1

Chapter1
CatalystFundamentalsin
IndustrialCatalysis

DraftChapterPreparedBy
RobertFarauto
LucasDorazio

2
ChapterOutline
1CatalystFundamentalsIndustrialCatalysis
1.1 Introduction

1.2 CatalyzedversusNoncatalyzedReactions
1.2.1 ExampleReaction:LiquidPhaseRedoxReaction
1.2.2 ExampleReaction:COOxidation
1.2.3 Homogeneousvs.HeterogeneousCatalysts

1.3 SelectivityandCatalyticReactions

1.4 PhysicalStructureofaHeterogeneousCatalyst
1.4.1 ActiveSpecies
1.4.2 Promoters
1.4.3 Carriers
1.4.4 StructureoftheCatalystandCatalyticReactor

1.5 AdsorptionandKineticModelsforHeterogeneousCatalysis
1.5.1 LangmuirIsotherm
1.5.2 ReactionKineticModels
1.5.2.1 LangmuirHinshelwoodKineticMechanism
1.5.2.2 MarsvanKrevelenKineticMechanism
1.5.2.3 EleyRidealKineticMechanism
1.5.3 EmpiricalPowerRateLaw

1.6 ChemicalandPhysicalStepsOccurringduringHeterogeneousCatalysis
1.6.1 MassTransferPhenomenaandSurfaceReaction
1.6.2 CatalystEffectivenessandConcentrationGradients
1.6.3 TheRateLimitingStep

3
1CatalystFundamentalsIndustrialCatalysis
1.1Introduction
Chemicalreactionsoccur bybreakingchemicalbondsofreactantsandformingnewbonds and
new compounds. Breaking stable bonds requires the absorption of energy while making new bonds
results in the liberation of energy. The combination of these energies results in either an exothermic
reactioninwhichtheconversionofreactantstoproductsliberatesenergyoranendothermicprocessin
which the conversion process requires energy. In the former case the energy of the product is lower
than that of the reactants with the difference being the heat liberated. In the latter case the product
energyisgreaterbytheamountthatmustbeaddedtoconservethetotalenergyofthesystem.Under
the same reaction conditions the heat of reaction (H) being a thermodynamic function does not
depend on the path or rate by which reactants are converted to products. Similarly the G of the
reactionisnotdependentonthereactionpathsinceittooisathermodynamicstatefunction.Thiswill
be emphasized once we discuss catalytic reactions. The rate of reaction is determined by the slowest
stepinaconversionprocessindependentoftheenergycontentofthereactantsorproducts.

1.2Catalyzedversesnoncatalyzedreactions
In the most basic sense, the purpose of the catalyst is to provide a reaction pathway or
mechanism that has a lower activation barrier compared to the noncatalyzed pathway, which is
illustrated in Figure 1.1. In any reaction, catalyzed or noncatalyzed, the reaction sequence occurs
through a series of elementary steps. In a noncatalyzed reaction, the species that participate in the
reactionsequencearederivedsolelyfromthereactants.Inacatalyzedreaction,thecatalystissimply
an additional species that becomes involved in the reaction sequence. More importantly, the
intermediate species involving catalyst species are easier to form (i.e. lower energy barrier) than if the

4
catalystspecieswerenot present.Finally,during thecatalyzed reactionsequence,thecatalystspecies
returnstoitsoriginalstate.Itistheregenerationofthecatalystspeciestoitsoriginalstatethatmakes
acatalystacatalyst,andnotareactant.Thus,acatalystisaspeciesthatbecomesinvolvedinthe
reactor sequence, reacts with the reactants to form an intermediate species, that than undergoes
further reaction to form the product and regenerate the catalyst species. This basic sequence of
eventsisillustratedinFigure1.2.

Figure1.1 Catalyzed and uncatalyzed reaction energy paths illustrating the lower
energy barrier (activation energy) associated with the catalytic reaction
comparedwiththenoncatalyticreaction.

Figure1.2 Illustrationofcatalyzedversusnoncatalyzedreactions

1.2.1ExampleReaction:LiquidPhaseRedoxReaction
Lets consider the simple solution redox reaction between Fe
+2
and Ce
+4
in aqueous solution.
The reaction below excludes the H
2
O present in the coordination sphere for each species since it does
notdirectlyparticipateinthereaction.
Fe
+2
+Ce
+4
Fe
+3
+Ce
+3
(1.1)
ThisreactioninvolvesadirectelectrontransferfromtheFe
+2
totheCe
+4
andbyitselfoccursveryslowly
becausetheelectrontransferprocessoccursslowly.However,inthepresenceofMn
+4
species,therate
dramatically increases because the electron transfer is now facilitated through the Mn
+4
/Mn
+2
couple.
The Mn
+4
species is a catalyst, not a reactant. While it does directly participate in the reaction, the
reactionpathwayresultsinnooverallchangeinthechemicalstateoftheMnion(Figure1.3).

Figure1.3 CatalyticFeCeredoxreactioncatalyzedbyMn

The reaction profile of both the catalyzed and noncatalyzed reactions can be described
kinetically by the Arrhenius profile in which reactants convert to products by surmounting the energy
barrier where the rate of reaction can be expressed by the Arrhenius Expression (Equation 1.2).
According to the Arrhenius expression, the rate of reaction is proportional to the exponential of
temperature (T) and inversely proportional to the exponential of the activation energy (E). The
remaining terms in Equation 1.2 include the gas constant (R), preexponential factor (k
o
), and the rate
constant(k).Thus,therateofreaction(i.e.rateconstant)willincreaseasthetemperatureincreasesor
theactivationenergydecreases.
k = k
o
cxp [
L
R1
(1.2)
Referringbacktoourredoxexample,thecatalyzedpathwayhastheloweractivationenergyand
therefore will have a higher reaction rate at a given temperature. The energy barrier was lowered by
theMncatalystprovidingachemicalshortcuttoproducts.Althoughtherateisgreaterforthecatalyzed
reaction,thereactantsandproductsarethesameasthenoncatalyzedreaction.Thus,thermodynamic
properties remain unchanged and both reaction pathways will have the same reaction enthalpy (H)
andreactionfreeenergy(G),anditsequilibriumconstant.Thecatalystcanonlyinfluencetherateof

7
whichreactantsareconvertedtoproductsinaccordancetotheequilibriumconstantandcannotmake
thermodynamically unfavorable reactions occur. In industrial practice reactions conditions, such as
temperature, pressure and reactant compositions are varied to bring the equilibrium to a desirable
valuetopermitthereactiontooccur.

1.2.2ExampleReaction:GasPhaseOxidationReaction
Nowwewillconsidertheconversionofcarbonmonoxide(CO),aknownhumanpoison,toCO
2
,
areactionofgreatimportancetothequalityofairwebreathedaily.Theoverallrateofthenoncatalytic
reactioniscontrolledbythedissociationoftheO
2
moleculetoOatoms(ratelimitingstep)whichrapidly
reacts with CO forming CO
2
. The temperature required to initiate the dissociation of O
2
is greater than
700
o
C and once provided the reaction rapidly goes to completion with a net liberation of energy (the
heat of reaction is exothermic). The requirement to bring about the O
2
dissociation and ultimately the
conversion of CO to CO
2
has an activation energy (E
nc
). Reaction occurs when a sufficient number of
molecules(O
2
)possesstheenergynecessary(asdeterminedbytheBoltzmanndistribution)tosurmount
the activation energy barrier (E
nc
) shown in Figure 1.4. The rate of reaction is expressed in accordance
withtheArrheniusequation(Equation1.2),wherekistherateconstant,k
o
isthepreexponentialfactor,
E is the activation energy for the reaction to occur, R is the gas constant, and T is the temperature at
which the reaction is occurring. Typically the activation energy for the noncatalytic or thermal
conversionofCOtoCO
2
isabout40KJ/mole.

Figure1.4 Activationenergydiagramfora)thermalreactionofCOandO
2
andb)the
same reaction in the presence of Pt. Activation energy for the non
catalyzed reaction is E
nc
. The Pt catalyzed reaction activation energy is
designatedE
c
.NotethatheatofreactionHisthesameforbothreactions.

