Pure & AppL Chem., VoL 46, pp. 71—90. Pergamon Press, 1976. Printed in Great Britain.

INTERNATIONAL UNION OF PUREAND APPLIED CHEMISTRY

PHYSICAL CHEMISTRY DIVISION

COMMISSION ON COLLOID AND SURFACE CHEMISTRY

MANUAL OF SYMBOLS ANDTERMINOLOGY FOR

PHYSICOCHEMICAL QUANTITIESAND UNITS—APPENDIX IIDefinitions, Terminology and Symbols in Colloid and Surface Chemistry

PART II: HETEROGENEOUS CATALYSIS

AdoptedbythelUPAC Council atMadrid, Spain, on9 September1975

Prepared for publication by

ROBERT L. BURWELL, JR.

PERGAMON PRESSOXFORD NEW YORK PARIS . FRANKFURT

PHYSICAL CHEMISTRY DIVISION

COMMISSION ON COLLOID AND SURFACE CHEMISTRY

DEFINITIONS, TERMINOLOGY AND SYMBOLS INCOLLOID AND SURFACE CHEMISTRY

PART II: HETEROGENEOUS CATALYSIS

(RULES APPROVED 1975)

PREFACE

This Part II of Appendix lit to the Manual of Symbolsand Terminology for Physicochemical Quantities andUnits t (hereinafter referred to as the Manual) has beenprepared by the Commission on Colloid and SurfaceChemistry of the Division of Physical Chemistry of theInternational Union of Pure and Applied Chemistry. It isthe outcome of extensive discussions within theCommission and its Task Force headed by ProfessorBurwell, with other IUPAC Commissions, and withpersons outside IUPAC during the period 1970-1975.Among the latter, special mention must be made toProfessors M. Boudart (USA), J. B. Butt (USA), and F. S.Stone (UK). A tentative version of these proposals wasissued as Appendix 39 (August 1974) on TentativeNomenclature, Symbols, Units and Standards to IUPACInformation Bulletin. The text has been revised in the lightof the criticisms, comments, and suggestions which werereceived, and the present version was prepared by theCommission and formally adopted by the IUPAC Councilat its meeting in Madrid, Spain, in September 1975.

It was felt that the use of unambiguous terminologywould promote communication and avoid misunderstand-ings among workers in heterogeneous catalysis and that alist of preferred symbols would be useful in many

tPart I of Appendix II, Definitions, Terminology and Symbols inColloid and Surface Chemistry, prepared for publication by D. H.Everett, Pure Appl. Chem., 31, 579—638 (1972).

lManual of Symbols and Terminology for PhysicochemicalQuantities and Units (1973 Edn.), prepared for publication by M.L. McGlashan and M. A. Paul, Butterworths, London (1975).

§The membership of the Commission during this period was asfollows:Chairman: —1973 D. H. Everett(UK); 1973—K.J.Mysels(USA)Secretary: H. van Olphen (USA)Titular Members: S. Brunauer (USA); R. L. Burwell, Jr. (USA);

R. Haul (Germany); V. B. Kazansky (USSR);1971— C. Kemball (UK); —1973 K. J. Mysels(USA); —1971 M. Pretre (France); G. Schay(Hungary).

Associate Members: R. M. Barrer (UK); —1973 G. K. Boreskov(USSR); A. V. Kiselev (USSR); —1973 H.Lange (Germany); 1973— J. Lyklema(Netherlands); A. Scheludko (Bulgaria); G.A. Schuit (Netherlands); 1971— K. Tamaru

(Japan).Observer: —1971 Sir Eric Rideal (UK).National Representatives: 1972— K. Morikawa (Japan); 1971—

1974 Sir Eric Rideal (UK) (deceased);1975 W. Schirmer (DDR).

respects. Heterogeneous catalysis is primarily a branch ofphysical chemistry but it has substantial overlap withorganic and inorganic chemistry and with chemicalengineering. The Commission agreed that no term orsymbol should be used in heterogeneous catalysis in asense different from that in physical chemistry in generalor, as far as possible, in a sense different from that inother branches of chemistry.

The present proposals are based on the same principlesas those used in the Manual and in Part I of thisAppendix and are consistent with them. The mostpertinent definitions of Part I are summarized and quotedin sections 1.2.1 and 1.2.2.

Historical and common usage of terms has beenretained as far as is compatible with the above principles.

Since the present proposals should be considered asone of the sub-sets of the set of terms and symbols ofphysical chemistry, the general principles are not repeatedhere. Attention must be called, however to one point,namely the restriction of the term "specific" to themeaning, divided by mass. This necessitates either therepetitive use of "per. unit area" or the introduction of anew term having this meaning. After careful considerationthe Commission recommends that the term areal, meaningdivided by area, be used. This is, however, at this time, aprovisional recommendation subject to a decision on thisand related terms by ICSU, the International Council ofScientific Unions.

CONTENTS

KAROL J. MYSELSChairmanCommission on Colloidand Surface Chemistry

Section 1. Definitions and Terminology1.1 Catalysis and Catalysts1.2 Adsorption

1.2.1 General terms1.2.2 Chemisorption and physisorption1.2.3 Types of chemisorption1.2.4 Sites for chemisorption1.2.5 Uniformity of sites1.2.6 Active site, active centre1.2.7 Adsorption isotherms1.2.8 Bifunctional catalysis1.2.9 Rates of adsorption and desorption

1.3 Composition, Structure and Texture of Catalysts

73

La Jolla, California29 December 1975

74 COMMISSION ON C0LL0ID AND SURFACE CHEMISTRY

1.3.1 General terms1.3.2 Porosity and texture

1.4 Catalytic Reactors1.5 Kinetics of Heterogeneous Catalytic Reactions

1.5.1 General terms1.5.2 Selectivity1.5.3 Rate equations1.5.4 Kinetic aspects of mechanism1.5.5 Non-uniformity of catalytic sites

1.6 Transport Phenomena in Heterogeneous Catalysis1.7 Loss of Catalytic Activity

1.7.1 Poisoning1.7.2 Deactivation: general1.7.3 Types of deactivation

1.8 Mechanism1.8.1 General1.8.2 Elementary processes in heterogeneous catalysis1.8.3 Nomenclature of surface intermediates

1.9 Nomenclature of Catalytic ReactionsSection 2. List of Symbols and AbbreviationsSection 3. Alphabetical Index

SECTION 1. DEFINITIONS AND TERMINOLOGY

1.1 Catalysis and catalystsCatalysis is the phenomenon in which a relatively small

amount of a foreign material, called a catalyst, augmentsthe rate of a chemical reaction without itself beingconsumed. Cases occur with certain reactants in whichthe addition of a substance reduces the rate of a particularreaction, for example, the addition of an inhibitor in achain reaction or a poison in a catalytic reaction. The term"negative catalysis" has been used for these phenomenabut this usage is not recommended; terms such asinhibition or poisoning are preferred.

A catalyst provides for sets of elementary processes(often called elementary steps) which link reactants andproducts and which do not occur in the absence of thecatalyst. For example, suppose the reaction

A=C

to proceed at some rate which might be measurable butmight be essentially zero. The addition of X might nowprovide a new pathway involving the intermediate B,

A + X—BB—C+X.

If reaction by this pathway proceeds at a rate significantwith respect to the uncatalysed rate such that the totalrate is increased, X is a catalyst. In this sense, a catalyticreaction is a closed sequence of elementary steps similarto the propagation steps of a gas-phase chain reaction.

The catalyst enters into reaction but is regenerated atthe end of each reaction cycle. Thus, one unit of catalystresults in the conversion of many units of reactants (butsee §1.7).

A catalyst, of course, may catalyse only one or some ofseveral thermodynamically possible reactions.

It is difficult to separate Nature into water-tight

tThe use of substrate for adsorbent or support is to bediscouraged because of its general use in enzyme chemistry todesignate a reactant.

Appendix II, Part I, recommends: The use of a solidus toseparate the names of bulk phases is preferred to the use of ahyphen which can lead to ambiguities.

compartments and probably no operational definition ofcatalysis can be entirely satisfactory. Thus, water mightfacilitate the reaction between two solids by dissolvingthem. This phenomenon might appear to constitute anexample of catalysis but such solvent effects are not, ingeneral, considered to fall within the scope of catalysis.The kinetic salt effect in solution is also usually excluded.Further, a catalyst must be material and, although an inputof heat into a system usually augments the rate of areaction, heat is not called a catalyst, nor is light a catalystin leading to reaction between chlorine and hydrogen.

A catalyst should be distinguished from an initiator. Aninitiator starts a chain reaction, for example, di-t-butylperoxide in the polymerization of styrene, but theinitiator is consumed in the reaction. It is not a catalyst.

In homogeneous catalysis, all reactants and the catalystare molecularly dispersed in one phase.

In heterogeneous catalysis, the catalyst constitutes aseparate phase. In the usual case, the catalyst is acrystalline or amorphous solid, the reactants and productsbeing in one or more fluid phases. The catalytic reactionoccurs at the surface of the solid and, ideally, its rate isproportional to the area of the catalyst. However, inpractical cases, transport processes may restrict the rate(see §1.6).

Most examples of catalysis can be readily characterizedas homogeneous or heterogeneous but there are examplesof catalysis which overlap the two types. Consider asystem in which intermediates are formed at the surfaceand then are desorbed into the gas phase and react there.Such intermediates might generate a chain reaction in thegas phase, i.e. chain initiation and chain termination occurat the surface but chain propagation occurs in the gasphase.

Enzyme catalysis may share some of the characteristicsof homogeneous and heterogeneous catalysis, as when thecatalyst is a macromolecule small enough to be molecu-larly dispersed in one phase with all reactants but largeenough so that one may speak of active sites on itssurface.

This manual deals with heterogeneous catalysis. Othertypes of catalysis will receive no further attention.

