1 2 INTRODUCTION TO CATALYSIS i -11 ■ CIS In Ihc last few chapters we have been discussing principles of reactions, with a particnlai ■ inphasis on reactions in the gas phase. However, now we will be changing topics and Marl to discuss how one can change the rate of a reaction by changing the chemical ■ nvironment where the reaction occurs. In particular, we will discuss how one can use ,i catalyst or solvent to promote a desired reaction. Our approach will be a little bil nonstandard in that wc start by asking what we can do to increase rates. We will develop tin ideas into predictive tools later in this book. 12.1 INTRODUCTION So far we have been discussing how reactions occur when the reactants are isolated from I||i ii environment. However, in most cases one does not run reactions that way. Instead, niie runs the reaction in solution and/or in the presence of a catalyst. The solvent ami i .iialyst are active participants in the reaction. If you run the reaction in the wrong solvent, you will not get as much desired product. Usually the solvent and catalysts are carefully i lioscn to maximize the rate of production of some desired product or to eliminate sonic ule reaction. In the remainder of this chapter, we will give an overview of how catalysts in solvents work. Additional details will be given in later chapters. Let's start with catalysts. < )stwald defined a catalyst as a substance that changes the rate of reaction but that is not itself consumed in the process. When students read Ostwald's definition, students often think that the catalysts are nol Ki live participants in the reaction. However, in reality, the catalysts are active participants 689 ill I 111.* ir,u lion Thr inllllysl usually ii'.H Is Willi Ihr leiu Innls In lonil ;i stable COinplOI Reactants i catalyst = > complex Then the complex rearranges to yield products and regenerate the catalyst: Complex products + catalyst Notice that the catalyst is regenerated at the end of the reaction so there is no mi consumption of catalyst. One way to think about a catalytic reaction is to consider the catalyst to be analo...... to a printing press. A printing press takes in reactants, paper and ink, goes through a cycll of steps, and produces a product: a printed page. The printing press is not changed di.....| the process. In the same way a catalyst takes in reactants goes through a series of step! and produces products. The catalyst is not changed in the process. People often talk about catalytic reactions in terms of a catalytic cycle. The calalylir cycle is the sequence of steps that the system goes through, starting with reactanls ami ending with products. For example, Figure 12.1 shows a catalytic cycle for the forma...... of acetic acid via the Monsanto process: ill 11 ii'I........Ii IHM CH3OH + CO => CH3COOH (12.1) CO 1,,,.....12.1 A schematic of the catalytic cycle for acetic acid production via the Monsanto process Mm .....w:. in this figure are as described in Chapter 5. There are two catalysts: HI and [Rh(CO)2I2]~. Both catalysts are active participants in llir reaction. The HI reacts with the alcohol in the first step of the reaction. The [Rh(CO: )l | reacts with the product of the HI reaction. Still, both catalysts cycle through the reaction and are regenerated. The catalysts are not consumed even though the catalysts are active participants in the reaction. Most catalytic reactions look like the reaction in Figure 12.1. The catalyst participate* in the reaction, but the catalyst is regenerated, so the catalyst is not consumed in ilu reaction. Table 12.1 lists the catalysts used to promote some common reactions. The rate enhancement seen under typical conditions is also included in the table. Rates typically increase by factors of 103-109 with gas-phase catalysts. Solid- or liquid-phase catalysis can often increase rates by factors of 10I0-1020, and there are cases where rati enhancements as large as 1040 are seen. Clearly, these effects are large enough that they need to be understood. Table 12.1 The rate enhancement of a number of reactions in the presence of a catalyst Reaction Catalyst Rate Enhancement Temperature, K Ortho H2 2NH3 = C2H4 + H2 : H2 + Br2 = 2NO + 2H2 CH3COH = CH3CH3 = => para H2 N2 + 3H2 =>C2H6 (CH3)3COH: > 2HBr =!> N2 + 2H20 >CH4 + CO C2H4 + H2 =s (CH3)2CH2CH2 + H20 Pt (solid) Mo (solid) Pt (solid) Pt (solid) Ru (solid) h (gas) N02 (gas) HBr (gas) IO40 IO20 1042 1 x 108 3 x 1016 4 x 106 1 x 109 3 x 108 300 600 300 300 500 500 750 750 1000 =a 1 c o 8 0.001 o E 1E-6 1 E 9 rr 1E-12 1E-15 I I i i Catalyst alone s Gas phase — no_ —^wall reactions --**^A^pro^umate / / / \effect of walls - \\ i i i 500 1000 1500 2000 2500 Temperature, K I lyure 12.2 The rate of hydrogen oxidation on a platinum-coated pore calculated with (a) only n.'inrogeneous (catalytic) reactions, (b) only radical reactions, and (c) combined radical-homogeneous ii.. ictions. Another key experimental observation is that catalysts do not work over a wide range nl conditions. For example, Figure 12.2 shows some rates of reaction calculated tor hydrogen oxidation in a small pore. Notice that the rate of reaction reaches a maximum .ii an intermediate temperature, and then declines. The catalyst is ineffective at high iiinperature. Interestingly, at very high temperatures, the gas-phase reactions are much faster than ihe catalytic reactions. At the highest temperatures, the catalyst slows down the reaction by promoting termination reactions. i ivl null W c il III iMi "li NI I lllfl i AIAI r'll' 12.1.1 Classes of Catalysts Next wc want to change topics and discuss the various types of catalysis. There are I two broad classes of catalysts: homogeneous catalysis and heterogeneous catalystl Homogeneous catalysts are substances you add to a reacting phase lo speed up | given reaction, while heterogeneous catalysts act at the boundary of a phase to promote reactions. In the next few sections we will list some of the key catalyst systems. The sections .uc written to provide a general overview of the kinds of materials that are calalytically active and not to teach people why the catalysts work. When I teach the course, my student! often find the breadth of the material overwhelming. However, I usually tell students to just memorize the material, particularly the material in the tables, and then reread the chapter. 12.2 OVERVIEW OF HOMOGENEOUS CATALYSTS Let's start with homogeneous catalysts. Homogeneous catalysts are substances that are added to a reacting phase to speed up some desired reaction. Examples of homogeneous catalysts include. . Acids or bases . Metal salts • Enzymes • Radical initiators • Solvents In the next several sections we will list several examples of acid catalysts, metal catalysts, and enzyme and radical initiators. The lists are not meant to be complete. However, we wanted to give the reader a picture of the wide variety of materials that show catalytic properties. 12.2.1 Acids and Bases Table 12.2 lists a number of reactions commonly catalyzed by acids or bases. Acid catalysis includes acids such as HF or H2S04 and organic acids such as acetic acid or triflouroacetic acid. Basic catalysts include compounds such as sodium hydroxide. Generally any strong acid or strong base can be used to catalyze reactions. Acids and bases act by interacting with various hydrocarbon species to form carbocations. Recall that a carbocation is a hydrocarbon ion with a positive charge. Carbocations are very reactive species. The production of the very reactive species allows the reaction to occur at an enhanced rate. For example, during the acid-catalyzed alkylation reaction Benzene + ethylene > ethylbenzene (12.2) ImIiIo 12.2 Som* reaction" ' n..........ly ■ ntnly/nil hy iii.IiIn mill hi mim M......on I <.....|>l< a proton reacts with the ethylene to form an ethyl ion: H+ + CH2CH2-► [CH3CH2]+ I ..in. ii/.Mi.hi (icaiianging the .....line ol a molecule) Ml . I iii.mi (making very small mi ilri ules into a bigger molecule) I ihi king (taking a big molecule and breaking il iiiln Iwo smaller ones) i |i iiu .iiinii (attaching an acid in a base eliminating water) Vlilnl condensation reactions (combining two aldehydes i.\ eliminating water) Mi uliol dehydration (removing a hydrogen and in i >l I from an alcohol, IHnilueing a double bond) I iiiiiiuc polymerization (il. (11(11(11, :• I Ii I II ( IK It, (11,(11 I IK II. I (II,( II.I IL( II, => ((•||t('llj)('II(('lli)(C'4tk)) C|2H24 =>C7H,4+C5Hio CH3CH2OH + CHiCOOH => CH3COOCH2CH3 + H20 2CH3CH2CH2CHO => CH3CH2CH2CH = C(CHO)CH2CH3 + H20 CH3CH2OH => CH2 = CH2 + H20 typical Application (Mane enhancement, inoiioniei production, paraxj leni |m,i.llli 11,111. Pharmaceutical produi lion monomer production Fins chemicals; hulane 1 olelin octane Crude-oil conversion, dip .1..... Soap production, frugram c production Fine chemicals phai inai 1 ,111. ,1 production Alternative fuels Propylene => polypropylene Polymer production I he ethyl ion reacts with benzene to yield an ethylbenzene ion: [CH3CH2]+ +CfiH6-► [CH3CH2C6H6]+ I hen the ethylbenzene ion loses a proton: [CH3CH2C6H6]+-> CH3CH2C6H5 + H+ d'li (12...) Reaction (12.2) has a significant Pauli repulsion, but reactions (12.3)—(12.5) do not I he absence of Pauli repulsions makes acid-catalyzed reactions much quicker than die 1111 responding uncatalyzed reactions. We discussed some of the reactions of carbocations in Chapter 5. Most industtial 1 aihoeation reactions are catalyzed by acids or bases. 12.2.2 Metal Atoms Metal atoms can also be used as homogeneous catalysts. Table 12.3 shows the mosl important examples. Generally the transition metal catalysts work by binding lo the intermediates of a reaction, thereby increasing the concentration of the key intermediates I or example, the Wilkinson catalyst, Rh(P(C6H5)3)3CI, catalyzes the hydrogenation of ethylene. The reaction occurs via the following steps: (12.3) H2 + 2S C2H4 + S C2H4(ad) + H(ad) C2H5(ad) + H(ad) -> 2H(ad) -> C2H4(ad) -> C2H5(ad) + S -+ C2H6 + 2S (I2.Í.) (12.7) (12.8) (12.9) 'vinvilW'il i |i iMi H ii Nl ( him Ai Al ymi'l 60S where s is an emptj site <>n the rhod......i lustet snd the (od)s denote Npci ii ■ iiiim hed to the cluster. In this case the clustei is stabilizing hydrogen atoms and ethyl groupi In Chapter 4, we showed thai the rate ol a cyclic reaction is proportional to ttv concentration of the intermediates of the reaction. In fact, the Wilkinson catalyst stabilized Table 12.3 Examples of reactions catalyzed by homogeneous transition metal catalysts Reaction Catalyst Olefin polymerization lni .(II 1 .., 1 . 1 \! ( II ■ (Ziegler-Natta catalyst) Olefin hydrogenation Rh(P(C6H5)3)3Cl (Wilkinson catalyst) c2h4 + h2o —> acetaldehyde (Wacker process) PdCl2(OH)2 c2h4 + h2 + CO —► propylaldehyde (hydroformylation) HCo(CO)4 CH3OH + CO -> CH3COOH (monsanto carbonylation [Rh(CO)2I2]1- process) Fe2+ H202 + ch3ch2oh -> ch3cho + 2H20 ill' 1111 > 111 in 11,1 u- Mini 1111 ses iIm 111111. nu.ih.ni nl (he miei niriliiili's As 11 result, the rule "i n .11 nun is tremendously enhanced tin mechanism Ol IMtll catalysts will he ihseussetl in detail 111 Cliaplei II llowevei, in. ilnii)' to leineinher 101 now is 11 in I metal atoms stabilize radical intermediates. The 1 it 1111 /,it 1011 speeds up rales. Industrially, hydroformulation is the largest application of homogeneous catalj 1 in. Monsanto process is also the main route that people use to produce acetii .11 ni Ii 1 Natta catalysts are used to produce polyethylene. There are many othei examples III reactions catalyzed by transition metal compounds in solution; however. Tahle I ' 1 ii 1 1 In- main reactions used commercially. 18.2.3 Enzymes I n/\ mes are another very important class of catalysts. Enzymes are complicated proteins 1 hi \ work principally by four different routes: Table 12.4 Some examples of enzymes listed in the Brookhaven National Labs' protein I ii.ili.ise Oxidoreductases" Transferases' Hydrolases'' nadh NADH + Dimethylallyl-di'- Dimethylallyl Carboxyleste- A carboxylic peroxidase H202 => transferase diphosphate + rase (promo- ester 4- H20 {oxidizes NAD<+) + (transfers isopentenyl tes hydrolysis an alcohol NADH with 2H20 dimcthylallyl diphosphate => of ester + a carboxylie peroxides) groups) diphosphate + dimethylallyl-cis-isopen-lenyldiphos-phate linkages) anion Ferroxidasc 4Fe2+ + 4H+ + Glycoaldehyde Sedoheptulose 1,4-o-Glucan Hydrolysis of (oxidizes 02 => transferase 7-phosphate + glucanohydro- 1,4-glucosidic Iron) 4Fe3+ + (transfers D-glyceralde- lase (also linkages in 2H20 glucoaldeydes); also called transketolasc hyde 3-phos-phate => D-ribose 5-phosphate + D-xylulose 5-phosphate called-amylase) oligosaccharides and polyasac-charides Glucose P-D-Glucose Alanine amino- i.