresponding change in rw of 0.001 A.For interest, the intermolecular van der Waals radii,12*13 rvw, were estimated for the n-alcohols and alkoxyethanols from the group and are compared with the corresponding hard-sphere radii in tsble 2. For all pure solutes orvw is greater than ro, with an average deviation of + 0.19 A (maximum deviation 0.29 A), whilst for aqueou? solutions r,, can be greatp or less than rw, with an average deviation of fO.07 A (maximum deviation 0.14 A). This means that rw is fairly close to that radius which is impenetrable to thermal collision.402 Limiting Volumes and Compressibilities Estimating A Pure Liquidr. A is easily estimated from the HSD and the molar volume, i.e.r,+A, = (V,/2.523)f (9) where ro equals HSD/2; 2.523 comes from (4/3)HNx Values of A, for a range of liquids were estimated from eqn (9) (tabl? 2). For water at "C, with & 18.07 cm3 mol-1 and r taken as 1.38 A, the packing distanceis 0.548 A which is smaller than any of the other liquids; A, ranges from 0.62 to 0.77 A. Accurate density data for most of the solids appear to be unavailable for 25 "C so that no estimates of V, could be made for these solutes. Solutions. From the scaled particle theory, the intrinsic volume is given by5 Vk = Ar3 + RTP,(A, + A , r + A , r2) (10) where A = 2.523; A , = y/(l - y ) ; A , = 6y/d,(l -y), and A, = [12y/(l -y)2+36y2/ (1 -~)~]/d:, where y is defined in eqn ( 5 ) and d, is estimated for water from eqn ( 5 ) (i.e.2.74 A). Eqn (10) may be compared with the expanded form of 2.523(r+A)3, i.e. VL = 2.523(r3 + A3 + 3A2r + 3Ar2). (1 1) The corresponding terms of eqn (10) and (1 1) (e.g. RTB, A , and 3 x 2.523A) are noJ exactly equivalent, giving somewhat different values of A.5 However, using dl = 2.74 A [from eqn (5)] and5he HSD values estimated above, with eqn (lo), A for solutes in water at 25 "C is 0.554 A for all the solutes considered. The valuoes from the terms contain- ing A,, A , and A,, respectively, are 0.558, 0.534 and 0.63 A, so that an estimate from RT/3,4,/7.569 would be generally satisfactory. This value of A is to be compared to 0.55 A used for the geometric packing distance of small ions in water.lo.l4 Values of (r,+0.554) are compared with (rw+Aw) from the V:n values of table 1, in table 2.For the non-polar solute 1,4-dioxane at 25 "C the two values are very close which could be expected if the SP theory gives reasonable values for (r, + A,), where A, is the geometric distance. For the polar solutes (r, + A,) is generally smaller. For electrolytes, the differences in Aw (from VP,) and the geometric value were shown to be due to electric distortion of the water molecules close to the solute.lo For small cations the difference was proportional to the number of water molecules that had lost their orientation polarization. Consequently it is assumed that (Aw - 0.554) or Ael is of polar origin and this assumption is supported by the following correlations.Correlations of Ael For those solutes for which data are available (table 3) Ael is simply related to the molar polarizability, P, of the pure solute, i.e. (12) where So = 0.046 A and C, = - 1.64 x (13) kO.01 A. P is given by: where E is the static dielectric constant and & the molar volume [data from ref. (1 5)]. The constant Co could indicate that the hard-sphere radii have been somewhat under- estimated ; however, the solubility results indicate that they could have been o~erestimated.~? More likely, C, represents a contribution from the polar solvent, only applying to polar molecules. Ael = C, + C,[P- P(dioxane)] P = (&- 1) &/(&+2) A cm3, with an average deviation ofJ . V. Leyendekkers 403 Table 3. Viscosity and molar polarizability of solutes at 25 "C 9 10 11 12 13 14 15 23 27 28 29 30 2.2 1 32.70 24.55 20.33 17.51 13.9 37.7 7.20 16.93 29.6 20 f 9.3 24.62 37.09 52.09 65.1 1 77.80 88.23 51.69 70.45 66.69 88.16 98.9 96.84 1.204 0.5445 1.078 1.989 2.640 3.347 17.73 0.455 1.60 1.85 2.56 3.15 - - - 0.0222 6 -0.0417 12 -0.0825 27 -0.1140 28 - 0.1949 35 a Static dielectric constant ; experimental data from ref.(1 5). Experimental dy- namic viscpsity, cP; ref. (15). In CP K-l, experimental data, ref. (1 5). In A K-l, this work. f Interpolated from E as a function of Molar polarizability, cm3 mol-1 ; eqn (1 3). n(CH2). The value of Ael becomes more positive with decreasing dynamic viscosity, 7, of the pure solute (table 3) and is correlated linearly with q, i.e. average deviation kO.01 A, 7 in cP.The diols have very large viscosities and ethane-1,2-diol, the only diol for which data are available, does not fit eqn (14). The viscosity is proportional to the shear relaxation time and the orientational relaxation (via the Debye equation16) so that eqn (14) suggests that Ae, derives from the dynamic space requirements of the solute as well as the static polar effects. Since dioxane is non-polar the coefficient 0.03 of eqn (14) (which dioxane fits) is not specific for polar molecules. This increment could, for example, derive from the Stark effect," which applies to both polar and apolar solutes in a polar solvent. A more general correlation involves the solid solutes as well, and was developed as follows. The influence of organic solute molecules on the dielectric relaxation time of water is closely correlated with N, the number of water molecules in the inner hydration sheath,18 i.e.those water molecules whose surfaces are directly adjacent to the solute molecule. N can be estimated using a simple model involving spherical solute molecules and cubic water molecules, from1* Ael(A) = 0.03 - 3.048 x 10-2q (14) Values of N calculated from eqn (1 5 ) are shown in table 4. These values show a close Ael(A) = 0.187- 1.763 x 10-2N correlation with Ael, i.e. with average deviation of 0.015 A. The ratio R = W/N, where N' is estimated from eqn (1 5 b) using the Ael values from table 2, indicates the deviations from eqn (15b). The solutes with hydrocarbon and hydrogen-bonding groups (which are not too large, or bulky) fit the correlation well, an$ R = 1 k0.2 (group A, table 4).The deviation for dioxane in terms of Ael is ca. 0.05 A which is consistent with the coefficient of eqn (12) and apparently only relates to polar molecules, thus dioxane belongs to group A rather than group B. Deviations for some404 Limiting Volumes and Compressibilities Table 4. Number of water molecules adjacent to solute no. solute N" N b N'/N KSc Group A 1 dextrose 2 D-ribose 5 1,3-dimethyl urea 7 acetamide 8 thioacetamide 10 methanol 11 ethanol 12 propan- l-ol 13 butan- 1-01 14 pentan-1-01 15 ethylene glycol 27 2-methoxy ethanol 28 2-ethoxy ethanol Group B 3 sucrose 9 1,4-dioxane 23 1,2-dimethoxy ethane 29 2-propoxy ethanol 30 2-butoxy ethanol Group C 4 urea 6 thiourea 10 methanol 11 ethanol 12 propan- l-ol 13 butan- l-ol 14 pentan- l-ol T = 25 "C 17.1 15.3 13.7 10.7 12.1 8.3 10.6 12.6 14.4 16.1 10.6 13.1 14.9 26.0 13.7 15.4 16.6 18.2 9.2 10.6 T=5OC 8.7 11.2 13.1 15.0 16.8 18.6 17.9 14.8 11.9 11.5 8.1 11.0 12.5 13.5 13.7 10.8 12.5 13.4 6.9 0.9 1.3 3.9 4.2 14.7 15.3 8.1d 11.4 14.2 15.1 16.0 1.1 1.2 1.1 1.1 1 .o 1 .o 1 .o 1 .o 0.9 0.9 1 .o 1 .o 0.9 0.7 0.8 0.7 0.8 0.8 1.6 I .4 0.9 1 .o 1.1 1 .o 1 .o 83.14 68.33 52.99 32.24 39.66 18.25 31.22 43.78 56.70 69.35 30.78 47.11 60.27 165.13 49.6 1 61.92 73.14 86.21 26.14 34.27 20.03 34.79 49.94 64.18 78.86 a Calculated from eqn (15a).Ael from table 2. and Ael. Calculated from eqn (15b) and Calculated from eqn (19) 2.523r: cm3 mo1-l. of the more bulky solutes probably arise because of inadequacies in the spherical approximation for the shape [eqn (1 5 a)] and/or because the diameters are too large to fit into the group A-type water cages.As discussed below, the deviations could also arise because only a fraction of N are affected by the polar characteristics of solute and solvent. The polar effects of group C must extend beyond the first hydration layer. An equivalent equation to eqn (1 5 a) is N = 2.928 + 5.303( Vhs/ Vw)$ (16) where is the hard-sphere volume (i.e. 2.523rEs) and Vw is the molar volume of water. Differences in N from eqn (1 5 a) and (1 6) only average 0.2 with a maximum deviation of 0.5. From solubility data for gases in water (using a statistical-mechanical theory of fluids applied to the case of hard-sphere molecules') theo limiting hard-sphere diameter, corresponding to zero polarizability for the gas, is 2.58 A.Substituting the correspondingJ . V , Leyendekkers 405 hard-sphere volume into eqn (16) gives N = 5.3f0.2, which may be considered tbe lowest possible value of N . From eqn o(l5a) this N value corresponds to Ael = 0.09 A, which is close to the value of 0.085 A from eqn (12) (for zero polarizability). This increment thus only applies to solutes with finite polarizability and in part [ca. 0.046 A, eqn (12)] only to solutes with non-zero dipole moments. Eqn (15a) and (16) give Ks as a function of Vin so that rhs could be estimated from the partial molal volume alone (i.e. without the need for K;). Calculations for the solutes in table 4 show that KS can be estimated with an average deviation of 1.6 c p 3 mol-1 in this way; the corresponding HS radius has an average deviation of 0.028 A.The correlation with N can be interpreted as follows. Alcohols (such as ethyl, t-butyl and isopropyl) are known to form hydrate clathrates, irrespective of the hydrophilic nature of the hydroxyl group. ''7 'O Furthermore, Raman spectroscopic studies of aqueous alcohol solutions have shown that the largest Raman linewidths (presumably indicating the largest number of hydrogen-bonded solute molecules possible in the solution) occur21 at the mole ratio of solute to water of 1 : 17, a ratio which is characteristic of many of the clathrates. Tanaka et aLZ2 recently made X-ray diffraction measurements on dilute aqueous solutions of tertiary butyl alcohol at 'room temperature ', together with molecular dynamic calculations on the simulated solutions.The experimental data and the simulations both indicated that a fairly stable clathrate- like structure of water formed around the solute molecule. A hexakaidecahedron formed from pentagons and hexagons seems likely. This cage structure contains 28 water molecules. Other techniques (n.m.r., ultrasonic velocity) infer 20 to 25 water molecules in the structure.22 This evidence suggests that bulky solutes such as 2-propoxyethanol, 2-butoxyethanol and 1,2-dimethoxyethane might fit large cage structures (for example of the 28 water- molecule type). The N'/N ratios of the group B solutes (table 4) indicate that this size would be appropriate. That is, to satisfy a Ael uersus N / N , correlation, with N , = 17 or 28, as appropriate. As will be shown below, the compressibility results also support this assumption. Of course, these are geometric packing type considerations and, in regard to polar influences, not all of the water molecules nearby might be affected for the larger solutes.For example, the 170 n.m.r. technique enables relaxation of different species to be separated and since the quadrupolar interaction dominates in the relaxation, only rotational motions need to be considered in the correlation function for relaxation. This makes the results more unequivocal than those from other types of relaxation studies. A recent 170 n.m.r. spin-lattice and transverse relaxation-time of aqueous solutions of sucrose (at 32-97 "C) found that each sucrose molecule slightly modified the correlation time of ca.16 water molecules which surround it. This number of water molecules is approximately correct (table 4) for eqn (15b), i.e. to account for Ael. From the above it is concluded that, as for the electrolytes, the electrical interaction effects (represented by the contribution Ael) on the intrinsic volume are closely related to the changes in the dynamic structure of the water molecules adjacent to the solute. The sign of Ael for methanol (i.e. positive) is the same as for cations, possibly indicating that this small alcohol resides rather centrally in the partial clathrate-like cage which is more loosely structured than for the other solutes. The latter have negative values for Ael, which is characteristic of an anion.This suggests that the polar group packs (the OH probably hydrogen bonding) towards the wall of the cage (in which the apolar group resides), behaving rather like a water molecule would, thus creating minimum disruption in the hydrogen-bonding structure. This sort of structuring is analogous to an anion with a hydrophobic part which forms a water-anion framework edge, the hydrocarbon part occupying one of the framework cages.24406 Limiting Volumes and Compressibilities Change of Temperature This was considered for the alcohols. The above equations were used to estimate the hard-sphere radii, the packing distance in water, A!, and hence Ael, at 5 "C (table 2). These results show that A,, becomes more negative as the temperature falls. The correlation with the viscosity of the pure liquid solute is maintained and (37) in A K-l, with average deviation of f 5 x T = A/q+C-Bq (18) equation15 using the coefficients listed in table 2.2 of ref.