FU~U~UY Discuss. 1995 100 C9-C27 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalysts John Meurig Thomas Davy Faraday Research Laboratory The Royal Institution of Great Britain 21 Albemarle Street London UK WIX 4BS and Peterhouse University of Cambridge Cambridge UK CB2 IQY 1. Introduction In view of the fact that the first part of the title of my talk has prompted some to comment on its sado-masochistic overtones I should point out that its origins lie deep in the poetry and insights of Shakespeare '. .. bitter torture shall winnow the truth from falsehood' and '. . . the torture of the mind (is) to be in restless ecstasy' In common with many others I pursue my chemistry with passion and commitment. And it seems to me to be within the bounds of poetic licence to describe one's oscillation of mood from euphoric exhilaration-when the act of probing is profitable-to satur-nine gloom-when it is not-as tortured ecstasy.Now catalysis as a phenomenon has been the subject of numerous Faraday Society Discussions. At the fiftieth held in Cambridge in 1928 T. M. Lowry in his opening address pursued a holistic approach in which he endeavoured to encompass homoge- neous heterogeneous biological and photo-catalysis. Looking back at that meeting as well as an earlier one in 1921 held in London also on catalysis we note that many of the ideas and concepts in vogue seventy years ago have fallen by the wayside broken mile- stones on a vanished road. But one of the most remarkable facts is that the sub- discipline of catalysis which was arguably the least understood seventy or so years ago-biological or enzymatic catalysis-is now almost certainly the best understood.2. Lessons from the Enzymologists and Organic Chemists Why is it that enzymologists can nowdays successfully design new biological catalysts such as artificial enzymes that rival (even surpass) the performance of the parent natural enzyme? And why is it that the inorganic catalyst scientist and engineer are a good deal less adroit in their endeavours? The principal reason is because those working with enzymes have for over thirty years been engaged in in situ structural studies of their catalysts and have consequently gained revealing insights into mechanism whereas those working with inorganic catalysis have not.In the mid 1960s D. C. Phillips et al. elucidated the structure of the enzyme lysozyme.' They identified the cavity within its structures that constitutes its active site and in particular characterized in atomic detail the stereochemistry of the site and the manner in which reactant molecules fitted snugly within it how as it were the reactant was held (Plate 1) poised for catalytic conversion. Their studies also revealed how product species and and/or inhibitors occupied the cavity of the active site (Fig. 1). The c9 c10 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst Plate 1 Packing of lysozyme with the cleft that defines the active site occupied by the reactant (substrate) sequence of the amino acid components lining the cavity active site was identified and the nature of the binding of reactants and inhibitors likewise elucidated.This in turn gave the enzyme chemist and molecular biologist equipped with the preparative and synthetic dexterity of the organic chemist a firm basis for the design of smaller (miniature) versions of naturally occurring enzymes. It became sensible to talk of biomimetic catalysts and artificial enzymes and with the later advent of site selective mutagenesis new dimensions of structural probing became possible. A particular amino acid close to or actually constituting the active site-which usually consists of a triad of amino acids-could be replaced thereby enabling the catalytic performance of the enzyme to be delicately monitored as a function of subtle change in stereochemical and electronic environment within the active site.(For an elegant account of such procedural Fig. 1 Mechanism of mode of catalytic action of lysozyme (from Phillips') J. M. Thomas c11 advances see the work of Knowles2 on the manner in which triose phosphate isomerase catalyses the interconversion of two triose phosphates known as DHAP and GAP.) Work on so-called artificial enzymes may be illustrated by reference to the proteo- lytic enzyme chymotrypsin the structure of the active site of which was elucidated by in situ X-ray crystallography several decades ago. This revealed that the key components of the catalytic reaction cavity of chymotrypsin are imidazolyl phenolic and carboxylic groups.Armed with this information a way of assembling a miniature artificial chy- motrypsin presented itself to Bender and co-worker~.