-Alanine + Interleukin 1- Release of oxidase + 02 => transferase 2-oxoglutaratc p-converting interleukin 1-0 (oxidizes D-glucono- (transfer amino =>• pyruvate + enzyme by specific glucose) 1,5-lactone + H202 groups from alanine) L-glutamate hydrolysis at 116-Asp-|-Ala-117 and 27-Asp-|-Gly-28 bonds "Promote oxidation reduction reactions. ''Promote transfer of functional groups. r Promote hydrolysis/cleavage reactions. J Promote addition of C02, H20 and NH3 to double bonds or formations of double bonds via elimination of C02, H20 or NH3. e Promote isomerization reactions. ^Promote bond formation; generally used to catalyze entothermic reactions requiring ATP. I vases'' Isomerasese L i gases' 1 nl.....ale dehydratase (dehydrates 1 nrhonates) 1 Urate dehydratase Pyruvate decarboxylase H2CO3 co2 - H20 Citrate cis- aconitate + H20 A 2-oxo acid an aldehyde + C02 Maleate isomerase (promotes cis-trans isomerization of Maleate) Cholestenol 8«-isomerase Man nose isomerase Maleate =>■ fumarate 5-a-Cholest-7-en-3-p*-ol => 5-a-cholest-8-en-3-P-ol i>-Mannose D-fructose Leucine-t-RNA ligase Pyruvate carboxylase Aspartate-ammonia ligase ATP + L-leucinc 1 t-RNA(leu) > AMP + diphosphate 1 i.-lcueyl-l-KNAlleu) atp + pyruvate | (HCOi) > adp h phoiphate i oxaloacctale atp + 1 -aspartate I nh3 => ami* 1 diphosphate I L-asparagine tmu líniu n >i......n i" < aiai r.ľ. 1 '»1 IIV/II W.......Ml n ,1 I I.....M 'v I 'M , M . (H)/ . En/ymcs bind It) llu- leaclanls in sin.li n way Ihal krv bonds in tlie teiKianls are stretched. Thai makes bonds easier to break, . Enzymes lower the energy of the transition State. The analysis In Chaptei 9 shows that lowering the transition slate energy enhances I ho rale ol reaction. . Enzymes work like metals to stabilize key intermediates. . When you have a bond-forming reaction, the enzymes push the reactants together, That also promotes reaction. Table 12.4 lists several different enzymes. Basically, enzymes are classified according to what they do rather than how they are constructed. For example, NADH peroxidase is an enzyme used to react NADH (nicotinamide adenine dinucleotide) with hydrogen peroxide and superoxide (H202") where NADH is a key molecule in the energy cycle of cells. In Table 12.4 we list "NADH peroxidase" as though NADH peroxidase were a single enzyme. In fact, NADH peroxidase represents a whole family of enzymes. For example, horseradish NADH peroxidase is slightly different from mammalian NADH peroxidase. Another important idea is that enzymes often work with "cofactors", which arc additional substances needed to get the enzyme to work. So, for example, FAD (flavine adenine dinucleotide) is a cofactor for NADH peroxidase. Table 12.4 lists six key classes of enzymes: • Oxidoreductases • Transferases . Hydrolases . Lyases • Isomerases • Ligases Oxidoreductases are enzymes that promote oxidation-reduction reactions. Transferases are enzymes that promote transfer of functional groups from a donor molecule to an acceptor molecule. Hydrolases are enzymes that promote hydrolysis reactions (i.e, scission of bonds with additions of water). Lyases are enzymes that promote elimination of C02, H20, and NH3 from organic molecules leading to the formation of double bonds, and the reverse reactions where water, C02, NH3, are added to double bonds. Isomerases are enzymes that promote unimolecular isomerization reactions. Ligases are enzymes that promote endothermic bond-formation processes, consuming ATP in the process. Enzymes are discussed in detail in elsewhere. See Faber (2000), Jencks (1987), Fersht (1999), or Sinnot (1997). The thing to remember for now is that there are many different enzymes, and they each perform a specific function. Another key point is that if you want to find an enzyme to do a specific reaction, you can look in a standard table and find out whether a suitable enzyme exists. I usually look things up in the protein database at which in 2000 was at http://www.rcsb.org/pdb/ or http://www.chem.qmw.ac.uk/iubmb/enzyme/ 12.2.4 Radical Initiators Radical initiators represent a fourth important class of catalysts. Recall that in many reactions, the slow step is the initiation step where radicals are formed. If one can find ■i wuy to liirtn llu- imltiitls inoiv easily. llie miř ol ivaelion wtll l>c cnlinmcil Rudieal Inltiatort are inolcciilcs th.n dcconiposc tuto nidicals vety easily, possibly in llie picsemc • il light. llie radicals llien miti.ilr Ihe icaclioii. I oi examplc, in (haplers 5 and III we considcrcd llie rcaclion C2H6 C2H4 + H2 i l • nu lind found that the slow step was the initiation step where the C-C bond broke. Well in Iodine iodine bond is much easier to break than a carbon-carbon bond. Consequently, ilie iodine can decompose at modest temperatures: X + I2 -► 21 + X I lien the iodine can react with ethane to start the reaction: I + CH3CH3 -► HI + CH2CH3 tl.' Ill 11.' 12) I here are many variations of this. For example, the carbon-carbon bonds in acetaldehydl lie much weaker than the carbon-carbon bonds in ethane. Consequently, acctaldchyde 1 1 useful initiator for ethane dehydration. Free-radical initiators are most important for polymerization reactions. Molecules like etliane are hard to decompose into radicals, and you need radicals or ions to start free radical polymerization. Consequently one adds a molecule that is easy to decompose, like benzoyl peroxide. The benzoyl peroxide decomposes into radicals: R-O-O-R-► 2RO. (12. Ml Then the radical reacts with the ethylene to start the polymerization process: RO. + CH2CH2-> /?OCH2CH2. (12,14) liee-radical initiators tend to give polymers with varying properties because ol Jinn transfer reactions, so they are less useful than the transition metal polymerization catalysts Still, the catalysts are used when one wants a soft material. Free-radical processes are also important to atmospheric chemistry. Chlorine atoms produced from photolysis of chlorocarbons catalyze the destruction of ozone via the process: CI + 03-► CIO + 02 (12.15) The CIO can then react via a number of processes to reduce the ozone layer. One particular reaction is CIO + O3 ==> CI + 202 (12.16) (Hhers reactions also occur. Table 12.5 gives several other examples of free-radical initiators. They are used extensively. li li 11' > I M Ii in it I Ii i III IM Ii IUI Nl ( Hl! 11 .Al Al ' l.ihli- I'.' !. Sonic i-x.impli'-. ol 10,11 lion-.....Ii.il.il in . .il.ily/ixl by In■•■ nulu iil'i iiikI similar species Reaction Initiator Real nun (alalysl Olefin Peroxides, 2S02 + 02 => SO, NO/NO; polymerization (Ph)3CC(Ph)3 (lead chamber process) Hydrocarbon Iodine, NO2, chlorine Ozone depletion CI dchydrogenation atoms Hydrocarbon [(CH3CH2)4N|[I|, — — oxidations [(CH3CH2)4N] [C6H5COO] Table 12.6 The rate of reaction (12.17) in several solvents3 Solvent Rate constant, liter/(molsecond) Gas phase Water Methol Methyl cyanide DMF *io-45 3.5 x irr3 3 x 10 6 0.13 2.5 "All measurements have been extrapolated to 25°C. 12.2.5 Solvents There is one other class of homogeneous catalysts that is usually discussed separately from all other types of homogeneous catalysts: solvents. Solvents can act just like catalysts. They speed up reactions and change selectivities. For example, Table 12.6 shows some data for the rate of the reaction CH3I + NaCI CH3C1 + NaT (12.17) Notice that the rate in DMF is a factor of 10'' larger than that in water, and 1045 larger than that in the gas phase. Clearly DMF is acting like a catalyst. In the literature, people seldom discuss solvents as catalysts, but solvents can act just like catalysts. Generally solvents speed up rates of ionic reactions by factors of 1030 or more. Increases in rates of nonionic reactions are more unusual. Solvents are less able to selectively catalyze reactions than are the other types of homogeneous catalysis. Still, many solvents show considerable catalytic activity. Therefore I consider solvents to be a class of homogeneous catalysts. 12.3 INTRODUCTION TO HETEROGENEOUS CATALYSTS At this point, we will be changing topics. So far we have been surveying homogeneous catalysts; however, now we move on and start to discuss heterogeneous catalysts. i inure 12.3 Pictures of some heten I mi mission.] ts. [From Wijngaarden and Westerterp (1991) With figure 12.3 shows pictures of some heterogeneous catalysts. Heterogeneous catalyltl ■ in be pellets, powders or other solids. I lelerogeneous catalysts are quite similar chemically to homogeneous catalysis The) lire substances that are added to a reacting mixture to speed up a reaction. However, while homogeneous catalysts are generally substances that dissolve into the reacting mixture, In krogeneous catalysts are solids that do not dissolve. Instead, reaction occurs al 01 net! I he interface between the solid and the reactant mixture. The fact that the reaction occurs at an interface means that reactants have to diflfuie 10 the interface before the reaction can occur. Consequently, heterogeneously catalyzed reactions are often a little slower than homogeneously catalyzed ones. Still, industrially, heterogeneous catalysts are almost always preferred over homo p-neous catalysts. Recall that with a homogeneous catalyst, the catalyst is added to the reacting phase. When the reaction is completed, either the catalyst must be separated from the products, or you end up losing the catalyst. Many catalysts contain expensive metals, such as rhodium. You would not want to throw rhodium away. Other catalysts contain acids; the acids are disposal problems. In contrast, a heterogeneous catalyst is a solid. The solid is easily separated from the reacting mixture. As a result, it is generally much easier lo use a heterogeneous catalyst rather than a homogeneous catalyst in a chemical process. Examples of heterogeneous catalysts include . Supported metals . Transition metal oxides and sulfides • Solid acids and bases . Immobilized enzymes and other polymer-bound species f W iin i 111 ii m m . iii in 11 m /m /m (m"i IM II n H M H Mi it J III Ml II III «11 NI I HIM AI AI v. I-. /Ol (lenerally, heterogeneous i atalysts are |usl like homogeneous i atalysts except lhal the active species are attached i" .1 solid. The solid allows the catalyst n> be separated from the reaction mixture quite easily. The easy separation generally makes heterogencoui catalysts preferred over homogeneous catalysis in industrial practice. In the next several sections we will list several examples of catalysts in each of thesi classes. The lists are not meant to be complete. However, we wanted to give the reada a picture of the wide variety of materials that show catalytic properties so lhal 1 lit- n-adi-i will have a general idea about what is available. 12.3.1 Supported Metal Catalysts The simplest heterogeneous catalyst is a supported metal catalyst. Figure 12.4 shows a diagram of a supported metal catalyst. The catalyst consists of a series of small platinum particles on an inert silica (SÍO2) support. Generally, the platinum provides most of llu-catalytic activity. However, the silica is there to both add mechanical strength to the catalyst, and to spread out the platinum over more surface area. Supported metals are fundamentally similar to the homogeneous transition metal catalysts discussed in Section 12.2.2. The catalyst contains an active metal that docs chemistry similar to that discussed in Section 12.2. However, the metal is attached to the support. The support spreads out the metal and holds the metal in place, so the catalyst is easy to separate from the reacting mixture. Supports tend to be high-surface-area materials: aluminum oxide, silicon dioxide, and activated carbon. Generally, most supported metal catalysts contain group VIII or group lb metals, although other transition metals and rare-earths also show catalytic activity. Table 12.7 shows a selection of the reactions commonly run on supported metal catalysts. Generally metals are used to hydrogenate and dchydrogenate hydrocarbons, to oxidize substances, and to generally catalyze hydrocarbon conversions. Supported metal catalysts are used in the catalytic converter in your car and your wood stove. People • • • • % m9 • m* m M •|!jf- .Mem,.— Platinum ♦ ^ • i all —Alumina Figure 12.4 A supported metal catalyst. i hIiIk \V I A •mini lion ill llui mm Ilium i iililly/iul hy •im|i|kmtoil im-l.il . Ml HI In m 11 \ ■ in n irbon li\iliii|'i'ii:ilinn. 