UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Heterogeneous catalysis of glucose mutarotation by alumina Dunstan, T. D. 1983

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1983_A1 D86.pdf [ 12.34MB ]
Metadata
JSON: 831-1.0060700.json
JSON-LD: 831-1.0060700-ld.json
RDF/XML (Pretty): 831-1.0060700-rdf.xml
RDF/JSON: 831-1.0060700-rdf.json
Turtle: 831-1.0060700-turtle.txt
N-Triples: 831-1.0060700-rdf-ntriples.txt
Original Record: 831-1.0060700-source.json
Full Text
831-1.0060700-fulltext.txt
Citation
831-1.0060700.ris

Full Text

HETEROGENEOUS CATALYSIS OF GLUCOSE MUTAROTATION BY ALUMINA by THANTHIRIMUDALIGE DON JOHN DUNSTAN B.Sc. (Hons. ) , U n i v e r s i t y of S r i Lanka, Colombo, 1973 M . S c , Da lhous ie U n i v e r s i t y , H a l i f a x , 1977 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept t h i s t h e s i s as conforming to the r equ i r ed s tandard THE UNIVERSITY OF BRITISH COLUMBIA August 1983 © Thanth i r imuda l i ge Don John Dunstan, 1983 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department of Clxg. <ufi.fr if The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date October s-, mi DE-6 (3/81) i i ABSTRACT Supervisor: Dr. R.E. Pincock The k i n e t i c s and mechanism of the heterogeneous c a t a l y s i s of the mutarotation of glucose by alumina have been investigated. Various types of aluminas held i n suspension, i n dimethyl sulfoxide, were used. At 25.0°C, the f i r s t order k i n e t i c p l o t s for mutarotation by alumina neutral (for t h i n layer chromatography; y-form) were curved due, f i r s t , to r e l a t i v e l y slow adsorption of glucose on alumina and, second, to progressive deactivation of the c a t a l y s t . P a r t i a l l y deactivated catalysts produced l i n e a r f i r s t order plots over three h a l f l i v e s and hence, glucose muta-ro t a t i o n by alumina i s a f i r s t order react i o n . Further, there were <1% side products formed during the surface reaction. Dehydration of the c a t a l y s t at low temperatures ( i . e . upto 600°C) decreased the c a t a l y t i c a c t i v i t y , unlike the other reactions studied on alumina surfaces. On further dehydration at higher temperatures the c a t a l y t i c a c t i v i t y increased, and the a c t i v i t y per unit area of a-alumina (= 3.6 x 10 -1 -2 sec m ) formed at 1250°C was about 26 times that of the standard alumina n e u t r a l . High c a t a l y t i c a c t i v i t y f o r the ct-form of alumina compared with the y-form was previously v i r t u a l l y unknown. Further, t h i s a-alumina did not deactivate during c a t a l y s i s and produced l i n e a r f i r s t order pl o t s over three h a l f l i v e s . -4 Adsorption studies showed the presence of (0.70 ± 0.02) x 10 moles of i r r e v e r s i b l e adsorption s i t e s on the surface of a gram of alumina n e u t r a l . The isotherm for adsorption of glucose on alumina neutral showed only mono-layer adsorption. The Langmuir pl o t f or r e v e r s i b l e adsorption of glucose on the surface showed the presence of two types of r e v e r s i b l e adsorption I l l -4 s i t e s ; (1.0 ± 0.1) x 10 moles of strong adsorption s i t e s with an equi-2 -1 librium constant for adsorption = (8.2 ± 1.2) x 10 l i t r e mole , and -4 (1.4 ± 0.2) x 10 moles of weak adsorption sit e s with an equilibrium constant for adsorption = 44 ± 3 l i t r e mole The study of the vari a t i o n of i n i t i a l rate with concentration of a-D-glucose showed that only the weak adsorption s i t e s are c a t a l y t i c a l l y active. Hence the active 13 2 s i t e density on alumina neutral was obtained as (5.4 ± 0.6) x 10 sites/cm . -3 The turnover number of a c a t a l y t i c s i t e was determined to be 2 x 10 molecules/site/sec. This i s one of the highest turnover numbers for a reaction catalyzed by an alumina surface. The observed f i r s t order rate constant for the surface reaction, ^ ^ k^ &2 G + C T~*" G -C ~7~*~ a.r. a + c a - a . 