~ They took a naturally occurring molecule such as P-cyclodextrin as the shallow cavity and grafted on to its rim- appropriately juxtaposed just as in the chymotrypsin itself-the three components (the triad of groups) that constitute the active site. Such a creation is shown schematically in the lower half of Fig. 2 and a scalar model is shown in Plate 2. Although there is now some doubt about the validity of the claims made in Bender and co-workers' work the notion that lies behind their overall strategy remains (Plate 2) a synthetic molecular entity of about a hundredth the size and mass of the parent enzyme can be made to function as a catalyst in such a manner as to rival the performance of the natural enzyme.Breslow4 has produced an even simpler piece of elegant analogous chemistry where the reality of the shape-selective catalysed chlorination of anisole is unmistakably E+S E-S (E -S)' EtP I t 1 catalyst substrate@) I (C -S) complex I products (C) stabilitv rate selectivity selectivity turnover artificial enzyme Fig. 2 (u) Schematic representation of the enzymatic process. The enzyme E processes the sub- strate (reactant) S into product P uiu the transition state E-S (after Lehn). (b) Strategy for the construction of an artificial chymotrypsin (miniature enzyme) based on cyclodextrin (from J. M. Thomas6) c12 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst .._-Plate 2 Model of artificial chymotrypsin.(after Render and co-workers’). (Colour graphics by R. G. Bell and J. M. Thomas.) Plate 3 Cyclodextrin-catalysed shape-selective conversion of anisole to p-chloroanisole. (Colour graphics by Dewi Lewis based on Breslow4.) J. M. Thornas C13 demonstrated (Plate 3). Only the para chloro isomer is produced in this cyclodextrin- steered reaction. 3. How Well can we Design Inorganic Catalysts? Although it seems highly unlikely that we shall ever have the benefit of an inorganic equivalent to site-selective mutagenesis great strides may be taken in designing shape- selective inorganic catalysts using molecular sieve solids typified by that shown in Plate 4.Here we see a zeolitic framework known as ZSM-5 (Zeolite Socony Mobil number 5) which is a highly siliceous aluminosilicate of general formula H '(Si -xAlx02)-nH20 where the Si/Al ratio may range from a low of ca. 10 to almost infinity. Aluminosilicate ZSM-5 is an excellent catalyst' for shape-selectively (a) converting benzene to monoethylbenzene (a precursor of styrene) by alkylation with ethene; (b) disproportionating two molecules of toluene to one of benzene and one of paruxylene; (c)isomerizing a mixture of xylenes preferentially to paraxylene; and (d) converting methanol to petrol (gasolineka mixture of predominantly benzene toluene and xylene-and water. By introducing certain key elemental substituents into the ZSM-5 framework numerous other highly selective catalytic reactions of profound petrochemical significance may be effected.6 Thus with gallium in the ZSM-5 propane and butane may be efficiently dehy- drogenated to yield benzene and toluene this being the basis of the British Petroleum cyclar process commercialized some five years ago.With zinc in the ZSM-5 framework dehydrogenation of isobutane to isobutene (known also as 2-methylpropene) is effected. This alkene first prepared in this building (and discovered by Michael Faraday) is of central importance industrially as it is the precursor of MTBE (methyl tertiary butyl Plate 4 ZSM-5 and its use as a modified catalyst for methane oxidation to methanol;* propane to benzene; and isobutane to 2-methylpropene (isobutene or isobutylene) C14 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst Fig.3 High-resolution electron micrograph of a zeolitic acid catalyst showing projected structure of pores lined with active sites. Large pores are 5.5 A in diameter [from J. M. Thomas,6 and J. M. Thomas and P. L. Gai-Boyes Nature (London),1993,364,4781. ether) a major constituent nowadays of motor car fuel since it both replaces the dele- terious lead-containing additives and facilitates7 the complete combustion of the fuel to CO rather than CO. With iron in the framework the ZSM-5 is reported to favour the oxidation of methane (with N,O as reactant) to methanol.' ZSM-5 itself (Fig. 3) was first produced (in the early 1970s) by accident. Workers at the laboratories of the Mobil Company USA decided to try a new organic molecule as a template for the production of synthetic zeolites using the general strategic prin- ciples first employed by the British scientist