catalysis on surfaces

When atoms on the surface of asolid interact with moleculesof a gas or liquid, they may al-

ter the structure of the molecules everso slightly, thereby promoting unusualchemical reactions. Indeed, by inves-tigating the interactions between mol-ecules and the surfaces of solids, re-searchers have learned to synthesize amyriad of novel substances, developchemical processes of unprecedentedeÛciency and remove pollutants fromthe environment. Yet the study of chem-istry on solid surfaces will have themost impact, in my opinion, on the tech-nology of catalystsÑsubstances that in-crease the rates of desirable chemicalreactions at the expense of others.

The rise of surface chemistry is a re-cent event. Some 15 years ago the sub-ject was just emerging as a branch ofscience, and only a few methods wereavailable for the analysis of complexsurface reactions. In recent years, I havewatched the Þeld advance rapidly astechniques were devised to create de-tailed pictures of surface reactions. Byusing the tools of surface chemistry,researchers can now view the action ofa catalyst at the molecular level.

To illustrate how surface chemis-try has advanced our understanding of catalytic reactions, I have focused on three cases that are simple yet havegreat scientiÞc, technological and so-cial signiÞcance. The Þrst is the synthe-sis of ammonia from airÑa process

that made possible the large-scale pro-duction of fertilizers and explosives.Although chemists invented the com-mercial process for ammonia synthesisin the early part of this century, theydid not understand the nuances of thecatalytic reaction until the techniquesof surface chemistry were applied sev-eral years ago.

The second case is the breakdown of nitric oxide in the exhaust of auto-mobiles. By reducing the emissions of nitric oxide from car exhaust, chem-ists have helped alleviate the problemof acid rain and mitigate the eÝects ofother harmful pollutants in the atmo-sphere. The Þnal case is the removal ofsulfur from fossil fuels, a process thatalso has implications for the atmo-sphere. The desulfurization of fuels in-volves particularly complicated surfacereactions, and the investigation of suchphenomena represents the frontier insurface chemistry.

In 1909 Fritz Haber, a German chem-ist, found an eÛcient way to synthe-size ammonia from nitrogen and

hydrogen gas. Five years later commer-cial production began at Badische Anil-in- & Soda-Fabrik (BASF)Ñan event thatlater proved to be a bane and a boon toall humanity. The Haber process pro-

vided Germany with the main ingredi-ent for manufacturing explosives, suchas nitroglycerine, for use in World WarI. Previously, Germany had made explo-sives from saltpeter mined in Chile. Ifthe supply had been blocked and theHaber process had not been invent-ed, some historians argue, World War Iprobably would have been shortened.

The Haber process also revolution-ized agriculture by fostering the pro-duction of mass quantities of cheap fer-tilizer. As a direct consequence, farm-ers achieved higher crop yields thanever before and greatly increased theglobal food supply.

To produce ammonia eÛciently, Ha-ber added iron as a catalyst to a mix-ture of nitrogen and hydrogen gas andthen subjected the reaction vessel topressures of some 100 atmospheresand to temperatures around 500 de-grees Celsius. As nitrogen and hydro-gen circulated over the catalyst, ammo-nia condensed out of the mixture.

HaberÕs major contribution was theidentiÞcation of a good catalyst for thesynthesis of ammonia. He tested morethan 1,000 materials before settling oniron. Today manufacturers of ammoniause a catalyst consisting of iron, potas-sium and calcium, which performs bet-ter than iron alone.

74 SCIENTIFIC AMERICAN April 1993

CYNTHIA M. FRIEND is professor ofchemistry at Harvard University and atrustee of RadcliÝe College. After earn-ing her Ph.D. in chemistry from the Uni-versity of California, Berkeley, in 1981,she conducted research in surface chem-istry at Stanford University. Her workhas earned her the Garvin Medal of theAmerican Chemical Society and the Pres-idential Young Investigator Award. Forthe past decade at Harvard, Friend hasbeen a pioneer in applying surface chem-istry techniques to investigate catalysis.