11, hydrogenation I 1 > oxidation, total uxidation of i. .h.H arbons ii m .mi. > en (Oft Cu/znO ICO I 2NO => '('().. f N2 N • I lib => 2NH3 < 11, +02 2 ethylene oxide < Mil s i Krm (n hi II, IM, Ni CO I H2 =» hydrocarbons (Fischer Tropsch) Pt, Rh, Ru (catalytic converter) Fe, Ru, Rh Ag, Cu i .ll.llv'.l Fe, Rh It, I'd, Cu, Ni, Steam reforming for production Ni plus additives Fe, Rh, Ru of hydrogen Reforming (isoineri/.alion of I'l/kr/AI '< 11 oil) 2NH3 + 02 N205 + 3H20 I'l Alcohols + 02 => Ag, ( n aldehydes + H20, e.g., 2CH3OH + 02 => 2H2CO + H20 R-R' + H2=> Ni, Co, Rh. Ru RH + HR'(hydrogenolysis) ire Marling to sell supported metal catalysts to clean the air in your home. Most majoi industrial chemicals are made with a supported metal or a metal oxide catalyst. ('hapu-i I I will be devoted to a detailed discussion of supported metal catalysts. However, the ilmi)' i" remember for now is that virtually all reactions where you add or subtract an atom limn a molecule are run on a supported metal catalysts or a metal oxide catalyst. Supported metal catalysts work just like homogeneous transition metal catalysis Foi example, platinum catalyses the hydrogenation of ethylene via the same series of steps as Ihc Wilkinson catalyst: H2 + 2S C2H4 + S C2H4(ad) + H(ad) C2H5(ad) + H(ad) -> 2H(ad) -> C2H4(ad) ■> C2H5(ad) + S -* C2H6 + 2S (12.1«) (12.10) (12.20) (12.21) w Mere S is an empty site on the surface of the platinum and the (ad) denote species attached to the surface of the platinum. In this case the platinum is stabilizing hydrogen aloms and ethyl groups. The stabilization tremendously speeds up the rate of reaction as discussed in Section 1.2. 12.3.2 Special Reactions on Supported Metal Reactions (12.18)—(12.21) can occur on either a cluster or a supported metal catalyst. I lowever, there are a few reactions that occur only on supported metal catalysts. For example, the conversion of acetylene to benzene occurs readily on a palladium catalyst, but at the time this book was written, the reaction had not yet been seen on a cluster compound. INIMnlHIi III IN IIi Mill IK Mil Nl I MINI AI Al YM'i 703 11 if overall reaction is U Ml, > C.ll,. (12.22) Physically, in order for reaction (12.22) to occur, the catalyst needs to simultaneously coordinate three acetylene molecules. That is easy on a metal surface because there art many atoms on the surface to hold the acetylene. In contrast, the reaction is difficult on a single transition metal atom in solution. I expect that as people begin to make larger cluster compounds, they will observe reaction (12.22). However, at present, metal clustd compounds with multiple metal atoms are difficult to make. In contrast, it is easy to get a cluster of atoms on a supported metal catalyst. The mechanism of metal catalysis will be discussed in detail in Chapter 14. However the thing to remember for now is that like metal atoms, metal surfaces stabilize radical intermediates. In Chapter 4 we found that an increase in the intermediate can tremendously enhance the rate of a reaction. 12.3.3 Transition Metal Oxides, Nitrides, Sulfides, and Carbides In Table 12.7, we listed reactions that are commonly run on pure metals. However, there are a number of reactions that are instead run on transition metal oxides or sulfides they arc-listed in Table 12.8. I usually think of a transition metal oxide, nitride, sulfide, or carbide catalyst as being a transition metal catalyst that has been poisoned (i.e., slowed down) Table 12.8 A selection of the reactions catalyzed by transition metal oxides, nitrides, and sulfides Reaction Catalyst Reaction Catalyst 2 S02 + 02 => 2 S03 Hydrodcsulfurization CH3CH=CH2 + 02 = CH2=CHCHO + H20 4NH3 + 4NO + 02 =! 4N2 + 6H20 (selective catalytic reduction) CH3CH2(C6H,) + o2 => CH2=CH(C6H5) + H20 (styrene production) Aromatiztion, e.g. heptane toluene H2 or H2Q v2o5 CoS, MoS, WS (Bi203)x(Mo03)y (bismuth molybate) uranium antimonate V205,Ti02 FeO Cr203/Al203 CO + H20 => C02 + H2 (water-gas shift) 2(CH3)3COH => (CH3)3COC(CH3)3 + H20 2 CH3CH = CH2 + 302 + 2NH3 => 2CH2=CHC=N + 6H20 (aminoxidation) Benzene + 02 => maleic anhydride + water naphthylene + o2 => phthalic anhydride + water Selective oxidation of hydrocarbons Hydrodenitrogenation FeO, CuO, ZnO Ti02 (FeO)x(Sb203)y (V205)x(P04)y NiO, Fe203, V205, TiOz, CuO, Co3, 04, Mn02 NiS, MoS 11 Hiklltlon ol oxygen, nllioyen, < uibon. oi sullm The siunc types ol chcnusiiy on in on ili, iiKi.il nii'l.il OXide, 11M111 It- siiIIhIi I, in i .iilnilc catalyst as on the pine iiicliil llowevi-i. ill, miction is slowci on the nicliil nul.il o\i H,0 Ml of these steps are the reverse of the steps mentioned in Section 12.3.1. In the literature, it is unusual to find the mechanism written this way. The reason Is that in a metal oxide catalyst, there are two types of sites: (1) metal sites, which I will i all SM sites; and (2) oxygen sites, which I will call S0x sites. Ethylene normally binds in ,i metal site while hydrogen prefers to bind to an oxide site. As a result one can rewrite iIn- mechanism as: C2Hg + Sm + Sox c2H5(ad,M) + Sox C2H4(ad,M) -> C2H4(ad M) + H(ad,ox) -*■ Sm + C2H4 (12.27) (12.28) (12.29) where the notation(adM) is used to denote a species attached to a metal atom while, H(adi0x) denotes a hydrogen attached to a lattice oxygen. Notice that the H(ad,0x) is basically a hydroxyl group. In Chapter 5, we found that hydroxyls can react with hydrocarbons like ethane via the reaction IM I IC MIHI, I IHN II I III II lim Ii Nl I II I'l i -A I Al VI I'i 70B The II,........i .in ir.u I in the »ume w.i\ H(ac|.o„ +C2Hh - (II I 11.() I I 1 (12.311 where the □ represents a lattice site thai has heen depleled of oxygen. Reaction (12.31) depletes lattice oxygen, which is replaced via ().. adsorption: 02 + 2 □-► S0, • So, (12.32) This example illustrates the one major difference between reactions on transition metals and on transition metal oxides. On transition metals, all of the reacting species are adsorbed on the metal. With metal oxides, the lattice oxygen is a major player in the reaction. People often think of reactions on metal oxides as oxidation reduction reactions, where metal is reduced, then reoxidized. Generally, there is more going on than jusi oxidation-reduction. However, oxidation-reduction is a key part of the catalytic cycle. 12.3.4 Solid Acids And Bases In Section 12.3.2, we discussed transition metal oxides. Transition metal oxides have weak metal-oxygen bonds, and so oxygen can be easily exchanged. By comparison, the metal-oxygen bond in alumina is so strong that the bond is not easily broken. As a result, alumina is not a good oxidation catalyst. Still, there is some important chemistry that occurs on metal oxides with strong metal-oxygen bonds. The objective of this section is to give an overview of the key chemistry. Generally, metal oxides with strong metal-oxygen bonds are acids or bases. For example, you know that H2S04 is a strong acid. Well, H2A103 is a strong acid, too. A1C13 and FeCI} are strong Lewis acids. There are strong Lewis acid sites on A1203. Silica (Si02) can stabilize the acid sites, so silica aluminas are quite effective solid acid catalysts. There are also special acids, like (SbF5) and sulfated zirconia, which are much stronger acids than H2S04. These compounds are also effective acid catalysts. Na20 and MgO are strong bases. They are used to catalyze basic reactions. Table (12.9) shows several other examples. All of the materials in Table (12.9) are strong acids or bases. Industrially, most acid-catalyzed reactions are run on a special class of solid acids called zeolites. Zeolites are highly porous silica/aluminas. The pore structure is very uniform. Generally, the structures consist of cages that are stacked one on top of another. Figure 12.5 shows one of the cages in a material called faugasite. The uniform pores Table 12.9 Some common solid acids and bases Material Type Material Type Silica/alumina Solid acid Mordenite Zeolite Alumina Solid acid ZSM-5 Zeolite Y-zeolite faugasite Zeolite VFI Large-pore zeolite Sodalite Zeolite Offretite Zeolite HF-SbF5 Superacid HSO3F Superacid H2[Ti604(S04)4(OEt)io] Superacid Sulfated zirconia Superacid MgO Solid base Na20 Base Figure 12.5 A diagram of the pore structure in faugasite. ..Mow people to design zeolites with special mass transfer properties thai enhance the iclcctivities of reactions. This idea will be discussed in greater detail later in this cliapici Industrially, most of the acid-catalyzed reactions listed in Table 12.2 are actuallj.....I on solid acids. The one major exception is butane alkylation, which is still commonl) t utalyzed with hydrofluoric acid. 12.3.5 Polymer-Bound Catalysts All of the heterogeneous catalysts we have discussed so far in this chapter were heterogeneous by their nature. The catalysts were solids. Solids have the advantage thai iliey are easy to separate at the end of a reaction, so they are generally cheapei to use than are homogeneous catalysts. In the 1970s and 1980s there was a lot of effotl devoted toward starting with a homogeneous catalyst, and anchoring the catalyst to a polymer to produce a material that is catalytically active and easy to separate from the products. For example, immobilized enzymes, that is, enzymes bound to polymer beads, were tried as reusable enzyme catalysts. People also tried bonding transition metal sails to polymers and using the resultant materials as catalysts. Acid groups were also bound 10 polymers. Acid-bound materials are currently being used as catalysts. When this hook was being written, none of the other materials had made it into industrial practice. Still, people continue to discuss polymer bound catalysts in the literature, and so I decided to mention them. 12.3.6 Photocatalysts There is one other class of catalysts that I did not have a place for anywhere else, so I decided to list separately: photocatalysts. Photocatalysts are catalysts that are inactive under normal conditions, but when the catalysts are irradiated with UV light, the catalysis become activated. At present, there are only three important photocatalysts: illNIMAI (iVI I Ml WOI CAIAI YIICACIION 707 . ho. and IISrl >, . Fe203 . Mercury vapor Ti02 is commonly used as a catalyst for destruction of hydrocarbon wastes in aqueous environments. The main mode of catalytic action is for water to react with the O2 in the titania surface to yield 20H (12.33) CT2 + H.O Ti02 is a semiconductor. When light hits the semiconductor, the light interacts with the semiconductor to produce free electrons and holes (positive charges). The charges can promote reaction. The holes can then react with the OH s to yield OH radicals: h+ + OH OH. where h+ is a hole. The OH radicals then oxidize hydrocarbons: OH. + C6H6 -> C6H5 + H20 Electrons react with 02 to regenerate the O2 lattice: 4e~ + 02 -> 20" (12.34) (12.35) (12.36) which can also oxidize species. Fe203 works similarly to Ti02 except that Fe2Oj is much less catalytically active than Ti02. Mercury vapor works via a different mechanism. Mercury is relatively unreactive in its ground state, but once the mercury is promoted into the 3Pi state, the mercury becomes quite reactive. For example, the excited mercury can extract hydrogen from a hydrocarbon: Hg(3P,) + H-/?-► HgH + R» (12.37) Then the mercury falls back to its ground ('So) state, releasing the hydrogen: HgH-►Hgt'SoJ + H (12.38) The result is that mercury vapor can initiate gas-phase radical chemistry and act much like the radical catalysts described earlier in this chapter. Industrially, people rarely use mercury anymore because mercury is so toxic. However, there are many papers on the topic. 12.3.7 Summary At this point, then, we have listed many different types of catalysts and briefly mentioned how they work. When I teach this course I find that students are often overwhelmed by all of the variations; and in reality there are reasons to say that the variations are overwhelming. In fact, that has been the main point of the sections so far in this chapter: there are many variations. Many substances can act as catalysts. V. I said euiliei, I teiommeml Huh the ivmlci imuioii/e rvoiylhuiH In i . I.'