R.M. Barrer. More will be said about Plate 5 Graphic illustrating how microporous titanosilicalite (TS-1) converts phenol to hydro- quinone using H,O ,and propene to propene oxide (Colour graphic by Dewi Lewis.) J. M. Thomas C15 templating later; what is important to note here is that it is possible to design new catalysts having a framework structure more-or-less identical to that of ZSM-5 but implanted with a 'designed' hetero-atom so as to endow the resulting solid with alto- gether new catalytic properties. This is precisely what researchers at the Enichem Company Italy succeeded in doing when they brought forth their so-called titanosilicalite-I (TS-1).This has the same framework structure as ZSM-5 and silicalite- 1 (which is the siliceous end-member of ZSM-5 i.e. x = 0 in the above-mentioned formula) but Ti4+ is accommodated in the framework. TS-1 is a highly selective and active oxidation catalyst capable amongst other things of facilitating the epoxidation of propene and the conversion of phenol and H,O to quinol (1,4 benzene diol) an important material (as a developer) in the photographic industry (Plate 5). 4. De Novo Synthesis of New Inorganic Catalysts Plate 6 shows the protonated (acidic) form of mordenite inside the pores of which are molecules of diisopropylnaphthalene (DIIPN) a vitally important building block in the synthetic polymer industry9*'* (DIIPN is formed in high yield when propene and naph- thalene are allowed to react in the presence of an acid catalyst).Over acidic silica- alumina gel numerous other undesirable products are formed as well as DIIPN. Over the environmentally harmful AlCl again there is a multiplicity of products. Acidic mordenite however is shape-selective in that it yields predominantly 2,6 DIIPN). In Plate 6 the snugly fitting product (DIIPN) gives a clue as to the mode of action of organic templates in the synthesis of new microporous catalysts. The degree to which templates are critical in the synthesis of a particular framework varies.' ',12 Certain templates (of which ethylenediamine is an example) may be regarded simply as void-filling species that do not contribute significantly to the preferential for- mation of a desired structure.We also find that a particular framework is formed by several different template molecules a process which might be more correctly termed Plate 6 Colour graphic representation of protonated mordenite of high Si/Al ratio. Inside one of the large pores a molecule of diisopropylnaphthalene (DIIPN) is shown. As described in the text (see also Cusumano') acidic mordenite is the catalyst of choice especially on environmental grounds in the conversion of naphthalene and propene to DTIPN. (After J. M. Thomas R. G. Bell and P. A. Wright Bull. Chim. SOC.Fr. 1994 131,482.) C16 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst structure-directing rather than genuine templating.In general however all the templates suitable for a particular framework will possess similar properties-shape size basicity etc.-which direct the gel medium (usually under hydrothermal conditions) from which the crystals nucleate toward the formation of particular structural motifs. A great deal of accumulated experience supplemented nowadays by computer graphics and modelling11p13 has been acquired to enable the inorganic preparative chemist to design a wide range of zeolitic material. In particular microporous aluminium phosphates and metal-substituted (in the framework) aluminium phosphates may now be routinely pre- pared. Some of these have structures similar to those that have been characterized in the aluminosilicate zeolites (natural and synthetic).Some however are quite novel. The framework structures of three entirely microporous solids synthesized in this Laboratory or in association with collaborators in Jilin University Peoples' Republic of China are illustrated in Fig. 4. It transpires that aluminium phosphates (ALPOs) may be quite readily produced as microporous analogues of naturally occurring and synthetic zeolites. Many MeALPOs (and with Me = Co Ni Zn Mn Cu Mg and other divalent cations partially replacing the trivalent A13+ of the parent ALPO) are very good solid acid and redox catalyst^.