摘要:

Contents 3663 3669 3675 3683 3693 370 1 3709 3717 3725 3737 Normal and Abnormal Electron Spin Resonance Spectra of Low-spin Cobalt(r1) IN,]-Macrocyclic Complexes. A Means of Breaking the Co-C Bond in B12 Co-enzyme M. Green, J. Daniels and L. M. Engelhardt The Interaction between Superoxide Dismutase and Doxorubicin. An Electron Spin Resonance Approach V. Malatesta, F. Morazzoni, L. Pellicciari-Bollini and R. Scotti Biomolecular Dynamics and Electron Spin Resonance Spectra of Copper Complexes of Antitumour Agents in Solution. Part 2.-Rifamycins R. Basosi, R. Pogni, E. Tiezzi, W. E. Antholine and L. C. Moscinsky An Electron Spin Resonance Investigation of the Nature of the Complexes formed between Copper(I1) and Glycylhistidine D. B. McPhail and B. A. Goodman A Vibronic Coupling Approach for the Interpretation of the g-Value Temperature Dependence in Type-I Copper Proteins M.Bacci and S. Cannistr aro The Electron Spin Resonance Spectrum of Al[C,H,] in Hydrocarbon Matrices J. A. Howard, B. Mile, J. S. Tse and H. Morris N; and (CN); Spin-Lattice Relaxation in KCN Crystals H. J. Kalinowski and L. C. Scavarda do Carmo Single-crystal Proton ENDOR of the SO, Centre in y-Irradiated Sulphamic Acid N. M. Atherton, C. Oliva, E. J. Oliver and D. M. Wylie Single-crystal Electron Spin Resonance Studies on Radiation-produced Species in Ice 1,. Part 1.-The 0- Radicals Single-crystal Electron Spin Resonance Studies on Radiation-produced Species in Ice I,. Part 2.-The HO, Radicals J. Bednarek and A. Plonka J. Bednarek and A. PlonkaContents 3663 3669 3675 3683 3693 370 1 3709 3717 3725 3737 Normal and Abnormal Electron Spin Resonance Spectra of Low-spin Cobalt(r1) IN,]-Macrocyclic Complexes.A Means of Breaking the Co-C Bond in B12 Co-enzyme M. Green, J. Daniels and L. M. Engelhardt The Interaction between Superoxide Dismutase and Doxorubicin. An Electron Spin Resonance Approach V. Malatesta, F. Morazzoni, L. Pellicciari-Bollini and R. Scotti Biomolecular Dynamics and Electron Spin Resonance Spectra of Copper Complexes of Antitumour Agents in Solution. Part 2.-Rifamycins R. Basosi, R. Pogni, E. Tiezzi, W. E. Antholine and L. C. Moscinsky An Electron Spin Resonance Investigation of the Nature of the Complexes formed between Copper(I1) and Glycylhistidine D. B. McPhail and B. A. Goodman A Vibronic Coupling Approach for the Interpretation of the g-Value Temperature Dependence in Type-I Copper Proteins M. Bacci and S. Cannistr aro The Electron Spin Resonance Spectrum of Al[C,H,] in Hydrocarbon Matrices J. A. Howard, B. Mile, J. S. Tse and H. Morris N; and (CN); Spin-Lattice Relaxation in KCN Crystals H. J. Kalinowski and L. C. Scavarda do Carmo Single-crystal Proton ENDOR of the SO, Centre in y-Irradiated Sulphamic Acid N. M. Atherton, C. Oliva, E. J. Oliver and D. M. Wylie Single-crystal Electron Spin Resonance Studies on Radiation-produced Species in Ice 1,. Part 1.-The 0- Radicals Single-crystal Electron Spin Resonance Studies on Radiation-produced Species in Ice I,. Part 2.-The HO, Radicals J. Bednarek and A. Plonka J. Bednarek and A. Plonka

ISSN:0300-9599    

DOI:10.1039/F198783FX009

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

Electrochemistry Group Workshop on Electrochemical Techniques and Instruments To be held at the University of Warwick on 6-7 January 1988 Further information from Dr P. N. Bartlett, Department of Chemistry, University of Warwick, Coventry CV4 7AL Surface Reactivity and Catalysis Group with the Process Technology Group and the Institute of Chemical Engineers Opportunities for Innovation in the Application of Catalysis To be held at Queen Mary College, London on 6-7 January 1988 Further information from Professor J. Pritchard, Queen Mary College, London Division with the Institute of Mathematics and its Applications Mathematical Modelling of Semiconductor Devices and Processes To be held at the University of Loughborough on 7-8 January 1988 Further information from the Institute of Mathematics, Maitland House, Warrior Square, Southend-on-Sea SS1 2JY Division London Symposium: Modern Electrochemical Systems To be held at Imperial College, London on 12 January 1988 Further information from Mrs Y.A. Fish, Royal Society of Chemistry, Burlington House, London W1V OBN 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 l w 1 1 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 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 40E Annual Congress: Division with Electrochemistry Group Solid State Materials in Electrochemistry To be held at the University of Kent, Canterbury on 12-15 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 GLI 3 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 302 (xiii)Electrochemistry Group Workshop on Electrochemical Techniques and Instruments To be held at the University of Warwick on 6-7 January 1988 Further information from Dr P.N. Bartlett, Department of Chemistry, University of Warwick, Coventry CV4 7AL Surface Reactivity and Catalysis Group with the Process Technology Group and the Institute of Chemical Engineers Opportunities for Innovation in the Application of Catalysis To be held at Queen Mary College, London on 6-7 January 1988 Further information from Professor J. Pritchard, Queen Mary College, London Division with the Institute of Mathematics and its Applications Mathematical Modelling of Semiconductor Devices and Processes To be held at the University of Loughborough on 7-8 January 1988 Further information from the Institute of Mathematics, Maitland House, Warrior Square, Southend-on-Sea SS1 2JY Division London Symposium: Modern Electrochemical Systems To be held at Imperial College, London on 12 January 1988 Further information from Mrs Y.A. Fish, Royal Society of Chemistry, Burlington House, London W1V OBN 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 l w 1 1 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 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 40E Annual Congress: Division with Electrochemistry Group Solid State Materials in Electrochemistry To be held at the University of Kent, Canterbury on 12-15 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 GLI 3 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 302 (xiii)

ISSN:0300-9599    

DOI:10.1039/F198783BX011

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ISSN 0300-9599 JCFTAR 83( 3) 571 -942 (1 987) 57 1 585 59 I 615 627 63 5 645 657 665 675 687 697 705 72 I 733 743 75 1 759 JOURNAL OF THE CHEMICAL SOCIETY Faraday Transactions I Physical Chemistry in Condensed Phases CONTENTS Hydrocarbon Formation from Methylating Agents over the Zeolite Catalyst ZSM-5. Comments on the Mechanism of Carbon-Carbon Bond and Methane Formation The Liquid-Vapour Transition in Monolayers of n-Pentadecanoic Acid at the Air/Water Interface N. R. Pallas and B. A. Pethica Surface Potential Measurements in Pentanol-Sodium Dodecyl Sulphate Micelles G. V. Hartland, F. Grieser and L. R. White Identification of the Space Group and Detection of Cationic Ordering in Iron Antimonate using Conventional and Convergent-beam Electron Diffraction F. J. Berry, J.G. Holden and M. H. Loretto Kinetics of N,O Decomposition on the Surface of y-Al,O, doped with Sodium Ions C. Kordulis, L. Vordonis, A. Lycourghiotis and P. Pomonis Standard Gibbs Free Energies of Transfer of NaCl and KCl from Water to Mixtures of the Four Isomers of Butyl Alcohol with Water. The Use of Ion-selective Electrodes to Study the Thermodynamics of Solutions D-Y. Chu, Q. Zhang and R-L. Liu Acrylonitrile Polymerization from Aqueous Solution. The Role of Surfactants E. Elbing, W. K. Tan, C. J. Lyons, B. A. W. Coller and I. R. Wilson Acrylonitrile Polymerization from Aqueous Solution. The Role of Particle Area E. Elbing, S. J. McCarthy, B. A. W. Coller, I. R. Wilson and D. F. Sangster Physicochemical Characterization of ZnO/Al,O, and ZnO-MoO,/Al,O, Catalysts A.Maezawa, Y. Okamoto and T. Imanaka Fourier-transform Infrared Investigation of Structures of Vanadium Oxide on Various Supports H. Miyata, K. Fujii, T. Ono, Y. Kubokawa, T. Ono and F. Hatayama Absence of Fast H+-Ion Motion in Aqueous HCl. A Neutron Scattering Study H. Bertagnolli, P. Chieux and H. G. Hertz Investigation of the Mechanism of Photocatalytic Alcohol Dehydrogenation over Pt/TiO, using Poisons and Labelled Ethanol P. Pichat, M-N. Mozzanega and H. Courbon Optical and Spectromagnetical Properties of Phosphate Glasses containing Ruthenium and Titanium Ions S. Pizzini, D. Narducci, D. Daverio, C. M. Mari, F. Morazzoni and A. Gervasini Dynamic Kinetics of CO Oxidation over Magnesium Oxide M. Kobayashi, T. Kanno and Y. Konishi Influence of Lithium on Reduction, Dispersion and Hydrogenation Activity of Nickel on Alumina Catalysts S.Narayanan and K. Uma The Use of Hydride-forming Rare-earth-Cobalt Intermetallic Compounds in the Dehydrogenation of Propan-2-01 H. Imamura, K. Nukui, K. Yamada, S. Tsuchiya and T. Sakai Direct Measurement of the Diffusion Coefficients of Hydrogen Atoms in Six Gases G. Blyth, A. A. Clifford, P. Gray and J. I. Waddicor INDO and CND0/2 Calculations for Substituted Benzene Cations H. Chandra, M. C. R. Symons and A. Hasegawa G. J. Hutchings, F. Gottschalk, M. V. M. Hall and R. Hunter 20 FAR 177 1 779 789 801 813 819 829 84 1 853 865 873 893 905 913 925 937 Contents A Polarimetric, Circular Dichroism and llB Nuclear Magnetic Resonance Study of the Reaction of the Tetrahydroxyborate Ion with Two Chiral 1,3-Diols J.G. Dawber Smectite Molecular Sieves. Part 2.-Expanded Fluorhectorite Sorbents R. M. Barrer and R. J. B. Craven Vibrational Spectra of Potassium 4-Pentenoate and Potassium 5-Hexenoate and Conformational Change on Micellization K. Tsukamoto, K. Ohshima, K. Taga, H. Okabayashi and H. Matsuura Dynamic Properties and Structure of Salt-free Polystyrene Sulphonate Solutions H. Vink Cross-relaxation of Hydrogen Atoms adsorbed on a Molybdena-Silica Catalyst S. Seyedmonir and R. F. Howe Thermodynamics of Mixtures with a Hexane Isomer. Excess Volumes of 1-Chlorohexane with a Hexane Isomer at 298.15 K A. Compostizo, A. C. Colin, R. G. Rubio and M. Diaz Peiia Diaphragm-cell Studies of Diffusion in the Four-component System Hydro- chloric Acid-Sodium Chloride-Sodium Iodide-Water D.G. Leaist The Dielectric Properties of Ethyl and Methyl Carbamates in Aqueous Solution J. B. Bateman and M. Thurai A Fourier-transform Infrared and Catalytic Study of the Evolution of the Surface Acidity of Zirconium Phosphate following Heat Treatment G. Busca, V. Lorenzelli, P. Galli, A. La Ginestra and P. Patron0 Kinetics of Hydrolysis of Phenyl Dichloroacetates in Aqueous Salt Solutions J. B. F. N. Engberts, M. J. Blandamer, J. Burgess, B. Clark and A. W. Hakin The Consolidation of Concentrated Suspensions. Part 1 .-The Theory of Sedimentation R. Buscall and L. R. White An Electron Spin Resonance Study of Single Crystals of X-irradiated L- Ascorbic Acid at Room Temperature. Experimental Results and Semiempirical Calculations J. T. Masiakowski, A. Lund and M. Lindgren Chemisorption and Catalysis by Metal Clusters. Hydrogenation of Carbon Monoxide and Carbon Dioxide catalysed by Supported Ruthenium Clusters derived from Ru,(CO),, and from H,Ru,(CO),, S. D. Jackson, R. B. Moyes, P. B. Wells and R. Whyman Influence of Metal-Support Interactions on the Hydrogenolysis of Methyl- cyclopentane over Supported Rh Catalysts J. B. F. Anderson, R. Burch and J. A. Cairns Oxidation of Crystalline UO, studied using X-Ray Photoelectron Spectroscopy and X-Ray Diffraction Reviews of Books J. C. R. Turner; R. A. Pethrick; B. J. Briscoe; J. S. Higgins; C. F. Wells; H. Vink; N. M. Atherton G. C. Allen, P. A. Tempest and J. W. Tyler

ISSN:0300-9599    

DOI:10.1039/F198783FP033

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

JOURNAL OF THE CHEMICAL SOCIETY 463 473 485 503 51 3 529 539 553 569 Faraday Transactions II, Issue 3,1987 Molecular and Chemical Physics For the benefit of readers of Faraday Transactions I, the contents list of Faraday Transactions 11, Issue 3, is reproduced below. Infrared-Radiofrequency Double-resonance Spectra and Pressure Broadening in the Enantiomers and Racemic Mixture of Gaseous 1,2-Dichloro-2- fluoroethanone D. Hennequin, P. Glorieux, E. Arimondo and M. W. Evans A Perturbational, RAM-like Approach for Adsorption of Non-spherical Mole- cules on Solid Surfaces L. Lajtar, A. Patrykiejew and s. Sokolowski Ab Initio Calculations relevant to the Mechanism of Chemical Carcinogenesis by N-Nitrosamines. Part 3.-Transition Structures in Nitrosamine Formation and Metabolism C.A. Reynolds and C. Thomson Preliminary Theoretical Search for Possible Conformational Signatures in the Valence X-Ray Photoelectron Spectra of Ordered Polyethylene Surfaces J. Delhalle, S. Delhalle and J. Riga Steady-state and Time-resolved Fluorescence Study of Excited-state Proton Transfer in 1 -Aminoalkyl-2-naphthols G. Kohler and P. Wolschann Kinetics of the Reactions of the Hydroxyl Radical with Molecular Chlorine and Bromine R. B. Boodaghians, I. W. Hall and R. P. Wayne Spontaneous Ignition and Thermal Runaway in Closed and Open Systems. Part 2.-Ignition and Extinction in the Adiabatic Continuous-flow Well Stirred Reactor (C.S.T.R.) for Reactions of Order m P. Gray and J. Mullins Energy Transfer in Cs,NaHoo~,,Ero~o,C16 and Assignment of the 516 and 514 Levels of HoCli- P.A. Tanner A Single-crystal Proton ENDOR Study of the C10, Centre in ?-Irradiated Barium Perchlorate at 120 K N. M. Atherton and D. S. Blackford The following papers were accepted for Publication in J . Chem. SOC., Faraday Trans. I during December 1986. 61619 61819 6/ 1072 6/ 1228 6/ 1289 A New Pressure-programmed Volumetric Method of Measuring Adsorption at the Gas-Solid Interface D. I. Hall, V. A. Self and P. A. Sermon Study of the Support Evolution through the Process of Preparation of Rh/Lanthana Catalysts S. Bernal, F. J. Botana, R. Garcia, F. Ramirez and J. M. Rodriguez-Izquierdo Corrosion of Ruthenium Dioxide Hydrate by CeIV Ions and other Oxidants A. Mills, S. Giddings, I. Pate1 and N. McMurray Crystal Structure of Sodium Diheptyl Sulphosuccinate Dihydrate and Com- parison with Phospholipids J.Lucassen and M. G. B. Drew Interfacial Tension Minima in Oil-Water-Surfactant Systems. Effects of Cosurfactants in Systems containing a Single- or Double-chain Surfactant R. Aveyard and B. P. Binks (06/ 1398 Hydrogenation of CO, over Co/Cu/K Catalysts H. Baussart, R. Delobel, M. Le Bras and J. M. Leroy 6/1422 The Molar VOhme of a Large Polymeric Cation J. W. Akitt, J. M. Elders and P. Letellier 6/ 1455 Photoinduced Reactions of Methane with Molybdena supported on Silica W. Hill, B. N. Shelimov and V. B. Kazansky 6/1457 Selective 'H-, l3C- and 'H-lH N.O.E. Studies of Adenosine-Thymidine Inter- action in Solution 6/ 1545 Origin of Boron Mobility over Boron-impregnated ZSM-5. An Infrared Study M. B.Sayed 6/ 1 572 Glutamic Acid-Hydrogen Phosphate Hydrogen Bonds. Proton Polarizability and Proton Transfer as a Function of the Cations Present and of the Degree of Hydration: Infrared Investigations U. Berget and G. Zundel C. Rossi, N. Niccolai and F. Laxhi 6/ 1607 Comments on the Mechanism of MTG/HZSM-5 Conversion M. B. Sayed 6/1608 Effect of Pretreatment Temperature on the Phase of Impregnated Boron. A Spectral Infrared/HRSS I1B-N.M.R. Study of Boron-impregnated ZSM-5 M. B. Sayed 6/ 1617 Adsorption, Decomposition and Surface Reactions of Methyl Chloride on Metal Films of Iron, Nickel, Palladium, Lead, Gold and Copper A.