..

.,

Many orlhe topiCS containld In Ihb bOok arc contained herein silould be Celi,':-,-'-illYcntion or prOCUS which ml)' Reason.ble ~tC has been \l$C:d i

...

" - "' ~' ." " ' "

'. :".

. . '.' ,~ :: .' '.' "J ., ' .1' "'-. .. . ~ ""'~I

. ", ..

, : ;3,

. , ..

'. , . ":':"; .: Copyri&liIQM.V. TwiU,I989 Published by Wolfe Pubtishlngl1d. 1911'J Printed by BUliu &: T~nnc r. Frome, En,I:llld ISBN U72J.4 UllH2

., ""If . i 1

AU filllls rc~rve4. No rcprooJuClion, ropyorlnnsmiuicmoflhls publiclu;on may be: m:uk withuut wrine" pcrmil4ion. .

t-Io patt Of lhili J'lUhtitulicn mDy be r~l'r""u(cd. copied or Ir~nll'nillC'" ,,"vc wilh wrincn rcrmiuion orin .C(ordancc wiLh the provisions of the CUf')'ri,1I1 Act 1~56 (I~ ammcndctl), Of under the tumlim ~ any liecn pcrmininsl;onilcll copying issued by the CtlJlyri,IIILiccnwnl

.. ...;AI~ncy. 33-34 Alfred l'I.ee-, LumJun, WCIE 7DP. 'A~~ pqr'son who lI(H:s ~11)' unmutho.iscll act in ,cl~!ion 10 thi' pllbliCliliun ~ . in.~'r be liable \0 criminal p",~cclltilln anll th'il diims (or .!i!Mlles.

A CIPC'~ I ~l"&uc r[cOIII for thh h"l.ll.: is "Y;lil~blc flom the British l,ibn,y. Fur:l full lisl of Wllift Sci~!I~ Atb,;cs. plus (ullhcominllidulnd dCI~ils of aUI oliK:r Aiiascl. plc;\S! wlile IU WlIlfe Publishi", UII , l16 TUlfi"IIQI, f!;Ke , L"nllun WCIE 1LT, Engl~n"'. .' . . ( , ,. -t

! ,

. . . ' ' j :- . :. 1..11;,] .... :...

.: -

:.,

Contributors G. W. Bridger BSc, MRSC

P.J.H. Carnell BSe, PhD p, Davies BSe

-- ' R .T. Do-n.l'!' -'! "

D.R. Good man BA N.H. Harbord BSc, PhD

J.R. Jennings BSc, PhD

L. Lloyd BSc, PhD W.J . Lywood BSc, ACGI . ~ ., ~.

B:.8. Pea~~e. BS~! PhD D.E. Ridler BSc ' ..

._-_ ._-_._-- -

." .

, .' . ,',

...

-. '.

-. .

M.S. Spencer BSc, PhD, C.Chem., FRSC M.V. Twigg BSc, PhD, CChem., FRSC

~y.~ILlZ.\> S.A. Ward BSc ""i ' "0 C. Woodward BSe, PhD ~~ ~

o '% ~ ENG INEE RI NG 0( ~ OfFICE !; .. "

FOR REFERENCE ONLY

I I

I I

Contents

Preface .................................................... .

ChapuT 1.

Fundamental Principles M.S. Spencer

.......... ..........

1.1. Fundamentals of Heterogeneous Catalysis . 1.1.1. introduction........... .. . ................................... . 1.1 .2. The Role oC.catahSI$ ............................. ............. .. ... ...

1. 1.2.1. Ammonia s)'~;i;;~i~:::::::::". 1.1.2.2. Ammonia Oxidation ....... .. ..........................

1.1.3. The Nature of the Dialytic p;~;~.~ ........... ' .......... .............. . 1.1.4. Catalyst Activity ....................... : ............. ....... : ..... ..... .... . 1.1.5. Catalyst Sel cf . . ............................ 1.1.6. Steps in the Ca~~~~kP;~~~~ .... .. .. .. .. .. .. .. .. .. .. 1.1.7. Adsorption and Desorption ......................... ;: ................. .. t 1.8. C~t~lyst Oe51&n ........... :::::::.::: ................................. ..

1.2. Catalyst Manufacture ................................. . 1.2. l. lnttoduction ... :: .. ................................................. . 1.2.2. Unsupported Me~~i~ .... .. .... .. .... .. .. .. .. .. ...... .. 1.2.3. Fused C,lIalysts ....... ::::::::::: .. .. .... .. .. .... ................ . 1.2.4. Wet Methods or Catalyst Man~r~~;~;~::: .. .. .. .. 1.2.5. FundamentOllsorPr~dpit;L1ion Proc~s ............................. .. :,'.7'. ICa talyst Manuf;U:lure by Pr~cipit;:uio;;r~~~; ...... ...... . . mpr~gnation PmC'esses. .. ................ ..

1.2.8. FormingStagcs ............ ::::::::::: ............................ . ...............

J6

17 17 18 18 19 2J 24 26 27 29 32 34 34 34

" 37

" 411

" " 13. CatOlJ)'S1 Tesling......... . . ............ .. .................... .

1.3 . l.intrOOuction . ........... ........ ..... ....... .... 48 1.3.2. ChemicOIJ andPh:i~~ip;~;~~;: .. .. .... .. .... ........ .. 48 1 33 B IkCh

. 'w I'W Les........... '9 U' emtc:1I PTOpI:nies ............................. ..

1.3.4. SurfaC'e Chemi~a l properti~;: .... .. .. .... .. .. .......... .... 49 1.3 5. Physica l Properties ........... :............................. ............. 50 1 3.6. {;atalyst Performan('~ ............. ................. ............ 52 1.3.7. Coarse LabOrntoryS(';~~;;i~g: .. .. ......... 55 1.3.8. Fine Labornto'" Scree . ............................. ............... 56 139 s' ., nmg.......... ...... ,.

. . emHechnicill Catalyst T ~'s tin ...................... .......... . 1.3.10. Reaction Kinetics II .......................................... 60 1.3.11 . Cmlyst Ageing .. :::::::::::::::::::: .. : .. ........... 6 1 1.3.12.MechanismoftheCalalyt'~R .; ..................................... 6S 1 ~OIctJOn ......... ........... ........ ..... 66

Contents

. . " .

1.3.12.1. Ammonia Synthesis ...... ;.:: ...... : .............. : . . :; ...... . 1.3.12.2. Methanol Synthesis , ...... ... ............................ .... .

C I . U . .. ' ' .. ' " ~ .

1.4. atayst m se ............... .... .. .. - .. .... .. .. ':-'" \.4.1. Introduction .............. : ........ : .. : .. .. : ...... : .......... . \.4.2. Pretreatment and Activation.: ............. .. .... .. 1.4.3. Loss of Catalyst Performance: ......... .......... ........ .. .... : .. .. .. 1.4.4. Physical Causes of Decay ......................... .. : ...... .. .. .. .. 1.4.5. p oisoning by Impuri ti~ in Feeds or catalysts ....................... . 1.4.6. poisoning by Reactantsor Produd$ .................. : ................ .. 1.4.1. Interact ions in Catalyst Deactivat ion ........... : ...................... .

. . ... - : , " ,:.;"'" . .... :- .. '

. ' .. ' :~!'i:' : Chapter 2.

Process Design, Rating and Performance .: . W.J. Lywood ...

67 68 (II (II (II 73 76 77 81 82

2.1. Design ofCalalytic Reactors ...................................................... B5 2.1. 1. Operating Temperature and Pre$ure .................................. 81

2.1.1. 1. Desulphurization Reactor ...................................... 87 2. 1.1 .2. Ste~m Refnrmers .:.: .::,,;.................................. .... . 81 2.1.1.3. Water-gasShilt Reactors ....... . ........ .... .... .. .. 88 2.1.1.4. MethanatiOn Reactor .................... ........................ 89 2. U.5. Ammonia and Methanol Synthesis Reactors ......... .... . 89

2.1 .2. Converter Types ............................................................. : 2.1.2.1. SinSle Adiabatic Sed ........... .. ...... .. .. 91 2.1.2.2. Quench Convener ............................................... . 2. 1.2.3. Int e r.b~d Coolin!:. ........................ .. .... .... 91 2.1.2.4. ICI High-convcrsion ReactOr .................................. 96 2.1.2.5. Tube-

I I I I

Comems.

2.2.1. Optimum Opcratin~ i ..:r.I.l~r;:ture , .................................. 121 2.3. Catalyst Performance ............................................................. .

2.3.1. Fall in Apparent C"WI)~t ,l.~ti'ily .................................. .. . 2.3. 1. I. PoisoninglSinh.: ring .................... : ....................... . 2.3.1.2. Poor Ga.s Distribution ........... .. .......................... .. 2.3.1.3. Poor Mixing of R..:actants ................................... ..

2.3.2. Increase In f rC3sure Drop ..... .......................... ............. .. . 2.3.2.1. Breakage or Ertl:ol () n uf Catalyst P&ttif;lC$ ... ............ . 2.3.2.2. Disintegration of Catalyst Particles ....................... . . 2.3.2.3. Deformation of Catalyst Particles ....... ..... ..... ..... ... . 2.3.2.4. ClIfryover onto Catalyst Bed ............. ........... .... ... . 2.3.2.5. Collapse of ncd Support ..... ........................... .... . .

2.3.3. Measurement of Perfurmance .............................. _ ........ .. 2.3.3. 1. Analysis ......... ................................................ .. 2.3.3.2. Mass Balante ............ y ..... : .............................. .. 2.3.3.3. Cat:llystbcd Temperature Rises .......................... .. 2.3.3.4 . C~ulystbe:d Temperature: Profiles ........................ . 2.3.'3:5. Ra.dioo.ct ive:Trating . ........ .. .............................. .. . 2.3.3.6: Pressurc: Dr~p ......................................... ......... .

2.3A. Quantifying Cafalyst Pc: rfo,rmanee .......... .. ...... ............ ..... . 2.3.4. 1. Composition at the Exi t from the: Reactor .............. .. 2.3.4.2. Approach to Equilibrium ............................. ....... . 2.3.4.3. Activity or Active: Volume: of Catalyst ................... ..

2.3.5. Calculation of Catalyst Performance ........ ......................... . 2.3.5.1. Rcuctor Exit Composi tion ..... _ ............ :-: ............. .. 2.3.5.2. C"reulation of Approach to Equilibritlm ................. . 2.3.5.3. C.IICt1lation of Activity or Active Volume from

Composi tion ........................................ .... ........ . 2.3.SA . C~leulation of Activity or Active: Volume from

Temper~ture Profiles . ... .... ............................. .. .. . 2.3.6. Application of Methods It) Ammonia and Methanol Cat~ l ysts

2.3.6. 1. Dcsulphurizcr ................................................... . 2.3.6.2. rrim~ry and Scrond~ry Reformer ...................... ... . 2.3.6.3. li igh Tempcnllur..: Shifl ..................................... .. 2.3.6. -1 . Low Tcmp.:rJture Shift ....................................... . 2.3 .6.5. Mcth~n~ tor ... .................................................. . .