1.2 Adsorption1.2.1 General terms

Although adsorption exists as a subject of scientificinvestigation independent of its role in heterogeneouscatalysis, it requires particular attention here because ofits central role in heterogeneous catalysis. Most or allcatalytic reactions involve the adsorption of at least oneof the reactants. Many terms related to adsorption havealready been defined in Appendix II, Part I, § 1.1. Theseinclude surface, interface, area of surface or interface, andspecific surface area. Appendix II, Part I, recommends Aor S and a or s as symbols for area and specific area,respectively. A, and a, may be used to avoid confusionwith Helmholtz energy A or entropy S where necessary.

Other terms are sorption, sorptive, sorbate [a distinc-tion being made between a species in its sorbed state(sorbate) and a substance in the fluid phase which iscapable of being sorbed (sorptive)], absorption, absorp -tive, absorbate, absorbent; and adsorption, adsorptive,adsorbate, adsorbent.t The term adsorption complex isused to denote the entity constituted by the adsorbate andthe part of the adsorbent to which it is bound.

Appendix II, Part I, § 1.1.5, treats the adsorbent/fluidsinterface as follows.

Partli: Heterogeneouscatalysis 75

"It is often useful to consider the adsorbent/fluidinterface as comprising two regions. The region of thefluid phase (i.e. liquid or gas) forming part of theadsorbent/fluid interface may be called the adsorptionspace, while the portion of the adsorbent included in theinterface is called the surface layer of the adsorbent."

When used to denote the process in which moleculestor dissociated molecules accumulate in the adsorptionspace or in the surface layer of the absorbent, adsorptionhas as its counterpart the term desorption which denotesthe converse process (see Appendix II, Part I, § 1.1.4).Adsorption is also used to denote the result of the processof adsorption, i.e. the presence of adsorbate on anadsorbent. The adsorbed state may or may not be inequilibrium with the adsorptive (see § 1.2.2(c)).

Adsorption and desorption may also be used to indicatethe direction from which equilibrium has been ap-proached, e.g. adsorption curve (point), desorption curve(point).

1.2.2 Chemisorption and physisorptionFor convenience, the relevant portions of §1.1.6 and

1.1.7 of Appendix II, Part I, are reproduced here.Chemisorption and physisorption

Chemisorption (or Chemical Adsorption) is adsorptionin which the forces involved are valence forces of thesame kind as those operating in the formation of chemicalcompounds. The problem of distinguishing betweenchemisorption and physisorption (see below) is basicallythe same as that of distinguishing between chemical andphysical interaction in general. No absolutely sharpdistinction can be made and intermediate cases exist, forexample, adsorption involving strong hydrogen bonds orweak charge-transfer.

Some features which are useful in recognisingchemisorption include:(a) the phenomenon is characterised by chemical specifi-

city;(b) changes in the electronic state may be detectable by

suitable physical means (e.g. u.v., infrared or mic-rowave spectroscopy, electrical conductivity, magne-tic susceptibility);

(c) the chemical nature of the adsorptive(s) may bealtered by surface dissociation or reaction in such away that on desorption the original species cannot berecovered; in this sense chemisorption may not bereversible;

(d) the energy of chemisorption is of the same order ofmagnitude as the energy change in a chemical reactionbetween a solid and a fluid: thus chemisorption, likechemical reactions in general, may be exothermic orendothermic and the magnitudes of the energychanges may range from very small to very large;

(e) the elementary step in chemisorption often involvesan activation energy;

(f) where the activation energy for adsorption is large(activated adsorption), true equilibrium may beachieved slowly or in practice not at all. For example,in the adsorption of gases by solids the observedextent of adsorption, at a constant gas pressure after afixed time, may in certain ranges of temperatureincrease with rise in temperature. In addition, wherethe activation energy for desorption is large, removalof the chemisorbed species from the surface may be

PAC Vol. 46, No. 1—F

possible only under extreme conditions of tempera-ture or high vacuum, or by some suitable chemicaltreatment of the surface;

(g) since the adsorbed molecules are linked to the surfaceby valence bonds, they will usually occupy certainadsorption sites on the surface and only one layer ofchemisorbed molecules is formed (monolayer adsorp-tion).

Physisorption (or PhysicalAdsorption) is adsorption inwhich the forces involved are intermolecular forces (vander Waals forces) of the same kind as those responsiblefor the imperfection of real gases and the condensation ofvapours, and which do not involve a significant change inthe electronic orbitalpatterns of the species involved. Theterm van der Waals adsorption is synonymous withphysical adsorption, but its use is not recommended.

Some features which are useful in recognisingphysisorption include:(a') the phenomenon is a general one and occurs in any

solid/fluid system, although certain specific molecularinteractions may occur, arising from particulargeometrical or electronic properties of the adsorbentand/or adsorptive;

(b') evidence for the perturbation of the electronic statesof adsorbent and adsorbate is minimal;

(c') the adsorbed species are chemically identical withthose in the fluid phase, so that the chemical nature ofthe fluid is not altered by adsorption and subsequentdesorption;

(d') the energy of interaction between the molecules ofadsorbate and the adsorbent is of the same order ofmagnitude as, but is usually greater than, the energyof condensation of the adsorptive;

(e') the elementary step in physical adsorption does notinvolve an activation energy. Slow, temperaturedependent, equilibration may however result fromrate-determining transport processes;

(f') in physical adsorption, equilibrium is establishedbetween the adsorbate and the fluid phase. In solid/gassystems at not too high pressures the extent of physicaladsorption increases with increase in gas pressure andusually decreases with increasing temperature. In thecase of systems showing hysteresis the equilibriummay be metastable.

(g') under appropriate conditions of pressure and temper-ature, molecules from the gas phase can be adsorbedin excess of those in direct contact with the surface(multilayer adsorption or filling of micropores).

Monolayer and multilayer adsorption, micropore fillingand capillary condensation

In monolayer adsorption all the adsorbed molecules arein contact with the surface layer of the adsorbent.

In multilayer adsorption the adsorption space accom-modates more than one layer of molecules and not alladsorbed molecules are in contact with the surface layerof the adsorbent.

The monolayer capacity is defined, for chemisorption,as the amount of adsorbate which is needed to occupy alladsorption sites as determined by the structure of theadsorbent and by the chemical nature of the adsorptive;and, for physisorption, as the amount needed to cover thesurface with a complete monolayer of molecules inclose-packed array, the kind of close-packing having to bestated explicitly when necessary. Quantities relating tomonolayer capacity may be denoted by subscript m.

The surface coverage (0) for both monolayer andtThe term molecules is used in the general sense to denote any

molecular species: atom, ion, neutral molecule or radical.

76 COMMISSION ON COLLOID AND SURFACE CHEMISTRY

multilayer adsorption is defined as the ratio of the amountof adsorbed substance to the monolayer capaclty

The area occupied by a molecule in a completemonolayer is denoted by am; for example, for nitrogenmolecules am(N2).

Micropore filling is the process in which molecules areadsorbed in the adsorption space within micropores.

The micropore volume is conventionally measured bythe volume of the adsorbed material which completelyfills the micropores, expressed in terms of bulk liquid atatmospheric pressure and at the temperature of measure-ment.

In certain cases (e.g. porous crystals) the microporevolume can be determined from structural data.

Capillary condensation is said to occur when, inporous solids, multilayer adsorption from a vapourproceeds to the point at which pore spaces are filled withliquid separated from the gas phase by menisci.

The concept of capillary condensation loses its sensewhen the dimensions of the pores are so small that theterm meniscus ceases to have a physical significance.Capillary condensation is often accompanied by hys-teresis."

1.2.3 Types of chemisorptionNon-dissociative, dissociative. If a molecules is ad-

sorbed without fragmentation, the adsorption process isnon -dissociative. Adsorption of carbon monoxide isfrequently of this type. If a molecule is adsorbed withdissociation into two or more fragments both or all ofwhich are bound to the surface of the adsorbent, theprocess is dissociative. Chemisorption of hydrogen iscommonly of this type.

H2(g)—÷2H(ads) or

The asterisk represents a surface site.Homolytic and heterolytic relate in the usual sense to

the formal nature of the cleavage of a single bond. If theelectron pair in the bond of the adsorptive A: B is dividedin the course of its dissociative adsorption, the adsorptionis homolytic dissociative adsorption. If A or B retains theelectron pair, the adsorption is heterolytic dissociativeadsorption. Examples follow.

(a) Homolytic dissociative adsorption of hydrogen onthe surface of a metal

(b) Heterolytió dissociative adsorption of hydrogen atthe surface of an oxide where the surface sites and02 are surface sites in which the ions are of lowercoordmation than the ions m the bulk phase

H2 + +O2—*WM"+H0

Where clarity requires it, the equation may be written

H2(g)+ M + + 0 2—*WM ++ HO-

where the subscript s indicates that the species indicatedare part of the surface.

The notation WM is used, as in conventionalinorganic terminology, to indicate that the oxidationnumber of M has not changed

H2 +2*—2H*.

(c) Heterolytic dissociative adsorption of water at thesame pair of sites as in (b):

H20+ + 02—H0M + H0

Reductive and oxidative dissociative adsorption involveusage analogous to that in coordination chemistry in whichone speaks of the following reaction as an oxidativeaddition

L4M(I) + H2—L4M(III)H2.

Here, M represents a transition metal atom and L a ligand.H as a ligand is given an oxidation number of —1. Ifreductive, the electron pair which constitutes the bond inthe sorptive, A : B, is transferred to surface species; ifoxidative, a pair of electrons is removed from surfacespecies. One would say that dissociative adsorption of Cl2on a metal is oxidative if chlorine forms CV ions on thesurface of the adsorbent. A dissociative adsorption wouldbe reductive if, for example, it occurred thus (note thatH2— 2W + 2e here),

H2(g) + 2[M(Ill)02],—+2[M(ll)(OH)i.

Charge transfer adsorption represents oxidative orreductive chemisorption where reductive and oxidativerefer to electron gain or loss on species in the solid. Insimple cases it is non-dissociative, i.e. there is a meretransfer of charge between adsorptive and adsorbent informing the adsorbate. Two examples follow.

Reductive X + *—,X*,

where X represents an aromatic molecule of lowionization potential sUch as anthracene or triphenylamineand * a site on silica-alumina.

Oxidative 02 + *O2*.