-Karim M. Ali, J. M. Saleh and N. A. Hikmat 6/ 1728 A Kinetic Study of the Self-reaction of Prop-2-ylperoxyl Radicals in Solution using Ultraviolet Absorption Spectroscopy J.E. Bennett 6/ 757 Sodium Ion Exchange on the 1,4,7,10,13, 16-Hexaoxactadecane-Sodium~1~ Cation in Several Solvents. A Sodium-23 Nuclear Magnetic Resonance Study S. F. Lincoln, A. White and A. M. Hounslow 6/ 847 The Inclusion of Tropaeolin 000 No. 2 by Permethylated Q-Cyclodextrin. A Kinetic and Equilibrium Study R. P. Villani, S. F. Lincoln and J. H. Coates 6/ 848 N.M.R. Self-diffusion Studies of Methanol-Water Mixtures sorbed in Pentasil- type Zeolites J. Caro, M. Bulow, J. Richter-Mendau, J. Karger, M. Hunger, D. Freude and L. V. C. Rees 6/ 188 1 Synthesis of Montmorillonite-Viologen Intercalation Compounds and their Photochromic Behaviour H. Miyata, K. Kuroda and C. Kato 6/ 1882 35Cl Nuclear Quadrupole Resonance and Infrared Spectroscopic Studies of Hydrogen Bonding and Proton Transfer in Solid Complexes of Trichloroacetic Acid with various Nitrogen and Oxygen Bases B.Nogaj, B. Brycki, Z. Dega- Szafran, M. Szafran and M. Mackowiak 6/ 1907 Dye-sensitized Photocatalyst for Generation of Oxygen from Aqueous Per- sulphate K. Tannakone, S. Wickramanayake and M. u. Gunasekara 6/ 1968 Oxidation of 2,4-Dibromo-6-nitroaniline in Aqueous Sulphuric Acid Solutions on a Platinum Electrode S. Arias, E. Brillas and J. M. Costa 6/2003 Electron-Donor-Electron-Acceptor Association of Pyrene with 2,4,6-Trinitro- toluene in Solution Determined by lH Nuclear Magnetic Resonance J. A. Chudek, R. Foster and F. Page 6/2034 A Small-angle Neutron-scattering Investigation of Rod-like Micelles aligned by Shear Flow P.G. Cummins, J. B. Hayter, J. Penfold and E. Staples (ii)6/2098 Infrared Study of the Adsorption of Non-ionic Surfactants on Silica Y. Lijour, J. Y. Calves and P. Saumagne 6/2111 An E.S.R. Study of Triplet Radical Pairs in Single Crystals of X-Irradiated L-Ascorbic Acid at 77 K J. T. Masiakowski and A. Lund 6/2144 Kinetic Modeling of Multiple Site Activity and the Kinetics of Inhibition Reactions in the Hydrogenolysis of C,H, on a Nickel Wire Catalyst S. Kristyan and R. B. Timmons 6/2145 Infrared Study of the Desorption of Polycondensed Aromatic Compounds by Non-ionic Surfactants at the Silica-Carbon Tetrachloride Interface Y. Lijour, J. Y. Calves and P. Saumagne 6/2350 Kinetic Models for the Development of Density in Radiographic Film by Visible-light Exposure B.W. Darvell (iii)Cumulative Author Index 1987 Agnel, J-P. L., 225 Alberti, A., 91 Allen, G. C., 925 Anderson, A. B., 463 Anderson, J. B. F., 913 Antholine, W. E., 151 Atherton, N. M., 37, 941 Axelsen, V., 107 Barratt, M. D., 135 Barrer, R. M., 779 Basosi, R., 151 Bastl, Z., 51 1 Bateman, J. B., 841 Battesti, C. M., 225 Becker, K. A., 535 Berclaz, T., 401 Berleur, F., 177 Berry, F. J., 615 Bertagnolli, H., 687 Berthelot, J., 231 Beyer, H. K., 51 1 Bianconi, A., 289 Blandamer, M. J., 559, 865 Blyth, G., 751 Borbkly, G., 51 1 Braquet, P., 177 Brazdil, J. F., 463 Briscoe, B. J., 938 Bruce, J. M., 85 Brustolon, M., 69 Budil, D. E., 13 Burch, R., 913 Burgess, J., 559, 865 Burke, L. D., 299 Busca, G., 853 Buscall, R., 873 Cairns, J. A., 913 Carley, A.F., 351 Cassidy, J. F., 231 Celalyan-Berthier, A., 401 Chalker, P. R., 351 Chandra, H., 759 Chieux, P., 687 Clark, B., 865 Clifford, A. A., 751 Colin, A. C., 819 Coller, B. A. W., 645, 657 Coluccia, S., 477 Compostizo, A., 819 Corvaja, C., 57 Couillard, C., 125 Courbon, H., 697 Craven, J. B., 779 Crossland, W. A., 37 D’Alba, F., 267 Chu, D-Y., 635 Daverio, D., 705 Davoli, I., 289 Dawber, J. G., 771 De Doncker, J., 125 De Laet, M., 125 De Ranter, C. J., 257 Declerck, P. J., 257 Delahanty, J. N., 135 Di Lorenzo, S., 267 Diaz Peiia, M., 819 Dodd, N. J. F., 85 Ducret, F., 141 Dudikova, L., 51 1 Dusaucy, A-C., 125 Elbing, E., 645, 657 Empis, J. M. A., 43 Endoh, A., 41 1 Engberts, J. B. F. N., 865 Evans, J. C., 43, 135 Fan, G., 323 Fatome, M., 177 Flint, N. J., 167 Formosinho, S.J., 431 Forrester, A. R., 21 1 Fraissard, J., 45 1 Fujii, K., 675 Galli, P., 853 Geoffroy, M., 401 Gervasini, A., 705 Gilbert, B. C., 77 Gottschalk, F., 571 Gozzi, D., 289 Grampp, G., 161 Gray, P., 751 Greci, L., 69 Grieser, F., 591 Grossi, L., 77 Grzybkowski, W., 281 Guardado, P., 559 Hakin, A. W., 559, 865 Hall, M. V. M., 571 Halpern, A., 219 Hamada, K., 527 Harrer, W., 161 Hartland, G. V., 591 Hasegawa, A., 759 Hatayama, F., 675 Heatley, F., 517 Hemminga, M. A., 203 Herold, B. J., 43 Hertz, H. G., 687 Higgins, J. S., 939 Holden, J. G., 615 Howe, R. F., 813 Hudson, A., 91 Hunter, R., 571 Hutchings, G. J., 571 Imamura, H., 743 Imanaka, T., 665 Ito, T., 451 Jackson, S. D., 905 Jaenicke, W., 161 Janes, R., 383 Kanno, T., 721 Kerr, C W., 85 Kobayashi, M., 721 Koda, S., 527 Konishi, Y., 721 Kordulis, C., 627 Korth, H-G., 95 Kowalak, S., 535 Kubelkova, L., 511 Kubokawa, Y., 675 La Ginestra, A., 853 Lambelet, P., 141 Lavagnino, S., 477 Leaist, D.G., 829 Lecomte, C., 177 Lin, C. P., 13 Lindgren, M., 893 Liu, R-L., 635 Loliger, J., 141 Lorenzelli, V., 853 Loretto, M. H., 615 Lund, A., 893 Lycourghiotis, A., 627 Lyons, C. J., 645 Lyons, M. E. G., 299 Maezawa, A., 665 Makela, R., 51 Maniero, A. L., 57, 69 Marchese, L., 477 Marcus, Y., 339 Mari, C. M., 705 Masiakowski, J. T., 893 Masliyah, J. H., 547 Matsuura, H., 789 McCarthy, S. J.. 657 McLauchlan, K. A., 29 Mehandru, S. P., 463 Miyata, H., 675 Monk, C. B., 425 Morazzoni, F., 705 Moyes, R. B., 905 Mozzanega, M-N., 697 Nair, V., 487 Narayanan, S., 733 Narducci, D., 705 Nomura, H., 527 Norris, J.R., 13 Nukui, K., 743 Nuttall, S., 559AUTHOR INDEX O’Brien, A. B., 371 Ohno, T., 675 Ohshima, K., 789 Okabayashi, H., 789 Okamoto, Y., 665 Ono, T., 675 Pallas, N. R., 585 Parry, D. J., 77 Patrono, P., 853 Pedersen, J. A., 107 Pedulli, G. F., 91 Pethica, B. A., 585 Pethrick, R. A., 938 Pichat, P., 697 Pilarczyk, M., 281 Pizzini, S., 705 Pogni, R., 151 Pomonis, P., 627 Priolisi O., 57 Purushotham, V., 21 1 Radulovic, S., 559 Raffi, J. J., 225 Riviere, J. C., 351 Roberts, M. W., 351 Roman, V., 177 Romiio, M. J., 43 Rosseinsky, D. R., 231, 245 Rowlands, C. C., 43, 135 Rubio, R. G., 819 Sakai, T., 743 Sangster, D. F., 657 Saraby-Reintjes, A., 271 Saucy, F., 141 Savoy, M-C., 141 Segal, M. G., 371 Segre, U., 69 Seyedmonir, S., 813 Sidahmed, I.M., 439 Steenken, S., 113 Stevens, D. G., 29 Suppan, P., 495 Sustmann, R., 95 Swartz, H. M., 191 Symons, M. C. R., 1, 383, 759 Szostak, R., 487 Tabner, B. J., 167 Taga, K., 789 Takaishi, T., 41 1 Tan, W. K., 645 Tempest, P. A., 925 ThiCry, C. L., 225 Thomas, T. L., 487 Thurai, M., 841 Tilquin, B., 125 Tomellini, M., 289 Tonge, J. S., 231, 245 Trabalzini, L., 151 Tsuchiya, S., 743 Tsukamoto, K., 789 Turner, J. C. R., 937 Tyler, J. W., 925 Uma, K., 733 Vachon, A., 177 van de Ven, T. G. M., 547 Vincent, P. B., 225 Vink, H., 801, 941 Vordonis, L., 627 Vuolle, M., 51 Waddicor, J. I., 751 Wells, C. F., 439, 939 Wells, P. B., 905 White, L. R., 591, 873 Whyman, R., 905 Williams, J. O., 323 Williams, W. J., 371 Wilson, I. R., 645, 657 Yamada, K., 743 Zhang, Q., 635THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION No.84 Dynamics of Elementary Gas-phase Reactions University of Birmingham, 14-16 September 1987 Organising Committee: Professor R. Grice (Chairman) Dr M. S. Child Dr J. N. L. Connor Dr M. J. Pilling Professor I. W. M. Smith Professor J. P. Simons The Discussion will focus on the development of experimental and theoretical approaches to the detailed description of elementary gas-phase reaction dynamics. Studies of reactions at high collision energy, state-to-state kinetics, non-adiabatic processes and thermal energy reactions will be included. Emphasis will be placed on systems exhibiting kinetic and dynamical behaviour which can be related t o the structure of the reaction potential- energy surface or surfaces.The preliminary programme may be obtained from: Mrs. Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W 1 V OBN THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY SYMPOSIUM N o . 23 Molecular Vibrations University of Reading, 15-16 December 1987 Organising Committee: Professor I. M. Mills (Chairman) Dr J. E. Baggott Professor A. D. Buckingham Dr M. S. Child Dr N. C. Handy Dr 6. J. Howard The Symposium will focus on recent advances in our understanding of the vibrations of polyatomic molecules. The topics to be discussed will include force field determinations by both ab initio and experimental methods, anharmonic effects in overtone spectroscopy, local modes and anharmonic resonances, intramolecular vibrational relaxation, and the frontier with molecular dynamics and reaction kinetics.The preliminary programme may be obtained from: Mrs. Y. A. Fish, The Royal Society of Chemistry, Burlington House, London W1V OBN~~ THE FARADAY DIVISION OF THE ROYAL SOCIETY OF CHEMISTRY GENERAL DISCUSSION N o . 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 The purpose of the Discussion is to compare solvation of ionicand non-ionic species in the gas phase and in matrices with corresponding solvation in the bulk liquid phase. The aim will be t o confront theory with experiment and t o consider the application of these concepts to relaxation and solvolytic processes.Contributions for consideration by the organising Committee are invited in the following areas: (a) Gas phase non-ionic clusters (b) Liquid phase non-ionic clusters (c) Gas phase ionic clusters (d) Liquid phase ionic solutions (e) Dynamic processes including solvolysis Abstracts of about 300 words should be sent by 31 May 1987 to: Professor M. C. R. Symons, Department of Chemistry, The University, Leicester LE1 7RH. Dr J. Yarwood Dr A. D. Pethybridge Professor W. A. P. Luck Dr D. A. Young 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 Organ king Committee: Professor A. C. Legon (Chairman) Dr P. 6. Davies Dr 6. 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. Contributions for consideration by the Organising Committee are invited. Titles should be submitted as soon as possible and abstracts of about 300 words by30 September 1987 to: Professor A. C. Legon, Department of Chemistry, University of Exeter, Exeter EX4 4QD. Full papers for publication in the Discussion volume will be required by May 1988.JOURNAL OF CHEMICAL RESEARCH Papers dealing with physical chemistrykhemical physics which have appeared recently in J.Chem. Research, The Royal Society of Chemistry's synopsis + microform journal, include the following: Gas-phase Pyrolysis of Nitroethene. Surface-promoted Formation of Nitrosoethene Helge Unified Theory of Metal-ion-complex Formation Constants Paul L. Brown and Ronald N. Empirical Relationships for Estimation of Enthalpies of Formation of Simple Hydrates, Part 1. Phillippe Egsgaard and Lars Carlsen (1987, Issue 1) Sylva (1 987, Issue 1 ) Hydrates of Al kali-metal Cations, of Hydrogen, and of Monovalent Cations Vieillard and H. Donald B. Jenkins (1986, Issue 12) Empirical Relationships for Estimation of Enthalpies of Formation of Simple Hydrates, Part 2. Phillippe Vieillard and H. Donald B. Jenkins Hydrates of Alkaline-earth-metal Cations (1986, Issue 12) Empirical Relationships for Estimation of Enthalpies of Formation of Simple Hydrates, Part 3.Hydrates of Transition Metal Cations (Cr2+, Fez+, Mn2+, Co2+, Ni2+, Cu2+, Zn2+, Cd2+) and of U02+ Phillippe Vieillard and H. Donald B. Jenkins (1 986, Issue 12) Formation of Methyl Radicals from Radical-cations of Lead Tetra-acetate and Thallium Acetates Harish Chandra and Martyn C. R. Symons (1986, Issue 11) Dynamic N.M.R. Investigation of Rotation about Alkyl-Olefin Bonds. Barriers and Populations for Some Substituted lsopropylethylenes J. Edgar Anderson, Bernt Bettels, Parviz Gharagozloo, and Kam-Hang Koon (1 986, Issue 1 1 ) Optical Study of the 'Face-to-Face' Complexation of Water-soh ble Metalloporphyrins and Stanislaw Radzki, Serge Gaspard, and Charles Giannotti Metallophthalocyanines (1 986, Issue 10) (Viii)FARADAY DIVISION INFORMAL AND GROUP MEETINGS Electrochemistry Group Spring Informal Meeting To be held at the University of Bristol on 1-3 April 1987 Further information from Dr A.R. Hillman, School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 ITS Polar Solids Group Ceramic Sensors To be held at the University of Keele on 6-7 April 1987 Further information from Professor C. R. A. Catlow, Department of Chemistry, University of Keele, Staffordshire ST5 5BG Colloid and Interface Science Group Rheology of Dispersions and Suspensions To be held at the University of Bath on 9-10 April 1987 Further information from Dr R. Aveyard, Department of Chemistry, University of Hull, Hull HU6 7RX Division - Annual Congress The Chemistry and Physics of Intercalation To be held at University College, Swansea on 13-16 April 1987 Further information from Professor J.H. Purnell, Department of Chemistry, University College, Singleton Park, Swansea SA2 8PP Polymer Physics Group Fundamental Aspects of Polymer Flammability To be held at Baden Powell House, London on 14-15 April 1987 Further information from Dr G. C. Stevens, CERL, Kelvin Avenue, Leatherhead KT22 7SE Division Full-day Endowed Lecture Symposium on Intramolecular Dynamics and Chemical Reactivity including the Centenary Lecture by S. A. Rice and the Tilden Lecture by M. S. Child To be held at the Scientific Societies Lecture Theatre, London on 6 May 1987 Further information from Mrs Y.A. Fish, The Royal Society of Chemistry, Burlington House, London WIV OBN Polymer Physics Group Electroactive Polymers To be held at the Geological Society, London on 14 May 1987 Further information from Dr G. C. Stevens, CERL, Kelvin Avenue, Leatherhead KT22 7SE - - - ~ ~~~ Electrochemistry Group with Macro Group UK Polymer Electrolytes To be held at the University of St. Andrews on 18-19 June 1987 Further information from Dr C. A. Vincent or Dr J. R. MacCallum, Department of Chemistry, University of St. Andrews, St. Andrews KY16 9ST Gas Kinetics Group Thermally and Photochemically Activated Reactions To be held at the University of Edinburgh on 9-10 July 1987 Further information from Professor R. Donovan, Department of Chemistry, University of Edinburgh, West Mains Road, Edinburgh EH9 3JJ Division Xlth International Symposium on Molecular Beams To be held at the University of Edinburgh on 13-17 July 1987 Further information from Dr J.F. Gibson, The Royal Society of Chemistry, Burlington House, London WlVOBNIndustrial Physical Chemistry Group The PhysScal Chemistry of Small Carbohydrates (as part of the International Symposium on Solute-Solute-Solvent Interactions) To be held at the University of Regensburg, West Germany on 10-14 August 1987 Further information from Dr F. Franks, Pafra Ltd, 150 Science Park, Milton Road, Cambridge CB4 4GG Industrial Physical Chemistry Group The Interaction of Biologically Active Molecules and Membranes To be held at Girton College, Cambridge on 8-10 September 1987 Further information from Dr T. G. Ryan, ICI New Science Group, PO Box 11, The Heath, Runcorn WA7 4QE Polymer Ph ysics Group Biennial Meeting To be held at University of Reading on 9-1 1 September 1987 Further information from Dr D. Bassett, Department of Physics, University of Reading, Reading RG7 2AD Neutron Scattering Group Applications of Neutron and X-Ray Optics To be held at the University of Oxford on 14-15 September 1987 Further information from Dr R. K. Thomas, Physical Chemistry Laboratory, South Parks Road, Oxford OX1 302 Colloid and Interface Science Group Polydispersity in Colloid Science To be held at the University of Nottingham on 15-16 September 1987 Further information from Dr. R. Buscall, ICI plc, Corporate Colloid Science Group, PO Box 11, The Heath, Runcorn WA7 4QE Polymer Physics Group New Materials To be held at the University of Warwick on 22-25 September 1987 Further information from Dr M. J. Richardson, Division of Materials Applications, National Physical Laboratory, Queens Road, Teddington, Middlesex TW11 OLW Division Autumn Meeting Spectroscopy of Gas-phase Molecular Ions and Clusters To be held at the University of Nottingham on 22-24 September 1987 Further information from Professor J. P. Simons, Department of Chemistry, University of Nottingham, Nottingham NG7 2RD Polymer Physics Group with the Institute of Marine Engineers Polymers in a Marine Environment To be held in London on 14-16 October 1987 Dr G. J. Lake, MRPRA, Brickendonbury, Herts SG13 8NL