(15). The change in A,, is in the opposite direction to Ag which becomes slightly larger, but it is the increase in rhs that mainly offsets the effect of A,,. As the structure of water becomes more open as the temperature falls to 5 "C the increase in the packing volume is expected. The number of water molecules adjacent to the solute were estimated from eqn (15) (with Vw = 18.016 cm3 mol-1 and using V:n values from table 1). There is only a slight increase in N as the temperature falls (table 4), ca. 0.03 per degree fall. Again A,, is linearly correlated with N, i.e. for 5 "C (19) with average deviation of kO.01 A.Ks and V:n are correlated via eqn (15) and (16) and with an average difference in N from the two equations of k0.03. Despite the fact that V: at 5 "C is now much larger than the molar volume of the pure liquid (table 2) it can be seen that the underlying mechanisms (affecting the value of V&) are the same as those at 25 "C. The rhs for the pure organic liquid incr2ases very slightly as the temperature falls. The packing distance decreases by ca. 0.04 A, reflecting the denser packing at the lower temperature; water shows the opposite trend (as the compressibility does relative to the compressibility of the organic liquid) due to increased bulkiness of structure. (aA,,/aT) = 1.4 x lo-* - 2.55 x 10-2(aq/aT) (aq/aT) at 25 "C was estimated from the AelA = 0.195- 1.991 x 10-2N N = 3.37+0.726 ViS (20) Components of A v In mixed functional solutes, the solvation structure of each group is not qualitatively perturbed by the presence of the other, even when the two are in close pro~imity,~~ so that there is qualitative (but not formal) support for the use of group additivity of thermodynamic quantities.For example, the contribution of a methylene group to AV may be obtained from the alcohol series, i.e. and this can be compared to the theoretical value obtained from, e.g. where A V for the hydrocarbon solute molecule is obtained from the relationship26. 27 A V" = - (43/24) Y"x; (23) where A V" is the decrease in volume when either aliphatic or aromatic hydrocarbons are transferred from a non-polar medium into water, Y" is the number of water molecules in the first layer around the hydrocarbon solute molecule and xl is the mole fraction of water in the first layer which is tetra-hydrogen bonded.xi is known for a givenJ . V. Leyendekkers 407 Table 5. Contribution of the methylene group to AVa T/"C: 1 5 20 25 source - 1.9 - 1.tIb - 1.5 - 1.4b theoretical, ref. (26) -1.1 - - 1.2 - estimate, ref. (27) - -2.2 - - 1.1 estimate, this work a Average values, cm3 mol-'. Interpolated. Table 6. Group contributions to V g and A V O a X V:n(X) range of n -AV(X) range of n source ~~ 17f0.2 17k0.3 17f0.1 CH,OH 31.8 fO.0 CH30CH3 63.0 f 0.0 OCH,CH, 39.5 k 0.1 39.6 f 0.1 38.9 & 0.1 HOH 24.7 +_ 0.1 24.7 & 0.1 OH 13.2 f 0.1 CH2 2-5 3-6 1-4 1-4 1-4 1-4 2-4 1-4 1-5 1-4 3-6 1.1 k0.05 1.1 f0.05 1.0 kO.1 2.1 kO.1 4.2 & 0.1 2.5 f 0.1 2.5f0.1 2.5 & 0.1 1.7 +_ 0.05 1.7 10.1 0.9 f 0.0 1-5 2-6 1-4 1-4 1-4 1-4 1-4 1 4 1-5 1-4 2-6 H(CH,),OH (CH,),(OH), H(CH,),OCH,CH,OH H(CH,),OCH,CH,OH CH,(OCH,CH,),OCH, CH,(OCH,CH,)OCH, H(OCH,CH,),OH H(CH,),OCH,CH,OH H(CH,),OCH,CH,OH H(CH,),OH (CH,),(OH), a 5 Quantities are deviations from the average.hydrocarbon in water and Yc can be estimated from molecular models.26 Eqn (23) was derived on the basis of a statistical-mechanical theory of aqueous solutions of hydrocarbons26 and assuming partial cages form around the solute. Computer simulations of aqueous systems containing non-polar solutes support the picture of enveloping cage structures. 28 The thermal distribution of orientations of water molecules that are near neighbours to a solute such as argon in liquid water is similar to that for the argon clathrate hydrate,28 although, unlike the clathrate, the hydration shells in the liquid are poorly defined, the molecules exchanging in picoseconds.Values of A G,, from the results here [using equations of the type of eqn (21)] and eqn (22) and (23) with data of ref. (26), are listed in table 5 for 1, 5, 20 and 25 "C. The values of A V estimated by Friedman and Scheraga,' are based on the assumption that V: is given by the molar volume, &, of the pure liquid solute. At 25 "C the results here show that V,: is close in value to &, but this is far from the case at 5 "C. This is why the AVCH, values calculated by Friedman and Scheraga show the wrong temperature dependence.The AGH, values estimated here show the correct temperature trends but are 0.3 cm3 mo1-I more positive at 25 "C and 0.4 cm3 rno1-I more negative at 5 "C. This represents a difference of ca. k0.5 in the value of [ Yc(propane) - Y"(ethane)], which is within the uncertainties of estimating Y". The values of x i at 25 and 5 "C given in ref. (26) are 0.410 and 0.498. At 25 "C Y" is 17.97 for propane and 15.93 for ethane, which gives AY&, as 2.04. At 5 "C, AYEH, is 2.13. These values of Y" are of the order of N (table 4). When the CH, contribution is subtracted from the total A V for the alcohols, diols and alkoxyethanols the resulting A V values for the polar remnants are directly proportional to the square of the dipole moment (from table l), i.e.AV(po1ar) = 0 . 1 7 2 ~ ~ (kO.1 cm3 mol-'). This indicates that the resolution of A V is consistent in terms of the408 Limiting Volumes and Compressibilities Table 7. Compressibility components of the solute no. n io4Ra 104~0,; I04qnd miscellaneous ~ I - - 17.5 2 - -12.46 3 - -18.56 4 - -3.2 5 - -0.24 6 - -2.34 7 - 5.94 8 - 8.35 9 - 9.6 H(CH,)nOH 10 1 12.6 11 2 10.0 12 3 6.3 13 4 4.6 14 5 2.6 16 3 9.5 17 4 9.4 18 5 8.1 19 6 6.6 H(OCH,CH,),OH 20 2 4.3 21 3 2.3 22 4 1.1 23 1 8.6 24 2 6.3 25 3 4.4 26 4 2.8 27 1 6.0 28 2 3.4 29 3 1.5 CH3(OCH,CH,),0CH3 H(CH,), OCH,CH,OH 30 4 -0.45 H(CH,),OH 10 1 7.3 11 2 -0.8 12 3 - 10.8 13 4 -19.0 14 5 -27.1 (CH,),(OH), 15 2 -3.8 16 3 -0.7 17 4 -5.3 18 5 -12.2 19 6 -18.9 - I3 -9 - 14 0.9 1.2 0.1 8 10 13.1 12.8 10.0 7.4 5.2 4.7 10.4 - - - - - - - 10.9 (8.6) (6.7) (3.1) 7.6 5.3 4.0 2.9 7.3 -1.1 -11.2 - 19.4 - 27.1 - - - - - T=25"C -25.5 -21.7 -48.0 -11.1 - 19.4 - 13.4 - 13.2 - 15.8 - 18.4 -9.2 - 12.9 - 16.4 - 19.9 -23.5 - 13.1 - 16.9 - 20.5 - 24.1 - 27.7 - 30.6 - 38.9 -47.3 -22.6 - 30.6 - 38.9 -47.3 - 18.0 -21.3 - 25.0 - 28.3 12.5 12.7 34.0 12 20.6 13.4 21.2 25.8 31.5 22.0 22.9 23.8 25.1 28.2 23.5 (17.8) (26.4) (29.9) (32.2) (34.3) (34.9) (41.2) (48.4) 33.5 (39.2) (45.6) (50.4) 25.6 26.6 29.0 31.2 T=5"C -20.0 27.3 -28.9 27.8 -37.0 25.8 -45.1 25.7 -53.2 26.1 -28.2 (24.4) -37.6 (36.9) -46.1 (40.8) -62.6 (43.7) -54.4 (42.2) 10 12 15 25 24 23 36 36 36 53.7 38.8 31.5 27.0 25.7 40.2 (30.4) (34.3) (31.7) (28.9) (26.6) (35.3) (29.8) (27.2) 32.7 (27.6) (25.1) (22.8) 31.7 27.2 25.3 23.7 62.6 44.0 32.0 26.1 22.5 (39.6) (45.1) (40.6) (35.6) (32.0) 0.2 0.3 0.3 0.6 0.5 0.5 0.8 0.8 0.8 1.2 0.9 0.7 0.6 0.6 0.9 (0.7) (0.8) (0.7) (0.6) (0.6) (0.8) (0.7) (0.6) 0.7 (0.6) (0.6) (0.5) 0.7 0.6 0.6 0.5 1.3 0.9 0.7 0.5 0.5 (0.8) (0.9) (0.8) (0.7) (0.7) a Experimental data: solutes 1-9, same as for footnote b in table 1 ; solutes 10-19, ref.(29); solutes 20-26, ref. (30). bSee table 2, bracketted values are estimates based on data for solute 23. (8"- VP,,)/(B,+ I), eqn (2): B,= 3004.5 bar (25 "C) and 2763.5 bar (5 "C). * From eqn (3), bracketted values indicate that K," has been used for K;. (KPn/VPn), bracketted values as for footnote d. Experimental isothermal compressibility : solutes 10-14, 0. Kiyohara and G. C. Benson, J. Solution Chem., 1981, 10, 281; M.Dim Peiia and G. Tardajos, J. Chem. Thermodyn., 1979,11,441; J. Timmermans, Physico-Chemical Constants of Pure Organic Compounh (Elsevier, New York, 1965), vol. 2; International Critical Tables (McGraw-Hill, New York, 1928), vol. 111, p. 41-2 and ref. (9). Bracketted values are estimates from eqn (7) and Tait equation', using data from ref. (15). vpB, is isothermal compressibility of water,' 45.52 x 10-o bar-' (25 "C); 49.50 x bar-' ( 5 "C); bracketted ratio: as for footnote d. Units: pressure in bar, volumes in cm3 mol-'.J . V. Leyendekkers 409 characteristics of the apolar and polar groups. However, as ki values for the solutes in table 1 are very small, relative to those for the amino acids, the predominant compressional effect on the solvent derives from the combined volumes of the polar and apolar groups i.e.AV = k, Vlt [from eqn (4), with ki small]. The cross correlations between hydroxy and apolar groups are generally small for the systems considered here, as can be seen from the additivity of the group contributions to the volumes (table 6). The first members of some series, e.g. the alcohols and diols, are exceptions. Limiting Partial Molal Compressibility Values of lC&mp were estimated from eqn (2) and thence Kin could be estimated using the experimental compressibility data.29-31 The values are listed in table 7. With the exception of methanol the intrinsic compressibility pi”, [ - (3 In Vi”,/3P)T] is lower than the compressibility of water, generally ranging from (25-35) x lop6 bar-’.The sugars are much less compressible, Bpn values being a third or less than that of water, whilst methanol is unique in having a En larger than Studies of aqueous solvation structure by computer simulations showz8 that for purely apolar solutes the mutual interaction among solvent molecules is not in balance with the much weaker solute-solvent interactions. The solvent orients to maintain the degree of hydrogen-bonding in the bulk (i.e. hydrogen bonding groups straddle the weakly inter- acting non-polar surface) and this resembles clathrate hydrate structures locally. 2 5 9 28 The frequency of relatively weak, distorted, hydrogen bonds is reduced so that for the clathrate-like structures near-neighbour interactions are, on average, slightly stronger than in the bulk.28 This ‘stiffening’ of the solvent around the solute could be expected to cause an effective decrease in the compressibility of the ‘solute space’.On the other hand, according to the simulations, polar hydrogen-bonding solute groups interact with water in a similar manner to that of another water molecule so that the hydrogen- bonding structure of water is hardly affected. For mixed functional solutes then the ‘stiffening’ would be the dominant effect, and if, as deduced from the Ael analysis, the hydrophilic groups bond into the framework of the cavity this would enhance the stiffening effect. The low values of Pp, relative to the compressibility of water could be explained in this way. The uniqueness of methanol in this respect (which also applies to A,J might be due to a greater mobility of this solute within its cavity in the water structure.A clathrate structure would affect the rotational motion of the solute molecule but it is not clear how this would influence En. If the low values of Dpn relative to water are due to a fairly rigid framework around the solute, then En should be correlated with N (table 4). For group A, excluding pentanol and 2-ethoxyethanol, Dpn is linear in N with N = 19 when Dp, is zero. For group B (plus pentanol and 2-ethoxyethanol) Dp, is again linear in N with N = 26 when Pp, is zero. for water. On the other hand, all the values of En correlate with N’, i.e. l.05En = 8.078 -0.3894” 105xn = 3.948 + 22.086Ae,. and from eqn (I 5 b) Eqn (24) and (25) give an average deviation of 0.15 x bar-’ in En with a maximum deviation for methanol of 0.43 x bar-’.From eqn (24), N’ = 20.7 when En is zero. These correlations support the conclusion above, that the compressibility of the solute in water