CATALYST made of the metal rhodiumis used in the exhaust systems of au-tomobiles to help transform nitric ox-ide (NO) and carbon monoxide (CO) intoharmless gases. The tools of surfacechemistry enabled researchers, in 1986,to learn how the rhodium catalyst worksat the molecular level. Carbon monoxide(1) binds to the rhodium surface (2 ).When nitric oxide does the same, it dis-sociates into oxygen and nitrogen (3 ).The bound oxygen reacts with the COto form carbon dioxide (CO2) (4 ). If an-other CO and NO land close to the re-maining bound nitrogen (5Ð8 ), the rhodi-um catalyst promotes the formation of asecond carbon dioxide molecule and anitrogen molecule (N2).

Catalysis on SurfacesScientists can now observe how solids interact with individual molecules

to speed reactions. Information about these catalysts is being usedto improve everything from materials synthesis to pollution control

by Cynthia M. Friend

OXYGEN

CARBON RHODIUM

NITROGEN

1

Copyright 1993 Scientific American, Inc.

Iron, like all catalysts, increases therate of a desirable reactionÑthe syn-thesis of ammonia, in this caseÑwhilesimultaneously diminishing the chanceof undesirable reactionsÑfor instance,the recombination of nitrogen atoms toform nitrogen gas. The catalytic actionoccurs when reactant molecules tem-porarily bind to the surface of the solid.This binding alters the forces betweenatoms, thereby changing the energeticrequirements for the reaction. Conse-quently, in the presence of a catalyst,some processes will be favored, andothers will not. Catalysts are never con-sumed or produced during the reaction.

For decades after the commercialprocess for ammonia synthesis was in-vented, researchers did not understandat the molecular level how the ironworked as a catalyst. The tools of sur-face chemistry have recently made itpossible, however, to conÞrm that theprimary function of the catalyst is to fa-cilitate the dissociation of nitrogen gas(N2). The bond that holds the N2 mol-ecule together is very strong and there-fore will not break unless the moleculecontains a substantial amount of en-ergy (941,000 joules per mole of nitro-gen, to be exact). As the energy contentof an N2 molecule increases, the aver-age distance between the nitrogen at-oms grows. The process is analogousto climbing a mountain. Just as a hikermust burn calories to get to the top ofa mountain, the N2 molecule must con-tain energy to surmount a high energybarrier. And just as a fast-moving hikerreaches the summit quickly, a moleculethat has a lot of energy will more readi-ly pass over the barrier. (The number ofhigh-energy molecules in a gas increas-es in proportion to the temperature.)

The rate at which N2 dissociates de-pends on the rate at which the mol-ecules can surmount the barrier, andthese rates, in turn, are related to theavailable energy, or temperature, of themolecules. Ammonia does not formreadily, however, if nitrogen and hydro-gen molecules are heated simply to thepoint at which they dissociate. At such

temperatures, ammonia itself breaks upas quickly as it is formed.

Yet if nitrogen molecules interactwith iron atoms on a surface, they willdissociate at relatively low tempera-tures. Iron catalyzes the dissociation ofN2; in other words, it lowers the energybarrier so that the two nitrogen atomswill separate easily. The iron atoms ac-complish this by donating electrons tothe nitrogen molecule. As a result, theiron atoms form a chemical bond withthe nitrogen molecule, and in a recipro-cal manner, the bond between the twonitrogen atoms in the molecule is weak-ened. The weak nitrogen-nitrogen bondcorresponds to a low energy barrier andfacilitates N2 dissociation. This processis typical of how catalysts work.

In 1984 Gerhard Ertl, Michael Grunze and Min-Chi Tsai and theirco-workers, then at the University

of Munich, obtained the Þrst direct evi-dence for the weakening of the nitro-gen-nitrogen bond. Their work involvedthe examination of samples made fromsingle crystals of iron under vacuumconditionsÑthat is, an environment freefrom extraneous gases. Unlike the at-oms in ordinary iron particles, those ina single crystal form a perfectly or-dered lattice. Because some chemicalreactions are extremely sensitive to thearrangement of atoms on the surface,the use of single crystals permits themethodical study of these eÝects. Re-searchers can even vary the surface ar-rangement by cutting crystals at vari-ous angles.