6 T B K2 KL, 1 (k^ + k^) [Catalytic Sites] was shown to be k = , where — = if . The o b s k2/k + [Glucose] k2 l c a t a l y t i c constant (k^ + k~) for the interconversion of a-D-glucose (ff Q) -3 -1 and 8-D-glucose (Ga) on the surface was determined to be 5 x 10 sec p Comparison with the c a t a l y t i c constant for mutarotation i n pure water -4 -1 (= 4 x 10 sec ) showed that the alumina surface offers a better medium for mutarotation than water. Further, the a c t i v i t y of the c a t a l y t i c s i t e s on alumina neutral i s about 9 times that of strong acids i n water. The in h i b i t o r y effects of 'neutral' molecules (water, methanol, methyl glucoside, i n o s i t o l etc.) indicate that the glucose adsorption s i t e s on alumina neutral are r e l a t i v e l y s p e c i f i c for adsorption of polyhydroxy compounds. In addition, aldehyde (e.g. hexanal) groups seem to interact p r e f e r e n t i a l l y with the c a t a l y t i c s i t e s on the alumina surface. Studies with a c i d i c (carbon dioxide) and basic (e.g. pyridine, i v n-butylamine) i n h i b i t o r molecules suggest that the c a t a l y t i c a c t i v i t y of alumina towards glucose mutarotation i s due to the presence of basic oxide ions and weak Bronsted acid s i t e s on the surface. About 85% of the a c t i v i t y of a-alumina formed at 1250°C i s due to these basic s i t e s , while the weak Bronsted acid s i t e s give r i s e to about 90% of the a c t i v i t y of alumina neutral. The observed high c a t a l y t i c a c t i v i t y of these weak Brbnsted acid s i t e s i s probably due to the s t a b i l i z a t i o n of the t r a n s i t i o n state leading to a c y c l i c intermediate by the polar alumina surface. Normal deuterium isotope e f f e c t s were observed with alumina neutral (k /, = 1.3) and the other aluminas prepared by dehydration of alumina neutral (e.g. kjj//^ = ^'^ w i - t n a-alumina formed at 1250°C) . There was no isotope e f f e c t on the adsorption-desorption process and hence, the observed isotope e f f e c t i s due to the c a t a l y t i c reaction on the surface. Therefore, the observed normal isotope e f f e c t s i n d i c a t e that glucose mutarotation on alumina surface i s a general acid-base catalyzed reaction, and occurs by a consecutive mechanism, v i a the a c y c l i c intermediate. There i s probably no b i f u n c t i o n a l c a t a l y s i s of the glucose mutarotation on the alumina surface. Further, the acid s i t e s seem to show an isotope e f f e c t of 1.2 and the basic s i t e s an isotope e f f e c t of 2.1. Hence, these studies have shown that glucose mutarotation d i f f e r s (e.g. higher c a t a l y t i c a c t i v i t y of the hydrated surface, high c a t a l y t i c a c t i v i t y of a-alumina) from many other reactions studied on alumina surfaces. This difference i n behaviour under mild conditions i s probably due to the high s e n s i t i v i t y of the mutarotation reaction to the weak a c i d i c and b a s i c s i t e s on the alumina surface. V TABLE OF CONTENTS I INTRODUCTION 1 1.1 C a t a l y t i c Aluminas 2 1.2 Mutarotation of Glucose 36 II CHARACTERIZATION OF ALUMINAS 63 III PRELIMINARY KINETIC STUDIES 71 IV • PRODUCT ANALYSIS 75 V ADSORPTION OF GLUCOSE ON ALUMINA SURFACE PART I PRELIMINARY STUDIES 77 V . l T h e o r e t i c a l Study of Adsorption on Reversible and I r r e v e r s i b l e S i t e s 79 V.2 Determination of the Number of I r r e v e r s i b l e Adsorption Sites per Gram of the Catalyst 88 V.3 Rate of Adsorption 90 VI DIFFUSION IN HETEROGENEOUS CATALYSIS 93 VII RATE EQUATION FOR THE HETEROGENEOUS CATALYTIC SYSTEM 100 VIII STUDY OF THE CURVATURE IN FIRST ORDER PLOTS 107 VIII.1 Tests f o r Deactivation of the Catalyst 109 VIII.2 Cause of Deactivation 111 IX THE EFFECT OF CATALYST CONCENTRATION ON THE OBSERVED RATE CONSTANT 118 X THE EFFECT OF SUBSTRATE CONCENTRATION ON THE OBSERVED RATE CONSTANT 121 v i TABLE OF CONTENTS (continued) XI REPRODUCIBILITY OF THE RESULTS AND COMPARISON OF CATALYTIC ACTIVITIES OF DIFFERENT ALUMINAS 125 XII EFFECT OF DEHYDRATION ON THE CATALYTIC ACTIVITY OF ALUMINA 132 X I I . l Cause of High C a t a l y t i c A c t i v i t y of a-Alumina 139 XII.2 Some Observations on the Nature of Active Sites 143 XIII EFFECT OF WATER ON THE CATALYTIC ACTIVITY 145 XIV ADSORPTION OF GLUCOSE ON ALUMINA SURFACE PART II ADSORPTION ISOTHERMS 156 XIV.1 Adsorption