The term, charge transfer adsorption, has also beenapplied to adsorption which resembles the charge transfercomplexes of Mulliken.

Immobile, mobile. These terms are used to describe thefreedom of the molecules of adsorbate to move about thesurface. Adsorption is immobile when kT is smallcompared to E, the energy barrier separating adjacentsites. The adsorbate has little chance of migrating toneighbouring sites and such adsorption is necessarilylocalized. Mobility of the adsorbate will increase withtemperature and mobile adsorption may be eitherlocalized or non -localized. In localized mobile adsorption,the adsorbate spends most of the time on the adsorptionsites but can migrate or be desorbed and readsorbedelsewhere. In non-localized adsorption the mobility is sogreat that a small fraction of the adsorbed species are onthe adsorption sites and a large fraction at other positionson the surface.

In some cases of localized adsorption the adsorbate isordered into a two-dimensional lattice or net in aparticular range of surface coverage and temperature. Ifthe net of the ordered adsorbed phase is in registry withthe lattice of the adsorbent the structure is calledcoherent, if not it is called incoherent (see also § 1.2.4).

Each of the various processes of adsorption may havedesorptions of theY reverse forms, for example, dissocia-tive adsorption may have as its, reverse, associative

Partli: Heterogeneouscatalysis 77

desorption. However, the process of chemisorption maynot be reversible (1.2.2(c)). Desorption may lead tospecies other than that adsorbed, for example, ethanedissociatively adsorbed on clean nickel gives little or noethane upon desorption, 1-butene dissociatively adsorbedto methylallyl and H on zinc oxide gives mainly 2-butenesupon desorption, and some W03 may evaporate fromtungsten covered with adsorbed oxygen.

Photoadsorption, photodesorption. Irradiation by light(usually visible or ultraviolet) may affect adsorption. In asystem containing adsorptive and adsorbent exposure tolight may lead to increased adsorption (photoadsorption)or it may lead to desorption of an adsorbate (photo -desorption).

1.2.4 Sites for chemisorptionSites may be classified according to their chemical

nature in usual chemical terminology. The following termsare simple extensions of ordinary chemical usage: basicsites, acidic sites, Lewis acid sites, proton or BrØnstedacid sites, electron accepting sites and electron donatingsites (possible examples of the last two appear undercharge transfer adsorption).

It is often useful to consider that sites for chemisorp-tion result from surface coordinative unsaturation, i.e. thatatoms at the surface have a lower coordination numberthan those in bulk. Thus, for example a chromium ion at thesurface of chromium oxide has a coordination number lessthan that of a chromium ion in the bulk. The chromium ionwill tend to bind a suitable adsorptive so as to restore itscoordination number. An atom in the (100) surface of aface-centered cubic metal has a coordination number of 8vs 12 for an atom in bulk; this, too, represents surfacecoordinative unsaturation. However, of course, there aresites to which the concept of surface coordinativeunsaturation does not apply, for example, Brønsted acidsites.

One is rarely sure as to the exact identity and structureof sites in adsorption and heterogeneous catalysis.However, some symbolism is needed for theoreticaldiscussion of possible sites. On the one hand one maywish to use a description which is general and non-specific. For this * and (ads) are recommended as, forexample, H* and H(ads). Or, one may wish to use asymbolism which is as specific as possible. Generalchemical symbols may be useful in this case. A symbolismuseful for metals involves the specification of C3 and Bwhere C3 denotes a surface atom with j nearest neighboursand B denotes an ensemble of n surface atoms whichtogether constitute an adsorption site, for example, theadsorption site lying above the centre of three surfaceatoms constituting the corners of an equilateral triangle isa B3 site (for details see van Hardeveld and Hartog,Surface Sci. 15, 189 (1969)).

Cases of chemisorption are known in which at highcoverages the net (two-dimensional lattice) of theadsorbate is not in registry with the lattice of theadsorbent. In such situations, the concept of sites ofprecise location and fixed number may not be applicable.Similar difficulties about the definition of sites will occur ifsurface reconstruction takes place upon interaction ofadsorbate and adsorbent.

Because of various difficulties which often appear inknowing the identity of surface sites, it is frequentlyconvenient, particularly for metals, to define the surfacecoverage 0 as the ratio of the number of adsorbed atomsor groups to the number of surface atoms (c.f. §1.2.2).

1.2.5 Uniformity of sitesVariations in the nature of the sites for adsorption or

catalysis can occur even with pure metals where there isno question of differences in chemical compositionbetween one part of the surface and another. Thesevariations arise not only because of defects in the metalsurfaces but also because the nature of a site depends onthe structure of the surface. Uniform sites are more likelyto be encountered when adsorption or catalysis is studiedon an individual face of a single crystal, but evenindividual faces may present more than one kind of site.Non -uniform sites will normally occur with specimens ofmetal exposing more than one type of crystal face. Thereare two main kinds of non-uniformities. Intrinsic non -uniformity is a variation due solely to the nature of theadsorbent. Induced non -uniformity arises when thepresence of an adsorbate molecule on one site leads to avariation in the strength of adsorption at a neighbouringsite. Thus, a set of uniform sites on an individual crystalface may become non-uniform if the surface is partiallycovered with a chemisorbed species.

When the catalytic properties of metals are examined,the importance of the non-uniformity of sites depends onthe reaction under study. For some reactions, the activityof the metal catalyst depends only on the total number ofsites available and these are termed structure -insensitivereactions. For other reactions, classified as structure -sensitive reactions, activity may be much greater on sitesassociated with a particular crystal face or even withsome type of defect structure. The alternative names offacile or demanding have been used to describestructure—insensitive or structure—sensitive reactions re-spectively.

The terms of § 1.2.5 have been discussed with referenceto metallic surfaces but they can be applied to otheradsorbents and catalysts and, in particular, to thepair-sites involved in heterolytic dissociative adsorption.

1.2.6 Active site, active centreThe term active sites is often applied to those sites for

adsorption which are the effective sites for a particularheterogeneous catalytic reaction. The terms active siteand active centre are often used as synonyms, but activecentre may also be used to describe an ensemble of sites atwhich a catalytic reaction takes place.

1.2.7 Adsorption isothermsAn adsorption isotherm for a single gaseous adsorptive

on a solid is the function which relates at constanttemperature the amount of substance adsorbed atequilibrium to the pressure (or concentration) of theadsorptive in the gas phase. The surface excess amountrather than the amount adsorbed is the quantity accessibleto experimental measurement, but, at lower pressures, thedifference between the two quantities becomes negligible(see Appendix II, Part I, § 1.1.11).

Similarly, when two or more adsorptives adsorbcompetitively on a surface, the adsorption isotherm foradsorptive i at a given temperature is a function of theequilibrium partial pressures of all of the adsorptives. Inthe case of adsorption from a liquid solution, anadsorption isotherm for any preferentially adsorbedsolute may be similarly defined in terms of the equilibriumconcentration of the respective solution component, butthe isotherm usually depends on the nature of the solventand on the concentrations (mole fractions) of other solutecomponents if present. Individual solute isotherms cannot

78 COMMISSION ON C0LLOID AND SURFACE CHEMISTRY

be derived from surface excesses except on the basis ofan appropriate model of the adsorption layer; whenchemisorption occurs it is generally adequate to assumemonolayer adsorption. Amounts adsorbed are oftenexpressed in terms of coverages O. In chemisorption, O isthe fraction of sites for adsorption covered by species i.Types of adsorption isotherms of interest to heterogene-ous catalysis follow.

The linear adsorption isotherm. The simplest adsorp-tion isotherm is the analogue of Henry's law. For a singleadsorptive, it takes the form

0=Kp or 0=Kc,

where p and c are the pressure and concentration of theadsorptive, 0 is the coverage by adsorbate and K thelinear adsorption isotherm equilibrium constant, orHenry's law constant. Most adsorption isotherms reduceto Henry's law when p or c becomes small enoughprovided that simple adsorption occurs, i.e. adsorption isneither dissociative nor associative. That is, at low enoughcoverages Henry's law usually applies to the first of thefollowing equations but not the second and third.

A+*A*;A2 + 2*=±2A*;2A + *=±A2*.

The Langmuir adsorption isotherm,

0= Kp or0

l+Kp (10)=Kor the equivalents in terms of concentrations, is com-monly taken to result from simple (non-dissociative)adsorption from an ideal gas on a surface with a fixednumber of uniform sites which can hold one and only oneadsorbate species. K is called the Langmuir adsorptionequilibrium constant. Further, the enthalpy of theadsorbed form must be independent of whether or notadjacent sites are occupied and consequently the enthalpyof adsorption is independent of 0. The second form ofLangmuir's isotherm given above, emphasizes that theconstant K is the equilibrium constant for A + *=±A*.Since the constancy of enthalpy with coverage isanalogous to the constancy of enthalpy with pressure inan ideal gas, the adsorbed state in a system followingLangmuir's isotherm is sometimes called an idealadsorbed state.

If chemisorption is dissociative,

A2 + 2*2A*,

Langmuir's equation takes the form

1/2 020 = K1'2p or K =(1_0)2.1+ K"

For simple adsorption of two adsorptives A and Bcompeting for the same sites, Langmuir's isotherm takesthe form

0— KApAA l+KApA+KBpB'

where KA and KB are the equilibrium constants for theseparate adsorption of A and B respectively. This

equation can be generalized to cover adsorption of severaladsorptives and to allow for dissociative adsorption ofone or more adsorptives.

In the Freundlich adsorption isotherm, the amountadsorbed is proportional to a fractional power of thepressure of the adsorptive. For a particular system, thefractional power and the constant of proportionality arefunctions of temperature. In terms of coverage theisotherm assumes the form

0 = ap

where n is a number greater than unity and a a constant.In the region of validity of the isotherm the (differential)enthalpy of adsorption is a linear function of ln 0.

In the Temkin adsorption isotherm, the amountadsorbed is related to the logarithm of the pressure of theadsorptive

0 =A ln p + B,

where A and B are constants. In the region of validity ofthe isotherm the (differential) enthalpy of adsorption is alinear function of 0.