is directly related to the number of adjacent water molecules. From purely geometric (packing) considerations (using N ) there could be two types of cavities with solute to water ratios around 19 and 26. These ratios are close to the 17 and 28 found from the Raman21 and X-rayz2 data discussed above. The polar effects cause changes in410 Limiting Volumes and Compressibilities the dynamic behaviour of the adjacent water molecules (all, or a fraction as for group B), the maximum number of water molecules that can be affected is ca. 21. The "0 n.m.r.results for sucrose, discussed above, indicate that the correlation times of 16 (i.e. about N') water molecules (table 4) are affected, which is within the limit of 21. Other Urea and thiourea En values versus N extrapolate to N = 27 En zero) so that they fall within group B in geometric terms. However, A,, indicates that polar influences extend beyond the first layer, suggesting that these solutes are hydrated species within the cavities. give 20 to 25 instead of 16. Discussion The partial molal volume is simply related to integrals over the radial distribution functions for the various components in a mixture and hence plays a fundamental role in the statistical-mechanical theory of It is obviously useful, therefore, to have a general model such as the chemical thermodynamic (CT) one used here and elsewhere which can describe in detail the volume characteristics of a wide range of solutes lo amino acids' and organic compounds such as those considered here) in a range of solvents, including water.The SP theory has provided an effective description of the solubilities of non-polar gases in water;33 however, the more recent Pratt-Chandler (PC) has also done this, yet the PC theory predicts a substantially higher radial distribution function,28 which indicates the importance of using a more realistic model for understanding solution structure. The SP theory has the advantage that it is simple to use for thermodynamic analyses and could be expected to predict the purely geometric or packing effects quite ''3 '* By combining this theory with the CT model, the charge effects can be accounted for.This approach greatly enhances the range of the SPT. As noted above, the errors involved in the extrapolation method used for gas solubility studies, which show up well in the temperature dependence of the diameter4 are not relevant here. The overall results show that the CT model can be applied to solutes of moderate polarizability. Use of the SPT enables extension of the analyses to a molecular level. In addition, SP theory simplifies the use of the model for other solvent systems, since Vk for 1,4-dioxane can be predicted fairly closely using the hard-sphere radius and the geometric packing distance so that k, can be estimated. Provided the dielectric and PVT characteristics of the solvent are known a whole range of solutes then can be studied using the methods outlined here and in parts 1 and 3.The effect of temperature can also be assessed. The rhs values, coupled with the CT model analysis, provide a quantitative structural picture that is generally in accord with results from computer simulations. A large number of studies on a variety of systems (e.g. aqueous solutions of alcohols and peptides) using computer modelling, indicates that the attractive interactions among polar solutes and water molecules are balanced and there is negligible hydrogen-bonding d i s r ~ p t i o n . ~ ~ The present analysis shows that the polar effects contribute negligibly to AV and generally less than 20 O h to the excess radius, Aw. The polar interactions appear to be linked more to dynamic mechanisms, the 'shrinking' of Vl' relative to an apolar volume (negative Ael) being associated with changes in the orientational polarisation and/or reorientational motions of solvent molecules near the solute.The computer simulations are somewhat limited in that they emphasise the short-range, time- independent structures. However, the results here show that the polar interactions, relatively small in terms of the volume effect, are nevertheless significant for time- dependent properties ; the various correlations for Ael, coupled with molecular data (experimental), show this. Other structural features that stand out are the dominance of the packing geometryJ. V. Leyendekkers 41 1 on the volume changes and the implied clathrate-like local bonding.This latter result is deduced from the analysis of A V, A and the reduced compressibility of the solute relative to that of water, and again is consistent with the computer simulation results (but giving more quantitative detail) and with results from X-ray and spectroscopic studies. 21, 22 The effect on the volume and compressibility of changing the concentration of solute will be considered in part 3 for the solutes used here. Because of the larger compressibilities of the latter (relative to amino acids) the slopes (av/arn) and (aK/arn) are slightly more complex than for the amino acids; the effect of concentration on the volume compression (part of AV) has to be taken into account. This makes the changes in the polarisation of water with solute concentration more significant than for the amino acids, for which the very large polar effects dominate over all other.References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 J. V. Leyendekkers, J. Phys. Chem., 1986, 90, 5449. J. V. Leyendekkers, J. Chem. SOC., Faraday Trans. 1, 1982, 78, 3383. R. A. Pierotti, J. Phys. Chem., 1967, 71, 2366. E. Wilhelm, J. Chem. Phys., 1973, 58, 3558. F. Hirata and K. Arakawa, Buff. Chem. Soc. Jpn, 1973, 46, 3367. N. Desrosiers and M. Lucas, J. Phys. Chem., 1974, 78, 2367. H. Heriland and E. Vikingstad, Acta Chem. Scand., 1976, A30, 692. H. L. Clever and R. Battino, in Solutions and Solubilities, ed. M. R. J. Dack, Techniques of Chemistry Series (John Wiley, New York, 1975), part 1, chap. 7, p. 425. H. S. Harned and B. B. Owen, in The Physical Chemistry of Electrolyte Solutions (Reinhold, New York, 1958). J. V. Leyendekkers, J. Chem. Soc., Faraday Trans. I , 1982, 78, 357. J. V. Leyendekkers, J. Chem. Soc., Faraday Trans. 1, 1983, 79, 1109; 1123. A. Bondi, J. Phys. Chem., 1964, 68, 441. H. Heriland and E. Vikingstad, Acta Chem. Scand., 1976, A30, 182. E. Glueckauf, Trans. Faraday Soc., 1983, 64, 2423. J. A. Riddick and W. B. Bunger, in Techniques of Chemistry vol. ZZ, Organic Solvents (Wiley- Interscience, New York, 3rd edn, 1970). J. B. Hasted, in Water: a Comprehensive Treatise, ed. F. Franks (Plenum Press, New York, 1972), vol. 1, chap. 7; A. Abragam, in The Principles of Nuclear Magnetism (Oxford University Press, Oxford, 1978). M. Jauquet and P. Laszlo, ref. (S), chap. 4, p. 213. J. B. Hasted, Aqueous Dielectrics (Chapman and Hall, London, 1973). D. N. Glew, H. D. Mak and N. S. Rath, in Hydrogen-Bonded Solvent Sysrems, ed. A. K. Covington and P. Jones (Taylor and Francis, London, 1968), p. 159. A. D. Potts and D. W. Davidson, J. Phys. Chem., 1965, 69, 996. K. Tanabe, J. Inclusion Phenom., 1984, 2, 267. H. Tanaka, K. Nakanishi and K. Nishikawa, J. Inclusion Phenom., 1984, 2, 119. P. S. Belton and K. M. Wright, J . Chem. Soc., Faraday Trans. I , 1986, 82, 451. Yu. A. Dyadin and K. A. Udachin, J . Inclusion Phenom., 1984, 2, 61. P. J. Rossky, Annu. Reu. Phys. Chem., 1985, 36, 321. H. A. Scheraga, J. Chem. Phys., 1962, 36, 3401. M. E. Friedman and H. A. Scheraga, J. Phys. Chem., 1965, 69, 3795. L. R. Pratt, Annu. Rev. Phys. Chem., 1985, 36, 433. T. Nakajima, T. Komatsu and T. Nakagawa, Bull. Chem. SOC. Jpn, 1975, 48, 788. S. Harada, T. Nakajima, T. Komatsu and T. Nakagawa, J . Solution Chem., 1978, 7, 463. S. Cabani, G. Conti and E. Matteoli, J. Solution Chem., 1979, 8, 11. J. G. Kirkwood and F. P. Buff, J. Chem. Phys., 1951, 19, 774. R. A. Pierotti, Chem. Rev., 1976, 76, 717. L. R. Pratt and D. Chandler, J. Chem. Phys., 1985, 67, 3683. Paper 71337; Received 23rd February, 1987

ISSN:0300-9599    

DOI:10.1039/F19888400397

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

J . Chem. Soc., Furaduy Trans. I, 1988, 84(2), 413431 Paramagnetic Rhodium Species in Zeolites Part 1.-RhNa-X and RhNa-Yf. AbdeLhamid Sayari,* John R. Morton and Keith F. Preston Division of Chemistry, National Research Council of Canada, Ottawa, Ontario, Canada KIA OR9 At least five different paramagnetic RhT1 species have been detected by e.s.r. following activation by 0, at 200-475 "C of ion-exchanged RhNa-Y prepared either from [Rh(NH,),Cl]Cl, (sample I) or from RhCl, - 3H,O (sample 11). The conditions of formation of these species, their stability and their possible locations have been studied. Almost no paramagnetic species were detected in samples activated in 0, at temperatures higher than 500 "C, indicating a total oxidation of rhodium. In samples activated in 0, at temperatures > 200 "C and then evacuated at increasing temperatures, all Rh'I species disappeared progressively, while a new species E(Y) with gll = 2.068 and g , = 2.674 developed.The same species,E(Y) was also generated in sample I1 following evacuation at temperatures > 200 "C without 0, activation. It is believed that species E(Y) is composed of isolated Rho atoms stabilized in the secondary channel of Y zeolite. Possible reduction mechanisms are discussed. A striking similarity between the overall chemistry of Rh in Na-X as revealed by e.s.r. and that of Rh in Na-Y was found. Supported Rh catalysts have been successfully used for -a wide variety of reactions including reduction of NO,'. hydrogenation of CO,, methanation of C0,,4 dimerization and hydroformylation of alkene~,~-' and carbonylation of alcohols.8* Of particular interest is the use of zeolites as a support.Indeed, in the latter case Rh can give rise, upon suitable treatment, to narrow distributions of small metallic particles throughout the zeolite'' as well as to numerous organometallic clusters inside the zeolite framework. 11-13 This makes zeolite-supported Rh catalysts very attractive both for heterogeneous l5 and for the so-called heterogenized homogeneous catalysis, where the zeolite is known to play the role of a 'solid solvent'.'' Several techniques have been used to characterize Rh-zeolite catalysts. These include infrared,l6' l7 ESCA,16. 18-,' temperature-programmed oxidation and reduction,,l and electron microscopy. lo In most reactions catalysed by rhodium, the active species have been identified as Rh', Rh'" or metallic rhodium.Moreover, there is growing evidence that the oxidation state of Rh plays an important role in directing the hydrogenation of CO towards the production of either hydrocarbons, methanol, or higher ~ x y g e n a t e s . ~ ~ - ~ ~ Quite a few processes are also believed to be catalysed by Rh1'.25-28 For these reasons, and because RhII is e.s.r.-active, several e.s.r. investigations of paramagnetic rhodium species in Na-X and Na-Y zeolites have been published in recent Despite the similarity between the structures of Na-X and Na-Y zeolites, the comparison between e.s.r. studies dealing with RhNa-X and RhNa-Y seems to indicate different behav- i ~ u r . ~ ' , ~ ' In particular, Naccache et al.,O reported the formation of RhII-CO and Rh"- 0, adducts when CO or 0, is admitted at room temperature onto a RhNa-Y sample activated under oxygen then evacuated at 500 "C.However, according to Goldfarb and t NRCC no. 27616. 413414 Paramagnetic Rhodium Species in Zeolites Kevan,,l none of these adducts is formed on RhNa-X. It seemed useful, therefore, to undertake a further e.s.r. study of Rh-exchanged Na-X and Na-Y and compare some of the rhodium chemistry in these zeolites. In this paper, we report on Rh-containing paramagnetic species generated by various activation treatments, while adsorption experiments will be discussed in a forthcoming paper.,, Preliminary results dealing with RhNa-Y have been reported in a previous p~blication.