To observe the reactions, the Munichgroup used two of the most importanttools in surface chemistry: x-ray photo-electron spectroscopy and high-resolu-tion electron energy loss (HREEL) spec-troscopy. By probing the forces betweenatoms, these techniques let chemistsÒseeÓ the changes that take place whena molecule interacts with the surface.

X-ray photoelectron spectroscopy,which was Þrst developed in the ear-ly 1960s, enables researchers to mea-sure the energy of strongly bound elec-

SCIENTIFIC AMERICAN April 1993 75

2 3 4

5

6

8

7



trons in molecules. In this method, x-rays hit the experimental sample, there-by knocking electrons from the atomsof the surface and from any moleculesbound to the surface. The energy ofthese electrons is sensitive to the lo-cal chemical environment, and thus thephotoelectron technique yields infor-mation about bonding (for instance,the breaking of the nitrogen-nitrogenbond in N2).

HREEL spectroscopy, which was in-vented in the late 1970s and reÞned in the mid-1980s, probes the forces between atoms in molecules by mea-suring the energies of molecular vi-brations. The atoms in a molecule can be envisioned as balls held together by springs. The springs represent thechemical bonds; the balls are the atom-ic nuclei. The strength of the bond isrelated to the stiÝness of the springand therefore depends on the energy re-quired to stretch the atoms apart or tochange the relative angle of the bond.

The results of HREEL spectroscopyare rather straightforward to interpretin the case of a nitrogen mole-cule. The only way the nitro-gen molecule can vibrate isthrough the stretching andcompression of the nitrogen-nitrogen bond; a nitrogen mol-ecule has a single vibrationalmode, to use the jargon ofchemistry. If a nitrogen mole-cule is bombarded with elec-trons, it will vibrate and thenrob them of a characteristicamount of energy. These elec-trons can be detected to revealthe energy of the vibrationsand thereby the strength of thespring, or bond, that holds themolecule together. The sametechnique can be used for ni-trogen molecules that bond to the surface, but the results are bit more complicated. Anitrogen molecule bound to asurface has more than one vi-brational mode because thereare additional ÒspringsÓ asso-ciated with the metal-nitrogenbonds.

HREEL spectroscopy enabledthe Munich group to demon-strate that when an N2 mole-cule is adsorbed on an ironsurface, the energy requiredto stretch the nitrogen atomsapart is decreased dramatical-ly compared with the energyneeded to lengthen the bondof a nitrogen molecule ßoat-ing freely.

Besides weakening the nitro-gen-nitrogen bond, the tech-

niques of surface chemistry have shownthat iron performs two other importantfunctions in the synthesis of ammonia.First, when hydrogen molecules (H2) in-teract with the iron surface, the hydro-gen-hydrogen bond is weakened. Hy-drogen atoms are thus easily freed andbind to the surface. Second, the iron at-oms conÞne the hydrogen and nitrogenatoms on the surface so that they mayreact to form NH, NH2 and Þnally NH3,ammonia, the desired product.

The chance that a bond will form be-tween the nitrogen and hydrogen at-oms depends on how well the nitrogenattaches to the iron. In fact, the iron-ni-trogen bond is suÛciently strong sothat nitrogen does not recombine toform N2, and it is suÛciently weak sothat the nitrogen can combine with hy-drogen. Several metals are more eÝec-tive than iron in dissociating nitrogenmolecules into their component atoms,but typically the atoms are bound sostrongly to the metal that they do notcombine with hydrogen to form ammo-nia. These insights and others are help-

ing researchers to Þgure out why onecatalyst is better than another and ul-timately may lead to the discovery ofpractical catalysts that will improve thecommercial synthesis of ammonia andenhance the eÛciency of related indus-trial processes.