~.3.6.6. Ammnni~ ~nu Methanol Synthesis Convater ......... .. 2.4. Computer Prn!; r:Jlns .... .. .. ....... ... ............................................ .

2.4. 1. Re~sons ror Using Computer Cllcul~tions ......................... . 2.-1.1. 1. Acrurate C ... lcula tions ........ ....... .......... .. ............. . 2A. I.2. Nonimthcrmal Reactors ..... ......................... ...... . 2A.\.3. Multiple Reactioll$ .......................................... . .. 2.4. IA . Optimization ............... ................... .................. . 2.4.1.5. Simul~tion ....................................................... .

2.4.2. Typ-cs ur Comput..:r rrnSr;'ms ........... ............................... .

1~3 123 12.' 12-1

12~ 125 I2S 125 126 126 126 126 126 [27 127 127 127 127 127 128 12" 128 128 129 130

IJ]

'" I3S 135 1)

Conlln.s

3,11.3. Stabilization of Low-temperature Shift Catalyst ................ . 3. 11 .4. Stabilization or Methanation Catalyst............................ . 3.11 .5. Stabilizalion or AmmoniaSynthesis Catalyst ............ ..... .. :::

3.12. Catalyst Discharge ............................................................ 3.12.1. General ....................... .. ...................... . . :: 3.12.2. Discharge of Pyrophoric Catalyst ................................... . 3.12.3. Top Discharge ............. _ .......................................... . 3.12.4. Blanketing Pyrophoric Dlalyst During Vacuum Extractio~ " 3.12.5. Discharge of Ammonia S)'nthesis Catalyst. ...................... ::

3.13. Reuse of Dischllrged Catal)"lt ................................................ . 3.14. DiHl('ul of Used Catalyst ......... : ...................... ............ ........ . 3.15. Safety Precautions ................... ................................... .........

Chap/Cf4 .

Feedstock Purification PJ.H. Carnell

4.1.lntroouction ..................................................... :: ........ ....... .

4.2. Feedstocks for Ammoni3, Methltlol and Hydrogen Production ...... . 4.2. 1. Natural Gas ................................................................ . 4.2.2. Associated Gas, Natural Gas Condensates and LPG ......... .. . . 4.2.3. Naphtha .. , ....................... ........................ ................... . 4.2.4. Refinery Off Gases ant! Electrolytic Hydrogen . .................. . 4.2.5. Coal Gasification and Coke Oven Gas .............................. . 4.2.6. Mixed Feeds .................. .. .. .. .. ....... .

4.3. Dcsulphuriution .................................................................. . 4.3. 1. Processes for Single-stage Sulphur Removal ..... ....... ........... . 4.3.2. Processes for Two-stage5ulphur Removal ......................... .

4.4. Thermal Dissociation or Sulphur Compounds .............................. .

4.5. Iiydrogenolysis or Sulphu r Compounds ..................................... .

4.6. Cilrbonyl Sulphide ................................................................ ,

4.7. Cobalt Molybdilte Calalysls .................................................... . 4.7.1. Presulphiding Cobilh /lloIyNlale Catalyst ......................... . . 4.7.2. Other Reactions O\'cr Cob' . .!t Molybdate Catalyst ............... .

4.8. Nickel Molybd~te ClIt~lyslS ............................... .... 4.9. Physical Form of Cahall am: ,'I ' ct hlolybdate Catalysts ............... . 4.10. Replacement and DischarGingof Cobal t and Nickel Molylxble

Catalysts ........... ......................... .. .. .. ... ............ ...... .

181 182 183 183 183 184 ISS 186 186 187 188 188

191 192 192 193 194 194 194 lOS

1% 1% 19' 199 200 203

21~ 2tlS 206 207

207

208

1 I

. : ContfJnb

" ; . 42 ... , :- , '., .' : '--"'" -r : . ': .1. Dechlonnauon ... :.:::: .. L.: .... t.:; ..... ;.:.:: ... ~:: .. :.: ... :; .... :.:.: ....... .

4. 12.1. Chloride Sources and Absorbents ... : ....................... ........ . 4.12.2. Oper.l.ting Conditions .. ; .. : .. : .... .-~.::.: ..... : .... .. ;: .. !;:.;~:: ... :.: .. .

, . .. . _ .. "~'.f _ ~ ... ,. . -

4. J3. Removal of Silica and Fluoride ... : ............ .. ........... :: ................ .

4.14. Demetal1ization .............. ................ .. ::.................................. 223

4.15. Denitrification ... :: ...... : .......... : .. .. ::::: .. :: .. : ................. :. ........... 224 '"

~ .. '- . .

Choptu5 . . . , ,",

Steam Reforming D.E. Ridler, M.V. Twigg

5.1. History ....... ........................... .................................. ............ 225

5.2. Feedslock and Fdstock Pretreatment........... ................ ........ .... 226 5.2. I. Natural Gas ................................................................. 227 5.2.2. N3phthas..................................................................... 228

5.3. Chemistry of Sleam Reforming .............. ...... ...................... .... ... 230 5.3. 1. Thermodynamics................... .......... ...................... ........ 230 5.3.2. Kinelics.... ......................... ............. ................... ....... ... 239

5.'1. Design or Steam Reforming COltalym.................................... .. ... 24~ 5.4.1.SeleClivity.................................................................... 244 5.4.2. Thermal Stability........................................................... 24~ 5.4.3. Physical Properties ..................... ....... .................... .... .... 244 5.4.4. Nickel as a Sleam Reforming Catalyst ................................ 244 5.4.5. Sup?Orts ror Nickel Steam Reforming Catalysts ......... ... ....... 249 5.4.6. u.rbon FormOltion on Reforming Catalysts ............... .... ...... 250

5.5. Secondary Reforming ............................................................. 253

5.6. Catalyst Dimensions... .............. ............................. ................. 254 5.7. Uses of Catalytic Steam Reforming............................................ 256

1 I' I "

I I I,

! "

; I ,I I

I ,

I I I I

Contents

5.7.1. Ammonia Synthesis ........ .... . .......... ............... . S.7 2. Methanol Svnthesis ..... ........................... ... ...... ::.: .... . s. 7.3. 0)[0 S~nlhesi;, Gns ..................................................... :: :: s. 7.4. Rcdu~mg Gas .... , ..................... ................................... . 5.7.5. Town Gas ......................... : ...... ............................ . 5.7.6. Sulntitule N'llural Gas (SNG) ............................. ...... :::::::

5.8. Practical Aspects of Stc:om Reformers .... ............ ............... . 5.8.1. Containing the Catalyst ........ .............. ................ . ...... . S.8.2. Reactant Gas Distribution ........ ... .. ....... .. 5.8.3. Firing the Reformer ... ..... .......... . :::::::::::::::::::::::: ........... . 5.8". Expansion and Contl'll;lion of ReformerTubes .... ...... .. 5.8.5. Faci.li[i~$ 10 Olarge and Discharge Catalyst ..... :::::::::::::::::::: 5.8.6. Des1gnlng a Reformer for Effieient Operation ................. .. 5.8.7. Cat~[)'st Reduction ... .. ......................................... ..

5.8.7.1. Redl.lctionwith Hydrogen ........................ : .. ::::::::: 5.8.7.2. Reduction with Ammonia ............. ............. . 5.8.7.3. RedLlction with Methanol .. .................................. 5.8.7.4 . Reduction wjlh Natlora! Gas ....................... .. : ...... 5.8.7.5. Redlldion with Other Hydrocarbons .............. :::::::: 5.8.7.6. Redll(lion After Shuloown ................... : ............... .

5.9. Factors Affeclinglhe Life of Reforming Catalyst ......................... .. 5.10. Catalyst Poisons ........................................ : .................. .

; .10.1. Su!ph~r ................................. ....... ..... ...... ~ ......... ::::::: . 10.2. Arsenle .. . ................... ..... ............ ....................... ....... .

5. 11 . Hot Bands in Naturll Gas Refonners .................. .................... ..

Chapltr6.

The Water-gas S hift Reaction L. Lloyd, D.E. RidJcr,M.V. T wigg

6.1. Int roduction ...... ............ ...... ... ..... ....... .... ........... ............... . ..

2l' 218 2l' 260 26 1 263

'" 26'

'" 270 271 27J

'" '" m m

'" '" 27' 217 m

27S 278 278

280

'" 6.2. Thermodynamics ................................................................... 285 6.3. Kinetics and Mechanism .. .......................... ,............ .. ..... ... 2SS

6.3. 1. Kinetics Over HT Shift C3talyst................ ........... ... ... .... ... 288 6.3.2. Kinetics Over L T Shift Catalyst ............................... .. ::::::: 2S9 6.3.3. Mechanism of the Cat:J!ytic Watergas Shift Reaction ... .. .. ..... ~I)t)

6.4. Converter Dl:si'n. ... ....... ... ............................ ... ....... ..... ......... . 29 1 6.5. Hil:h lempcr~turc Shift ... ...................... ...................... .... 29.3

6.5. 1. Hightemperature Shi ft Catalyst Formulation ......... : .... :::: ::: 293 6.5.2. Diffusion Effects und Pellet SIze: .. ... .............. ................. ... 295

CDntenlS

6.5.2.1. Eliec! of Pellet Size on Activity ......... ..... , ....... , .. :, .~ 6.5.2.2. Effect of Pellet S~eon PIC5Sure Drop ................... ..

6.5.3. Reductianof HT Shift Catal)'5t .................. .. ...... .. ........ 6.5". Operation.of HT Shift Catalyst ....................... .. .. : .... .. .. 6.5.5. Poisoning and Deactivation ......... ........................... .... .... . 6.5.6. RCOli~on and Discharge ...................... .. .... .... .... ..

6.6. Low.temperalw:~Shirt ........ ~ ......................................... : ....... . 6.6. 1. General __ ........ .... .... ...................... .. ...... .. .. .. .. 6.6.2. Low Temperature Shirt Catalyst fOMulation ... ..... .. .. .. .. 6.6.3. DiffusiOll Effects and Pellet Size ............... .. .... .... .... .. 6.6.4. Reduction of LT Shift Catalyst .......................... ........ .. ..

6.6.4.1. General Considerations ........ .... .... .... .. .. .... .. .. 6.6".2. Oncethrough Reductions ............................ .. ..... . 6.6.4.3. Recycle Reduction SystCJl'\1 ........................ . : .... . : . 6.6".4 .. Commissioning Reduced catalyst ..... ... ......... .. .. ..

6.6.5. Operation and Monitoring Performance ......... ......... .. .. .. 6.6.6. Dcactir.ttion and Poisoning .......................... .................. .

6.6.6.1. Deactivation ........................................... ... ...... . 6.6.6.2. S'ufphur Poisoning ........................... ................. .. 6.6.6.3. Chloride and Other Poisons .................. .. .... .. ..

6.6.7. Oxidationl and Diseharge ..................... .. ...... .. .. ~ .... .. I"'!I':.' ,.': 6.6.8. Guard Beds ........................... ;' ........ ...... ...... .... .. .. ..