ISSN:0300-9599    

DOI:10.1039/F198783BP035

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1987, 83, 571-583 Hydrocarbon Formation from Methylating Agents over the Zeolite Catalyst ZSM-5 Comments on the Mechanism of Carbon-Carbon Bond and Methane Formation Graham J. Hutchings," Frank Gottschalk, M. V. Michele Hall and Roger Hunter* Department of Chemistry, University of Witwatersrand, 1 Jan Smuts Avenue, Johannesburg 2001, Republic of South Africa The mechanism of hydrocarbon formation from methanol has been studied by reacting methylating agents MeX (X = OH, I, OS0,Me) over the acidic form of the pentasil zeolite ZSM-5 and the sodium form of its conjugate base. At > 0.1 % conversions the product distributions obtained are similar, indicating that similar mechanisms for carbon-carbon bond formation operate with the three reagents. Hydrocarbon formation (CH, and C,H,) from Me,SO, over Na-ZSM-5 was observed and is considered to be strong evidence against the involvement of dimethyloxonium methylide as an intermediate since such a species cannot be formed with this reactant under basic conditions.Further evidence against the involvement of a dimethyloxonium methylide is given using reactivity comparisons of oxygen and sulphur-containing reagents. Me1 and Me,SO, were found to be less reactive than MeOH and at low conversion the primary hydrocarbon products observed, CH, and C,H,, suggest that these are the primary products formed in methanol conversion. Under conditions where conver- sion increased with reaction time, the methane/alkenes ratio decreased markedly and this is considered to be inconsistent with a mechanism involving a surface carbene intermediate formed in a one-step process from the reagent MeX.Instead it is proposed that the crucial first step in the mechanism is the formation of a surface methoxyl species which is primarily responsible for CH, formation at low conversions. Deprotonation of the surface methoxyl species generates a surface-bonded oxonium methylide which subsequently reacts to form the initial carbon+arbon bond. This mechanistic proposal is both consistent with and unifies the available experimental data obtained for this reaction to date. ________________ In the methanol to gasoline-range hydrocarbons process catalysed by the pentasil zeolite H-ZSM-Y and bifunctional acid-base catalysts,2 considerable research effort has gone into elucidating the mechanism of formation of the initial carbon-carbon bond, but an agreement on it has yet to be reached.Current experimental evidence indicates the involvement of either an oxonium ion ylide3~ or a surface-bound carbene species1 as likely reaction intermediates and experimental evidence to support the involvement of both species has been presented re~ently.~? In addition, the nature of the primary hydrocarbon products is still to be resolved. A number of workers have considered the possibility that both ethene and propene might be primary alkene products3* and Haag* has demonstrated that as the conversion of methanol to hydrocarbon decreases, so does the propene/ethene ratio, indicating that ethene alone may be the primary alkene product.Subsequent studies by Chu and Chang9 and Wu and KaedinglO have supported this proposition. However, Espinoza and Manderslootll have demonstrated that the evidence of Haag is inconclusive since the propene/ethene ratio depends only on the 57 1 20-2572 Hydrocarbon Formation over Zeolites dimethyl ether/methanol ratio and not on conversion. The major problem in the identification of the primary product(s) from methanol is that even at low conversion, significant quantities of ethene, propene as well as higher hydrocarbons are ObservedlO and extrapolation of the results to determine the nature of the primary products is inconclusive. Conversion of the primary products to higher hydrocarbons has been explained using carbocationic intermediates1, involving methanol alkylation. The auto- catalytic nature of the process is well documented13 and is considered to be a result of the rapid reaction of the primary alkene product(s) with methanol to give the broad range of products typically observed.However, to date no studies have specifically addressed the formation of methane in the methanol conversion reaction. At high methanol conversions methane is only formed in small quantities, e.g. ca. 1 mol% , and most workers consider that it originates mainly from thermal cracking of higher hydrocarbons catalysed by the highly acidic sites of H-ZSM-5. However, as indicated by HaagTs methane is the"first and major product formed at very low conversion, i.e. < 0.5 mol % . At such low conversions it is extremely doubtful that thermal cracking can account for the significant methane yields, e.g.26.5 mass % at 0.03 % methanol conversion,1o since sufficient concentrations of the higher hydrocarbons are not present. It is therefore clear that any mechanism proposed for the formation of the initial carbon+arbon bond must also account for the formation of methane in agreement with the experimental observations. It is known that the zeolite H-ZSM-5 converts a range of different organic precursors to hydrocarbons' and we have shown in our recent comm~nication~~ that the zeolite can also convert a range of methylating agents to hydrocarbons. In this paper we extend our preliminary mechanistic studies using methylating agents, and in particular we consider the mechanistic origins of methane in addition to the formation and nature of the primary hydrocarbon product(s).Experimental Zeolite Preparation The sodium form of the pentasil zeolite ZSM-5 (Na-ZSM-5) was prepared according to the method of Howden15 with SiO,/Al,O, mole ratios of 35 and 70. Na-ZSM-5 was then converted into the acidic form (H-ZSM-5) by the following procedure. Na-ZSM-5 (100 g) was stirred in an aqueous solution of ammonium sulphate (1 dm3, 1 mol dm-3) for 30 min at 25 "C, then filtered, washed with distilled water and the procedure repeated twice. The zeolite was thoroughly washed with water to ensure removal of all the sulphate ions, dried and then calcined at 550 "C for 3 h and stored in a desiccator when cool. Catalyst Testing Catalytic reactions were carried out using a fixed-bed Pyrex glass reactor with a temperature control of 1 "C.Zeolite catalyst (1 g) was dried in situ in the reactor using a stream of dry N, at 350 "C for several hours prior to use. MeOH, Me,SO,, Me1 and Me,S were individually reacted over the zeolite by vaporising them at a regulated temperature using dry N, carrier gas at a constant flow rate of 1 cm3 s-l. The weight hourly space velocity (w.h.s.v. = g reactants per g catalyst h-l) of the reactants could be varied by control of the vaporisation temperature, which ranged from -60 "C for MeOH and Me1 to 130 "C for Me,SO,. Products were analysed using gas chromato- graphy. MeOH (AR, BDH) and Me1 (Carlo Erba, stored over 4A molecular sieves) were used without further purification. Me,S (analytical reagent, Aldrich) was distilled prior to use.Me,SO, (Hopkins and Williams) was distilled from CaH, and stored over 4A molecular sieves. Prior to use, Me,SO, was analysed by n.m.r. spectroscopy to ensure that no MeOH was present. Thermal blank reactions of the reagents in the absence ofG. J . Hutchings et al. 573 Table 1. Reaction of Me,SO, and Me1 over the zeolite catalyst H-ZSM-5 at 250 "C time on total product selectivity (molx )" methylating w.h.s.v. line conversion agent /h-l /min ( m o l z ) CH, C,H, C3H, C,H, C,, Me,SO, 0.054 15 30 45 60 75 Me,SO, 0.075 15 60 100 190 230 Me1 0.6 15 30 60 75 MeIb 0.8 60 100 Me1 2.0 20 60 0.007 0.010 0.024 0.90 1 .oo 0.12 3.3 12.9 16.6 21.2 0.01 0.02 0.10 0.1 1 0.02 0.13 1.2 1.7 24.0 76.0 0 0 0 21.3 78.7 0 0 0 17.6 82.4 0 0 0 2.4 14.8 29.6 38.4 14.8 1.6 28.0 28.2 30.2 12.0 81.5 18.5 0 0 0 20.5 44.9 28.2 6.4 0 2.5 28.9 11.2 18.6 38.8 1.6 36.6 1212 10.5 39.1 1.5 45.2 15.1 10.8 27.4 58.8 41.2 0 0 0 51.7 48.3 0 0 0 1.9 7.3 38.8 40.0 12.0 0.5 3.5 78.9 13.1 4.0 7.0 44.1 32.5 13.6 2.8 1.3 8.5 32.1 32.6 25.4 1 .o 5.8 24.4 28.5 40.1 66.1 33.9 0 0 0 a Only trace amounts of alkanes were observed.Zero conversion at 15 min time on line. Table 2. Reaction of methanol over zeolite catalyst H-ZSM-5 time on conversion to product selectivity (molx ) w.h.s.v. line hydrocarbons /h-l T/"C /min (mol%) CH4 "ZH4 C2H6 C3H6 C3H8 '4 c,+ 0.005 250 15 30 60 0.060 2 50 0.044 250 15 30 45 60 75 15 30 45 75 90 0.150 250 15 60 100 0.03 300 40 90 0.03" 3 50 40 86.2 85.3 86.6 78.2 87.0 97.1 86.5 90.7 16.5 81.8 86.2 91.4 95.2 7.1 16.7 22.6 29.0 51.0 57.0 1.6 5.5 0 20.5 4.6 63.4 4.5 1.1 6.6 0 19.2 3.7 49.1 20.3 0.8 8.8 0 17.1 2.6 35.6 35.1 2.2 2.8 0 9.9 2.5 38.0 44.6 1.8 5.7 0 9.5 2.7 38.4 41.9 0.8 5.0 trace 7.0 1.7 28.1 57.4 0.9 7.6 trace 9.4 1.9 27.8 52.4 0.5 5.1 trace 4.3 1.0 14.5 74.6 2.1 9.9 0 11.6 1.5 17.2 57.7 1.8 5.6 0 13.0 2.5 35.0 42.0 2.0 10.2 trace 13.7 2.8 38.2 33.1 1.3 12.4 trace 12.4 3.1 30.1 40.7 1.1 13.3 trace 10.7 2.8 29.3 42.8 4.0 3.9 0 2.6 1.5 36.1 52.0 1.1 14.1 trace 3.6 2.0 28.8 51.3 0.9 24.3 trace 3.9 2.4 29.3 39.2 5.2 34.7 0 24.4 3.0 10.4 22.3 1.8 35.2 0 28.2 3.9 10.2 20.7 0.7 5.7 0 9.2 3.0 35.3 53.9 No hydrogen observed at 250 and 300 "C, only a trace of hydrogen was observed at 350 "C.574 Hydrocarbon Formation over Zeolites catalyst for the reaction conditions studied did not produce any hydrocarbons.The water gas shift activity of the catalyst was tested using the following procedure. Synthesis gas (CO/H, = 1 by volume) was passed through water saturators at a constant flow rate and temperature, and the resultant gas mixture (typically CO: H, : H,O = 1 : 1 : 4 by volume) was then reacted over the zeolite. The products were cooled to 0 "C to trap out unreacted water and products were analysed by gas chromatography. Results Reaction of Methylating Agents over Zeolite H-ZSM-5 MeOH, Me1 and Me,SO, were reacted individually over H-ZSM-5 (SiO,/Al,O, mol ratio 35) for a range of weight hourly space velocities at 250 "C and the results are shown in tables 1 and 2. All reagents showed increasing conversion with time on line as has been previously observed for methanol conversion by Ono and M o i 8 Of the three reagents, MeOH was found to be the most reactive with the order of reactivity being MeOH > Me,SO, > MeI, suggesting that hydrophilicity of reagents for entry into the internal zeolite structure and ease of formation of the protonated species are important factors in reactivity.With Me1 and Me,SO, the products initially formed at low conversion were methane and ethene, whereas with MeOH considerable quantities of higher hydrocarbons were observed even at low conversions. However, for all reagents and conditions investigated, the methane yield steadily decreased with increasing reaction time and/or conversion. It is important to note that MeOH could be a by-product of Me,SO, reacting as a methylating agent according to scheme 1.However, any MeOH produced in this way would be insignificant compared with the concentration of Me,SO, around it within the zeolite pore structure. 0 0-H H II Me I I1 II ____) 0 0 I + Me-0-S-OMe I + OrS-OMe 0 - H I I1 0 O=S-OMe - SO3 + MeOH Scheme 1. MeOH was reacted over zeolite H-ZSM-5 (SiO,/Al,O, molar ratio 70) at 350 "C, w.h.s.v. = 2.5 h-l and the results are shown in table 3. Under these conditions the zeolite gives total MeOH conversion for an appreciable time before a steady decrease in conversion is observed which has been ascribed to loss of catalyst acidity together with deposition of carbonaceous deposits within the zeolite pore structure.13 The striking feature of these results is that as the conversion decreases the methane selectivity increases by a factor of ca.3. At the same time the selectivity to higher hydrocarbons steadily decreases.G. J . Hutchings et al. 575 Table 3. Methanol conversion over zeolite H-ZSM-Y time on methanol product selectivity (mol % ) line conversion /min (mol%) CH4 c2 c3 c4 c5+ 45 142 255 365 475 605 1024 1186 100 100 99.8 92.0 77.8 51.1 29.0 6.2 13.9 25.2 31.4 23.4 6.1 11.5 30.4 36.4 19.4 2.3 22.4 42.8 23.8 8.1 2.9 18.2 44.7 24.8 9.5 2.8 22.1 43.6 23.9 7.5 2.9 22.3 45.7 23.4 6.6 2.0 36.3 44.0 11.7 8.0 0 40.0 45.2 12.8 2.0 0 ' . a (Si02/A1203 = 70), w.h.s.v. = 2.5 h-l, 370 "C. Table 4. Reactions of methylating agents over zeolite Na-ZSM-5 time on product selectivity (mol % ) methylating w.h.s.v. line conversion agent /h-l T/"C /min (molx) CH, C2H, c3+ MeOH" 0.075 300 200 0.001 64.4 35.6 0 Me1 0.10 250 15 0.05 66.0 34.0 0 0.10 300 90 0.07 70.0 30.0 0 Me2S0, 0.075 250 90 0.06 52.4 47.6 0 0.075 300 130 2.0 48.1 51.9 trace a No CO or H, detected.Reaction of Methylating Agents over Zeolite Na-ZSM-5 MeOH, Me1 and Me,SO, were individually reacted over the sodium form of the conjugate base of the zeolite ZSM-5 (SiO,/Al,O, molar ratio 35) at temperatures 250 to 300 "C and the results are shown in table 4. The products observed were mainly methane and ethene and only a trace of propene was observed with Me,SO, after an extended reaction period. The formation of ethene at a conversion of ca. 2% with Me,SO, under these conditions is considered to be highly significant since formation of an oxonium ion, the central feature of the oxonium ion-ylide mechanism proposed by Olah4 and van den Berg,3 is not considered possible.Reaction of Dimethyl Sulphide over Zeolite H-ZSM-5 An important feature of both the zeolite catalyst H-ZSM-5l and the bifunctional acid-base catalysts, e.g. W0,/Al,0,,2 is their ability to convert a range of small organic molecules to petroleum-range hydrocarbons, but this general feature has not been fully exploited in mechanistic studies. Previous studies have shownl7 that sulphur- containing reagents are less reactive than the corresponding oxygen analogues. Few data are available and direct comparison of activities and selectivities has not been possible. Me,S and MeOH were individually reacted over the zeolite H-ZSM-5 at a range of temperatures, and the results are shown in table 5 together with the previous literature data for comparison.Me,S is considerably less reactive than MeOH at comparable576 Hydrocarbon Formation over Zeolites Table 5. Reaction of dimethylsulphide over H-ZSM-5 w.h.s.v. reagent /h-l Me,S 0.8 Me,S 0.3 MeOH 0.3 MeOHa l.Ob MeSHa 1 .Ob MeOHC 2.5d MeOHf 509 Me,@ 509 Me,Sf 509 Me20c 3.9d T/"C conversion to hydrocarbons (mol% ) 250 300 375 415 475 250 300 325 37 1 482 450 450 325 320 380 trace 0.1 5.3 19.4 52.7 18.8 31.2 48.4 72.7 100 100 100 69.0 70.0 32.0 product selectivity (% by mass) CH4 C2H4 'ZH6 C3H6 C3H13 '4 trace - - - - - 8.0 19.3 0 6.1 0 60.6 9.5 22.4 0 7.1 0 58.2 9.3 23.1 1.2 11.4 3.8 49.6 19.2 29.5 2.5 11.4 5.7 31.6 0.9 34.5 trace 16.6 7.1 7.1 0.6 28.8 0.2 20.4 9.2 15.5 0.6 20.2 0.2 16.9 8.7 9.3 1.0 0.5 0.6 1.0 16.2 25.6 6.6 6.7 8.3 1.3 15.3 12.3 3.2e 5.2 0.1 32.1 0.7 24.6 1.0" 2.7 0.1 31.0 0.4 25.2 46.2 29.8 0.8 19.3 trace 3.9 39.8 28.6 1.2 20.6 0.8 7.9 63.8 15.4 1.0 18.5 trace 1.3 - 6.0 2.8 1.6 0.2 33.8 25.3 44.1 55.1 55.5 34.1 39.7 trace 1.1 - a Data taken from ref.(l), catalyst H-ZSM-5. 1.h.s.v. = volume liquid reactant (cm3) per volume catalyst cm3 h-l. Data taken from ref. (lo), catalyst H-ZSM-5. Space time (s) = catalyst volume/gaseous flow rate. " CH4+CO+C02. f Data taken from ref. (2), catalyst W03/A1203. 9 g.h.s.v. = volume gaseous reactant (cm3) per g catalyst h-l. Table 6. Conversion of methanol to hydrogen over the zeolite Na-ZSM-5 methanol w.h.s.v. conversion T/"C /h-l (mol% 1 250 0.08 0 250 14.8 0 300 0.1 0 3 50 0.1 0 350 10.4 0.9 380 15.1 0.1 1 400 16.2 0.22 conditions and in general the reaction temperature has to be increased by ca.150 "C to obtain comparable reactivity. In addition there are significant differences in product selectivity, the most important being that CH, formation is much higher for Me$ under all conditions tested. Methanol Decomposition and Water-gas Shift Reaction over Zeolite ZSM-5 In a recent study Mihail et aZ.16 proposed that methane formation could result from the known reaction of a carbene with molecular hydrogen" in addition to the cracking of higher alkanes. This latter reaction is only expected to become significant at the high conversions required to obtain significant concentrations of the higher alkanes.G.J . Hutchings et al. 577 Therefore, at low methanol conversions when methane concentrations can be very high8T lo, l3 a significant source of hydrogen is required if methane is generated from a carbene intermediate. Mihail et al.ls postulated the hydrogen source to be a combination of the methanol decomposition reaction (CH30H e CO + 2H,) and the water-gas shift reaction (CO + H,O e CO, + H2), but they provided no experimental evidence to support this proposal. Methanol decomposition was therefore studied over the sodium form of the zeolite Na-ZSM-5 since the complicating reaction of hydrocarbon formation over this zeolite is negligible.14 The results, shown in table 6, indicate that at 250 "C methanol decomposition is not observed for a wide range of reactant feed rates.Methanol decomposition is only observed to a limited extent at higher temperatures (350-400 "C). The water-gas shift activity of the acidic form of zeolite H-ZSM-5 was studied and for a wide range of experimental conditions (T = 35WOO "C) the reaction was not observed. These results indicate that this reaction cannot be considered to occur over the zeolite catalyst at typical hydrocarbon formation conditions. These experiments demonstrate that methanol decomposition and water-gas shift reactions are not viable sources of hydrogen to account for the high concentrations of methane formed at low conversions via direct reaction of hydrogen with a carbene intermediate. If this reaction does proceed then an alternative source of hydrogen must be cited.However, monitoring of hydrogen concentrations during methanol conversion (table 2) indicated that only traces of hydrogen were observed at 350 "C, whilst at 250 and 300 "C hydrogen was not detected as a product, although significant methane yields were observed. Discussion Mechanism of Carbon-Carbon Bond Formation Of the mechanistic proposals to account for the formation of the initial carbon-carbon bond from a C, precursor, only two have received significant experimental support, namely involvement of a carbene intermediate' (scheme 2) and the dimethyloxonium methylide proposal of van den Berg13 and Olah4 (scheme 3). The evidence supporting the carbene mechanism includes hydrocarbon formation from diazomethane5 and analysis of the product distribution,18 while the main support for the dimethyl- oxonium methylide proposal comes from the studies of deuterium exchange by M ~ l e ~ ~ ~ ~ ~ and the production of methyl ethyl ether from a trimethyloxonium salt by reaction with the strong, hindered, non-nucleophilic base 2,2,6,6- tetrame t h ylpiperid yl lithium.21 However, neither mechanism satisfactorily accounts for all of the experimental data available.A common feature of both proposals is that they involve the conjugate base of the zeolite acting as a base towards a non-bonded reactant species. A crucial question therefore arises as to whether the conjugate base of H-ZSM-5, the latter being a very strong acid, is sufficiently basic to deprotonate a non-surface-bonded reactant. In a previous study22 we investigated this aspect by modelling the conjugate base using LiA1(OPri), and studying the reactions of this model compound with a range of methylating agents, including a trimethyloxonium salt.The results of our studies showed that these reagents reacted solely as methylating agents, which is what might be predicted from their known chemistry, and that the A1-0 bond is far more nucleophilic than basic in its reactions with these non-bonded reagents. These results indicate therefore that if the trimethyloxonium ion were to be formed during the reaction it would be expected to react primarily as a methylating agent, i.e. it would methylate the zeolite surface to generate surface methoxyl groups which have been proposed on the basis of experimental data by Ono and Mori.13 The results of reaction of a range of methylating agents with the zeolite ZSM-5 now reported by us confirm the previous results of the model studies using LiA1(OPri),.578 Hydrocarbon Formation over Zeolites Formation of surface carbene species ..8-H CH2 H 2 0 B .. where B basic form of ZSM-5 B-H acidic form of ZSM-5 Formation of methane CH2 + CH3OH - CH4 + CH20 Formation of alkenes CH2 + CH3OH - CH2= CH2 + H2O CH2 + CH2=CH2 - CH3CH=CH2 CH2 + RCH=CH2 - RCHzCH=CH, Scheme 2. + CH,OH + H + e CH,OH, + CH30H, + CH30H CH,0CH3 + H,O CH,0CH3 + H ' b (CH,),&H (CH,),i)H + CH,OH 3 (CH,),O++ H 2 0 CH, I Stevens * CH,-0 -CH2CH, 0 CH, I /+\cH2 rearrangement j I <'Hj H I \ Si p\;/o, A l Si / \ Si /"\ At /?\ Si / / \ /-\ / \ / \ /a\ / \ Scheme 3.G. J .Hutchings et al. 579 Reactions of MeOH, Me1 and Me,SO, with the acidic and basic forms of zeolite ZSM-5 show that these reagents give similar product distributions, strongly indicating that they react uia a common pathway as methylating agents of the type MeX. Formation of hydrocarbons from Me,SO, is considered to be highly significant. Under acidic conditions Me,SO, demonstrated similar reactivity to MeOH and gave similar reaction products, while under basic conditions Me,SO, proved to be several orders of magnitude more reactive than MeOH, which is consistent with their relative methylating ability. Under acidic conditions the oxygens of Me,SO, are far less nucleophilic than the oxygen of MeOH, therefore it is unlikely that under such conditions Me,SO, will preferentially generate an oxonium ion, an essential feature of the oxonium ion-ylide mechanism proposed by van den Berg3 and Olah., Moreover, under basic conditions formation of an oxonium ion with Me,SO, is impossible, suggesting even more strongly that the trimethyloxonium ion ylide is not an intermediate in initial carbon-carbon bond formation for MeOH conversion, providing that a common mechanism operates for both acidic and basic conditions.These results emphasise that if the trimethyloxonium ion is formed from MeOH within the zeolite structure its role is more likely to be that of a methylating agent rather than an ylide source. Further evidence against the involvement of trimethyloxonium ion as an ylide source is obtained from the relative reactivity of sulphur- and oxygen-containing reagents. If the trimethyloxonium ion is activing as an ylide source in the methanol conversion reaction then assuming deprotonation to be rate-limiting, substitution of sulphur for oxygen would be expected to enhance the reagent reactivity. However, this is not observed and oxygen-containing reagents are far more reactive than the sulphur- containing analogues.The results are more consistent with the relative methylating ability of MeSHi, Me,SH+ and Me,S+ compared to MeOHi, Me,OH+ and Me30+. These experimental findings are also consistent with our previous model studies2, in which we demonstrated that the oxygens of LiAl(OPri), were not sufficiently basic to deprotonate Me,S+I- nor sufficiently nucleophilic to be methylated by it. The relative reactivity of these compounds therefore tends to confirm that the dimethyloxonium methylide is not a viable intermediate for the formation of the initial carbon-carbon bond.In the light of these results we propose that the initial step of the reaction involves methylation of the zeolite active site to give what Ono and Mori call a surface methoxyl (surface-bonded methyloxonium This surface-bonded species is then depro- tonated via reaction with a nearby basic zeolite site (e.g. an adjacent A1-0 site) to give a surface-bonded oxonium methylide. This surface bonded species is isoelectronic with the surface carbene intermediate originally proposed by Chang and Si1vestri.l However, our evidence indicates that surface methylation precedes generation of the active intermediate, i.e.a two-stage process (scheme 4), and not a one-stage process as proposed by Chang and Silvestril (scheme 2). More recently Lee5 has shown that diazomethane can form hydrocarbons over H-ZSM-5 and considered that this evidence supported the existence of carbenoid- surface-bound ylide species as the active intermediate in carbon-carbon bond formation. However, since under acidic conditions diazomethane is known to react as an electro- philic methylating agent: CH2-N, + H-Z + +CH3-N, + Z- we consider that their results are consistent with our finding, namely that the initial action of the reagent is as a methylating agent. We consider that the proposed two-stage mechanism is consistent with all experimental evidence available to date since it unifies the central features of both the carbene and oxonium ion-ylide proposals.In particular, it is consistent with the deuterium-exchange580 H Hydrocarbon Formation over Zeolites acid form of Z S M - 5 Conjugate base form of ZSM-5 C H I + C H 2 O H X=OH,0S03Me, I Me surface "methoxyl" /protonat ion CH, I I surface incorporated ylide I I - R CH 2C H = C H 2 H I further methylation C CH2= CH2 initial carbon corbon bond formation surface carbene Scheme 4. experiments of Mole,l99 2o since both the surface methoxyl and the surface-bonded oxonium methylide are active surface species. In addition, it accounts for the 13C data of P e r ~ t , ~ ~ which indicate that ethene is formed intermolecularly from dimethylether, and the product distribution data of Chang and Chu,18 since the surface-bonded oxonium methylide is chemically identical to an adsorbed carbene species.Nature of the Primary Products The data for the reaction of MeOH, Me1 and Me,SO, over the acidic form of the zeolite, as stated previously, indicate, since the product distributions are similar for conversions > 0.1 mol % , that the reagents react via a common mechanism. The data also clearly show that the initial products with the less-reactive reagents at low conversion (i.e. Me,SO, and MeI) over H-ZSM-5 are always methane and ethene under our conditions. Since similar products and product distributions are observed at higher conversions for both these reagents and methanol we suggest that these results support the proposal that methane and ethene are the primary hydrocarbon products formed during methanol conversion, in agreement with the extrapolated results of Haag.8 Our finding that methane is a significant primary product with Me,SO, and Me1 is in agreement with the previous results obtained for MeOH by Haag,8 Wu and KaedinglO and Ono and Mori.13 Reaction of Me,SO, and Me1 over the basic form of the zeolite Na-ZSM-5 also confirms these findings and a trace of propene observed with Me,SO, (table 4) is evidence that propene is found as a secondary product via methylation of ethene.HenceG. J . Hutchings et al. 58 1 the use of the less-reactive model reagents Me,SO, and Me1 does not require the use of extrapolation methods to determine the primary products from methanol conversion. The findings of Espinoza and Mandersloot,ll that the propene/ethene ratio at low conversions correlates well with the (dimethyl ether)/methanol reagent ratio, are not inconsistent with our findings.We consider that the propene/ethene ratio observed in their experiments reflects the relative methylating ability of protonated dimethyl ether versus protonated methanol for the alkylation of ethene within the zeolite catalyst. Formation of Methane Since the adsorbed carbene and surface-bonded oxonium methylide species (schemes 2 and 4) are electronically identical, differentiation between these mechanistic proposals on the basis of chemical reactivity alone is not possible. However, we consider that the two mechanisms have different implications for production of methane, a major primary product at low conversion.Consideration of the reactions occurring in scheme 4 indicates that, in the initial stages, ethene is formed by reaction of the surface-bonded oxonium methylide species with methanol (or suitable methylating agent) and we propose that methane is formed from the surface methoxyl species via hydride abstraction from methanol. The by-product formaldehyde would be expected to polymerise rapidly to coke under the acidic conditions or could undergo a Cannizarro reaction as proposed by Olah.2 Under these conditions formaldehyde would not be an expected final product and has not been reported to date as a product in methanol conversion reactions over H-ZSM-5. Higher alkenes are formed by methylation of alkenes by the surface methoxyl via carbocationic intermediates which is consistent with a previous proposal.12 It is apparent from scheme 4 that the rate of formation of ethene, and all higher alkenes, is dependent on the concentration of the surface-bonded oxonium methylide species, whereas the rate of formation of methane is dependent on the concentration of the surface methoxyl species.Since in the initial stages the former will be steadily increasing since its formation is the slowest reaction step we consider that the ratio of methane/C,+ alkenes should decrease as the conversion increases, and this is what is observed. By comparison, consideration of the reactions occurring in scheme 2 for the carbene mechanism indicates that the surface carbenoid species is the central intermediate for the formation of both methane and alkenes.Therefore the ratio [methane]/[alkenes] should remain constant regardless of reactant concentrations and conversions and this is not experimentally observed. Furthermore, the absence of significant CO and hydrogen concentrations during methanol decomposition over Na-ZSM-5 is strong evidence against methane formation being via a reaction involving methylene carbene and hydrogen, and abstraction of two hydrogens by the methylene carbene from methanol to generate methane would appear to be an unlikely possibility compared with direct insertion to ethanol. Moreover the changing ratio [methane]/[alkenes] is also not consistent with the direct insertion of surface methoxyl into methanol to give ethanol which dehydrates to ethylene as proposed by Ono and Mori,s since then the surface methoxyl species would be giving rise to both methane and alkenes.The results of a series of methylating agents MeX (X = OH, I, OS0,Me) clearly show that as conversion increases in the initial stages of the reaction the ratio [methane]/[alkenes] decreases (table 7). The effect is most marked for the less-reactive methylating agents Me1 and Me,SO,. Consideration of literature data for low methanol conversions also shows a decrease in this ratio with increasing conversion.1o The effect is, however, more pronounced as the catalyst ages (table 3), since as the conversion drops the methane yield increases by a factor of ca. 3. We consider this indicates that the rate of formation of the surface-bonded oxonium methylide species decreases relative to the rate of formation of the surface methoxyl species.Hence, the formation of methane observed experi-582 Hydrocarbon Formation over Zeolites Table 7. [CH,]/[alkene] mole ratio uersus conversion time on total [CHJ methylating w. h.s.v. line conversion [alkene] agent /h-l /min (mol% ) mole ratio Me,SO, 0.054 15 30 45 60 75 60 100 190 230 Me,SO, 0.075 15 Me1 0.6 MeOH 0.044 15 30 60 75 15 30 45 60 75 MeOH 0.150 15 60 100 0.007 0.010 0.024 0.90 1 .oo 0.12 3.3 12.9 16.6 21.2 0.0 1 0.02 0.10 0.1 1 78.2 87.0 97.1 86.5 90.7 7.1 16.7 22.6 0.32 0.27 0.21 0.024 0.016 4.41 0.26 0.026 0.0 16 0.01 5 1.43 1.07 0.0 19 0.005 0.025 0.020 0.009 0.010 0.005 0.043 0.012 0.010 mentally is consistent with a two-step formation of a surface-bonded oxonium methylide species as the active intermediate for initial carbon-carbon bond formation in the methanol conversion reaction (scheme 4).However, the one-step concerted mechanism for the formation of the surface carbenoid species proposed by Chang and Silvestril (scheme 2) is clearly not consistent with methane formation and hence our findings are strong evidence against this mechanism. With regard to the difference in reactivity between MeSH and MeOH or Me$ and Me,O, lower conversions at a particular temperature for the sulphur analogues can be accommodated by both the Olah-van den Berg ion-ylide and surface-bonded methyl- oxonium ion mechanisms, since sulphur, being a softer nucleophile than oxygen, will react slower in generating the key trimethylsulphonium ion intermediate.However, if a dimethylsulphonium methylide mechanism is operative one would expect the sulphonium ylide to be generated faster than its oxonium ion counterpart, resulting in a greater ratio of alkenes to methane since the latter would be formed at the sulphonium ion stage prior to deprotonation. However, this is not observed. In fact, comparing data at similar conversions in table 5 shows that the methane concentration is always relatively higher for S us. 0. By comparison, a surface methylation mechanism, favoured by us, accommodates the data more satisfactorily since methane could be generated by reduction of either the sulphonium or surface-bonded methyloxonium ions and the demethylation of the former would always be relatively slower than with the oxygen analogue ensuring a relatively higher concentration of it at any one time.The results of this study using a range of methylating agents and dimethyl sulphide as less-reactive model reagents for methanol presents strong evidence in favour of aG. J. Hutchings et al. 583 surface-bonded oxonium methylide as being the reactive intermediate responsible for initial carbon-carbon bond formation to give ethene. In addition, it is proposed that this reactive species is generated from a surface methoxyl species derived from protonated methanol or other suitable methylating agents. We thank D. Glasser and S. W. Orchard for useful discussions and also the Senate Research Council, University of the Witwatersrand and the Foundation for Research Development, CSIR for financial support. References 1 C. D. Chang and A. J. Silvestri, J. Catal., 1977, 47, 249. 2 G. A. Olah, H. Doggweiler, JJ. D. Felbeg, S. Frohlich, M. J. Grdina, R. Karpeles, T. Kevmi, S. Laba, 3 J. P. van den Berg, J. P. Wolthuizen and J. H. C. van Hoof, Proc. 5th ConJ: Zeolites, Naples, 1980, 4 G. A. Olah, Pure Appl. Chem., 1981, 53, 201. 5 C. S. Lee and M. M. Wu, J. Chem. SOC., Chem. Commun., 1985, 250. 6 G. A. Olah, G. K. Prakash, R. W. Ellis and J. A. Olah, J. Chem. SOC., Chem. Commun., 1986, 9. 7 C. D. Chang, W. H. Lang and R. L. Smith, J. Catal., 1979, 56, 169. 8 W. 0. Haag, R. M. Lago and P. G. Rodewald, J. Mol. Catal., 1982, 17, 161. 9 C. T-W. Chu and C. D. Chang, J. Catal., 1984,86, 297. 10 M. M. Wu and W. W. Kaeding, J. Catal., 1984, 88, 478. 11 R. L. Espinoza and W. G. B. Mandersloot, J. Mol. Catal., 1984, 24, 127. 12 J. R. Anderson, T. Mole and V. Christov, J. Catal., 1980, 61, 477. 13 Y. Ono and T. Mori, J. Chem. SOC., Faraday Trans. 1, 1981, 77, 2209. 14 R. Hunter and G. J. Hutchings, J. Chem. Soc., Chem. Commun., 1985, 1643. 15 H. G. Howden, CSZR Report C. Eng. (CSIR, Pretoria, South Africa, 1982), no. 413. 16 R. Mihail, S. Straja, G. Marla, G. Musca and G. Pop, Znd. Eng. Chem. Proc. Des. Dev., 1983,22, 532. 17 W. Kirmse, in Carbene Chemistry (Academic Press, New York, 1964). 18 C. D. Chang and C. T-W. Chu, J. Catal., 1982, 74, 203. 19 T. Mole and J. A. Whiteside, J. Catal., 1982, 75, 284. 20 T. Mole, J. Catal., 1983, 84, 423. 21 P. Rimmelin, H. Taghavi and J. Sommer, J. Chem. Sac., Chem. Commun., 1984, 1210. 22 R. Hunter and G. J. Hutchings, J. Chem. SOC., Chem. Commun., 1985, 886. 23 G. Perot, F. X. Cormerais and M. Guisnet, J. Mol. Catal., 1982, 17, 255. W. H. Ip, K. Lammertsma, G. Salem and D. C. Tabar, J. Am. Chem. SOC., 1984, 106, 2143. ed. L. V. C. Rees (Heyden, London, 1981), pp. 649. Paper 5/2158; Received 9th December, 1985