The Brunauer—Emmett—Teller (or BET) adsorptionisotherm applies only to the physisorption of vapours butit is important to heterogeneous catalysis because of itsuse for the determination of the surface areas of solids.The isotherm is given by the following equation,

n — c(p/p°) —

nm-(l-p /p°)(1 + (c - l)(p /p°))

-0,

where c is a constant which depends upon the tempera-ture, the adsorptive and the adsorbent, n is the amountadsorbed, nm is the monolayer capacity and p° is thesaturated vapour pressure of the pure, liquid adsorptive atthe temperature in question. According to this equation,which is based on a model of multilayer adsorption, 0exceeds unity when p/p° is sufficiently large.

1.2.8 Bifunctional catalysisSome heterogeneous catalytic reactions proceed by a

sequence of elementary processes certain of which occurat one set of sites while others occur at sites which are ofa completely different nature. For example, some of theprocesses in the reforming reactions of hydrocarbons onplatinum/alumina occur at the surface of platinum, othersat acidic sites on the alumina. Such catalytic reactions aresaid to represent bifunctional catalysis. The two types ofsites are ordinarily intermixed on the same primaryparticles (1.3.2) but similar reactions may result evenwhen the catalyst is a mixture of particles each containingbut one type of site. These ideas could, of course, beextended to create the concept of polyfunctionalcatalysis.

1.2.9 Rates of adsorption and desorptionSticking coefficient is the ratio of the rate of adsorption

to the rate at which the adsorptive strikes the totalsurface, i.e. covered and uncovered. It is usually afunction of surface coverage, of temperature and of thedetails of the surface structure of the adsorbent.

Sticking probability is often used with the samemeaning but in principle it is a microscopic quantityconcerned with the individual collision process. Thus thesticking coefficient can be considered as a mean sticking

Partli: Heterogeneouscatalysis 79

probability averaged over all angles and energies of theimpinging molecules and over the whole surface.

The mean residence time of adsorbed molecules is themean time during which the molecules remain on thesurface of the adsorbent, i.e. the mean time intervalbetween impact and desorption. While residing on thesurface the molecules may migrate between adsorptionsites before desorption. If the residence time of anadsorbed species refers to specified adsorption sites itwould be called the mean life time of the particularadsorption complex. When the rate of desorption is firstorder in coverage the residence time is independent ofsurface coverage and equal to the reciprocal of the rateconstant of the desorption process. In this case it can becharacterized unambiguously also by a half-life or by someother specified fractional-life of the desorption process. Ifthe desorption process is not first order, e.g. due to mutualinteractions of the adsorbed molecules and/or energeticheterogeneity of the surface, the residence time dependsupon surface coverage and the operational definition of"residence time" needs to be specified precisely.

Unactivated and activated adsorption. If the tempera-ture coefficient of the rate of adsorption is very small, theadsorption process is said to be unactivated (i.e. to have anegligible activation energy). In this case the stickingcoefficient at low coverages may be near unity particularlyfor smaller molecules. If the temperature coefficient of therate of adsorption is substantial, the adsorption process issaid to be activated (i.e. to have a significant activationenergy). In this case, the sticking coefficient is small. Ingeneral, the activation energy of activated adsorption is afunction of coverage and it usually increases withincreasing coverage.

A number of relations between rate of activatedadsorption and coverage have been proposed. Of these,one has been particularly frequently used, the Roginskii—Zeldovich equation sometimes called the Elovich equa -tion,

dO_ -bO—ae

where 0 is the coverage, and a and b are constantscharacteristic of the system.

1.3 Composition, structure and texture of catalysts1.3.1 General terms

Catalysts may be one-phase or multiphase. In the firstcase, they may be composed of one substance (forexample, alumina or platinum black) or they may be a onephase solution of two or more substances. In this case, thecomponents of the solution should be given and joined by ahyphen (for example, silica-alumina).

Support. In multiphase catalysts, the active catalyticmaterial is often present as the minor component dispersedupon a support sometimes called a carrier. The supportmay be catalytically inert but it may contribute to theoverall catalytic activity. Certain bifunctional catalysts(1.2.8) constitute an extreme example of this. In namingsuch a catalyst, the active component should be listed first,the support second and the two words or phrases shouldbe separated by a solidus, for example, platinum/silica orplatinum/silica-alumina. The solidus is sometimes replacedby the word "on", for example, platinum on alumina.

Promoter. In some cases, a relatively small quantity ofone or more substances, the promoter or promoters, when

tSee Appendix II, Part I, §1.1.5.

added to a catalyst improves the activity, the selectivity, orthe useful lifetime of the catalyst. In general, a promotermay either augment a desired reaction or suppress an un-desired one. There is no formal system of nomenclaturefor designating promoted catalysts. One may, however, forexample, employ the phrase "iron promoted with aluminaand potassium oxide".

A promoter which works by reducing the tendency forsintering and loss of area may be called a textural promoter(see §1.7.3).

Doping. In the case of semiconducting catalysts, a smallamount of foreign material dissolved in the originalcatalyst may modify the rate of a particular reaction. Thisphenomenon is sometimes called doping by analogy withthe effect of similar materials upon semiconductivity.1.3.2 Porosity and texture

Many but not all catalysts are porous materials in whichmost of the surface area is internal. It is sometimesconvenient to speak of the structure and texture of suchmaterials. The structure is defined by the distribution inspace of the atoms or ions in the material part of thecatalyst and, in particular, by the distribution at thesurface. The texture is defined by the detailed geometry ofthe void space in the particles of catalyst. Porosity is aconcept related to texture and refers to the pore space in amaterial. With zeolites, however, much of the porosity isdetermined by the crystal structure.

An exact description of the texture of a porous catalystwould require the specification of a very large number ofparameters. The following averaged properties are oftenused.

With respect to porous solids, the surface associatedwith pores may be called the internal surface. Because theaccessibility of pores may depend on the size of the fluidmolecules, the extent of the accessible internal surfacemay depend on the size of the molecules comprising thefluid, and may be different for the various components ofa fluid mixture (molecular sieve effect).

When a porous solid consists of discrete particles, it isconvenient to describe the outer boundary of the particlesas external surface.

It is expedient to classify pores according to their sizest(i) pores with widths exceeding about 0.05 j.m or 50 nm

(500 A) are called macropores;(ii) pores with widths not exceeding about 2.0 nm (20 A)

are called micropores;(iii) pores of intermediate size are called mesopores.

The terms intermediate or transitional pores, whichhave been used in the past, are not recommended.

In the case of micropores, the whole of their accessiblevolume may be regarded as adsorption space.

The above limits are to some extent arbitrary. In somecircumstances it may prove convenient to choosesomewhat different values.

Pore size distribution is the distribution of pore volumewith respect to pore size; alternatively, it may be definedby the related distribution of pore area with respect topore size. It is an important factor for the kineticbehaviour of a porous catalyst and thus an essentialproperty for its characterization (see § 1.6).

The computation of such a distribution involvesarbitrary assumptions and a pore-size distribution shouldalways be accompanied by an indication as to the methodused in its determination. The methods usually involveeither or both of the following (i) adsorption—desorptionisotherms of nitrogen or other adsorptives in conjunctionwith a particular model for conversion of the isotherm

80 CoMMIssIoN ON C0LL0ID AND SURFACE CHEMISTRY

into a pore-size distribution, (ii) data obtained by themercury porosimeter. The isotherm gives a pore-sizedistribution for mesopores. The mercury porosimetergives a distribution covering macropores and largermesopores. In both cases what is measured is, strictlyspeaking, not the exact volume of pores having a givenpore size, but the volume of pores accessible throughpores of a given size. The relationship between these twofunctions depends on the geometrical nature of the poresystem.

The specific pore volume is the total internal voidvolume per unit mass of adsorbent. Some of the porevolume may be completely enclosed, and thus inaccessi-ble to molecules participating in a catalytic reaction.

The total accessible pore volume may be measured bythe amount of adsorbate at the saturation pressure of theadsorptive, calculated as liquid volume, provided theadsorption on the external surface can be neglected or canbe evaluated. The accessible pore volume may bedifferent formolecules of different sizes. A method whichis not subject to the effect of the external surface is thedetermination of the dead space by means of anon-sorbable gas (normally helium) in conjunction withthe determination of the bulk volume of the adsorbent bymeans of a non-wetting liquid or by geometricalmeasurements.

Primary particles. Certain materials widely used ascatalysts, or supports consist of spheroids of about 10 nm(100 A) in diameter loosely cemented into granules orpellets. The texture of these resembles that of a cemented,loose gravel bed. The 10 nm (100 A) particles may becalled primary particles.

Percentage exposed in metallic catalysts. The accessi-bility of the atoms of metal in metallic catalysts,supported or unsupported, dopends upon the percentageof the total atoms of metal which are surface atoms. It isrecommended that the term percentage exposed beemployed for this quantity rather than the term dispersionwhich has been frequently employed.

Pretreatment and activation. Following the preparationof a catalyst or following its insertion into a catalyticreactor, a catalyst is often subjected to various treatmentsbefore the start of a catalytic run. The term pretreatmentmay, in general, be applied to this set of treatments. Insome cases the word activation is used. It implies that thematerial is converted into a catalyst or into a very muchmore effective one by the pretreatment. Outgassing is aform of pretreatment in which a catalyst is heated invacuo to remove adsorbed or dissolved gas. Calcinationis a term which means heating in air or oxygen and is mostlikely to be applied to a step in the preparation of acatalyst.

1.4 Catalytic reactorsThe vessel in which a catalytic reaction is carried out is

called a reactor. Many different arrangements can beadopted for introducing the reactants and removing theproducts.

In a batch reactor the reactants and the catalyst areplaced in the reactor which is then closed to transport ofmatter and the reaction is allowed to proceed for a giventime whereupon the mixture of unreacted materialtogether with the products is withdrawn. Provision formixing may be required.

In a flow reactor, the reactants pass through the reactorwhile the catalysis is in progress. Many variations arepossible.