Let us now discuss the effect of passing the same gaseous reactants, CO and O
2
, through a
reactorcontainingasolidcatalyst.Sincetheprocessisnowcarriedoutintwoseparatephasestheterm
heterogeneous catalytic reaction is used. In the presence of a catalyst such as Pt, the O
2
and CO
moleculesadsorbonseparatesitesinaprocesscalledchemisorptioninwhichachemicalpartialbondis
formed between reactants and the catalyst surface. Dissociation of chemisorbed O
2
molecules to
chemisorbedOatomsisveryrapidoccurringessentiallyatroomtemperature.Highlyreactiveadsorbed
OatomsreactwithchemisorbedCOonadjacentPtsitesproducingCO
2
whichdesorbsfromthePtsite
completingthereactionandfreeingthecatalyticsiteforanothercycle.ThepresenceofthePtcatalyst
greatlyfacilitatesthedissociationofoxygen,whichwas theslowenergyintensivestepassociatedwith
the gas phase oxidation of CO. Thus the activation energy for the Pt catalyzed reaction (E
c
), shown in
Figure1.4(b),isconsiderablysmallerthanthatforthenoncatalyzedreaction,enhancingtheconversion

9
kinetics. Typically the activation energy for the Pt catalyzed CO to CO
2
is lessthan about 20 Kcal/mole.
Thisdifferenceinactivationenergycanbeeasilyobservedwhencomparingthelightofftemperatureof
the catalyzed versus the gas phase reaction, which is illustrated in Figure 1.5. The noncatalyzed
reaction has a considerably higher lightoff temperature (around 700
o
C) due to its higher activation
energy.Moreinputenergyisnecessarytoprovidethemoleculesthenecessaryenergytosurmountthe
activationbarriersolightoffoccursathighertemperatures.Itshouldbenoted,however,thatthenon
catalyzed reaction has a greater sensitivity to temperature (slope of plot). Thus, the reaction with the
higher the activation energy has the greater sensitivity to temperature making it increase to a greater
extent with temperature than that with a lower activation energy. This is a serious problem for highly
exothermicreactions,suchasCOandhydrocarbonoxidation,wherenoncatalyticfreeradicalreactions,
with large activation energies, can lead to very undesirable products. Thus the reactor temperature
mustbecarefullycontrolledwithinthereactor.

Figure1.5 ConversionofCOvs.temperatureforanoncatalyzed(homogeneous)and
catalyzedreaction.

10
Equations relating reaction rates to activation energies will be discussed in considerable detail in
Chapter 4 but for now it is sufficient to understand that a relationship exists between the activation
energyandreactionrate.
Relativetoonethatisnotcatalyzed,kineticratestudiesindicatethattheratelimitingstepisthe
reactionofchemisorbedCOwithchemisorbedOatomsonadjacentPtsites.Thereactionoccursaround
100
o
C far below the 700
o
C required for the noncatalytic process described above. Thus the catalyst
providesanewreactionpathwayinwhichtheratelimitingstepisalteredfromoneofhightemperature
dissociation of O
2
to that of the reaction between two adsorbed moieties on adjacent Pt sites at a
significantly lower temperature. This shows the great importance of catalysis in enhancing rates of
reaction allowing them to occur at moderate temperatures as indicated in this example. Lower
operating temperature translate to energy savings, less expensive reactor materials of construction,
preferred product distributions with greater rates of production with smaller size reactors. For this
reasoncatalystsarecommonlyusedinmanyindustrialapplicationsrangingfrompetroleumprocessing,
chemicalandenergysynthesisandenvironmentalemissioncontrol(BartholomewandFarrauto2006).
Inspection of Figure 1.4 (b) indicates an energy decrease associated with the adsorption of CO
and O
2
on the Pt surface ( H
a
) due to its exothermic nature. This is a consequence of the decreased
entropy (S) when the molecules are confined in an adsorbed state with the commensurate loss in a
degreeoffreedom.SinceG
a
mustbenegativeandTS
a
ispositiveH
a
mustbenegativeinaccordance
withG
a
=H
a
TS
a
.Desorptionisalwaysendothermic.
Noncatalyzed reactions have much higher activation energies and therefore conversion is
initiated by at much higher temperatures by a different mechanism. This is clear from Figure 1.5 that
showsthehighertemperaturerequiredforthehomogeneousornoncatalyticreactions.Alsotheslope

11
ofthisreactionismuchlargerthanthatofthecatalyzedreaction.Thisclearlyillustratesthevalueofa
catalystfromtheenergypointofview.
1.2.3HomogeneousversusHeterogeneousCatalysis
In the past two sections, we presented two examples of catalysis. The first involved a redox
reaction between ions of Fe and Ce in the presence of Mn ions. In this example, the catalyst and
reactantswereinthesamephase.Whenthecatalystandreactantsexistinthesamephase,thecatalyst
isclassifiedasahomogeneouscatalyst.Mostofteninhomogeneouscatalysis,thecatalystiscodisolved
intoasolventwiththereactants,whichinthisexamplethesolventiswater.Inourredoxexample,the
homogeneouscatalystwasasimpleMnion.However,homogenouscatalystscanincludeawidevariety
of complex molecules such as metalligand complexes, organometallic complexes, and enzymes.
Homogeneous catalysts are typically limited to mild reaction conditions, however where they can be
used,theycanoffermanyadvantagessuchastheabsenceofdiffusionlimitationsandhigheffectiveness
of catalytic materials. One significant issue associated with homogeneous catalysis is separating the
catalyst from the reaction species, which typically requires significant additional processing steps
downstream of the catalytic reactor. Over the last five decades, the application of homogeneous
catalysishasgrowngreatlyandhasasignificantpotentialforthefuture.Homogeneouscatalysiswillbe
discussedindepthinChapterX.
When the catalyst and reaction species exist in different phases, this catalyst is classified as a
heterogeneous catalyst. Heterogeneous catalysts were the first to be used commercially and have
dominatedindustrialpracticeforoveracentury.Forpollutionabatementinautomobilesandfactories,
heterogeneous catalysis is exclusively used. For these reasons, this book focuses on heterogeneous
catalysis. The platinum catalyst used in our CO oxidation example is considered a heterogeneous
catalyst.Inthiscase,theCOmoleculesdiffusefromthegasphaseandadsorbontothesolidplatinum

12
surface where chemical reaction can occur. Unlike the homogeneous catalyst that is well dispersed
into the reaction mixture, heterogeneous catalyst can only occur on the solid surfaces exposed to the
gas or liquid phase containing the reactants. Thus, the surface area of exposed catalyst becomes a
criticalimportant.Inthecaseofpreciousmetalcatalyst,suchasourPtoxidationcatalyst,theplatinum
isdispersedintoahighsurfaceareasolidtomaximizetheexposedsurfacearea.Dispersionisoneissue
associated with heterogeneous catalysis. Another issue is the presence of diffusion limitations. To
diffuse from the bulk gas stream to the surface of the catalyst, the CO molecule must diffuse through
the thin boundary layer surrounding catalyst particle and diffuse through the pore structure of the
catalyst particle itself. In some cases, the rate of diffusion can be actually slower than the rate of
reaction. The physical processes occurring during heterogeneous catalysis and the concept of the rate
limitingstepwillbereviewedinmoredetaillaterinthischapter.
1.3SelectivityandCatalyticReactions
The selectivity of a reaction is generally defined as the ratio of the amount of desired product
formed to the amount of undesired product formed in a reaction. Mathematically, this ratio can be
expressedinmanydifferentterms,whichisgiveninEquation1.3.
Sclccti:ity =
moIc cscd ()
moIc 0ndcscd (0)
=
r

r
0
, (1.3)
The catalyst has the potential to greatly influence selectivity by preferentially lowering the activation
energyforaparticularstepinthereactionsequenceandincreasingtherateatwhichthisstepproceeds.
For the same reactants at the same conditions, different catalysts will influence the rate of different
reactionsthatarepossible.Thus,thecatalystcangreatlyinfluencereactionselectivity.Bychoosingthe
propercatalysttheselectivitytowardsthedesiredproductcanbeenhancedevenifotherproductsare
morethermodynamicallyfavored.Forexample,considertheoxidationofammoniafortheproduction

13
of nitric acid (Chapter X). Two ammonia oxidation reactions are possible, which are given in Equation
1.4and1.5.Fortheproductionofnitricacid,itistheformationofNOthatisthedesiredreaction.From
a thermodynamic perspective, the equilibrium constant for the N
2
forming reaction is four orders of
magnitude greater than the NO forming reaction, which means the formation of N
2
is greatly favored
thermodynamically. However, when this reaction occurs in the presence of a PtRh catalyst, the
formation of NO greatly exceeds that of N
2
. Thus, the PtRh catalyst greatly favors the reaction
sequencetoformNOandselectivelyenhancestherateofthisreactionovertheformationofN
2
.
4NH
3
+5O
2
4NO+6H
2
O (1.4)
4NH
3
+3O
2
2N
2
+6H
2
O (1.5)
Another example of a catalyst influencing product selectivity is the oxidation of ethylene.
Consider the comparison between the products produced when Pt is used as a catalyst as opposed to
oxides of vanadium for the oxidation of ethylene. In the case of Pt, the carbon in the ethylene is
completelyoxidizedtoCO2withaselectivityof100%.
Pt
C
2
H
4
+3O
2
>2CO
2
+2H
2
O (1.6)
However,theV
2
O
5
catalystfavorstheselectiveoxidationononlyonecarbonintheethyleneyieldingan
aldehydebecausethisreactionpathwaywiththiscatalysthasthelowestactivationenergycomparedto
thecompletecombustiontoCO
2
andH
2
O.
V
2
O
5