ISSN:0300-9599    

DOI:10.1039/F19878300571

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. I, 1987,83, 585-590 The Liquid-Vapour Transition in Monolayers of n-Pentadecanoic Acid at the Air/Water Interface Norman R. Pallast and Brian A. Pethica*$ Departments of Physics and Chemistry, Clarkson University, Potsdam, New York 13676, U.S.A. Using recently developed methods for surface manometry, the first-order two-dimensional phase transition from vapour to liquid in monolayers of ultra-pure n-pentadecanoic acid at the interface between air and aqueous mol dm-3 HCl has been studied at temperatures from the triple point (17 "C) to 40 "C. Contrary to previous reports, the critical temperature is well above 40°C. It was not possible with the apparatus and methods developed to date to determine the critical temperature, which is probably in the range of 5G60 "C.The latent heat of the phase transition was obtained as a function of temperature. The phase transitions in monolayers of n-pentadecanoic acid have been studied in detail as part of a program to establish definitive criteria for surface manometry with insoluble monolayers at the air/water interface. Results published to date have established the equilibrium spreading pressures' and the characteristics of the liquid-solid (l/s) and liquid-vapour (l/v) transitions for n-pentadecanoic acid monolayers at 25 "C on aqueous 1 OP2 mol dmP3 HC1.2. The spreading transitions for n-hexadecanoic and n-tetradecanoic acids on mol dm-3 HCl have also been described,l and the thermodynamically predicted effects of electric fields applied normal to the interface in two-phase and single-phase regions of the isotherms for n-pentadecanoic and n-octadecanoic acids have been demonstrated.* These results have accounted for published variations in equilibrium spreading pressures and have demonstrated that the well known degenerate form of the so-called liquid-expanded to liquid-condensed transition is the result of a number of experimental artefact^.^ This transition is, in fact, simply first-order in behaviour, and will usually correspond to a liquid-solid transition.In this paper we examine the liquid-vapour transition in monolayers of n-penta- decanoic acid on loP2 mol dmP3 HCl as a function of temperature. It will be shown that earlier estimates of the two-dimensional critical temperaturel27 l3 are too low owing to incorrect estimation of the densities of the coexisting phases as the result of incomplete experimental techniques.The new results are used to estimate the enthalpy of the transition. Experimental The techniques for surface manometry and for the purification of hexane, HC1, n-pentacanoic acid and water have been described in detail, as have the methods for controlling humidity and These techniques have proved adequate for monolayer studies up to 40 "C. At higher temperatures we have not so far obtained reliable results because of problems in spreading and difficulties with the behaviour of photocells and related electrical circuitry at high water vapour pressures. Salient features of the experiments presented here are the use of clear silica troughs 7 Present address: Standard Oil of Ohio, Warrensville Research Center, Cleveland, Ohio 44128, U.S.A.1 Present address: Electro-Biology, Inc., Fairfield, New Jersey 07006, U.S.A. 585586 260 240 220 - 200 180 2 160 $ 140 120 I 8 100 d 2 8 0 - 60 40 20 Phase Transitions in n-Pentadecanoic Acid Monolayers I I I cll I 1 1 1 , 1 I ( ! I I 1 1 1 - - - - - - - - - - - - 4 1 1 1 1 1 I I I l l I I I l l I 1 1 1 50 100 500 1000 5000 10000 area/A2 molecule-' Fig. 1. 71 us. A isotherms for monolayers of n-pentadecanoic acid at the air/10-2 mol dm-3 HCl interface in the gaseous and liquid-vapour transition regions at several temperatures: 0, 20; 0 , 2 5 ; 0 , 3 0 ; 0 , 4 0 "C. Single-shot spreadings indicated by ' as in 6. Results by spreading from a crystal at 25 "C and subsequent expansion indicated by filled circles.2 The isotherm at 15 "C (A) is for the solid-vapour transition below the triple point (data from Iwahashi, using the same experimental systems' 5 ) .which were roasted clean after each short series of experiments on a given day, the rigorous purification of water and monolayer substance (99.95 % pure), spreading from n-hexane by single-shot or successive aliquots of fatty-acid solution, and the use of monolayer expansion methods using sliding barriers. The shapes of the isotherms were also confirmed by spreading monolayers from the solid in the absence of solvents followed by expansion using the barriers. Most of the results presented here were obtained by the Wilhelmy method, with glass plates in the absence of waxes or Vaseline.Some were measured with paper plates or by the use of a horizontal float and threads, with excellent agreement in surface pressures between the methods. None of the data were obtained by the commonly used monolayer compression method, which is a major source of large errors at low monolayer pressures. The reproducibility of the monolayer pressure is k 1 pN m-l, and the monolayer areas were reproducible to k 2%, corre- sponding to the precision of the horizontal float method as described previously.2 Results and Discussion The experimental results of the study of the l/v phase transition region for n- pentadecanoic acid monolayers on mol dm-3 HCl at temperatures from 15 to 40 "C are collected in fig. 1. The experimental surface pressures (n) range is below 300 pN m-l, and results were obtained out to 3 .0 ~ lo4 A2 molecule-l. The triple point for n-pentadecanoic acid is close to 17 oC.6-8 The isotherm at 15 "C thus represents a sublimation from the two-dimensional solid to vapour. In all cases the phase transitions are first-order, a conclusion confirmed by the large fluctuations in the surface potentials observed in the transition regions over the entire temperature range.4* The transitionN . R. Pallas and B. A . Pethica 587 Table 1. Experimental results for the liquid- vapour transition in n-pentadecanoic acid mono: layers at the air/lOP2 mol dm-3 HC1 interface, and for the vapour-solid transition at 15 "C (molecular areas in the coexisting phases given in A21 T/"C q/pN m-l A(vapour) A(1iquid) 20 132 1500 41.5 25 162 1300 43.5 30 192 1200 47.2 40 252 850 51.0 15 102 2000 - pressures and the molecular areas for the coexisting liquid and vapour monolayers are shown in table 1.Adam and Jessopg measured the l/v transition pressure at 20 "C, finding a value in fair agreement with the results given here. Their range of constant x was much narrower than shown in fig. 1, probably owing to impurities causing to rise as the dense end of the transition is approached. Hawkins and BenedeklO studied the transition for monolayers on water, which implies that the n-pentadecanoic acid in the monolayer was ionised to an extent depending on the surface density." It seems probable that their measurements were also complicated by artefacts arising from impurities in their system. Kim and Cannell12 examined the transition over a temperature range from 15 to 35 "C, and some features of their data agree well with the results presented here, particularly in the values of the transition pressure at lower temperatures.The densities of the coexisting phases given by Kim and Cannell are, however, different from those shown in fig. 1 and table 1. The range of the region of constant pressure they describe is much narrower than we find, giving high densities to the vapour and low densities to the liquid phases in coexistence. Kim and Cannell also give surface potentials for n-pentadecanoic acid, showing the range of fluctuations observed in the l/v transition region.13 The upper and lower limits of these fluctuations can be used to estimate the highest density of the vapour phase and the lowest density of the liquid phase in coexistence at the one temperature at which surface potentials were measured (20.1 "C).We estimate these limiting densities as 10 x and 150 x molecule as against 16.5 x interpolated from their tabulated values obtained from surface pressure isotherms. Our results at 20 "C are 6.7 x and 240 x molecule for the densities of the coexisting vapour and liquid phases, as indicated by our pressure isotherms and confirmed by surface potentials in our system.* We conclude that the results of Kim and Cannell are affected by residual impurities, particularly in their sample of n-pentadecanoic acid, for which no analysis was given. Their isotherms show a rise in surface pressure as the dense end of the phase transition is approached, much as we have observed for less pure samples than that studied in this paper.2 Because of their incorrect identification of the range of the 1/v transition, Kim and Cannell concluded that the two-dimensional critical temperature (T,) for n-pentadecan- oic acid monolayers on lop2 mol dm-3 HCl is 26.27 "C and went on to calculate several critical exponents from their data, concluding that mean field theory gives a good account of the coexistence envelope for the transition.As will be clear from fig. 1, the critical temperature is much higher than 40 "C, and we have not been able to reach the critical point with the apparatus used in this series of experiments. Attempts to work at 50 "C have given very unsatisfactory results. Another apparatus will be necessary either to protect the photocells and circuitry from the high water vapour pressures or to use and 122 x588 Phase Transitions in n-Pentadecanoic Acid Monolayers I I I I 10 20 30 40 50 60 70 TI" C Fig.2. The heat of vaporisation for the liquid-vapour transition in monolayers of n-pentadecanoic acid at the air/10-2 mol dm-3 HC1 interface as a function of temperature. Note that at 15 "C the monolayer sublimates. The triple point is at ca. 17 "C. alternative manometric methods. Spreading will also need further study at higher temperatures, and even if all these practical problems are solved the monolayer may prove either too soluble or too volatile to give useful results. Over the range 20-40 "C the l/v transition pressure varies very linearly with tempera- ture by 6.0 pN m-' K-l.Kim and Cannel give 6.7 pN m-l K-l over a smaller range of temperatures up to 26 "C. From the two-dimensional form of the Clausius equation for the transition pressure (n,) dz, AH dT - TAA where AA is the difference in molecular area across the transition, the enthalpy of evaporation (AH) for the transition may be calculated, as shown in fig. 2. AH varies linearly with temperature, corresponding to a difference in the constant-pressure heat capacity between the vapour and liquid states of 300 J mol-1 K-l. The linear variation of the transition pressure continues to 15 "C, where the transition is a sublimation. The estimated sublimation heat is shown on the figure. It lies off the line for the heats of the l/v transition, allowing a rough estimate of the heat of the solid-liquid transition at 3.5 kJ mol-l.From the results shown in fig. 2, an extrapolation may be made to zero AH, as would apply at the critical temperature. The temperature so obtained is 71 "C. If the usual pattern of AH variation with temperature near T, applies, then T, will be well below 71 "C. Another estimate of T, can be made by assuming that the critical exponent p for the dependence of the density difference across the transition ( A r ) on the temperature near the critical point is 0.5, as required by mean-field theory,'* and using the data from table 1 to extrapolate to zero Ar'. From the densities at 30 and 40 "C the value of T, so obtained is 51 "C. A value 10 "C or more above the highest temperature for which a reliable experimental isotherm has been measured seems acceptable from a consider- ation of the isothermal compressibilities ( K ) of the two-dimensional vapour at theN .R. Pallas and B. A . Pethica 589 16.0 - I z E $ 14.0 8 ki E s x * .- - .- v) Q 12.0 5 5 .- 10.0 8.0 10 20 30 40 TI" C Fig. 3. Isothermal compressibilities for n-pentadecanoic acid vapour at the transition pressure at several temperatures. Compressibilities estimated graphically as the limiting slope at the transition from the isotherms of fig. 1. The data are insufficient for an estimate at 20 "C. transition pressures, as estimated graphically and shown in fig. 3. Near the critical point K will rise rapidly to infinity. At the experimental temperatures K falls with increasing temperature, with some indication of reaching a minimum.Taking all these leads, T, is probably in the region 50-60 "C, near the melting point (54 "C) for the solid fatty acid and in a difficult range for future experiment. There is little guidance in the data on the applicability of mean field or Ising theory to the transition. We thank Clarkson University for laboratory facilities and support, and Electro-Biology, Inc. for financial assistance in the course of this work. The apparatus was constructed in part with funds from N.I.H. and N.S.F. grants CHE 7827566 and ROIGM25068, gratefully acknowledged. We thank Dr Makio Iwahashi for permission to quote his results for the solid-vapour transition at 15 "C. It is a pleasure to thank Dr David S. Cannell and Dr Mah Won Kim for valued discussions of our results. References 1 M. Iwahashi, N. Maehara, Y. Kaneko, T. Seimiya, S. R. Middleton, N. R. Pallas and B. A. Pethica, J. Chem. SOC., Faraday Trans. I , 1985, 81, 973. 2 S. R. Middleton, M. Iwahashi, N. R. Pallas and B. A. Pethica, Proc. R. SOC. London, Ser. A , 1984,396, 143. 3 N. R. Pallas and B. A. Pethica, Langmuir, 1985, 1, 509.590 Phase Transitions in n-Pentadecanoic Acid Monolayers 4 S. R. Middleton and B. A. Pethica, Faraday Symp. Chem. SOC., 1981, 16, 109. 5 N. R. Pallas and B. A. Pethica, Colloids Surf., 1983, 6, 221. 6 W. D. Harkins, T. Fraser-Young and E. Boyd, J. Chem. Phys., 1940, 8, 954. 7 N. R. Pallas, Doctoral Thesis (Clarkson University, 1983). 8 S. R. Middleton, unpublished results. 9 N. K. Adam and G. Jessop, Proc. R . SOC. London, Ser. A , 1926, 423. 10 G. A. Hawkins and G. B. Benedek, Phys. Rev. Lett., 1974, 32, 524. 11 J. J. Betts and B. A. Pethica, Trans. Faraday SOC., 1956, 52, 151. 12 M. W. Kim and D. S. Cannell, Phys. Rev. A, 1976, 13,411. 13 M. W. Kim and D. S. Cannell, Phys. Rev. A, 1976, 14, 1299. 14 H. E. Stanley, Introduction lo Phase Transitions and Critical Phenomena (Oxford University Press, 15 M. Iwahashi, unpublished results. New York, 1971). Paper 5/2246; Received 20th December, 1985