6.6.9. Econoillics of Operation ................. : ............................. ..

6.7. R'ecent Devclopments .............. ~ ......................... .. .. .. : .. .. 6.7. 1. SulphUl'-lolerant Shirt CatalysIS ................ ...... .. .. .. 6.7.2. Operalion at Very Low Steam Ratios ............ ...... .. .... .. ..

1

Chap/if 7.

Methanati'on B.B. Pearce, M.V. Twigg, C. Woodward

7.1. Introduction __ ...... .... ..... ..... .. ........................ .. ....... ... .... ........ . 7.2. Methanation in.Ammonia and Hydroge~ PlantS .................. ....... ..

7.2.1. MethanaOOnlEqulibria .. ....... ... ............ .................... ... .. .. 7.2.2. Kinetir;:s:allldIMechanisms ..... ........ .. ...... .. .. .. .. 7.2.3. Catalyst Formulation .. ....... ........................ .......... ... ...... . 7.2.4. Physic3li Properties of Methanation Catalysts . ... ........ .... .. .. 7.2.5. Catal)"tReduction .................................. ............ ......... . 7.2.6. Catalyst Poisons .......................................................... .. 7.2.7. Prediction of Catalyst Life ...................................... ........ . 7.2.8. Operating:Experience ................................. ........ .......... .

7.3. Methanatio~ .. Aydrogen Streams forOlefin Plants ..................... .

'" 296 298 302 304 306 308 J08 309 312 31' 31' 317 318 319 320 32' 324 326 328 330 JJI 335 335 ,35 338

340 341 34< 34' 352 358 359 360 362 365

36'

7.4. Substitute Natural Gas (SNG) .... .. .. .. ........................................ 368 7.4.1. Oilbased Routes to SNG ........................................ ........ 368

1.4.2. Co:1lbased Routes to SNG ............................. .. ............... 372 7.4.2.1. Lurgi CoaVSN9 Process ................... ..... ... ........... 373 7.4.2.2. HICOM CoaUSNG Proccu ..... ,.. .......................... 374 7.4.2.3. Other D~elopments........................................... 376

7.5. Heat Transfer Applica tions ......... :.. ........ ............ .. ................. ... 378 . '. . 7.5. 1. The EVA-ADAM Project ..................... ......................... 378

Chapter8.

Ammonia Synthesis l .R . lennings.S:~. Ward

' ..

8. 1. Introduction ~. : ..... ; ..... .... . ~ ..................................... :............... 384 8.2. Thermodynamics of Ammonia Synthesis...................... ............. .. ]88

8.7.1'. Theoretical Aspects .......... . , .................. :.. ...................... 388 8.2.2 .. Process Consequences . .......................... :. ....................... 390 8.2.] . The Synthesis Loop. ........................................................ ]91

8.3. Am~onia Synthesis Catalysts ................ .......... ... ...................... 393 8.] .1. The Iron Component ............... ............................ .......... ]94 8.].2. Promolers .. ............................ .... .. ...... .. .. .. ...... ]95

8.3.2.1. Structura l Promoters .......................... . .......... . ..... ]95 . 8.].2.2. Elcctronic Promotcrs .............................. .. ...... .... 398

8.4. Catalyst Reduction .................................................. .... ..... ...... 400 8.4.1. Typical Plant Procedure........................................... ..... .. 4(10 8.4,2. Prereduced Catalysts................................................. .. ... 4U:! 8.4.'3. Economics of Prereduced Catalyst .................................... 4~

8.5. Poisoning and Deactivation .................... ... .... ... ...... ........... ..... 404 8.S.1. lntroduclion.............................. ................................... 4~ 8.5.2. Temporary Poisoning in Ammonia Converters .................... 406 8.5.3. Permanent Poisoning in Ammonia Conveners............. . ....... 407

8.6. Kine tics and Mechanism ...... ........ ............................................ 409 8.6.1. Temkin Kinetics ........... .... ............................. ... ............ -109 8.6.2. Effect of Catalyst Size ..................... ............................... 411 8.6.3. Implications on Process Design ........................................ 412 8.6.4. Reaction Mcchanism...................................... ....... ......... 413

8.7. Plant Operation ............... : .......................................... .... ..... 415 8.7.1. General Considerations .......................................... .. ...... 415 8.7.2. Circulation ................................................... .. .. .. .... 418 8.7.3. HydrogeniNi lrogen Ralio ............................................... 420 8.7.4. Influence of Inert Gas Concentration and Purge Rale............ 420

Cont~nt$

8.8. Commercial Ammooia Conveners ............... ...... .. .. .. .. ...... .. 8)~.: ~~~:.~~:~~~~.~~. :::::::::::::~::::::::: : ::::::::::::::::::::::: : :: ' '. 8.8.1.2. Temperature Control al\d Heat ReCXIvery ............. . ..

8.8.2. Quench Conven er ................. : ... .. .. .. .. ...... .... ...... .. .e.8.3. Indirectly Cooled Mult ibed Converter ............................ .. 8.8.4. TubcH::uokdCOnverter ............................ . .... .. ...... ; .... : . ~;' !.:~ .. <

I ,',' ; -. t ~.;~ ;,. ~.9. ~e F~lu.~ ...... : ....... ; ... ::. :::.: .... ::::~ .... ~ ..... :;: .: ..... : ........ ~ .~ ..... ~ "1:" -' , ... ~. "- . -",." ... '. . ..~:S\ ".ro.: . ;:;.,..~ ~ ~:; :. '. ', . .::t~3t"il:":' l" C/tapter9" .~ .':' . : ', .~.:olY;l.'-> . '.: .: ;.

~ tf;"[:~~; .. " ' . ..... . " . ~ .' .; ';'.-:. ,,;,f c,', Methanol SynthesiS

1, ~. }~.\:::.~~~ : :';. ~:: '; .. ~. ' .'" ... :.: >-' .. 'I- ';'l'~'; G .W.Bridger.M.S.Spencer

'.,4"; . ': ,. ;~ . ~ h: . ~ .. ' ..

, ~... . . .

.! ~;:~Z ... :,9}. In~~uctiO~ .................. :.::.: ........ :.: .. ; .. : .:.:.:;:: ... ;:; .. ... . ; ............. .. .:. r. ~::/::~~~~ 9.2. ThennodynamicAspects ....................... .... ; .......... ...... ...... . r .... : .. ~~. i.:: 9.2.1: Meth,,:"?IFonn.alion .................................................... ..

:. .. ~;: .. ;;.:~ ... ;~'. .9.2 .. 2. ScJeCI1~11J ............. ................ ; ... ...................... : .............. . ,-.~J~I!;;':~~ ~ 9 . ]~ The Melhans ol ~~.{,hel.oopsis Proc.eu ..................... .......... .. .... .. .. ,,:,'1; .-,:;";':", 9.3. I . The ynu~s ..................................................... ..

~ i~:;-i~:., " 9.3.2. MakeupGas Composition ........... : .............. .. .... .... .. !' '! .:-: ..

9.4. Methanol Syntl!disCalalysts ............... .. ..... .. .. .. .. .... .. .. .. 9.4.1. High.prezufe Catalysts .... ............................................ . . 9.4.2. Lowpressure Calalysts ................................. .. .. .... .. ..

9.5. Select ivity and Poisons ............................... ........................... ..

9.6. Mechanisms and Kinelics ................... .... .. .. .. .... .... ........ .. 9.6.1. Reaction Mechanism .................... .. ...... .... .. ........ ..

. 9.6.2. Kinetics ..................... .. .. : .... .... .. ...... .. .. .. .. 9.7. Recent Developau:nts ........................ .. .. .. .. .. .... .. ..

Chapter 10.

Catalytic Oxidations P. D avies, R .T . OOClald-, N.H . Harbord

423 . 423

424 425 426 433 433

439

441

442 442 444

446 446 452

453 453 453

,~)

462 462 467 467

10. 1. Introduction .......................... .. .. .. .... ""'"'''''''''''''' ........... 469

, . .

~ ~

I ,

-

I I

. ' I

,

I I ! I

ContenfS

10.2. Ammonia Oxidation ............................................... ...... 10.2.1. HislOry of Nitric Acid Production ..................................

10.2.1.1 . Routes from Atmospheric Nitrogen ..................... : 10.2.1.2. Ammonia Oxidation .......... ..................... .......... .

10.2.2. Chemistry of the Modem Process ........................... ....... 10.2.3. The Chemistry of Ab$orplion .... .................................... . 10.2.4. Nitric Olide Oxidation Chemistry ............................ ..... .. 10.2.5. Ammonia Oxidation Q\C:mistry ..................................... . 1':1.2.5. Mo.d.:m Pbnu .... .. .... , .. , .. , ............ .................. ... .... ....... . 10.2.1. The BurnerGauze-PlatinumIRhodium Catalyst .............. .

10.2.1.1. Gauze Activation ............ ... ........................ .. .... : 10.2.1.2. Gane Dea.ctivation and aeaning .............. .. ........ . 10.2.1.3. MetDI Recovery ......... _ ............ ......... ................ .

10.3. Methanol Oxidation .......................................... .. ......... ....... .. 10.3. 1. lntrOOuctioll ._ ............................. ........ .. ................ . 10.3.2. TheSilver-catalysed Pr~ess ........................ ..................

10 . .3.2. 1. Silvcr-: ....... ; ;i!. . . ..... )':: ,",: !~;:i;;jl't';' I, ' ;. '''''';il~ tJ ~{ 7::;!3 .. '. .:.~.~. ,:;-~~;:~~- ;> " . ... , ~ .. ,:;':\ , ;',

~'''_:''''-: ;':;!l- W' ", . . . :'-."', ~M~'-

-

Preface The first edition of CattJiyst Handbook was published twenty yea~ ago. It contained fundamental infonnation on heterogeneous catalysIs and practical details about the catalysts ~nd ,processes employed ~n the n:oduction of hydrogen and ammoma VIa the stream refomung of

!drocarbons. It was used extensively by industrialists, and also by lose working in research and teacbing institutions, -:vha roun~ it 'Iluable because it was one of the very few easily accessible .llhoritativc sources of infonnatic;m I'bout industrial catalysis and the aeration of catalytic proCesscs. . . . . '. Since the publication of the fint edition, ther~ h~ve been significant. ivances in areas of catalysis and the technologies It covered. many o~ hieh originated from ICl's operations at Billingham, in the North East f England. As a result t~re have been numerous requests for an pdated edition incorporating!.h~ ?ew d~v.elopments . . . This seCond edition is very . different fr9J!l the first edition. The

sections concerned with the catalysts and cat!llyti!= proCesses employed in the production of synthesis gas for ,":a~ing ammonia and hydrog:" have been completely rewritten. In addition t~ these broadened n.'~n sections. there is a chapter on methanol s~nthesls as the technology ~.!n many ways similar to that used in ammoma and. hydrogen pla~ts, haVing as it docs the same initial steps for the productlon of synthesIs gas from hydrocarbons. Conversion of methanol to form~ldehyde is the larg~t outlet for methanol. and as it is often operated with methanol synthesIs on the same site it is appropriate for it to be covered here. Similarly, sulphuric acid and nitric acid pl~nts have long been assoc:i~ted with ammonia for the manufacture of mtrogen and compound fertilIZers, and so brief sections dealing with them are included.