Isotherm f o r Alumina Neutral 160 XIV.2 Adsorption Isotherm f o r Sintered Alumina 160 XIV.3 Maximum Amount of Glucose Adsorbed and the Surface Areas of Samples 164 XIV. 4 Strength of Adsorption of Glucose 165 XV CATALYTIC ACTIVITY OF REVERSIBLE ADSORPTION SITES AND THE ACTIVE SITE DENSITY OF ALUMINA NEUTRAL 176 XV. 1 Comparison with Other Methods of Determining Active S i t e Density 182 XV. 2 C a t a l y t i c I n a c t i v i t y of Strong Adsorption Sit e s 185 XVI KINETIC PARAMETERS OF THE CATALYTIC SYSTEM 187 XVI. 1 Determination of the C a t a l y t i c Constant f o r the Surface Reaction (k + k^) 188 XVI.2 C a t a l y t i c Constant for Heterogeneous Ca t a l y s i s of Glucose Mutarotation 190 XVI.3 Turnover Number of a C a t a l y t i c S i t e 191 v i i TABLE OF CONTENTS (continued) XVII NATURE OF ADSORPTION ON ACTIVE SITES 194 XVII. 1 E f f e c t on C a t a l y t i c A c t i v i t y of Pretreatment of Alumina with In h i b i t o r s 201 XVIII NATURE OF ACID/BASE FUNCTIONAL GROUPS ON ACTIVE SITES 206 XVIII. 1 Test f o r Basic Sites - I n h i b i t i o n by Carbon Dioxide 207 XVIII. 2 Tests f o r A c i d i c S i t e s on Alumina Neutral 210 XIX DEUTERIUM ISOTOPE EFFECT ON GLUCOSE MUTAROTATION BY ALUMINA 224 XIX. l Deuterium Isotope E f f e c t f o r Catalysis by 800°C Alumina 225 XIX.2 Deuterium Isotope E f f e c t f o r Catalysis by Sintered Alumina 239 XIX.3 Deuterium Isotope E f f e c t with Alumina Neutral 240 XIX.4 Relationship between the Observed Deuterium Isotope E f f e c t and the Percentage of A c t i v i t y due to Basic Sit e s 242 XX SUMMARY 248 XXI EXPERIMENTAL 257 BIBILIOGRAPHY 290 APPENDIX A : CHARACTERIZATION OF ALUMINAS 300 APPENDIX B : PREPARATION OF A BATCH OF ALUMINA WHICH ACTIVATES ON PYROLYSIS AT HIGH TEMPERATURES, SIMILAR TO THE FIRST BATCH OF ALUMINA NEUTRAL 311 v i i i LIST OF TABLES I Reactions Catalyzed by Aluminas 4 II Possible OH Configurations on Alumina Surface 17 III C a t a l y t i c Constants (10^ ^ c a t l i t r e mole ^ sec ^ expressed i n natural logarithms) f o r Mutarotation of Glucose or Tetramethylglucose 45 IV BET Surface Areas 65 V C h a r a c t e r i s t i c s of the P a r t i c l e Size D i s t r i b u t i o n Plots 67 VI Cumulative Percentages Greater than Stated Diameters for Ground and Unground Samples of Alumina Neutral 98 VII E f f e c t of Dehydration on C a t a l y t i c A c t i v i t y 134 VIII pH's of 10% S l u r r i e s in Water 140 IX E f f e c t of Water on Equilibrium Optical Rotation 152 X Data for the Adsorption Isotherm f o r Glucose onto Alumina Neutral at 25.0°C 161 XI Data for the Adsorption Isotherm for Glucose (at 25.0°C) onto Alumina Heated at 1250°C for 6 hours " 163 XII Data for the Langmuir Plot of [GC]/[G] versus [GC] Mole L i t r e f o r Adsorption of Glucose on Alumina Neutral at 25.0°C 167 XIII Data for the Langmuir P l o t of [GC]/[G] versus [GC] Mole L i t r e for Adsorption of Glucose (at 25.0°C) on Alumina Sintered at 1250°C 173 XIV Data for V a r i a t i o n of I n i t i a l Rate with Equilibrium Concentration of a-D-Glucose at Constant Catalyst Concentration (26.7 mg/ml) 179 XV Turnover Numbers for some Surface Catalyzed Reactions 192 XVI E f f e c t of I n h i b i t o r s on the C a t a l y t i c A c t i v i t y and on the Amount of Glucose Adsorbed 200 XVII E f f e c t of C0 2 on D i f f e r e n t Aluminas 208 i x LIST OF TABLES (continued) XVIII Data for the Relationship between the Observed Deuterium Isotope E f f e c t and Percentage of A c t i v i t y due to Basic Sites XIX Data for Adsorption of Glucose by 1.6 g of Alumina Neutral from a 60 ml 0.05 M Solution of a,6 Mixture i n DMSO at 25.0°C XX Trace Metals Present i n Different Aluminas XXI X-ray D i f f r a c t i o n Data for Alumina Neutral XXII X-ray D i f f r a c t i o n Data for Alumina Sintered at 1250°C for 6 hours 242 279 286 287 289 X LIST OF FIGURES 1 Schematic Representa t ion of Formation of va r i ou s A^O^ -hyd ra te s 9 2 ( I I I ) - Face of Alumina at D i f f e r e n t Stages of Dehydroxy la t i on 21 3 Pore S i ze D i s t r i b u t i o n s of (a) Alumina N e u t r a l and (b) Alumina Py ro l yzed at 800°C f o r 4 hours 66 4 P a r t i c l e S i z e D i s t r i b u t i o n s