ISSN:0300-9599    

DOI:10.1039/F19878300585

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. I , 1987, 83, 591-613 Surface Potential Measurements in Pentanol-Sodium Dodecyl Sulphate Micelles Gregory V. Hartland and Franz Grieser Department of Physical Chemistry, University of Melbourne, Parkville 3052, Australia Lee R. White* Department of Mathematics, University of Melbourne, Parkville 3052, Australia The surface potential of sodium dodecyl sulphate (SDS) micelles has been measured by the apparent pK, of a micelle solubilized acid-base indicator. The indicator used was 4-heptadecyl-7-hydroxycoumarin. The surface poten- tials were measured under conditions of changing surface charge density and bulk ionic strength and the results analysed by a Gouy-Chapman-Stern model for the double layer of the micelle. The results indicate that the sodium counterions in the micelle’s Stern layer bind between the sulphate head groups.When pentanol is solubilized by the micelle the experimental potentials are smaller than predicted. The difference is attributed to an increase in the number of sites for counterion binding at the micelle surface. Ionic micelles have long been considered an excellent model system for studying colloidal chemical phenomena as they can provide a reproducible surface of high purity.l? Thus it is of some interest to measure the surface potential of a micellar system, under conditions of changing surface charge density and bulk ionic strength, to establish if surface charge density-surface potential relationships derived for solid colloidal particles can be applied to micelles. Studies of this kind have not been attempted in the past owing to the lack of reliable experimental methods.It is only recently that spectroscopic probe techniques have been developed that enable the measurement of the surface potential with varying charge density of a micelle. The surface charge density of an ionic micelle can be varied by the solubilization of a non-ionic surfactant. Sodium dodecyl sulphate (SDS) micelles were chosen for the present study because they have been well characterized in the literature. The non-ionic surfactant chosen was pentanol and sodium chloride was used to alter the ionic strength. This system is well defined and exhibits none of the experimental problems (such as questionable purity) of solid colloidal systems (e.g. latex particle^).^ Theory Theoretical Double Layer Models The SDS micelles have been considered as monodisperse, spherical aggregates with a free monomer concentration equal to the c.m.c.The assumption that the micelles are spherical, or near1 so, needs some justification. The limiting length of a twelve-carbon chain is 16.7 l: the typical radius of the hydrophobic core of a SDS micelle is 17 A and so by packing constraints it cannot be a However, the surface of a micelle is not perfectly smooth, surface fluctuations occur which decrease the volume of the hydrophobic core.*t9 The average protrusion of 59 1592 Pentanol-SDS Micelles d d c .- 4 +-' 0 Q electrokinetic plane of shear Fig. 1. Charge separation at the surface of an anionic micelle. monomers from the micelle surface, calculated by Anianssorqs is about one CH, group.The protrusion of monomers will be varied so that those that do not protrude may now reach to the centre of the micelle. This means that the deviation in the time-averaged shape from spherical may not be great, even for quite large micelles.s The model chosen for the present study to represent the electrical double layer at the surface of the SDS micelles is the Gouy-Chapman-Stern model, derived for spherical colloidal particles. The charges on the sulphate head groups at the micelle surface are balanced by sodium counterions in a Stern (or inner) layer and a diffuse layer that extends into the aqueous solution. The Stern layer is considered as a relatively immobile layer of adsorbed counterions (see fig.1). The charges from the sulphate head groups are assumed to be spread evenly, in an infinitesimally thin layer, over the surface of the spherical hydrophobic core of the micelle. Similarly, the charges from the counterions in the Stern layer are assumed to be spread evenly, in an infinitesimally thin layer, over the spherical shell formed by the locus of centres of the counterions. The distance between the two layers of charge is p (fig. 1). The double-layer charges and potentials are defined such that o,, and ly,, o6 and tyyg and o d and lyd, are the surface, Stern layer and the diffuse-layer charge densities and potentials, respectively. The zero-potential (c) is the potential at the electrokinetic plane of shear between the micelle and bulk solution (see fig.1). The equations defining the double layer are lyyg = vo-oo/K1 (1) where K , is the inner-layer capacitance: 81 80 Kl = - PG. V. Hartland, F. Grieser and L. R . White 593 where E , is the dielectric constant of the inner layer. 2 8 In [cosh (&/4)] + K r COSh2 ( & / 4 ) ( K r ) 2 sinh2 ( & / 2 ) sinh (&/2) (1 + 2 ~ , co uk T e Cd = where u-l is the Debye length and yd = elyd/kT [see ref. (1 l)]. v/p= vd oo+op+od = 0 oo = -eNs/4nr2 where N, is the number of SDS molecules per micelle, r is the radius of the hydrophobic core, aNa+ is the bulk sodium ion activity and & is the dissociation constant for the reaction : Stern layer bound counterion + vacant site + counterion Kd and omax is the maximum Stern layer charge density. omax equals the charge on the counterion multiplied by the total number of sites per unit area.Eqn (7) represents Langmuir isotherm control of the Stern layer ions.l0 The models of the Stern layer have been considered. (I) Site Binding Model i.e. the sodium counterions bind to the sulphate head groups of the micelle.12 omax = - 00 (8) (11) Classic Stern Layer omax = e / A ~ a + (9) where ANzt+ is the cross-sectional area of a sodium ion, i.e. there are no specific sites for binding at the micelle surface. The set of equations can be solved iteratively for v0 and v/8 given that oo, K,, Kd and the bulk sodium ion concentration are known. By equating the experimental potential to the surface potential, Kd can be calculated given that oo, K , and the bulk ionic strength are known.Further elaboration of the double-layer model to include a second compact layer and discreteness of charge effects is unnecessary for the present calculations and probably pointless because of the dynamic nature of the micelle surface.** For this reason we have used the simplified picture of the double layer as described by eqn (4). It should be noted that eqn (3) and (7) only work at moderately high ionic strengths where the particles are non-interacting. Spectroscopic Probe Methods Surface Potential Measurements The surface potential of SDS micelles can be measured from the pKa of a pH indicator solubilized at the micelle surface.13 In the absence of specific molecular interactions and interfacial salt effects, the equilibrium of an indicator at the surface of a charged micelle will be affected by the microenvironment it experiences, i.e.the reduced effective dielectric constant of the micelle surface and the change of proton activity at the micelle-water594 Pentanol-SDS Micelles interface, owing to the surface potential of the mice1le.l The ‘apparent’ pK of an indicator solubilized by a micelle can be split into an ‘intrinsic’ pK (due only to the local environment of the micelle surface) and a part due to the surface potential of the (10) micelle : where pK, is the ‘apparent’ pK of the indicator, pKi is the ‘intrinsic’ pK of the indicator and ry, is the surface potential of the micelle. The pK, of the indicator can be determined by the Henderson-Hasselbalch equation: (1 1) where a, the degree of ionization of the indicator, can be measured spectroscopica11y.l A lipoidal pH indicator was used to eliminate the problem of partitioning of the indicator between the micelle interface and bulk l5 The indicator chosen for the present study was 4-heptadecyl-7-hydroxycoumarin (HC).This indicator was chosen because its degree ofionization can be measured by either U.V. absorption or fluorescence16 and the pKi of the indicator in an SDS micelle can be determined using a non-ionic polyoxyethylene surfactant micelle as a neutral reference state,l77 l8 thus PK, = pKi - Fy/,/2.3 RT pH = PKa + log [a/( 1 - a)] where pKZ is the pK, of the indicator in the non-ionic micelle. The non-ionic surfactant used was n-dodecyl octaxyethylene glycol monoether (C12 E8). This treatment presumes that the pKi for HC in SDS micelles does not vary with the addition of NaCl or pentanol.The potential measured by the indicator (tyexpt) is the potential at the position of the ionizable hydroxy group of the HC indicator. ryexpt is considered to be equal to ry,. Surface Charge Density Calculations The solubilization of pentanol molecules into SDS micelles reduces the surface charge density of the micelle because the molecules act as spacers between the SDS head groups. 19-21 The surface charge density has been calculated from the number of SDS and pentanol molecules per micelle ( N , and N,, respectively) and the known volumes of the hydrocarbon chains. cr, = -Nse/4.n[3(N, V,+N, <)/4n]z (14) where yS is the volume of an SDS hydrocarbon chain (357 A3 4 ? 19) and & is the volume of a pentanol hydrocarbon chain (161 A3 4 9 19).In the present work the number of SDS molecules per micelle was either taken from the literature or measured by a fluorescence-quenching technique (following section). To determine the number of pentanol molecules per micelle another fluorescence-quenching technique was used to measure the partitioning of pentanol between the micelles and bulk solution. Measurement of the Number of SDS Molecules per Micelle A full description of this technique is given by Turro and YektaZ2 and more recently by Almgren and L o f r ~ t h . ~ ~ The technique relies upon the quenching of Ru(bipy)i+ fluorescence by 9-methylanthracene (9-MeA). Assuming that the fluorescent probe [Ru(bipy):+] and quencher (9-MeA) are both entirely partitioned into the micellar phase, that fluorescence is only observed fromG .V. Hartland, F. Grieser and L. R . White 595 micelles containing no quencher and that a Poisson distribution of probe and quencher molecules exists, among the micelles, then NJQl ln(Io/I) = ( C - c.m.c.) where Io is the fluorescence intensity of Ru(bipy)g+ with no added quencher, I is the fluorescence intensity with the quencher concentration equal to [Q] and (C-c.m.c.) is the concentration of micellized surfactant (C is the total surfactant concentration and the c.m.c. is assumed equal to the free monomer concentration). Thus by measuring the relative fluorescence intensity of a series of solutions with different 9-MeA concentrations but a fixed Ru(bipy)g+ concentration the number of SDS molecules per micelle can be determined by plotting ln(Io/I) us.[Q]. Measurement of the Partitioning of Pentanol in SDS Micellar Solutions A fluorescent probe technique for the measurement of the partitioning of alcohols in SDS micellar solutions has recently been published by Abuin and The technique is based on observing the decrease in fluorescence from the probe, solubilized in the SDS micelles, when alcohol is added. Two probes were used to measure the partitioning of pentanol : pyrene (as used by Abuin and L i s ~ i ~ ~ ) and 1 -pyrenecarboxalde- hyde. Both probes gave equivalent results (within experimental error); however, 1 - pyrenecarboxaldehyde is easier to use owing to its less structured fluorescence. The equations (first derived by Hayase and Hayano26) relating Xa the mole fraction of alcohol in the micelle phase, Y, the mole fraction of alcohol in the aqueous phase, C, the total concentration of alcohol and Kp the partition coefficient are Xa = B/(C-c.m.c.+B) Y, = 18 A/(1000-#) Kp = xa/Ya (18) where A is the fraction of alcohol in the aqueous phase, B is the fraction of alcohol in the micelle phase and # is the volume of the micelle phase in a litre of Combining eqn (1 6)-( 18) gives eqn (1 9) :24 The assumption is made that the relative fluorescence intensity of the probe is determined by the mole fraction of alcohol in the micelle phase.To measure Xa and Kp the relative fluorescence intensity ( I o / I ) us. Ca at different surfactant concentrations is measured (fig. 2). A set of Ca values with the same Xa (and hence K,) can be obtained as shown.Plotting Ca us. (C-c.m.c.) (fig. 3) and using eqn (19) allows the evaluation of Xa (from the slope of the line) and Kp (from the intercept). From these data a graph of Xa us. Ca can be obtained for the different surfactant concentrations. The number of alcohol molecules per micelle can be determined from Xa and N,: Na = - xa N,. 1 -x, Experimental Reagents SDS (Specially pure, BDH) was recrystallized twice from an ethanol-water mixture (95: 5 , v:v), washed with anhydrous ether (analytical, May and Baker) and dried596 Pent anol-SD S Micelles 2.05 total pentanol concentration Fig. 2. The relative fluorescence intensity I o / I of 1 -pyrenecarboxaldehyde us. the total pentanol concentration for three SDS concentrations (0.10 mol dm-3 NaCl).0, 0.02 mol dm-3; x , 0.05 mol dm-3; @, 0.10 mol dm-3. 0.26 0.301 0.22 0.18 0.14 0.10 0.06 0.02 t / 0.02 0.04 0.06 0.08 0.10 C-c.m.c. Fig. 3. The values of C, obtained at a constant I o / I plotted against C-c.m.c.G. V . Hartland, F. Grieser and L. R. White 597 under a vacuum. The recrystallizations were performed to remove any dodecanol. Surface tension measurements gave a c.m.c. = 6.1 x lop3 mol dmP3 (literature value = 8.3 x lod3 mol dm-3 2 7 9 ,*) with a small 1 to 2 mN m-l minimum. The low c.m.c. and minimum were attributed to C14 and c16 sulphates present in the sample (0.5 and 0.05 % , re~pectively).~~? 30 Octaoxythylene glycol monoether (C12 E8) was obtained from the Nikko Chemical Co. and was found to have a purity > 99% by gas chr~matography.~~ 4-Heptadecyl- 7-hydroxycoumarin (HC) was obtained from the Victorian Institute of Drug Technology and was used without further purification. 9-Methylanthracene (9-MeA), pyrene, pentan- 1-01 (Tokyo Kasei), Ru(bipy),Cl, (Pfaltz and Bauer) and 1 -pyrenecarboxalde- hyde (Py-CHO) (Aldrich) were all high-purity grades and used without further purific- ation.Pyrene and Py-CHO solutions were made using ethanol (Spectrosol, Ajax), Ru(bipy),Cl, solutions were made using Millipore filtered water (conductivity < 1 x SZ-l cm-l and surface tension = 72.0 mN m-l at 25 OC31) and 9-MeA solu- tions were made using methanol (Spectrosol, Ajax). The pH was adjusted by NaOH (analytical, Ajax) or HCl (analytical, May and Baker).NaCl (analytical, Ajax) was used to change the ionic strength. All surfactant solutions were prepared on the day of use with Millipore filtered water and thermostatted at 25 "C during measurement. Fluorescence measurements were performed on a Perkin-Elmer MPF-44A spectro- fluorimeter. U.V. absorbance measurements were performed on a Varian Carey model 2 10 spectrophotometer. Surface Potential Measurements The concentration of 4-heptadecyl-7-hydroxycoumarin (HC) used was ca. 1 0-5 mol dm-3. The pH of the solutions was measured using a glass electrode (Titron, type A) and a double-junction reference electrode (Titron, calomel electrode). The inner compartment of the reference electrode was filled with saturated KCl and the other compartment with NaCl solution. (The concentration was adjusted to match the ionic strength of the solution being studied.) The pH was recorded on a Radiometer PHM84 research pH meter and the electrodes were standardized at the start of each day using pH 4 and 9 buffer solutions (obtained from either May and Baker, BDH or Merck).Solutions were titrated from high to low pH by the addition of small aliquots of concentrated NaOH or HCl solutions. After each addition the pH of the solution and degree of ionization of the indicator was measured. The degree of ionization was measured by either U.V. absorption (scanning around 365 nm) or fluorescence (at an excitation wavelength of 367 nm and scanning around 450 nm). Using the above procedure, the ionic strength, I, of the bulk solution was fixed at I = 10(PHi-14) +c.m.c.+ [NaCl] (21) where pHi is the initial pH and [NaCl] is the concentration of added NaC1. Eqn (21) neglects the contribution to the ionic strength from the micelles and the dissociated counterions. At high concentrations of added NaCl this contribution is insignificant and eqn (21) can be used for the ionic strength. However, this is not true at low NaCl concentrations and a correction for the micelles and dissociated counterions must be made. It is hoped that this correction will be made in a future paper. The surface potentials obtained were found to be independent of the technique (u.v. absorption or fluorescence) used to measure the degree of ionization.Pen tanol-SDS Micelles Aggregation Number Measurements Stock solutions were prepared with the required concentration of SDS, salt and pentanol and a Ru(bipy),Cl, concentration of 2 x lo5 mol dm-3.Four solutions with 9-MeA concentrations of (1,2,3 and 4) x loe4 mol dm-3 were then made from the stock solution. The intensity of Ru(bipy)i+ fluorescence with and without quencher added was measured using an excitation wavelength of 415.5 nm and scanning around 640 nm. Partitioning Measurements Stock solutions with SDS concentrations of 0.02, 0.05 and 0.10 mol dm-3, any added NaCl and a probe concentration of mol dma3 were prepared. From each stock solution, solutions with different pentanol concentrations were made, usually six concentrations in the range 0-0.3 mol dm-3. For each SDS concentration the fluor- escence quantum yield of the probe in the solutions with pentanol was measured relative to the solution with no pentanol.For pyrene the area under the fluorescence band was measured, using an excitation wavelength of 340nm and scanning the complete spectrum, i.e. from 350 nm to ca. 500 nm. For Py-CHO the intensity of the fluorescence peak was measured, using an excitation wavelength of 360 nm and scanning around 400 nm. Results and Discussion Before discussing the experimental results it is important to define certain aspects of the micelle and the location of the probe in the micelle. To calculate the surface charge density of the micelle the radius of the hydrophobic core is needed. Owing to shape fluctuations of the micelle surface the radius is not known precisely.*Y9 The effect of changing the radius on the surface charge density and potential is shown in fig.4 for two ionic strengths. An error of 1 A in calculating the radius will give a 5% error in the surface charge density. The corresponding error in the surface potential is minimal and since the surface potential is the major focus of this study, any error in calculating the radius is insignificant. The exact position of the HC indicator with regard to the micelle surface is not known. If the indicator sits outside the micelle surface then it will sense a potential lower than the surface potential. N.m.r. spectroscopic evidence from aromatic molecules solubilized in SDS micelles indicates preferential solubilization at the micelle inte~face.~, 4- Octadecyl- 1 -naphthoic acid, an indicator with a similar size and hydrophobicity to HC, has also been shown by n.m.r.spectroscopy to sit at the surface of SDS micelles and at the interface between the hydrophobic core and ethylene oxide head group region in C,, E, micelle~.~~ In addition solubilization studies of neutral arenes in ionic micelles show preferential solubilization at the micelle surface.34 From the n.m.r. spectroscopic evidence and molecular models it can be shown that the ionizable hydroxy group on the HC indicator sits in the plane of the sulphate head groups in the micelle. The HC indicator will, therefore, respond to a surface potential. The Surface Potential of SDS Micelles with Added NaCl The radius of the hydrophobic core, surface charge density and experimental potentials for 0.02 mol dm-3 SDS micellar solutions at different ionic strengths are shown inG.V. Hartland, F. Grieser and L. R. White 599 C t % 0.165 .r( Y Fig. 17 18 19 20 radius/A 4. The magnitude of the surface charge density and surface potential us. radius micelle at I = 0.01 mol dm-3 ( N , = 64) (-) and I = 0.10 mol drn+ (Ns = 75) for an SDS (---). Table 1. Radius, surface charge density and experimental potential of 0.02 mol dm+ SDS at different ionic strengths 0.007 0.020 0.025 0.065 0.102 0.202 0.302 0.382 0.48 1 64 65 66 70 75 87 101 112 129 17.6f 1 17.7f 1 17.8 f 1 18.1 f 1 18.6+_ 1 19.5 f 1 20.5 f 1 21.2f 1 22.2 f 1 - 0.263 f 0.005 - 0.265 f 0.005 - 0.266 & 0.005 -0.271 f0.005 - 0.278 f 0.005 - 0.292 f 0.005 - 0.307 f 0.005 - 0,.3 17 f 0.005 - 0.333 f 0.005 - 141 + 5 - 125f5 - 122f5 -1lOf5 -95f5 -85f5 -73f5 -72+_5 -67f5 a Taken from ref.(22). table 1. The potentials were calculated using eqn (12) based on the pK, of HC in 5 x lop3 mol dm-3 C,, E,, which was found to be 9.10+0.05 (pKi). This value was unaffected by the presence of up to 4 mol dm-3 NaCl or 0.2 mol dm-3 pentanol. The values of 'yexpt and 'yo with no Stern layer [Le. Kl = Kd = 00 and so tp0 = 'yg = 'yd, see eqn (1)-(7) and fig. I] are shown in fig. 5 . The calculated 'yo values are much higher than 'yexpt. The inclusion of a Stern layer with ion binding introduces two adjustable parameters, Kl [the inner-layer capacitance, eqn (1) and (2)] and Kd (the ion dissociation constant). Decreasing Kd decreases the magnitude of 'yo and 'yg and increasing Kl decreases the difference between i,u0 and 'yg.By equating 'yexpt to 'yo the Kd needed to give charge neutrality [eqn (5)] for a particular Kl can be calculated: The calculated Kd values do not change significantly as the ionic strength increases. This is not surprising; the addition600 Pentanol-SDS Micelles -45.0 - 15.0 -2.1 -1.9 -1.7 -1.5 -1.3 -1.1 -0.9 -0.7 -0.5 -0.3 log (ionic strength) Fig. 5. The experimental potentials (0) and calculated surface potentials (-) (for no Stern layer) DS. log (ionic strength). Table 2. The zeta-potentials of SDS micelles at different sodium chloride concentrations mobility I/mol dm-3 Nsa 4 / 1 0 - ~ cm2 s-' V-l c/mvc 0.008 64 0.5 4.55 -110 0.033 67 1.1 3.84 -91 0.101 75 1.9 3.42 - 77 a Taken from ref. (22). Taken from ref. (2). Calculated using the numerical results of ref.(35). of NaCl to the micellar solution would not be expected to change the chemical binding of the sodium counter-ions to the micelle surface. By averaging the values of Kd for a particular Kl over the whole ionic strength range, a line of best fit to the experimental results can be obtained. The zeta potentials of SDS micelles at different NaCl concen- trations have been calculated from mobilities measured by Stigter and Mysels2 using data obtained by Ottewill and S h a ~ . ~ ~ The zeta potentials are recorded in table 2 and plotted against ionic strength in fig. 6. If the electrokinetic plane of shear is considered to be at the Stern layer (i.e. c = yP, see fig. l), then the inner-layer capacitance needed to give the correct potential drop is Kl = 8.5 F m+.However, the electrokinetic plane of shear may not occur at the Stern layer but at a position further from the micelle surface, in which case c < yP < ly,. This implies that the value of the inner-layer capacitance calculated above is a lower limit. A typical inner-layer capacitance of a solid colloidal particle is Kl z 1 F m-2.36 The average dissociation constants (Kd) for Kl = 10, 20 and 00 F m-2 are shown in table 3 for both models of the Stern layer. Plots of yo and yp, obtained from the average values of Kd, us. log1 are shown in fig. 7 for model (I). Model (11) (the classical Stern-layer model) gives very similar results. The present study will concentrate on model (I) (the site-binding model) because it contains more detail about the micelle surface.G .V . Hartland, F. Grieser and L. R . White 60 1 -150.0 -130.0 - 1 10.0 > E B -90.0 % 1 - + - 70.0 a - 50.0 - 30.0 - 10.0 O O A 0 0 0 A 0 0 0 -2.1 -1.9 -1.7 -1.5 -1.3 -1.1 -0.9 -0.7 -0.5 -0.3 log (ionic strength) Fig. 6. The experimental potentials (0) and zeta potentials (A) us. log (ionic strength). Table 3. Best-fit values of binding constant K,(mol dm-3) for different values of the inner layer capacitance Kl (F m-2) for models (I) and (11) model 10 20 a3 I 0.47 1.30 3.92 I1 0.80 1.92 5.36 The magnitude of wexpt at low ionic strength is smaller than predicted. This is because the ionic strength given by eqn (21) is smaller than the actual ionic strength at low NaCl concentrations owing to the contribution from the micelles and the dissociated counterions.As stated earlier, it is hoped that a correction for this will be made in a future paper. The large values of Kl needed to explain the experimental data imply that the distance separating the Stern-layer charge density and surface charge density is very small, i.e. the sodium counterions sit amongst (or between) the sulphate head groups. This agrees with the ‘rough’ micelle surface proposed by Stigter and Mysels2 to explain their zeta-potential measurements. Fluctuations at the micelle surface may cause the average positions of the Stern-layer counterions and micelle head groups to be close together. An alternative explanation of the experimental results is that the probe senses a potential some distance from the micelle surface.For no Stern layer the surface potential and the potentials at distances of 1, 2 and 4 A from the surface us. log1 are shown in The experimental potentials can be ex lained by assuming that the ionizable group realistic distance. However, because of work done on solid colloidal particles,1° it is fig. 8. of the HC indicator sits between 1 and 2 f from the micelle surface, which is a physically 21 F A R 1602 Pentanol-SDS Micelles - -12;;- 150 90 - 60 / .r( Y B -90- a -60- I I I 1 I 1 I I 1 -2.0 -1.8 -1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 log (ionic strength) Fig. 7. The calculated surface (0) and Stern layer (-) potentials for different values of the inner-layer capacitance Kl and the best-fit values of the binding constant K d (see table 3), model (I).