~~ Experimental Materials The zeolites Na-Y(L2Y-52) and Na-X( 13X) were obtained from Union Carbide. The rhodium compounds, [Rh(NH,),Cl]Cl, and RhC1, - 3H,O, were supplied by Strem Chemical Company.Two RhNa-Y samples were prepared for this study by cation exchange of trivalent Rh for Na+ in Na-Y. The first sample (2%w/w Rh), referred to as RhNa-Y(I), was prepared by ion exchange of zeolite Na-Y with a 0.002 mol dm-, aqueous solution of [Rh(NH,),Cl]Cl,. The mixture was stirred continuously for 48 h at room temperature. The solid was then filtered and thoroughly washed with distilled water until free of chloride ions. After drying in air at room ter,perature, the sample was white. The sample designated RhNa-Y(I1) was prepared from RhC1, - 3H20 according to a method recently suggested by Shannon et aL2' The cation exchange was carried out in a dilute solution of RhC1;3H2O (0.002 mol drn-,) at 90 "C for 24 h.The subsequent steps are the same as for RhNa-Y(1). The rhodium content of this sample was also 2% w/w (ca. 3.5 Rh cations per unit cell) and its colour was light yellow. The RhNa-X (2% w/w Rh) sample used in this study was prepared as for RhNa- W). Treatment The standard activation treatment consisted of heating the zeolite samples in a grease- less Pyrex microreactor under flowing oxygen. The temperature was slowly raised to a preset value, typically in the range 200-500 "C. When higher temperatures were needed, a quartz reactor was used. The samples were kept overnight under these conditions, cooled to room temperature under oxygen, then flushed with dry nitrogen.For e.s.r. measurements, the samples were transferred into 4 mm 0.d. Suprasil tubes and evacuated overnight at various temperatures in the range 25-500 "C. The transfer of samples into e.s.r. tubes was carried out in a glove box under dry nitrogen. For some experiments, fresh samples were thermally treated under helium or under vacuum instead of oxygen. Where applicable, the activated samples will be referred to as RhNa- Y(I or 11) or RhNa-X x / y / z , where x, y and z stand, respectively, for the temperature of 0, activation (in "C), and the temperature (in "C) and the time (in h) of subsequent evacuation. Most of the e.s.r. spectra were recorded at 77 K using a Varian E-12 spectrometer equipped with magnetic field and microwave frequency measuring devices. Occasionally, e.s.r.spectra were recorded at 4 K or at room temperature. Results RhNa-Y (I) Oxygen Treatment Fig. 1 shows the e.s.r. spectra of several RhNa-Y(1) x/25/ 18 (x = 200-540 "C) samples. Upon oxygen activation at 200 "C [fig. 1 (a)], the sample turned from white to yellow and a first paramagnetic species, A(Y), was formed; it had an orthorhombic e.s.r. spectrumA . Sayari, J . R. Morton and K. F. Preston 41 5 w"" t 2.011 1.986 Fig. 1. First-derivative e.s.r. spectra of RhNa-Y(I)x/25/18: (a) x = 200 "C, (a') x = 200 "C, spectrum recorded at 25 "C, (b) x = 250 "C, (c) x = 300 "C, ( d ) x = 400 "C, (e) x = 475 "C, (f) x = 540 "C. All spectra but (a') were recorded at 77 K with a microwave frequency of 9.1 GHz.The numbers on the right-hand side indicate the relative gain. with g, = 2.088, g, = 2.048 and g , = 1.968. In the presence of oxygen (> 60 Torr?), the spectrum broadened beyond detection. A sharp signal at g = 2.002 always appeared at these low activation temperatures; however, species A(Y) was observed alone when the e.s.r. spectrum was recorded at room temperature [fig. 1 (a')]. It therefore seems that the signal close to the free-electron g-factor corresponds to minute impurities that were destroyed at higher activation temperatures. This species will not be discussed further. Upon oxygen activation at 250 "C, the sample turned dark brown and species A(Y) vanished while two new species developed [fig. 1 (b)]. The first, B(Y), can be better seen in spectra of samples activated at higher temperatures (see below).The other species, designated C(Y), exhibited a broad e.s.r. signal with g,, = 2.285 and g , = 2.082. Samples activated in 0, at 300-400 "C displayed even more complicated e.s.r. spectra, and at least three different species could be distinguished [fig. l(c) and (41. The most important species, B(Y), began to appear at 250 "C; it has an axially symmetric g matrix with g,, = 1.881 and g , = 2.550. The high-field component exhibited a hyperfine splitting ( A , / = 34 G) due to interaction with lo3Rh(Z = i, natural abundance = 100 YO). As seen in fig. l(c) and (d), the central regions of the spectra were very complicated and no precise assignments could be made. The problem of signal assignment was not solved either by changing the temperature of the e.s.r.measurements or by changing the microwave power. We believe, however, that these signals correspond to the same paramagnetic species experiencing different crystal fields. Based on measurements on Rh -f 1 Torr = 101 325/760 Pa.416 Paramagnetic Rhodium Species in Zeolites 2.265 x 16 / \ \ A 1.881 2.068 Fig. 2. First-derivative e.s.r. spectra of RhNa-Y(1) 250/y/z (a) y = 25 "C, z = 18 h; (b) y = 250 "C, z = 18 h; (c) y = 300 "C, z = 3 h; ( d ) y = 300"C, z = 20 h; ( e ) y = 350 "C, z = 3 h; cf) y = 350 "C, z = 20 h; ( g ) y = 400 "C, z = 18 h; ( h ) y = 450 "C, z = 18 h; ( i ) y = 500 "C, z = 18 h. The same sample was used for all experiments. Spectral conditions as in fig. 1. The numbers on the right-hand side indicate the relative gain.in conventional supports,34 as well as on literature data, we tentatively assign g, = 2.266, g, = 2.1 14 and g, = 1.980 to a species C'(Y), and g,,, = 2.186 to a species D(Y). In contrast to species A(Y), none of these new paramagnetic species [B(Y), C(Y), C'(Y), D(Y)] could be observed at room temperature. However, like species A(Y), the e.s.r. spectra of B(Y), C(Y), C'(Y) and D(Y) were reversibly broadened by oxygen. Heating the RhNa-Y(1) sample in 0, at 475 "C [fig. 1 (e)] brought about a decrease in the intensities of species C'(Y) and D(Y), and the appearance of new minor species with g values between 2.1 and 1.95. We believe these to be several 0; species; they were not, however, studied in detail. Note the formation of a very weak signal with g,, x 2.00 and g , = 2.67 [species D'(Y)].When the sample is heated under 0, at temperatures > 500 "C (up to 620 "C), species B(Y), C'(Y) and D(Y) vanished completely, while only very weak e.s.r. signals persisted [fig. 1 (f)]. Vacuum Treatment It should be emphasized that activation of RhNa-Y(1) with 0, was essential for the observation of any e.s.r. signals whatsoever. In 0,-activated samples, changes in the e.s.r. spectra were obtained by evacuating the samples at various temperatures. Evacuation at 200 "C for 18 h of RhNa-Y(1) 200/25/ 18 resulted in a significant decrease of signal A(Y) without the appearance of any new paramagnetic species. Evacuated at higher temperatures, the sample turned dark brown, while species A(Y) vanished without further changes in the e.s.r.spectra. Quite different behaviour was exhibited by samples activated in 0, at temperatures between 250 and 475 "C. Typical changes in the e.s.r. spectra are shown in fig. 2 for a sample heated in flowing 0, at 250 "C and evacuated at different temperatures forA . Sayari, J. R. Morton and K. F. Preston 417 Fig. 3. First-derivative e.s.r. spectra of RhNa-Y(1) 540/y/z: (a) y = 25 "C, z = 18 h; (b) y = 200 "C, z = 18 h; (c) y = 300 "C, z = 18 h; ( d ) y = 500 "C, z = 3 h. The same sample was used for all experiments. Spectral conditions as in fig. 1. The numbers on the right-hand side indicate the relative gain. different periods of time, usually overnight (I 6-1 8 h). Similar behaviour was observed for samples pretreated in 0, at temperatures between 250 and 475 "C.The changes in the e.s.r. spectra clearly demonstrated that during the evacuation of the sample at high temperature, events other than the mere desorption of oxygen took place. The resulting changes in the e.s.r. spectra can be summarized as follows. First, all signals in the middle region of the spectrum disappeared gradually as the temperature or the time of evacuation increased. Secondly, the intensity of the e.s.r. signal of species B(Y) passed through a maximum at ca. 300°C before it eventually vanished at 500°C. Sim- ultaneously, a very strong signal corresponding to a new species E(Y) developed. As seen in fig. 2(i), species E(Y) has an axially symmetric g matrix with g,, = 2.068 and g , = 2.674. Similar results were found for a sample activated in 0, at 400 0C.33 It has already been mentioned [fig.l(e) and (f)] that 0, activation at temperatures > 475 "C generated a weak signal with g,, = 2.00 and g , = 2.67 [species D'(Y)]. Fig. 3(b) and (c) show that the intensity of species D'(Y) did not change upon evacuation overnight at 200 or 300 "C. However, following evacuation at 200 "C, all signals in the central region of the spectrum disappeared, leaving a weak background due to Mn2+ impurities. The g,, feature of species D'(Y) could hardly be distinguished. Heating the same sample under vacuum at 500 "C for 3 h gave rise to species E(Y) with an intensity 20 times higher than species D'(Y). Note that no other species were observed during the evacuation process. It is also important to draw attention to the fact that for RhNa-Y(1) activated in 0, at 250 "C, species E(Y) begins to develop at an evacuation temperature as low as 250 "C [fig.2(b)], whereas for RhNa-Y(1) activated in 0, at 540 "C, temperatures > 400 "C are required. Table 1 summarizes the e.s.r. parameters of the various species and the conditions under which these species are generated as a result of oxygen activation and subsequent evacuation. In addition, fig. 4 provides a graphical representation of such conditions.418 Paramagnetic Rhodium Species in Zeolites Table 1. Summary of experimental results for RhNa-Y(1) T/"C paramagnetic 0, activation evacuation speciesa comments none 200 - - 250 300-475 250-475 500-620 - - 25-500 25 200 250-500 25 25 250-500 25 200-300 > 400 none g , = 2.088 g3 = 1.968 A intensity decreases none - - 2-08' species B begins to form C (: 2.285) as temperature increases, all signals become sharper, intensities of C and D decrease, intensity of B increases g, = 1.980 D gisO = 2.186) ? as temperature increases, intensities of C' and at ca.400 oC,b E appears and becomes stronger at higher temperatures; at ca. 450 "C, species B, C' and D disappear D decrease, intensity of B passes through a maximum B, C', D L - 2.67 D' (k- 2*oo 1 all signals are very weak and other minor species J D' all other minor species disappear E intensity of E reaches a maximum at evacuation temperature of 500-600 "C ~~ ~~~ a In text, all paramagnetic species P are referred to as Po(). Depending on temperature of 0, treatment.RhNa-Y (11) Activation of RhNa-Y(I1) under flowing oxygen at 200 "C followed by evacuation at room temperature [fig. 5(a)] generated only a weak and broad signal at g,, = 2.268 and g , = 2.073 [species C(Y)]. The intensity of this signal increased when the 0, activation was carried out at 300 "C; however, no new species appeared except for a few weak features at g = 2.00 [fig. 5(b)]. In particular no species B(Y) was formed as in the case of RhNa-Y(1) under similar conditions. Species B(Y) with slightly different g values (g,, = 1.854, g , = 2.575) was generated when RhNa-Y(I1) was activated with 0, at 400 "C and evacuated at room temperature [fig. 5(c)J. In this case, the high-field component is rather broad and no hyperfine splitting is observed.Traces (d)--(g) of fig. 5 represent the evolution of the e.s.r. spectrum of RhNa-Y(I1) 400/25/18 upon further evacuation for periods of 18 h at 100 "C increments up to 400 "C. Evacuation at 100 or 200 "C did not give rise to any noticeable change. However, the e.s.r. signal of species B(Y) was significantly weaker following evacuation at 300 "C [fig. 5 0 1 . Further increase of the temperature to 400 "CA . Sayari, J . R. Morton and K. F. Preston 419 E 5 00 0 \ .