of (a) Alumina N e u t r a l and (b) Alumina S i n te red at 1250°C f o r 6 hours 68 5 E l e c t r o n Micrograph of Alumina N e u t r a l f o r TLC (x 4000) 69 6 E l e c t r o n Micrograph of S in te red (1250°C) Alumina (x 8000) 69 7 F i r s t Order K i n e t i c P l o t s f o r Mu ta ro t a t i on of (a) a-D-Glucose and (b) g-D-Glucose by Alumina N e u t r a l 73 8 P l o t of Change i n O p t i c a l R o t a t i o n versus Weight of Alumina f o r Adso rp t i on o f Glucose on Alumina N e u t r a l 80 9 P l o t of ( E q u i l i b r i u m O p t i c a l Ro ta t i on ) versus Weight of Alumina f o r Ad so rp t i on of Glucose on Alumina N e u t r a l 81 10 T h e o r e t i c a l P l o t of 1/[G ] versus [C ] , the Concent ra t i on eq o ' o f Ad so rp t i on S i t e s of a S o l i d C a t a l y s t Con ta in i ng Only R v e r s i b l e Ad so rp t i on S i t e s 83 11 T h e o r e t i c a l P l o t s of 1/[G ] versus [C ] , the Concent ra t i on eq o ' o f Ad so rp t i on S i t e s of a S o l i d C a t a l y s t Con ta in i ng (a) Only I r r e v e r s i b l e Ad so rp t i on S i t e s , (b) Both I r r e v e r s i b l e and R e v e r s i b l e Ad so rp t i on S i t e s , and (c) Only R e ve r s i b l e Ad so rp t i on S i t e s 85 12 T h e o r e t i c a l P l o t s of 1/[G ] versus [C ] the Concent ra t i on eq o of Ad so rp t i on S i t e s of a S o l i d C a t a l y s t Con ta in i ng (a) Both Weak and Strong R e v e r s i b l e Ad so rp t i on S i t e s , (b) Only S t rong Ad so rp t i on S i t e s , and (c) Only Weak Reve r s i b l e Adso rp t i on S i t e s 87 13 P l o t of E q u i l i b r i u m O p t i c a l Ro t a t i on versus Weight of Alumina f o r the Determinat ion of the Number of I r r e v e r s i b l e Ad so rp t i on S i t e s on Alumina N e u t r a l 89 14 Rate of Ad so rp t i on o f Glucose on Alumina N e u t r a l 91 15 F i r s t Order K i n e t i c P l o t f o r a-D-Glucose Mu ta ro t a t i on by Alumina N e u t r a l and Rate of Ad so rp t i on of Glucose on A l um ina . N e u t r a l 92 LIST OF FIGURES (continued) The E f f e c t of Rate of S t i r r i n g on the Observed Rate Constant P a r t i c l e Size D i s t r i b u t i o n s of (a) Unground and (b) Ground Samples of Alumina Neutral Deactivation of Alumina Neutral by DMSO and by Glucose During C a t a l y s i s of Mutarotation The E f f e c t of Water on the C a t a l y t i c A c t i v i t y of Alumina Neutral Tests f o r the Presence of Inhibi t o r y Products i n Solution and on the Surface of the Catalyst The E f f e c t of Catalyst Concentration on the Observed Rate Constant The E f f e c t of a-D-Glucose Concentration on the Observed Rate Constant Relation of (Observed Rate Constant) to the Concentration of a-D-Glucose The E f f e c t of Drying and D i s t i l l i n g of DMSO on the C a t a l y t i c A c t i v i t y of Alumina Neutral Re p r o d u c i b i l i t y of the Ki n e t i c s with Alumina Neutral and Comparison of the C a t a l y t i c A c t i v i t i e s of D i f f e r e n t Chromatographic Aluminas (Neutral) Comparison of the C a t a l y t i c A c t i v i t i e s of Some Non-Chromatographic Aluminas The E f f e c t of Dehydration Temperature of Alumina Neutral on Its C a t a l y t i c A c t i v i t y The E f f e c t of Dehydration Temperature of Alumina Neutral on Its C a t a l y t i c A c t i v i t y per Unit Area Comparison of the C a t a l y t i c A c t i v i t i e s of A c i d i c , Basic, and Neutral Aluminas (for TLC) and the E f f e c t of Dehydration at 1250°C for 6 hours on Their C a t a l y t i c A c t i v i t i e s The E f f e c t of Water on Alumina Dried at Room Temperature The E f f e c t of Water on Alumina Dried at 150°C Relation of the Observed Rate Constant to the Amount of Water Added to Alumina Dehydrated at 150°C x i i LIST OF FIGURES (continued) 33 Isotherms for Adsorption of Glucose on (a) Alumina Neutral and (b) Alumina Sintered at 1250°C f o r 6 hours 162 34 The Langmuir P l o t f o r Adsorption of Glucose on Alumina Neutral 168 35 The Langmuir P l o t f o r Adsorption of Glucose on Alumina Neutral at High Concentrations of Glucose 169 36 The Langmuir P l o t f o r Adsorption of Glucose on Alumina Sintered at 1250°C for 6 hours 