C
2
H
4
+1/2O
2
>CH
3
CH=O (1.7)
ForPtthereactionproductsareexclusivelyCO
2
andH
2
Othusselectivityisessentially100%makingita
good catalyst for pollution abatement. While for V
2
O
5
the selectivity is about 8090% towards the

14
aldehyde with the balance 2010% being CO
2
and H
2
O. Clearly V
2
O
5
would not be desirable for
conversion of hydrocarbons such as ethylene to harmless CO
2
and H
2
O but is used commercially for
selective partial oxidation reactions to desirable chemicals. So it is the function of the catalyst with
optimum reaction conditions to reduce the activation energy of the path that will yield the most
desirableproduct.Theabilityto1)enhancereactionratesand2)directreactantstospecificproducts
makescatalysisextremelyimportantintheenvironmental,petroleum,andchemicalindustries.
AnimportantreactionintheautomotivecatalyticconverteristhereductionofNObyH
2
during
a specific driving mode that will be discussed later in the automobile converter chapter. Two parallel
reaction pathways are possible one desirable leading to N
2
formation and the other undesirable
producingtoxicNH
3
.
Rh
2H
2
+2NO>N
2
+2H
2
O (1.8)
Pt
5H
2
+2NO>2NH
3
+2H
2
O (1.9)

Clearly Rh is more selective and dominates the NO to N
2
pathway with a rate considerable
higherthanthatundesiredpathwayleadingtoNH
3
formationwhenPtisused.
The reaction conditions also have a pronounced effect on product distribution
dependingonactivationenergiesforallpossiblereactions.ForexampleNO(componentofacid
rainandacontributortoozoneformation)emittedfromautomobileenginesand powerplant
exhausts can be reduced using a V
2
O
5
containing catalyst with high selectivity provided the
temperatureismaintainedbetween250300
o
C.

15
250300
o
C
2NH
3
+NO+O
2
>3/2N
2
+3H
2
O (1.10)
>300
o
C
2NH
3
+NO+5/2O
2
>3NO+3H
2
O (1.11)
AmmoniaalsodecomposestoN
2
above300
o
CandthusisnotavailabletoreducetheNO

>300
o
C
2NH
3
+3/2O
2
>N
2
+3H
2
O (1.12)
The desired reaction is favored below 300
o
C since it has the lowest activation energy of the
other two reactions. Once the temperature exceeds 300
o
C the reactions with the higher
activationenergy(greatertemperaturesensitivity)becomefavoredandmixedproductsform.

ExamplesofSelectivitycalculationsforreactionswithmultipleproducts
Generalequationforselectivity
xA+x/2ByC
SelectivityofC=(molesofCproduced)(x/y)x100
(molesofAconsumed)

2NO+2H
2
N
2
+3H
2
O
(1mole) (0.3moles)
(notbalanced)
2NH
3

(0.4moles)

SelectivitytoN2=(molesofN
2
formed)(2)x100=(0.3molesN
2
)2x100=60%
(molesofNOconverted) (ImoleofNOconsumed)

SelectivitytoNH3=(molesofNH
3
formed)x100=(0.4molesNH
3
x100=40%
(molesofNOconsumed) (1moleNOconsumed)

16
Carbonbalance=(SumofCatomsinallproducts*)x100=Shouldbeclosetocloseto100%
(SumofCAtomsinallreactants)

=includesunreactedreactants

1.4 PhysicalStructureofaHeterogeneousCatalyst
All heterogeneous catalysts are solid substances that can be classified into two general
categories: unsupported or supported. Most heterogeneous catalysts fall under the supported
category,wheretheactivematerialisdispersedintoacarrier(i.e.support).Theprimaryfunctionofthe
carrieristomaintainthedispersionoftheactivephase.Asanexampleofsupportedandunsupported
catalysts, consider the process for ammonia synthesis (Equation 1.13) where both supported and
unsupported catalysts are used. One catalyst used for this reaction is unsupported iron oxide. In the
preparationofthiscatalyst,theironoxideismeltedwithsmallamountsofotheradditives,thenpoured
into water to form small particles of the iron oxide catalyst. Thus, the bulk of the catalyst structure is
comprised of a porous iron oxide. Another catalyst used for this reaction is ruthenium supported on a
high surface area carbon carrier. In the preparation of this catalyst, a small amount of ruthenium is
dispersed into the carbon carrier. Thus, the bulk of the catalyst is comprised of carbon with small
islandsofwelldispersedruthenium.Bothtypesofcatalyststructuresarewidelyusedandtheoptimal
catalyststructureifoftenafunctionofseveralvariablesincludethenatureoftheactivespecies,catalyst
manufacturingcost,thedesiredcatalystformtobeusedintheindustrialprocess.
3H
2
+N
2
2NH
3
(1.13)
Whether supported or unsupported, the catalyst will be comprised on two basic components: active
species and promoters. In the case of the supported catalyst, a carrier material is also present. The

17
species which adsorbs the gas phase reactant and on which the surface reaction occurs is called the
active species. Within the active species, the specific atomic location where adsorption and reaction
occursiscalledtheactivesite.Itisontheactivesitewheretheshortlivedreactionintermediatesform
on the catalyst surface, which are called the active center. Beyond the active species, there are often
otherspeciesincorporatedintothecatalyststructure,calledpromoters,thateitherimprovestheactivity
oftheactivespeciesbypositivelyparticipatinginthesurfacereactionsequence(chemicalpromoter)or
helpsmaintainthecatalystactivityovertimebystabilizingthecatalyststructure(texturalpromoter).

1.4.1ActiveCatalyticSpecies
GroupVIIIBmetalsandtheiroxidessuchasFe,Co,NiarecatalyticasareCuandAg(Group1b),
V(GroupVb)andCrandMo(GroupV1b)andinspecificcombinationsaremainlyusedinthechemical
andpetroleumindustries.Thesearereferredtoasbasemetals(oxides).PreciousmetalssuchasPt,Pd,
RuandRharealsoinGroupVIIIareverycommonlyusedbroadlyinallindustries.Ironicallythe
preciousmetalsarealsoreferredtoasthenoblemetalsfortheirresistancetooxidation,variouspoisons
andhightemperaturesyettheyaresomeofthemostcatalyticallyactiveelementsinnatureduetotheir
abilitytochemisorbandconvertadsorbedspecieswithhighrates.Theyarerare,veryexpensiveand
thuswhennolongerperformingsatisfactorilytheyarerecycled,purifiedandreused.Theyareprimarily
minedinSouthAfricaandRussiawithsmalldepositsinCanadaandtheUS.Inmostcasesbothbase
andpreciousmetals(ortheiroxides)aredepositedonhighsurfaceareacarriersinordertomaximize
theiraccessiblecatalyticsites.
Itshouldbeunderstoodthatoftentheactivecatalyticcomponentisnotalwayspresentinits
nativeelementalstatebutmaybepresentasanoxide.FortheoxidationofmanyhydrocarbonsPdis

18
mostcatalyticallyactiveasPdOwhileforthehydrogenationreactionsNi,Cu,Pdmetalaremostactive.
Vanadiumpentoxide(V
2
O
5
)isanactivecatalystforoxidizingSO
2
toSO
3
inthemanufactureofsulfuric
acid.IncontrastNimetalisactiveforhydrogenationreactions.
Throughout this book you will see the breath of applications for all of these metal and their oxides as
catalyststhatenhanceactivityandselectivity.