(a) Kl = 10 F m-2, K d = 0.47 mol dm-3; (6) Kl = 20 F m-2, Kd = 130 mol dmW3; (c) Kl = 00 F m-2, K d = 3.92 mol dm-3. reasonable to suggest that micelles have a Stern layer (as proposed by Stigter2f 379 38). The inclusion of a Stern layer lowers the magnitude of yo and hence lowers the magnitude of the potential at a fixed distance from the surface. The experimental potentials would now be expected to lie between the surface potential and the potential 1 A from the surface, i.e. if the probe does sit out from the surface the distance must be less than 1 A. This indicates that there is no real error involved in considering the HC indicator as measuring the surface potential, as was implied by n.m.r. spectroscopy and molecular model evidence. Fluctuations at the micelle surface will cause the charge from the sulphate head groups to be spread over a certain region.89 The timescale of these fluctuations is 10-lo s , ~ whereas the timescale of the HC fluorescence is s.The HC indicator will, therefore, sense some ' average' surface charge density. Aniansson8 has shown that allowing for the distribution of head groups away from the micelle surface gives a lower effective surface charge density than calculated for a static micelle. The reduced effective surface charge density may mean the average values of Kd obtained are too low; however, the magnitude and trends in K d should be correct.G. V. Hartland, F. Grieser and L. R. White 603 t -225.0 0 I 1 I I I I I 1 I -2.1 -1.9 -1.7 -1.5 -1.3 -1.1 -0.9 -0.7 -0.5 - 0 . 3 log (ionic strength) Fig.8. The calculated surface potential (a), and the potentials at distances of (b) 1, ( c ) 2 and (d) 4 A from the surface and experimental potentials as a function of log (ionic strength) (no Stern layer). Table 4. The degree of dissociation of the sulphate head groups for different ionic strengths for models (I) and (11) and Kl = GO Z/mol dm-3 model I model I1 0.007 0.020 0.025 0.065 0.102 0.202 0.302 0.382 0.48 1 0.55 0.55 0.55 0.55 0.54 0.52 0.50 0.49 0.47 0.54 0.54 0.53 0.53 0.52 0.51 0.50 0.50 0.49 If the assumption is made that only the ions in the Stern layer can be considered bound to the micelle, then the degree of dissociation (a) of the sulphate head groups is (22) The degree of dissociation with changing ionic strength is shown in table 4 for both models of the Stern layer and Kl = 00.There is a slight decrease in a as the ionic strength increases, i.e. a larger number of counterions are present in the Stern layer at high ionic strength. Choosing a finite Kl would mean a decreased degree of dissociation. The value of a z 0.5 agrees with calculations performed by Stigter on the number of counterions in the micelle Stern layer. 38 The degree of dissociation, calculated for Kl = co, is cc)nsiderably higher than the values calculated from micelle mobility measurements, usually a = 0.3 to 0.4.2 Both these a = 1 - lq/a,l. 21-2604 2.90 2.80 2.70 2.60 ?5 on 2 2.50 2.40 2.30 2.20 Pentanol-SDS Micelles - - - - - - - - 0.05 0.15 0.25 0.35 0.45 0.55 0.65 0.75 0.85 xi4 Fig. 9. Log (K,) as a function of X,, the literature values are marked by a solid line.A, No NaCl; 0, 0.1 mol dm-3 NaC1. are higher than the values obtained from sodium ion-specific electrode measurements which are usually a = 0.20 to 0.25,39-41 which is comparable to the value extracted from a thermodynamic treatment42 of micelle formation based on c.m.c. data. The differences in the degree of dissociation obtained from the different methods can be explained by considering what each technique measures. The pH indicator technique of this work only includes ions actually in the micelles Stern layer as bound to the micelle, whereas micelle mobility measurements include all ions within the electrokinetic plane of shear between the micelle and bulk solution (see fig. 1) as bound. Not surprisingly, mobility measurements give lower values for the degree of dissociation.The sodium ion-specific electrode measures the bulk sodium-ion concentration. Hence, a measure- ments from a sodium ion-specific electrode will include all those ions within the electrokinetic plane of shear and some of the ions in the diffuse layer as bound, thus giving an even lower value for the degree of dissociation. The same considerations also apply to a obtained from the c.m.c. The stability of the micelle structure depends on the partial neutralization of the sulphate surface charge. This occurs by sodium ions being bound ‘next to’ the sulphate head groups, as well as ions in the diffuse region of the double layer. Therefore the value of a determined from c.m.c. data will again be smaller than the a we have calculated, which only considers the Na+ ions ‘directly’ bound to the sulphate charges.The Surface Potential of SDS Micelles with added Pentanol To calculate vexpt of SDS micelles with added pentanol, the pKi of the HC indicator was assumed to be 9.10 regardless of Xa (the mole fraction of pentanol in the micelle). The partitioning of pentanol into SDS micelles (with no added NaCl) was determined by averaging results obtained from pyrene and 1 -pyrenecarboxaldehyde fluorescence intensity experiments. Graphs of log Kp us. Xa and Xa us. Ca for 0.02 mol dm-3 SDS and 0.05 mol dm-3 SDS are shown in fig. 9 and 10. The values of Kp obtained agree with the values determined in the literature at low0.65 0.55 0.45 xa 0.35 0.25 0.15 0.05 G .V . Hartland, F. Grieser and L. R . White 605 1 I I 0.02 0.06 0.10 0.14 0.18 0.22 0.26 0.30 0.34 total pentanol concentration Fig. 10. X, as a function of C, for (A) 0.02 mol dm-3 and (0) 0.05 mol dm-3 SDS solutions with no added NaCl. 105.0 c Fig. 11. The number of SDS (filled symbols) and pentanol (unfilled symbols) molecules per micelle plotted against X, for 0.02 mol dm-3 (circles) and 0.05 mol dm-3 (triangles) SDS with no added NaCl. Xa and at s a t ~ r a t i o n . ~ ~ ~ 4 3 9 44 The number of SDS and pentanol molecules per micelle ( N , and N,, respectively) are plotted against Xa in fig. 1 1. The raw data are contained in the Appendix. N, was taken from ref. (21). The radius of the hydrophobic core, the surface charge density and yexpt for 0.02 mol dmV3 SDS and 0.05 mol dm-3 SDS with added pentanol are shown in table 5.606 Pentanol-SDS Micelles Table 5.The radius, surface change density and experimental potential of 0.02 and 0.05 rnol dmP3 SDS with added pentanol and no NaCl XlZ I/mol dm-3 radius/A oo/C mP2 ‘Yexpt 0.38 0.38 0.53 0.66 0.69 0.69 0.10 0.19 0.27 0.36 0.43 0.46 0.61 0.61 0.66 0.70 0.006 0.005 0.003 0.003 0.002 0.00 1 0.007 0.008 0.008 0.006 0.008 0.005 0.003 0.002 0.001 0.001 [SDS] = 0.02 mol dm-3 17.8 f 0.2 - 0.209 f 0.007 17.8 f 0.2 - 0.209 f 0.007 16.1 f0.2 - 0.158 f 0.007 16.3 k0.2 - 0.130 f 0.007 16.4 k0.2 - 0.123 f 0.007 16.4f0.2 - 0.123 f 0.007 [SDS] = 0.05 mol dm-3 18.0k0.2 - 0.255 & 0.007 18.2k0.2 -0.246 f0.007 18.3 k 0.2 - 0.235 f 0.007 18.5 k0.2 - 0.221 f 0.007 18.5 f 0.2 - 0.208 f 0.007 18.5f0.2 - 0.200 f 0.007 17.5 f 0.2 - 0.154 f 0.007 17.5 f 0.2 - 0.154 & 0.007 17.0 f 0.2 - 0.136 f 0.007 18.5 f 0.2 - 0.135 f 0.007 -92.0f5 - 102.0 f 5 -83.0+5 - 52.0 f 5 - 37.0 t 5 -43.0 f 5 -115.Of5 - 105.0 f 5 - 105.0 f 5 - 101.0f5 -93.0f5 -85.0f5 - 67.0 f 5 - 70.0 f 5 - 40.0 & 5 -28.0f5 a The c.m.c.values needed to calculate the ionic strength were taken from ref. (26). The experimental potentials are plotted against the mole fraction of pentanol in the micelle phase in fig. 12. No attempt has been made to analyse the data in terms of electrical double-layer theory owing to the correction needed for the ionic strength when no NaCl is present in the system. The Surface Potential of SDS Micelles with added Pentanol and 0.10 mol dm-3 NaCl By adding 0.10 mol dm-3 NaCl the ionic strength of the solution is fixed at a value where the contribution due to the micelles and dissociated counterions can be ignored, i.e.eqn (21) is correct. The partitioning of pentanol in 0.05 mol dm-3 SDS with 0.10 mol dm-3 NaCl was determined using 1-pyrenecarboxaldehyde as the probe. Graphs of log Kp us. Xa and Xa us. Ca are shown in fig. 9 and 13. Comparison of SDS micellar solutions with no added NaCl indicates a slight decrease in partitioning (ca. 6% ), however, this is within experimental error. The number of SDS and pentanol molecules per micelle for 0.05 mol dm-3 SDS with 0.10 mol dm-3 NaCl are plotted against Xa in fig. 14 (the raw data are contained in the Appendix). Comparing N , and N, for 0.05 mol dm-3 SDS with 0.10 mol dm-3 NaCl to 0.05 mol dm-3 SDS with no NaCl (fig.11) shows an increase in micelle size when NaCl is added. The radius of the hydrophobic core, surface charge density and tyexpt are shown in table 6. Experiments performed with a sodium ion-specific electrode show there is only a 6% increase in the sodium ion activity at Xa = 0.5. Hence, the release of sodium ions from the micelle’s double layer when pentanol is solubilized does not significantly affect the bulk ionic strength. The magnitude of tyexpt and ty,, calculated for K , = co and two values of Kd, are plotted against CT, for model (I) [see eqn (1)-(9)] in fig. 15. Kl = co was chosen because it is numerically easy to work with and the results from the first partG. V . Hartland, F. Grieser and L.R . White 607 -150 c 0.05 0.15 0.25 0.35 0.45 0.55 0.65 0.75 0.85 pen tanol mole-frac tion Fig. 12. The experimental potentials us. the mole fraction of pentanol in the micelle phase for (A) 0.02 mol dm-3 and (0) 0.05 mol dm-3 SDS with no added NaC1. 0.75 0.65 0.55 0.45 *a 0.35 0.25 0.15 0.05 0.02 0.06 0.10 0.14 0.18 0.22 0.26 0.30 0.34 total pentanol concentration f error on each point Fig. 13. X, as a function of the pentanol concentration for 0.05 mol dm-3 SDS and 0.10 mol dm-3 NaCl. of this section indicate that a large value of Kl may give the best description of the Stern layer. Using a finite Kl will not change the trend of the calculated potentials. The values of Kd chosen were Kd = 10 and the average Kd value obtained from the ionic strength data. The magnitude of yYexpt and yo start within experimental error of each other and show608 105.0 91.0 Pen tanol-SD S Micelles - - 63.0 D 49.0- 35.0 21.0 7.0- Table 6.The radius, surface charge density and experimental potential of 0.05 mol dm-3 SDS with 0.10 mol dmW3 NaCl and added pentanol - - P - / / / 10 / / each error point On -E - P’ ,,/’ /’I I I I 1 I I I I xa I/mol dm-3 a radius/A a,/C m-* rye*,t/mV - 0.15 0.26 0.34 0.40 0.44 0.47 0.50 0.54 0.57 0.60 0.103 0.104 0.103 0.103 0.103 0.103 0.103 0.104 0.104 0.103 0.103 19.4f0.2 19.3 f0.2 19.7f0.2 20.2 f 0.2 20.2 f 0.2 20.2 & 0.2 19.5 & 0.2 19.3 f 0.2 19.1 f 0.2 18.9 k0.2 19.0 f 0.2 - 0.29 1 f 0.007 - 0.267 f 0.007 -0.255 f 0.007 - 0.246 f 0.007 - 0.233 f 0.007 - 0.222 f 0.007 - 0.207 f 0.007 - 0.199 & 0.007 - 0.188 f 0.007 - 0.176 f 0.007 - 0.170 f 0.007 - 83.0 f 5 - 78.0 & 5 - 72.0 f 5 - 64.0 f 5 - 57.0 f 5 - 51.0f 5 -45.0f 5 -45.0&5 - 37.0 f 5 - 30.0 & 5 -25.0f5 a The contribution of the c.m.c.can be neglected at 0.10 rnol dm-3 NaCl (especially in the presence of added pentanol). the same trend until Xa z 0.25; after Xa = 0.25, tyexpt drops far more quickly than tyo. The same result is obtained for model (11). The possibility that the deviation between the experimental and calculated potentials arises from a change in the interfacial dielectric constant with increasing solubilized pentanol can be excluded based on results from a separate In that particular study it was found that the absorbance maximum of the HC spectrum was sensitive to the dielectric constant of the solvent.In experiments in which pentanol was added to SDS solutions containing the alkyl coumarin probe virtually no shift in the HC absorbance maximum was observed, which implies that there is no large change in the environment of the indicator and hence the interfacial region.G. V. Hartland, F. Grieser and L. R . White -130.0. -110.0 > E -90.0, > E: .r( Y Y -70.0, a -50.0. - 30.0, - 10.0. 609 A A A A a A A A 0.16 0.18 0.20 0.22 0.24 0.26 0.28 0.30 charge density/C rn-? Fig. 15. The experimental potentials (A) and surface potentials (-) plotted against change density. The surface potentials have been calculated from model (I) using Kl = co and Kd = 10 mol dmP3 (a) and 3.92 mol dm-3 (6) (the average value of &, see table 3).Five possible reasons have been considered to explain the difference between the predicted and experimental surface potentials. (1) Solubilization of pentanol increases the chemical binding of the counterions to the micelle surface, i t . Kd decreases. The increase in binding needed to produce the required change in Kd is 2-3 kT. A possible reason is a decrease in the dielectric constant of the micelle surfaces, favouring ion-pair formation. However, there appears to be only a small, if any, decrease in the dielectric constant when pentanol is solubilized, therefore 2 to 3 kT seems to be an unrealistically large change in the binding energy. (2) An increase in the fluctuations at the micelle surface could cause a lower surface charge density than expected and may explain the results. A decrease in the surface tension at the micelle interface would cause an increase in the surface fluctuations of the mi~elle.~ However, the number of molecules per unit area at the micelle surface stays reasonably constant with the addition of pentanol, hence there is not expected to be a large increase in the surface fluctuations.(3) In polydisperse systems the distribution of the probe will be biased towards the large micelles. The large micelles may also contain a higher mole fraction of pentanol and therefore have a lower surface charge density than the smaller micelles. If enough large micelles are present the potential sensed by the probe may be significantly reduced below that expected for a monodisperse system of spherical micelles.The SDS aggregation numbers recorded in this paper were determined by a steady-state method which cannot detect p~lydispersity.~~ However, time-resolved measurements on pentanol-SDS systems indicate that there is little polydispersity in the mixed system.46 Hence polydispersity can be ruled out as a possible explanation for the experimental results. 23 (4) Solubilization of pentanol causes a change in micelle shape. The calculations at ty,, and typ are based on spherical micelles. If the micelle is not spherical it will have a smaller surface charge density and hence the magnitude of yo will be smaller. Allowing for a610 I I I I I I I 0.16 0.18 0.20 0.22 0.24 0.26 0.28 0.30 Pen tanol-SDS Micelles I -150.0 - 130.0 - 110.0 > 5 -90.0 5 - ._ t l -70.0 a - 50.0 - 30.0 - 10.0 0 change of shape to a capped cylinder produces an approximate 2% decrease in B,, which is too small a change to account for the experimental potentials.(5) The last possibility considered in this work is that as Xa increases the number of sites per unit area for counterion binding increases. For omax = -0, [model (I)], binding to the sulphate head groups, the number of sites per unit area decreases as pentanol is added. For om,, = e/ANa+, [model (11)], the number of sites per unit area stays constant. If it is assumed that the sodium counterions bind to the free micelle surface between the sulphate head groups, then cmaX has the form: A-26 N , Omax = ( 45.5 A ). where A is the total surface area of the micelle (which stays reasonably constant), 26 N, is the amount of surface area (A2) occupied by the sulphate head groups (the cross-sectional area of a sulphate head group, determined from molecular models, is 26 A2) and the cross-sectional area of a hydrated sodium ion is 45.4 A2.Note that omax increases as Xa increases. for Kd = 10 and the average value of Kd, us. B,, are shown in fig. 16. The average value of Kd is determined by equating By using the above form of omax, tyexpt can be explained by a single Kd, which indicates there is no change in the chemical binding of the counterion to the micelle surface when Xa increases. The reason why the counterions would bind to spaces between the SDS head groups is that by sitting between the head groups the counterions would benefit electrostatically from the high surface potentials, but not interfere with the hydration of the head groups or the counterions themselves.Sodium-ion binding between the sulphate head groups corresponds to the idea of a ‘rough’ micelle surface, proposed by Stigter and Mysels2 to explain their zeta-potential The magnitudes of tyexpt and yo, calculated using Kl = tyexpt and y o .G. V. Hartland, F. Grieser and L. R . White 61 1 measurements of SDS micelles. Carol and Haydon4’ have also used the idea to explain the zeta-potential of hydrocarbon droplets stabilized by cationic surfactants. The above form of a,,,, however, does not account for the experimental potentials with added NaCl described in this paper. As NaCl is added, the area per head group decreases and, therefore, from eqn (23), the maximum Stern-layer charge density decreases. A constant Kd is not obtained and at high ionic strengths omax is less than the Stern-layer charge density needed for charge neutrality, which is physically impossible.A possible explanation is that there are two sites for counterion binding: (i) binding to the sulphate head groups and (ii) when pentanol is solubilized, producing additional binding between the head groups. Considering the presence of surface fluctuations the sodium counterions would see an average site for binding. The physical picture of the micelle surface that emerges is that with no added pentanol the sodium counterions bind to the sulphate head groups, in the same plane as the head groups. When pentanol is solubilized by the micelle the area between the head groups increases.The increase in area allows more sodium counterions to bind between the head groups and om,, now has a form given by eqn (23). The reason for counterion binding to the micelle is still electrostatic attraction between the sodium ions and the sulphate head groups, but now the total number of sites depends on the area between the head groups. Conclusions The major conclusion of this study is that classical electrical double layer theories give a good description of the behaviour of the surface potential of SDS micelles in the presence of added NaCl. The results indicate that the counterions in the Stern layer bind between the sulphate head groups. When pentanol is solubilized by the micelle the magnitudes of the experimental potentials are lower than predicted using values of Kl and Kd obtained from the ionic strength data.The difference between the predicted and experimental potentials can be explained by assuming that the number of sites for counterion binding in the micelles Stern layer is determined by the area between the sulphate head groups. This result further indicates that the sodium counterions bind between the head groups. This work was supported by the Australian Research Grants Scheme. We thank Dr Derek Chan and Professor T. W. Healy for their comments and suggestions on various aspects of this study. Appendix Raw Data for the Partitioning of Pentanol into SDS Micelles and the Number of SDS and Pentanol Molecules per Micelle Table A 1. Partitioning of pentanol into 0.02 mol dm-3 SDS and 0.05 mol dm-3 SDS solutions (no NaCl) [pen tanol] /mol dm-3 xa Nsa Na [SDSI /mol dme3 - - 65.0 f 2 - 0.02 0.02 0.05 0.38 52.0_+ 1 32.0+ 1 0.02 0.10 0.53 32.0 & 2 37.0 f 2 0.02 0.20 0.66 27.0f 1 52.0 & 2 0.02 0.27 0.69 26.0 & 1 58.0+4 0.05 - - 66.0 & 3 -612 Pen tanol-SDS Micelles Table A 1 (con?.) [SDSI [pentanol] /mol dm-3 /mol dm-3 xa NSa Na 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.0 1 0.03b 0.04b 0.06b 0.08b 0.10 0.20 0.30 0.40 0.10 0.19 0.27 0.36 0.42 0.46 0.61 0.66 0.70 65.0 f 2 62.0 f 2 62.0 f 2 59.0 & 2 56.0 f 2 54.0 f 2 37.0 f.2 31.0f 1 30.0 f 2 8.0& 1 15.0f 1 23.0+ 1 33.0f 1 41.0f 1 46.0 f. 2 58.0 _+ 3 60.0 f 2 84.0 f. 5 ~ ~ ~~ a N, values taken from ref. (21). N, values at these concentrations have been interpolated.Table A 2. Partitioning of pentanol into 0.05 mol dmP3 SDS solutions with 0.10 mol dm-3 NaCl - 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.18 0.22 0.25 - 0.15 0.26 0.34 0.40 0.44 0.47 0.50 0.54 0.57 0.60 86.0 78.0 78.0 79.0 75.0 71 .O 62.0 58.0 54.0 49.0 48.0 - 14.0 27.0 41.0 50.0 57.0 56.0 58.0 63.0 66.0 72.0 References I P. Mukerjee and K. Banerjee, J. Phys. Chem., 1964,68, 3567. 2 D. Stigter and K. J. Mysels, J. Phys. Chem., 1955, 59,45. 3 I. Harding, Ph.D. Thesis (University of Melbourne, 1984). 4 C . Tanford, The Hydrophobic Eflect (Wiley-Interscience, New York, 1973). 5 J. W. Israelachvilli, D. J. Mitchell and B. W. Ninham, J. Chem. SOC., Furuday Trans. 2, 1976,72, 1525. 6 D. J. Mitchell and B. W. Ninham, J. Chem. SOC., Furuday Trans.2, 1981,77, 601. 7 B. Jonsson and H. Wennerstrom, J. Colloid Znterfuce Sci., 1981, 80, 482. 8 G. E. A. Aniansson, J. Phys. Chem., 1978,82, 2805. 9 S. Ljungren and J. C. Eriksson, J. Chem. SOC., Furuduy Trans. 2, 1984,489. 10 G. R. Wiese, R. 0. James, D. E. Yates and T. W. Healy, Electrochemistry (MTP Int. Rev. Sci.) 11 H. Ohshima, T. W. Healy and L. R. White, J. Colloid Znterfuce Sci., 1982, 90, 17. 12 T. W. Healy and L. R. White, Adv. Colloid Interface Sci., 1978, 9, 303. 13 G. S. Hartley and J. W. Roe, Trans. Furuduy SOC., 1940, 36, 101. 14 P. Fromherz, Biochim. Biophys. Acta, 1973, 323, 326. 15 P. Fromherz, Chem. Phys. Lett., 1974, 26, 221. 16 R. F. Chen, Anal. Lett., 1968, 1, 423. 17 P. Fromherz and B. Masters, Biochim. Biophys. Actu, 1974, 356, 270. 18 M. S. Fernandez and P. Fromherz, J. Phys. Chem., 1977,81, 1755. (Butterworths, London, 1976).613 G. V. Hartland, F. Grieser and L. R. White 19 M. Almgren and S. Swarup, J. Colloid Interface Sci., 1983, 91, 256. 20 M. Almgren and S. Swarup, J. Phys. Chem., 1982,86, 4212. 21 F. Grieser, J. Phys. Chem., 1981,85,928. 22 N. J. Turro and A. Yekta, J. Am. Chem. Soc., 1978, 100, 5951. 23 M. Almgren and J. E. Lofroth, J. Colloid Interface Sci., 1981, 81, 486. 24 E. B. Abuin and E. A. Lissi, J. Colloid Interface Sci., 1983, 95, 198. 25 M. V. Encinas and E. A. Lissi, Chem. Phys. Lett., 1982, 91, 55. 26 K. Hayase and S. Hayano, Bull. Chem. Soc. Jpn, 1977,50, 83. 27 P. Mukerjee and K. J. Mysels, Natl Bur. Stand. Ref. Data Ser., 197 1, 36, 5 1. 28 P. H. Elworthy and K. J. Mysels, J. Colloid Interface Sci., 1966, 21, 331. 29 S. P. Harold, J. Colloid Sci., 1960, 15, 280. 30 B.D.H. Pty Ltd, personal communication. 31 C. J. Drummond, G. G. Wan, F. Grieser, B. W. Ninham and D. F. Evans, J. Phys. Chem., 1985,89, 32 K. N. Ganesh, P. Mitra and D. Balasubramanian, J. Phys. Chem., 1982, 86, 279. 33 B. Lovelock, F. Grieser and T. W. Healy, J. Phys. Chem., 1985,89, 501. 34 M. Almgren, F. Grieser and J. K. Thomas, J. Am. Chem. Soc., 1979, 101, 279. 35 R. H. 0. Ottewill and J. N. Shaw, J. Electroanal. Chem. Int. Electrochem., 1972, 37, 133. 36 G. R. Wiese, R. 0. James and T. W. Healy, Discuss. Faraday Soc., 1971, 52, 302. 37 D. Stigter, J. Colloid Interface Sci., 1967, 23, 379. 38 D. Stigter, J. Phys. Chem., 1975, 79, 1008. 39 A. S. C. Lawrence and J. T. Pearson, Trans. Faraday Soc., 1967, 63,495. 40 C. T. Dulin and K. J. Mysels, J. Colloid Sci., 1955, 10, 461. 41 S. G. Cutler, P. Meares and D. G. Hall, J. Chem. Soc., Faraday Trans. 1, 1978,74, 1758. 42 D. G. Hall, J. Chem. Soc., Faraday Trans. 2, 1977, 73, 897. 43 K. Hayase and S. Hayano, J. Colloid Interface Sci., 1978, 63, 446. 44 M. Almgren, F. Grieser and J. K. Thomas, J. Chem. Soc., Faraday Trans. 1, 1979, 75, 1674. 45 C. J. Drummond and F. Grieser, to be published. 46 G. G. Warr, Ph.D. Thesis (University of Melbourne, 1985). 47 B. J. Carol and D. A. Haydon, J. Chem. Soc., Faraday Trans. 1, 1975, 71, 361. 2103. Paper 6/01 1 ; Received 2nd January, 1986