175 37 Adsorption Isotherm for Reversible Adsorption of Glucose on Alumina Neutral and P l o t of I n i t i a l Rate on Alumina Neutral versus Concentration of Glucose 180 38 Comparison of the Theoretical Isotherms for Adsorption of Glucose on (a) Strong and (b) Weak Adsorption Sites on Alumina Neutral with (c) Plot of I n i t i a l Rate versus Glucose Concentration 181 39 The E f f e c t of Pretreatment of Alumina Neutral with Methyl a-D-Glucoside on Its C a t a l y t i c A c t i v i t y 203 40 The E f f e c t of Pretreatment of Alumina Neutral with Hexanal on I t s C a t a l y t i c A c t i v i t y 204 41 The E f f e c t of Carbon Dioxide on the C a t a l y t i c A c t i v i t y of Alumina Sintered at 1250°C f o r 6 hours 209 42 The E f f e c t of n-Butylamine on the C a t a l y t i c A c t i v i t y of Alumina Neutral 213 43 The E f f e c t of Tetramethylammonium Hydroxide on the C a t a l y t i c A c t i v i t y of Alumina Neutral 219 44 Deuterium Isotope E f f e c t with 800°C Alumina Under Anhydrous Conditions 226 45 Deuterium Isotope E f f e c t with 800°C Alumina i n the Presence of H 20/D 20 228 46 Deuterium Isotope E f f e c t on the Rate of Adsorption of Methyl a-D-Glucoside on 800°C Alumina 233 47 Deuterium Isotope E f f e c t on the Rate of Adsorption of Glucose on 800°C Alumina 236 x i i i LIST OF FIGURES (cont inued) 48 49 50 Deuterium Isotope E f f e c t w i t h Alumina S i n te red at 1250°C f o r 6 hours 241 C o r r e l a t i o n of the Observed Deuterium Isotope E f f e c t w i t h the Percentage D e a c t i v a t i o n by Carbon D iox ide 243 Comparison o f the Exper imenta l and T h e o r e t i c a l C o r r e l a t i o n P l o t s 247 51 P ro ton NMR Spectrum of 0 -deuterated a-D-Glucose i n DMSO-dg 274 52 Proton NMR Spectrum of a-D-Glucose i n DMS0-d& 275 53 P ro ton NMR Spectrum of B-D-Glucose i n DMSO-dg 276 54 R e l a t i o n of the O p t i c a l Ro t a t i on to the Concent ra t i on of a,6 M i x tu re i n DMSO 281 55 Ad so rp t i on and Desorpt ion Isotherms of N i t r o gen on Alumina N e u t r a l 305 56 Adso rp t i on and Deso rp t ion Isotherms of N i t rogen on 800°C Alumina 306 57 Adso rp t i on and Deso rp t ion Isotherms o f N i t r ogen on Alumina S in te red at 1250°C f o r 6 hours 307 58 Ba s i c Mechanism of the P r i n c i p l e of E l e c t r o zone Ce l l o scope 308 59 C a t a l y t i c A c t i v i t i e s of D i f f e r e n t Aluminas Prepared from Ac id Treated Bas i c Alumina 313 x i v ACKNOWLEDGEMENTS I would l i k e to extend my s i n c e r e g r a t i t u d e to P ro fe s s o r R ichard E. P incock f o r h i s gu idance, encouragement, and support throughout t h i s research and du r i ng the p r epa r a t i on o f t h i s t h e s i s . I am a l s o g r a t e f u l f o r the freedom I enjoyed dur ing the course of t h i s work. I would a l s o l i k e to thank my w i f e , L a l a n t h a , and son, Ro sh i t ha , f o r t h e i r l o v e , understanding and cont inuous encouragement dur ing the course of t h i s work. I remain most g r a t e f u l . I would l i k e to thank P ro fe s so r s H a r r i s o n , S tewart , and Do lph in of the Chemistry Department f o r d i s c u s s i n g s e v e r a l aspects of t h i s p r o j e c t w i t h me and f o r t h e i r v a l u a b l e suggest ions . S p e c i a l thanks are a l so due to P ro fe s s o r J . L e j a and Ms. S. F i no r a of the Department of M in ing and M i n e r a l Process Eng ineer ing f o r adv ice and a s s i s t ance i n c h a r a c t e r i z i n g the alumina samples. Thanks are a l s o due to Dr. A. Cheung f o r a s s i s t ance w i t h computer work and to Mr. S. Karunan i thy f o r a s s i s t ance w i t h X - ray powder a n a l y s i s . I would l i k e to express my a p p r e c i a t i o n to Mr. S h i r v e l S t an i s l a u s f o r p l o t t i n g the f i g u r e s on computer and to Mrs. Rani Theeparajah f o r t yp i n g the manuscr ipt . Messrs. L. T a l a g a l a , V. Karunaratne, V. Rajanayagam, L. Kuan and Miss P. Levy are thanked f o r p roo f read ing the t y p e s c r i p t . F i n a l l y , I would l i k e to thank the U n i v e r s i t y o f B r i t i s h Columbia Graduate Fe l l owsh ip Fund f o r award