When applying the results ofsurface studies to commercialcatalysis, investigators face a

diÛcult problem. The tools of surfacechemistry require vacuum conditions,whereas most practical catalysts mustwork in a high-pressure environment.

Indeed, surface chemists have an ex-treme deÞnition of clean. By their stan-dards, the surface of a sample must remain relatively free of contaminantsfor at least a few hours. It must there-fore be stored under ultrahigh vacuumconditionsÑthat is, pressures 10 tril-lion times lower than that of the atmo-sphere. The extremely low pressuresenable chemists to study surfaces ofknown composition and to introduce re-actants in a purposeful way. Such stud-

ies could never be done, say, inthe open air because a perfect-ly clean metal surface wouldbe covered with gas moleculesin about a billionth of second.Even if the pressure were re-duced to a billionth of an at-mosphere, the surface would becovered in about one second,insuÛcient time to perform anexperiment.

Researchers need to be care-ful about how they apply theÞndings of surface chemistryto high-pressure catalytic re-actions. The great disparity inpressure means a vast diÝer-ence in the number of gas mol-ecules hitting a catalyst at anyone time, and therefore the kinetics of the laboratory re-action may diÝer from the dynamics of the high-pressurereaction. Although chemistscould not initially Þgure outhow to compensate for the so-called pressure gap, they re-cently learned to extrapolatethe behavior of high-pressurecatalytic reactions from ideal-ized surface studies.

An elegant example of howthe pressure gap can be bridgedis the story of a set of surfacechemistry experiments aimedat improving the technology of catalytic converters for auto-mobiles. The primary functionof the converters is to removenitric oxide (NO) and carbonmonoxide (CO) from automo-

FRITZ HABER developed, in 1909, the Þrst commercialprocess for synthesizing ammonia. One of his importantinsights was that iron catalyzes the reaction that trans-forms hydrogen and nitrogen gas from ammonia.


bile exhaust. Nitric oxide reacts rap-idly with air to form nitrogen-oxygencompounds (NOx), which are harmfulto the environment, notably becausethey contribute to acid rain. Carbonmonoxide is also extremely toxic tomost forms of life. By transforming NOand CO from car exhaust into lessharmful products, catalytic convertershave dramatically reduced the levels ofthese contaminants. The technology hasbeen very successful, yet chemists con-tinue to search for better catalysts, re-alizing that small improvements canmean tremendous beneÞts for the en-tire environment.

A typical catalytic converter consistsof particles of platinum (Pt) and rhodi-um (Rh) deposited on a ceramic honey-comb. The platinum and rhodium parti-cles catalyze the reactions that removeNOx, CO and uncombusted hydrocar-bons from car exhaust. The ceramichoneycomb and small particle size servethe dual function of maximizing theexposure of the metals to the exhaustfumes and minimizing the quantity of

platinum and rhodiumÑtwo very ex-pensive metals.

In the mid-1980s researchers at Gen-eral Motors and elsewhere set out to in-vestigate how rhodium interacts withthe nitric oxide and carbon monoxidefrom car exhausts. To do so, they stud-ied reactions of NO and CO on singlecrystals of rhodium. Employing HREELspectroscopy and other methods, theworkers identiÞed the key steps in thebreakdown of nitric oxide, and they de-termined how the arrangement of rho-dium atoms on the surface inßuencedcatalysis.

Yet it was not clear at Þrst whetherthese results were applicable to thetechnology of catalytic converters: theexperiments were conducted under vac-uum conditions, whereas the rhodiumparticles used in catalytic converters areexposed to gases at high pressure. Todemonstrate that practical informationcould be gained from the surface chem-istry studies, the GM workers testedthe rate of NO reduction on a rhodiumsurface in an environment that repli-

cated the pressure conditions in a cat-alytic converter. By analyzing the actionof rhodium in vacuum conditions andunder high pressure, the researchers de-vised a mathematical model of the cat-alytic process. The model has allowedthe results of the surface studies to beused in determining how new kinds ofcatalytic materials will operate underhigh pressures.