The catalyst may be held in a packed bed and thereactants passed over the catalyst. A packed bed flowreactor is commonly called a fixed bed reactor and theterm plug-flow is also used to indicate that no attempt ismade to back-mix the reaction mixture as it passesthrough the catalyst bed. The main modes of operation ofa flow reactor are differential involving a small amount ofreaction so that the composition of the mixture isapproximately constant throughout the catalyst bed, orintegral involving a more substantial amount of reactionsuch that the composition of material in contact with thefinal section of the catalyst bed is different from thatentering the bed.

In a pulse reactor, a carrier gas, which may be inert orpossibly one of the reactants, flows over the catalyst andsmall amounts of the other reactant or reactants areinjected into the carrier gas at intervals. A pulse reactor isuseful for exploratory work but kinetic results apply to atransient rather than to the steady state conditions of thecatalyst.

Several alternative modes of operation may be used toavoid the complications of the changing concentrationsalong the catalyst bed associated with integral flowreactors and each of these has a special name. In a stirredflow reactor, effective mixing is achieved within thereactor often by placing the catalyst in a rapidly-rotatingbasket. If the mixing achieved in this way is efficient, thecomposition of the mixture in the reactor will be close tothat of the exit gases. The same result can be reached byrecirculation of the gas around a ioop containing a fixedbed of catalyst, provided that the rate of recirculation isconsiderably larger then the rate of flow in and out of theloop. Under these circumstances, a substantial conversionto products can be obtained even though conditions in thebed correspond more closely to those associated with adifferential rather than with an integral reactor. Anothermode of operation involves a fluidized bed in which theflow of gases is sufficient to cause the bed of finely dividedparticles of catalyst to behave like a fluid. In a fluidizedbed, the temperature is uniform throughout, althoughmixing of gas and solid is usually incomplete. It hasspecial applications in cases where the catalyst has to beregenerated, e.g. by oxidation, after a short period of use.Continuous transfer of catalyst between two vessels (oneused as reactor and the other for catalyst regeneration) ispossible with a fluidized system. The stirred flow and therecirculation reactors are characterized ideally by verysmall concentration and temperature gradients within thecatalyst region. The term, gradientless reactor, may beused to include both types.

All reactors, batch or flow, may be operated in threemain ways in regard to temperature. These are isothermal,adiabatic and temperature-programmed. For the last, in abatch reactor the variation of temperature with time maybe programmed, or in a fixed bed reactor the variation oftemperature along the length of the bed may becontrolled.

When reactors are operated isothermally the batchreactor is characterized by adsorbate concentrations andother aspects of the state of the surface which areconstant in space (i.e. uniform within the catalyst mass)but which change with time. In the integral flow reactorwith the catalyst at steady state activity, the surfaceconditions are constant with time but change along thebed. In the gradientless reactorat steady state, the surfaceconditions are constant in space and, if the catalyst is at asteady state, with time. In the pulse reactor, the catalyst is

often not in a condition of steady state, concentrationschange as the pulse moves through the bed, and there maybe chromatographic separation of reactants and products.

In general, if heterogeneous catalytic reactions are to beconducted isothermally, the reactor design must providefor heat flow to or from the particles of catalyst so as tokeep the thermal gradients small. Otherwise, tempera-tures within the catalyst bed will be non-uniform. Thedifferential reactor and the various forms of the gradient-less reactors are advantageous in this regard.

The types of reactors described above can, in principle,be extended to reactions in the liquid phase although thepulse reactor has been little used in such cases.

Reactions in which one reactant is gaseous, the other isin a liquid phase, and the catalyst is dispersed in the liquidphase, constitute a special but not unusual case, forexample, the hydrogenation of a liquid alkene catalysedby platinum. A batch reactor is most commonly employedfor laboratory scale studies of such reactions. Masstransport from the gaseous to the liquid phase may reducethe rate of such a catalytic reaction unless the contactbetween the gas and the liquid is excellent (see § 1.6).

1.5 Kinetics of heterogeneous catalytic reactions1.5.1 General terms

Consider a chemical reaction

0= BVBB,

where i/B is the stoichiometric coefficient (plus forproducts, minus for reactants) of any product or reactantB. The extent of reaction is defined (see § 11.1 of theManual)

d = VBdnB,

where nB is the amount of the substance B.If rate of reaction is to have an unambiguous meaning,

it should be defined as the rate of increase of the extent ofreaction

=d/dt = VBdnB/dt

whereas the quantity dnB/dt may be called the rate offormation (or consumption) of B.

To facilitate the comparison of the results of differentinvestigators, the rates of heterogeneous catalytic reac-tions should be suitably expressed and the conditionsunder which they have been measured should be specifiedin sufficient detail. If the rate of the uncatalyzed reactionis negligible, the rate of the catalyzed reaction maybegivenas

r=d/dt.If Q, the quantity of catalyst, is in mass,

r=rm td/dt

and rm is the specific rate of reaction which may be calledthe specific activity of the catalyst under the specifiedconditions. If Q is in volume,

r = r, =3d/dt.

The volume should be that of the catalyst granules

tTheterm area! meaningperunitareaistentative(seePreface).

81

excluding the intergranular space. If Q is in area,

r = ra kd/dt.where ra S the arealt rate ofreaction. If the total sUrfacearea of the catalyst is used, it should be preferably a BETnitrogen area. 'However, other types of specified areasmay be employed, for example, the exposed metal area ofa supported metallic catalyst. The exposed metal area isoften estimated by selective chemisorption of a suitablesorptive, e.g. 'hydrogen or carbon monoxide.

The turnover frequency, N, (commonly called theturnover number) defined, as in enzyme catalysis, asmolecules reacting per active site in unit time, can be auseful concept if employed with care. In view of theproblems in measuring the number of active sitesdiscussed in §1.2.4, it is important to specify exactly themeans used to express Q in terms of active sites. Arealistic measure of such sites may be the number ofsurface metal atoms on a supported catalyst but in othercases estimation on the basis of a BET surface area maybe the only readily available method. Of course, turnovernumbers (like rates) must be reported at specifiedconditions of temperature, initial concentration or initialpartial pressures, and extent of reaction.

In comparing various catalysts for a given reaction or incomparing various reactions on a given catalyst, it may beinconvenient or impracticable to compare rates at aspecified temperature since rates must be meUsured attemperatures at which they have convenient values.Therefore, it may be expedient to compare the. tempera-tures at which the rates have, a specified value.

In reactors in which the concentrations of reactants andproducts are uniform in space, the rate is the same on allparts of the catalyst surface at any specified time. Inintegral flow reactors, however, the rate on each elementof the catalyst bed varies along the bed.

1.5.2 SelectivityThe term selectivity S is used to describe the relative

rates of two or more competing reactions on a catalyst.Such competition includes cases of different reactantsundergoing simultaneous reactions or of a single reactanttaking part in two or more reactions. For the latter case, Smay be defined in two ways.' The first of these defines afractional selectivity SF for eachproduct by the equation

SF =iI i.

The second defines relative selectivities, SR, for each pairof products by:

SR = 1IJ.

In shape selectivity, which may be observed incatalysts with very small pores, the selectivity, is largelydetermined by the bulk or size of one or more reactants.On zeolites, for example, the rate of reaction of alkaneswith linear carbon chains may be much greater than thatof those with branched chains.

1.5.3 Rate equationsGaseous systems in which all concentrations are

uniform in space and in which the reaction is irreversiblewill be considered first. ' . , .

Partil: Heterogeneouscatalysis

82 COMMISSION ON COLLOID AND SURFACE CHEMISTRY

The rate besides being proportional to the quantity ofcatalyst, Q, is also in general a function of temperature Tand the concentrations c1 or partial pressures Pi ofreactants, products and other substances if present:

r = = f(T, c) or r = f(T, p1).

The statement of this equation is commonly called therate equation or the rate law. Frequently, in heterogene-ous catalysis, the function I is of the form

r = k[Jcja

where k is the rate constant which is a function oftemperature but not of concentrations and a (integral orfractional; positive, negative or zero) is the order of thereaction with respect to component i. This form of therate law is called a power rate law. Often, however, a rateexpression of different form is used. For example, for areaction A + B — products, the rate equation might be

kKAKBCACB

(1+KAcA+KBcB+ Kc)2

This equation can be interpreted in terms of Langmuiradsorption isotherms. It is assumed (see § 1.5.4) that bothreactants must be adsorbed in order to react and that KAand KB are the respective Langmuir adsorption equilib-rium constants. The denominator allows for competitionfor sites between reactants and other substances (di-luents, poisons and products) present in the system atconcentrations c with related adsorption equilibriumconstants K. A rate law of this type is appropriatelycalled a Langmuir rate law although it was made popularby Hinshelwood, Schwab, Hougen, Watson and others.Such rate laws are frequently used for systems in whichthe adsorptions may not obey the Langmuir adsorptionisotherm. Under these circumstances, the rate laws canstill provide a useful means of correlating experimentalresults but the values of the derived constants must beinterpreted with caution.

For a single elementary process,

k = A exp (—E/RT),

where A is the frequency factor and E the activationenergy. Even though heterogeneous catalytic reactionsrarely if ever proceed by a single elementary process, thesame relation often applies to the overall rate constant. Insuch a case, however, A is not a frequency factor butshould be called the pre -exponential factor and E shouldbe called the apparent activation energy.

Sometimes A and E exhibit compensation, i.e. theychange in the same direction with change in catalyst fora given reaction or with change in reaction for a givencatalyst. A special case of compensation called the 0-ruleoccurs when, at least approximately,

ElnA =const+—,

where T0 is the isokinetic temperature, the temperature atwhich all k's would be identical.

These considerations can be extended to reversible

processes. They also apply to single phase, liquid systems.For the case, rather common in heterogeneous catalysis,in which one reactant is in agas phase and the others andthe products are in a liquid phase, application of theprinciples given above is straightforward provided thatthere is mass transfer equilibrium between gas phase andliquid phase, i.e. the fugacity of the reactant in the gasphase is identical with its fugacity in the liquid phase. Insuch case, a power rate law for an irreversible reaction ofthe form

= kpgflci

may apply where the quantities have the same significanceas before except that the gaseous reactant g is omittedfrom the ci's and entered as a pressure term with order ag.