1.4.2ChemicalandTexturalPromoters
Themaincatalyticcomponentdominatestheactivityandtheselectivitybutoxidesormetalpromoters,
thatmayormaynotbecatalyticforthereactionofinterest,canpromotetheactivityorfurtherenhance
one reaction over another. Promoters are generally classified as two types: chemical promoters and
textural (structural) promoters. In the case of the chemical promoter, the material facilitates the
surfacereaction.Inthecaseofthetexturalpromoter,thematerialstabilizesthemetaldispersionorthe
structure of the carrier, which prevents or slows catalyst deactivation over time. There are many
examplesforawidevarietyofpetroleum,chemical,environmentalandalternativeenergyprocesseswe
willdiscussinsubsequentchaptersbutjusttogiveapreviewherearesomeexamples.Theadditionof
CeO
2
toapreciousmetalcatalystsuchasPtorPdintheautomotivecatalystpromotestheoxidationof
hydrocarbons and carbon monoxide decreasing the minimum temperature needed to initiate catalytic
oxidation during the cold start portion of the automobile cycle requirement. Adding Cl to a naphtha
reforming catalyst PtRe/Al2O3 used to make high octane gasoline enhances the acidity of the catalyst
andcatalyzesisomerizationreactionsgeneratingbranchedhydrocarbonswithhighoctane.Theaddition
of sulfur to a hydro desulfurization catalysts (Co, Mo/Al2O3) diminishes over activity leading to
undesired gaseous products during the desulfurization of crude oils. The addition of alkali (K2O) to a

19
Ni/Al2O3steamreformingcatalystdecreasesacidityandpromotessteamgasificationminimizingcarbon
formation.Onemoreexamplewillbeinstructiveindemonstratingtheimportantroleofsmallamounts
of promoters in altering catalyst selectivity. In the generation of fuel cell quality H
2
the gas stream is
enriched in H
2
by promoting the oxidation of CO by water in the water gas shift reaction using a Pt
containingcatalyst.ForaPtonlycatalystappreciableamountsofundesiredmethaneareformedduring
the reaction at 300
o
C. The addition of 510% ZnO to the catalyst suppresses the methanation reaction
andavoidstheconsumptionofH
2
andthelargeexothermassociatedwiththereaction(Korotkikhetal
2003).
CO+H
2
O>H
2
+CO
2
(1.14)
CO+3H
2
>CH
4
+H
2
O (1.15)

1.4.3CarrierMaterials
Inmostindustrialreactions,thenumberofreactantmoleculesconvertedtoproductsinagiventimeis
directly related to the number of catalytic sites available to the reactants. It is, therefore, common
practice to maximize the number of active sites by dispersing the catalytic components onto a porous
material.Maximizingthesurfaceareaofthecatalyticcomponents,suchasPt,Fe,Ni,Rh,Pd,CuO,PdO,
CoO, and so forth increases the number of sites upon which chemisorption and catalytic reaction can
occur.Thecatalyticcomponentsareintroducedintothecarrierbyimpregnationfromaqueoussolutions.
ts on a high surface area carrier, such as Al
2
O
3
, SiO
2
, TiO
2
, SiO
2
Al
2
O
3,
zeolites, CeO
2
, etc. In most cases
carriersthemselvesarenotcatalyticallyactiveforthespecificreactioninquestion,butdoplayamajor
role in promoting the activity, selectivity and maintaining the overall stability and durability of the
finishedcatalyst.Suchisthecaseinthehydrogenationoforganicfunctionalgroupswherehighsurface
area carbons are used as carriers for precious metals and Ni. In these cases the carrier plays a role in

20
activity and selectivity by adsorbing the organic compound prior to hydrogenation by the metallic
catalytic component. For now when we speak of the catalyst it is understood to be composed of
catalyticcomponentsdispersedonhighsurfaceareaporouscarriers.Exceptionstothisbroaddefinition
willbediscussedintheapplicationsection.

1.4.4StructureoftheCatalystandCatalyticReactor

In chemical and petroleum applications many different catalyst structures can be used. The structure
depends on a number of factors such as volume of product to be produced, addition or withdrawal of
heat, thermodynamic equilibrium, need for frequent regenerations, life etc. The rationale for each
reactor type will be discussed in the specific application section. For now we will discuss packed bed
reactors(Figure16)whichcontainafixedvolumeofcatalystparticulates.

Figure1.6Packedbedreactorcontainingparticulatestructuredcatalysts.

21
Thecatalystispresentinapreshapedparticulatecontainingthecatalyticcomponentdispersedonthe
carrier. The size and shape of the catalyst depends on the nature of flow, permitted pressure drop,
requirement for heat management (exothermic or endothermic reactions), mechanical strength, etc.
Spherical and tablet shapes 310 mm are commonly used. The proper flow of inlet reactant gas is
established by individually controlled mass flow meters. The mixed reactants are continuously
preheated and the conversion of reactants and appearance of products observed using analytical
instrumentation such as a gas chromatograph or other suitable equipment calibrated for specific for
reactantsandproductsamounts.Theinlet,midbedandoutlettemperaturesarecommonlymeasured
usingthermocouples.
Thepreparationandpropertiesofthesematerialsandtheirinfluenceoncatalyticreactionswill
bediscussed inChapter2,but,fornow,Al
2
O
3
(the mostcommonlyusedcarrierincatalysisespecially
forchemicalandpetroleumapplications)willbeusedtodevelopamodelofaheterogeneouscatalyst.
TheporesofAl2O3thecarrieraretypically210nmindiameterbutirregularlyshaped.Thesimplistic
drawingshowninFigure16hascircularcatalyticcomponentsdispersedonthewallssimilartoraisinin
acake.Thephysicalsurfaceareaofthecarrieristhesumofallinternalareasoftheoxidefromallthe
walls of each and every pore. It is upon these internal walls that the catalytic components are bound.
Thecatalyticsurfaceareaisthesumofalltheareasoftheactivecatalyticcomponentsinthisexample.
The smaller the individual size of the active catalytic material (higher catalytic surface area), the more
sitesareavailableforthereactantstointeract.Asaroughapproximation,oneassumesthehigherthe
catalytic surface area, the higher the rate of reaction for a process controlled by kinetics. This is often
the case but there are exceptions in which a particular reaction is said to be structurally sensitive and
therateisamaximumwheninteractingwithacatalyticcrystalsizeofaspecificsizerange.

22
Thisproceduremaximizesthecatalyticareabutalsointroducesotherphysicalprocessessuchas
masstransferofthereactantstothecatalyticsites.Eachoftheseprocesseshasarateinfluencedbythe
hydrodynamics of the fluid flow, the pore size and structure of the carrier, and the molecular
dimensionsofthediffusingmolecule(FarrautoandBartholomew2006,Morbidellietal.2001).Theseare
discussedbelow.

1.5 AdsorptionandKineticModelsforHeterogeneousCatalysis
Inheterogeneouscatalysis,theinitiatingandterminatingstepsforthesurfacereactioninvolves
the adsorption of the reactant onto and the desorption of the product from the surface. In chemical
terms, adsorption is the formation of chemical bonds between the adsorbing species (adsorbate) and
theadsorbingsurface(adsorbent).Ingeneral,therearetwokindsofadsorptions,chemicalandphysical,
typically referred to as chemisorption and physisorption. Physisorption is the relatively weak,
nonselectivecondensationofgaseousmoleculesonasolidatrelativelylowtemperatures.Incontrast,
chemisorptionistherelativelystrong,selectiveadsorptionofchemicallyreactivegasesonavailablesites
of metal or metal oxide surfaces. In chemisorption, the adsorbentadsorbate interaction involves the
formationofchemicalbondsandheatsofreactionontheorderof50300kJ/mol.Itistheformationof
thesestrongbondsthataltersthechemicalnatureoftheadsorbate,makingthenewlyformedstructure
morereactiveandmoreeasilytransformedthanthegaseousmolecule.
The maximum number of reactant molecules adsorbed onto a unit area of catalyst surface will
bethenumberofavailablecatalyticsitesperunitarea.Inthecasewhereallcatalyticsitesarecovered
with an adsorbate molecule, the surface is said to be saturated and the fractional site coverage () is

23
unity. Adsorption and desorption is continuously occurring even when the net number of adsorbed
species reaches a steady state. Thus, an equilibrium coverage is established, where the fractional
coverage at equilibrium is a function of temperature, adsorbate partial pressure, and the chemical
natureoftheadsorbate.
In heterogeneous catalysis, it is essential to know the amount of reactants that adsorb onto a
given area of catalyst surface. The quantity of adsorbed species is generally characterized by an
isotherm, which is a mathematical relationship between fractional site coverage () and temperature.
The rate of reaction on the catalyst surface is going to be a function of the surface concentration of
reactants, not necessarily the gas phase reactant concentrations. However, we can only measure the
gasphaseconcentrationofthereactant.Thus,theisothermprovidesuswiththelinkbetweenthegas
phaseconcentration(measurable)andthesurfaceconcentration(immeasurable).