- 400 2 0 > "0 300 E E Y g 200 Y 100 none 100 200 / B C / B C' D I 6 D ' 300 400 ! D' - I0 temperature of oxygen activation/'C Fig. 4. Graphical representation of the conditions of formation of the various paramagnetic species in RhNa-Y(1). The figure does not attempt to indicate the relative intensities of the various spectra and it should be remembered the 0,-activation step always preceded the evacuation process.suppressed all previous e.s.r. signals and species E(Y) developed. Slight variations in the g values of species E(Y) (g,, = 2.064, g , = 2.685) were also observed. As seen, the parallel feature of E(Y) is always weak and broad, leading to a large uncertainty for g,,. In contrast to RhNa-Y(I), heating RhNa-Y(I1) under vacuum without prior activation in oxygen generated species E(Y) at temperatures > 200 "C. No intermediate paramagnetic species were obtained at any temperature. Fig. 6(a) gives the relative intensity of species E(Y) as a function of evacuation temperature. In another experiment RhNa-Y(I1) evacuated at 400°C was exposed to 220Torr 0, and heated overnight at increasing temperatures up to 400 "C.Fig. 6(b) gives the relative amount of species E(Y) measured under 0, pressure as a function of temperature. At the end of the experiment, the oxygen was pumped off at room temperature ; however, hardly any B(Y) species was observed. RhNa-X E.s.r. spectra of samples treated under oxygen at various temperatures and evacuated overnight at room temperature are shown in fig. 7. The g values of the most significant species along with the conditions of their formation are summarized in table 2. Such assignments were made essentially by comparison with RhNa-Y. Changes in the e.s.r. spectra upon evacuation at increasing temperatures are very much the same as for RhNa-Y(1). Typical results are shown in fig.8 for a sample activated under oxygen at 400 "C. Evacuation of this sample at 400 "C for 16 h brought about an increase in the intensity of B(X), while B'(X) disappeared. At 450 "C [fig. 8(c)], the intensity of B(X) decreased significantly and a new species, E(X), began to develop. The g matrix of E(X) is axially symmetric with gll = 2.029 and g , = 2.610. When the sample was further evacuated at 500 "C only species E(X) persisted. The intensity of420 Paramagnetic Rhodium Species in Zeolites Fig. 5. First-derivative e.s.r. spectra of RhNa-Y(I1) x/y/18: (a) x = 200 "c, y = 25 "c; (b) X = 300 "C, y = 25 "C; (c) x = 400 "C, y = 25 "C; ( d ) x = 400 "C, y = 100 "C; (e) x = 400 "C, y = 200 "C; v") x = 400 "C, y = 300 "C; (g) x = 400 "C, y = 400 "C. The same sample was used for experiments (c)-(g).Spectral conditions as in fig. 1. The numbers on the right-hand side indicate the relative gain. signal E(X) was followed as a function of evacuation temperature. Fig. 9 shows that this intensity passes through a maximum at ca. 550 "C. Ultimately, almost no signals are left after evacuating the same sample at 750 "C for 18 h [fig. 8(e)]. Note that, as in the case of species E(Y) in RhNa-Y samples, species Em) was never generated by oxygen treatment alone, i.e. a subsequent evacuation at high temperature is needed in order to generate E(X). However, as we will show in a forthcoming paper,32 E(X) can be obtained by chemical reduction of RhNa-X after an appropriate treatment. Discussion RhNa-Y Three oxidation states of Rh, namely Rho (ds), Rh" (d7) and RhIV (d5), are known to be paramagnetic with S = f.The most important Rh species with respect to homogeneous and heterogenized homogeneous catalysis, i.e. Rh'" and Rh', are diamagnetic and thus cannot be directly studied by e.s.r. As for heterogeneous catalysis, the most interesting species, Rho, tends to agglomerate in the course of the reduction process into larger metallic particles undetectable by e.s.r. In several cases, however, particularly whenA . Sayari, J. R. Morton and K. F. Preston 42 1 t/"C Fig. 6. (a) Relative concentration of species E(Y) in RhNa-Y(I1) as a function of evacuation temperature. The sample was not pretreated, and was held for 18 h at each temperature. (b) Relative concentration of species E(Y) in RhNa-Y(I1) exposed to 220Torr of 0, at room temperature, then heated for 18 h at various temperatures. The sample was previously evacuated at 400 "C for 18 h.zeolites are used as supports, several transition-metal elements can be stabilized in unusual oxidation states or as isolated atoms. Examples of unusual oxidation states of transition-metal ions include Pd1,35-38 Pd111,36-38 Nix,39940 Cu14' and RhI' . 29-33 As for isolated atoms stabilized inside the zeolite framework, several studies have been reported on Pt,42 Pd,43944 Ni45 and R u . ~ ~ Isolated Rho atoms, if stabilized, can be directly studied by e.s.r.; however, no such claims have been made so far. The first question to be dealt with concerns the nature of the paramagnetic species generated by oxygen activation in the temperature range 200-475 "C followed by evacuation at low temperature, i.e.< 200 "C. The presence of RhIV is not compatible with the observed e.s.r. parameters. Indeed in the few cases where RhIV has been positively identified, at least one g value was quite 1 0 ~ ; ~ ~ ~ ~ ~ moreover, its e.s.r. spectrum can be seen only at very low temperatures (20 K). The remaining possibilities, Rh" and Rho, cannot, unfortunately be distinguished by e.s.r. The many conflicting assignments of paramagnetic rhodium species (table 3)28-311 47-91 b ased on e.s.r. 'evidence' substantiate this statement. The e.s.r. parameters of Rh" itself are very sensitive to environment, as seen in table 3. Nevertheless, in our case all paramagnetic species generated by oxygen activation alone are probably Rh"; the occurrence of neutral atoms under such oxidizing conditions is highly improbable.Species A(Y) generated by oxygen activation at 200 "C has e.s.r. parameters very close to those observed for some RhII species in organometallic compounds having a square- planar geometry, such as Rh(TPP),67 and Rh(diphos)C12.82 Since A(Y) is the first para- magnetic species formed, we speculate that it is either [Rh1I(NH3),I2+ or [Rh"(NH,),Cl]+, which arises from [Rh(NH3),C1I2+ through the loss of one ligand and concomitant reduction. Because A(Y) has rhombic symmetry, it is probably the latter. Moreover, the fact that A(Y) could not be generated in RhNa-Y(I1) is strong evidence that in A(Y),422 Paramagnetic Rhodium Species in Zeolites t 1 954 7.First-derivative em-. spectra for RhNa-X x/25/18: (a) x = 200 "C, (b) x 300 "C, ( d ) x = 400 "C, (e) x = 460 "C, df) x = 500 "C. Spectra were recorded a microwave frequency of 9.1 GHz. Fig. X = = 250 . at 77 Rh" is still attached to some of its original ligands. Owing to the size of such complexes, they can be accommodated only in the a cage.3o Heating RhNa-Y(1) under 0, at 250°C with subsequent evacuation at room temperature generates species C(Y) with g,, = 2.285 and g , = 2.082 (gav = 2.150). We believe that this species is a bare Rh" ion linked to the zeolite framework, i.e. all original ligands have been removed. Such a conclusion stems from e.s.r. studies of Rh/Al,O,,"* where a similar paramagnetic species with g,, = 2.240, g , = 2.091 (gav = 2.14) was generated by oxidation at 500 "C followed by evacuation at room temperature.Fig. l(c) and ( d ) obviously correspond to several species. Goldfarb and Kevan31 found in RhNa-X an Rh" species having rhombic symmetry with g, = 2.516, g , = 2.560 and g , = 1.883. As in the case of our B(Y) species, the g , feature exhibited a hyperfine interaction of 32 G due to interaction with lo3Rh. This species was assigned to Rh" located in the /I cage.31 Our species B(Y) is clearly closely related to this, but it has axial symmetry instead of a slightly rhombic symmetry. Moreover, the e.s.r. signals of our species B(Y), C(Y), C(Y) and D(Y) are all reversibly broadened in the presence of oxygen, suggesting that these spFcies are located in the a cage not the p cage.Indeed, with a kinetic diameter of 3.46 A,92 0, cannot penetrate the cage. We believe, there- fore, that B(Y), C(Y), C(Y) and D(Y) are Rh" at different locations within the a cage.A . Sayari, J. R. Morton and K. F. Preston 423 25 4 Table 2. Summary of experimental results for RhNa-X g, = 2.565 B g, = 2.526 g, = 1.885 g, = 2.600 B' g , = 2.491 I I g, = 1.995 g, = 2.100 g,, = 2.289 c l T/"C paramagnetic 0, activation evacuation speciesa comments 250 1 other minor species are present - 400 25 B, B' 500 25 minor species with species B and B' disappear low intensity a In text, all paramagnetic species P are referred to as P(X). Species A(Y), as detected in RhNa-Y(1) 200/25/18, is fairly stable under vacuum at 200 "C, but at higher temperature it is likely that more ligands will dissociate, leaving a more reduced rhodium species either as Rh' or more probably as metallic particles that cannot be detected by e.s.r.The colour change from yellow to dark brown supports such a suggestion. Evacuation of RhNa-Y(1) 250/25/ 18 at increasing temperatures yields a totally different picture (fig. 2). The intensity of E(Y) increases as the evacuation temperature increases up to 500 "C, at which temperature E(Y) is the only paramagnetic species left. The same trend applies for RhNa-Y(1) .x/25/18 (x = 300-475 "C). The e.s.r. spectrum of E(Y) does not broaden in the presence of oxygen, indicating that E(Y) is located in hidden sites such as the D cage or the hexagonal prisms of the zeolite. Species with very similar parameters which have been r e p ~ r t e d ~ ~ - ~ l for RhNa-X and RhNa-Y have been assigned to Rh" and mixed-valence Rh-Rh dimers.The possibility of such species being Rho was ruled out either because of the oxidizing atmosphere during thermal treatment,30 or because all e.s.r. signals disappear upon reduction with H, at elevated temperature^.^^ However, as seen in fig. 1, species E(Y) was never formed under 0,: it was observed following subsequent evacuation at high temperatures. Moreover, whatever its oxidation state, H, treatment at high temperature would result in its disappearance either by further reduction or by agglomeration. Therefore the possibility that E(Y) is actually Rho cannot be lightly dismissed. Our results show that vacuum treatment of RhNa-Y(I1) at 400 "C (18 h) leads to a highly dispersed metallic phase.Preliminary experiments on hydrogen adsorption carried out in a conventional vacuum line gave an H/Rh ratio of 1.1. At the same time a strong e.s.r. signal of species E(Y) was observed, suggesting that species E(Y) are isolated Rh atoms. This interpretation is in agreement with the findings of Okamoto424 Paramagnetic Rhodium Species in Zeolites 2.565 100 G 2.600 H 2.526 xl v t t 1885 1995 xl Fig. 8. First-derivative e.s.r. spectra for RhNa-X 400/y/18: (a) y = 25 "C, (b) y = 400 "C, (c) y = 450 "C, ( d ) y = 500 "C, (e) y = 750 "C. The same sample was used for all experiments. Spectral conditions as in fig. 1. t/"c Fig. 9. Relative concentration of species E(X) in RhNa-X 400/y/ 18 as a function of y (evacuation temperature in "C).et aZ.,5 who detected only Rh', Rh"' and metallic rhodium in RhNa-Y prepared from RhCl, and outgassed at 25-700 "C. It has been suggested elsewhere,,l that in RhNa-X species B had migrated towards the hexagonal prisms at high temperature, becoming E. This hypothesis can be ruled out for the following reasons.A . Sayari, J. R. Morton and K. F. Preston 425 (a) Fig. 1 shows that species B(Y) was generated in RhNa-Y(1) x/25/18 with x = 250-475 "C. Under 0, at higher temperatures, B(Y) as well as all other Rh" species disappeared, while no species E(Y) was formed. (b) Fig. 5(f) shows that following evacuation of RhNa-Y(I1) 400/25/18 at 300 "C, species B(Y) has disappeared almost completely, while species E(Y) has not yet formed. (c) Comparison between fig.1 and fig. 6(b) shows that species E(Y) is much less stable in oxygen than species B(Y). Therefore E(Y) cannot be a B(Y) species in an even more hidden site. We believe that the higher reactivity of E(Y) with respect to 0, is due to the fact that E(Y) is a more reduced species than B(Y). This is also in agreement with the much higher reactivity of B(Y) in the presence of reducing agents such as H, and ( d ) For RhNa-Y(1) activated in 0, at 50S620 "C, as well as for RhNa-Y(II), species E(Y) is generated by vacuum treatment without the intermediate formation of any other paramagnetic species. Owing to the much higher chemical and thermal stability of species E(Y) as compared to all other species, and also to its conditions of formation, we believe that E(Y) is a species chemically different from the others.We assign it to an Rho species stabilized inside the /? cage or the hexagonal prisms. Once the assignment of E(Y) to Rho and all other species to RhI' is accepted, one may attempt to assess the various processes that are involved in the formation of these species. Indeed, several reduction processes have been shown to take place during the thermal treatment of Group VIII (Groups 8-10) and Group IB (Group 11) cations in zeolites. The most studied is the process involving ammonia ligands that occurs during thermal treatment of transition-metal ammine complexes in zeolites.2'* 3 7 . 