The production of this book was truly multi-disciplinary involving specialists with extensive experience of cat.alys t d:velopment, catalyst manufacture. plant design, and plant operauon. TIns brought together a team of research workers and technologists comprising chemists, physicists. metallurgists, ch~ical. engineers and oth~rs .. The result ~s a handbook that is intended pnmanly for people working 10 the chemical industry . However. it has been written with other readers in mind,. so that it can be used as a reference source for research and teaching purposes in universities.

In addition to the named contributors, many other people at le I both at Billingham and at Chicago, USA, helped in compiling this book, and the editor would like to extend his sincere thanks to them.

I

1 I.

of

" of 8' si

,

m o

u

::1

:1

.,

. . :

'.;

f

. ; ,

fJ '.'

LV ==-- . Chapter J '. '~ """~. , ... , ,", ., ' ~,

. . ' - . ~ .' .....

Fundamental Principles .

11.1. Fundamentals of . Heterogeneous Catalysis . '. ' . '.

:', .;. r . I . ,

: , -, .~:/ ...... :;." , ) ' .. ;' ; .' . ;.-

... : .!l, ! 1 ~ . . ;4/ ' '; '1~ .. ...,."' : : .~ .

" . , .... l o{. ~: ~ . 1.. c .,

,il" .. ; ,': : ;;'. ~ '" . r , . ~':

I

I I I

I I

Ch,pttlt J. Fund,ment,J Principles

concerns the gasphase species only but thO d - . involvement of the solid in th' f IS, oes Dot ~reclude the

. . e ormanon of )nte d' speaes-mdeed, this is essential for helerog' eo A' ,rme late n us cat .... ysls.

2. The 'products of the catalysed reaction can, at least in . . obtam~d from an uncatalysed reaction under the pnncl~I~ , be There IS., therefore , no way of using catalysis to "ch:~e CO~?IIi.ons. In pracllce, however, the uncatalysed reaction may be' equdlbnum. slow or may give v:-ry different products. Immeasurably

3. Any useful catalY5t must have a high productivity TIt tonne of catalyst to "'make" many lonnes of rod us, we ex~ct 1 to look at catalysis on the atomic scale the :atal uctd Alternatively, must ~r many times at -the reactio~ site on yse reaction steps

. before catalYlic act!vity is lost. the ~talyst surface 4, The catalysed reaction st~ps take place very cI ' ." .

surface. These steps may be between ose to the solid catalyst surfaCe, or exte~sive reactio;:sa;~!~C:les ~cts:orbed.on the topmost a~omic !~yersof the catalysts. The influe:race mvol~ng the not effectively extend more than an t . d' ce of the solid does P

h. d h' . a omlc lameter into th . dSe, an t e direct involvement of atoms bel h e gas IS not usually possible; ' . o~.~~ ~opmost layers

1.1.2. The Role or Catalysis

The synthesis of ammonia and its subse 'd . . give ilIustr.ltjons of two ways in who hquentioxi at lon to mtric oxide chemical industry. IC cata ysts are essential to the

1.1.2.1. A.mmoniD synthesis Ammonia synthesis from nitrogen and h dro reaction shown in equation (J). y gen occurs by the overall

N2 +3 H1 _ 2NH . ,U) No reactIOn takes place with the reactants at a b' even with an active catalyst a temp""t fm lent temperatures, and

bt . .. ure 0 some 400"C is d d

o am commercially useful rates of ' nee e to . reaction. Nothing h,p . h

system Without a catalyst as Ih pens In t e temperatures higher than l(J()()OCe tem~r.~ture is raised until, at hydrogen molecules are dissociated' a slgm Icant proponion of the (2). For example at l430"C with Into atoms, as shown in equation 150 bar, the paTti'al pre!iSure' of ;J.lo~~re~s~re of molecular hydrogen of H

2. pressure, i.e. 0.l5 bar. Even this :oeY rogen w~uld be 0.1%. of the

fixmg nitrogen, for the reaction of h s dnot

proVide a m~chamsm for y rogen atoms With nitrogen 18

1.1. Flmd'm'ntalsofHlu,rogBn~ c.t,tysis -.:

molecules is very slow? Only above JOOOC, where tbe even more strongly.bound nitrogen molecules stan dissociating 10 atoms, equation (3), does nitrogen faation become possible.

Hz H+H (2) ., N~ ~: N+N (3)

If each nitrogen atom formed in this way gave a molecule of ammonia by subsequent reactions with hydrogen molecules or atoms, tben the rate of ammonia formation can be calwlated from the rate of reaction (3).1 A 100 m) reactor containing a 3:1 hydrogen/ni trogen mixture at a total pressure of 200 bar and a temperature of 31S()"C would give 1300 lonnesofammonia Per day. .' ... :; ", : ',\:;; . . ' ;"~ . ~owever, ~his simple analysis ignores the reverse reaction. Reaction

'_'""'. (I) is an equilibrium reaction, and ammonia synthesis is favoured by low temperatures and.higb pressures. Thermodynamic datal can be used to

., sbow tbat the partial pressure of ammonia under the conditions above cannot exceed 0.07 bar. so with a gas space velocity of 10

4 b-

I the make ',.~~:.; rate ohmmonia would be only 6 tonl}e5 day-I. '.

.'.

I ,

J

\ ~ ,

~. . The role of the' catalyst in ammonia synthesis is therefore that of making the reaction go sufficiently fast (by facilitating the dissociation

.... of molecular nitrogen) so that significant rates are obtained under conditions whcre the equilibrium conversion' is large enough to be useful ~ . (figurc 1.1). :",,',' . ' ~ :".;. ~:;..!

In Ihe on phlSe

.... ---~; Nitrogen ( ~

.1 Amm,," ,r=-H -::H~_'-:::"--':~' HydrOQen

Ammonil~ . c*.>

@Nitrogen . t On the catalyst suriate

I 'H'@'H' ,B. -lL_:c---:-~

@; hydrogen ~.: ' .. ,: -I ' . . a-,~

RgUf' 1.1. Comparison 01 'h' r.action steps in the eynthesl, of ammonl. by homogeneous and heterogenou, (catalytic] tOutlS. Sulllble calat'(lIC make the helerogeneous routl filsllnough to be useful.

When it is realized thai reaction (5), giving the more stable products. nitrogen and water I has the stronger thermodynamic driving force. Moreover, it can be seen from the stoichiometries of Ihe reactions that ruction (4) requires more oxygen than the unselective reaction. Selectivity cannot therefore be achieved by limiting the extent of oxidation.

Despite the wide range of chemica] processes in use which would not be possible without catalysis, catalysts are not quite the modern equivalents of the philosophers' slone. There are a number of transformations which would be desirable industrial processes iC only one could get them 10 go, or, in some cases, get them to go Caster or give different products. This can be demonstrated by three examples, all related to nitric acid manufacture (Figure 1.2).

The :direct oxidation of ammonia to nitric acid is theoretically possible, since r~action (6) lies well on the. righthand side. In practice,

20

I N"H'o Uncatalvsed. e.g. flame 0, . 0, NH, --=-=':-i.~'NO

Plcala/ySI

N,

O,.H,

O .. HNO~ Nitdc acid prDCeSS untalalysed ,:,1. _::~,::-., .. ' . .:., c-"~"\::~~"'~' -r.:.,t:,. ... P;"f1,'"!e .. .

;:~~~~i~~~~~~ . ?~~~~~::~~ .. Single-$ltp ructions theoreliully .' ';: '" possible,but 1'16 ulalYSls knOWn . : .. "'":;~ ,', ..

I:)" .' ;. .,

,. . ,'. ,. " ..

Figure 1.2: Various process routes. all theoreticallv possible. to make nitric acid from ~mmonia or molecular nitrogen. Suitable catelysts achieve the necessary selectivity in the conveBion of ammonia but nCHIe has yel been found for the nitrogen based reactions.

, ., .. ~. .. I. '.

Tlbl. 1,1. P'OrtlU ateps In the manufactur. 0' ammonl. ~n. _ nllur"9as n.u.c lad from

Process'tep

HVdrodesulphurlutlon

CH,SH + H. ... CH. + HlS Primary steam reforming C~+ H~O - CO.CO, + H~ Secondary steam reforming CH. + 0.+ H,O ... CO.CO,+H,

High temperature ' water-gas shift CO + H,O -CO,+H,

low temperature Watergas shift CO + H.O - COl +H2

CO.remov.1

Metna"ation CO, CO, + H. - CH. + H.O

Ammo"ia synthesis

Ammonia oxidation 2NH. + '/. 0. -2NO+3H.0

Nitric acid formation 2NO + H,O + '/10 -.2HNO,

22

Hltetoveneou. caulyst

SuJp~I~d CoIMoIAI.O +2,,0 I

flCl Catalysts 416 and 6t-1I NilAl,O,

UCI C4ilalyJ15 4&-1 46--4 46-9.nds1.3) ,

NilAJJO,

{lCI Catalysts 54-3 .nd S4~1

(lCICatalysts 15-4 end 15-5)

CulZnOlAl,~

(lCICata1vSl53.1)

None

NilAl,O, flCI Calelvstl!-3)

Fe. promoted by K, Ca, AI,O~ UCICatillvslS 35-4 and 358) F'tIRh

None

Cetlllytic properties tequired

ActiVity, tife

Activity. life

Activity,life

Activity. seleC'livi'" life . Y.

..

AC'livity, selectivity,li fe

Activity, life

Aetivity,life

ActivitV. selectivitv.life

l Sometimes; however, the failure to achieve any catalysis is a result ot the lack of reactivity of the raw materials, and then the possibility of achieving a new process pe~islS as a gleam in the catalyst ~esearcher's eye. In 1923. Lewis and Randall~ calcula ted.rr~m the."energetics of reaction (8) that slarting with water and air it should'be possible to form

:ts whIch .:.!! can actIVate nitr.ogen arc poisoned ,by water and ,o:tygen. 50 !' IS hardly

" ,' ._j_, surprising that no-one in the past 60 years has succeeded in making any .... ;1. ". . nitric acid by reaction (8). Finally "it is to be" ho~d that n'ature will not . r' !:: ::-. !, diSCl?ver a catalyst for t~is reaction , which would pc.rmit all the oxygen

f .. ;'; and part of the nitrogen or the air to 'tum t~~ -oCeans into dilute nitrie

. ':!:' .. acid", - . , : ").tI, .. :.:_"',"' ...... . !._ .... ;~. ,;: . ...... ,.~ .:.~.; ) '-' .. : ... ~;;

. ... .:i.i! i ... ~. f,; ~; ... I . :3.!h,-:.~~tuf.e. oft~~CataI~~e~?C~ :::. ~ '; ~;,.'. . ~ . The good catalyst has three cardinal virtue;;; those of activity, seit!criviry

. t :' ,1..:; -. and life. While the general meaning of these tenns is obvious, it is useful t:~-!\:.l.\ to define them a little more closely, especially with respect to plant

; .r.:,: )'7' . performance (Table 1.2) ;.C.:: '.-.~ ' . 'i.-f

J I

I I

, ,

I

Chlp/er 1. FUfld,m.n(1I1 Principl.s

a very simple approach, using single reactions. A computer model III of the catalysed reaction under working conditions, derived (rom the kinetics of the individual reaction steps. agrees well with pilln! otiservations.