.' I.") I In- in.ii.ii.il in these tables will be necessary background lot Ihc thing! that follow, 12.4 GENERAL OVERVIEW OF CATALYTIC ACTION \l this point, we will be changing topics. So far we have been discussing what catalysis in like Now, we will start to discuss how catalysts work. i .il.ilysls work by changing the local environment around the reactants. The change in Ihc local environment stabilizes intermediates and modifies the lours between the n ii i.nils. These changes can be used to promote a desired reaction. I he idea that rales of reactions can vary with the local environment around the rem hint* ... . hack to the early days of kinetics. In 1817, sir Humphry Davy found that In .....Id in. \iiit explosions in coal mines if he surrounded the candles used to illuminate the mined nil a platinum shield. A few years later, Michael Faraday examined Ihc liiudatiu null processes in Davy's lamp. In 1834 Faraday proposed that the platinum was catalyzing the termination reactions in the flame by acting in two key ways. The platinum was holding iln icactants in close proximity so that they could react. The platinum was also modi I \ ll)| ihc forces between the reactants to stimulate the reaction. A lew years later, Jons Berzelius (I836a,b) did extensive studies on catalysts At dial time people did not know about molecules. However, Berzelius proposed thai the . .iialyst changed the rate by modifying the forces between the reactants to si......late relictions. Another idea in the literature came from the work of Jean-Uaplisle IVinn i I'M')), who suggested that catalysts were able to transfer energy into the readmits and thereby overcome the energy requirements needed to activate molecules, finally. Paul Sabatier (1913) suggested that the catalysts stabilized intermediates, thereby promoting ■ reaction. Over the years there have been many arguments about whether Faraday. Berzeliui Sabatier, or Perrin were correct. However, we now know that they were all coins I i iialysts can work in a variety of different ways, and if you look at different reactions, sou can find many different modes of catalyst action. When I teach my catalysis comse I give the following list of ways a catalyst can be designed to work: • Catalysts can be designed to help initiate reactions. . Catalysts can be designed to stabilize the intermediates of a reaction. . Catalysts can be designed to hold the reactants in close proximity. . Catalysts can be designed to hold the reactants in the right configuration to react. . Catalysts can be designed to block side reactions. . Catalysts can be designed to sequentially stretch bonds and otherwise make bondl easier to break. . Catalysts can be designed to donate and accept electrons. . Catalysts can be designed to act as efficient means for energy transfer. II is also important to realize that . One needs a catalytic cycle to get reactions to happen. . Mass transfer limitations are more important when a catalyst is present. I ft I ft I . I ■ -I I hl I H .III I ' n ' INI IIA 11 iii Al III IN'I /I ľ I I here are many Ideal here so ll Is useful to consldei an example < onsldei the rea< i..... li i r.i 1 III:. (12 19] In chapter 5 we found that the reaction goes via the following mechanism: Br2-► 2Br Br + H2 H + Br2 2 Br -> HBr + H ■+ HBr + Br Br, (12.40) In actual practice, the reaction goes through a catalytic cycle, as will be discussed latei in this chapter. However, for now we will assume that the reaction goes only once, and try to see how a catalyst can work. Figure 12.6 shows how the free energy of the system changes during the reaction in the gas phase. The system goes uphill during the initiation step, and more uphill during the first reaction step. Then everything is downhill to products. The first way a catalyst can work is to help initiate the reaction by reducing the initial steep rise in energy as shown in Figure 12.6b. According to the Polanyi relationship, when you reduce the initial rise, you will decrease the barriers to reaction. The next way that a catalyst works is to stabilize the intermediates as shown in Figure 12.6c. The idea is to bind to the reactants just enough that the reaction never has to go uphill. That way the overall barrier for reaction is reduced. 50 E o S CD C E o -Q "o Ô E 5? o 50 50 it 0 -50 / \h+hb ŕ~ \ (a) r y2Br2 / \ +Br2 1 Br+2HBr '/4Br2+2HBr Gas phase 50 Reaction progress (c) i/2Br2+2HBr Stabilize intermediates 50 50 +H2 (b) A H+HBr ' \ +Br2 V*r2 j A Br \ Br+2HBr 1__J i/sBr2+2HBr Initiate reaction Reaction progress 50 +H (d) 54Brz íl Br +Br, 1 H+HBr Br+2HBr Modify intrinsic barriers '/2Br2+2HBr Figure 12.6 reaction. Reaction progress Reaction progress An illustration of some of the ways a catalyst can affect the free-energy changes during a Hu- third in.hm wny u catalyst wink1, I* In adjust the iiilnnsu lumieis ul each slujte .•I I lit* leaition .is nulu atcd in I ifiit i- I.'íhI (ieneiiilly you wunl In lowci iho minusu li.iiiicis for dcsiicd renclions ami laisc the minusu haulers lot iiiulesualile reactions. One i mi lower I he intrinsic hanuis hy changing eilliei the entropy or the enlhalpy ol activation ion i an lower the entropy ol activation l»v holding the reactants close logclhci and in Ihe right configuration to read. You can lower the enthalpy of activation hy either slicuhuif I.....ds. thereby making them easier to break, or by adjusting the charge to moderate I he ľ.mil repulsions between the reactants. I lirsc ideas will be discussed in delail in the next several sections, ll would be unusual I'ot ,i catalyst to do all of these things. The very best catalysts iniglii help initiate the reaction, stabilize the reactants, hold the reactants in close proximity and in the light .....figuration to react, and possibly block side reactions. However, few catalysis wink lhal cľliciently. I view the list above as a list of things lhat you would ideally wain the ■ ii.ilvsl to do. However, few catalysts do all of those things. In the nexl several SCI tlona will choose examples to illustrate each of these effects. Another point is lhat much of the material from here on will be qualitative. People know in a qualitative way how catalytic reactions work. However, very few ol the Ideal have been quantified enough that one can use them to make practical calculations. I used the word "catalyst" in the last few paragraphs, but I do not want to restrii i the discussion to conventional catalysts. People now know, for example, that a solvent i an Bi I like a catalyst. Solvents can modify the forces between reactants, stabilize intermediates, hold the reactants in close proximity, and act as efficient media for energy transfer. ( Khet • iindensed phases can also act as catalysts. In the literature it has become common to treat the effects of solvents separately from the effects of catalysts. However, we now know lhat the principles of catalytic action and the principles of solvation are almost the game Therefore, I find it convenient to treat catalysts and solvents as having l'uiidaiiieniall\ dusáme types of influence on rates. In the next several sections we will discuss how a solvent or a catalyst can modify the rate of a reaction. We will concentrate on the general principles that apply to a large number of systems. Specific information about different catalyst systems will be given in ( hapters 13 and 14. 12.5 CATALYSTS CAN BE DESIGNED TO INITIATE REACTIONS The basis for all of our analysis will be a simple observation, discussed in Masel (IW(>) Catalysts seldom change the mechanisms of reactions; catalysts change only the initiation processes and the rates and selectivities of reactions. For example, McKenney et al. (I (>(\) | found that at 200 K the rate of the reaction C2H6 C2H4 + H2 (12.41) goes up by a factor of about 109 if one adds N02 to the reacting mixture. The mechanism of reaction (12.41) does not change substantially in the presence of the catalyst. The only major change is in the initiation step. Recall from Chapter 5 that reaction (12.41) goes via the following mechanism: Initiation C2H6 -> 2CH, (12.42) I IW in1mi m mm iii in ii i i /vi ai , I i.AIAI Y*nO«d W WNHlNI li lii-ilAIMII/l IN 11 MMI I HAII '■ 711 I I .in.In Clli I < Ml,, (Ml, I (II (12 43) Propagation Termination C2H4 + H C2H5 H + C2H6->C,H5+H2 2CH, -► C2H„ 2C2H5 -> C4H10 CH3 + C2Hj -> CjHg ( 12,11) (12.4.5) (12.46) (12.47) (12.48) If fact, N02 changes the reaction in a very simple way. Recall that N02 is paramagnetic. It has an unpaired electron. The presence of the unpaired electron makes N02 a "stable radical". N02 can do radical chemistry even though N02 is a stable species. Now, let's use the material in Chapter 5 to guess how the presence of N02 would affect the rate. Let's consider the reaction N02 + C2H6 products (12.49) According to the analysis in Chapter 5, when an N02 molecule collides with ethane, the main reaction will be a hydrogen transfer process: N02 + C2H6 -> C2H5 + HN02 (12.50) Once one forms the C2H5 the C2H5 can react via reactions (12.44) and (12.45). Reaction (12.50) is about 20 kcal/mol endothermic, and according to the analysis in Chapter 5, it should have an intrinsic barrier of 14 kcal/mol. If one plugs the intrinsic barrier into the Polanyi relationship, one finds that reaction (12.50) should have an activation barrier of about 25 kcal/mol. By comparison, reaction (12.42) has an activation barrier of about 90 kcal/mol. The result is that reaction (12.50) is usually much faster than reaction (12.42). The increase in the initiation process means that reaction (12.41) is much faster in the presence of N02 than in the absence of N02. This example illustrates a key point: Catalysts can initiate reactions. The mechanisms are similar to the mechanisms without a catalyst, but the initiation process is much faster with the catalyst. The increase in the initiation rate produces a tremendous increase in the overall rate of reaction. One can quantify the effect using analysis similar to that in Section 12.7. In the case above, the presence of N02 increases the rate by a factor of 109 at 750 K. Table 12.10 shows several other examples where catalysts initiate reactions. Notice that a wide range of reactions can be initiated by catalysts. Not all catalysts work this way, of course. However, catalysts can be designed to initiate chemical reactions. i >l>!•■ VJ 1(1 Siiinti <>»iini|il«i>. nl mm limn, liilllnlixl l>y catnlyntM I', hi ilon .li ml (ii| \ ii> I 'II t('Oil > (Hi I CO I thvlenc > polyethylene III Hi.. > 2HBr Propylene polypropylene ' II (III => C2H4 + H20 !()., > 302 (lítaly!) NO. I. .AOOA Metallic platinum Ti+ CI Mechanism ol Initiation N(l. i (II,(II, • UNO. I CH3I M- X + h -* 21 + X I I (II,(Oil • III I (II,CO. MX)A' . 2 AO. AO. I (If. C,,II,()CII..CII... I nClliCII. polyethylene CH2 -> ROCH2CIL. Br2 + 2S -> 2Br.ad T|+ + propylene -»■ CH3CHTÍCH5 1 CH3CHTiCH2+ + nC,H,, => polypropylene C2H5OH + H+ -» IC.IIsOII I fC2H5OH3]+ -> [C2H5|1 I II.O [C2H5]+^C2H4 + H o3 + CI -» o2 + CIO 12.6 CATALYSTS CAN BE DESIGNED TO STABILIZE INTERMEDIATES Another way that catalysts can change rates of reaction is by stabilizing intermediates. In < hapter 2, we noted that at 200°C the rate of the reaction H2 + Br2 ==> 2HBr (12.51) roes up by a factor of 108 in the presence of a platinum catalyst. Well, the mechanism <>i reaction (12.51) does not change dramatically in the gas phase and on the catalyst. H ll lust that the rates of many of the elementary reactions are much slower in the gas phase than on the catalyst. Platinum is able to catalyze reaction (12.51) mainly because the platinum stabilize! die intermediates of the reaction. In Chapter 5 we found that in the gas phase reaction (12.51) goes by the following mechanism: X + Br2 —1—> 2Br + X Br + H2 —HBr + H H + Br2 —U- HBr + Br X + 2Br-> Br2 + X H + HBr —> H2 + Br X + Br + H —HBr + X (12.52) oil) IO Ml AI1II l/l INIIMMIIMAir. III Chuptci V wc round lluil dining icmlion (12.52), most lllii is being produced ill .1 catalytli cycle First .1 bromine atom reacts wnil .1 11 to produce an ilUi and an 11 atom. The hydrogen atom then reacts with the Iti • to generate more niti and regeneratl (he bromine atom. Noie that in Section 4.3 we found that the rate ol reaction (12.52) is directly proportional to the bromine concentration in the gas phase, so if we could find .1 way to increase the bromine atom concentration, we would increase the rale. Now, it should not take much to enhance the rate. One can show that at 2()()"C, the bromine atom concentration should be about 10~17 mol/liter. Bromine atoms are the active centers during reaction (12.52), and with only 10 11 mol/liter of bromine atoms, the rale is limited by a shortage of bromine atoms. One could imagine trying to find a way to increase the concentration of bromine atoms using a solvent. After all, we know that many different molecules will dissociate in solution. If we could find a solvent that dissociated the Br2 to yield bromine atoms, we would increase the rate of reaction. I do not know of any simple solvent that will dissociate Br2. However, Br2 docs dissociate when Br2 adsorbs on platinum to yield chemisorbed bromine atoms. Physically, the platinum is acting just like a solvent when it dissociates the Br2. The free electrons in the platinum form a cage around the bromine atoms in exactly the same way that a solvent forms a cage around an ion in solution. The solvent cage is different in a metal than in a liquid because electrons behave differently than solvent molecules. Still, the process is fundamentally just like solvation. The electrons in the platinum solvate the bromine atoms, so that bromine atoms are stable on the platinum surface. Just to quantify things, at saturation, the bromine atom concentration in the adsorbed layer is about 50 mol/liter. By comparison under similar conditions, the bromine atom concentration is about 10~17 mol/liter in the gas phase. This is a general finding. The first key role of any catalyst is to increase the concentration of the active species. Increases of 20 orders of magnitude in the intermediate concentration are typical. That has a tremendous influence on the rate. In the literature it is common to view the stabilization of intermediates in terms of the changes in the enthalpy of the system. To orient the reader, Figure 12.7 shows the enthalpy changes during the gas-phase reaction H2 + Br2 =>■ 2HBr. The curve is very similar to that in Figure 12.6 except that the vertical axis is enthalpy, not free energy. Reaction progress Figure 12.7 The enthalpy changes during the gas-phase reaction H2 + Br2 => 2HBr. The enthalpy is measured as the enthalpy per mole of bromine atoms produced in the initiation step. I in using on llie llgllic. wi m 1 ill ii In .I llit- III. dissociates In 11IIMI Immune iiloins I bill insls .'