of a Graduate F e l l o w s h i p (1979-81), and P r o f e s s o r R.E. P incock f o r awarding me a Research A s s i s t a n t s h i p (1981-83). Dedicated to My Parents 11 I INTRODUCTION 2 I. INTRODUCTION Cat a l y s i s i s the phenomenon In which r e l a t i v e l y small amounts of foreign material c a l l e d a c a t a l y s t , augments the rate of a chemical reaction without the c a t a l y s t i t s e l f being consumed. In heterogeneous c a t a l y s i s the catalyst forms one phase, usually a s o l i d , and the reactants and products are present i n one or more f l u i d phases (gas or liquid)'''. The c a t a l y t i c reaction occurs on the surface of the s o l i d and the c a t a l y s t provides a mechanistic pathway not present i n i t s absence. Heterogeneous c a t a l y t i c systems have many advantages and also some disadvantages over homogeneous c a t a l y s t s . Work up i s easy with a s o l i d c a t a l y s t since f i l t r a t i o n would remove the c a t a l y s t and for reactions at gas/solid i n t e r f a c e there i s no solvent involved. Because of strong surface forces, reactions normally not possible i n s o l u t i o n , can take place on a s o l i d c a t a l y s t . They are also e a s i l y adapted to continuous processes which i s a great advantage f o r i n d u s t r i a l a p p l i c a t i o n s . In s p i t e of these advantages heterogeneous c a t a l y s t s are e a s i l y poisoned and they may not be used as e f f i c i e n t l y as a homogeneous ca t a l y s t since only the atoms on the surface are a v a i l a b l e for the c a t a l y t i c reaction. I.1 C a t a l y t i c Aluminas Aluminas ( i . e . various forms of A^O^) have been used extensively as 2 adsorbents, a c t i v e c a t a l y s t s , and c a t a l y s t supports . Already i n 1797 the alumina-catalyzed dehydration of ethanol was discovered by Dutch chemists 3 4 and as early as 1914, Sabatier ' reviewed the use of aluminas as a c t i v e c a t a l y s t s f o r various reactions. Since that time the a p p l i c a t i o n of aluminas i n c a t a l y t i c processes have increased tremendously. In i n d u s t r i a l c a t a l y t i c 3 5 6 processes, aluminas are mostly used as c a t a l y s t supports ' . Oxides and mixed oxides as w e l l as t r a n s i t i o n metals and noble metals are supported on alumina. Thus chromia-alumina c a t a l y s t s are being used for the conversion of p a r a f f i n s and o l e f i n i c hydrocarbons, i n hydrodealkylation of aromatics and to a lesser extent i n c a t a l y t i c reforming. The l a t t e r process i s also catalyzed by molybdena-alumina, a c a t a l y s t system which i s also a c t i v e for making toluene and other aromatics from saturated hydrocarbons. It also catalyzes the isomerization of p a r a f f i n s . Another c a t a l y t i c system of enormous importance i s cobalt oxide-molybdenum oxide-alumina which i s widely used for hydrodesulfurization, hydrodenitrogenation and hydrocracking reactions. In a l l these cases there i s ample evidence that 7 8 alumina i s f a r from a passive i n e r t support ' . The a p p l i c a t i o n of pure aluminas as c a t a l y s t s i n i n d u s t r i a l processes i s of le s s importance, although they are used, for example, i n the c a t a l y t i c conversion of the side products i n the oxo process, such as 2 a l k y l esters and high b o i l i n g condensation products . In academic research, pure aluminas are used widely for several groups of reactions, some of which are summarized i n Table 1. The reactions compiled i n the table show that aluminas are able to activa t e hydrogen-hydrogen, carbon-hydrogen, carbon-carbon, carbon-oxygen and oxygen-hydrogen bonds, although with varying e f f i c i e n c y . Thus o-p-H^ and 11^-1)^ e q u i l i b r a t i o n reactions occur at very low temperatures, and C-H bond a c t i v a t i o n i n exchange and isomerization reactions i s e f f e c t i v e near room temperature. For C-C bond a c t i v a t i o n , for example i n s k e l e t a l isomerization, aluminas are less a c t i v e . These reactions require much higher temperatures roughly above 325°C. 4 TABLE 1 REACTIONS CATALYZED