1.1.5. Catalyst Selectivity

Catalyst selectivity is of equal imponance to activity in most catalytic reactions (not ammonia synthesis, of course), and smalle r changes in energy barriers than those found in ammonia sYnihesis arc sufficient to give large changes in selectivity. As an oversimplication (Chapter 10) we can write two processes in which methanol is oxidized:

CH)OH + !h0l_ HCHQ + H20 (9a) . CH)OH ..... HCHO + Hz (9b)

(/0) The uncatalys.ed" cOmbustion of I;tlclhanol with air at about 500C corresponds broadly with" reaclion (10), although some CO, together with traces of formal~ehyde, i~ formed. l1te silver catalyst used in the formaldehyde process (Chapter ~O) 'gives formaldehyde, by reactions (9.a) and (9b). The presence of the catalyst increases the formaldehyde yield from, say~O.l% to 95%; To achieve t~is the rate of reactions (9) , have to be increased rc:lativc to reaction (10) by a factor of about 2 x 114. This requires a change in the energy barrier to reaction of about 60 kJ mol-i-well within the range 01 decreases in energy barriers brought about by catalysts. A catalyst may change the rate of the undesired reaction as well as the rate of production of the wanted product. but as lon~ as the necessary change in relative rates is achieved . then the catalyst will be selective.

As energy barriers arc dominant in determining catalyst selectivities, various gcne ral patterns of selectivity can be deduced. By-products are almost always the products of reactions with higher energy barriers than those of Ihe main reaction. so accounting for their relatively slow formation. It follows that an increase in temperature usually accelerates by-product formation more than main product formation, Le. product selectivity decreases. A comparison of the lCllow-pressure methano l process with the older, high-pressure process illustrates this point.

Methanol synthesis can be described by the two overall equations (J J) and (12).

CO/C02+ H:r+CH30H (J I) COICOl + H:r+by-products (/1)

(CH., HCOOCH1 , CzH~OH, elc.)

I [~' - ' ; ' I . ". ] ! .~ 11 L

I I I

1 ,

J. J. FumhmBnt.ls of HlIltlfOf/eneous C.,./ysis

The: CulZnO/A lzOl catalyst developed by leI has a very much higher activity for reaction (J J) than the earlier ZnO/Cr201 catalys t had, and as the rate of reaction (/2) was not accelerated as much as that of reaction (J /) there was an immediate improvement in selectivity (Chapler 9). However, the high activity allowed operation at lower temperatures and pressures, so this in tum brought about a further improvement in selectivity_

The relationship between activity and sele!:tivily can be more Complex than thcse examples show_ In ammonia oxidation high selectivity depends on high activity. but for reasons which are different from those in methanol synthesis (Chapter 10). In some other reactions, especially organic oxidations. high activity is frequently associated with low selectivity. We can again take methanol oxidation as an illustration, and add the reaclion . -- .

(13) to reactions (9) and (10). Thus, the requirement for high selectivity is that reaclions (9) should be faster than reactions (10) and (/3) (Figur~ 1.4). A very active oxidation catalyst, such as platinum, has low energy barriers for all three reaclions. so sclectivty is poor and oxidation proceeds to CO , COl and HzO. Less-active oxidation catalYSis-for instance, silver or Fe/Mo oxides-are used in formaldeh yde processes, and with these catalYSIS the differences in energy barriers are high (nough 10 ensure selectivity for formaldehyde.

1_1.6. Steps in the CatalytiC Process

So far we have made the implicit assumption that process rates are controlled solely by the rates o f key reactions on the catalyst surface, but this is not always so. For example, suppose the addi tion of a new promoter to tht: iro n catalys t for ammonia synthesis greatly increased the proportion of nitrogen molecules which dissociated on colliding with the surface. If the proportion rose from I in lot' to almost all, would this give a million-fold increase in the 'elte of ammonia synthesis-and a reduction in the size of plant converter from 100m' to, say, 100 cm)? E\'cn apart from the problc ms of heal transfer and fl uid now in trying to make 1500 IOnnes of NH) per day in a reactor the size of a wine glass. the answcr is no. The separate stages of a heterogeneously catalysed gas-phasc reaction arc shown in Table I.J.

Any of these stages , if slow, may limit the overall ra te of a catalytic reaction. Distinctions are often drawn between catalysts which are "film-diffusion controlled" (i.e. limited by stages 1 andlor 7), "pore-diffusion controlled" and "reaction controlled" (i.e. limited by

27

[b)

-------------. -----r--

393

676

co, + 2H~O.

Figure 1.4. Schematic representation 01 the various energy barriers in the selective a~d unselective oKidation of methenol. A suitable choice 01 catalyst (fot example SilverI with appropriate energy barriers {al and (bl can give high selectiviry \0 formaldehyde. Numerical values are kJ mol-I.

stages 3, 4 andlor 5). However, the development of the best industrial catalysts usually leads, either by design or empiricism, to the elimination D,f any stronglylimiting constraint all catalyst perfonnance, so these Simple distinctions rarely apply in practice. The concentration profiles of reae,lanl,s and products in and around a Iypical catalyst pellet are

s~o~n In FIgure 1.5. The combined effects of diffusion and reaction give stgmficant concentration gradients, both within and outside the catalyst pellets.

. We ctln now see why a large increase in the rate of dissociation of mtrogen molecules on an ammonia synthesis catalyst would not give a

,

, , / . ~

.,

Table 1:3. Sequence 01 stages in the eatalysis of I glSphlise ruction bV' heterogeneous utalyst . : ~ _. ~

. Transport of reactants through thegas phase to the exterior of the catalyst pellet 2 Transport of reactants through tile pore-system olthe ~taly~ pe11at to a ."

catalytically-activesil8 , .. '.: : : ;. '; .! .: '.' 3 Adsorption of reactants at the Catalytically-active site 4 Chemical reections between reactan'~ attha catatytically-acti~e sitll (frequantly

several steps) 5 Desorption of products ~mthecatal~ic~lIy.ac:tiv8 site .

, . _ . . -. ' .,., . ; .. '}' ~ ''; . 6 Transport of products through the catalYst pore system from the

catalytica11y.active sitetolhe exteriorol tha catalyst pellet .'. ,0' 7 Transport of prOd~ctS lnlo the gaS" phase from 'the i"~eri?r 01 the catalyst pellet

Notes ',,, In this description the catalyst is assumed to be in the form of a pellet, The seme stages apply to othertypesofcatalyst particle (e.'iI. spheres mada by granulation. or eKlrusions), but stages 2 and 6 clearLy do not take place in catalysts made of wire gauzes or nonporous metal crystals, unless surface roughening becomes so advanced that in elf~ a pore system Is formed.

2 Transport throu~h Iha caU,lyst dore svs1;~ ~n b;, 'eiiherlhrough the gas phasa or across the interior surfUlI oIthe catalyst surface.

3 Several different catalyticalLY'active sites may be involved. Adsorption, possibly followed by reaction, mayoccur alone $lta, followed by transport alan intermediate product to a differenl site lor lunher raactions. Transport of the Intermediate product canbe eilherthrou'ilh the gas phase, il the intermediate is a stable molecule, or aCtoS!l the catalyslsurfaC8 (the 'spillover" effect}.

proportional increase in overall rate: other stages would become of more significance in controlling the process. Indeed, even with the present catalyst there is experimental evidence that under some induslrial condilions this occurs. There is a complex interaction between the relative importance of these different slages and the resulling selectivity when several products are formed.

1.1.7, Adsorption and Desorption Table 1.3 shows thai adsorption and desorption are both essential and critical stages of the overall catalytic process. In the act of adsorption, a molecule approaches the solid surface from the gas phase and is held dose 10 (or in) Ihe surface. This is different from the- collision and rebound that occurs at all solid surfaces in contact wilh gases. Thus, a molecule which stays on the surface for a time longer than that of a collision is said to be adsorbed. The removal of an adsorbed molecule is known as desorption. A distinction is drawn between adsorplion, where the adsorbed material slays on (or alleast dose to) the solid surface,

a_C2ML

RUc\Jnl

ProduC!

Reactants flow into Jjellets

Gn

I 1iItl\

Catalyst I pellet I

1'-/ I ~ J

I i

l i , '

'-.Qu fil "!. .. / --

ProduCls lIow out of plillets

Figure 1.5. Concentration profilhi of rUClanls lind products around and In I typical cat.lyst ptlilel under reacUon conditionlwhere mass-transfer contrails significant.

and absorption, where the absorbed material "soaks" into thc bulk of the solid.

Different types of adsorption are listed in Table 1.4 and are shown for carbon monoxide in Figure 1.6. We can see the differences by considering the adsorption of carbon monoxide on a-alumina. copper metal and nickel metal surfaces. There is no special bonding between CO and an a-Ah03 surface, so physical adsorption lakes place. The energy of adsorption is o f the same order as Ibat between CO molecules in liquid CO , so the physisorption is extensive at low temperatures only. Physisorplion is often import::ml in catalysis as a precursor to chemisorption. Associative chemisorplion occurs with CO on copper because the energy of interaction is greater than that of physisorption. Various adsorbed Slates have been identified, with different adsorption energies, and although the C-O bond strength is weakened no breaki ng or this bond lakes p lace on copper surfaces. This is significant in the hydrogenation of CO 10' methanol, when dissociation of the C-O bond

30

, ,

"

Table 1." Typ"ol .d~rption

Typa of edsorption

Physical . adsorption (physisorpt ion)

2 Associative chemical adsorptio n ~

I. I. FundilfJlenrels of HetBrogeneous c.telytis

ChIl"fc:t.triItiCS ; .. , ',~.

U~se!lICtivlJ. Lo .... en~gy of .dsorplion. Extenl of adsorption relaled to boiling poinl 01 guo not nature ol,olld surface. No breaking of bonds in molecules and negiiglbl. changes in bond

. 8n.rgl8$ , ... ~. ,. ~;. .:;-i .:.:.. ~:. SelllCtive.slfonglydepend. nt on both gas Ind solid surface. tflgher energies Olldsorption than those of physisorption. Bonds In the.cisorbed .

. _ ... (chemisorption]

., . ;: .. ;. ,,.;. '. rnoleeul8$lr.charigedin strength but not broken

_ . l.e.moIeeuleadi.Orbedwho le .. J~it!'.vl ~ ,_ Selec:tive. strong~ dependent on ~th DO and ' solid surflce. Higher energies of Id5QfPlio.n than those of physisorption. Bonds in the adsorbed mol8elJles are bro ken. i.e. molecule adsorbld IS twoormoremdeeularfragments :

J

"

3 Dlssoeiative chemical .dsorption !chemisorptioo]

'-0 I ,~

liml)j ~ Molecule Physical

.pproaches . , ,){jsorptlon surface

'",

"~'5!:-,!:., ," !; ;:;:,!,., .. . ',:,,~'i-', . . l !"