The determination of rate of reaction in a flow systemrequires knowledge both of the feed rate, v, of a givenreactant and of the fraction converted, x. The definition offeed rate as the amount of reactant fed per unit time to theinlet of the reactor is consistent with 1.5.1. The rate ofreaction is then given by

dt VB

where PB is the stoichiometric coefficient of the reactantof which the fraction x is converted. Alternatively, onemay proceed from rm, r, and ra rather than d/dt bydefining the space velocities, vm, v,, and va where the v 'srepresent the rate of feed of the given reactant fed perunit mass, volume or surface area of the catalyst. Therelation,

xrm = Vm

PB

gives the specific rate of reaction or, under specifiedconditions, the specific activity of the catalyst. Substitu-tion of Ua or v, gives the areal rate of reaction or the ratedivided by volume of the catalyst, respectively. Alterna-tively, space times, Tm, Ta, and r, the reciprocals of thespace velocities, may be used. "Contact time" and"residence time" are terms which may be misleading forflow systems in heterogeneous catalysis and should beavoided.

1.5.4 Kinetic aspects of mechanismOf general convenience in the treatment of mechanisms

are the notions of rate determining process or step and mostabundant surface intermediate. The rate determiningprocess is defined, as is usual in kinetics in general, as thatsingle elementary process in the catalytic sequence which isnot in equilibrium when the overall reaction is significantlydisplaced from equilibrium. If the surface of a catalyst hasone set of catalytic sites, a particular intermediate is said tobe the most abundant surface intermediate if the fractionalcoverageby thatintermediate is muchlarger than coveragesby the other intermediates. Of course, there is no guaranteethat either a rate determining process or a most abundantsurface intermediate will exist for any particular reactionunder a particular set of conditions.

The term reaction centre may be used to include bothvacant and occupied catalytic sites. The sum of thesurface concentrations of reaction centres on the surfaceof a catalyst is a constant L. Thus, if species m at a surface

Partli: Heterogeneous catalysis 83

concentration L is the most abundant surface inter-mediate, Lm + L —L, where L is the surface concentra-tion of vacant reaction centres.

Langmuir—Hinshelwood mechanism. This represents asomewhat anomalous use of the term mechanism tospecify relative magnitudes of rate constants. In aLangmuir—Hinshelwood mechanism, all adsorption-desorption steps are essentially at equilibrium and asurface step is rate determining. Such a surface step mayinvolve the unimolecular reaction of a single adsorbatemolecule or the reaction of two or more molecules onadjacent sites with each other. Where the adsorptionprocesses follow Langmuir adsorption isotherms, theoverall reaction will follow some kind of a Langmuir ratelaw (1.5.3). However, the term Langmuir—Hinshelwoodmechanism may cover situations in which Langmuiradsorption isotherms do not apply.

1.5.5 Non-uniformity of catalytic sitesA characteristic of a catalytic surface is that its sites

may differ in their thermodynamic and kinetic properties.In the kinetic description of catalytic reactions onnon-uniform surfaces, a parameter a is frequently used toconnect changes in the activation energy of activatedadsorption with the enthalpy of the adsorption

Eads—Ed,= a(q — q°),

where Ed, is the energy of activation and —q° is theenthalpy of adsorption on the uncovered surface. Eads andq apply to the surface with the same value of 0. Inpractice the equation may apply only over a restrictedrange of 0. Sometimes a is defined as in the equationabove but in terms of Gibbs energies of activation andadsorption respectively. The name transfer coefficient hasbeen used by electrochemists to represent a in anotherrelated situation.

1.6 Transport phenomena in heterogeneous catalysisThis section will not attempt to cover the more

technical aspects of chemical reactor engineering.A unique feature of heterogeneous catalytic reactions is

the ease with which chemical kinetic laws are disguised byvarious transport phenomena connected with the exis-tence of concentration and/or temperature gradients inthe hydrodynamic boundary layer surrounding the cata-lyst particles (external gradients) or in the porous textureof the catalyst particles themselves (internal gradients).Additional difficulties arise in batch reactors and in stirredflow reactors if agitation is inadequate to maintainuniform concentrations in the fluid phase. Agitation isparticularly critical where one of the reactants is a gas andthe catalyst and other reactants and products are incondensed phases for example, in the hydrogenation of aliquid alkene. Here the agitation must be adequate tomaintain the fugacity of the dissolved gaseous reactantequal to that in the gaseous phase.

When external gradients correspond to substantialdifferences in concentration or temperature between thebulk of the fluid and the external surface of the catalystparticle, the rate of reaction at the surface is significantlydifferent from that which would prevail if the concentra-tion or temperature at the surface were equal to that in thebulk of the fluid. The catalytic reaction is then said to beinfluenced by external mass or heat transfer respectively,and, when this influence is the dominant one, the ratecorresponds to a regime of external mass or heat transfer.

Similarly, when internal gradients correspond to differ-ences in concentration or temperature between theexternal surface of the catalyst particle and its centre, therate in the particle is substantially different from thatwhich would prevail if the concentration or temperaturewere the same throughout the particle. The catalyticreaction is then said to be influenced by internal mass orheat transfer, and, when this influence is the dominantone, the rate corresponds to a regime of internal mass orheat transfer.

Terms such as diffusion limited or diffusion controlledare undesirable because a rate may be larger in regimes ofheat or mass transfer than in the kinetic regime ofoperation, i.e. when gradients are negligible.

1.7 Loss ofcatalytic acitvity1.7.1 Poisoning and inhibition

Traces of impurities in the fluid to which the catalystis exposed can adsorb at the active sites and reduce oreliminate catalytic activity. This is called poisoning andthe effective impurity is called a poison. If adsorption ofpoison is strong and not readily reversed, the poisoning iscalled permanent. If the adsorption of the poison isweaker and reversible, removal of the poison from thefluid phase results in restoration of the original catalyticactivity. Such poisoning is called temporary. If adsorptionof the poison is still weaker and not greatly preferred toadsorption of reactant, the reduction in rate occasionedby the poison may be called competitive inhibition orinhibition. Here, of course, the poison may be present inmuch larger than trace amounts. There are, of course, nosharp boundaries in the sequence permanent poisoning,temporary poisoning, competitive inhibition.

In selective poisoning or selective inhibition, a poisonretards the rate of one catalysed reaction more than thatof another or it may retard only one of the reactions. Forexample, there are poisons which retard the hydrogena-tion of olefins much more than the hydrogenation ofacetylenes or dienes. Also, traces of sulphur compoundsappear selectively to inhibit hydrogenolysis of hydrocar-bons during catalytic reforming.

A product of a reaction may cause poisoning orinhibition. The phenomenon is called self-poisoning orautopoisoning.

1.7.2 Deactivation—generalThe conversion in a catalytic reaction performed under

constant conditions of reaction often decreases with timeof run or time on stream. This phenomenon is calledcatalyst deactivation or catalyst decay. If it is possible todetermine the kinetic form of the reaction and, thus, tomeasure the rate constant for the catalytic reaction k, it issometimes possible to express the rate of deactivation byan empirical equation such as

—dk/dt = Bk,

where t is the time on stream, n is some positive constant,and B remains constant during a run but depends uponthe temperature and other conditions of the reaction.Alternatively, the decline in k may be assumed to resultfrom elimination of active sites and L may be substitutedfor k in the preceding equation where L is considered tobe the effective concentration of surface centers. It is thencommon practice to define a time of deactivation (ordecay time) as the time on stream during which k falls to aspecified fraction of its original value, often 0.5. Times of

84 COMMISSION ON C0LL0ID AND SURFACE CHEMISTRY

deactivation may vary from minutes as in catalyticcracking to years as in hydrodesuiphurization.

Catalytic deactivation can sometimes be reversed andthe original catalytic activity restored by some specialoperation called regeneration. For example, coked crack-ing catalyst is regenerated by burning off the coke (see§1.7.3, 1.9).

If the catalytic reaction is a network of variousprocesses, deactivation can lead to a change in thedistribution of products. In such cases, the deactivationnot only reduces the overall rate but it changes theselectivity.

1.7.3 Types of deactivationCatalyst deactivation can result from deactivation of

catalytic sites by poisoning either by impurities or byproducts of the catalytic reaction (l.7.l). Many reactionsinvolving hydrocarbons and particularly those run athigher temperatures lead to the deposition on the catalystof high molecular weight compounds of carbon andhydrogen which deactivate the catalyst. This phenomenonis called coking or fouling. Catalysts so deactivated canoften be regenerated.

Catalyst deactivation may also result from changes inthe structure or in the texture of the catalyst. Changes ofthis kind are usually irreversible and the catalyst cannotbe regenerated. This type of deactivation is often calledcatalyst ageing.

Sintering and recrystallization. Catalysts often sufferduring use from a gradual increase in the average size ofthe crystallites or growth of the primary particles. This isusually called sintering. The occurrence of sintering leadsto a decrease in surface area and, therefore, to a decreasein the number of catalytic sites. In some cases, sinteringleads to a change in the catalytic properties of the sites,for example, for catalysts consisting of highly dispersedmetals on supports, catalytic properties may change otisintering due to a change in the relative exposure ofdifferent crystal planes of the metallic component of thecatalyst or for other reasons. Thus sintering leads to adecrease in rate and perhaps also to a change inselectivity. Similar phenomena can occur in oxidecatalysts as used in catalytic oxidation. The crystal sizeincreases, or the initial structure of the crystals changes.For example, a binary solid compound may decomposeinto its components or an amorphous mass may crystal-lize. These processes may be called recrystallization. Insome cases the terms sintering and recrystallization mayrefer to the same process. The removal of surface defectsmay accompany these processes.

In some cases, as for example in catalytic cracking onsilica—alumina, processes similar to those involved insintering and recrystallization can lead to a change in thetexture of the catalyst. Surface areas are diminished andthe pore-size distribution is changed.