BY ALUMINAS Reaction Temperature, °C o-H2 * p-H2 -195 H2 + °2 2 W -125 CH^/CD^ isotopic scrambling 25 Alkene + D 2 -»• alkene-d + HD 25 Benzene + D 2 benzene-d + HD 25 Double-bond isomerization of alkenes 25 Cis/trans isomerization of alkenes 25 Cyclopropane ->• propene 100 Alcohols -*• alkenes + H20 75 2 alcohols -»- ether + H20 125 Skeletal isomerization of alkenes 325 o-Xylene isomerization 500 In addition to those mentioned above many other reactions, often involving complex organic compounds, have been observed on alumina during 9 chromatography • This led to the use of chromatographic alumina to cause various adsorbed organic molecules to undergo many different types of often unanticipated chemical reactions^' Interest has been rekindled in this area by recent developments involving deliberate placement of different 5 reagents on s o l i d alumina and use of t h i s doped alumina to cause diverse organic reactions heterogeneously at the alumina surface under unusually 12 mild conditions • A few of the more i n t e r e s t i n g examples are discussed below. ( i ) The reaction of (+)-2a,3a-epoxypinane I on exposure to ac t i v e alumina i n hexane gave three main products I I , I I I and IV i n the r a t i o 1 : 4 : 2 besides traces (^2%) of trans g l y c o l ^ . The alcohol IV resulted »» 0 hexane . 25-30°C, 24 hrs CH2OH I II I I I IV from Cannizzaro type reaction of aldehyde V on alumina, the aldehyde i t s e l f being f i r s t formed from the oxide v i a a carbonium ion rearrangement. The occurrence of a Cannizzaro reaction was confirmed by i s o l a t i o n of the Cannizzaro a c i d VI as the methyl ester from the spent alumina. CHO C0 2H V VI 12 13 ( i i ) Posner e t . a l ' have shown that a s t i r r e d s l u r r y of commercially a v a i l a b l e Woelm 200 neutral chromatographic alumina catalyzes opening of a very wide v a r i e t y of epoxides by only a few equivalents of RZ-H nucleophiles reproducibly and under exceedingly mild conditions (10 min to 1 h 25°C, d i e t h y l ether solvent). Nucleophiles s u c c e s s f u l l y incorporated under these conditions include alcohols, t h i o l s , benzene s e l e n o l , amines and a c e t i c acid. 6 Alumina impregnated with a few equivalents of these nucleophiles opened epoxides r e g i o s e l e c t i v e l y at the l e s s substituted epoxide carbon atom and s t e r e o s p e c i f i c a l l y (trans) to give the corresponding 3-functionalized alcohol cleanly and i n good y i e l d . For example cyclohexene oxide reacted with alumina carrying 4% of RZ-H to give the corresponding trans-2-functionalized cyclohexanols reproducibly i n good y i e l d as the only product (Eq. 1). No a l l y l a l c o h o l , 1,2-diol or cis-isomer was detected. In several cases cyclohexene oxide reactions with RZ-H doped alumina were superior to homogeneous methods f o r synthesis of the corresponding 2-RZ-cyclohexanols. ZR Y i e l d (%) OMe 66 OCH2Ph 47 In addition to the reactions mentioned above, alumina has been used for intramolecular addition of OH groups, i n t r a - and intermolecular addition of CH groups, i n oxidation-reduction reactions, s u b s t i t u t i o n reactions, elimination reactions, decarboxylation reactions and s k e l e t a l rearrange-12 ments In s p i t e of the wide v a r i e t y of organic reactions catalyzed by alumina i n s o l u t i o n , there have been l i t t l e or no studies of the k i n e t i c s and mechanism of alumina catalyzed organic reactions i n s o l u t i o n . Most of 7 the mechanistic studies of organic reactions on alumina surface are confined to gas phase reactions of alcohols and a l k e n e s ^ ' T h e only k i n e t i c study of a substrate i n s o l u t i o n reacting on an alumina surface i s by L e f f l e r and M i l l e r ^ who investigated the reaction of d i a c y l peroxides with chromatographic alumina. They observed that several d i a c y l peroxides as wel l as perbenzoic acid and hydrogen peroxide reacted r a p i d l y with alumina surfaces to give a nonextractable oxidant of equivalent o x i d i z i n g power. Surface hydroperoxides Al g-OOH were suggested as the t i t r a b l e oxidant groups formed on alumina because the decay