1.5.1LangmuirIsotherm:
There are many different isotherm forms, but the Langmuir isotherm is the simplest and most
widely used form that is applicable to many reactions. The Langmuir isotherm is based on the key
assumption that all sites on the adsorbent surface are of equal energies. It also assumes the rate
limitingstepisthesurfacereactionbetweenadsorbedspecieswithallothersfastandinequilibrium.In
spiteofthisidealviewitsapplicationyieldsreasonablepredictions.
ConsidertheexamplereactionpresentedearlierwhereCOisoxidizedonaplatinumcatalyst.In
thisreaction,COandO
2
mustadsorbontotheplatinumsurfaceforreactiontooccur.Further,theO
2
Pt
must dissociate to form two PtO species. We can use the Langmuir isotherm to predict the surface

24
concentration of each reactant and how the surface concentrations will vary with reaction conditions.
WefirstconsiderthestrongadsorptionofCOinequilibriumwiththesurfaceofPt.
CO+PtCOPt (1.16)
TheforwardrateofEquation1.16(COadsorption)isgivenby
(Rate)
fCO
=k
fCO
P
CO
(1
CO
) (1.17)

k
fCO
= the forward rate constant of CO adsorption on Pt
,
P
CO
= the partial pressure of CO and
CO
=the
fractionof thesurfaceof PtcoveredbyCO. Theterm (1
CO
) isthefractional numberofsitesavailable
foradditionalCOadsorptiononthePtsurface.Theisothermalsoassumeseachsiteisoccupiedbyonly
oneadsorbatemoleculeandfullcoverageisamonolayer.
TherateorreverseofEquation1.16(COdesorption)isgivenby
(Rate)
dCO
=k
dCO

CO
(1.18)
Atequilibriumtheforwardanddesorptionratesareequalandtheratioof theforwardratetoreverse
rateisdefinedastheadsorptionequilibriumconstant(Equation1.19)
k
fCO
/k
dCO
=K
CO
(1.19)
k
fCO
P
CO
(1
CO
)=k
dCO

CO
(1.20)

CO=
K
CO
P
CO
/(1+K
CO
P
CO
) (1.21)
Thus, Equation 1.10 provides the surface concentration of adsorbed CO as a function of CO partial
pressure and the adsorption equilibrium constant. The adsorption equilibrium constant captures the

25
influence of temperature and chemical nature of the adsorbate since the forward and reverse rate
constantsareafunctionoftheseparameters.PlottingEquation(1.10)generatesFigure17

Figure1.7 Adsorption isotherm (

CO
)

for CO on Pt for large, moderate and low

partialpressuresofCO.TheslopeatlowpartialpressuresofCOequals
theadsorptionequilibriumconstantK
CO
.

Wecannowconsidertwolimitingcases:onewherethepartialpressureofCOisverylow(P
CO
issmall)
andonewherethepartialpressureofCOisverylarge(P
CO
islarge).
WhenP
CO
islarge,thedenominatorinEquation1.21(1+K
CO
P
CO
)isapproximatelyequalto(K
CO
P
CO
)since
K
CO
P
CO
>>1. Thus, Equation1.21reducesto
CO
=1 thatsignifies allthecatalyticsitesarecoveredwith
COmoleculesandwesaythesurfaceissaturatedwithCO.
When P
CO
is small, the denominator in Equation 1.21 (1 + K
CO
P
CO
) is approximately equal to 1 since
K
CO
P
CO
<<1andEquation 1.21 reduces to
CO
=

K
CO
P
CO
. Under these conditions, the surface
concentration responds linearly with increasing gas phase concentration where the adsorption
equilibrium constant, K
CO
. Recall the equilibrium constant is defined as the ratio of the forward to
reverseadsorptionrates,Equation1.19.Anadsorbatethatstronglyadsorbsontothesurfacewillhavea

26
relatively large equilibrium constant and surface coverage will increase rapidly with increasing gas
partial pressure. This is the case for CO on Pt, which is illustrated in Figure 14 as the steeply sloped
traceatlowP
CO
.

AtmoderatevaluesofP
CO
,Equation1.21appliesaswrittenwiththecurvebecomesmoreshallowasP
CO

increases.
TheisothermfordissociativechemisorptionofO
2
onPtissimilarlygeneratedwhere
O
refersto
thefractionalcoveragebyOatomsconsistentwiththestoichiometryforCOoxidationofoneOatomfor
eachCO.
O
2
+Pt2OPt (1.22)
Therateoftheforward(adsorptionofO
2
onPt)
(Rate)
fO2
=k
fO2
P
O2
(1
O
)
2
(1.23)
Therateofthereversereaction(OdesorptionfromPt)is
(Rate)
dO2
=k
dO2

O
2
(1.24)
Thesquaretermforboththeforwardandreverseratesisduetothelowerprobabilitythattwoadjacent
Pt sites will be available to accommodate two oxygen atoms resulting from the dissociative
chemisorption of O
2
on Pt. Similarly two adsorbed O atoms on Pt must be adjacent for desorption and
recombination of diatomic O
2
to occur. Also the adsorption equilibrium constant (K
O2
)

for O
2
on Pt is
k
fO2/
k
dO2
.
Equatingforwardandreverserates
k
fO2
P
O2
(1
O
)
2
=k
dO2

O
2 (1.25)

27

O
=K
O2
1/2
P
O2
1/2
/(1+K
O2
P
O2
)
1/2
(1.26)
PlottingthefractionalcoverageofoxygenatomsversesP
O2
1/2
generatesasimilarplotasFigure1.5but
theslopeatlowP
O2
=K
O2

AthighP
O2

and

O
=1.
The isotherm provides us with the relationship for reactant surface concentration as a function of
reaction conditions (i.e. temperature and reactant partial pressure). The next step is to use this
information to estimate the rate of chemical reaction occurring on the catalyst surface. For this, we
needtomakesome

1.5.2ReactionKineticModels
Theisothermprovidesuswiththerelationshipforreactantsurfaceconcentrationasafunction
of reaction conditions (i.e. temperature and reactant partial pressure). The next step is to use this
informationtoestimatetherateofchemicalreactionoccurringonthecatalystsurface.Todevelopan
expression for reaction rate, we need to know or assume a mechanism for the surface reaction.
Specifically, the mechanism considers which reactants adsorb onto the surface and on which active
species do they adsorb? While anything is possible, three general mechanism are used to describe
adsorptionandreactioninheterogeneouscatalysis:
LangmuirHinshelwoodMechanism Reactantsadsorbontothesameactivespeciesandthuscompete
foravailableactivesites.

28

Figure1.8 IllustrationofLangmuirHinshelwoodReactionMechanism

MarsvanKrevelenMechanism Reactants adsorb onto different active catalyst species, then

the two surface intermediates interact and reaction occurs.
Since each reactant adsorbs onto different surface species,
thereisnositecompetitionbetweenreactants.

Figure1.9 IllustrationofMarsvanKrevelenReactionMechanism

EleyRidealMechanism One reactant adsorbs onto the catalyst surface forming an

intermediatesurfacespeciesthatthenreactswithagasphase
reactant.Thus,thereisnocompetitionforcatalyticsites.

29

Figure1.10 IllustrationofEleyRidealReactionMechanism

Each of these three mechanisms is described in more detail in the following sections. The critical
question is determining which mechanism a particular reaction follows. For each mechanism, a rate
equation can be derived in terms of the appropriate surface and gas phase reactant concentrations.
Once derived, experiments can be conducted to see which rate equation best describes the
experimental data. One possibility will be that neither of these three mechanisms describes the
experimental data, which suggests the existence of a combination of mechanisms occurring
simultaneouslyonthesurfaceortheexistenceofanentirelydifferentmechanism.