The GM group also found that thedissociation of NO was sensitive to thearrangement of atoms on the rhodi-um surface. They arrived at this con-clusion by using infrared spectrosco-py. This technique is similar to HREELspectroscopy, but it uses infrared ra-diation, instead of electrons, to causemolecular vibrations. Infrared spectros-copy has an advantage, however: it canbe used both under vacuum conditionsand at high pressures. Researchers atGM were therefore able to study, athigh and low pressure, how nitric ox-ide interacts with irregular particles ofrhodium and how it binds to a rho-dium surface in which the atoms are ar-


a

1a 1b

1 2

rrangement of atoms on the surfaceof a catalyst depends on how the

atoms are packed together and how thematerial is cut. Two typical atomic lattic-es are the face-centered cubic (1) and thebody-centered cubic (2). Different cuts (a,b) through such lattices yield differentsurface arrangements (1a, 1b, 2a, 2b).

ASurface Geometry

b

2a 2b


ranged in a hexagonal pattern. Infraredspectroscopy identiÞed diÝerences inthe vibrations of NO on the two typesof surfaces, proving that surface struc-ture inßuences bonding and ultimatelythe catalytic process.

To study, in even Þner detail, howsurface structure aÝects such reactionsas the dissociation of nitric oxide onrhodium, chemists have recently begunto turn to the scanning tunneling micro-scope. In 1986 Gerd K. Binnig and Hein-rich Rohrer won the Nobel Prize for in-venting the device because it revolu-tionized the study of surface structure.To produce atomic-scale images of asurface, a scanning tunneling micro-scope positions a Þne metal stylus onlya few angstroms above a sample andthen moves the stylus across the sur-face. The microscope senses electronsas they pass between the stylus and thesurface. Because the ßow of electrons isrelated to the height of the atoms onthe surface, the information can betranslated into images.

Scanning tunneling microscopy is justbeginning to provide insights into howsurface structure inßuences catalysis. Itshould permit chemists to identify thesurface structures that optimize the per-formance of catalysts.

Surface chemistry has also helpedresearchers understand the cat-alytic process for removing sul-

fur from fossil fuels. Traces of sulfur infossil fuels harm the environment intwo ways. First, when fuel is burned inan engine, some sulfur reacts with air toform sulfur-oxygen compounds, whichcontribute to acid rain. Second, sulfursticks to the platinum and rhodium in catalytic converters, thereby shuttingdown their activity and indirectly in-creasing the emission of NO and CO.

Sulfur is removed from petrochemi-cals at reÞneries as crude oil is trans-formed into such useful hydrocarbonsas octane. Ideally, the desulfurizationprocess should extract all the sulfurwithout destroying the valuable hydro-carbons. The process therefore requiresa catalyst that encourages desulfuri-zation but discourages the breakdown of pure hydrocarbons. At present, the best catalyst for desulfuriza-tion is a mixture of molyb-denum, cobalt and sulfur it-self. Because the material has a very complicated structure,chemists have had diÛcultyÞguring out how petrochemi-cals interact with the catalyst.It is not clear whether all com-ponents of the catalyst are ef-fective in promoting desulfuri-zation. For example, one com-

ponent, cobalt sulÞde, is not thought tobe the active catalytic material.

My colleagues and I at Harvard Uni-versity have investigated the role of mo-lybdenum in desulfurization. We are in-terested in the question of how petro-chemicals interact with molybdenum invarious forms and in combination withcobalt and sulfur. Our goal is to con-struct a general model that can be usedto predict the products and rates of re-actions for all the common sulfur-con-taining hydrocarbons.