~' ;,"'~?:~~ ' AssoeIllioi. Oissoci.tiycv Reaction.nd

dlemisorption chemisorption:~' .bsorption

locrnsing inte/action with metal surfaee

Figure 1.6. SChl!mllle representation of the adsorption nd possible subsequent ruction, of carbon monoxide on vlrious solid surfaces.

cou ld le:ld to the fo rmation of methane and waler (Chapler 9). The removal of CO in the melhanalion reactioo (Chapler 7) does require its conversion to methane and water as in equation (/4)

(14) which is catalysed by nickel. Here dissociative chemisorption of CO occurs and the CO molecule is broken on adsorption as in equation (15) thus fa ciiilating the methanation reaction.

(15)

31

1.1.8. Catalyst Design

I~ the p rev~ous ~cdon we ~aw how the pallems of adsorption on d~[ferent solids lei:! to the chOIce of appropriate metals as catalysts f dlff~ren t reactions. The design of a catalyst covers all aspects from I~r chOice of catalytically-active material to the method of {o,m", '

rt " 1 Th" " b d"1 " "' pOl It es. IS exercIse can e etar ed, ngarous and extensive as described by Dowden,lI starting from fundamen tals to get the best catalyst for a ne~ proces~; bu~ sometimes,lh,e design of a new catalyst

ma~ be only a ml~or modificatiOn of an ex:tstmg industrial catalyst. The opttmum catalyst IS the one that provides the necessary combination of properties, including activi ty. selectivity and life, at an acceptable cost. Th~e requirements always ~ut c,onflicting. demands on the catalyst

de~lgner, and much of the deSigner s art COnSiSts of the achievement of a sUitable compromise

",

At lhe start of catalyst design a consideration of both d~sirable and undesirable reactions in the overall process leads to a choice of catalytically-active materials fo r possible catalysts, Suitable promoters are ~requently needed to gel adeq,uate performance: these may either modify the catalyst structure, so Improving stability, or enhance the ~ata lytic reactions to'give better activity or selectivity, The relative Import~nce of som,e fac,tors '2 in~uencing the activity of metal catalYSIS in some Simple reaclions IS shown I~ ,!~b!e 1.5 .. Th~ nature of the metal is

' - , .. ,, - ~ . ,

Tlblal.S. Effeeu of m~i. 1 struct~re. ~r~~~t;~'~nd ty~~ ~f ~atat on IIttivity for l some .... I .. ,ytie re~ctlon (from referlnee 12)

Reaction ._ ..... - .. Eff~of SlnIr;ture Prom011,. Type of met.l

H.+O .. 2HO VS S M C:H. + H .. CIH. VS S M Cyc/~H. + H .. C,H, VS S M c.H. + lH .. C.H .. VS S M C:H. + H:-2CH. S l Vl N. + lH .. 2NH l M l Vl

"The order of magnitude of the effect is classified as follows: I

,

I

, I I I

I I f

c.c zi~c oxide and al~mina ~ refractory suppons greatly limits the slnt~n?g ~f the co~per crYSlalht~. and it gives two further benefits. The preClpltauon of mIXed copper/nnc carbonates (Section 1.2.6) leads to the eventual formation of very small and well dispersed co crystallites. pper

Zinc, oxi~e also a~ts as an a~rbaol (or cal~lYSI ~isons: thus further extendmg catalyst hfe . The Opll~u~ proportion of COpper depends on the copper an.d support crystallite sues.1S A typical reduced IT shift catalyst contams about 20% Cu by volume, with both Cu and ZnO crystallites about 5 nm in size. Jt.: catalyst with about 40% Cu by volume wou.ld need the support crystallites to be

the use of a fusion method is the way in which the iron oxide, Fe304, is reduced by hydrogen. On reduction iron metal is formed as a porous solid of the same overall dimensions as the original iron oxide. For pure Fe304 the removal of oxygen corresponds to a weight loss of 28%, but because melallic iron is deQser than Fe304 the resulting theoretical porosity is 52%. In practice, ammonia synthesis catalysts contain small amounts of other solids, so the weight loss and porosity achieved on reduction are less than theoretical. ,.

Ammonia synthesis catalysts are made" by the fusion (in an electric furnace at -16OO"C) of magnetite, Fe304, of a suitably pure grade, together with the promoters, typically calcium, potassium and alumina, in a "triply-promoted" catalyst. The appropriate purity of the raw materials is of critical importance in fused catalysIS because thcre is no subsequent stage in catalyst manufacture during which poisons can be removed (as can be done' with precipitated catalysIS), The only opportunity for the ' ~c;.'!l,o"al of poisons is during catalyst activation (Section 1.4), Conve'nlional forming techniques (pelleting, extrusion and granulation) are nOI practical with fused catalysts, so the common practice is cooling of the melt, followed by crushing and size grading, to give particles of the required dimension ranges (Chapter 8). A flow sheet for the process is shown in Figure 1.7. Very good mixing of the components is attained in the fusion stage , but some segregation, on a micro-scale, occurs in the'cooled solid. n This separation of components is not sufficient to affect the final cat .. lysts. since funher migration to the optimum distribution for a~tivi ty occurs on activationlreduction,21

Figure 1,7, Flow sheet for II plant to make an ammonium synlhnis catatyst by the fusion route, .

'.2. C8tlJ/yst M,nu(actlJTIJ

Rane~ nickel21 is. prepared as a nickel-aluminium .. alloy , by conyentJOnal metallurgical techniques. The aluminium is removed with

aq~eoussod!um hydroxide. to leave porous nickel (of high surface area) which ~ntalns a small amount of stabilizing alumina .. formed in the '

~~tlra ctlon proc~. The resulting catalyst is of high area (e,g. 100 m~ g ) and, actIVIty, and is useful for mild,'" low-temperature '

hyd~ogenatlons, but it is ~nsitive to poisons and is rapidly deactivated at higher temperalUres. Raney cobalt and Raney copper catalysIS can

-'. also be made by the same techniques. ' ,

, ' . '~ .~:" '." '~:~'; ', : , ' :, "~;~'; _~~;~',~t~/~~':~:,'; ',I~~.!~,~,~ : ~ ~;/:p,r~,:::'~ .. ~:,~~~, 1.1,4, Wet MclhodsotCalalyst Manuradure ", ' j ~,;; ~ " ~ ' : ;-:' ,~'>~ '; ~.'J'_, ' ~ Most indu~t~al ~lalYSts are ~~de (Fi~:;'e i~)~iI~~:-by ~recipila;ion, :i:'~

~hen act~ve phase' and sUPP'?rt are made iogether, or by the Impregnation of a,n active phase on to a preformed SUpport ,I6--19. 21 The p.rocess used depends on many. factors. such as the chemistry of the ~talyst compo.nents, and their possible precursors, 'the concentrations of

-, dJfre~e,nt c~mP:OnenlS required. : physical strength required, reaction condlll~ns ,pc. cal~lyst in' use ' ,a'n~ .,the ~eed 'and ease' ,~f .removing ' contammants. HIlS usually easier to get a high concentration ' of the

r catalyt ically active phase by precipitation than, b'y ' impregnation p~ocesses, but ,t~e developmen,t of, ~~equate'! str~ngt~ .' can '~ more diffiCUlt., SomctUT':cs the p~cc~s a~c ,oolllbi~ed, ~~ ~hen, a prt:cipitated catal!st 15 With potasslUJ;11 hydroxide ' solution to give the required I ' .. , ",' . , ' . '

, '.: ' .~,._ ,,. , ' :~i' ';' .:,: ~ rj ' " '!', :,. ' ,.r::.;;..fji-'!;r.,:~-:t., ,--; ~

,

i 'I I

Ch.pter I. FundlJmtlnr.1 P,indpl,.

1.2.S. Fundamen~orPrKipltatlon Processes Precipitated C3talystll are generally prepared by rapid miXing of. concentrated solutions of melai salts, and a typical nowsheet is shown in Figure 1.9. The product precipitates in a finely divided form of high surface area. Precipitated mixed hydroxides or carbonates are most frequently prepared. The reaction , .

2Ni(NOlh + 2NaOH + Na1C03 - Nh(OHhC03 + ~NaN03 (I8) is typical, but the precipitates arc usually non-stoichiometric and often amorphous. After filtration and washing tbe precipitate is dried and heated to decompose the hydroxides/carbonates to the corresponding : . . oxides. The final size of CT}'stal lj tes present in precipitated catalyst are , tyrically in the range 3-15 nm, while overall surface areucan be 50-200 ., . m g- I or more, 1'hese ~alues c!ln change markedly in use (Section 1.4); but the actIvity of precipitated catalyst is nevertheless usually high. . .

The main aims in the use o f a precipitation proi:es;!' f~~ . catalyst , .. manufacture are the inti.mate mi;{ing ~f the 'catalyst components and the .~< . formation of very small particles to give a high surfaci area. 11'e . '~~ . neceMary degree of mixing can be achieved either bY,lhe onnalion 'o{ ,-very small crystallites, in close proximity, of Ibe different components or by the formatio n of mixed crystallites ' containing"~'.the ealalyst constituents. Hydroxides, carbonates or basic' "Carbonates ' are the :.l favoured precipitated intermediates. for the foll~~i;lg' rc~ris: .... ~:.. '. ':.'.~ .' . . ~ . . ,' _n .. . ,' .... .. . ':,._. ' ....

1. The solubili ties of these salts of transition metals and' other catalytic. components are very low. Conseqently, very high supersaturations. leading to very small precipitate particlc si1es, can be reached.

2. The solubiliti~s of the precu;sors: typi~~ly ;;'etal nit;ate~ :i~d sOdium ... . hydroxide or carbonate, are highj' sO concentrated solutions can be used, again giving high supersaturatio ns.

3. Hydroxides and carbonates are readily. decomposed , by heat, to oxides of high area without leaving catalyst poisons as, for example, sulphur residues from the calcination of sulphates.

, -4. Many mixed hydroxides, carbonates and hydroxycarbonates are '

known, so there is a good chance of getting a mixed compound of the required composition for given components.

S. Environmental difficulties arising from the calcination of hydroxides and carbonates arc minimal.

38

"

' ~ ,

100

~ . ~ BO

160 & I 40 : 20

ln , .~. ~.~ ,; ".

.... , .. '.

. .' : ~ .. , ,'.' :." .... : . .:

""'\ Zn ' .::" --:';;;- , , . '. . ! ,...-.'1-:

Ai . , " . 1 : , .';.' ~ ..

. ....... ,'.. l.