1.5.2.1 LangmuirHinshelwoodKineticsforCOoxidationonPt
The most widely accepted kinetic model for the CO oxidation reaction on Pt is based on the Langmuir
isothermfromwhichisderivedLangmuirHinshelwood(LH)kinetics(Hinshelwood1940andHougenand
Watson1943).TheadsorptionisothermsforCOandO
2
wereconsideredseparatelyin1.5.1butforthe
oxidationofCObyO
2
itisnecessarytoconsiderbothgasespresentwitheachcompetingforthesame
sitesonPt.WewillusekastherateconstantfortheoxidationofCOwherethenetrateofreactionfor
COoxidationwillbe

30
[Rate]
CO
=k
CO

O
(1.27)
The surface coverage of CO and O,
CO
and

O
respectively
,
must be modified to account for the
competitiveadsorptionthatoccursontheactivePtsites.Specifically,thefractionofopensitesusedto
determinetherateofadsorptionmustincludetermsforbothOandCOaswritteninEquation1.28.
(Rate)
fCO
=k
fCO
P
CO
(1
CO

O
) (1.28)
However, the desorption rate for CO is still only dependent on the sites occupied by CO and remains
unchangedfromEquation1.18.
(Rate)
dCO
=k
dCO

CO
(1.29)
Equatingadsorption(1.28)anddesorption(1.18)ratesatequilibriumandrecognizingthat
k
fCO
/k
dCO
=

K
CO
yieldsanexpressionforthesurfaceconcentrationofCO.
K
CO
P
CO
=
CO
(1
CO

O
) (1.30)
However, Equation 1.30 contains two unknowns,
CO
and

O
. Before we can solve Equation 1.30 any
further,weneedanexpressionfor
O
.Followingasimilarapproachtothatusedpreviously,fortherate
ofadsorptionanddesorptionofO
2
weobtain
K
O2
1/2
P
O2
1/2
=
O2
/

(1
CO

O
) (1.31)
Simplificationisachievedbydividing(1.30)by(1.31)

O
=
CO
K
O2
1/2
P
O2
1/2
/K
CO
P
CO
(1.32)
Wenowsubstitute(1.32)into(1.30)toyieldanexpressionfor
CO
intermsofknownquantities.

CO=
K
CO
P
CO
/(1+

K
CO
P
CO
+

K
O2
1/2
P
O2
1/2
) (1.33)

31
Similarly,wesubstitute(1.33)into(1.32)toyieldanexpressionfor
O
intermsofknownquantites.

O
=K
O2
1/2
P
O2
1/2
K
CO
P
CO
/(1+

K
CO
P
CO
+

K
O2
1/2
P
O2
1/2
) (1.34)
Nowwehaveanexpressionfor
CO
and
O
thatcanbesubstitutedintoEquation1.27todeveloparate
expressionforCOoxidationintermsofthesurfaceconcentrationofCOandO.
[Rate]
CO
=k
CO

O
(1.27)
[Rate]
CO
=kK
CO
P
CO
K
O2
1/2
P
O2
1/2
/(1+K
CO
P
CO
+K
O2
1/2
P
O2
1/2
)
2
(1.35)
LetsnowconsidertheconcentrationextremeswhentheconcentrationofCOisverylowandveryhigh.
ForlowP
CO
,Equation1.35reducestoEquation1.36.
[Rate]
CO
=kK
CO
P
CO
K
O2
1/2
P
O2
1/2
/(1+K
O2
1/2
P
O2
1/2
)
2
(1.36)

ThisshowsadirectrelationshipbetweentherateandP
CO
whenP
O2
isconstant
Forlarge

P
CO
,

Equation1.35

reducestoEquation1.37.
[Rate]
CO
=kK
O2
1/2
P
O2
1/2
/(P
CO
K
CO
)

(1.37)
Equation 1.37 indicates that the rate of reaction will decrease as the gas phase concentration of CO
becomes high. This means high CO concentrations inhibit the rate of reaction, which is due to CO
saturatingthecatalystsurfaceanddisplacingadsorbedoxygen.Thus,thereactionrateapproacheszero
a high and low CO concentrations, which indicates a maximum in rate exists when
O
=
CO
, which is
showngraphicallyinFigure1.11

32

Figure1.11 LHkineticsappliedtoincreasingP
CO
atconstantP
O2
.Maximumratewas
achievedwhenanequalnumberofCOmoleculesandOatomsareadsorbed
(
O
=
CO
)onadjacentPtsites.

Applying this model was very useful in designing an optimum system for the first gasoline
oxidationcatalystfortheautomobileconverter.WhentheCOwashighduringthecoldstartportionof
thedrivingcycletheadditionofextraO
2
(fromair)decreasedtheP
CO
morethanP
O2
andtherateofthe
reactionforCOoxidationincreased.Thusunderstandingkineticsandtherateexpressionshelpeddesign
aworkablesystemtomeetregulations.
From a fundamental point of view it should be noted that the assumption of uniform energy
sitesonthecatalystintheLangmuirisothermisnotcorrect.Aheterogeneouscatalyticsurfaceconsists
of a distribution of strong, moderate and weak sites upon which the reactant molecules adsorb.
Naturally the weak sites are the least strongly held so with increasing temperature only the stronger
sites retain adsorbed molecules and the fractional coverage decreases. This results in a change in the
overall energy of the adsorbed states on the activation energy profile of Figure 1.2 b). Fundamentally
with increasing temperature this causes a small change in the activation energy but for all intensive

33
purposes can be ignored when making activation energy measurements. Measurements of activation
energieswillbediscussedlater.

1.5.3MarsvanKrevelenKineticMechanism
Thereareanumberofreactionsthatproceedbyasurfaceredoxmechanismwhereoxygenisprovided
by one catalytic component to an adsorbed reactant on another catalyst site leading to oxidation.
ExamplesforthismechanismapplywhenV
2
O
5
isthecatalystfortheoxidationofsulfurdioxidetosulfur
trioxide. It also applies to oxidation of benzene to benzoquinone. We will use an example of the
oxidation of CO over copper oxide and iron oxide to demonstrate the mechanism and rate equation.
These mechanisms are known as MarsVan Krevelen (MvK). In this case gas phase CO reduces Cu
+2
to
Cu
+1
(Equation1.38).TheFeOcontributesitsOtoreoxidizetheCuto+2withreducedFeforming
(Equation1.39).GasphaseO
2
thenadsorbsandoxidizestheFetoFeO(Equation1.40).
CO+2CuOCu
2
O+CO
2
(1.38)
Cu
2
O+FeO2CuO+Fe (1.39)
O
2
+FeFeO (1.40)
NetreactionCO+1/2O
2
CO
2

TheoverallrateofreactioncanbewrittenrecognizingthatequalsthefractionalcoverageofOatoms
ontheFespecies.
OveralloxidationofCO
Rate
CO
=k
1
P
CO
(1.41)

34
RateofreoxidationoftheFe,
Rate
o
=k
2
PO
2
1/2
(1) (1.42)
The overall oxidation rate of the CO can be no faster that the rate at which Fe is reoxidized at steady
statesowecanequatethetwoequations
k
1
P
CO
=k
2
PO
2
1/2
(1) (1.43)
=k
2
P
O2

1/2
/k
1
P
CO
+k
2
P
O2
1/2
(1.44)
TheMvKrateequationforCOoxidationcanbewritten
R=k
1
k
2
P
CO
P
O2
1/2
/

k
1
P
CO
+k
2
P
O2
1/2
(1.45)
ThisexpressionrecognizesthattherateisdependentonthegasphasepartialpressureofbothCOand
O
2
.
AtlowCO/O
2
theratereducestok
1
PCO
AthighCO/O
2
theratereducestok
2
PO
2
1/2

1.5.4EleyRidealKineticMechanism
Thismodelrepresentsacasewhereagasphasereactantreactswithareactantthathasadsorbedonto
thecatalystsurface.
A
(s)
+B
(g)
C
(g)
+D
(g)

AssumingCandDdonotadsorbandAadsorbswithanequilibriumconstantK
A
andgasspeciesBdoes
notcompetewithAforcatalyticsitesandonlyadsorbsonadsorbedA,theexpressionreducesto

35
Rate=k
A
P
B
(1.46)
Rate=kK
A
P
B
/(1+K
A
) (1.47)
This rate model has not been proven to apply to any reaction to date but is only shown in hopes that
someday it will be useful. There is some speculation that the oxidation of ethylene to ethylene oxide
withaAgcatalystmightfollowaEleyRidealmechanism.IfsoO
2
dissociativelychemisorbsontotheAg
andethylenethenadsorbsontheadsorbedOAg.
Itshouldbeunderstoodthatempiricalmodels(seechapter4)areoftenmoreusefulthankineticmodels
forindustrialapplicationssincereactionsarecarriedoutofteninregimesinwhichkineticsanddiffusion
aresignificant.