<> / I kt.nl/tiiol I Inn 11 it' Ii.....Hue atoms rend with II. In produce liydiogcn ...... mil mil ih,it it-,11 turn is iiiioiiu'i (1 x kcul/iiioi cntloihcrmlc, Then there is a cycle n. 11 .1 Mi., reads Willi .1 liMlmgi'ii 11I0111 lo pioducc 111 it anil a bromine atom Thai .....1 it hi is 44 kcal/mol exothermic. Then the bromine atom reads with II.. lo yield lllli 111.1 ill. Iiviliiigeii atom. That icaclion is (1 X kcal/mol endolhcrmic. I In cycle repeals several limes, liach lime the system goes through the cycle, Iwo .....1.1 iilcs ol 11 Hi are produced. Consequently, the enthalpy of the system goes down by IWlt'C the heal of formation of HHr or 37.16 kcal/mol. We indicated two cycles 111 ilic ...... hul the system actually goes through the cycle hundreds of times. Fvcnlually, die n .11 Hon terminates. In the literature people often redraw Figure 12.7, assuming thai the reaciiou Ic.......lit 1 1111 1 completing one cycle. The diagram assuming that the reaction te.......tttei sftei ......plclmg one cycle is shown in Figure 12.8. The key feature of Figuie I.' X r. 1l1.1i ii.....it lion needs to go uphill twice for the reaction to proceed. That makes the r< 111 1..... nilht'i slow. I Igure 12.9 shows how the enthalpy changes that occur during the reaction v.ut il ton run Ihc reaction on a platinum surface. The platinum binds to Ihc intermediates ol the II 11 1..... The enthalpy of formation of the intermediate is lowered by the strength ol dn id 1 lull- surface bond. This stabilization of intermediates is a key process on plal........ Notice that on platinum all of the steps in the mechanism except the termination step lowci the enthalpy of the system. There is only one uphill step, and its barrier is sin,ill I hul makes the reaction H2 + Br2 => 2HBr very rapid on Pt(l 11). Most catalysts speed up reactions by binding to some key intermediate and thercbj 1 ibilizing the intermediate. That increases the intermediate concentration, and thereb) e o 5 E Br 50 t+Br, 1/2Br2 y2Br2+2HBr Reaction progress ^*i^n H„ + Bro => 2HBr assuming that tlv Rgure 12.8 The enthalpy changes during the gas-phase reaction H2 + Br2 + reaction terminates after one cycle. Reaction progress Figure 12.9 The enthalpy changes during the Rideal-Eley surface reaction H2 + Br2 => 2HBr on Pt(111) assuming that the reaction terminates after one cycle. increases the rate. One has to design the catalyst to selectively bind to the desired intermediate, and that is not always possible. However, most catalysts do increase the concentration of key intermediates. 12.6.1 Stabilization of Multiply Bound Intermediates An alternative is for the catalyst to open up a new reaction pathway that would not be seen in the gas phase. In Section 12.6, we stated that catalysts seldom change the mechanism of a reaction. However, there are exceptions. The exceptions occur because the primary mechanisms of a reaction can vary with conditions. Catalysts can also stabilize intermediates that would be present only at very low concentrations in the gas phase. As a result, a minor reaction pathway in the gas phase can become a major reaction pathway in Ihe presence of a catalyst. Generally in the gas phase, one sees only monoradicals, that is, radicals with one dangling bond. However, on a solid catalyst one can see di- or triradicals, namely, species with multiple bonds to the catalyst. The ability to form multiple bonds allows reactions to occur that would not occur at reasonable rates in the gas phase. For example, consider a simple reaction: N2 + 3H> ■ 2NH, (12.53) 1.....iin.11 Propagation \ 1 11. .Ml I X 1. 1 mination II 1 N.i —► NH + N N I II. . Nil + H NH + H2 -► NH2 H2 + NH2-> NH3 + H X + 2H-► H2 + X 1 1 ■ i 11 (12.55) (12.56) (12.57) (I2.57u) (12.58) However, according to data in the CRC, reaction (12.55) is about 150 kcal/mol . ndolhermic. Consequently, in the gas phase, reaction (12.55) is extremely slow 11 in) 11 1 onable conditions. I verything changes in the presence of an iron catalyst. On iron, the nitrogen readll) h 111 i;iies via the reaction N2 +6S-> 2N(ad) 1 I ' 39) 1 Inc can then hydrogenate the adsorbed nitrogen via the following reactions: N(ad) + H(ad)-> NH(ad) + S (12.60) NH(ad) + H(ad)->NH2(ad) + S (12.61) NH2(ad) + H(ad)-► NH3(ad) + S (12.62) Notice that when reaction (12.59) occurs, six nitrogen-surface bonds form. According in Table 6.5, on iron, AHr for forming a single nitrogen surface bond is —14 kcal/mol lluielbre, the heat of reaction of reaction (12.59) is (—14) x 6 =-84 kcal/mol (exothermic). Consequently, there is no thermodynamic barrier to reaction (12.59). (Ion* i|iien(ly, reaction (12.59) can occur on an iron catalyst even though reaction (I2.51)) 0C( m negligibly slowly in the gas phase. One can use the data in Table 6.4 to show that reactions (12.60)—(12.62) are c« li 'I kcal/mol endothermic on iron. If we substitute the 21 kcal/mol into the Polanyl relationship, we iind that each reaction should have an activation barrier ol .ilu.ui .'N kcal/mol. Such reactions are quite feasible at reasonable temperatures, which is why one uses an iron catalyst. One does not see a similar reaction in the gas phase since according to dala in the < !Rl handbook, in the absence of a catalyst, reaction (12.59) is 225 kcal/mol endothermic. This example illustrates an important point. Nitrogen radicals with three unpaired bonds are fairly unstable. However, when nitrogen adsorbs onto an iron catalyst, the Iron can form multiple bonds to the nitrogen. The possibility of forming multiple bonds allows reaction (12.53) to occur. One can generalize these results to many other situations. One can design a catalyst that stabilizes covalently bonded di- and triradicals. That allows the catalytic reactions to i AIAI VMI'H «11 HI III 'III INI I) III III Allll l/l IN 11 IIMI I HA 11 0 717 go vín n reaction piiiliwiiy III.il is mil available m ihc gas phase Generally, the abiliiy to form multiply bound species allows a catalysl to produce species that would not be produced at reasonable rates in the gas phase. 12.6.2 Stabilization of Ionic Intermediates Another way that a catalyst can work is to stabilize ionic intermediates, thereby allowing ionic pathways to occur. Consider a simple isomerization reaction: /?HC=C/?H RRC=CHH (12.631 It is hard to find a feasible mechanism for reaction (12.63) in the gas phase. One possibility is for the R group to break off: X+ /?HC=C/?H /?HC=CH +R. + X Then there could be an addition reaction: R. + RUC=CRH + X-► R2HCCRH. + X Then a ^-hydrogen elimination: /?2HCCH. + X-► R2C=CH + H. + X (12.64) (12.65) (12.66) Reaction (12.64) is 120 kcal/mol endothermic. Further, this mechanism does not have a catalytic cycle. Consequently, according to the results in Section 5.4, the mechanism will have a very slow rate. On the other hand, if one starts with a proton, H+ in an acid catalyst, it is easy to find a catalytic cycle. First the proton reacts with the olefin to yield a carbocation: RHC=CRH + H+-► [/?HC=CRHH]+ (12.67) Then the carbocation isomeriz.es: [/?HC=C«HH]+ -► [RRHC=CHH]+ (12.68) Then the new ion loses a proton to form the products: [/?/?HC=CHH]+-► RRC=CHH + H+ (12.69) Reactions (12.67)—(12.69) have barriers of 20 kcal/mol or less. Notice that the addition of the acid catalyst has allowed the reaction to occur via an ionic species. None of the reactions have large barriers. The result is that the reaction is quite facile (speedy). This example shows that an acid catalyst can facilitate reactions by helping to create ionic intermediates. Again, one does have to design the catalyst properly to stabilize the ionic intermediates. Not all catalysts will work. Still, one can design solid acid catalysis that stabilize ionic intermediates and thereby catalyze reactions like reaction (12.63). 11.6.3 The Effect of Intermedlnte Stabilization on Rate* — Oin Wo Have Too Mm Ii ill .1 (■.....I 11.....i ' línie is a subtle point in all ol this it is possible to stabilize an intermediate so much ilhil the intermediate becomes unieactive. Iheielbre, one needs to be careful to choose ili. i .ilalysl carefully. in tins section, we quantify how much rates go up in the presence "i a catalyil and In.w that one can overstabili/e an intermediate and thereby decrease the rale. We will go Inn k to the example in Section 12.6, namely H, + Br2 2HBr i I 1 /m iiiul derive an expression for the reaction. To put the discussion in perspective, in Sci lion 12.6, we found that the concentration of bromine atoms increases by a factOI "I Hi" in the presence of a platinum catalyst. Yet, the data in Table 12.1 show lliat the i.iic mi iiiscs by a factor of only 108. Other catalysts show much smaller rale enhaiu innni even though the other catalysts increase the bromine atom by a factor of more ili.m III" I In ii lore, il appears that there is something else going on that prevents a Id" increase in Ihc intermediate concentration from producing a 1017 increase in rate. In the remainder of this section, and in the next section, we will try to explain why Ihe i.ite does not show a 1017 increase . To start off, it is useful to recall thai a stead) Itati treatment of mechanism (12.52) shows that the rate of HBr formation via mechanism (12.52) is rHBr = 2k2[H2][Br] (12.71) v here rMn, is the rate of HBr formation; [H2] and [Br] are the concentrations of hydrogen molecules and bromine atoms, respectively; and k2 is the rate constant for the second reaction in mechanism (12.52). Therefore, if everything were equal, and k. did nOl i linage, one would expect the rate to increase by a factor of 10'7 when the bromine aiom concentration goes up by a factor of 1017. However, experimentally, the rate goei up by a factor of only 108. There are several reasons why the rate changes by a factor of IO8. There are lOrnl mass transfer limitations, and there is a second reaction pathway involving a direct reai til in between adsorbed hydrogen atoms and adsorbed bromine's. However, the key effet t II that when you stabilize the bromine atoms on the platinum surface, the bromine atomi become less reactive so k2 goes down. This is also a general effect. Radicals are very reactive in the gas phase. You can stabilize the radicals by solvating them on a metal surface. However, then the radical becomes less reactive. In the next section, we will quantify this effect. Our main tool will be the I'olnnvi relationship, which we derived in Chapter 10: E° + yPAHr I I .' / 1 i where Ea is the activation barrier, E° is the intrinsic barrier to reaction, AHr is the In«1 of reaction, and y> is the transfer coefficient. According to equation (12.72), when you change the heat of reaction, you also change the activation barrier of the reaction. Next, let's work out the implications of the Polanyi relationship for the reaction: Br + H2 HBr + H (12.73) I II I (IVA I II IN I II 'lAimill niNi.irii riw Winn sun sl.ihili/c Hi. Ihuiiiiih' .iIchiis, 11 it- lire energy ul the hiiinuni- atoms gut's down, The enthalpy ol the bromine ulomi goes down, loo, which means thai the All, ul reaction (12.73) becomes more positive. According lo equation (12.72), when All, becomes more positive, the activation barrier Tor reaction (12.73) will increase Consequently, k.. will go down. Therefore, the implication of equation (12.72) is that when we stabilize the bromine atoms, the rate constants for reaction (12.73) will always go down. Again, this is also a general effect. Whenever you stabilize a reactive intermediate, you increase the concentration of the intermediate. That generally increases the rate. However, there is always the secondary effect that the intermediate becomes less reactive. The loss of reactivity of the intermediate tends to partially counteract the effects of the increased stability of the intermediate, so the rate does not increase as much as the intermediate concentration increases. Interestingly, there are some cases where one stabilizes the intermediate so much that the reaction slows down. For example, Sachtler and Fahrenfort (1958) and Fahrenfort el al. (1960) examined the decomposition of formic acid on a number of catalysts. Additional data is in Sachtler (1960). They found that the main reaction is HCOOH H2 + C02 (12.74) Sachtler and Fahrenfort (1958) suggested that reaction (12.74) occurred via a very simple mechanism: The formic acid first adsorbs to form a formate intermediate. Then the formate intermediate decomposes to yield C02 and H2. HCOOH H(ad) + HCOO.7, -> HCOO-d) + Had C02 + H2 (12.75) Figure 12.10 shows how the rate of formic acid decomposition changes as one changes the binding energy of the formate intermediate. Notice that the rate increases reaches a maximum, and then declines. People call plots with a maximum like those in Figure 12 10 volcano plots. Most catalytic systems follow volcano plots. o CD cd q- E 350 400 450 500 550 600 50 1 1-1 Pt" lr" 1 Ru ■ i i ■ / ■ / Rh Cu*\ A*— \CoB nN| _Fe -X M— 4 An —L- 1 1 i i i 60 70 80 90 100 110 120 Heat of formation of formate Figure 12.10 The rate of formic acid decomposition changes as a function of the binding energy of the formate intermediate. ' I he idea lli.il i.ilc. mich u iiiinlin..... was discovered by Siibullci (l')l.l), Snbnliei •li|iycslccl that all reactions will lulhiw a volcano plot Allei all. an increase in die iiiiciinedi.ite Concentration will usually inciease the rale Still, Sabalier noted thai you do not want the intermediate lo be so stable thai you produce the intermediate and ......he product. Therefore, Snbnliei proposed that volcano plots would be a univeis.il phenomenon. Sllhatiei also proposed what is now called I lie principle of Sabalier I Ii. best catalysts are substances that bind the rcactants strongly, but not loo strongl) II you have a weakly bound intermediate, increasing the binding energy will IncfMM ii.....lermediate concentration, thus increasing the rate. On the other hand, il you base i itrongly bound intermediate, the intermediate concentration is already high. In iui h I • . increasing the binding energy of the intermediate docs not increase the intermediate .