1.8 Mechanism of catalytic reactions1.8.1 General

A chemical reaction proceeds by a set of elementaryprocesses (the Manual, § 11.3) which are in series andperhaps also in parallel. These processes start andterminate at species of minimum free energy (reactants,intermediates and products) and each elementary processpasses through a state of maximum free energy (thetransition state). To specify the mechanism, one mustspecify the elementary processes. This specifies theintermediates. One must also give the nature (energetics,

structure, charge distribution) of the transition state. Somuch is true for chemistry in general. The special featuresof mechanism in heterogeneous catalysis are those whichinvolve reactions between sorptives and active sites,reactions among adsorbates, and processes which regen-erate active sites to give a type of chain reaction.

In general, only partial approaches to the specificationof mechanism as given above have been possible.

Mechanism is sometimes used in different senses. Forexample, consider the two situations.

A+B=±C A+B—*Cvs

C =D

It may be said that the two situations have differentmechanisms or that they are two variants of the samemechanism.

1,8.2 Elementary processes in heterogeneous catalysisThere are many more types of elementary processes in

heterogeneous catalysis than in gas phase reactions. Inheterogeneous catalysis the elementary processes arebroadly classified as either adsorption—desorption orsurface reaction, i.e. elementary processes which involvereaction of adsorbed species. Free surface sites andmolecules from the fluid phase may or may not participatein surface reaction steps.

There is no generally accepted classification of elemen-tary processes in heterogeneous catalysis. However,names for a few types of elementary processes aregenerally accepted and terminology for a partial classifica-tion (see M. Boudart, Kinetics of Chemical Processes.Chap. 2 (1968)) has received some currency. Theparticular reactions used below to exemplify this ter-minology are ones which have been proposed in theliterature but some have not been securely established asoccurring in nature at any important rate.

Adsorption—desorption. This includes the process ofphysical adsorption as well as non-dissociativechemisorption.

* +NH3(g)==H3N*

Dissociative adsorption and its reverse, associativedesorption.

2* + CH4(g)==CH3*+H*

The methane might be supposed to react either from thegas phase or from a physisorbed state.

Dissociative surface reactiQn and its reverse, associa-tive surface reaction.

2* +C2H5*=±H* + *CH2CH2*

This involves 'dissociative adsorption' in an adsorbate.Sorptive insertion. This is analogous to the process of

ligand insertion in coordination chemistry.

H* + C2H4g)—*C2H5

This reaction might also be imagined to proceed byadsorption of C2IL followed by ligand migration (anassociative surface reaction).

Reactive adsorption and its reverse, reactive desorp -

Partil: Heterogeneouscatalysis 85

tion. This resembles dissociative adsorption but onefragment adds to an adsorbate rather than to a surfacesite.

CH2D

H2C=CH2 + D—D(g) H2C" D** * 4,

In abstraction and extraction processes, an adsorptiveor adsorbate species extracts an adsorbed atom or alattice atom respectively.

Abstraction process *H + H(g)—* + H2(g)

Extraction process O,2_ + CO(g)—2e + C02(g)

The following elementary process occurring either onone site or, as shown, on two sites is called a Rideal or aRideal-.Eley mechanism:

D-D(g) H—D(g)H D* * —p * *

D2 may also be considered to be in some kind of a weaklyadsorbed state. It will be noted that one D atom is neverbonded to the surface in any minimum Gibbs energyintermediate. It is recommended that the term Rideal orRideal—Eley mechanism be reserved for this particularelementary process. However, the term has been used foranalogous processes in which there is a reactant moleculeand a product molecule of nearly the same energy in thefluid phase or in some weakly adsorbed state and in whichone or more atoms are never bonded to the surface. Anexample is the following elementary process

H3C—CH=CH2(g)--- H2C=CH—CH2D(g)D H*

which has been called a switch process. The term mightwell be used generically for similar processes. The termRideal or Rideal—Eley mechanism has been furtherextended to include all elementary processes in which amolecule reacts from the fluid phase or from some weaklyadsorbed state. Even the sorptive insertion process andthe abstraction process illustrated above fall within thisextended definition.

1.8.3 Nomenclature of surface intermediatesSurface intermediates should be named in ways

compatible in so far as possible with chemical nomencla-ture in general.

Adsorbed species may be treated as surface compoundsanalogous to molecular compounds. For example, *H maybe called surface hydride, *=C=O may be called a linearsurface carbonyl and

0

may be called a bridged surface carbonyl. H2N* may becalled a surface amide and H3C*, a surface methyl or asurface o-alkyl.

The species *H may also be called an adsorbedhydrogen atom and *C0, adsorbed carbon monoxide.

Organic adsorbates pose a particular problem becausequite particular structures of some complexity are

regularly discussed. A nomenclature is recommended inwhich the surface is treated as a substituent whichreplaces one or more hydrogen atoms. The degree ofsubstitution is indicated by monoadsorbed, diadsorbed,etc. This terminology does not specify the nature of thechemical bonding to the surface nor does it restrict, apriori, the valency of the surface site *. Thus, both of thefollowing species

H2 H2

H2CH2 H2ç,CH.are named 1,3-di-adsorbed propane. Other examples

CH3

CH3——CH3 2-monoadsorbed 2-methyipropane

*

*OCH2CH3 0-monoadsorbedethanol*CH2CH2OH 2-monOadsorbed ethanolCH3—CH—CH2-—CH—CH2_-CH3 2,4-diadsorbed

* * hexane

H1/,.H eclipsed 1,2-diadsorbed ethane

*= CH—CH3 or (*)2CH—CH3 1,1-diadsorbedethane

* = NH or (*)2NH diadsorbed ammonia*COCH3 1-monoadsorbed acetaldehydeSpecies adsorbed as ir-complexes are described asIT -adsorbed:

H2CCH2ir-adsorbed ethylene

*H

HH2C=C=CH2 or H2C <CH2 IT-adsorbed allyl*

The substitution system of nomenclature should beviewed as showing only how atoms are connected and notas indicating the precise electronic structure. Thusir-adsorbed ethylene is one representation of 1,2-diadsorbed ethane.

Nomenclature based upon the process of formation of aparticular adsorbate is to be discouraged. Thus, H* maybe 'dissociatively adsorbed hydrogen' but the samespecies is formed in dissociative adsorption of CIL, NH3,H20.

1.9 Nomenclature of catalytic reactions

In general, a catalytic reaction may be named by addingthe adjective "catalytic" to the standard chemical termfor the reaction, for example, catalytic hydrogenation (or,if clarity demands, heterogeneous catalytic hydrogena -tion), catalytic hydrodesulphurization, catalytic oxidativedehydrogenation, catalytic stereospecific polymerization.

In general, special terminology for reactions is to bediscouraged. However, certain catalytic processes of

*CH3*CH2CH2CH3

are:

monoadsorbed methane1-monoadsorbed propane

* *

86 CoMMissioN ON COLLOID AND SURFACE CHEMISTRY

technological interest have special names in common use. Surface siteWhere such processes involve the simulteneous occurr- Ion (or atom M) ofence of two or more different chemical reactions, special adsorbent or catalyst 'names for the processes are probably inevitable. Some at the surface MS (or M,)important examples of such processes of technological Constant in Henry's law Kinterest are: Constant in Langmuir's

Catalytic cracking. In this process, a higher boiling cut adsorption isotherms Kof petroleum, for example, gas oil, is converted substan- Constant in Langmuir'stially into a lower boiling material of high octane number. adsorption isotherms forAmong the processes which appear to be involved are substance i Kskeletal isomerization of alkanes followed by their Constants in Freundlichcleavage into alkane and olefin, and hydrogen transfer isotherms a, nreactions which reduce the amount of olefin formed and Constants in Temkinwhich lead to coke and aromatic hydrocarbons. isotherms A, B

Catalytic hydrocracking. This is similar to catalytic Constant in BETcracking in its industrial purpose but it is effected under isotherms chydrogen pressure and on a catalyst containing an Monolayer capacity nmingredient with a hydrogenating function. Constants of Roginskii—

Catalytic reforming. Catalytic reforming is a process Zeldovich equation a, bfor increasing the octane number of naphthas. It involvesisomerization of alkanes, dehydrogenation of cyclohex- 2.3 Composition, structure andanes to aromatic hydrocarbons, isomerization and dehyd- texture of catalystsrogenation of alkylcyclopentanes, and dehydrocydizationof alkanes. 2.4 Catalytic reactors

The following reactions may be mentioned becausethey are rare except as heterogeneous catalytic reactions 2.5 Kinetics of heterogeneousand have somewhat specialized meanings in catalysis. catalytic reactors

Catalytic methanation. This is a process for removing Stoichiometric coefficientcarbon monoxide from gas streams or for producing of substance B VB

methane by the reaction Extent of reactionRate of catalysed

CO +3H2—CH4+ H20 reactionQuantity of catalyst Q

Catalytic dehydrocyclization. This is a reaction in Specific ratewhich an alkane is converted into an aromatic hydrocar- of reaction rmbon and hydrogen, for example, Specific activity

of the catalyst rm

heptane—*toluene +4H2 Rate of reactionper unit volume

Catalytic hydrogenolysis. This is ordinarily used for of catalyst r,reactions in which nC—Ca + H2 gives aCH + HCa, for Areal rate ofexample, reaction Ta

Turnover frequencypropane +H2—ethane +methane (turnover number) Ntoluene + H2—+benzene +methane Selectivity S, SF, SR

butane + H2—2ethane Rate constant kOrder of the reaction a,

However, it may also be used for cleavage of bonds other Frequency factor Athan aC—Ca, for example, Activation energy E

Isokinetic temperaturebenzyl acetate +H2—toluene +aceticacid (Kelvin scale) T0

benzylamine + H2—toluene + NH3 Fraction converted xFeed rate v

Catalytic hydrodesulphurization. This is a process in Space velocities Vm, Vv, Vawhich, in the presence of hydrogen, sulphur is removed as Space times Tm, T,, Tahydrogen sulphide. Sum of surface

concentrations ofSECTION 2. LIST OF SYMBOLS AND ABBREVIATIONS reaction centres L

2.1 Catalysis and catalysts Surface concentration ofsurface intermediate m Lm

2.2 Adsorption Surface concentration ofArea of surface A, A,, S vacant reaction centres LSpecffic surface area a, a,, s Energy of activationSurface coverage 0 for activated adsorptionArea per molecule in Energy of activatjon

complete monolayer for activated adsorptionof substance i am(i) on uncovered surface E°,