^,'^ Thanks to the efforts of many over the past decade about a hundred distinct micro- porous structures with channel apertures falling in the range from 4 to 14 A have been prepared.Into such structures about a third of the elements of the Periodic Table may be incorporated at framework sites. And into the sites occupied by extra-framework usually exchangeable cations a further third of all the known elements may be placed. These architecturally elegant inorganic repositories have a prodigality of chemical com- position that defies precise description; millions of distinct compositions are possible. Many thousands have already been prepared. The scope for delicate variation is enor-mous. DAF-I DAF-2 (b) JDF-3 Fig. 4 Framework structures of DAF-1 (magnesium aluminium phosphate) showing (a) the wide channel containing supercages and (b) the narrower channel; DAF-2 (cobalt phosphate) contain- ing T (T = Co P) viewed along (a) [OOl] and (b) [loo] axes; and JDF-3 (zinc phosphate) contain- ing T (T = Zn P) viewed along [loo] axis.DAF stands for Davy Faraday arid JDF Jilin Davy Faraday. J. M. Thomas C17 H Plate 7 Quantitative description (of bond lengths and coordination) at the Co" active site in the CoALPO-18 solid acid catalyst for the selective dehydration of methanol. (From J. M. Thomas and G. N. Greaves Sciences 1994,265,1675.) C18 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst 5. Uniform Heterogeneous Catalysts A Vast Family of Inorganic Solids Amenable to in situ Structural Investigation ZSM-5 mordenite and all the other inorganic including solid-acid catalysts mentioned above are all examples of uniform heterogeneous ~~talysts,'~~'~ where the active sites such as bridging hydroxys of the type Si-O(H)-A1 are distributed in a spatially uniform and accessible (to reactants) fashion throughout the bulk of the solid.They have other attributes notably exceptionally large surface areas often in excess of 600 m2 g-' ninety-five percent (or more depending on crystal size) of which is inside the solid. In effect they are solid catalysts with three-dimensional surfaces permeating the entire material and accessible to all reactant (and product) molecules small enough to enter the apertures at their outer surfaces. This being so these uniform heterogeneous catalysts are amenable to the entire panoply of spectroscopic diffraction and scattering techniques refined over the years by physicists and chemists.Such catalysts present unrivalled-in the experience of inorga- nic and physical chemists-opportunities to explore the relationship between structure and catalytic performance much in the way that modern enzymologists pursue their investigations (described in Section 2). In situ X-ray crystallography is well suited to such work (as it was originally for lysozyme and is for other enzymes). But there are some serious practical obstacles that have to be surmounted when as is often the case the concentration of the active site is low. X-Ray diffraction alone even when imaginative use'7 is made of anomolous scat- tering is not sufficiently sensitive. Take for example CoALPO-18 (Plate 7) which is a good catalyst for converting methanol to light alkene~.'~.'~ The concentration of the Co" ions in the framework of the Plate 8 Experimental equipment of the combined Qu-EXAFS-XRD setup.The spectroscopy and diffraction data obtained for cordierite are shown by the two inserts relating to the two detector systems. (From J. M. Thomas et d2l) J. M. Thomas C19 parent ALP0 seldom exceeds a few percent too low to be directly addressable by X-ray diffraction. We may however use X-ray absorption thanks to the availability of syn- chrotron radiation. Moreover we have evolved methods of probing catalysts of this kind under operating conditions using combined X-ray diffraction and X-ray absorption spectroscopy (XRD and XAS) recorded with the same sample under realistic conditions of catalytic use (Plate -22 Combined studies in parallel with infrared absorption spectroscopy yield results such as those shown in Plate 7 where the changed bond lengths associated with conversion of the Co" to Co"' valence states while still retained in the framework may be quantitatively specified.Typical results obtained by the com- bined in situ XRD and XAS techniques recorded during the calcination of the CoALPO- 18 are shown in Fig. 5. There is every reason to believe that this dual in situ approach to the study of open- structure (as well as ~ther'~.