4 6 7 9 4 , 9 5 For RhNa-Y, the complex [Rh(NH3),C1I2+ in Y zeolite has been reported to start to decompose under vacuum at 50 "C; at 500 "C all the RhTT' is reduced to the metallic state.21 In our case, we did not detect any paramagnetic species, suggesting that under thermal evacuation alone, neither Rh" nor isolated Rh atoms can be stabilized, i.e.the final metallic rhodium particles must be too large to be detectable by e.s.r. Reduction of zeolite supported metallic cations can also occur via other processes involving water or oxide ions (02-) of the zeolite framework. Such reduction processes are well documented in the case of zeolite-supported AgQ6 and cu. *l As for our RhNa-Y(I1) thermally treated under vacuum, three reduction schemes may be contemplated: (i) reduction by water, (ii) reduction by C1- and (iii) reduction by 0'-. The first scheme does not seem to play a major role because E(Y) forms in significant amounts only at relatively high temperatures at which most of the water has already been lost.However, water may intervene in an earlier stage of reduction of Rh"' into diamagnetic Rh'. The reduction scheme involving C1- may also be discarded as a predominant mechanism on the grounds of ESCA measurements and chemical analysis performed by Shannon et a1.'' Furthermore, RhNa-Y samples exhibit the same behaviour with respect to thermal treatment under vacuum, whether prepared from RhCl, or Rh(NO,), .19 We are left with mechanism (iii), which implies the evolution of a certain amount of oxygen during evacuation. This reduction scheme has been shown to apply for Cu- and Ag-zeolite systems; however, there are indications in the literature that such a process is of much wider applicability.Table 4 summarizes some examples of transition-metal cations supported on zeolites in which autoreduction of cations under vacuum has been reported.5.18,19,35,37 We suggest that in all these examples, 02- is a common reducing agent of the supported cations. Although we have not yet a direct proof of 0, formation under vacuum treatment, there is some indirect evidence for this mechanism. This lies in the dependence of the appearance temperature of E(Y) on oxygen activation temperature, suggesting that the mobility of 0,- depends on the temperature of 0, activation. c0.93Table 3. E.s.r. parameters of paramagnetic rhodium species material activationa g1 g2 g,b Rl glib assignment Rh/y-Al,O, Rh/y-Al,O, Rh/y-Al,O, Rh/SiO, Rh/TiO, Rh/TiO, Rh/MgO Rh/NaY Rh/NaY Rh/NaX Na,RhCl; 12H,O RhOOH Rh/TiO, Rh/MgO Rh/NaCl Rh/ AgCl Rh/NH,Cl Rh/ Ag Br Rh/ Ag Br Rh/ZnWO, in powders air, 300; vac., 500-800 2.65 2.00 vac., 200 (2 h) 2.21 2.1 1 vac., 200-500 (0.5 h) 2.68 2.22 2.00 2.43 2.00 - 2.14 - - - H,, 320 2.14 0,, 25 (5-6 days) 2.092 2.031 1.963 air, 1000 (5 h) 2.308 2.149 2.042 2.224 (6) 2.185 (9) 2.064 (21) 0,, 400; H,, 25 2.16 2.29 H,, 200 2.38 2.05 0,, 500; vac., 500 2.68 2.006 vac., 500; 0,, 500; vac., 2.69 2.18 (177) 0,, 320-400; vac., 2.516 2.560 1.883 (32) 0,, 450-500; vac., 2.61 1 2.028 - - - - 0,, 210; vac., 210 2.09 2.06 1.97 500 320-400 450-500 y-irradiation 2.479 ? 2.260 2.150 2.006 2.25 2.08 1.94 in single crystals 2.07 2.39 o,, 1000 2.265 (22) 2.053 (0) 1.349 (16) 2.087 (3) 2.004 (2) 1.977 (0) - H,, 700 y-irradiation 2.171 (12) - 2.245 (16) 2.023 (7) 1.668 (32) -- - y-irradiation 2.453 2.019 y-irradiation 2.131 2.474 actinic light 2.359 2.030 2.199 2.375 N,, 390 2.34 2.03 y-irradiation 2.422 2.01 1 2.585 (15) 2.522 (15) 1.989 (9) 2.375 (12) 2.130 (8) 2.111 (7) Rh" RhIVC Rh" Rh" Rh" Rh" Rh" Rh" Rh" Rh"-OH Rh" Rh" Rh" Rho-Rh' Rh" Rh" Rh" Rh" Rh" Rh" Rh" Rh" RhlId Rh"' RhIVf Rh" Rh" Rh" Rho' Rh1Ig Rh" Rho Rh" P o\ ref.h, 28 50 51, 52 53 54 56 57 r, 5' S' 9 47 % 5 48, 58 58, 59 48 56 56 60 61 62 63Rh/KCl Rh/LiH Rh/LiD Rh/AgCl 2 MeV electrons U.V. light Rh(TPP)' Rh(TPP)k u.v.-irradiation [Rh,CI,(DMA),I2- (nBu,N),Rh(MNT),"< (nBu,N),Rh(MNT),"." Rh(C,H,), [Rh,(O,CMe),(H,O),I+ electrochemically or y-irradiation y-irradiation reduced by NOBF, [Rh,(OzCEt), (PPh,),]' Rh2(02CR)4(PY3): Rh,(O,CMe), .2MeOH Rh(cod) (P-PZ),' [Rh2(OZCMe),., (acam),l+' Rh(PPh,),Cl,, t~ = 1,2 Rh(PPh,),(NO)Cl, Rh(NOC1,) (EtOH),, t~ = 1,2 Rh(diphos)Cl, Rh(0-Me-Cys-), RhXX'(PR,), Rh[P(OPr),], Rh(2,2'-dipyridyl), Rh( 1,IO-phenanthroline), Rh(PPh3), Rh/DPBP" Rh/polyPPh," R h / po 1 y PP h , L' 2.297 (36) 2.128 2.160 2.038 (14) 2.417 2.207 2.075 (15) 2.060 (12) 2.005 (1 1) 2.287 - - as organometallic clusters 2.089 2.029 1.990 2.103 2.031 1.971 2.447 2.019 1.936 2.35 2.015 1.950 2.17 2.087 2.019 1.988 2.035 2.020 1.955 3.96-4.32 0.98-1.68 0.8-1.33 2.105 2.016 2.011 2.22 2.10 1.99 - - 2.46 2.003 1.2 2.148 2.088-2.203 2.24 2.24 2.I 1-2.21 2.05 2.0 1 2.0 1 2.01 in polymers 2.103 2.026 1.975 1.99 2.085 (20) 2.045 2.008 1.975 1.995 (40) Rh" 1.997 Rh" 1.990 Rh" 2.1 12 (26) Rh" RhIIi 2.015 Rho' 2.436 Rh'j Rho' Rh" 2.00 Rh" Rh" 2.033 Rh" Rh" Rh" 3.76 Rh;' 1.996 ( 1 3) Rhi' 2.00 Rh:+* 1.89-1.92 Rh;' 2.19 Rh" 2.03 Rh" Rh" Rh" Rh" Rh" Rho Rho 1.98 Rho 1.97 Rho 1.995-2.003 ( 1 3-20) Rh:'" 2.13 (18) WT" Rh" 2.12 Rh" 2.000 Rho Rho 64 65 66 67 68 69 70 71 72 73 74, 75 74 76 77 79 80 81 81 82 83 84 85 86 87 87 26, 88 89, 90 91 ~ ~ _ ~ _ ~______ ~ 'Temperatures in "C.Hyperfine splitting (G) in parentheses. Assignment uncertain, it is probably Rh". At 77 K. At 4 K. 'At 20 K. Below 100 K. MNT: maleonitriledithiolate. ' In isomorphic single crystals of (nBu,N)Ni(MNT),. * Polycrystalline. E x . parameters depend on the nature of R and Y. n = 0-4. At 10&200 K. ' Below 120 K. j Reassigned as Rh", see ref.(61). TPP: tetraphenylporphyrin. DMA: N,N-dimethylacetamide. q Several other species form by annealing. ' cod : cyclo-octa- 1 S-diene, pzH = pyrazole. ' For other mixed-valence Rh see ref. (78). ' acam = HNOCCH;, h, 4 DPBP = polymeric diphenylbenzenephosphine. ' polyPPh, = phosphinated styrene-2 % divinylbenzene copolymer.428 Paramagnetic Rhodium Species in Zeolites Table 4. Literature data on autoreduction of transition-metal cations in zeolites samplesa activation techniques final state ref. RhCl,/NaY vacuum, 0.5 h, e.s.r., 25-700 "C X.P.S. RhBr,/NaY vacuum, 2 h, 300 "C X.P.S. X.P.S. Rh(NO,),/NaY vacuum, 3 h, 350 "C e.s.r., Pd(NH,),Cl,/NaM 0,, 500 "C + e.s.r. Pd(NH,),Cl,/CaX vacuum, 500 "C e.s.r. vacuum, 500 "C +O,, 3 h, 500 "C +vacuum, 12 h, 500 "C Rh'I', Rh', 5 Rho Rho 18 Rh'' : species E(Y), 19 Rh' Pd' 35 PdO Pd"' Pd' 37 a All samples were prepared by ion exchange.Table 5. Rhodium paramagnetic species in activated RhNa-X and RhNa-Y(1) RhNa-X RhNa-Y(1) g1 g2 g3 g1 g II 81 g2 g3 81 gll X y z species X y z species 200 25 18 A&) 2.092 2.031 1.976 200 25 18 A(Y) 2.088 2.048 1.968 250-450 25 18 B(X) 2.565 2.526 1.885 250450 25 18 B(Y) 2.550 1.881 2.100 2.289 - - - W ) 2.082 2.285 C'Cy) 2.266 2.114 1.980 B'(X) 2.600 2.491 1.995 - - -- - - C(X) D(X) &so = 2-16 - - - D(Y) g,,, = 2.18 > 500 25 18 nonea - - - > 500 25 18 nonea - - - 200-600 500 18 E(X) 2.610 2.029 250-600 500 18 Em) 2.674 2.064 a Minor species not assigned. RhNa-X As seen in table 5, there is a striking similarity between the overall chemistry of Rh in Na-X zeolite as 'seen by em.' and that of Rh in Na-Y. We assume, therefore, that the origin and nature of the paramagnetic species detected in activated RhNa-X are the same as for RhNa-Y(1).However, it is interesting to note that despite this close relationship between the behaviour of Rh in both zeolites some minor differences do exist. These differences are summarized below. Species B'(X) has no equivalent in RhNa-Y zeolite. Species B(X) and B(Y), although closely related, do not have the same symmetry. Species B(X) is more stable than B(Y), as it survives higher evacuation temperatures. This conclusion is reinforced by the fact that BO() can be generated by evacuating RhNa-X 200/25/ 18 at high temperature, while no paramagnetic species are generated by evacuating RhNa-Y(1) 200/25/18.These two samples were shown to contain residual ammonia.A . Sayari, J . R. Morton and K. F. Preston 429 Conclusions By closely monitoring the activation procedure a strong parallel between the conditions of formation, stability and g values of Rh paramagnetic species in both zeolites X and Y can be drawn. At least five different Rh" species are stabilized in samples activated in flowing 0, at 200-475 "C. The occurrence of these species as well as their stability depend on the temperature of both 0, activation and subsequent evacuation. All these RhII species are located in easily accessible sites, probably in the a cage. Oxygen activation at temperatures > 500 "C leads to fully oxidized samples, with the rhodium probably being trivalent.During the oxidation step, RhIII freed from its ligands migrates to the secondary channel. A new paramagnetic species, E(Y) or E(X), is generated by thermal treatment under vacuum of all RhNa-Y(I) or RhNa-X samples activated in 0, at 250 "C (200 "C for RhNa-X) or higher. The same species E(Y) is generated by heating RhNa-Y(II) under vacuum. In this case the sample does not have to be activated in 0,. Species E(Y) and E(X) are assigned to isolated Rho atoms stabilized inside the secondary channel of the zeolite. References 1 K. C. Taylor, in Catalysis, Science and Technology, ed. J. R. Anderson and M. Boudart (Springer- 2 N. K. Pande and A. T. Bell, Appl. Catal., 1986, 20, 109. 3 E. K. Poels and V. Ponec, in Catalysis, ed.G. C. Bond and G. Webb (The Royal Society of Chemistry, 4 F. Solymosi, I. Tombacz and J. Koszta, J . Catal., 1985, 95, 578, and references therein. 5 Y. Okamoto, N. Ishida, T. Imanaka and S . Teranishi, J . Catal., 1979, 58, 82. 6 N. Takahashi, Y. Fujuwara and M. Akihilo, Zeolites, 1985, 5, 363. 7 E. J. Rode, M. E. Davis and B. E. Hanson, J . Catal., 1985, 96, 563. 8 B. K. Nefedov, N. S . Sergeeva and Ya. T. Eidus, Izv. Akad. Nauk SSSR, Ser. Khim., 1976, 2271. 9 S. L. T. Anderson and M. S. Scurrell, J . Mol. Catal., 1983, 18, 375. 10 N. Kaufherr, M. Primet, M. Dufaux and C. Naccache, C.R. Acad. Sci., Ser. C, 1978, 286, 131. 11 P. Gelin, F. Lefebvre, B. Elleuch, C. Naccache and Y. Ben Taarit, in Intrazeolite Chemistry, ed. G. D. Stucky and F. G. Dwyer, ACS Symp. Ser.218 (Am. Chem. SOC., Washington D.C., 1983), p. 455. 12 F. Lefebvre, A. Auroux and Y. Ben Taarit, Stud. Surf. Sci. Catal., 1985, 24, 411. 13 E. J. Rode, M. E. Davis and B. E. Hanson, J. Catal., 1985, 96, 574. 14 L. Tebassi, A. Sayari, A. Ghorbel, M. Dufaux, Y. Ben Taarit and C. Naccache, in Proc. 6th Int. Zeolite Conf, ed. D. Olson and A. Bisio (Butterworths, London, 1984), p. 368. 15 L. Tebassi, A. Sayari, A. Ghorbel, M. Dufaux and C. Naccache, J . Mol. Cataf., 1984, 25, 297. 16 M. Primet, J. C. Vedrine and C. Naccache, J . Mol. Catal., 1978, 4, 411. 17 M. Primet, J . Chem. Soc., Faraday Trans. 1, 1978, 74, 2570. 18 S. M. Kuznicki and E. M. Eyring, J . Catal., 1980, 65, 227. 19 T. Iizuka and J. H. Lunsford, J. Mol. Catal., 1980, 8, 391. 20 R. D. Shannon, J.C. Vedrine, C. Naccache and F. Lefebvre, J . Catal., 1984, 88, 431. 21 H. Van Brabant, R. A. Schoonheydt and J. Peigrims, Stud. Surf. Sci. Catal., 1982, 12, 61. 22 M. Kawai, M. Uda and M. Ichikawa, J . Phys. Chem., 1985, 89, 1654. 23 F. G. A. Van den Berg, J. H. E. Glezer and W. M. H. Sachtler, J . Cataf., 1985, 93, 340. 24 G. van der Lee, B. Schuller, H. Prost, T. L. F. Favre and V. Ponec, J. Catal., 1986, 98, 522. 25 E. K. Poels, P. J. Mangnus, J. v. Walzen and V. Ponec, in 8th Int. Congr. Catal. (Verlag Chemie, 26 T. Imanaka, K. Kaneda, S. Teranishi and M. Terasawa, in Proc. 6th Znt. Congr. Catal. (The Chemical 27 T. R. Felthouse, Prog. Znorg. Chem., 1982, 29, 73. 28 H. C. Yao and M. Shelef, Stud. Surf Sci. Catal., 1981, 7A, 329. 29 V. D. Atanasova, V. A. Shvets and V.B. Kazanskii, Kinet. Katal., 1977, 18, 628. 30 C. Naccache, Y. Ben Taarit and M. Boudart, ACS Symp. Ser., 1977, 40, 156. 31 D. Goldfarb and L. Kevan, J. Phys. Chem., 1986, 90, 264; 2135. 32 A. Sayari, J. R. Morton and K. F. Preston, to be submitted. 33 A. Sayari, J. R. Morton and K. F. Preston, J . Phys. Chem., 1987, 91, 899. Verlag, Berlin, 1984), vol. 5 , p. 119. London, 1983), vol. 6, p. 196. Weinheim, 1984), vol. 2, p. 59. Society, London, 1977), vol. 2, p. 509. 