■in eulration substantially. However, k2 decreases. According to equation (I2.7I), if k,> ill i icascs. and the intermediate concentration does not increase by a comparable amoiinl iii. rate should decrease. In actual practice, however, Sabatier's principle is not quite a universal phenomenon iii. n are lew reactions where one observes a leveling off rather than a decline ol the i.iic when you stabilize the intermediates too much. However, this is the exception. Most italytic reactions follow Sabatier's principle. i. 1 DERIVATION OF SABATIER'S PRINCIPLE Next, we want to show how the maxima arises in detail. Our approach will be to work "in an equation for the rate, and use the equation to see if there is a maximum, Let's go bill k lo reaction (12.51), HBr formation, and assume that we are running the reaction in hi adsorbed layer. There are two reaction pathways for HBr formation: a Kidcal Elej mechanism Br2 + 2S-► 2Brad Brad+H2 -> HBr + Had Had+Br2-► HBr + Br (12.76) ami a Langmuir-Hinshelwood mechanism H2 + 2S -Br2 + 2S -Had + Brad 2Had 2Brad -> HBr (12.77) Where the names Rideal-Eley and Langmuir-Hinshel were discussed in Chapter 5 I l,c Rideal-Eley mechanism is very similar to the gas-phase reaction, while the I angmuir-Hinshelwood reaction usually occurs only on a surface. For the purposes ol this section we will ignore the Langmuir-Hinshelwood reaction, (12.77), even though on a real catalyst, the Langmuir Hinshelwood reaction is faster than the Rideal-Eley reaction at moderate pressures. The Rideal-Eley reaction dominates at high pressures, however. Following the discussion in Chaptei 5, wr will assume dial llu- ndsoipiion ill hi i....... goes via llu- n-ailiiin ißr2 +S ^==i Hi,,, (12.78) where S is a site on the surface thai is available to hold gas as described in Chaptei 3 We will also assume that reaction (12.78) is in equilibrium. Al equilibrium, the bromine concentration is given by the following equation: Kßr = [Brad] (12.79) In equation (12.79), KBr is the equilibrium constant for the adsorption process; (Bn|, [Brad|, and [S] are the concentrations of Br2, adsorbed bromine atoms, and bare-surface sites. Solving equation (12.79) for [Bra(l], and substituting the result into equation (12.71) yields rHBr = 2k2KBr[H2][S][Br2]'/2 (12.80) Next, we will attempt to consider how changes in the binding energy of the bromine atoms changes the rate of HBr formation. Well, when the binding energy of the bromine changes, KBr goes up and k2 goes down. In order to quantify the effect, we will assume that KBr is given by ^g*^ - ~p (r<*»« - TASgi)' (12.81) where AGad is the free energy of adsorntion ah ; ,u u entropy of adsorption, kB is Boltzman? clnsi T of adsorption, ASad is the According toUenius' law IT" giv^n ^ " "* abS°'Ute : k2 exp (12.82) ^^iuUnJ^ZT^^"?"2 'S aCtiVati°n "™ for ««*■ 2-luiiufe equation (12.72) into equation (12.82) yields k2=k?exp (iKiJ^Al^) (12.83) AHr,2 = AH0 - AHai (12.84) .uli.i II ni ni)'. equal ions (I2.HI) mill (12.HI) llllo equiillou ( I.' HID and llu-n subsuming in . 111 .ii m ti i (12.84) yields hin, 2k|| exp ( (I V|,..')AH„,i |H.||S||li. I vi Mil 2k? exp -(Eüi2-yp,2AHo + TAS.d II.' S'i I ( I ' HOI I here arc two interesting cases to consider: (1) the case where |S| is constant, so thai there i no limit to the adsorbed bromine concentration; and (2) a case where |S| decrcai i liiomine adsorbs, so that the surface can hold only a linite amount of bromine. Figure 12.11 shows a plot of the rate of HBr formation as calculated from equalion i 12.85), assuming that [S| is constant. Notice that the rate of HBr formation uu i. i . nionoionically as the binding energy of the Br increases. Physically, if we assume lhal |S|, Ihc number of bare sites, is constant, we are in effect assuming that there is nothing in slow down the adsorption of bromine. In such a case, the bromine concentration continues i" Increase as the binding energy of the bromine increases, and so the reaction rale ini rcusi moiiotonically with coverage. Quite a different effect occurs if one assumes thai the reaction follows a Langmuil adsorption isotherm. According to the analysis later in this chapter, if one adsorb hydrogen and bromine onto the platinum catalyst, the hydrogen and bromine lake up sites. [S], the number of free sites, decreases as hydrogen and bromine adsorb, One i an quantify this effect using an equation that we will derive in Section 12.17: [S] = So 1 +K Br2 V^br2 + ^/Ph (12.8/) figure 12.12 shows a plot of the rate of HBr formation calculated from equalion (12 13) with [S] from equation (12.87) and yp = 0.5. Notice that Figure 12.12 shows a volcano plot similar to that in Figure 12.10, where the rate reaches a maximum a heal of adsorption (i.e., heats of formation of the intermediate) and then declines. Physically, when All,, ~ 1E+31 2 o E a> ra 40 -20 0 20 40 60 Heat of formation of intermediate, kcal/mol 80 Figure 12.11 The rate of HBr formation as calculated from equation (12.85), with [S] = 1 x 10" cm2 and yp = 0.5, T = 500 K, Ph2 = Pbi2 = 1 atm. Ill IMVAIK IN HI •lAKAIII II'M I'IMNI ll'l I 723 rl y li i n, 1E+14 CD (T> Cxj" 1E+12 E o 1E+10 1E+8 ecu 1E+6 o E 1E+4 Rate, 1E+2 1E+0 80 Figure 12.12 yp = 0.5, T = "40 -20 0 20 40 60 Heat of formation of intermediate, kcal/mol JnnV n6 °ff0man°n calc"'^ted from equation (12.85), with [S] from equation (12.87) and suu K, = pBf2 = 1 atm. is small, increases in AHa increase the intermediate concentration substantially. The increases in AHa also decrease the reactivity of the intermediate since the intermediate is being stabilized, but that is a smaller effect. The net result is that the rate increases. In contrast, once the surface fills up with reactants, further increases in AHa do not produce substantial increases in the intermediate concentration. However, the increases in AHa still decrease the reactivity of the intermediate. The net result is that the rate decreases. If one works through the numbers, one finds that there is an maximum rate at intermediate values of AHa. One also finds that for a given reaction on a given catalyst, there is an optimal temperature where the rate is maximized. That is why there is a maximum in Figures 2.16 and 12.2. In actual practice, one almost always observes a maximum in the rate of reaction with increasing intermediate bond strength. Physically, with a heterogeneous catalyst, there are always a finite number of sites on the catalysts to hold reactants. Once all of the sites are filled, further increases in the bond strength of the intermediates mainly decreases the reactivity. With a homogeneous catalyst, there are only a finite number of complexes that you can make, before the catalysts are saturated. Once all the catalyst molecules are attached to reactants, further increases in bond strength decrease the activity. In practice, Sabaticr's principle works for most real catalysts. The one exception is in polymerization catalysts, such as a free-radical initiator. For example, a peroxide can initiate a free-radical polymerization via the mechanism ROOR-RO + CH2=CH2 -/?OCH2CH2. + CH2=CH2 - -» 2 AO. -» /?OCH2CH2. ■+ A>OCH2CH2CH2CH2. (12.88) In this case, one wants the initiator, RO., to have as strong of a bond with the reactants as possible. Phys.cally, the reactivity of the radical does not decline with increasing bond strength, so a high bond strength is helpful. There is one other detail of note. One does not need to know the heat of adsorption of the key intermediates to construct a plot like Figure 12.12. Instead, one can use any measure of the bond strength as the x axis in Figure 12.12, and one will still get a iniiliii curve This is very uiipoiiimi bet mine il often is i|iuie dilln nil in know llie henl ol 1 ti u m ol a reactive inlerinedliile However, it one can find some othei measure ol the I......I strength thai is piopinlnni.il tO th« heal Ol formation ol the reactive intermediate. ..... i an still construe) a volcano plol In llie literature, people often plot volcano plots as a function of the heal ol lormalion nl the bulk oxide per mole of oxygen. Figure 12.1 í shows a plol ol the heal ol adsoiplmii i ........bei oi adsorbed intermediates as a function of the heat of lormalion ol the bull ,i pii mole of oxygen. Notice that the heal of formation of the bulk oxide pel moll III iixyj'en is proportional to llie heat of adsorption ol the reactive intermediate, A surface 111 ,ii hinds some intermediates strongly tends to bind most adsorbates strongly as well I In ivlore, one can construct a volcano plot using the heat of formation of the bulk oxides M mi x axis (abscissa). Mi,- advantage of this approach is that one can construct the volcano plot from datl ih a ,ne readily available, and so one can construct volcano plots for a wide iiiiinlui ol n i. nous Table 12.11 lists the heats of formation of a number of oxides. One can use llie iI mi hi the table to construct volcano plots. I line is one subtlety in all of this. Table 12.11 gives the total heat of formal..... pi • Mini, ol oxide. We actually want to scale the heat of formation to moles ol .nl\,nl>.iir Wi II m the literature, there are two approaches to go to moles of adsorbale One idea i in assume that the heat of formation of the intermediate is proportional to the heal ol i.....i.ilion of the oxide per mole of oxygen. The second idea is to assume llial llie lie.it ol lormalion of the intermediate is proportional to the heat of lormalion of the oxide ;•. i mole of metal. One also customarily assumes that if there are multiple oxides, the iiongcst oxide is the most important. If one makes these assumptions, one can plot die rule of a catalytic reaction versus the heat of formation of the oxide, and expect to obscixe a volcanolike curve. I 'here are two key types of plots in the literature: . Plots of the log of the reaction rate versus the heat of formation of the oxide /><; mole of oxygen 10 20 30 40 50 60 70 80 90 100 110 120 Heat of formation of oxide per mole of oxygen, kcal/mol Figure 12 13 The heat of formation of adsorbed oxygen, hydrogen, and carbon monoxide on a number ol metals as a function of the heat of formation of the corresponding oxide per mole of oxygen. I'il»lc II* I I Ihr h.Ml nl Ioiiii.iIkiii i>l -.iuii.il m.l.il oxliln-i Oxide Heal nl Formation, kcal/mol ( Ixidc Heal ol Formation, kcal/mol Oxide IKal nl I m malum, Iii in, ■! AgO Ag02 AS2O5 A112O3 BÍ2O3 CdO CoO Cr203 CuO Cu20 Ga203 In203 -7.3 -6.3 -218.6 + 19.3 -137.9 -60.9 -57.2 -269.7 -37.1 -39.8 -258 -222.5 Ir02 -40.1 PtO -9.7 FeO -63.7 Re207 -297.5 Fe203 -197.5 RhO -21.7 PbO -52.4 kuli -52.5 Mn02 -125.5 Se02 -55 Mn^Oi -232.1 SnO -68.4 HgO -21.7 T1O2 -218 M0O2 -130 v2o5 -373 M0O3 -180.3 V2O2 -200 NiO -58.4 WO2 -136.3 PbO -52.4 W20, -337.9 PdO -20.4 ZnO -83.17 Source: Dala from the CRC or Landolt-Bornstein. • Plots of the log of the reaction rate versus the heat of formation of the oxide per mole of metal Plots of the log of the reaction rate versus the heat of formation of the oxide per mole of oxygen are called Sachtler-Fahrenfort plots (Sachtler and Fahrenfort, 1958). Plots of the log of the reaction rate versus the heat of formation of the oxide per mole of metal are called Tanaka-Tamaru plots. (Tanaka and Tamaru, 1963) Figure 12.14 shows a Sachtler-Fahrenfort plot and a Tanaka-Tamaru plot for the hydrogenation of ethylene. Both plots show some correlation to the data, although clearly there arc lots of variations. (Without the line, the figures look like scalterplots.) Generally, Sachtler-Fahrenfort plots fit better than do Tamaru-Tanaka plots. One has to be careful not to extrapolate these plots too far. Selenium, mercury, and lead are inactive for ethylene hydrogenation even though they have the heat of formation of the oxides is similar to those for the transition metal. Still, the Sachtler-Fahrenfort plots are a useful way to correlate data, even though they do not lit exactly. CAIAI VI PI CAN Ml III 'III It I.............Mil III Al I AN I'. Ill 1 I < I'll I'III KIMIIY lii summary, then, the Willi! 