~~) catalysts will provide fresh insights into the mechanisms of heterogeneous catalysis. A specific example of such new insights has recently been seen in the study of metal-ion catalysts anchored onto the interior surfaces of meso- porous supports (see below).The dual in situ XRD-XAS approach is also of great value in tracking the synthesis-the actual act of crystallization from a precursor gel-of new zeolitic cata- lysts as outlined in the next section 7900 Fig. 5 Combined Qu-FLEXAFS-XRD following the calcination of CoAPO-18. (a)Shift in the K edge of cobalt (measured in fluorescence) during initial heating in air to burn off the template and during reduction to activate the catalyst. (h)XRD patterns during initial heating showing virtually no change except for a drop in background on removal of the template. (From J. M. Thomas et a1.2l) c20 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst 6.Synchrotron-based Studies for Tracking the Synthesis of Uniform Heterogeneous Catalysts Progress in our understanding of uniform heterogeneous catalysts rests almost as much on our ability to monitor in atomic detail the formation or synthesis of such catalysts as well as the structural changes they undergo during operation. A start has recently been made by my group working closely with Professor G. N. Greaves and his colleagues at the EPSRC Daresbury Laboratory. Measurements that are experimentally demanding with conventional laboratory- based X-ray sources often become readily accessible when carried out with synchrotron radiation. This is so in the study of structural changes in solid catalysts under operation conditions.It is also true in the study of the nucleation and growth of crystals from solutions melts and gels. With photon fluxes from a typical synchrotron exit port of the order of 1013 s-' and more importantly with high-energy X-rays (up to 140 keV) capable of penetrating stainless-steel containers it is relatively straightforward to study condensed phases using appropriate high-temperature and/or high-pressure cells without serious attenuation in the intensity. Moreover taking advantage of solid-state detectors the white radiation of synchrotrons permits the rapid buildup of time-resolved diffraction patterns (under isothermal conditions) that allows the ready tracking of structural change occurring in a cell that is held in a fixed position.Fig. 6 shows the energy-dispersive X-ray diffraction (EDXRD) geometry available on station 16.4 at the 2 GeV Daresbury Synchrotron Radiation Source. The EDXRD facility is illuminated with radiation from a 6 T superconducting wiggler. The stainless-steel autoclave shown in the figure can be operated up to temperatures of ca. 250"C permitting the in situ character-ization during hydrothermal synthesis of most of the zeolitic materials. The solid-state detector is inclined at a fixed 20 angle chosen so that from the photon energies available (7-1 40 keV) representative crystallographic spacings for zeolite structures are readily identifiable. With SRS electron beam currents of around 200 mA energy-dispersed pat- terns were integrated every 2 min in order to achieve adequate statistics for phase identi- fication and the recording of reaction kinetics for the synthesis rates encountered here.To piressure gauge4 Oven -A drlu valve / Solid-state detect0r CoIIirnat0r Post Sample A Polychromati Synchrotron Radiation liquor-gel) Window-hole in oven liner Ct-;nlnee rtnnl otaiiiicaa ~LCCI autoclave Fig. 6 Schematic diagram of the experimental set-up for energy-dispersive X-ray diffraction. Syn- thesis takes place in the heated stainless-steel autoclave and the crystallization is detected with a single-element (Canberra) Ge solid-state detector. This is inclined at a fixed 28 angle of 1.4556 f0.0001" and calibrated using NBS 640b Si. The integration time for each frame is 2 min. (From F.Rey et J. M. Thomas c21 Time-resolved X-ray diffraction patterns of the catalyst CoALPO-5 as well as of its parent ALPO-5 have been collected (and publi~hed~~) and these clearly show the change from structureless diffraction patterns exhibited by the mother liquor to highly structured ones corresponding to the appearance of the ALPO-5 phase along with a co-existent chabazitic equivalent of CoALPO-5. Very recently my colleagues and I have used the combined XRD-XAS approach schematized in Fig. 7. With the capillary reaction vessel housed in an appropriately designed furnace both XRD patterns (comparable to those obtained with the set-up shown in Fig. 6) as well as richly detailed X-ray absorption spectra are obtainable (see Fig. 8).The pre-edge X-ray absorption spectra and also the XANES and EXAFS infor- mation (taken in conjunction with the XRD results) unmistakably reveal25 that prior to the onset of crystallization octahedral Co" ions in the precursor (templated) gel become tetrahedrally coordinated. 