ISSN:0300-9599    

DOI:10.1039/F19878300591

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

J . Chem. SOC., Faraday Trans. I , 1987, 83, 615-626 Identification of the Space Group and Detection of Cationic Ordering in Iron Antimonate using Conventional and Convergent - beam Electron Diffraction Frank J. Berry,*? John G. Holden?$ and Michael H. Loretta$ Department of Chemistry? and Department of Metallurgy and Materials,$ University of Birmingham, P.O. Box 363, Birmingham B15 2TT The structure of rutile-related iron antimonate, FeSbO,, has been deter- mined by electron diffraction. It is shown, in agreement with earlier studies by X-ray diffraction, that the space group is P4,lmnm. However, and in distinct contrast to the conclusions drawn from the X-ray diffraction data and more recent neutron diffraction studies which have described the structure of the compound in terms of the random distribution of iron and antimony over the cationic sites in the oxygen octahedra, it is shown here that the iron and antimony ions are ordered such that c z 9.23 A as opposed to 3.07 A determined by X-ray diffraction. The extent of cationic ordering appears to be influenced by the nature of the cationic oxidation states.The means by which the structure of small, ca. 200 8, diameter, particles of iron antimonate have been determined by electron-diffraction techniques is presented in detail to demonstrate the value of such methods for the structural exami- nation of inorganic solids. We are currently involved in investigations of iron antimonate, FeSbO,, by a range of techniques. In agreement with the results of other studie~,l-~ we confirm that the X-ray diffraction data reveal no evidence for the ordering of the cations. Indeed, our results are consistent with previous descriptions of this compound as a rutile-related solid with tetragonal lattice parameters a = b x 4.63 A and c x 3.07 A, in which the cations are randomly distributed over the oxygen octahedra.This model has recently been endorsed by neutron-diffraction studiesP Transmission electron diffraction is a more suitable technique for the detection of weak superlattice reflections than conventional X-ray diffraction. The superiority of electron diffraction in this particular application derives in part from the better signal-to- background ratio for electron-diffraction patterns than for X-ray diffraction patterns. In view of this the present work has been undertaken using conventional and convergent- beam electron diffraction in an attempt to identify any Occurrence of cationic ordering in iron antimonate. Conventional transmission electron diffraction uses an approximately parallel beam of electrons, so that the diffraction patterns show sharp diffraction maxima which define the directions of strong diffracted waves and hence the spacing of the planes giving rise to these rays.Modern transmission electron microscopes are equipped with facilities for tilting the specimen, typically through k45" about two orthogonal axes, and it is therefore relatively straightforward to obtain successive diffraction patterns correspon- ding to various incident electron beam directions from individual particles.Measure- ments of the angles between these crystallographic directions and the angles between the various diffraction maxima in any one diffraction pattern enable the crystal system to be identified. The reflections on the various patterns can thus be indexed in a self-consistent manner and the magnitude of the various diffracting vectors, Ig*(hkl >I used to obtain the interplanar spacings d{hkZ) giving rise to the diffracted beams.5 The size 615616 Iron Antimonate studied by Electron Diflraction of the unit cell can then be readily calculated. Although it is not easy to obtain additional crystallographic information, such as the point or space group of the sample, from conventional electron diffraction patterns, the technique of convergent-beam electron diffraction allows this additional information to be obtained.Convergent-beam electron diffraction gives rise to discs of diffracted intensity rather than sharp maxima which characterise conventional electron diffraction patterns. The symmetry information contained within these discs together with the symmetry of the high-angle diffraction information, which can appear in the form of a series of diffraction maxima arranged in circles centred on the incident electron beam direction,6 allows the identification of point groups and space groups.' These circles of maxima are first or higher order Laue zones (HOLZ) and correspond to reciprocal lattice points which lie on successive planes of the reciprocal lattice along the electron beam direction. These circular arrays of diffraction maxima therefore provide three-dimensional information, since the radius R of the circles is a measure of this interplanar spacing, H, of the reciprocal lattice through the relation R = (2H/A)&, where A is the wavelength of the electrons and R is the radius of the diffraction circle in reciprocal space.6$ Conventional electron diffraction has been used in this paper to show that iron antimonate is ordered and convergent-beam electron diffraction has been used to determine the point group and space group of this ordered form of FeSbO,.Experimental Iron antimonate was prepared from an antimony-rich precipitate formed by the addition of aqueous ammonia (relative density 0.88) to stirred mixtures of iron(1Ir) nitrate nonahydrate and antimony(v) chloride (relative density 1.42). The iron antimonate A was prepared by calcination of the precipitate at 1000 "C (24 h) in an evacuated sealed silica tube and was shown by 57Fe and lZISb Mossbauer spectroscopy to contain iron(m), iron(@, antimony(v) and antimony(II1). Iron antimonate B was formed by the cal- cination of the precipitate at 1000 "C (96 h) in air and was shown by Mossbauer spectro- scopy to contain only iron(Ir1) and antimony(v).Although the occurrence of two forms of iron antimonate differing in the nature of the cationic oxidation states is not new,8* we shall be reporting on our studies of the conditions under which such materials are formed in a subsequent paper.l0 Samples for examination in the electron microscope were prepared by dispersing 10 mg of each iron antimonate in 15 cm3 n-butanol by ultrasonic vibration and transferring the sample to a glass slide. The butanol was allowed to evaporate and a carbon film evaporated onto the slide. Small portions of this film were cut out, removed by flotation in distilled water, and an individual portion of the film containing a large number of iron antimonate particles collected on a 3 mm diameter copper grid.Electron diffraction and energy-dispersive X-ray analysis (EDX) was carried out using an EM400T transmission electron microscope operating at 100 kV interfaced to an EDAX 9100/60 X-ray system. For the energy-dispersive X-ray analysis (EDX) the samples were placed in a beryllium holder to minimise any artefacts in the measured spectrum and the microscope was operated in the probe mode with only the particle of interest being irradiated with electrons.The counting time used in the analysis was selected to generate sufficient X-ray counts to obtain an accuracy of f 2% in the analysis. Background stripping was performed by using the software available on the EDAX 9100/60 instrument, but quantification of the data was carried out on a separate computer using a program which has been checked against many standard samples. Convergent-beam electron diffraction patterns were taken at a variety of camera lengths and with a range of convergence angles. In order to determine the point group it is usually necessary to obtain the symmetries of diffraction patterns in several electronJ . Chem. SOC., Faraday Trans.1, Vol. 83,part 3 Plate 1. Individual particle of iron antimonate imaged using 100 kV electrons. Plate 1 F. J. Berry, J. G. Holden and M. H. Loretto (Facing p . 616)J . Chem. SOC., Furuday Trans. I , Vol. 83, part 3 Plate 2 Plate 2. Transmission electron diffraction patterns (a)-(c) taken with a small probe (ca. 400 A diameter) and a small convergence angle. F. J. Berry, J. G. Holden and M. H. Lorettot h U m .r( 3 +., .r( $ X I;. J. Berry, J . G . Holden and M . H. Loretto 617 t- i fe K r I 4 6 X-ray energy/keV Fig. 1. Typical energy dispersive X-ray spectrum taken from an individual particle of iron antimonate of the type imaged in plate 1 using 100 kV electrons. The peaks visible correspond to Fe K and Sb L X-rays. beam directions (zone axes), and this requires controlled tilting of the sample about defined tilt axes.This was done by tilting about selected diffraction maxima, and the angles between the zone axes were obtained from the tilt readings on the double-tilt specimen holder. All indexing was done using the convention that the electron beam direction B is taken as upwards from the diffraction pattern as seen on the screen, and all patterns were printed emulsion-side up in order to maintain the sense of indexing carried out during the actual e~periment.~ Results and Discussion All particles which were analysed by EDX gave spectra similar to that shown in fig. 1, which was obtained from a particle of the type imaged in plate 1. The relative intensities of the iron and antimony peaks showed that, within experimental error, the Fe : Sb ratio in all particles was 1 : 1.Determination of the Crystal System and Unit Cell of Iron Antimonate Electron diffraction patterns taken from an individual crystal of iron antimonate are shown in plates 2(a)-(c) together with a schematic diagram in fig. 2, which shows the tilt angles and tilt axes between the three patterns. The X-ray measurements demon- strating the tetragonal nature of the crystal system of iron antimonate were consistent with the electron diffraction pattern shown in plate 2(a), which has four-fold symmetry. It is also relevant to note that patterns exhibiting three fold-symmetry were observed. On this basis the patterns-shown in plates 2(a)-(c) were indexed as [OOl], [I1 11 and [Ol 11 for a tetragonal system with a = b = 4.63 A and c = 9.23 A.The ratio c / a for this crystal is therefore ca. 2.0, which is three times that calculated from X-ray diffraction data. The fact that the value of c is tripled is immediately apparent from the [I1 11 pattern shown in plate 2(b); not only are there reflections corresponding to a c-lattice spacing of 9.23 A, but also the tilt angle of only ca. 35" between [OOl] and [l 1 11 shows that c % a. Note618 Iron Antimonate studied by Electron Diflraction Fig. 2. Schematic diagram showing the angles between the diffraction patterns in plate 2 and the spatial relationship between them. that tilting a crystal of rutile titanium dioxide (c = 2.96 A, a = 4.59 A) through 35" from [00 11 generates a [ 1 131 pattern as shown in plate 3.It is clear from plate 2 that the diffraction maxima which show that the c-lattice spacing is tripled are far weaker than the other maxima. Relatively weak reflections would be expected if the crystal is ordered, with the weak reflections arising from the differences between, rather than the sums of, scattered intensities. In order to describe the structure more fully, convergent-beam electron diffraction has been used to determine the point and space group of iron antimonate. Determination of the Point Group of Iron Antimonate Electron diffraction patterns taken with a larger convergence angle than those used to obtain the diffraction patterns shown in plate 2 reveal symmetry information from which the crystal point group can be obtained.' For example, plate 4 shows convergent-beam diffraction patterns (CBDP) taken with the electron beam direction, B, precisely along [OOl].The important information is present both within the discs in the zero-order Laue zone [plate 4(a)], and in the distribution of diffraction maxima in the high-order zone [plate 4(b)]. From the CBD patterns the symmetry of the zero-order zone and of the high-order zone can be seen to be 4mm. In the case of the zero-order zone this symmetry can be seen on the original negatives more clearly than on prints. No HOLZ lines can be seen in the zero-order disc, and reference to table 1 shows that the observation of 4mm symmetry both in the zero-order and in the whole pattern leads to the conclusion that the possible diffraction groups are 4mm and 4mm1,.Zero-order disc information does not enable a distinction to be made between these two groups (cf. table 1). The possible crystal point groups which show these symmetries are 4mm, 4/mmm or m3m (cf. table 2). Since m3m is not a possible point group for a tetragonal system only the 4mm and 4/mmm point groups need be considered. Distinction between these can be obtained by tilting the sample to, for example, [131], which shows two-fold symmetry in the zero-order layer and no symmetry in the first-order zone [plate 4(c))]. Reference to table 1 shows that the diffraction group which exhibits this behaviour is 2R.9 which is not possible for a point group of 4mm (cf. table 2), and it follows that the point group is 4/mmm. The symmetry observations from plates 4 and 5 and the deductions made from them, which lead to the conclusion that the point group is 4/mmm, are summarised in table 3.J .Chem. SOC., Faraday Trans. I , Vol. 83, part 3 Plate 3 Plate 3. Diffraction patterns taken from a sample of rutile titanium dioxide: (a) [OOl] and (b) [113]. The tilt angle between these patterns was ca. 35" [cf. plates 2(a) and (b)]. F. J. Berry, J. G. Holden and M. H. Loretto (Facing p . 61 8)J . Chern. SOC., Faraday Trans. I , Vol. 83,part 3 Plate 4 Plate 4. Convergent-beam electron diffraction patterns (a) and (b) taken with B = [OOl] and ( c ) [131] from an individual particle of iron antimonate. The [OOl] pattern shows 4mm symmetry when (a) only the zero order of information is considered and (b) the whole pattern symmetry is considered; ( c ) shows no symmetry in the whole pattern.F. J. Berry, J. G. Holden and M. H. LorettoF. J . Berry, J. G. Holden and M. H. Loretto 619 Table 1. Relationship between the observed symmetries in conv- ergent beam diffraction patterns and the 31 diffraction groups which correspond to the 32 different three-dimensional point groups [after ref. (7)] observed symmetries of high-order information symmetry in possible whole pattern zero-order diffraction (zero-order + zone groups high-order) zero-order disc 1 2 m 2mm 4 4mm 3 3m 6 6mm 1 2 1 , 2, 21, mR m mlR 2 m ~ m ~ 2mm 2 ~ m m ~ 2mm 1 4 4R 41R 4mRmR 4R.mmR 4mm 4mm1, 3 31R 3mR 3m 3mlR 6 6 , 61 R 6 m ~ m ~ 6mm 6 ~ m m ~ 6mm 1 1 1 2 1 2 1 m m 2 2mm m 2mm 4 2 4 4 4mm 2mm 4mm 3 3 3 3m 3m 6 3 6 6 6mm 3m 6mm 1 2 2 1 2 m m 2mm 2mm 2mm m 2mm 4 4 4 4mm 4mm 4mm 4mm 3 6 3m 3m 6mm 6 3 6 6mm 6mm 3m 6mm The radius of the high-order zone in [131] corresponds to that expected for the first-order zone for a tetragonal crystal with a = b = 4.63 A and c = 9.23 A.The radius of the high-order zone visible in plate 4(b) can also be used to obtain the apparent spacing of the reciprocal lattice planes along the electron beam direction, which for this pattern corresponds to the spacing along [OOl] (i.e. along the c axis). Measurements carried out on plate 4(b) result, however, in a c-spacing of only 3.07 A rather than the value of 9.23 A derived from the sizes of diffracting vectors in the zero-order zones of plates 2 (b) and (c) and from the HOLZ in 11311. This suggests that the HOLZ visible in plate 4(b) corresponds to the third order rather than the first-order Laue zone.The implications of this will be discussed later.620 Iron Antimonate studied by Electron Difraction Table 2. Relationship between the diffraction groups and crystal point groups7 diffraction groups 6mm1, 3m1, 6mm 6m~m, 61, 6 ~ R M ~ R 3m 6 , 31R 3mR 3 4mmlR 4mm 4RmmR 4mRmR 41R 4R 4 . . . . . . . . . . . . . . . . . . . . . . . . . ' X . . . . . ' X " " " ' X " " " ' " X " " " " ' X " " " " ' ' X " " " " " ' X " " " " " ' ' X " " " " " - x . . . . . . . . . . . . . . . . . . x . . . . . . . . . . ' X ' . . . . . . . . . . . . . . . . . . x . . . . . . . . . . . x . . " X " " " " " ' X ' " ' X " " " ' * " ' X " " * x ' X ' . . . . . . . . . . . . X . " . . . . .. . . . . . . . . . . ' X " . . . . . . . . . . x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x . . . . . . . . . . . . . . . . x . . . . . . . . . . . ' . . . . . . . . . . . . . . . . . . . . . . . . . . . ' X . . . . . . . . . x . . . . . .x . . . . . . . . . . ' X ' X " X . . . . x . . x . . ~ . . . x . . " X . . x . . . x . x . . x ' X " ' - " ' X " ' X ' X " ' X X " ' ' X ' ' X ' . . ' X " X " " ' X X " . . x . . x . . x x . . . - - X ' * x x * x x * x x x * * ' x " x " x x x ' x ' x x ' . . . . x . . . . . . . . . . . . . . x . . . . . . . . . . . . . x . . x . . x . . x . . . x . x . . x . . x . . . x . x . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2mmlR 2 ~ m m ~ 2mm 2 m ~ m ~ ml, m mR 21, 2, 1, . . . . . . . . . . . . . . . . . . . . . . . x . . . . ' X ' X . . . . . . . . ' X " " . . . . . . . . . . . . . . ' X " " ' X X . . . . . . . . . . . 2 1 x ' x x ' x x ' x x ' x x x ~ x ' x x ' x x " x ' x x x ~ x * x x x Table 3. Summary of the symmetries observed in the convergent-beam electron diffraction patterns taken with B = [OOl] and [131] from a sample of iron antimonate and the deductions which can be made from these patterns using tables 1 and 2 observed possible observed possible possible zero-order diffraction symmetry in diffraction tetragonal B symmetry groups whole pattern groups point groups - 4mm 1 , - [0011 4mm 4mRmR 4RmmR 4mm 4mm 4mm 4/mmm - - 4mm - - - 4mm1, 2 2, 1 2, 4/mmm - - - - - - 21 R 2J .Chern. Sac., Faraday Trans. I , Vol. 83,part 3 Plates 5 and 6 Plate 5. Convergent-beam electron diffraction patterns taken (a) with B = [301] and (b) with B = [loll. Note the presence of the 010* reflection in (a) and its absence in (b). The dark line across the 010* reflection in (a) confirms it as a kinematically forbidden reflection which is visible in (a) through double diffraction.Plate 6. Convergent-beam diffraction pattern taken with B = [210] showing that all reflections of the type hkl and hkO* are allowed. F. J. Berry, J. G. Holden and M. H. Loretto (Facing p . 620)J . Chem. Soc., Faraday Trans. I , Vol. 83,part 3 Plates 7 and 8 Plate 7. Convergent-beam diffraction pattern taken with B = [loo] showing that for reflection of the type Okl, [k+l] must be even. Plate 8. Convergent-beam diffraction pattern taken with B = [332] showing that all reflections of the type hkl are allowed. F. J. Berry, J. G. Holden and M. H. LorettoF. J . Berry, J . G. Holden and M. H. Loretto 62 1 Determination of the Space Group of Iron Antimonate The diffraction patterns shown in plate 4(b) can also be used to show that the lattice of iron antimonate is primitive.Thus, projection of the high-order diffraction maxima along [OOl] onto the zero-order layer shows them to project precisely onto the zero-order maxima rather than between them. This shows that the reciprocal lattice is primitive and demonstrates that the lattice of iron antimonate is also primiti~e.~ The number of possible space groups for iron antimonate is, on the above basis, limited to those for which the point group is 4/mmm, and the lattice is primitive. This number can be reduced further by identifying the presence (or absence) of screw and glide axes. These manifest themselves in electron diffraction patterns by the presence of dynamic absences which appear as dark lines or crosses in kinematically forbidden reflection^.^^ ' 9 l1 An example of such a dynamic absence in the 010* reflection with B = [301] is shown [plate 5(a)].This reflection is totally absent when B = [loll [plate 5(b)]. The 010* reflection therefore arises by double diffraction when B = [301] and its absence in [ 1011, together with the fact that there is no mirror perpendicular to [ 1031 which demonstrates that the presence of the 010* reflection when B = [301] cannot be due to the presence of a glide plane, shows that there is a screw axis along [OlO]. On the basis of the above observations there are eight possible space groups12 which may be assigned to iron antimonate. The number of possibilities can be reduced by recognising the conditions defining the kinematically allowed reflections.Examination of electron diffraction patterns taken with B = (210) show that 123* and 243* reflections are present and this, taken together with maxima observed in other beam directions, shows that there are no conditions limiting reflections of the type hkl*. Similarly, plate 6 shows that 120* and 240* reflections are present, and it follows that all reflections of the type hkO* are allowed. Plate 7 shows a diffraction pattern obtained with B = [loo] which contains 01 1 *, 042* and 024* reflections but does not contain 012*, 032* and 014* reflections. Thus for reflections of the type Okl* it is necessary that (k+l) must be even. On the other hand, the diffraction pattern shown in plate 8 corresponds to B = [332], and this contains 110*, 113* and 226* reflections such that there appear to be no conditions limiting reflections of the type hhZ*.The conditions which govern allowed reflections are summarised in table 4, and it can be deduced that, of the eight possible space groups identified earlier, only P4,lmnm has conditions governing allowed reflec- tions which agree with those shown in table 4. Hence the analysis of convergent beam diffraction patterns taken from an individual particle of iron antimonate shows that the material is primitive tetragonal with a = b x 4.63 A, c x 9.23 A and has a structure with a space group P4,/mnrn. Analysis of the CBDP recorded from iron antimonate B showed them to be identical in all respects to those obtained from iron antimonate A, with the exception that the intensities of reflections of the type 002*, hkl* and hk2* were significantly lower in the diffraction patterns recorded from iron antimonate B.Determination of the Ordered Arrangement of Iron and Antimony Cations in Iron Antimonate The fact that the c-lattice spacing and the ratio c / a have been found to be three times larger when using electron diffraction than when using X-ray diffraction suggests that the true structure of iron antimonate consists of three tetragonal cells of similar dimensions to those found by X-ray diffraction, stacked along the c-direction. Ordering of the cations along the c-axis would lead to weak reflections which are not detected by X-ray diffraction, and hence X-ray observations would not lead to the detection of the superlattice (see later calculation of intensities).It is well known that the signal- to-background ratio is far better for electron diffraction patterns than for X-ray622 Iron Antirnonate studied by Electron Difraction Table 4. Summary of the conditions for observeda reflections in iron antimonate type of reflection present or absent hkI* present for all hkI* hkO* OkI* hhl* present for all h and k present if (k+ I ) even present for all h and I a Note that some reflections, such as 100* and 010*, are not observed except by double diffraction. diffraction patterns, the improvement arising both from the larger elastic scattering factor for electrons and from the larger background intensity on X-ray films. The large elastic scattering factor for electrons also means that diffracted intensities are usually influenced by dynamical events, i.e. by rediffraction of diffracted rays, so that intensities based on kinematical structure factors, i.e.when there is no rediffraction, cannot be used to determine precise atomic coordinates as is conventionally done with X-ray diffraction data. Hence it is not immediately obvious how the electron diffraction data obtained in the present work, which suggest the presence of cation ordering, can be used to investigate further the arrangement of the iron and antimony cations within the large unit cell. However, the fact that analysis of the diffraction data has shown that the 001* and 010*, and therefore the 1 OO*, reflections are kinematically forbidden allows possible atomic positions, consistent with the P4,lmnrn space group, to be tested to see if they are consistent with the structure factor for these reflections.In this respect it is important to consider the contribution which both anions and cations make to the structure factor for all reflections. The structure factor, F, for kinematically forbidden reflections, such as 001 * and 010*, must equal zero and we can therefore write12 Fool* = = Z fi exp 2.ni(hui + kui +- Zwi) = 0 where fi is the electron scattering factor for the atom, which in the present case corresponds to the appropriate electron scattering factor for oxygen, antimony or iron; hkl are the Miller indices of the reflections, i.e. 001* and 010*, and uc vi wc are the atomic coordinates of the atoms within the unit cell listed in table 5.The coordinates of equivalent positions have been taken from the values reported for structures with the P4,lrnnm space group,12 and these have been used to deduce possible atomic coordinates referred to the tripled unit cell. As indicated 'in table 5 the specific sites for iron and antimony cannot be differentiated from the X-ray data, and the cation sites ( a ) + ) are therefore labelled (Fe or Sb). The position of the oxygen atoms are taken to be precisely those given in the 1iterat~re.l~ If we consider the contribution of the oxygen to the 001* reflection we can write Fool+ = 2fo(eo + ,Xi + ezniI6 + e2W3 + &xi13 + ebnil3) = 0 where fo is the atomic scattering amplitude for oxygen. Similarly, considering the contribution of the oxygen atoms to the 010* reflection it follows that The oxygen atoms therefore contribute zero intensity to the 001* and 010* reflections, and it therefore follows that the sum of the contributions of the antimony and iron atoms to these reflections must also be zero.= 3fo(e0.63ni +e-O.63ni + e1.83ni +e0.37Xi) = 0.F. J. Berry, J . G. Holden and M . H. Loretto 623 Table 5. The atomic coordinates of iron, antimony and oxygen in the large tetragonal cell with c = 9.23 %, and a and b = 3.07 %, using atomic positions appropriate to the space group P4z,mnm given in ref. (3) atomic species atomic coordinates and reference to atomic site position iron or antimony (equal probability) a Ref. (3). If we now consider the sites for the metal atoms ( a m (see table 5 ) and represent the atomic scattering amplitude appropriate to each atomic site as fla),flb) .. .fly) we can write the structure factor for the 001* reflection = 0 = f(a) eo +f(b) e21ri/3 +f(c) e41ri/3 +Ad) e21ri/6 +f(e) elri + f ( f ) eSiI3 (1) 2fla) +Ad) +flf> = fib) +m + 2 f e ) (2) and m +Ad) = flc> +flf)- (3) fl4 +m +m = f l d ) +f(4 +m) (4) from which it follows that Similarly for the 010* reflection it follows that which on addition to eqn (2) reveals thatfla) = f i e ) , i.e. that the electron scattering factor for atoms in the (a) sites equals that of atoms in the (e) sites. Further simple substitutions show thatflb) = flf) andflc) =Ad), so that the observed zero intensities of the 001* and 010* reflections can be explained in terms of three pairs of sites in which the scattering factors for the atoms in each site in a given pair are equal.Since there are only two atom types which can be placed in these sites, one possible arrangement of atoms involves antimony atoms on (a) and (e) sites, iron atoms on (c) and ( d ) sites and iron and antimony distributed with equal probability on (b) and (f) sites. Thus the 001* and 010* reflections have zero intensity if antimony occupies the sites at 000 and E, iron atoms occupy the sites at O@ and a and with equal probabilities, either antimony or iron occupy the sites at Oq and #, as shown in fig. 3. The structure illustrated in fig. 3 conforms to the requirement that there are equal numbers of iron and antimony atoms, and it is clear, by inspection of this figure, that the 010* and 001* reflections would have zero intensity, since there are equivalent planes of atoms arranged with spacing one half of the (010) and (001) planes.The structure shown in fig. 3 corresponds to perfect order, but the degree of order in the samples is not known and cannot be obtained from the electron diffraction patterns presented here. Indeed, in view of the method of preparation it is highly likely that the samples are not perfectly ordered. Calculations of the structure factor for reflections expected to lie on the first-, second-624 Iron Antimonate studied by Electron Digraction c @=Sb 0 = Fe 8 = Sb/Fe Fig. 3. Diagram illustrating the positions of the iron, antimony and oxygen atoms in the tripled unit cell of ordered iron antimonate. See table 5 for precise coordinates.and third-order Laue zones with B = [OOl] and the first-order zone for B = [131] have been performed since only the third-order zone is obvious when B = [OOl] [see plate 4(b)], whereas the first-order zone is obvious in plate 4(c) with B = [131]. The results of these calculations are shown in table 6, together with results for other reflections, and it is clear that the observations are compatible with these calculations. The reflections in the first- and second-order zones with B = [OOl] are far weaker than those in the third-order zone, which are of the same intensity as the first-order reflections when B = [ 13 11. Thus, although these kinematic calculations can be used only qualitatively, the trend, especially for these high-order weak reflections, is expected to be in agreement with the observations, and the observed agreement is further evidence that the structure shown in fig.3 is the correct structure of iron antimonate. The importance of the contribution of the oxygen atoms and of double diffraction to the observed intensities of the reflections can be appreciated by considering the diffraction patterns shown in plates 2 and 7. Thus in plate 2(a), for which B = [OOl], the kinematically forbidden reflections 010* and 100* are visible, but when B = [loo] (plate 7) neither the 010* nor the 001* reflection appears. This difference in the visibility of kinematically forbidden reflections arises because of the different intensities of 2 lo*, 120*, 012* and 021* reflections.All of these reflections would be expected to have zero intensity from the cations because of the special condition for reflections when the specific cation sites are occupied.12 However, the 210* and 120* reflections will have significant intensities, and indeed calculations show (see table 6 ) that these reflections will be of the same order of intensity as the superlattice reflections such as 101*, with all the intensity coming from the oxygen atoms; the 012* and 021* reflections are expected to have zero intensity whatever atomic sites are occupied in this space group,12 and calculations confirm that the structure factor for the anions and cations is zero for these reflections. Thus the significance of the influence of the electron beam directions on the visibility of 010*, 100* and 001* reflections lies in the fact that the 210* and 120* reflections, which are excited with B = [OOl], provided the necessary condition for the generation of diffracted intensity into the 100* and 010* diffraction maxima by double diffraction, i.e.that the sum or the difference of two kinematically allowed reflections equals the kinematically forbidden reflection, e.g. 210* - 200* = 010*. It is clear that this condition is not fulfilled with B = [loo] since the 012* and 021* reflections are themselves forbidden.F. J . Berry, J. G . Holden and M . H . Loretto 625 Table 6. Intensities of electron and X-ray diffraction maxima for various types of reflections from ordered iron antimonatea intensity of reflection indices and nature of reflection electron X-ray 110 fundamental reflection 10 1 superlattice reflection 002 superlattice reflection 210 reflection due to scattering first-order zone reflections B = [OOl] second-order zone reflections B = [00 I] third-order zone reflections B = [OOl] first-order zone reflections B = [131] by oxygen ions 100 100 1.3 3.8 1.1 3.8 7.0 3.4 <0.01 - <0.01 0.2 0.3 - - - a The intensities are expressed as percentages of the intensities of the 110 fundamental reflection for electron and X-ray diffraction.Coordinates of atoms used for these calculations are given in table 5. Thus the observed diffraction maxima are all compatible with the intensities derived from the structure shown in fig. 3. The fact that the superlattice reflections, i.e.those for which the intensities are determined by the difference in scattering factors between iron and antimony, such as 101* and 202*, are weaker in electron diffraction patterns taken from iron antimonate B, which contains only FelI1 and SbV rather than FeII, FeIII, SbIII and SbV, suggests that the degree of long-range order is influenced by the valency state of the cations. The absence of visible superlattice reflections in X-ray diffraction patterns can be understood on the basis of the structure shown in fig. 3. Thus X-ray structure-factor calculations, based on the assumption that iron antimonate is perfectly ordered, have been carried out for the fundamental and superlattice reflections listed in table 6, where it can be seen that the strongest superlattice reflections will have an intensity of ca.4% of the fundamental reflections. Although these may just be visible on an X-ray film, any departure from perfect order would reduce the relative intensity even further, and this taken together with particle size broadening would make these reflections too weak to detect. The corresponding calculations for electron diffraction (table 6) show that the superlattice reflection would actually be a smaller fraction of the intensity of the fundamental reflection. These maxima are nevertheless easily visible because of the good signal- to-background ratio. Conclusions (1) Iron antimonate crystallises as an ordered tripled rutile-type structure with a = b x 4.63 A and c z 9.23 A, with a P4,lmnm space group. (2) Superlattice reflections generated from this ordered cell will not be easily visible and are not observed in X-ray diffraction patterns. (3) The scattering of electrons by the oxygen ions must be considered in order to account for the presence of certain kinematically forbidden reflections. (4) The degree of order appears to be influenced by the oxidation state of the cations.626 Iron Antimonate studied by Electron Diflraction References 1 K. Brandt, Arkiv. Mineral. Geol., 1943, A17, 15. 2 J. D. Donaldson, A. Kjekshus, D. G. Nicholson and T. Rakke, Acta Chem. Scand., 1975, A29, 803. 3 J. Amador and I. Rashes, J. Appl. Crystallogr., 1981, 14, 348. 4 R. G. Teller, J. F. Brazdil, R. K. Grasselli and W. Yelon, J. Chem. Soc., Faraday Trans. I , 1985, 41, 5 M. H. Loretto, Electron-beam Analysis of Crystals (Chapman and Hall, London, 1984). 6 P. B. Hirsch, A. Howie, R. B. Nicholson, D. W. Pashley and M. J. Whelan, Electron Microscopy of 7 B. F. Buxton, J. A. Eades, J. W. Steeds, G. W. Rackham, Philos. Trans. R. SOC. London, Ser. A, 1976, 8 H. Kriegsmann, G. Ohlmann, J. Scheve and F. J. Ulrich, Proc. 6th Int. Congr. Catal., ed. G. C. Bond, 9 N. Burriesci, F. Garbani, M. Petrera and G. Petrini, J. Chem. SOC., Faraduy Trans. I , 1982, 78, 817. 1693. Thin Crystals (Butterworths, London, 1973). 281, 171. P. B. Wells and F. C. Tompkins (The Chemical Society, London, 1976), p. 836. 10 F. J. Berry, J. G. Holden, M. H. Loretto and D. Urch, unpublished results. 11 J. Gjonnes and A. F. Moodie, Acta Crystallogr., 1965, 19, 65. 12 International Tables for X-ray Crystallography I (Kynoch Press, Birmingham, 1952). 13 W. B. Pearson, A Handbook of Lattice Spacings of Metals and Alloys (Pergamon, Oxford, 1967), vol. 2. Paper 611 10; Received 14th January, 1986