.: .~. ..-., ,

0 4 5 . 6 . 1 .... , 9 .. ~O ... ,. p!-4olpreelpitation . ~

. . . f eopPerhinc: oldde CIItalysl$ Figure 1.10. Variation with pH of Ihl propertles.o I of pH , ffeels com-

re ared by pre

Ch8prer 1. Fund,mertr8' P,;rrciple,

examples are given where impiegnation techniques have marked advantages over other procedures. The flowsheet for a plant for making an impregnated catalyst is shown in Figure 1.12. The great virtue of impregnation processes is the separation of the making of the active phase and the support phase, which is clearly not possible in the case of precipitated catalyst manufacture. The support is normally a porous refractory oxide, usually fired at a high temperature to give stability. High-area carbons are used for some impregnated catalysts, especially metals of the platinum group. The disadvantage of impregnation lies in the limited amount of material which ca~ be incorporated in a support by an impregnation stage. Multiple impregnation/dryinwfiring cycles can be used to increase the levels of active components, but only at a significant cost in catalyst manufa~ure.

Finished catalyst pellets

Support prltCUr50rs and cements

Hopper

Pdettlng . L,,,,~ machine ' .. PeUet Drier

pre-treatment Mellt sah Promoter solution solution

Calciner

Figure 1.t2. Flow sheet lor. plant lor the manulacture 01 catalysts by the Impregn-Ilion of pre-formed supports with. compound of the active metal.

In the absence of specific interactions between the preformed support and the ?lmponents of the impregnating solution, the impregnation process s1ages can be described quite simply. A solution is made up containing the component to be put on the catalyst. In the next stage either the support is dipped into a tank of this solution or the solution is sprayed on (or added to) the support-lhe latter procedure is sometimes

) ,

,

1.2. C,I"",t M.nufat;/utl

' .... '

... o -0 ..

~ <

: ~

~ 0 ~

'a U

t ;; > ..

3 E g E .. !! ; <

g <

, ,

it

39

-, :

Ch.pt.r r. Fund.men(.' FrlnCip"J

1.1.6. t:lllal),st ~1anuracture b)' Predpitation Processes An example of the complexity of development work with precipilated catalysts can be taken from the ICI work on the copperlzindalumina catalysts for methanol synthesis.u .lS Earlier work bad shown thai copper catalysts were active but . unstable, so activity was rapidly lost. The copper particles in these catalysts are the essenlial catalytic material, while the olher components are the stabilizers. In earlier c:ltalysts made by conventional batcb methods these stabilizers were not effective because the particles were nOI small enough and were nOI uniformly mixed. In the batch methodinvolving the addition of alkali to an acid soluliqn, the catalyst which is obtained at the beginning of the precipitation is formed under acid conditions, whereas laler, as more alkali is added, it is produced under alkaline conditions. This affects both the composition of the catalyst and the size of the particles (Figure 1.10). The catalyst produced under the acid conditions at the beginning of the precipiiation is rich in aluminium and deficient in zinc, while at the end. und~r. alkaline Conditions, the catalyst is low in copper. The catalyst particle s!ze is lilso affected by the pH: under acidic and alkaline conditions large particles are produced, but under neutral conditions they are much smaller. Th~ best catalyst is obtained by precipitation at around pH 7. and this was achieved "by a precipitation procedure in which the acidic and alkaline solutions were mixed continuously.

Very fine precipilales can, however, cause severe problems in the later stages of processing and much skill and empirical development is needed to reach the optimum process. If the pr~cipitate is too fine it may pass through the filter or block the filter pores. Even when a satisfactory filter cake is made. j)e removal of sodium and other unwanted ions can be dirficult. Salt solutions held in the filter cake are readily displaced, but salts can be adsorbed on the high surface area of a fine precipitate . Washing with water may not the n be enough 10 reduce sodium levels to the specification for the final catalyst, and special techniques have to be used.

. The precipitation process outlined above represents the simplest form in which it is used, many catalyst production routes involve more complex or addi tional steps. The additio,:,.of a solid component either during o r just after precipitation is common. Ageing periods after precipilation, or after washing out unwanted materials, may be needed to achieve the desired chemical or physical state of the catalYSI intermediate. This may not always be advantageous because some crystal growth may occur in a precipitate in contact with the filtrate. Crystallization, especially from concentrated solutions, is one of the less-well understood chemical processes. so scale-up (Figure 1.11) from laboratoty experimt:nts has 10 bc t;"ne with care.

40

. ~]

l~ ~ LJ

,.2. C.f,I,..$( M,nuf,cture

T.bl. I.G. Catalysts mad. by imprqnetionprocesM'

Catalyst

Supponed platinum-oroup metat catalysis .

2 Catatysts IOf ~htemperature processing .

J C.talyslswithan uneven distrl buliol"l 01 COlll9Onel"lts, 8.g. active phase on eKtefiOf

4 Alkalidopedc;atalysts

5 Metall{eotiteaJtafysts

G Supponed mohe-n-5elt catalysts

Renon for use of Imprelll"lation f .. techniquu ., :.

Good melal dispenion .nd no loss 01 metal within support phase! Support can be $!abililed to ',:sist reaction conditions before active phlSl incorpofllted ' ., ... Plecipitation inethods give uniform tat,lysls, , t lelSton. macro-5CilI,; impregnation can concal"llflte c:cmponents where requIred ADcall ~uld be rem;;:;ed at ;""shing stages of precipitotlon route Usu.lIymore convenient to edd . melal by ion-exchatlge .flerzeotite synthesis ratherthan include metal In leolite SYl"lthesil Usually more convenient to add motten nil phasa (e.g. V20~~O. fOf sulphuric acid eatalystsllO preformed support, e.g. silia!

-' -'.

described as the "incipient wetness" technique. in both methods the take. up of the solution is governed by the porosity ofthe. support, ~ the level of active component i ncorporat~d in the catalyst IS a funcllon of the solution concentration and support porosity. After absorpt ion of the solution into Ihe porc S}'5 tem of the support, a drying stage is used to remove water. This bas to be carri~d out so that Ihe impregnated componcnt rcmains within the support pore system and docs not migrate to the e);terior surface of the support. If this sUlge is done correctly, the support then has crystalli tes o( the impregnated component. typically a metal ni trate , in the iOlerstices of the pore system. Most impregnatcd catalysts are calcined in air after drying, thus converting the soluble saitta insoluble oxide. Calcination can also have othcr cffects; for instance, thc firing of a potashdoped catalyst fo r naphtha-stcam rcforming ~clps to :'~X H the pOla~sium by r.eact.i~n wi~ thc rcfrolctory phases to gtve kalslhtc, a potassium alumlnoslhcate. This is resistant to hydrolysis by steam, and without this "anchor" the polash would be lost rapidly as volatile potassium hydroxide in sleam during use.

Interactions betwcen impregnating solution and support are commonplace. and indeed uscful. in .the impregnation process. Three

43

.

types of interaction can be recognized although the boundaries of the classification are somewhat arbitrary.

1. Specific adsorption. If the impregnating component (e.g. a metal ion) is more strongly adsorbed on Ihe support surface than the molecules of the solvent (usu~lIy water). metal ions are removed from the impregnating solution as it passes into the support, with the result that the impregnated :'tI~lal is concentrated in the OUler layer of the support pellet. This "shell" catalyst can ensure that maximum use is mad", of the catalytic metal in reactions that are 5t:'ongly limjted by pon .. j i(fusion (Section 1.1). Further treatment with water and/or other solutions can give different distributions. For example, in coking or poisoning c.onditions there can be advantages in having the active phase in a layer belo~ the pellet surface.

2. Some high-area solids (zeolites form the most important class) have significant' itJn-ucf.ange C:1pacity. Thus, metal ions in the impregnating solution will exchange with ions already present in the support suiface befor~ imp.regnalion: Since this gives almost an atom-by-atom depositiQn over the surface, the metal distribution, initially at least, is very goad: As with adsorption, the amount of metal taken up by the support is not simply determined by concentration and support porosity, but is a complex function of

i on~xchange equilibria and ' diffusio'.l rates in the liquid phase.21 lon-exchange, like less-spe

j

I ,-

I I

0' I , ,

1-

I-

Figurt 1.15. A typical m'aclline used 10 extrude eaIBI'f$ts.

p

ell I--[ J [~ [rJ L[J W [-2 [ J

0

" LJ i. [-:J r

u ~ " ~ ~ ~

I "

J , ,

~ !! i , ,

, ,

I Figure 1.16. A typical machine used!" form calalysi particles bV granulation.

these conditions only the outside layer of the catalyst particle is effectively used, so the maximum activity from a reactor is obtained with catalyst partides of maximum superficial area. As smaller particles can be made more easily by granul21ion or extrusion than by pellcting. these are thc preferred methods of forming. Cross-sectional shapes of longer periphery than a circle (e.g. a "clover lear') can give even larger superficial areas on extruded catalysts. and, these have been used in some hydrodesulphurization catalysts. Occasionally just a decrease in pellet size ean give a sufficient increase in activity under pore-diffusion limited conditions. With the rise in temperature moving down a HT watergas shift reactor the process moves from almost pure reaction contra lto substantial pore-diffusion control. Jl Consequently, there are advantages in tbe use of large. dense pellets in tbe inlet regions of the reactor and small pellets in the exit region of the reactor. 2~

47

"

- _ ........ _ ... _ .. _ .. .... -... -

Table 1.7. Relative meriU of differ,nt types of ~tal'(lt particle

Type 01 particle

Pellet

ElCtruslons '

Granules

Proplrties

Cyllndrieat 'shape. Denser and stronger than elrtrusions at granulas. COlwenlent size range from about S mm to abou t 20 mm. RaschIg rings as well os cylinders can bit made. but otharwlselimited in shape. Two stages 01 .comPliction olton needed UsulIlIy lo"o:;rregul", ev\in.drical shllpa. but othar shapes of largo. e"telMI surface lIree le.g. cloverleal- ) pouible. Lessdense than pallets. Convenient cizelrom about 1 mm diameter upwards. Also possible 10 make hollow matrices le.g. ear exhaust ClIllilys ts) Sphericllshape only. LIWI dense than pelleo.

. Corlvenienl slle from ,bout 2 mm diamellr upwards

The choice between extrusion a~(l granulation- for ' small calalyst panicles depends primarily On. t~e .nai~re of the powder precursor: not a ll powders can be either extru~e~ or granulate" satisfactorily, even with the help of special addi!iy-es (e.g. plasticizers or cements). The requirements of the forming process !ead to constraints orrearlier stages of the process. IS Thus, in the pelleting of methanol synthesiS catalystsl6 it is important to control the amount of hydroxide and carbonate remaining fro m the calcination stage. An 'excess weakens the pellets after reduction, whereas :In inadequate proportion produces fragi le pellets because the hydroxide and carbonate have a role in binding the oxides under prcssure in the pclleting machine.

.,.3. Catalyst II Testing 1.3.1. Introduction "Commit your blunders on a small scale and make your profits on a large scale." This principle, given by Baeke land2~ in his Perkin Medal add ress in 1916, summarizes the whole basis of catalyst testing. In catalyst testing there are two obvious features which can be emphasized.

I~ ~., ), . ..