7. Mesoporous Catalysts Plate 9 summarizes progress made in the last forty years in synthesizing open-structure solids with potential as catalysts. Zeolite A produced by the Linde (Union Carbide) Company in the early 1950s has pore dimensions around 4 A (depending upon the exchangeable cation). These are of great commercial applicability-in detergency and water-softening especially-but they have essentially zero prospects as viable catalysts. DAF-1 (Davy Faraday One) already described above and discovered here in 1992 is in quite a different category and has demonstrated value-but not much staying power- as an acid catalyst for the isomerization but-1 -ene to isobutene (2-methylpropene) the material first synthesized by Faraday.MCM-41 (Mobil Catalytic Material number 41) first reported in 199226) is of even greater potential in a new generation of shape-selective catalysts and as supports for metal-based catalytic centres. Fig. 7 Schematic representation of set-up used by Sanker Rey Thomas and Greaves (in ref. 24) for combined XRD-XAS in situ measurements to follow crystallization of molecular sieve cata- lysts from liquids and gels. (See also Fig. 3 of ref. 21.) c22 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst 15.9 ~"c"7.69774 Fig.8 Stacked (a)XRD patterns and (b)X-ray absorption spectra during the crystallization of a -.--. .. .. * .. . . . . _. -_ LoALYW-3 catalyst from the liquid phase studied using the set-up shown In k'ig. 7. From the pre-edge (Is + 3d) transition the EXAFS and the shape of the XANES peaks there is no doubt that the cobalt ion originally in octahedral coordination becomes tetrahedrally bonded prior to the onset of crystallization. Mesoporous siliceous solids with channel apertures from 25 to 100 A have opened up new possibilities in heterogeneous catalysis. The large diameter (ca. 30 A) channels of the so-called MCM-41 mesoporous silicas that we and others have used for selective oxidation permit in principle the direct grafting of complete metal complexes and organometallic moieties onto the inner walls of these high-surface-area (typically >800 m2 8-l) solids.This opens routes to the preparation of novel catalysts consisting of J. M. Thomas C23 Plate 9 Zeolite A first prepared in the mid 1950s is a synthetic molecular sieve (ca. 4 A diameter apertures) of little catalytic significance. DAF-1 (Davy Faraday one) first synthesized in 1992 (diameter of apertures >7 A) as well as MCM-41 (Mobil catalytic material 41) a mesoporous solid has great potential as a catalyst. Plate 10 An example of a tethered metallic catalyst prepared by T. Maschmeyer (unpublished) at the Davy Faraday Research Laboratory (see also ref. 27). C24 Tales of’ Tortured Ecstasy Probing the Secrets of’Solid Catalyst large concentrations of accessible well-spaced and structurally well defined active sites.Maschmeyer et al. in these laboratories have described the production of titanocene- derived catalyst precursors anchored to the inner walls of the MCM-41 and its conver- sion (monitored by X-ray absorption spectroscopy) to a powerful catalyst for the epoxidation of cy~lohexene.~~ Furthermore this catalyst is also active in the oxidation of other important bulkier reactants such as pinene. X-Ray absorption spectroscopy not only helps (as does in situ FTIR) to identify key intermediates in the epoxidation catalysis but also reveals that in the activated state of the precursor Ti is four coordi- nated with no evidence for a titanyl (Ti=O) bond.During catalytic reaction it is six coordinated. With MCM-41 based catalysts it is possible not only to anchor a reactive (catalytic) centre designed to order according to the principles of organometallic chemistry but also to tether them as schematized in Plate 10. Here the active site is situated at the extremity of the tether and is free to flutter in the molecular breeze during the process of catalytic conversion. By deliberately restricting the spatial freedom in the vicinity of the active centre (see Plate 11) it should be possible to design highly stereo-selective (enantiometric) catalysts. Such possibilities are now under active investigation at the Davy Faraday Research Laboratory. One other highly convenient property of MCM-41 mesoporous solids is that they may be modified in subtle and catalytically useful ways during synthesis.Plate 12 taken from the work of my colleagues,28 shows how hetero-atoms may be incorporated into the siliceous framework of MCM-41. When titanium is anchored in the wall of such mesoporous hosts facile epoxidation (of cyclohexene as shown in Fig. 9) ensues. Plate 11 Deliberate restriction of spatial freedom at the active centre grafted onto the inner walls of a mesoporous silica offers scope for enantiomeric catalysis (Maschmeyer et al. in preparation.) J. M. Thomas C25 Plate 12 Idealised metal-containing MCM-41 structure generated by computer modelling methods (from Frey et ~1.