1.6 ChemicalandPhysicalStepsinOccurringduringHeterogeneousCatalysis
1.6.1MassTransferPhenomenaandSurfaceReaction
To maximize reaction rates, it is essential to insure accessibility of all reactants to the active catalytic
component sites dispersed within the internal pore network of the carrier. Once again, lets consider
thephysicalandchemicalstepsoccurringduringheterogeneousCOoxidationinapackedbedcatalytic
reactor.
COandO
2
moleculesareflowingthroughabedofasolidparticulatecatalyst.TobeconvertedtoCO
2
,
thefollowingphysicalandchemicalstepsmustoccur:
1. CO and O
2
must make contact with the outer surface of the carrier (or
washcoat in the case of a monolithic supported catalyst) containing the
catalytic sites. To do so, they must diffuse through a stagnant thin layer of

36
gas or boundary layer forming around the solid catalyst particles. Bulk
moleculardiffusionratesvaryapproximatelywithT
3/2
andtypicallyhavean
apparentactivationenergies,E
1
=24Kcal/mole.
The term apparent activation energy is used here to distinguish the
physical phenomena of diffusion from the truly activated chemical
processes that occur at the catalytic site. Diffusion reactions are a physical
phenomenaandthusarenotanactivatedprocesses.Thustermapparent
activation energy is a convenient term used to give a figure of merit for
reactionsensitivitytotemperature.
2. Since the bulk of the catalytic components are internally dispersed, the
majority of CO and O
2
molecules must diffuse through the porous network
towardstheactivecatalyticsites.Theapparentactivationenergyforpore
diffusion,E
2
,isapproximately1/2thatofachemicalreactionorabout610
Kcal/mole.
3. Once the molecules CO and O
2
arrive at the catalytic site, O
2
dissociates
quickly and chemisorption of both O and CO occurs on adjacent catalytic
sites.Thekineticsgenerallyfollowexponentialdependenceontemperature;
i.e.,exp(E
3
/RT),whereE
3
istheactivationenergy,whichforchemisorption,
istypicallygreaterthan10Kcal/mole.
4. AnactivatedcomplexformsbetweenadsorbedCOandadsorbedOwithan
energyequaltothatatthepeakoftheactivationenergyprofilesincethisis
the rate limiting step. At this point the activated complex has sufficient

37
energy to convert to CO
2
which remains adsorbed on the catalytic site
Kinetics also follow exponential dependence on temperature; i.e., (E
4
/RT)
withactivationenergiestypicallygreaterthan10Kcal./mole.
5. CO
2
desorbs from the site following exponential kinetics; i.e., exp(E
5
/RT)
withactivationenergiestypicallygreaterthan10Kcal./mole.
6.The desorbed CO
2
diffuses through the porous network towards the outer
surfacewithanapparentactivationenergyandkineticssimilartostep2.
7. CO
2
must diffuse through the stagnant layer and, finally, into the bulk gas.
Reaction rates follow T
3/2
dependence. Apparent activation energies are
alsosimilartostep1)lessthan24KCal/mole.

Steps1and7representbulkmasstransferwhichisafunctionofthespecificmolecules,thedynamicsof
the flow conditions, and the geometric surface area (outside or external area) of the catalyst/carrier.
Porediffusion,illustratedinsteps2and6,dependsprimarilyonthesizeandshapeofboththeporeand
the diffusing reactants and product. Steps 3, 4 and 5 are related to the chemical interactions of reac
tantsandproducts(i.e.CO,O
2
andCO
2
)atthecatalyticsite(s).Thesequenceofthesethreeprocessesas
illustratedinFigure1.12.

38
Figure1.12: Illustration of the sequence of chemical and physical steps
occurringinheterogeneouscatalysis

Anyofthe7stepslistedabovecanberatelimitingandcontroltheoverallrateofreaction.The
ratecontrollingprocessthendictatestheoverallrateofreaction observedexperimentally. Incatalytic
processes,itisessentialtounderstandwhatprocesslimitstheoverallrateofreaction.Therateofeach
oftheseprocessesisafunctionofreactionconditions,thedesignofthecatalystbed,andpropertiesof
the catalyst. As reaction conditions change, so will the rate of each of these processes and the rate
limitingstepmaychange.Forexample,considertheeffectoftemperatureontherateofahypothetical
reactiondescribedinthetemperatureconversionplotinFigure1.13.

Figure1.13 Conversionversetemperatureprofileillustratingregionsforchemicalkinetics,
porediffusionandbulkmasstransfercontrol.

Atalowtemperature,therateofreactionischemicallycontrolled.However,chemicalreactionrateis
verysensitiveto temperature (highactivationenergy),sotherateofchemicalreactionincreasesmore
with temperature than the diffusion processes. Pore diffusion then becomes rate limiting as the
temperature increases but eventually the least temperature sensitive bulk mass transfer process

39
becomesratelimiting.ThisisalsodemonstratedinFigure114whichdepictsthethreerelativeratesof
reaction. It shows that of the three rate limiting phenomena, bulk mass transfer (BMT) is the fastest
process at low temperatures but has a shallow dependence on temperature due to its low apparent
activation energy. Pore diffusion has a lower reaction rate than BMT due to its higher apparent
activation energy but its temperature dependence is greater. The highest temperature dependence
occursforareactioncontrolledbychemicalkineticsbutbecauseofitshigheractivationenergytherate
islowatlowtemperaturesrelativetothosecontrolledbydiffusion(Morbidellietal.2001).

Figure114 Relative rates of bulkmass transfer, porediffusionandchemicalkinetics asa

function of temperature. Chemical kinetics controls the rate between
temperatures A to B. Pore diffusion controls from B to C while bulk mass
transfercontrolsattemperaturesgreaterthanC.

1.6.2 Reactantconcentrationgradientswithinthecatalyzedwashcoat
In the chemical kinetic control region the reaction of chemisorbed CO with chemisorbed O is
slowrelativetodiffusionandthusisratelimiting.Asthetemperatureisfurtherincreasedcontrolofthe
overallratewillshift toporediffusion.Here thesurfacereactionbetweenCOand Oisfasterthanthe

40
rate in gaseous CO and O
2
can be supplied to the sites and a concentration gradient exists decreasing
within the washcoat. This is referred to as intraparticle diffusion in which the catalytic components
deepwithinthewashcoatarenotbeingcompletelyutilizedorhaveaneffectivenessfactorlessthan1.
Theeffectivenessfactoristheratioof theactualrateverses the theoretical maximumrate and canbe
thought of as a measure of the utilization of the catalytic component(s). At higher temperatures the
rate of diffusion of the CO and O
2
from the bulk gas to the external surface of the washcoat is slow
relativetotheotherprocessesandtheratebecomescontrolledbybulkmasstransfer.Inthisregimethe
CO and O
2
are converted to CO
2
as soon as they arrive at the external surface of the washcoat. The
concentration of reactant and product is essentially zero at the external interface of the washcoat and
the bulk fluid. The effectiveness factor is close to zero. Figure 1.9 graphically shows the relative
gradientsinconcentrationforreactantsforthethreeratecontrollingprocessesinacatalyzedwashcoat.

Figure1.9 Reactantconcentrationgradientswithinasphericalstructuredcatalyst
forthreeregimescontrollingtherateofreaction.

41
1.6.3 TheRateLimitingStep
Theefficiencywithwhichacatalystfunctionsinaprocessdependsofwhatcontrolstheoverall
reaction rate. If the kinetics of a process are measured and found to be in a regime where chemical
kineticsareratecontrollingthecatalystshouldbemadewithashighacatalyticsurfaceareaaspossible.
This is accomplished by increasing the catalytic component loading and/or dispersion so that every
catalytic site is available to the reactants. Furthermore, the catalytic components should be dispersed
uniformly throughout the interior of the carrier; with an effectiveness factor approaching 1. Process
parameters,suchasanincreaseintemperature,promotesareactioncontrolledbychemicalkinetics.
Whenitisknownthataprocesswillhavesignificantporediffusionlimitationsthecarriershould
beselectedwithlargeporesandtolocatethecatalyticcomponentsasclosetothesurfaceaspossibleto
improve the effectiveness factor. To enhance the transport rate one can decrease the thickness of the
washcoat to decrease the diffusion path of reactants and products. A temperature increase will have
someeffectofenhancingtheratebuttoalesserextentthanforthosereactionscontrolledbychemical
kinetics.
Therateofmasstransferisenhancedbyincreasingturbulenceinthebulkgasandbyincreasing
the geometric surface area (i.e. external area) of the washcoated monolithic catalyst. This can be
accomplishedbyselectingamonolithwithahighgeometricsurfaceareaordensityofchannelsperunit
area.Clearlyincreasingthecatalyticcomponentssurfacearea,theloadingofthecatalytic components
or the size of the pores will have no effect on enhancing the rate of mass transfer since these catalyst
propertiesdonotparticipateintheratelimitingstep.Alsotemperaturewillhavevirtuallynoimpacton
theBMTrate.