,

" " 7 ,' . "

" ,

"

. ,.3. C.tlIlysr Testing

The first is that .some simulat ion of what will (or could) happen to the catalyst is required for each test, but it is not possible to do all of the tests simultaneously. The results from various tests are combined in order to predict- full-scale behaviour. The second point to be emphasized ts that all of these figures have to be ~right ", or very nearly '0,

A ll of the properties which Can affect catalyst activity, selectivity and life need to be investigated, no mauerwhether a novel process or a new catalyst for an existing prOCess .is being examined--or . indeed, for ensuring the cOnstant quali ty of a production catalysC A range of experimental teChniques is uscd in the small-scale simulation of the different aspects of the full scale process, but the equipment rarely resembles a full seale reactor. Baekeland stated that the principle quoted above "should guide everybody who enters into a new chemical entcrprise. even if it taxes the patience of some men who cannot conceive that one single apparent ly minor detail in a chemical process may upset all the good points and lead to ruin".2Y

. ~ '. -, !. ,. ,

1.3.2. Chemical and Physical Properties

Many of the chemical ilnd physical properties of a catalyst innuence the perfonnancc: of the catalyst in a full-scale reactor. A wide range of properties, classified as bllik chemical properlies, surface chemical properries and physicQI proper/ies, is detcrmincd, usually o n a routine basis, as part of catalys t testing. In contrast with the various techniques dcveloped for Ihe assessment of catalyst performance, mOSt of thc analytical procedures used 10 dctermine other catalyst propert ies arc standard tcchniques which require little modification for thcse ilpplications.

1.3.3. Bulk Chemical Propc!rlies

Techniques used to detennine these properties are given in Table 1.8. Thc first three mcthods-elemcntal analysis, Xray diffraction and clectron microscopy-arc of widest ilpplication in the determination of Ihe chcmical nature of a catalyst. Thc advances in electron microscopy over the past decade hilve made this potentially the most powe rful technique available for uncovering the chemical nature of catalys ts.:W1 Both the clement analysis and crystal structure of particles identificd in the field can be determined.3 1 The other experimentallechniques given in Table 1.8 lend to be used in special cases only. Thus, in an investigation of a poisoned cata lyst, GClMS can identify poisons present, and electron probe analys is or radiography will give the distribution ofthe poison across the catalyst structure.

49

Table 1.8. Techniques used for bulk chemical properties of c.atalysts

Tedlnique

Element analysis (qualitative and quantitative) Xray diffraction

Eleetron microscopy and associated X'ray fluorescence and elactron diffraction Microprobe analysis

Radiography with radioactive isotopes . Nuclear magnetic resonance (NMR) Mossbauer spectroscopy

~sible ) spect~scopy UV Extended Xray adsorption fine-structure analysis (E){AFS) Thermal analysis IOTA, TGA)

Temperatureprogrammed reduction (TPR) Combined gas chromatography and mass Spectrometry IGCJMS)

1.3.4. Surface Chemical Properties

Properties detarmined

Bulk elemenla! composition

Cry$talline phases prennt, crystallite sizes Panicle shapes and sizes, panicle compOsitions, panicle crystal structures Variation in composition across pellet Di$tribution of radioactive component Chemical environmentof element Chemical environmant ofelamenl

Types alchemical band present

Types of chemical bond present

Phasa chang as. weight changes on heeting Size and temperature range of reduction stages Analysis 0' volatile components

As heterogenous catalysis is a su rface phenomenon, the detennination of surface properties (Table 1.9) p!ays a large pan in catalyst characterization. Some techniques (e.g. LEEO) can be used only with single crystal surfaces, and nOl with practical catalysts, and are not included in Table 1.9 (for further details see, for example, reference 32). No single technique can give a full description of the nature of the surface , so the use of several techniques is essential. As differenl equipment is used under various conditions, it is also essential to ensure that it is really the same catalyst surface which is being examined in each experiment. Some of the modern techniques of surface physics, listed at

'"

I < eo ~. ! . , , .

I b , [--! -

Table 1.9. Som,techniques used for determ[nltionofAlrflce properties of Clltatyru

Tedmique

Photoelectron spectroscopy IUPS,XPS) Auger spectroscopy Secondary ion mass spectrometry (SIMS) Temperatureprogrammed desorption, nash desorption Ph'(5isorption of gasesle.g. N2) Chemisorption 01 CO, H2'or O2 Surface reaction of N20

Chemisorption of bases (e.g. NH,. pyridine) Chemisorption of acidic gases le.g.C011

~si~le ) ~pecuo~~py. UV Highresolution electron energy loss spectroscopy (HREElS) Extended X-ray absorption fine-structure analysis IEXAfS) Wo.k function detetmin3tion NMR Electron microscopy

Isotopic labelling of reagents or catalysts

Properties determined

Chemiealidentityof surface leyers . .. .

Chemical identity of surface layers Chemieal idonlityof surface layers

Chemical Identity of adsorbed surface speciBl: ~ .; .. .. . ' . ' Total surfaceerea ~:: .. : Surface area of metal components Surface area of metal components Surface concentralion of acidic sites

Surface coru:entriiltion of basic sites

Types of chemical bond present

Type of chemical bond prese~t

Atomie struetur. of surfaces and adsorbates Surface ionization Chemicalenvironmenl of element Chemical identity lind strueture of surface layers Chemiealorigin of adsorbed species

the lop of Table 1.9. provide a ..... ider range of chemical s~rface information. especially concerning minor compone".ts such as.polso.ns, than could be obtained by the older, selectIVe chemisorption mcthods. Jl-36

For example in recent work21 on the surface composition and topography of' both unreduced and reduced ammonia synthesis catalySIS. scanning Auger electron spectroscopy, X-ray photoelectron spectroscopy and scanning electron microscopy were used. The

SI

,.

Ch'pter 1. Fund.m.nr./ Principles

application of these techniques has given much detailed information on the microstructure of the reduced iron component. and the distribution. (In a micro-scale. of the catal~"S t promoters. Nevertheless, Emmell and b.i~ ~ ........ ..... y1:~C'5. r \Io1tb ~ fe\\. So,)m~wh.at primiti\'e tcchruq~ .a\ail3b~ $c)m~ 50 \'~an a20. ach.ie\~d a broad l.LDdemandiog both of the nature .. ,j lh~ rr .. ~m()l~d~ir,,'n 3mmoni.1 ~~T1lh~sis catal~"St. and of hOloll il ..... orked . ..... hi~h slill holds looay.

1.3.5. Physkal Properties

A reJali\~ly small range of techniques is used 10 monilor physical propen ieS"311 of calaly.SIS. and these are lisled in Table 1.10. The slungth of CD/Dlyst pilfliclu is needed to a~ss the possiblity of failure in use. Apan from chemical causes such as pellet rupture upon excessive coking. two .different physical stresses are imposed on industrial catalysts: first, in the loading of a catalyst into a reactor (Chapter 3) and second , when the catalys t bed is installed in the reactor. Even under steady operating c.Pnditions catalyst particles are stressed by the weight of the catalyst bed abev"e , and by the pressure drop across the catalyst bed due to gas flow. Catalyst break-up during loading is simulaled (Figure 1.17) by the use o f a standard tumbling test to determine attri tion loss,J9 but all estimates a re done by comparison_with standard

Tabl. 1.10. S.lmed t.ehniques used for detetminetion of physiul properties of c.talysts

Tec:hnique

Pellet crushing Peliet lUmbling Phy,ilorption isotherms Porosimetry Controlled packing

Inert gas flow through catalyst bed

Single pellet reactor Nonsteady slate gas flow Ihr01Jgh CII"yst bed

5'

Properties determined

Compression strenglh Antilion toss Textural properties Pore size distribution Bulk density and packing ch8t8C1eristics Auid flow prope:rtia, including pressure drop POIIdiffusion coefficlerlt Dif/usion characteristics

" ."' ; .. ' ,.3. C1111/y:Jt Te.ring

53

Ch.pter 1. Fund.m.fII.,Prlncip/es

materials. since there is no direct relationship between the test and losses in plants. Similarly, comparative measurements are used in the crushing tests which simulate static loads in a reactor. Obviously, the way in which a crushing strength is determined depends on Ihe shape of the catalyst particle. Spheres and long extrudated cylinders can be crushed only along a diameter, but pellets can be crushed either along the axis or along a diameter of the cylinder. The lauer, the horizontal crushing strength. conelates better with behaviour in the plant.

Many of the physical properties required can be described as proper(i~s of the luture of the catalyst or properties (e.g. pore diffusion coefficients) which are determined' by catalyst texture. Average properties are generally used (a complete description of the texture of a catalyst pellet is not possible): .. ' (a) specific surface area (i.e~lotal accessible area); (b) specific porosity (i.e. total ilccessible pore volume); (c) pore size distribution (in dist ribution of pore volume as a function of

pore radius); .. . ' . ~ .... - ,. ' . .' ,,_" . ' _f" ~':. ... t.: .-' -(d)'rneanpore radi~a~d ,-' (e) particle ,i.e di_,riibu,lion .

Pores or mesopores of physical" obtaining ' a. combined " '

. " .~ . ~'.''->'~'" ''

(>30-35 nm), mi~ropOi-eS (

,

Chllpter 1. FundllmllnflJ/ Pr;ncip!t,

The flow sheet in Figure 1.19 shuws the stages of catalyst development from exploratory work through to catalyst production. Few research programmes follow the iull sequence: most speculative research projects ,fail and many development projects start from a considerable body of knowledge. SlUdiM of the reaction mechanism (considered below) can be carried out al Dn)' $Iage. For example, mechanistic studies may be used to interpret unexpected results in the coarse screening at the start of catalyst development, but work is still being done on the mechanism of ammonia s)'nthesis some 70 years after the successful development of a commerci~1 C3ttllyst.

1.3.7. Coarse Laboralory5a'ttning At each stage of development v~rious catalyst formulations are rejected as unsuftable, and this rejected fraction is largest at the initial laboratory stage, which may be considered "coarse screening". The catalysts tested here are either speculative catalysts for an existing, possibly even well-established, il)dustrial process, or a range of plausible catalysts for a new process. The1'rimary aim in coarse testing is the Separation of the promising catalysts from the useless catalysts. It is important at this stage that, as (ar as possible, no promising form ulation is rejected. The experimental requirements to achieve eo,rse screening at a reasonable cost are given in Table 1.1 t: Although the criteria for aceeptance at this stage are coarse, i.e. sel'!liquilhiitative or even qualitative. successful coarse testing required much experimental finesse.

Tabla 1.11 . Experimental require!tlenls lor eo_rle , crunlng 01 ubltysts

Simple criteria le.g. minimum tctlvity Or selectivity to desired productlto reject uselen formulations

2 Rapid t~sts. pref.rablywith sever.1 call1ysts simultaneously 3 Small cafalyst samples only nudild lor eveluation 4 Determination of intrinsiccatalys\ performance 5 Able 10 cope with as wide a v .... i.tyolc.1l1ysls possible

The requirements in Table 1.11 are most easily achieved with micro reactor systems. A modern installation with full computer control of operation and analysis is show