Partil: Heterogeneouscatalysis 87

(Differential) enthalpy BET adsorptionof adsorption —q isotherm 1.2.7

(Differential) enthalpy bifunctionalof adsorption on catalysis 1.2.8uncovered surface —q° Brønsted acid

Transfer coefficient a site 1.2.4Brunauer—Emmett—

2.6 Transport phenomena in Teller adsorptionheterogeneous catalysis isotherm 1.2.7

calcination 1.3.2

2.7 Loss of catalytic activity capillaryConstants in equation condensation 1.2.2

for rate of deactivation B, n carrier 1.3.1

Time of run (on stream) t catalysis 1.1

catalyst 1.1

2.8 Mechanism catalyst ageing 1.7.3

catalytic2.9 Nomenclature of catalytic reactions cracking 1.9

catalystSECTION 3. ALPHABETICAL INDEX deactivation

Symbol Term Section (decay) 1.7.2absorbate 1.2.1 catalyticabsorbent 1.2.1 dehydrocyclization 1.9absorption 1.2.1 catalyticabsorptive 1.2.1 hydrodesuiphurization 1.9abstraction catalytic

process 1.8.2 hydrocracking 1.9

accessible catalyticpore volume 1.3.2 hydrogenation 1.9

acid site 1.2.4 catalyticactivated hydrogenolysis 1.9

adsorption 1.2.2, 1.2.9, 1.5.5 catalyticactivation 1.3.2 methanation 1.9

E, Eads activation energy 1.5.3, 1.5.5 catalyticactive centre 1.2.6 oxidative

*, (ads) active site 1.2.2, 1.2.6 dehydrogenation 1.9adiabatic reactor 1.4 catalytic reaction 1.1

adsorbate 1.2.1 catalytic reactors 1.4adsorbed state 1.2.1 catalytic reforming 1.9adsorbent 1.2.1 catalyticadsorption 1.2.1 stereospecificadsorption— polymerization 1.9

desorption process 1.8.2 charge transferadsorption complex 1.2.1 adsorption 1.2.3, 1.2.4adsorption isotherm 1.2.7 chemical adsorption 1.2.2

*, (ads) adsorption site 1.2.2 chemisorption 1.2.2, 1.2.3, 1.2.4adsorption space 1.2.1 coherent structure 1.2.3

adsorptive 1.2.1 coking 1.7.3

ageing 1.7.3 compensation 1.5.3E apparent competitive

activation energy 1.5.3 inhibition 1.7.1

am area occupied by composition ofmolecule in catalyst 1.3

complete monolayer 1.2.2 contact time 1.5.3

A, A, s area of 0, 0 coverage 1.2.2, 1.2.4, 1.2.7interface 1.2.1 deactivation 1.7.2

A, A, s area of surface 1.2.1 decay time 1.7.2areal preface demanding

ra areal rate reaction 1.2.5of reaction 1.5.1 desorption 1.2.1, 1.2.3

associative diadsorbed 1.8.3

desorption 1.2.3, 1.8.2 differential flowassociative reactor 1.4

surface reaction 1.8.2 diffusion limitedautopoisoning 1.7.1 (controlled) 1.6basic site 1.2.4 dispersion 1.3.2batch reactor 1.4 dissociative

COMMISSION ON COLLOID AND SURFACE CHEMISTRY

intrinsic1.2.3, 1.8.2 non-uniformity 1.2.5

T9 isokinetic1.8.2 temperature 1.5.31.3.1 isothermal

reactor 1.41.8.3 kinetic aspects

of mechanism 1.5.41.2.4 kinetics of

heterogeneous1.2.4 catalytic

reactions 1.51.1, 1.8.2 kinetic regime 1.61.1 K, K Langmuir1.2.9 adsorption1.1 equilibrium1.5.1 constant1.6 Langmuir1.3.2 adsorption

isotherm 1.2.71.8.2 Langmuir rate1.2.5 law 1.5.31.5.3 Langmuir—1.4 Hinshelwood1.4 mechanism 1.5.4

Lewis acid site 1.2.41.4 linear adsorption1.7.3 isotherm 1.2.7

K linear adsorption1.5.3 isotherm1.5.3 equilibrium

constant 1.2.7localized adsorption 1.2.3

1.2.7 loss of catalyticactivity 1.7

1.4 macropores 1.3.2mean life time

1.2.7 of adsorptioncomplex 1.2.9

mean residence1.9 time 1.2.9

mechanism of1.1 catalytic

reactions 1.8

mesopores 1.3.21.2.3 micropore filling 1.2.2

micropore volume 1.2.21.1 micropores 1.3.2

mobile adsorption 1.2.3molecular sieve

1.2.3 effect 1.3.2monoadsorbed 1.8.3

1.2.7 monolayer1.2.3 adsorption 1.2.2

flmS, flm Vm monolayer1.2.3 capacity

most abundant1.2.5 surface1.7.1 intermediate 1.5.41.1 multilayer

adsorption 1.2.21.4 negative catalysis 1.11.2.1 net 1.2.3

nomenclature of1.3.2 catalytic1.6 reactions 1.91.3.2 nomenclature of

88

V

x

A

K

1.2.7, 1.5.3

adsorption(chemisorption)

dissociativesurface reaction

dopingeclipsed

diadsorbedelectron

accepting siteelectron

donating siteelementary

processelementary stepElovich equationenzyme catalysisextent of reactionexternal gradientexternal surfaceextraction

processfacile reactionfeed ratefixed bed reactorflow reactorfluidized bed

reactorfoulingfraction of

reactant convertedfrequency factorFreundlich

adsorptionisotherm

gradientless reactorHenry's law

constantheterogeneous

catalytichydrogenation

heterogeneouscatalysis

heterolyticdissociativeadsorption

homogeneouscatalysis

homolyticdissociativeadsorption

ideal adsorbedstate

immobile adsorptionincoherent

structureinduced

non-uniformityinhibitioninitiatorintegral flow

reactorinterfaceintermediate

poresinternal gradientinternal surface

1.2.2, 1.2.7

Partli: Heterogeneouscatalysis 89

surface reactor 1.4intermediates 1.8.3 recirculation

non-dissociative reactor 1.4chemisorption 1.2.3 recrystallization 1.7.3

non-localized reductiveadsorption 1.2.3 dissociative

non-uniform site 1.2.5, 1.5.5 adsorption 1.2.3a1 order of reaction 1.5.3 regeneration 1.7.2

outgassing 1.3.2 regime ofoxidative external mass

dissociative (or heat)adsorption 1.2.3 transfer 1.6

packed bed regime ofreactor 1.4 internal mass

percentage exposed 1.3.2 (or heat)permanent transfer 1.6

poisoning 1.7.1 residence time 1.2.9, 1.5.3photoadsorption 1.2.3 Ridealphotodesorption 1.2.3 (or Rideal—Eley)physical mechanism 1.8.2

adsorption 1.2.2 Roginskii—physisorption 1.2.2 Zeldovichir-adsorbed 1.8.3 equation 1.2.9plug-flow reactor 1.4 selectivepoison, poisoning 1.7.1 inhibition 1.7.1

polyfunctional selectivecatalysis 1.2.8 poisoning 1.7.1

V pore volume 1.3.2 SF, S selectivitypores 1.3.2 (as fraction) 1.5.2pore-size SR, S selectivity

distribution 1.3.2 (as ratio) 1.5.2porosity 1.3.2 self poisoning 1.7.1power rate shape

law 1.5.3 selectivity 1.5.2pre-exponential sintering 1.7.3

factor 1.5.3 *, (ads) site forpretreatment 1.3.2 chemisorption 1.2.4primary particles 1.3.2 sorbate 1.2.1

promoter 1.3.1 sorption 1.2.1proton acid sorptive 1.2.1

site 1.2.4 sorptive insertion 1.8.2pulse reactor 1.4 Tm, Ta, r. spacetime, per

Q quantity of unit mass,catalyst 1.5.1 area, volume

k rate constant 1.5.3 of catalyst 1.5.3rate Vm, Va, V space velocities

determining per unit mass,process (step) 1.5.4 area, volume

rate equation 1.5.3 of catalyst 1.5.3rate law 1.5.3 Tm specific activityrate of of catalyst 1.5.1

adsorption and specific poredesorption 1.2.9 volume 1.3.2

dnB/dt rate of Tm specific rate offormation catalysed reaction 1.5.1(consumption) a, s,a specific surfaceof B 1.5.1 area 1.2.1

rate of stickingreaction 1.5.1 coefficient 1.2.9

rate of reaction stickingper unit volume probability 1.2.9of catalyst 1.5.1 stirred flow

reaction centre 1.5.4 reactor 1.4reactive structure of

adsorption 1.8.2 adsorbentreactive (catalyst) 1.3, 1.3.2

desorption 1.8.2 structure

90 COMMISSION ON COLLOID AND SURFACE CHEMISTRY

insensitive temporaryreaction 1.2.5 poisoning 1.7.1

structure texture ofsensitive adsorbent (catalyst)1.3, 1.3.2reaction 1.2.5 0-rule 1.5.3

substrate 1.2.1 time ofsupport 1.2.1, 1.3.1 deactivation 1.7.2surface 1.2.1 time of run

L surface (on stream) 1.7.2concentration ofreaction centres 1.5.4 a transfer

surface coefficient 1.5.5coordinative transitionalunsaturation 1.2.4 pores 1.3.2

0, 0 surface coverage 1.2.2, 1.2.4, 1.2.7 transportsurface layer processes in

of the adsorbent 1.2.1 heterogeneoussurface step 1.5.4 catalysts 1.6switch process 1.8.2 N turnover frequency

(turnover number) 1.5.1Temkin unactivated

adsorption adsorption 1.2.9isotherm 1.2.7 uniform sites 1.2.5

temperature van der Waalsprogrammed adsorption 1.2.2reactor 1.4