~') (4 (d Fig. 9 Schematic representation of the preparation of the grafted catalyst (a) the support (b) anchoring reaction (c) calcination and (d) reversible reaction of the organometallic-derived Ti-MCM-41 epoxidation catalyst with water upon exposure to the atmosphere C26 Tales of Tortured Ecstasy Probing the Secrets of Solid Catalyst 8.Epilogue In summary we observe that both microporous and mesoporous crystalline inorganic solids have greatly extended the scope and nature of heterogeneous catalysis. Homoge- neous catalysts may be readily heterogenized so that the separate advantages of homo- geneous catalysts on the one hand-their high specificity-and heterogeneous catalysts on the other-their built-in ease of separation of product from reactant-may be jointly harnessed in a synergistic mode.We also note that in situ methods of probing the structure of catalysts under oper- ating conditions are rendered feasible owing to the availability of synchrotron radiation. All this enables greater insights to be gained into the mechansim of catalytic reaction insights which are often enriched by modern computational chemical techniques. Table 1 summarizes my own group’s overall strategy. In conclusion I cannot help recalling what Humphrey Davy standing at this spot said in 1803 Of modern chemistry it may be said that its origins are pleasure its growth know- ledge its objects truth beauty and utility. I hope that I have demonstrated that this admirable definition is as valid now as it was when it was first enunciated. Table 1 Overall strategy of work on heterogeneous catalysis at the Davy Faraday Research Laboratory 0 Synthesis of new micro- and meso-porous catalysts 0 Development of techniques and tools 0 Characterisation 0 Performance 0 Mechanism +computation 0 Applications (patents?) 0 Collaboration with industrial partners References I D.C. Phillips Proc. Natl. Acad. Sci. USA 1967 57 484. 2 J. R. Knowles Nature (London) 1991,350 121. 3 V. T. D. Souza K. Hanahusa T. O’Leary R. C. Gradewood and M. L. Bender Biochem. Biophys. Res. Commun. 1985 129 727. 4 R. Breslow Acc. Chem. Rex 1995 28 146. 5 P. B. Weisz Proc. 6th Int. Conyr. Catalysis Kodansha Tokyo 1990 p. 1. 6 J. M. Thomas Anyew. Chem. Int. Ed. Enyl. 1994,33,913. 7 J. M. Thomas Sri. Americun 1992 266 85. 8 K.I. Zamaraev 7op. Catal. 1996 in the press. 9 J. A. Cusumano CHEMTECH. 1992,22,482. 10 J. M. Thomas and K. I. Zamaraev Angew. Chem. 1994 106,316. 11 R. G. Bell D. W. Lewis C. R. A. Callow J. M. Thomas P. Voigl and C. M. Freeman in Zeolites and Related Microporous Muterial\ Slate of lhe Art ed. J. Weitkamp H. G. Karge H. Pfeiffer and W. Hoderich Elsevier 1994 vol. 84 p. 127. 12 D. W. Lewis C. R. A. Catlow and C. M. Freeman J. Phys. Chem. 1995,99 11 194. 13 D. W. Lewis C. R. A. Catlow and J. M. Thomas submitted. 14 J. Chen and J. M. Thomas J. (‘hem. Soc. Chem. Commun. 1994,603. 15 J. M. Thomas Angew. Chem. Inl. Ed. Engl. 1988 27 1673. 16 J. M. Thomas J. Chen and A. R. George Chem. Rr. 1992,28,991. 17 A. K. Cheetham and A. P. Wilkinson Anyew. Chem.Int. Ed. Engl. 1992,31 1559. 18 J. M. Thomas G. Sankar P. A. Wright J. Chen L. Marchese and G. N. Greaves Angew. Chem. Int. Ed. Engl. 1994 33 639. J. M. Thomas C27 19 J. W. Couves J. M. Thomas G. N. Greaves R. H. Jones D. Waller A. J. Dent and G. E. Derbyshire Nature (London) 1991,354,465. 20 G. Sankar P. A. Wright S. Natarajan J. M. Thomas G. N. Greaves A. J. Dent B. R. Dobson C. A. Ramsdale R. H. Jones J. Phys. Chem. 1993,97,9550. 21 J. M. Thomas G. N. Greaves and C. R. A. Catlow Nucl. Instruments acd Methods in Physical Research 1995 B97 1. 22 P. A. Barrett G. Sankar J. M. Thomas and C. R. A. Catlow J. Phys. Chem. Solids 1995,56 1395. 23 B. S. Clausen L. Grabek G. Steffensan P. L. Hausen and H. Topsrae Catal. Lett. 1993,20,23. 24 F.Rey G. Sankar J. M. Thomas P. A. Barrett D. W. Lewis C. R. A. Catlow S. M. Clark and G. N. Greaves Chem. Muter. 1995,7 1435; see also J. S. 0.Evans R. J. Francis D. O’Hare S. J. Price S. M. Clark J. Flaherty J. Gordon A. Nield and C. C. Tang Rev. Sci. Instrum. 1995 66 2442. 25 G. Sankar F. Rey J. M. Thomas and G. N. Greaves J. Chem. SOC. Chem. Commun. 1995 in the press. 26 C. T. Kresge M. E. Leonowicz W. J. Roth J. C. Vartuli J. S. Beck Nature (London) 1992,359 710. 27 T. Maschmeyer F. Rey G. Sankar and J. M. Thomas Nature (London) 1995,378 159. 28 F. Rey G. Sankar T. Maschmeyer J. M. Thomas R. G. Bell G. N. Greaves Top. Cataf. 1996 in the press. Faraday Discussion 100 Celebration Paper Presented 24th April 1995