42
References
Avgouropoulos,G.,Ioannides,T.,Papadopoulos,C.,Hocevar,S,Matralis,H.AcomparativestudyofPt/
Al
2
O
3
, Au/- Fe
2
O
3
and CuOCeO
2
catalysts for the selective oxidation of carbon monoxide in excess
hydrogenCatalysisToday75(2002)157.
Bartholomew, C., and Farrauto, R. and Fundamentals of Industrial Catalytic Processes, Second edition,
WileyandSons,Hoboken,NJ(2006).
Bollinger,M.andVannice,M.A.AkineticsandDriftstudyoflowtemperaturecarbon
monoxideoxidationoverAuTiO2catalystsAPCATB:Environmental8(1996)417.
Bond,G.,Lous,CandThompson,D.CatalysisbyGoldImperialCollegePress,London(2006).
Grisel,R.andNieuwenhuys,B.AcomparativestudyoftheoxidationofCOandCH
4
over
Au/MO
x
/Al
2
O
3
catalystsCatalysisToday64(2001)69.
Hinshelwood,C.N.Thekineticsofchemicalchange,Oxford,ClarendonPress,London1940.
Hougen,O.andWatson,K.SolidCatalystsandReactionRatesGeneralPrinciplesIndustrial
EngineeringChemistry35,(1943)529.
Hurtado, P. Ordonez, S., Sastre, H. and Diez, F. Development of a kinetic model for the oxidation of
methane over Pd/Al2O3 at dry and wet conditions Applied Catalysis B: Environmental 51 (2004) 229
238.
Korotkikh,O.andFarrauto,R.SelectivecatalyticoxidationofCOinH
2
:FuelcellapplicationsCatalysis
Today62(2000)2.

43
Korotkikh,O.,Ruettinger,W.andFarrauto,R.(2003)Suppressionofmethanationactivityby
watergasshiftreactioncatalystUS6,562,315.
Liu X, Korotkikh, O and Farrauto, R. (2002) Selective catalytic oxidation of CO in H
2
: a structural of Fe
oxidepromotedPt/aluminacatalystAppliedCatalysisB:Environmental226,293.
Mars, P. and Van Krevelen, D. Oxidations carried out by means of vanadium oxide catalysts Chemical
EngineeringSeries3(1954)41.
Morbidelli,M.,Garvriilidis,A.andVarma,A.,CatalystDesign:Optimaldistributionofcatalystinpellets,
reactorsandmembranes,CambridgeUniversityPress,Cambridge,England(2001).

44
Chapter1Questions

1. Giveeverydayexamplesofhowyouencounterratelimitingstepsusinganalogiestointrinsic,
porediffusionandmasstransfercontrolledreactions.
2. Giveexamplesofhowinyoureverydaylifeyoumakedecisionsregardingselectivity.
3. List3examplesforeachofpetroleumandchemicalproductsproducedbycatalyticprocesses.
4. Distinguishahomogeneouscatalystandprocessfromaheterogeneouscatalystandprocess.
5. Whyisitimportanttomaximizethenumberofactivesitesinaheterogeneouscatalyst?
6. Whatisthepracticalvalueinpreparingacatalystoradjustingtheprocessconditionsby
knowingtheactivationenergyofareaction?
7. Howwouldyouchangetheprocessconditionsbyknowingthatinahydrocarbonoxidation
reactionthehydrocarbonhasaverylargeinhibitioneffectasintheequation:
Rate~k[(Kads
HC
PP
HC
)(Kads
O2
PP
O2
)]/(KadsPP
HC
)
8a. Howdoesthepresenceofacatalystchangethethermodynamicequilibriumconstantofa
reaction?
b. Tworeactionsarethermodynamicallyfeasible(bothhavenegativefreeenergies)butoneis
muchmorenegativethantheother.Canacatalystdirectthereactantstotheleastfavorable?
9. Whatarethebenefitsofusingacatalystforabatingemissions?

10.DerivetheLHrateexpressionforthedecompositionofPH3onaWmetalliccatalystwhereno
productsadsorb.

45
W
PH3 P+3/2H2
Ans:Rate=k
PH3
=kK
PH3
P
PH3
/1+K
PH3
P
PH3

11.WritetheLHforN2OdecompositiononPtwhereboththeN2OandtheproductO2stronglyadsorb
ontothePt.MakeaplotofRatevsconversionofN2O

Ans:Rate=k(K
n2O
P
N2O
)/1+K
N2O
P
N2O
+K
O2
1/2
P
O2
1/2

PlotofRatevsConversionshowsasteepdroptozeroduetostrongO2adsorption.

13.Whatcatalystproperties(assumethecatalystisPtdispersedonporousAl2O3presentasathin
coating(washcoat)onaceramicmonolith)andprocessconditionswillenhancetherateofagasphase
reactionifthereactioniscontrolledby
a. chemicalkinetics?
b. porediffusion?
c. bulkmasstransfer?

Ans: a.IncreasethePtconcentration,increaseitscatalyticsurfacearea(increasedispersion
byincreasinginternalsurfaceareaofthecarrier),increasethetemperature
b.largerporesofthecarrier,catalyticcomponentclosertothesurface,thinner
washcoat,increasetemperature(enhancementissmallerthanforchemicalcontrol)
c.Increasegeometricsurfacearea(thinnercoating)ofthemonolithbutdepositthe
sameamountofcatalystonalargersurface,increaseturbulencebyrougheningsurface
and/orenhancingvelocityofgasreactantsbyincreasinglinearvelocity.

14Whattemperatureisrequiredforacatalystwithanactivationenergyof15,000cal/moltooperate
withthesamevolumeasonewithanactivationenergyofE=12,000cal/mol(operationtemperatureis
227oC)withthesameconversion?Assumekoisthesameforbothcatalysts.R=2cal/oK

46
Answerto14
ko/SV21.2Exp[12,000cal/mol/(2cal/oK)(500oK))]=ko/SV12Exp[15,000cal/mol/(2cal/oK)(T15))]

exp(6000/500oK)=exp(7500/T15)

lnofbothsides

6000/500=7500/T15orT15=625oK
15. Howwilltherateconstant(andhencethereactionrate)changewhenthetemperature
isincreasedfrom100oCto110oC?Assumeanactivationenergyof20Kcal/molewith
theuniversalgasconstant=2cal/degmole.Feelfreealsotousejoulesforactivation
energywith(R=8.3J/mol
o
K)
Ans:
100oC=373oK
110oC=383oK
A. k
383
=koexp[20,000/2(383)]
B k
373
=koexp[20,000/2(373)]

A/B
=exp[10,000/383+10,000/373]
=exp[26.10+26.80]=exp{0.7}
k
383
/k
373
=exp[0.7]
k
383
/k
373
=2

47
16.a.Considerareactionwithtwopossibleproductdistributions;onedesiredand.theothernot.Whyis
itimportanttoknowtheirrespectiveactivationenergies?
Ansa:Byknowingtheactivationenergyyoucanestimatetheratelimitingstep.Thiswillallowyouto
makeadjustmentsinthecatalystandprocesstemperaturetofavoroneovertheother.Low
Temperaturewillfavortheonewiththesmallestactivationenergy.
Youcancalculatetheratioofrateatanytemperatureforallreactions.Byusingratiosthek
0
cancel
andtherelativerateswillbedeterminedbythetemperaturesusingtheknownactivationenergy.

b..Whatconditionswouldyouuseintheprocesstofavorthedesiredreaction?Plottheirrespective
relativeratesvstemperature
Answerb:Operateatalowtemperaturetofavorthedesiredreaction.Thefavoredreactionwill
increasewithtemperatureslowerthantheundesiredone.Itallowsyoutoworkinatemperature
rangebestforthedesiredreaction.AplotofRatevsTwillshowahigherrateatthelowerTforthe
desiredreactionwiththelowerE.AstheTincreasestheundesiredRatewillovercomethemore
desired.
17.
Whatpropertiesoftheparticulatecatalystandprocessconditionscanbechangedtoenhancetherate
limitingstepforeachofthefourseparatecasesofratecontrol?
1.BulkMassTransfer
2.PoreDiffusion
3.Chemical
4.Heattransfer
Ans: 1.BMT:IncreasetheGSAbydecreasingparticulatesize,increaseturbulence,increasesurface
roughness.Verysmalleffectduetotemperature
2.PoreDiffusion:Increasetheporediameterofthecarrier,decreasetheparticlesizewhich
decreasesthediffusionpath,changetheshapeandgeometryoftheparticletodecreasethediffusion
path,depositcatalyticcomponentclosertotheoutersurface,mildeffectwhenthetemperatureis
increased.
3.Chemical:Increasecatalyticsitesbyincreasedloadinganddispersion,increasethe
temperature.

48
4.HeatTransfer:Selectacarrierwithalowheatcapacity(lowSA,moredense),useareactor
withasmalldiameter(smallerradius),Smallerparticlesinpackedbedtomaximizecontactand
radialconductivity,reconsideranduseametalmonolithuponwhichthecatalystisdeposited.