15 FAR I430 Paramagnetic Rhodium Species in Zeolites 34 A. Sayari, J. R. Morton and K. F. Preston, to be submitted. 35 Y. Ben Taarit, J. C. Vedrine, J. F. Dutel and C. Naccache, J. Magn. Reson., 1978, 31, 251. 36 C. Naccache, J. F. Dutel and V. M. Mathieu, A&. Chem. Ser., 1973, 121, 266.37 J. Michalik, M. Narayana and L. Kevan, J. Phys. Chem., 1985,89,4553. 38 J. Michalik, M. Heming and L. Kevan, J . Phys. Chem., 1986, 90, 2132. 39 E. Garbowski, M. Primet and V. M. Mathieu, ACS Symp. Ser., 1977,40, 281. 40 J. Michalik, M. Narayana and L. Kevan, J. Phys. Chem., 1984,88, 5236. 41 P. A. Jacobs, W. de Wilde, R. A. Shoonheydt, J. B. Uytterhoeven and H. Beyer, J. Chem. SOC., 42 P. Gallezot, Catal. Rev. Sci. Eng., 1979, 20, 121, and references therein. 43 M. Primet and Y. Ben Taarit, J. Phys. Chem., 1977,81, 1317. 44 P. Gallezot, in Catalysis, Science and Technology, ed. J. R. Anderson and M. Boudart (Springer Verlag, 45 E. D. Garbowski, C. Mirodatos, M. Primet and V. M. Mathieu, J. Phys. Chem., 1983, 87, 303. 46 J. R. Pearce, W. Mortier and J.B. Uytterhoeven, J. Chem. SOC., Faraday Trans. I , 1979, 75, 1395. 47 K. W. Blazey and F. Levy, Solid State Commun., 1986, 59, 335. 48 A. Raizman, J. T. Suss and S. Szapiro, Phys. Lett., 1970, 32A, 30. 49 A. Abragam and B. Bleaney, in Electron Paramagnetic Resonance of Transition Ions (Clarendon Press, 50 H. F. J. van’t Blik, J. B. A. D. van Zon, T. Huizinga, J. C. Vis, D. C. Koningsberger and R. Prins, 51 T. Beringhelli, A. Gervasini, F. Morazzoni, D. Stramolo, S. Martinengo and L. Zanderighi, J. Chem. 52 A. Gervasini, F. Morazzoni, D. Strumolo, F. Pinna, G. Strukul and L. Zanderighi, J. Chem. SOC., 53 V . E. Shubin, V. A. Shvets and V. B. Kazanskii, Kinet. Katal., 1978, 19, 1026. 54 M. Valigi, D. Gazzoli and D. Cordishi, J. Mater. Sci., 1982, 17, 1277. 55 S. J.DeCanio, J. B. Miller, J. B. Michel and C. Dybowski, J. Phys. Chem., 1983, 87, 4619. 56 J. R. Shock and M. T. Rogers, J. Chem. Phys., 1975, 62, 2640. 57 E. Moran Miguelez, M. A. Alario Franco and J. Soria, J. Solid State Chem., 1983, 46, 156. 58 J. T. Suss, A. Raizman, S. Szapiro and W. Low, J. Magn. Reson., 1972, 6, 438. 59 Z. Luz, A. Raizman and J. T. Suss, Solid State Commun., 1977, 21, 849. 60 M. D. Sastry, K. Savitri and B. D. Joshi, J. Chem. Phys., 1980,73, 5568. 61 R. S. Eachus and R. E. Graves, J. Chem. Phys., 1974,61,2860; 1973, 59, 2160. 62 W. John and W. Windsch, Phys. Stat. Solid., 1973, B55, K39. 63 M. G. Townsend, J. Chem. Phys., 1964, 41, 3149. 64 R. P. A. Muniz, N. V. Vugman and J. Danon, J. Chern. Phys., 1971, 54, 1284. 65 G. C. Abell and R. C.Bowman Jr, J. Chem. Phys., 1979, 70, 2611. 66 J. Wilkens, D. P. Deraag and J. N. Helle, Phys. Lett., 1965, 19, 178. 67 B. R. James and D. V. Stynes, J. Am. Chem. SOC., 1972, 94, 6225. 68 M. Hoshino, K. Yasufuku, S. Konishi and M. Imamura, Inorg. Chem., 1984, 23, 1982. 69 (a) B. R. James, F. T. T. Ng and E. Ochiai, Can. J. Chem., 1972, 50, 590; (b) B. R. James and G. 70 H. J. Keller and H. Wawersik, J. Organomer. Chem., 1967, 8, 185. 71 A. H. Maki, N. Edelstein, A. Davison and R. H. Holm, J. Am. Chem. SOC., 1964,86,4580. 72 E. Billig, S. I. Ehupack, J. H. Waters, R. Williams and H. B. Gray, J. Am. Chem. SOC., 1964, 86, 73 T. Kawamura, H. Katayama and T. Yamabe, Chem. Phys. Lett., 1986, 130, 20. 74 T. Kawamura, K. Fukamachi, T. Sowa, S. Hayashida and T. Yonezawa, J. Am. Chem. SOC., 1981,103, 75 T. Kawamura, K. Fukamachi and S. Hayashida, J. Chem. SOC., Chem. Commun., 1979, 945. 76 G. W. Eastland and M. C. R. Symons, J. Chem. SOC., Dalton Trans., 1984, 2193. 77 D. 0. K. Fjeldsted and S. R. Stobart, J. Chem. SOC., Chem. Commun., 1985, 908. 78 N. G. Connelly, C. J. Finn, M. J. Freeman, A. G. Orpen and J. Sterling, J. Chem. SOC., Chem. 79 M. Y. Chavan, T. P. Zhu, X. Q. Lin, M. Q. Ahsan, J. L. Bear and K. M . Kadish, Inorg. Chem., 1984, 80 J. A. Osborn, F. H. Jardine, J. F. Young and G. Wilkinson, J. Chem. SOC. A, 1966, 171 I. 81 M. C. Baird, Inorg. Chim. Acta, 1971, 5, 46. 82 F. Pruchnik, Inorg. Nucl. Chem. Lett., 1974, 10, 661. 83 G. Pneumatikakis and P. Psaroulis, Inorg. Chim. Acta, 1980, 46, 97. 84 H. L. M. Van Gaal, J. M. J. Verlaak and T. Posno, Inorg. Chim. Acta, 1977, 23,43. 85 G. N. George, S. I. Klein and J. F. Nixon, Chem. Phys. Lett., 1984, 108, 627. 86 G. Zotti, S. Zecchin and G. Pilloni, J. Electroanal. Chem., 1984, 175, 241. Faraday Trans. I , 1976, 72, 1221. Berlin, 1984), vol. 5, p. 221, and references therein. Oxford, 1979), chap. 8. J. Am. Chem. Soc., 1985, 107, 3139. SOC., Faraday Trans. I , 1984, 80, 1479. Faraday Trans. I , 1986, 82, 1795. Rosenberg, Coord. Chem. Rev., 1975, 16, 153. 926. 364. Commun., 1984, 1025. 23,4538.A . Sayari, J . R. Morton and K. F. Preston 43 1 87 H. Caldararu, M. K. DeArmond, K. W. Hanck and V. E. Sahini, J. Am. Chem. SOC., 1976, 98, 88 K. Kaneda, T. Terasawa, T. Imanaka and S. Teranishi, Chem. Lett., 1976, 995. 89 F. Pinna, M. Bonivento, G. Strukul, M. Graziani, E. Cernia and N. Palladino, J. Mol. Catal., 1976, 1, 90 G. Strukul, M. Bonivento, M. Graziani, E. Cernia and N. Palladino, Inorg. Chim. Acra, 1975, 12, 91 T. Beringhelli, A. Gervasini, F. Morazzoni, F. Pinna and G. Strukul, J. Catal., 1984, 88, 313. 92 D. W. Breck, in Zeolite Molecular Sieves (J. Wiley, New York, 1974), chap. 8. 93 A. Sayari, J. R. Morton and K. F. Preston, to be submitted. 94 D. Exner, N. I. Jaeger, K. Moller, R. Nowak, H. Schrubbers, G. Schulz-Ekloff and P. Ryder, Stud. 95 W. J. Reagan, A. W. Chester and G. T. Kerr, J. Catal., 1981, 69, 89. 96 J. Texter, R. Kellerman and T. Gonsiorowski, J. Phys. Chem., 1986, 90, 2118, and references 4455. 309. 15. Surf. Sci. Catal., 1982, 12, 205. therein. Paper 71358; Received 24th February, 1987 15-2

ISSN:0300-9599    

DOI:10.1039/F19888400413

出版商:RSC    

年代:1988

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. I , 1988, 84(2), 433-439 Raman Spectra of Aniline adsorbed on an Ag Electrode in Acidic Solutions Hitoshi Shindo and Chizuko Nishihara National Chemical Laboratory for Industry, Tsukuba Research Center, Yatabe, Ibaraki 305, Japan The adsorption of aniline on an Ag electrode in acidic aqueous solutions has been studied by surface-enhanced Raman scattering. Adsorption, both as neutral and protonated forms, was observed to depend upon the electrode potential. The choice of halide ions used as the supporting electrolyte as well as the pH of the solution affected the stability of the protonated form on the surface. The geometry of adsorption was also studied by the change in intensities of the Raman bands upon adsorption. It was concluded that the anilinium ion is adsorbed Coulombically in at least two different geometries on halide ions which remain on the electrode surface.Surface-enhanced Raman scattering (SERS) has been widely applied to the studies of molecular structures of solid/solution interfaces as well as solid/gas interfaces.' This technique will greatly help electrochemists to draw realistic pictures of molecules and atoms on electrode surfaces. We have employed SERS in studying the structures of adsorbates on an Ag electrode during the reduction of nitrobenzene in neutral and alkaline aqueous solutions. 2-5 The Raman spectrum of weakly adsorbed aniline was observed in the more cathodic potential range, while at more anodic potentials the electrode was covered with other adsorbates such as trans-azobenzene.2 The geometry of adsorption of aniline was studied by the change in intensities of Raman bands of each symmetry species upon adsorption.Using SER electromagnetic enhancement factors derived by Creighton6, and assuming approximately C,, symmetry for the adsorbate, it was concluded that aniline is adsorbed face-on on the ele~trode.~ In the present work adsorption of aniline was studied in acidic conditions, where the molecule takes a protonated form in the solution phase. Experimental Sample solutions were prepared by dissolving aniline (0.005 or 0.01 mol dm-3) in de- ionised and distilled water and HX(X = C1, Br or I) was added. The solution was purged of dissolved oxygen by bubbling with pure nitrogen. Aniline (Nakarai Chemicals, SP grade) was purified by distillation.The following chemicals were used as purchased : HC1 (Kokusan Chemical Works, GR, 35 YO); HBr (Nakarai Chemicals, GR, 47 YO); HI (Nakarai Chemicals, GR, 55 YO) ; D,O (Wako Pure Chemical Industries, 99.75 YO D) ; C,D,NH2 (Aldrich, Gold Label, 99 % D) ; DCl (Aldrich, Gold Label, 99 % D); DI (Aldrich, Gold Label, 99 YO D). Anilinium halides (powder) were synthesized from chemicals described above by a conventional method. The design of the electrochemical cell used for Raman measurements was based on that introduced by Watanabe.g As a modification the counter-electrode, a platinum wire, was held in a compartment separated from the bulk solution by a glass frit. Mixing of by-products at the counter-electrode with the sample solution was thus avoided.For example, oxidation of halide ions and polymerization of aniline occurs at the counter- electrode. Experiments with D20 solutions were performed in a smaller cell with a design 43 3434 Aniline adsorbed on an Ag Electrode used by Fleischmann et a1." The counter-electrode was separated in this case, too. A saturated calomel electrode (SCE) was used as the reference electrode. The electrode potential is quoted us. SCE throughout this paper. Origin and pretreatments of the Ag working electrode (99.99 % purity) were described previ~usly.~ The oxidation and reduction cycle for the activation of SERS was made by potential steps. Formation of silver halides was performed at + 0.2,O.O and - 0.2 V (vs. SCE) for solutions with C1-, Br- and I-, respectively.An argon-ion laser (Coherent, CR-8) was used for the excitation of Raman scattering. In most cases the 514.5 nm line was used at 50 mW. The scattered light was observed in the 90" direction and was analysed with a double-dispersion Raman spectrometer (JASCO, R-800). Resolution of 5 cm-' was usually used to obtain the spectra shown in this paper. However, all the frequencies given in the figures were obtained by separate measurements with a resolution of 2cm-'. Other details are given in our previous Results and Discussion Adsorption of Aniline In the first experiment a 0.01 mol dm-3 aqueous solution of aniline with 0.1 mol dm-3 HCl was used. The spectrum shown in fig. 1(a) was observed when the electrode potential was set to -0.3 V immediately after the oxidation of Ag at 0.2 V for 3 s.Three sharp and strong bands were observed at 1027, 1006 and 793 cm-l. They are easily assigned to v18a, vI2 and v, modes of anilinium ion adsorbed on the electrode. Throughout this paper Wilson notation'' is used for the numbering of vibrational modes of the phenyl ring. For the anilinium ion in the solution phase these modes are observed at 1030, 1008 and 797cm-', respectively. The frequency shifts upon adsorption are rather small. Contribution by the Raman bands of the ion in the solution is almost negligible in fig. 1 (a), since they are weaker than those of the adsorbed ion by more than one order of magnitude. The broad band at around 840 cm-' is assigned to the vlOa mode. This mode is weakly observed at 837cm-' in the Raman spectrum of C,H, NHlCl- in the solid phase, although the band is too weak to be detected in the solution phase.The enhancement and broadening of the vlOa band was observed previously in the case of adsorbed aniline.3 When the electrode potential was stepped to a more cathodic range, deprotonation of the ion occurred. The two bands in fig. 1 (b) at 1023 and 996 cm-' are assigned to v~~~ and v,, bands of adsorbed aniline in the neutral form. The broad vlOa band is observed at 822 cm-l. The deprotonation occurs at ca. -0.5 V. At this potential, coadsorption of aniline in neutral and protonated forms is observed. The Raman band of adsorbed C1- is observed at ca. 240 cm-'. Fig. 2 shows the change in intensity of the Raman band with the electrode potential in a blank experiment without aniline.The curves (a)-(e) were obtained in a series while the voltage was changed stepwise from -0.2 V towards the more cathodic range. A marked decrease in intensity occurred at ca. -0.5 V. At -0.6 V the peak nearly disappeared. The result is in accordance with the results of Wetzel et aZ.12 in neutral solution. In our experiment addition of aniline to the HCl solution did not change the behaviour of the Raman band of c1-. The fact that the decrease in intensity of Raman bands of the anilinium ion and C1- occurs in the same potential range suggests a relation between the two ionic species. As reported by Wetzel et aZ.,12 I- is adsorbed on Ag more strongly than C1- and Br- in the more cathodic potential range. When we used 0.1 mol dmW3 HI in place of HCl the protooated form of aniline remained the main adsorbate down to -0.9 V.The Raman band of adsorbed I- was also observed at this voltage. It is very likely that the anilinium ion is absorbed on the Ag electrode by Coulombic interaction with halide ions which remain on the surface, depending upon the potential.H. Shindo and C. Nishihara 435 n I 1 I 1 1 1000 900 800 Raman shiftlcm-' Fig. 1. Raman spectra of aniline adsorbed in acidic conditions. Aniline 0.005 mol dm-3; HCl 0.1 mol dm-3, (a) at -0.3 V us. SCE adsorption of C,H,NHi was observed; (b) at -0.6 V adsorption of C,H,NH, was observed. 