111 llns sec lion show 11 ml il a miIisi.hu c hinds the ..... 11111 ill,lies ol ,1 1 c.i. lion .11111 iid \ lull mil loo si mi if.lv. I he siibslam e will show some 1i.1l s In .ulivils llowcvci. one does not vs .ml to overdo the bond Strength, 01 else the nil.1,line will be calalylieally inactive. I .• II CATALYSTS CAN BE DESIGNED TO HOLD THE REACTANTS IN CLOSE I'll! iXIMITY \niiihci way thai a catalyst can work is to hold (he reactants in close proximity to each III hi 1 For example, in Section 12.1 we noted that in 1817, Davy discovered thai d one iiiiniiiidcd a candle with a platinum gauze, the platinum would prevent the Maine limn 111 me an explosion. In this case, the platinum is acting to catalyze the terminal..... 1 111 uis in the Maine so the candle does not cause an explosion. In 1834, Faraday proposed that the main role of the catalyst is to hold the icact.inis m I. . .1 proximity so they can react. I he reactions in a flame are pretty complex. There are many catalytic cycles Still I hi most important intermediates are the hydroxyls and the hydrogen atoms Figure 12.13 Iums a catalytic cycle for the hydroxyls. \s noted above, Davy found that platinum can catalyze the termination reactions m .1 II. line. In the gas phase, the main quenching reactions arc (12.89) (12.90) 2H- H + OH-> H20 limb reactions arc slow in the gas phase because the concentration of intermediates is low-When you add a platinum catalyst, the catalyst concentrates the intermediates. Hydrogens adsorb on the catalyst and wait for another hydrogen atom or hydroxyl to hit the catal) 1 and react with the hydrogen. The catalyst in effect concentrates the reactants. That speeds up the termination reaction. There are some subtleties here because the platinum also speeds up the initiation reactions. For example, if you put a platinum wire into a hydrogen/oxygen mixture, the platinum speeds up the initiation process more than the platinum speeds up the lei.....liiuhi pmeess. In that case, the (lame is initiated by platinum. On the other hand, with a candle 1e+20 o c 8 1e+18 8 V i. IE+16 lá § 1E+14 S E o 1e+12 rr 1e+10 Sachtler-Fahrenfort Ru ■ RhA \ pi /■pd „--- Ni -■- W ■ Ta "1 Cu ■ Tanaka-Tamaru -10 -30 -50 -70 -90 -110 0 -20 -40 -60 -80 -100 Heal of formation of oxide per mole ol oxygen, kcal/mol 0 -100 -200 -300 -50 -150 -250 Heat of formation of oxide per mole of metal, kcal/mol OCmHn_2 Figure 12.14 A Sachtler-Fahrenfort and Tanaka-Tamaru plots for the hydrogenation of ethyle H02 CmHn_, xo CmHn-2 2 Figure 12.15 A simplified version of the reactions in a flame. "' ■ '' .......m «P"*.....he termini.........,........,......,. „,.,„ J 'nation reactions. As a result, the plal.....m quenches the flame the L^ľntľľ8' ""i 'S concentratin« intermediates oi me reac............J he reactants are tn close proximity to react. It is just thai in one case the platinum speej ^itľScTctlons more than the lcrmination rcaĽ,ii,ns while......' °the- 3 12.9 CATALYSTS CAN BE DESIGNED TO HOLD THE REACTANTS IN THE CORRECT CONFIGURATION TO REACT In the previous section, we noted that catalysts hold the reactants in close proximity and thereby speed up rates of reaction. Still, it is interesting to note that one could gel even higher increases in rate if the catalyst could also hold the reactants in just the righl configuration to react. Pushing the reactants together would also produce an additional increase in rate. It is, in fact, possible to design a catalyst that holds the reactants in just the right configuration to react. Very few catalysts do this, but the ones that do are especially efficient. There is one specific example that illustrates the effects particularly well: the conversion of acetylene to benzene: 3C2H2 CM. (12.91) Reaction (12.91) is very rapid on a palladium catalyst. Figure 12.16 shows what is called the active site on the catalyst. The active site is defined as the arrangement of surface atoms where the reaction occurs. It happens that when acetylene adsorbs on palladium, the acetylene binds in what is called a "bridge bound state," where the acetylene is held off center of what is called a "threefold hollow" on the palladium surface. The bonding position is shown in Figure 12.16. It happens that if you put three acetylenes onto the palladium, the three acetylenes form a hexagonal structure similar to the hexagonal structure in benzene. The bond lengths are also about right for benzene. That promotes benzene formation. , aiai .......HIMlltl'.ANimmiMWinl MAIM IUIIMI. II I in..... 12.17 A cartoon of the reaction of ethanol and NAD' on the active site of liver alcohol i Im n |<' 11111 ,i i. |Adapted from Oppenheimer and Handlon (1992).] II one changes the geometry by, for example, going to a square surface, Ihe reaction is mil observed. This example clearly illustrates the idea that if a catalyst holds ihe ic.n...... m the nghl configuration to react, the reaction will occur at an unusually high rate I Ins process commonly happens in enzyme-catalyzed reactions. For example, the Uril 11 p m the destruction of alcohol by your liver is a hydrogen transfer from Ihe alcohol lo an ion called NAD+ [nicotinamide adenine dinucleotidc (oxidized form)]: NAD+ + CH3CH2OH NADH + [CH3CHOHl i I "i 'i Figure 12.16 The active site for reaction (12.91) on a palladium catalyst. i i. non (12.92) occurs on an enzyme called liver alcohol dehydrogenase, figure 12.17 hows how the reaction occurs in liver alcohol dehydrogenase. The diagram is adapted from Oppenheimer and Handlon (1992). The NAD+ fits into a pocket in the en/\nie I In alcohol sits over the NAD+ in a bent configuration, with a hydrogen in the alcohol pushing into the NAD+. The close proximity of the alcohol and the NAD1 facilitates ihe reaction. 12.10 CATALYSTS STRETCH BONDS AND OTHERWISE MAKE BONDS LASIER TO BREAK Some enzymes work in another way, too. They stretch bonds and otherwise make bonds easier to break. For example, Figure 12.18 shows a diagram of an enzyme called Lyso: ynti loll.. The name Lysozyme means that the molecule is part of a wide class of cn/ymes made in a part of a cell called a lysozome. The number 161L is the listing in Ihe protein database. (The listings go in order.) Generally, lysozymes are enzymes that animals use to kill bacteria. For example, there are lysozymes in your tears that kill bacteria in your cms Microbiologists say that lysozymes work by catalyzing the hydrolysis of polysaccha rides in the cell walls of the bacteria, causing the cell walls to rupture. Let me translate. Polysaccharides are carbohydrates (sugars). In this case the polysaccharides consist oi NAM (/V-acetylmuramic acid) and NAG (/V-acctylglucosamine) units. Both molecules ire shown in Figure 12.19. Lysozyme catalyzes the hydrolysis of the bond between ihe NAG and NAM units: -NAG-NAM-NAG-NAM-NAG-NAM- + H20-► -NAG-NAM-NAG-NAM-OH + H-NAG-NAM- (12> '.I AHM IMMUN I II IIIAN'IIIH IN 'UAH 'I 7M Active site i'll SIAIIII l/AIION ol IIIAN'......NMAIIS \nolhei iiiihIc ul en/yniatlc iicllon is Iti stabilize tin- 11 iinsil ion sllllc loi a icai lion When vmi stabilize Ihe transition stale, you lowei Ihe intiinsic harrier lo reaction ami llieiehy i ■ ■ ■ i up the reaction Recall from equation 9.2 thai anything one does to in< rcase <\ . ihe iniiiiliiiii function lor Ihe transition stale, will increase Ihe rale of reaction. One way lo .......ise i|' is lo stabilize ihe transition state. I lie lysozyme example illustrates this effect as well. Note that in Ihe discussion above. m .ml thai ihe NAM is stretched into a planar configuration when the polysaccharide hlllds to the en/.ymc. In the transition state lor reaction (12.93), NAM has a plunut .....figuration. The enzyme stabilizes that planar configuration. As a result, die binding "I Hi. polysaccharide to the enzyme has stabilized the transition state for reaction (12.93) Most enzymes are thought to stabilize the transition stales for reactions. Thai i purl ul the reason why enzymes are so catalytically active. Figure 12.18 A picture of Lvsozvme 1 fill Thio , using data in the protein database frorr^ an x Jav Jlaetion ' "T^ USin9 3 Pr°9ram called ™SMOL, (•987). m ^ diffraction spectrum generated by Weaver and Matthews HOH2C HÖ H3C\ ^COOH HOH,C CH"' 9^0 0H HN \ HN \ iC-CH, H°\ ^COOH HOH2C PH ^L-J~^ HN HOH2C HÖ HN \ NAM NAG Figure 12.19 The structure of %C—CH, —NAM—NAG-NAM, NAG, and a NAM-NAG repeat unit. The active site for the hydrolysis reaction is indicated by the arrow in Figure 12.18. Generally a six-sugar -NAG-NAM-NAG-NAM-NAG-NAM-unit binds to the active site. The bonds in the NAM that are being hydrolyzed are stretched, and the NAM is distorted to a near-planar configuration. Then a proton is donated from the enzyme, breaking the polysaccharide bond. Then water is added to the resultant fragment. The first step is for the enzyme to stretch. This makes it easy for the proton to come in. In addition, the lysozyme has a breathing mode where the opening (cleft) in the lysozyme opens and closes like a Pac-Man. When the Pac-Man closes, the proton is pushed into the stretched bond in the sugar, which also facilitates reaction. Note that both of these interactions are modifying the forces between the reactants, to facilitate the reaction. The enzyme is also stabilizing the ionic intermediate formed when the proton reacts with the polysaccharide, cleaving the NAG. This example shows that enzymes can stretch bonds. Bond stretching makes bond scission very likely to occur. 12.11.1 Catalytic Antibodies I here is a separate branch of chemistry called catalytic antibodies that tries lo exploit iliese findings. The idea is to build an antibody that selectively binds to Ihe transition State lot a reaction. Note that by definition, when an antibody binds to a specific mole, ul. Iluil molecule will be stabilized. Well, when the antibody binds to a transition stale, the transition state will be stabilized. That increases q* and thereby speeds up reactions For example, Gouverneur, et al. (1993) synthesized a series of catalytic antibodies Im ihr I )iels-Alder reaction: c c I + II C\ c. c I I X (12.94) The antibody was designed to bind to the transition state for the reaction. Thai lowered ilie energy of the transition state. Fxperimentally, the catalytic antibody was quite a good catalyst for the reaction. I Ins i .ise is particularly interesting since the catalytic antibody does chemistry that is nol found in nature, and unlike a standard metal catalyst, it produces enantiomerically pure products The result is that these compounds look quite interesting. There is a new branch of chemistry developing on the basis of these findings. Generally, people first synthesize a hapten, which has a structure similar to that of the transition stale for a reaction. They then feed the hapten to mice or bacteria. The mice or bacteria treal the hapten as an invading molecule, and build antibodies to the hapten. Those antibodies ne separated, and some of them show catalytic activity. These results show that materials that stabilize transition states can enhance reaction rates. 