During the past eight years, we for-mulated such a general model. It de-scribes how thiols, a major class of sul-fur-containing petrochemicals, interactwith various molybdenum surfaces. Tobe precise, thiols consist of a hydrogenatom attached to a sulfur atom, whichin turn is bonded to some combinationof carbon and hydrogen. To study theinteraction between thiols and molyb-denum catalysts, we used crystals ofpure molybdenum so that the atomson the surface would form a highly or-dered pattern. The regular structure lim-its the number of diÝerent kinds ofsites available for bonding. By introduc-ing various components to the molyb-denum surface in a systematic way, wecould then infer the role of each com-ponent in desulfurization.

Many years ago chemists made anastonishing discovery about molybde-num-induced desulfurization. The per-formance of pure molybdenum is actu-ally enhanced when it is contaminatedwith sulfur, whereas most metals losetheir ability to catalyze reactions whensulfur bonds to them. To study the role

of surface sulfur, we compared how thi-ols react on clean molybdenum surfaceswith how they perform on molybdenumcovered with an ordered array of sulfur.Thiols interact with molybdenum cata-lysts to produce hydrocarbons, hydro-gen gas, surface carbon and sulfur. Forexample, ethanethiol (CH3(CH2)SH) isbroken down into sulfur and one of twohydrocarbons: ethane (CH3CH3) andethene (CH2CH2). Ideally, the catalystshould promote only the production ofhydrocarbons and the removal of sul-fur; it should discourage the synthesisof surface carbon and hydrogen gas,which have little value.

The molybdenum surface coated withsulfur removes sulfur from thiols moreslowly than the clean surface can, but at the same time, the sulfur on the mo-lybdenum surface decreases the rate of reactions that lead to undesirable prod-ucts, thereby increasing the yield of use-ful hydrocarbons. Indeed, sulfur depos-ited on a molybdenum surface is bene-Þcial for the desulfurization not only ofthiols but also of other petrochemicals.

One major objective of our re-search has been to discover theimportant intermediate mole-

cules that are produced as sulfur is re-moved from thiols using a molybde-num catalyst. We attempted to identifythe intermediates by applying a combi-

nation of x-ray photoelectronand HREEL spectroscopies.Our analysis was complicatedby the fact that thiols consistof a large number of atomsand may react with the sur-face in many diÝerent ways.Furthermore, the large size ofthe reactants made it diÛcultto interpret spectroscopic re-sults; molecules composed ofmany atoms have many vibra-


SULFUR-CONTAINING MOLECULE binds to a molybdenumsurface. The orientation of the molecule on the surfacewas deduced using electron spectroscopy.

MOLYBDENUM catalyzes a reaction thatremoves sulfur from fuels. When a mol-

1

SULFUR HYDROGEN

CARBON

MOLYBDENUM


tional and electronic states. Neverthe-less, we were able to distinguish sever-al of the steps in desulfurization of thi-ols on molybdenum.

A detailed analysis of spectroscop-ic data can reveal the identity of an in-termediate molecule on the surface.The data can also provide informationabout how the bonds within an inter-mediate change as the molecule inter-acts with the surface. In some cases, wecan even deduce the orientation of theintermediates from spectroscopic stud-ies. Such structural information is ex-ceedingly important in creating theo-retical models of the bonding of thiolsand their associated intermediates tosurfaces. Infrared spectroscopy is par-ticularly versatile as a structural probebecause the vibrations within a mole-cule that are detected in an infrared ex-periment are sensitive to the symmetryof the molecule.

Such successes have encouraged sur-face chemists to extend their experi-mental reach by inventing additionaltools for determining molecular struc-ture on surfaces. Moreover, this eÝort isone of the frontiers in surface chemis-try. Several of the emerging techniquesrequire radiation from a huge electronaccelerator known as a synchrotron. Byaccelerating electrons around a circu-lar track to speeds approaching that oflight, a synchrotron can generate pow-erful beams of x-rays and other kindsof radiation. To utilize the x-ray beamfor spectroscopy, researchers must de-sign their experiments and measure-ment apparatus so that they interfacewith the x-ray source.