300 200 Raman shiftlcm-' Fig. 2. Raman spectra of C1- adsorbed on Ag at various potentials. The spectra were obtained in the order ( a x e ) . (a) -0.2 V, (b) -0.3 V, (c) -0.4 V, ( d ) -0.5 V, (e) -0.6 V us.SCE.436 Aniline adsorbed on an Ag Electrode " - I 1 1 1600 1400 1200 1000 800 600 400 Raman shift/cm-' Fig. 3. Raman spectra of anilinium ion in solid and adsorbed states. (a) C,H,NH,Br (solid); (b) C,H,NHi adsorbed at -0.4 V from a 0.005 mol dm-, aqueous solution of aniline with 0.1 mol dm-, HBr; (c) C,H,NHi adsorbed at -0.7 V in a 0.005 mol dm-, solution of aniline with 0.1 mol dm-, HI. The pH of the solution has a large effect on the stability of the anilinium ion on the surface. In the case of 0.1 mol dm-3 HC1 solution, the coadsorption of aniline in the two forms was observed at -0.5 V. When 0.01 mol dm-3 HC1 was used, while keeping [Cl-] = 0.1 mol dm-3 by addition of KC1, the coadsorption was observed at -0.4 V.On the other hand, only the ionised form was observed even at -0.8 V for a solution with 1.0 mol dm-3 HCl. Raman Spectra of the adsorbed Anilinium Ion In fig. 3 (b) and (c) are shown the Raman spectra of anilinium ions adsorbed on Ag from 0.005 mol dm-3 solutions of aniline in 0.1 mol dm-3 HBr and HI, respectively. In order to study the structure of the adsorbed ion in detail, the vibrational spectra of anilinium halides in solid and solution phases were also studied. As the Raman bands of out-of- plane modes are very weak in the solution spectra, the Raman spectrum of anilinium bromide in powder form is shown in fig. 3(a). The assignments in fig. 3(a) were made by studying Raman and infrared spectra of C,H5NH3X (X = C1, Br, I), C6H5ND3X and C6D,NH3X in solid and solution phases.H.Shindo and C. Nishihara 437 Raman shift/cm-' Fig. 4. Raman spectra of C,H,NDi adsorbed on an Ag electrode observed at -0.9 V in a 0.005 mol dm-3 solution of aniline in D,O with 0.1 mol dm-3 DI. Virtually all the H atoms of the amino group are replaced with D atoms. (a) 1750-400 cm-l; (b) 2600-2050 cm-'. - - tn 0- a w > 7. vv N Otn N o , . . 1 I 1 1 1 1600 1200 800 400 Raman shift/cm-' Fig. 5. Raman spectrum of C,D,NHi adsorbed on an Ag electrode observed at -0.9 V in a 0.01 mol dm-3 aqueous solution of C,D,NH, with 0.1 rnol dm-3 HI. Ab initio calculations of vibrational frequencies were also performed. The details of the assignment will be reported in a separate paper and only the results are shown here. Most Raman bands in fig. 3(b) and ( c ) are easily assigned by comparing the spectra with fig.3(a). In order to confirm the assignment, experiments with deuterated compounds were performed also. The Raman spectra of adsorbed C,H,NDi are shown in fig. 4. The spectra were observed at - 0.9 V for a 0.005 mol dm-, solution of aniline in D,O with 0.1 mol dm-, DI. Virtually all H atoms of the amino group were quickly replaced with D atoms. The Raman band of the N-D stretching vibration is observed at ca. 2150 cm-l. A corresponding band was observed for adsorbed C,H,NHi in the experiment in H,O. However, we cannot conclude that this is the N-H stretching band of the anilinium ion, since the observation of C-H stretching bands of contaminants has been reported', in the same frequency range.On the other hand, the broad band in fig. 4 is safely assigned to the N-D stretching band, since it is improbable that the C-H bonds of contaminants are easily replaced with C-D. The N-D stretching band is observed in the 2350-2100 cm-' range for C,H,ND,I powder. The NH, bending band (6,) is observed at ca. 1540 cm-l in fig. 3(b), while the ND,438 Aniline ahorbed on an Ag Electrode bending mode (S,?) is observed at 1125 cm-l in fig. 4(a). The C-N stretching mode is clearly observed in both spectra. Compared to the band at 1629 cm-l in fig. 3(a) the vst, mode of the adsorbed ion at 1653 and 1660 cm-l in fig. 3 (b) and (c), respectively, seems to be too high in energy. Chen et aL1* reported that the bending mode of adsorbed water is sometimes observed at a frequency as high as 1640 cm-l when halide ions are used in large concentrations.However, the bands in fig. 3(b) and (c) do not belong to water, since a similar band is observed at 1650 cm-l, as shown in fig. 4(a) in a D,O solution. The bands do not come from contaminants either, since they are shifted to a lower frequency, as indicated in fig. 5 when C6D,NHi was used with 0.1 mol dm-3 HI. The band certainly belongs to the phenyl ring. The reason for the large shift to higher frequencies upon adsorption is still under investigation. All other assignments in fig. 3-5 are in good agreement with each other. Geometry of Adsorption In the previous r e p ~ r t , ~ geometry of adsorption of aniline was discussed by assuming that the molecule has nearly C,, symmetry. The assumption is more readily applicable to the anilinium ion since the -NHi group rotates freely around the C-N axis just as in the case of the methyl group of toluene.Here, again, the (T, plane was chosen to be perpendicular to the plane of the phenyl ring. Thus, phenyl in-plane vibrations belong to a, or b, symmetry species, while out-of-plane modes belong to b, or a, species. Creighton' calculated the surface Raman enhancement factors for molecules adsorbed at the surface of a metal sphere. He also suggested that the result applies qualitatively to adsorption on roughened electrode and aggregated colloid surfaces. According to his calculations for a C,, molecule, the b, modes are least enhanced when the molecule lies flatly on the surface, while the a, modes are least enhanced when the molecule stands up on the metal with the molecular axis perpendicular to the surface. The former is the case for aniline adsorbed on Ag in the neutral and alkaline condition^.^ As shown in fig.6(a), the molecule most probably adsorbs with n-electrons interacting with the metal atoms on the surface. The results in acidic conditions showed a marked difference. In fig. 3 (c), b, bands such as vet,, v,, and v l g t , are clearly visible, while a, modes such as vlea and vtOa are relatively weak. In neutral conditions the results were opposite. It is suggested that the anilinium ion stands up rather than lies flatly on the electrode surface. The result shown in fig. 3 (b) is an intermediate case, where a, modes as well as b, modes have considerable intensities.It is noted that b, Raman bands such as v16, and v,,, are observed neither in fig. 3(b) nor in (c). They were observed for neutral aniline lying flatly on the s~rface.~ According to the calculation by Creighton,' the b, modes are least enhanced if the molecule lies side-on on the electrode. The side-on adsorption does not seem very likely, but it is not at all impossible considering an interaction between an H atom in the ortho position of the phenyl ring and a halide ion on the surface. However, the b, Raman bands of the anilinium ion are weak in the first place, as shown in fig. 3(a), and it is difficult to evaluate the surface Raman enhancement factors upon adsorption. As is seen in fig. 3(a), the a, Raman bands generally have smaller intensities than the b, modes.The fact that b, Raman bands are stronger than a, modes in fig. 3(c) does not necessarily mean that 'the enhancement factor' for b, modes is larger than that for a, modes. However, it is safe to say that the anilinium ion takes at least two different forms of adsorption and that the case of fig. 3(b) is closer to the face-on adsorption. Let us now consider the reason why the adsorption geometry differs with the choice of halide ions. As discussed above, I- is adsorbed more strongly than Br- and CI- on the surface of Ag.12 If compared at the same potential, the surface concentration of halide ions is largest when I- is used. When a halide ion has a large concentration on theH . Shindo and C . Nishihara 439 Fig. 6. Proposed geometries of adsorption of aniline: (a) face-on adsorption in the neutral form; ( c ) end-on adsorption in the protonated form making bonds with three halide ions on the surface ; (b) an intermediate case in which the anilinium ion makes bonds with two (or one) halide ions.surface, it is probable that the anilinium ion makes bonds with up to three halide ions as demonstrated in fig. 6(b) and (c). Neutron diffraction analyses of the structures of crystalline anilinium bromidel5- l6 support the structure shown in fig. 6 (c). In the case of fig. 6(b) the angle of the molecular plane to the electrode surface is not fixed because of the rotation around the C-N axis. On the other hand, the angle is fixed at 90" in the case of fig. 6(c). In this case the u2 Raman bands are least enhanced.We propose that the adsorption geometry of the anilinium ion is close to fig. 6 ( c ) when the surface concentration of the halide ion is very large. When the potential is swept to a more cathodic range, the halide ion desorbs and the geometry of adsorption of aniline becomes closer to face-on adsorption. In the most cathodic potential range no halide ion remains on the surface, and aniline does not absorb stably in the protonated form. Face- on adsorption in the neutral form prevails instead. The presence of adsorbates on the surface very often has a strong effect on the selectivity of electrode reacti~ns,~' although the mechanisms are not clear in most cases. We have studied the adsorption of aniline in relation to the reduction of nitrobenzene. It is generally said that nitrobenzene is reduced to phenylhydroxylamine (four-electron reduction) in alkaline solutions, while six-electron reduction to aniline prevails in acidic solutions." It is very likely that the adsorbed anilinium ion is working as an effective proton donor on the surface, facilitating the six-electron reduction.References 1 See, e.g. Spectroscopic Studies of Adsorbates on Solid Surfaces, Surf. Sci., 1985, 158. 2 H. Shindo and C. Nishihara, Surf. Sci., 1985, 158, 393. 3 H. Shindo, J. Chem. SOC., Faraday Trans. I , 1982, 82, 45. 4 C. Nishihara and H. Shindo, J. Electroanal. Chem., 1986, 202, 231. 5 H. Shindo, in Recent Advances in Electro-organic Synthesis, Proc. 1st Int. Symp. Electro-organic 6 J. A. Creighton, Surf. Sci., 1983, 124, 209. 7 J. A. Creighton, Surf. Sci., 1985, 158, 211. 8 T. Sakai and H. Terauchi, Acta Crystallogr., Sect. B, 1981, 37, 2101. 9 T. Watanabe, J. Metal Finish. SOC. Jpn, 1982, 33, 96. Synthesis, Kurashiki, 1986, ed. S. Torii (Kodansha-Elsevier, Tokyo-Amsterdam, 1987), p. 405. 10 M. Fleischmann, P. J. Hendra and A. J. McQuillan, Chem. Phys. Lett., 1974, 26, 163. 11 E. B. Wilson, Phys. Rev., 1934, 45, 706. 12 H. Wetzel, H. Gerischer and B. Pettinger, Chem. Phys. Lett., 1981, 78, 392. 13 B. Pettinger, M. R. Philpott, J. G. Gordon 11, J. Chem. Phys., 1981, 74, 934. 14 T. T. Chen, J. F. Owen, R. K. Chang and B. L. Laube, Chem. Phys. Lett., 1982, 89, 356. 15 G. Fecher, A. Weiss, W. Joswig and H. Fuess, Z. Naturforsch., Teil A , 1981, 36, 956. 16 G. Fecher, A. Weiss and G. Heger, Z. Naturforsch., Teil A, 1981, 36, 967. 17 R. Jansson, Chem. Eng. News, 1984, 62 (47), 43. 18 W. Kemura and T. M. Krygowski, in Encyclopedia of Electrochemistry of Elements, ed. A. J. Bard (Dekker, New York, 1979): vol. XIII, chap. 2. Paper 7/369; Received 26th February, 1987

ISSN:0300-9599    

DOI:10.1039/F19888400433

出版商:RSC    

年代:1988

数据来源: RSC