12.11.2 Transition Metal Catalysts Transition metal catalysts (e.g., platinum) can occasionally work similarly to catalytic antibodies in stabilizing the transition state for a reaction. Note that during a reaction 'IIAIIH l/AII<>N(H MIANHIIH IN Ml All 'I one is mixing excited si.iics mln ihc gmuiiil slate ol n system II one inn stabilize llittse aittibonding orbituls, one will also slahili/e the liansilion stale As a icsuli. one can lower lbe energy of the transition state anil thereby facilitate a reaction, Generally il electron are able to interact with anlibontling orbituls to stabilize them. S anil p electrons arc less able to facilitate reactions. For example, consider the dissociation of hydrogen on platinum: H2 + 2S -> 2Ha (12.95) Reaction (12.95) is a classic four-centered symmetry-forbidden reaction. So, on the basis of the analysis in Section 10.9, one would expect reaction (12.95) to have a large intrinsic barrier. Well, on aluminum the reaction does have a large barrier. People observe little dissociation of energies up to 50 kcal/mol even though the reaction is 60 kcal/mol exothermic. If one plugs these measurements into the Polanyi relationship with a yv 01 0.5, one finds that the intrinsic barrier is at least 80 kcal/mol. An intrinsic barrier of 80 kcal/mol is consistent with our expectations for a symmetry forbidden reaction. On platinum, however, the reaction is unactivated. The heat of adsorption is 13 kcal/mol, and so if we plug into the Polanyi relationship, we find that the intrinsic barrier is less than 7 kcal/mol. Clearly, H2 dissociation on platinum does not have the large-intrinsic barrier one expects for a symmetry-forbidden reaction. This occurs because the reaction is no longer symmetry-forbidden on platinum. Figure 12.20 shows a correlation diagram for reaction (12.95). In the absence of the platinum, the correlation diagram is identical to the diagram for reaction (10.11). During the reaction the oo* state is lost and a a*a state forms. This is a standard symmetry-forbidden reaction. When the platinum is present, everything changes. The d bands in the platinum add some extra states, which I have labeled A and E in the diagram, both of which are partially filled with electrons. In the material that follows we will show that during the reaction, the electrons from the aa* state can flow into the A band, while electrons from the E band can flow into the a*o state. That allows the reaction to occur with a minimal barrier. A good way to understand this interaction is to look at what happens when electrons flow into the a*a state. The o*a state consists of an antibonding state in the H2 plus a corresponding state in the metal. Figure 12.21 shows a diagram of the antibonding state. Platinum Reactants Products Reactants Products Reaction coordinate Reaction _ coordinate •.....I ul I hit i- is a sign i 11.111 )•• in tin......I... 11.1111 > • si.tic ( onsequeiillv, itccniillllg In the ........il in Section 10.11. d one h.e. ,i Kuli' in the metal with a s.....I.n sign change, and in lake electrons out ol ihul stale, ilicu one can gel the reaction In occui imoothl) hi sinm ii in figure 10.35. I iguic 12.21 shows a diagram ol Ihc key oibiials during the reaction. There are two .....I ..I oibiials in platinum: s orbilals and d orbitals. The s orbituls spread out ovei ih. whole surface. There ure no big sign changes, so the s band cannot iniciaci Willi ihc Kill......ling oibiials on the II... In contrast, the I i states in the d bands have the same sign i . in ihc antibonding orbilals in the II... Consequently, one can iransfei clcclions ........In d hands lo ihe antibonding orbilals in the H2 with minimal difficulty, In 1,1.1, Ihere is some subtlety to the arguments, because one wants lo break die II II i noi simply stabilize the antibonding orbilals. Well, the d bands do that, loo Rei ill ih a orbilals of the same sign attract while orbitals of different sign repel. If one looks al lln s hands, one finds thai there is an attractive interaction on one side of the hydrogen bul ii repulsive interaction on the other side. The net effect is thai the s bands do not .iiii.n i ih. .iniihonding orbitals. On the other hand, if one orients the hydrogen as shown in the i.i diagram in Figure 12.21, both sides of the antibonding orbitals on the hyilingcu III In- attracted to the d bands on the platinum. The overlap increases as the hydrogi n ... i.iic\. The net effect is that the antibonding orbitals in the hydrogen arc pulled I. hi .ml the platinum atoms. That rips the hydrogen apart. The net effect is that the intrinsic barrier for the reaction has been reduced h\ ovci I i I . al/mol. Generally, the 70 kcal/mol or more reduction in intrinsic barriers occurs only loi symmetry-forbidden reactions: dissociation of diatomic molecules into two atoms, 01 i urn of double or triple bonds. Most other reactions show similar intrinsic barriers on In Mli transition metals and nontransition metals. s Bands Figure 12.20 A correlation diagram for reaction (12.95) with and without platinum. Figure 12.21 A diagram of the key interactions during the dissociation of hydrogen on platinum. 12.11.3 Acid Catalysts .........:......:......*.....■........i In order ,„ See how lha, w„rts. consider *. rcacli.....,isr„»d ...... ,, „ 2. ........ii. •i ii-M......iuiaiim 733 /?HC=C/?H /?/?C=CHH (12.96] yiell^tfot2 " "°ted th3t m 3Cld S°1Utl0n the ^ wi* the olefin J /MC=C/?H + H+-, [/?HC=C/?HH]+ Then the carbocation isomerizes: [/?HC=C/?HHr [ff/?HC=CHH]^ (12.97) (12.98) Then the new ion loses a proton to form the products: [RRHC = CHH]+-> tf/?C=CHH + H+ (12.99) In this case the acid is doing several things. The acid is initiating the reaction. However, the acid solution is also stabilizing the charged species. The stabilization of the charges is important to the reaction. In the material that follows we will discuss how the charges affect the barriers to reaction. In order to determine how the charge on the molecule affects the barriers to reaction, it is useful to consider a reaction where a neutral hydrogen atom reacts with the olefin to yield a radical: X+ «HC=C/?H + H. Then the radical isomerizes: . R X+#HC-CH2- Then there is loss of a hydrogen atom: R /?HC-CH2 + X R . /?HC-CH2 + X R X+/?HC-CH2-► RC = CH2 + H. + X (12.102) First, let us consider reaction (12.100). Figure 12.22 shows some of the key orbitals during reaction (12.100). The reactant, RHC=CRH, starts with a 7r-bond. Recall that a 7r-bond forms when two p orbitals line up and bind together. The rr bond looks like an extended P orbital as shown in Figure 12.22. The hydrogen starts out with a spherical orbital as we saw in Chapter 10. Now consider what happens when reaction (12.100) starts. When reaction (12.100) starts, the hydrogen approaches the ethylene. In Chapter 10, we found that there will be a bonding and antibonding interaction. Figure 12.22 shows the bonding and antibonding molecular orbitals. The two MOs look almost the same, except that in the bonding MO (12.100) (12.101) - —v^ Antibonding Bonding Figure 12.22 A diagram of the key MOs during reaction^ 2.100). iii. lop lobe on the ethylene has the same sign as the s orbital while in the antibonding M«». the lobe has a sign different from that of the s orbital. Now consider what happens when the orbitals come together. This case is very similai in ihe cases we discussed in Section 10.6.3. Notice that in the antibonding stale, a positive .iini.il is pushing up against a negative orbital. Physically, we are pushing one balloon ..I electrons into another, so there is a repulsion exactly like the repulsions discussed in i hapter 10. The repulsive interaction pushes up the energy of the system. Consequently ilu reaction is activated. A more detailed analysis shows that the repulsion occurs because there are three i In lions in the system. If there were only two electrons, we could put both electrons into the bonding state, and so we would not have to put electrons into orbitals of different nlgns. Consequently, one could eliminate the repulsive antibonding interaction. However, 11 ii- bonding MO can hold only two electrons. If one has three electrons, one of Ihe . 11vIrons must be put into the antibonding MO. That produces the repulsive interaction. Notice that if one starts with H+, there will be only two electrons in the system, so [here will be no repulsion. Consequently, the fact that one has protons or hydroniums, rather than neutral hydrogen atoms, in acid solution eliminates most of the barriers to icaction (12.100). One can also run reaction (12.100) on a solid. Solid catalysts are great because one . in modify the charges on the protons. Recall that in a covalently bonded molecule such us water, the hydrogen has a small net positive charge. That is why water has a net dipole, In a solid, one can increase the charge. On certain solids, called superacid catalysts the charge is nearly +1. That promotes easy proton transfer. Consequently, superacid catalysis lire able to lower the intrinsic barriers to reaction (12.100) The superacid catalysts also promote reaction (12.101). Reaction (12.101) is more i oniplex than reaction (12.100). Reaction (12.101) starts with a radical. There is a hall Idled P orbital on one carbon, and a C-R bond on the other carbon. Figure 12.23 shows a very approximate diagram of the key orbitals during reaction (12.101). I li lili I. II it1 - I III.,! I 'llllllll Is I mpty IV III IV Filled Ó8 / /// Figure 12.23 A rough diagram of the key MOs during reaction (12.101). During reaction (12.101) the system starts out with the orbitals in the left ol Figure 12.23. There are four MOs labeled I, II, III, and IV. In the diagram, we have arbitrarily assigned the lobe on the R group to have a positive sign, and then considered all possible signs on the p hybrids on the carbon. In MO I, both the p's are positive; in MO II, the left p is negative, while the right p is positive. In the MO III the right p is negative and the left p is postitive, while in MO IV, both p's are negative. In MOs I and II, the 71 group has a bonding interaction with the carbon, while in III and IV, there is a sign change in moving from R to C and so the interaction is antibonding. As a result, orbitals I and II are bonding orbitals while orbitals III and IV are antibonding orbitals. In contrast, at the end of the reaction, the R group migrates from the right to the left of the molecule. In this case, orbitals I and III are bonding while orbitals II and IV arc antibonding. Now consider moving the R group. Notice that the R group needs to move across the molecule for reaction to occur. However, in order to move the positive orbital on the R group in orbital II, it will need to displace the negative nonbonding orbital on the carbon. In Chapter 10 we found that such orbital displacements have large barriers. The net effect is that 1,2 displacements have large barriers with neutral radicals. Notice that the repulsion occurs only because the electrons in the R group are pushing up against the nonbonding orbital in the molecule. If one modifies the charges on the molecule, one can remove the electrons from the nonbonding orbital. If one puts a +1 charge on the molecule, there will be no repulsions. Well, again on superacid catalysts the charge is nearly +1. That promotes easy isomer-ization. The net result is that the superacid catalyst is able to promote reaction (12.101). These results show that catalysts can modify the intrinsic barriers to reaction. These modifications allow very selective reactions to occur. These results show that catalysts can be designed to modify the changes on the reactants in a way that facilitates reaction. 12.12 CATALYSTS CAN BE DESIGNED TO BLOCK SIDE REACTIONS Another thing that catalysts can do is to block side reactions. The idea is simple. You design the catalyst so that it is shaped in such a way that the reactants can get together fiATAI V*T> ľ! AN in m M'.ni li i" m'" ľ I II III Al / I't ......„„,i,,,,,„..................................."'.......'•;...... ^^rrrv^^ÄR.,*................... ■titnylene: CH3 nH2C=C!ICHi [-C-C-]n (12.103) Polypropylene undergoes free-radical orcationic polymerization as discussed in ( haplei "i 11 the free radical reaction occurs, the methyl group can go on the top 01 the bol lom 1 I hi molecule. If the methyl groups arc distributed randomly, the polymei has pool .....h.uiical properties. As a result, the polymer cannot be used in many applii atioitl In ......1.1 il one can control the positions of the methyl groups, one can produce polymi IK nil much better mechanical properties. Such polymers are very valuable I here are (wo key forms of oriented polypropylene: isotactic polypropylene, when all ..1 ílu methyl groups are on one side HpH H£h HCH HCH h£h h£h /c-c" CscxC"c" ^ iml ,„d,o,actic polypropylene, where the methyl groups alternate from side I.....le u H H H HCHHCH HCHHCH V x x x h h HCH H(jH \^ v 1(1-1) 1 .insider making isotatic polyethylene. Figure 12.24 shows a diagram of a step during Hi. production of isotactic polypropylene where a single propylene unit is added to 1 flowing polymer chain. Notice that one can add the propylene, with a methyl group I.a nig in the correct direction or in the wrong direction. If one would add the propylene .Mill the methyl group facing the wrong way, polymerization would still occur, bin one would not end up with isotactic polypropylene. HqH h£h h£h h^h h„hh h£h h»h v?" Cc \ h£h h£h h£h r h^h Methyl on wrong side h*h h*h h''h Figure 12.24 A rough diagram of one step during the production of isotatic polypropylene.