During the past few years, we haveused synchrotron, infrared and othertechniques to Þgure out what the stag-es of the desulfurization are and whatsteps are crucial in setting the overallrate of the reaction. The Þrst step is the

cleavage of the sulfur-hydrogen bond.We found that this step occurs veryrapidly and is favored because bothsulfur and hydrogen form strong bondsto the molybdenum surface. In sub-sequent steps, the carbon-sulfur bondmust break, and one carbon-hydrogenbond is either formed or broken to yieldthe hydrocarbon products from the thi-ol. For example, ethanethiol bonds tothe catalyst, breaking the sulfur-hydro-gen bond and becoming ethyl thiolate(CH3(CH2)S). The carbon-sulfur bond isbroken next. Finally, the formation of acarbon-hydrogen bond yields ethane,whereas the breaking of such a bondleads to ethene.

The intermediates that form as ethylthiolate becomes ethane or ethene aretoo short-lived to be detected by meansof surface spectroscopy. Therefore, indi-rect probes must be used. These probesindicate that the breaking of the car-bon-sulfur bond is the rate-limiting stepin the desulfurization of ethyl thiolate.More important, we have found evi-dence that for any thiol molecule, thestrength of the carbon-sulfur bond de-termines the overall rate of the desul-furization reaction.

By using our general model and in-formation about the strength of thecarbon-sulfur bond, we have been ableto predict the rates of reaction and the types of products formed duringthe desulfurization of thiols on mo-lybdenum surfaces. Surface-bound car-bon and gaseous hydrogenÑthe unde-sirable productsÑare formed at a ratethat depends largely on whether a car-bon-hydrogen bond can be broken be-fore the breaking of the carbon-sulfurbond. According to fundamental prin-ciples of chemistry, therefore, the frac-tion of thiol intermediates that lead to useful hydrocarbons is proportion-al to the rate of carbon-sulfur bond

breaking relative to the rate of carbon-hydrogen bond breaking. Hence, we pre-dicted that thiols with low carbon-sul-fur bond strengthsÑthat is, high ratesof bond breakingÑwould yield a largefraction of hydrocarbons. This conclu-sion has been borne out by all our experiments.

The desulfurization of fossil fuels, theremoval of nitric oxide from car exhaustand the synthesis of ammonia are justthree of the many areas that have ben-eÞted from surface chemistry. Yet re-searchers need to continue to maketechnological and conceptual advancesif we are truly to understand how sur-faces inßuence the bonding and struc-ture of complex intermediates. The nextdecade promises major advances in ex-perimental and theoretical methodsthat should give a tremendous boost toour knowledge of the complicated cat-alytic reactions that occur on surfaces.


FURTHER READING

p-BONDED N2 ON FE(111): THE PRECUR-SOR FOR DISSOCIATION. M. Grunze, M.Golze, W. Hirschwald, H.-J. Freund, H.Pulm, U. Seip, M. C. Tsai, G. Ertl, J. K�p-pers in Physical Review Letters, Vol. 53,No. 8, pages 850Ð853; August 20, 1984.

MOLECULAR TRANSFORMATIONS ON SIN-GLE CRYSTAL METAL SURFACES. R. J. Ma-dix in Science, Vol. 233, pages 1159Ð1166; September 12, 1986.

MECHANISM OF THE NITRIC OXIDEÐCAR-BON MONOXIDEÐOXYGEN REACTION OVERA SINGLE CRYSTAL RHODIUM CATALYST.Galen B. Fisher et al. in Proceedings ofthe 9th International Congress on Catal-ysis, Vol. 3: Characterization and MetalCatalysts. Edited by M. J. Phillips and M.Ternan. Chemical Institute of Canada,1988.

SURFACE CHEMISTRY. John T. Yates, Jr.,in Chemical & Engineering News, Vol. 70,No. 13, pages 22Ð35; March 30, 1992.

ecule (1 ) based on sulfur, carbon and hydrogen interacts withmolybdenum, it loses a hydrogen and binds to the surface

(2 ). The bond between carbon and sulfur then weakens (3 ),and the remaining atoms may recombine to form ethane (4 ).

2 3 4


catalysis on surfaces

Documents