ISSN:0300-9599    

DOI:10.1039/F19878300615

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1987, 83, 627-634 Kinetics of N20 Decomposition on the Surface of y-A120, doped with Sodium Ions Christos Kordulis, Leonidas Vordonis and Alexis Lycourghiotis* Department of Chemistry, University of Patras, Patras, Greece Phillipos Pomonis Department of Chemistry, University of Ioannina, loannina, Greece The kinetics of N,O decomposition on a series of specimens prepared by doping y-Al,O, with various amounts of Na+ ions has been studied at various temperatures using a flow-bed reactor working under atmospheric pressure. This doping promotes the adsorption of oxygen anions produced from surface decomposition, presumably uia the formation of surface species “a+ - * 0- * - Na+], bringing about a transformation of the rate equation from R = k into R = k b , 2 0 P , z o / b ~ z P ~ 2 (where bNzO and boz are adsorp- tion coefficients and PNzO and PO2 are partial pressures).Moreover, a decrease in catalytic activity, expressed either as fractional conversion or rate of reaction, was observed on increasing the surface coverage C, of y-Al,O, with Na+ ions determined by X-ray photoelectron spectroscopy. Specifically, the dependence of the catalytic activity on the surface coverage of y-Al,O, is described by the relationship In (1/R) = 15.4+(281/K) C (where K is a proportionality constant) and it was concluded that the deactivation observed is due to the promotion of the 0, adsorption caused by the Na+ ions. Finally, the linear dependence of the surface coverage of y-Al,O, on the sodium content strongly suggests that the dispersion of the sodium supported species is constant irrespective of the surface concentration of sodium.The decomposition of nitrous oxide has been extensively used as a probe reaction for studying many catalytic systems based on aluminas, such as solid solutions and supported metal-oxide catalysts.1-6 In most cases interest was focussed on the relationship between the nature of the active ion (its symmetry, valence, concentration etc.), which is a transition-metal cation, and the catalytic activity. Moreover, the elucidation of the mechanism of the title reaction, by performing detailed kinetics, helps establish relationships between the nature of the active phase and the kinetic parameters involved in the surface process. However, in the above-mentioned studies the possibility that impurities in the alumina, the most important of which are sodium ions, could influence its activity, and the kinetic law of N20 decomposition has been completely disregarded.This possibility, at least concerning Na+ ions, must not be considered negligible because it has been recently observed that Li+ ions, like Na+ ions, considerably influence both the activity of Fe203/A1203 catalysts and the kinetics of N20 decomp~sition.~ The purpose of the present work is to study the influence of sodium ions on the catalytic behaviour of alumina. This is attempted by studying the kinetics of N20 decomposition on the surface of a series of specimens prepared by doping alumina with various amounts of sodium ions. Kinetic experiments at various temperatures were performed using a flow-bed reactor working under atmospheric pressure. In addition X-ray photoelectron spectroscopy (X.P.S.) and specific surface area (B.E.T.) measurements were carried out to characterize the specimens used. 627628 N,O Decomposition on y-Al,O,-Na Catalysts Experiment a1 Preparation of the Specimens The doping of alumina (y-A1203, Houdry Ho 41 5, 100-1 50 mesh) with various amounts of Na+ ions was performed by pore-volume impregnation with aqueous solutions of NaNO, [Merck p.a.1 followed by drying at 110 "C for 2.5 h and air-calcination at 600 "C for 12 h.Reference to the specimens prepared will be made using the formula Na-X-A1203, where Xdenotes the Na+ content in mmol g-1 Al,O, (X: 0.226,0.309,0.392 and 0.621).The sodium in the specimen with X = 0.226, which was prepared by impregnation with pure solvent (distilled water), originates from the y-Al,O, used. X-Ray Photoelectron Spectroscopy The surface of the catalyst was characterized by X.P.S. The X.P.S. spectra were recorded using a Vacuum Generators ESCA 3 spectrometer equipped with an aluminium anode [A1 Ka = 1486.6 eV] operating at 20 mA and 14 kV. The residual pressure inside the spectrometer was ca. Torr.? All the X.p. spectra presented a small C(1s) contamination peak to which a binding energy of 285 eV was assigned. This peak served as a reference for determining the binding energies of all other peaks in the spectra. The surface coverage of y-Al,O, by Na+ ions, denoted by C, is defined as the ratio of the surface area covered by the supported ion, s, to the total surface area, S : c = s/s (1 4 Moreover, the dispersion of the supported Na+, denoted by D, is defined as the ratio of the surface area to the amount of supported Na+: D = s / X (1 4 On the other hand, in a recent publications it was demonstrated that the dispersion of the Na+ ions is given by the following relationship: = [(INa(l~)/~Al(lp)) (A1/Na)l (1 c) where 1Na(ls)/1A1(2p) is the true intensity ratio of the X.P.S.signals from the Na(1s) and Al(2p) photoelectrons, Al/Na denotes the atomic ratio of these elements and K is a proportionality constant. By combining the above relationships the following equation is derived: Surface Area and Pore Volume Determination The specific surface areas of the samples studied were determined by the B.E.T. method.The experimental details were described in a previous paper.g The total pore volume of the specimens studied was determined by measuring the amount of water necessary for filling the pores. Catalytic Tests The catalytic tests were carried out in a plug-flow reactor working under atmospheric pressure. The catalyst sample, 400 mg, was placed on a perforated-glass bed occupying a volume of ca. 0.35 cm3 and having a depth of 0.2 cm. A mixture of He and N,O was t 1 Torr = 101 325/760 Pa.Ch. Kordulis, L. Vordonis, A . Lycourghiotis and P . Pomonis 629 passed through the bed at a rate of 180 5 cm3 min-l. Of this mixture two-thirds was He and one-third was N,O. Under these conditions the contact time corresponds to 0.12 s.A Varian 3700 gas chromatograph equipped with a thermal conductivity detector was used to analyse both reactants and products. Two gas valves with 1 cm3 loops were used for sampling. The column (dimensions 0.5 m x b in i.d.) was of stainless steel and filled with molecular sieve 5A. Preliminary tests showed that no decomposition took place up to a temperature of 700 "C in the reactor without catalyst. The temperature range examined was from 525 to 625 "C. Determination of the Kinetic Laws The conventional description of the whole process on the surface of most oxides and solid solutions includes the following 6~ ' 9 2N,O(g) + 2s f 2N20-(ads) + 2S+ (9 2N,O-(ads) 5 2N,(ads) + 20-(ads) (ii) 20-(ads) + 2S+ f O,(g) + 2s 2N,(ads) 2N,(g). Here S represents a surface site.equation : The rate of the surface reaction depicted by step (ii) can be d[N,O-(ads)] dt = k@NzO-(ads) R = (iii) (iv) described by the following (2) where R represents the reaction rate, k is the intrinsic rate constant and @N@-(ads) is the fraction of catalytically active sites on the surface occupied by N20. Provided that the adsorption of the reactant, products and carrier gas follow a Langmuir type isotherm, BN20-(ads) is given by where b,,, bNzO, bN, and bo, represent the adsorption coefficients and PHe, PNzO, P,, and Po, the partial pressures of these components in the reaction mixture. Since adsorption of He and N, is generally accepted to be negligible, the products bH,PH, and bNZPNZ in the denominator of the right-hand side of eqn (3) can be neglected.Thus eqn (3) is reduced to (4) b N ,OPN '20 @NzO-(ads) = + b N 2 0 P N 2 0 + b&'!z' According to the magnitude of adsorption of the N,O- and 0- on the surface of the catalysts the following extreme cases may be considered: (i) weak adsorption of both N,O- and 0-, (ii) strong adsorption of N,O- and weak adsorption of 0-, (iii) strong adsorption of 0- and weak adsorption of N,O-. For each case eqn (4) is reduced to a certain simpler equation. By combining each of the resulting equations with eqn (2), we derive the following kinetic equations corresponding (1) (ji) (iii) to the above-mentioned cases: ( 5 ) (6) (7)630 N,O Decomposition on y-Al,O,-Na Catalysts Taking into account the stoichiometry of the reaction, the fact that under our experimental conditions the total pressure in the stream is equal to 3PO,,, (where P i 2 , is the partial pressure of N,O in the feed) and the molar ratio of He to N,O = 2, we derive the following relationships between the partial pressures PNzO and Po, in the reaction mixture and the degree of conversion: The kinetic equations ( 5 ) and ( 7 ) are transformed into eqn (10) and ( 1 1) by substitution of PNzO and Po, from eqn (8) and (9): (iii) R = kb,,, 6P&,O (1 -4 6 + x The kinetic results are analysed using the well known equation of the plug-flow reactor : Fdx = RdS (12) where P i s the reactor feed in N,O (mol s-l) and S the surface area of the catalyst used (m2).Substitution of the rate obtained from eqn (6), (10) and ( 1 1 ) into eqn (12) and integration of the resulting relationships gives the expressions (1 3), (14) and (1 5), respectively : (13) - k 6 P > 2 0 S b N 2 0 F (0 -[71n(l l - x I ) + x ] - (ii) x = k S / F (14) 2[7x(6+x)]i+8x+6 (iii) 4 In {I 2[x(6 + x)]i - 2x - 6 I> + d 7 In bNzO 3.46 PG2o d s bk, * ( 1 5 ) F - { [ x ( 6 + x ) y - 11.91) = k- The following well known relationships are valid for both N,O and 0,: b = ka/kd ka = Aa exp ( -Ea/RT) kd = Ad exp (-Ed/RT) AHa = I?-Ed k = A exp (-Ea(,,/RT) Ea(a, = Ea(t) + AH!%~o where (ka, P, A") and (kd, Ed and Ad) are the Arrhenius parameters for adsorption and desorption respectively, AHa is the heat of adsorption and Ea(,) and are the trueCh.Kordulis, L . Vordonis, A . Lycourghiotis and P. Pomonis 63 1 and apparent activation energies of reaction, respectively. Using these relationships we can rearrange eqn (1 3), (14) and (1 5) into the expressions (1 6), (1 7) and (1 8), respectively : (17) A S Ea(t) 1 lnx = In---- F R T (ii) I -x(~+x)- 11.91 2 [7x(6 + x)# + 8x + 6 4ln{(2[~(6+~)]4-2x-61)+ d71n To decide which of the above models describes more satisfactorily the experimental data, plots of the left-hand side of eqn (15)-(17) vs.1/T were constructed, and the standard errors of the slope and intercept as well as the correlation coefficients were determined. The model selected for each catalyst had the minimum percentage standard error in both the slope and intercept as well as the maximum correlation coefficient. The above procedure was performed using a BBC microcomputer. Results and Discussion Fig. 1 illustrates the variation with sodium content of the surface coverage of y-Al,O, and the specific surface area of the specimens studied.The specific surface area remains practically constant, whereas the surface coverage increases with the X . The lack of variation in the specific surface area suggests that the sodium doping does not provoke any considerable change in the y-Al,O, texture. This is corroborated by the values determined for the total pore volume. In fact, the total pore volume remains practically constant (0.50 0.02 cm3 g-l) irrespective the Na+ content. The linear increase of the surface coverage of y-Al,O, with the concentration of deposited sodium demonstrates that the size of the crystallites of the supported Na species is constant irrespective of the sodium content. Furthermore, the slightly lower value of the surface coverage obtained for the sample with X = 0.226 compared to that predicted by the straight line simply reflects the fact that the repartition of the deposited Na+ ions along the radius of an alumina grain favours the surface of the grain rather than the more uniform microdistribution of the Na+ ions resulting from the original material used (undoped y-Al,O,).At this point it must be emphasized that the quantities C / K and X plotted in fig. 1 have been determined experimentally. According to the literature17 the deposited Na+ ions can diffuse into the y-Al,O, lattice during calcination, neutralize the surface hydroxyls, forming =A1-0-Na surface groups, and form islands of Na+ compounds supported on the y-Al,O, surface. These islands are aggregates of sodium, e.g.NaNO, and NaNO,, or tridimensional compounds between Na and Al, e.g. Na,A1,0,. In most cases these compounds are poorly defined by X-ray analysis, presumably because of their low concentration and/or their poor crystallinity. The relative extent of the above- mentioned surface processes depends on the concentration of sodium and the calcination temperature. The relatively low sodium concentration and temperature at which the specimens were calcined favour the neutralization of the surface hydroxyls. Table 1 illustrates the models which follow from N,O decomposition on the surface of the specimens examined. The experimental data are fitted better by model (ii) in the case of the undoped support, thus suggesting a zero-order kinetic law.This behaviour implies strong adsorption of N,O and weak adsorption of oxygen. The doping of y-Al,O, with Na+ ions causes a transformation of the model from (ii) to (iii).632 N,O Decomposition on y-Al,O,-Na Catalysts Fig. 1. Variation of the surface coverage, C, of y-Al,O, (A) and the specific surface area, S, of the specimens studied (0) with the sodium content. Table 1. Model selection sample model Na-0.226-A1,03 (ii) Na-0. 309-A1,0, (iii) Na-0.392-AI20, (iii) Na-0.621-A1,03 (iii) The change in the kinetic law may suggest: (i) promotion of 0, adsorption or (ii) promotion of N,O adsorption but much stronger promotion of 0, adsorption. In view of the fact that both N,O and 0, are retained on the surface as negative ions the second suggestion seems more realistic.The catalytic activity, expressed as fractional conversion, is illustrated in table 2. One observes a continuous drop in activity with increasing dopant content. Moreover, fig. 2 shows the dependence of the rate of the N,O decomposition, determined at a typical temperature (575 "C), on the surface coverage of y-Al,O, with Na+ ions. This rate has been calculated using the relation Fx = RS resulting from eqn (12), for values of conversion lower than 0.15, where the reactor could be considered as a differential one. An inspection of fig. 2 shows that a good correlation is obtained between In (l/R) and the surface coverage, C. This correlation is described by the following empirical relationship : (19) where K is the proportionality constant of eqn (1 d ) . In order to explain this relationship we assume that the promotions of the 0, and N,O adsorption are described by the expressions In (l/R) = 15.4+281 C / K b&Pto, = a exp ( b ~ ) (20) (21) bNzOPN20 = a' exp (b'c)Ch.Kordulis, L . Vordonis, A . Lycourghiotis and P. Pomonis 633 Table 2. Degree of conversion for the decomposition of N20 temperature of reaction/”C sample 525 550 575 600 625 Na-0.226-A120, 0.047 0.076 0.140 0.220 0.460 Na-0. 309-A120, 0.03 7 0.065 0.1 10 0.160 0.320 Na-0.621 -A120, - - 0.045 0.078 0.140 Na-0.392-A120, - 0.037 0.070 0.100 0.210 15.5 I 1 1500 3000 4 500 (C/K)/nmol Na m-2 6 000 Fig. 2. Dependence of the rate of the N20 decomposition determined at 575 “C on the surface coverage of 7-A120, by Na+ ions.where a, a’ and b 9 6’ are proportionality constants. Based on these expfessjons eqn (22) is derived by combining eqn ( 2 ) and (4), provided that 1 + bNzoPNzo + bB,P&,. The latter relation is implied in the assumptions already made for the derivations of eqn (6) and (7) : ka’ exp (b’C) a’ exp (b’C) + a exp (bC)‘ R = On the assumption that 1 / k 4 (a/ka’) exp [(b - b‘)q, an expression similar to eqn (1 9) can be easily achieved by transforming eqn (22): (23) In (1 / R ) = In (alka’) + (b - b’) C. The above equation clearly demonstrates that the stronger promotion of 0, adsorption, compared to that of N20 adsorption, caused by the doping is responsible for the decrease in the catalytic activity with Na+ concentration. As to the mechanism whereby the Na+ ions promote oxygen adsorption it seems to us reasonable to assume, in agreement with assumptions for the Fe,O,/y-Al,O,-Li sy~tem,~ that the retention on the surface of the oxygen anions produced from the N20 decomposition occurs via the formation of an unstable surface species between Na+ and 0- (Na+ * 0- - * - Na+).22 FAR 1634 N20 Decomposition on y-Al,O,-Na Catalysts In conclusion, the modification of y-A120, by sodium ions brings about a promotion of oxygen adsorption produced from the decomposition of N20, which in turn lowers the kinetic rate and decreases the catalytic activity. These effects must be taken into account when various aluminas containing different amounts of sodium ions have to be utilized as supports or matrices in catalytic systems where N,O decomposition is used as a probe reaction.References 1 T. A. Egerton, F. S. Stone and J. C. Vickerman, J. Catal., 1974, 33, 299. 2 T. A. Egerton, F. S. Stone and J. C. Vickerman, J. Catal., 1974,33, 307. 3 P. Pomonis, Ph. D. Thesis (UMIST, 1977). 4 J. C. Vickennan in Catalysis (Special Periodical Report, The Chemical Society, London, 1978), vol. 2, p. 107. 5 P. Pomonis, D. Vattis, A. Lycourghiotis and Ch. Kordulis, J. Chem. Soc., Faraday Trans. I , 1985,81, 2043 6 A. Lycourghiotis, D. Vattis, Ch. Kordulis and P. Pomonis, Proc. 2nd Czechoslovak ConJ on the Preparation and Properties of Heterogeneous Catalysts, ed. V. Zapletal (Czechoslovak Academy of Sciences, Bechync, 1985), p. 102. 7 A. Lycourghiotis, Ch. Kordulis, D. Vattis and P. Pomonis, presented at the Anglo-Dutch Catalysis Conference, September 18-20, 1985, Noordwijkerhout, The Netherlands. 8 Ch. Kordulis, S. Voliotis, A. Lycourghiotis and D. Vattis, Appl. Catal., 1984, 11, 179. 9 L. Vordonis, P. G. Koutsoukos and A. Lycourghiotis, J. Catal., 1986, 98, 296. 10 A. G. Keeman and R. D. Iyengar, J. Catal., 1966,5, 301. 1 1 A. Cimino, R. Bosco, V. Indovina and M. Schiavello, J. Catal., 1966, 5, 271. 12 M. L. Volpe and J. F. Ready, J. Catal., 1967, 7, 76. 13 A. Cimino, V. Indovina, F. Pepe and M. Schiavello, Proc. 4th Znt. Congr. Catal. (Mir, Moscow, 1968), 14 A. Cimino and V. Indovina, J. Catal., 1970, 17, 54. 15 A. Cimino and F. Pepe, J. Catal., 1972, 25, 362. 16 P. Pomonis and J. C. Vickerman, J. Catal., 1978, 55, 88. 17 Ch. Kordulis, S. Voliotis and A. Lycourghiotis, J. Less Common Met., 1982, 84, 187. p. 187, paper 12. Paper 6/125; Received 16th January, 1986

ISSN:0300-9599    

DOI:10.1039/F19878300627

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

J. Chem. SOC., Faraday Trans. 1, 1987,83, 635-644 Standard Gibbs Free Energies of Transfer of NaCl and KCl from Water to Mixtures of the Four Isomers of Butyl Alcohol with Water The Use of Ion-selective Electrodes to Study the Thermodynamics of Solutions De-Ying Chu,* Qian Zhangt and Rui-Lin Liu Department of Chemistry, Peking University, Beijing, China Measurements have been made on NaCl and KCl in water and in mixtures of water with the four isomers of butyl alcohol using the following cell, which has no liquid junction and which consists of ion-selective electrodes : cation(M) ; responsive glass or ion-selective electrode H 0 C1--selective The standard Gibbs free energies of transfer of NaCl and KC1 from water to mixtures of water with the four isomers of butyl alcohol were calculated from cell potentials.This paper compares measurements of A G and AG: using ion-selective electrodes with those obtained from amalgam cells. The results obtained using ion-selective electrodes are reliable, and the differences between AG: of NaCl and KC1 from water to butyl alcohol-water mixtures are very small. AG: is proportional to the amount of butanol present over the experimental concentration range. MC1, I '1 electrode. The use of ion-selective electrodes to study the thermodynamic properties of solutions is safe, simple and quick, although its accuracy is not as high as that of the classical amalgam method (see later). We have carried out a series of experiments with ion-selective electrodes to study the thermodynamics of salt solution in mixed solvents in order to examine the laws of solvation of ions in mixed solvents. The standard Gibbs free energy of transfer between water, w, and another solvent or mixed solvent, s, is the difference between the molal standard Gibbs free energies of solvation of the electrolyte in the two different solvents: AG,"(MX) = 'G"(MX) - "G"(MX).AGY is an important measure of the differences in interactions between the ions of the electrolyte and the solvent molecules in the two media. It may be measured by e.m.f. methods. Lowe and Smith2 reported the use of sodium-responsive glass electrodes to obtain the free energy of transfer of sodium chloride between H,O and a H20-D,O mixture. Pointud et aL3 used cell (I), which consisted of an electrode responsive to monovalent cations and a silver-silver halide electrode : glass electrode responsive to M I MC1 in solvents I AgCl, Ag.(1) The free energy of transfer of the alkali-metal halides from water to water-t-butyl t Present address : Department of Physical Chemistry, School of Pharmaceutical Sciences, Beijing Medical I-Jniversity, Beijing, China. 635 22-2636 Transfer Energies of NaCl and KCI in Water-Alcohol Mixtures 14.0 12.0 " n 0 0 5 10.0 4.0 2-0 0.0 0.00 0.02 0.04 0-06 0.08 0.10 m Fig. 1. Plots of AG' against rn for the system NBA-H,O-NaCl. Values on plots are wt% NBA. 14.0 12.0 10.0 L b a - 0 - 0 " > 8.0 E 1 4.0 0.0 0.00 0.02 0-04 0.06 0.08 0-10 m Fig. 2. As fie. 1 for NBA-H-0-KCl.D-Y. Chu, Q. Zhang and R-L. Liu cation(M)-responsive 637 (11) anion(X)-responsive 20.0 18.0 16.0 14.0 12.0 ;> 5 10.0 %- a 8.0 6.0 4.0 2.0 0.0 1 0.00 0.02 0.04 0.06 0-08 0.10 m Fig.3. Plots of AG’ against rn for the system SBA-H,O-NaCl. Values on plots are wt% SBA. Previously we reported the use of selective electrodes to determine the free energy of transfer of alkali-metal halides from water to water-methanol mixtures and to water- dimethylformamide (DMF) mixture^.^ This paper reports the determination of Gibbs free energies of transfer of NaCl and KCl from water to water-n-butyl alcohol (NBA), water-s-butyl alcohol (SBA), water-isobutyl alcohol (IBA) mixtures using selective electrodes. We also compare these results with those of other workers in our laboratory using the amalgam method.6 Principles The work reported in this paper used cell (11) with no liquid junction.The electromotive force, of the cell was measured for a solution of MCl (mW) in water and in the mixed solvent, SE(rnS).638 Transfer Energies of NaCl and KCI in Water-Alcohol Mixtures 18.0 16.0 14.0 12.0 - - - A - - 0 7 - - I ti 3 0.00 0.02 0-04 0-06 0.08 0.10 m Fig. 4. As fig. 3 for SBA-H,O-KC1. The difference in e.m.f. (AE,) between the two cells is given by AE, = sE'-wE"+2 x 2.303(RT/F) logl,(mSy~/mWy~) = AEY + 2k log,, (msy$ /mwy$) (1) where k = 2.303 (RT/F), R = 8.314 J k-l mol-l and F = 96500 C mol-l; m and y+ are the molality and the mean ionic activity coefficient of the salt, respectively. When ms = mw then A 4 = AG + 2k log,, (7% I72 >. log,, y* = [ -Am;/( 1 + BGm;)] + bm - log,, (1 + 0.0O2Mz, m) (2) y$ - and y? - can be calculated from a corrected Debye-Huckel expression:8 ( 3 ) in whichQ Almol-4 dmj K: = (1.8246 x 1 06)/(DT)3 B/cm-l mol-4 dmt Ki = 50.29/(0T);D-Y.Chu, Q. Zhang and R-L. Liu 16-0 14.0 639 - Y A - ,l " - ,. 12.0 0.0 1 I 1 I I 0.00 0.02 0.04 0.06 0.08 0.10 m Fig. 5. Plots of A q ' against rn for the system IBA-H,O-NaCl. Values on plots are wt% IBA. - Y U - 16.0 10.0 14.0 " " - - 12.0 10-0 > E 0 - 8.0 6-0 4-0 2.0 I I L 1 1 0.00 0-02 0.04 0-06 0.08 0.10 0-0 m Fig. 6. As fig. 5 for IBA-H,O-KCl.640 Transfer Energies of NaCl and KCl in Water-Alcohol Mixtures 40 $ 30- 2 20: 1 o r - " - c 15 " 10 A n P w 1 0 ' 0 no0 0 -01 0.02 0.03 m Fig. 7. Plots of AG' against rn for the system TBA-H,O-KCl. 5 c - tJ Table 1. The physical properties of the butyl alcohol isomers and water refractive index, ng dipole moment, species measured lit.lo p( CGSE)lo dielectric constant, D (25 "C) NBA 1.3970 1.3973 1.75 SBA 1.3949 1.3950 1.66 IBA 1.3937 1.3939 1.79 TBA 1.3850 1.3851 1.66 H,O 1.3325 1.3325 1.84 17.51 16.56 17.93 12.47 78.36 where D is the dielectric constant of the solvent, 6 is a Debye radius set as 4.4A and b is a parameter.Mzy is the mean molecular weight of the mixed solvent; it can be expressed as Mxy = (1 -x)M,+xM, where x is the mole fraction of the organic component. Let f(m) = ( - Ami/ 1 + B6mi) - log,, (1 + 0.002Mx, m) then eqn (3) becomes logy, =JTm)+bm. Substitution of eqn ( 5 ) in eqn (2) gives AEt = A G + 2k[f(ms) - Amw)] + 2k(bs - bw) m AEt - 2k[f(ms) -Amw)] = A G + 2k(bs - bw) m. AG' = AEt - 2k[f(ms) -Amw)] then Let hence AK' = AG+2kAbm where Ab = b"-bw.D-Y.Chu, Q. Zhang and R-L. Liu 64 1 Table 2. AG and Gibbs free energies of transfer of NaCl and KCl from H,O to H,O-NBA mixtures (298.15 & 0.1 K) Ag/mV AG,"(rn)/kJ mo1-I NBA (wt % ) I" IIb I" IIb 4.4 6.8 8.8 11.1 13.2 4.4 6.8 8.8 11.0 13.4 NaCl 4.40 6.72 8.92 11.17 13.49 4.30 6.69 8.88 1 1.04 13.53 KC1 0.42 0.66 0.85 1.07 1.27 0.42 0.66 0.85 1.06 1.29 0.43 0.65 0.86 1.08 1.30 0.42 0.65 0.86 1.07 1.31 a This paper. Ref. (6). Table 3. AG and Gibbs free energies of transfer of NaCl and KCI from H,O to H,O-SBA mixtures (298.15kO.l K) ~~~ Ac/mV AG,"(rn)/kJ rno1-I SBA (wt%) I" IIb I" IIb 2.1 6.5 11.0 15.4 19.7 2.1 6.6 11.0 15.4 19.7 NaCl 2.17 6.58 11 .oo 15.44 19.85 2.1 1 6.45 10.94 15.38 19.83 KC1 0.20 0.63 1.06 1.49 1.90 0.20 0.64 1.06 1.49 1.90 0.2 1 0.64 1.06 1.49 1.92 0.20 0.62 1.06 1.48 1.91 a This paper.Ref. (6). A& is the measurable quantity. Using the experiment data, we plot AG' against m for all systems; at m = 0 AG is obtained as the intercept (see fig. 1-7), and 2kAb is obtained from the slope. The value of b for the salt in pure water is obtained from experiment, so that values of b for the salt in various mixed solvents may be obtained. The mean ionic activity coefficient is calculated from eqn ( 5 ) . The standard Gibbs free energy of transfer of a monovalent salt from water to a mixed solvent is given by AGt(m) = FAG(m). (9)642 Transfer Energies of NaCl and KCl in Water-Alcohol Mixtures Table 4. A G and Gibbs free energies of transfer of NaCl and KC1 from H,O to H,O-IBA mixtures (298.15 & 0.1 K) ~ ~~ Aq/mV AG,"(m)/kJ mol-l IBA (wt%) Ia IIb IU IIb 2.0 6.6 11.2 13.4 15.4 2.1 6.5 11.0 13.4 15.4 NaCl 2.09 6.64 11.24 13.47 15.8 1 2.07 6.56 11.17 13.41 15.79 KCI 0.19 0.64 1.08 1.29 1.49 0.20 0.63 1.06 1.29 1.49 0.20 0.64 1.09 1.30 1.53 0.20 0.63 1.08 1.29 1.52 a This paper.Ref. (6). Table 5. A G and Gibbs free energies of transfer of NaCl and KCl from H,O-TBA mixtures (298.15 & 0.1 K) AG@)/kJ mol-l AG:(m)/kJ mol-l AG/mV TBA (wt % ) NaC1, IIa KC1, Ib KCl, Ib KCl, IIIc 5 10 15 20 1.1 1 11 1.1 1.08 2.18 22 2.1 2.17 3.32 34 3.3 3.25 4.38 45 4.3 4.28 a Ref. (6). This paper. Ref. (7). Experiment a1 We used cell (11), constructed of a SAC-ION glass electrode responsive to sodium ions or a 40 1 -type potassium-selective electrode and a C1--selective electrode (made in China).The calibration curves of the electrodes were measured using a 217 calomel electrode with double salt bridge (with the internal chamber filled with a solution of saturated KCl and the outer chamber filled with a solution of saturated NH,NO,). The slopes are 58.7 (Na+ glass electrode), 54.7 (K+ electrode) and 57.9 (Cl- electrode), respectively. The solvents were purified by methods described by Riddick.lo The purity was controlled by refractive-index measurements (see table 1); solutions were prepared with redistilled water [conductivity ca. (1-2) x C2-l cm-l ] by gravimetry. E.m.f. measurements were at 25.0 f 0.1 "C. In order to check the correctness of this method, the free energies of transfer of NaCl and KCl in CH,OH-H,O systems were measured by using selective electrodes, and the results were compared with data obtained using an amalgam electrode by Feakins et aL5~ *D-Y.Chu, Q. Zhang and R-L. Liu 643 With the concentration of salt fixed, a group of aqueous solutions of butyl alcohol whose concentrations increased in a given proportion were prepared. Then the con- centration of salt was changed, another group of solutions was prepared at the new salt concentration, and so on. The potentials were measured in the order of increasing concentration of salt and organic solvent. Each group of experiments was repeated at least twice. We took the value of steady potential as the equilibrium value, i.e. any change did not exceed 0.2 mV in 5 min.In this paper we report measurements of the Gibbs free energy of transfer of NaCl and KC1 from H,O to NBA-H,O, SBA-H,O and IBA-H,O mixtures which have not been reported previously, and we compare the data for the KCl-TBA-H,O system with those given in the literature.' The experimental data are listed in tables 2-5. Results and Discussion The data obtained using ion-selective electrodes were compared with the results of ref. (6) using the amalgam method (see tables 2-5). The mean deviation in "E by the amalgam method is 0.05 mV; the total error in the standard e.m.f. was kO.1 mV and the Gibbs free energy of transfer was 10 J, less than that reported by Feakins et al.' The difference between the e.m.f. values obtained using ion-selective electrodes and those obtained using an amalgam electrode was 0.1-0.2 mV for values of "E or sE, and for "Go or SGo the errors were zt 10-97 J.The results for AG," are as follows. AG,"(m) increase in direct proportion to the butyl alcohol content for NaCl and KCl in butyl alcohol-water systems provided that the butyl alcohol content is < 20 wt % , as AG," = kC, where k = 0.21-0.22 kJ mol-1 and C is the wt% of butyl alcohol in the solvent mixtures. The Gibbs free energy of transfer of the ion is the difference between the free energies of solvation of the ion in the different solvents. AG: varied as the content of the second component changed. This is due to the fact that the solvation of NaCl or KCl in the butyl alcohol-water mixtures was affected by the second component. Some water molecules in the primary solvation shell were substituted by butyl alcohol molecules; however, the composition of the primary solvation shell is dependent on the bulk butyl alcohol content.The differences between the Gibbs free energies of transfer of NaCl or KCl from water to butyl alcohol-water mixtures were not large, and did not exceed 30 J (to 0.3 mV). For the four isomers of butyl alcohol, the molecular structures and physical properties are slightly different (see table 1). These differences influence the free energy of solvation, but their influence is small, so that the difference in AG: is also small (tables 2-5). It provides a foundation for the theoretical treatment of the free energy of solvation and for the choice of theoretical model. The hard-sphere model can be used to treat the four isomers, regardless of the fact that they have different structures.The experimental results show that the influence of butyl alcohol on the solvation of Na+ or K+ in mixed solvents is the same. References 1 D. Feakins and P. Watson, J. Chem. SOC., 1963, 4686. 2 B. M. Lowe and D. G. Smith, J. Chem. SOC., Chem. Commun., 1972,989. 3 Y. Pointud, J. Jillard, J-P. Morel and L. Avedikian, Electrochirn. Acta, 1974, 19, 229. 4 K. Covington and J. M. Thain, J. Chem. SOC., Faraday Trans. 1, 1975, 71, 78. 5 Chu Deying, Liu Ruilin et al., Communication of Ion Selective Electrodes (China), 1984, 4, 49. 6 Fang Liqi and Liu Ruilin, Acta Chim. Sin., 1985, 43, 415. 7 T. A. Clune, D. Feakins and P. J. McCarthy, J. Electroanal. Chem., Interfac. Electrochem., 1977, 84, 199.644 Transfer Energies of NaCl and KCl in Water-Alcohol Mixtures 8 D. Feakins and P. J. Voice, J. Chem. Soc., Faraday Trans. I , 1972, 68, 1390. 9 Huang Ziqing, Introduction to the Theory of Electrolytic Solutions (The Science Publishing House of 10 J. A. Riddck and W. B. Bunger, Techniques of Chemistry (Wiley, New York, 3rd edn, 1970), vol. 11, I 1 A, K. Covington, Ion-selective Electrode Methodology (CRC Press, Boca Raton, Fla, 1979), vol. I, Paper 61264; Received 6th February, 1986 China, 1983), pp. 82-83. pp. 834-840. p. 67.

ISSN:0300-9599    

DOI:10.1039/F19878300635

出版商:RSC    

年代:1987

数据来源: RSC