reaction released molecular oxygen. F i r s t order plots f or decomposition of the surface oxidant were curved with high i n i t i a l rates which decreased r a p i d l y both within a run and as a function of concentration of the i n i t i a l surface oxidant. They explained these r e s u l t s i n terms of surface s i t e s with d i f f e r e n t p r operties; the s i t e s were divided into categories I and I I . According to the model, Type I s i t e s reacted with d i a c y l peroxide more r a p i d l y than the Type II s i t e s (Eqs. 2 and 3). However,the oxidant at Type I s i t e s was considered to decompose more slowly than the oxidant at the Type II s i t e s (Eqs. 4 and 5) giving r i s e to the observed k i n e t i c s . R-C-0-O-C-R + A 1 2 0 3 ( I ) fa s t 2RC00H + A l OOH(I) (2) s slow 2RC00H + A l 00H (II) (3) A l OOH(I) slow *• A l OH + JgO. s Z (4) A l 00H (II) fas t A l OH + (5) s In order to obtain a more complete study of the k i n e t i c s and 8 mechanism of an alumina catalyzed organic reaction the study of glucose mutarotation by aluminum oxide was systematically undertaken f o r t h i s t h e s i s . Further, the study of glucose mutarotation by alumina should provide information on the nature of act i v e s i t e s on the surface and hence, might f i n d uses as a method f o r cha r a c t e r i z i n g alumina c a t a l y s t s . To gain an in s i g h t i n t o the action of alumina as a c a t a l y s t i t i s important to understand both the p h y s i c a l and chemical properties of alumina, s p e c i a l l y of i t s surface. The following sections w i l l review the present knowledge of the preparation, structure and nature of act i v e s i t e s on alumina. I.1.1 Preparation and C r y s t a l l i n e Structure of Alumina Aluminas are usually obtained by dehydration of aluminum trihydroxides (gibbsite, bayerite and nordstrandite) obtained by p r e c i p i t a t i o n from aqueous solutions containing aluminum i o n s ^ . Depending on the method of aging the gelatinous hydroxide and the p y r o l y s i s temperature i t can have several 6 8 17 c r y s t a l l i n e forms (Fig. I) ' ' . The method of formation of the hydroxide generally determines the impurities present i n the resultant alumina and hence w i l l a f f e c t i t s c a t a l y t i c p r o p e r t i e s ^ ' ^ . Aluminum hydroxide obtained by hydrolysis of aluminum isopropoxide which has been d i s t i l l e d i n vacuum w i l l produce 18 alumina v i r t u a l l y free of i o n i c impurities . But the hydroxide produced by adding ammonium hydroxide to a so l u t i o n of an ammonium s a l t contains entrapped anions. On the other hand, alumina prepared from gibbsite or from potassium or sodium aluminate contains a l k a l i i n the amount of 0.08 to 14 0.65% . During dehydration of the trihydroxide adjacent hydroxyl groups I n i t i a l P r e c i p i t a t i o n .^ 0- Nordstrandite K, S^e, Bayerite 20c • Boehmite s!0 gel 300c Y-alumina ^ PH >12 Gibbsite 230c ^ C r y s t a l l i n e boehmite r i-alumina 900c 6-alumina 1000° 6 + a-alumina 1200c 850c 0-alumina 1200° 450° y-alumina 600° 6-alumina 1050° 9 (+a)-alumina 1200° 250c X-alumina 900' K-alumina 1200c a-alumina F i g . 1 Schematic Representation of Formation of Various A^C^-hydrates 8,17 Formula A1 20 3.3H 20 A1 20 3.H 20 Al 20 3.nH 20 o<n<0.6 A l 2 0 3.nH 20 n low A1 2 0 3 (Temperature i n °C). vo 10 combine forming water vapour which gives r i s e to small pores within the c r y s t a l l a t t i c e . As dehydration proceeds the pores from d i f f e r e n t parts of the l a t t i c e j o i n forming a network and eventually the water vapour finds i t s way out of the l a t t i c e connecting the i n t e r n a l surface with the external surface. Thus a porous sample of alumina with high surface area per unit weight i s produced. Since a heterogeneous c a t a l y t i c reaction proceeds on the surface of the c a t a l y s t , the presence of a high surface area increases i t s e f f i c i e n c y and the presence of active s i t e s i n pores can also lead to a 14 'solvating e f f e c t ' of alumina and reactions normally not possible on a surface may proceed w i t h i n i t . 1.1.2 C l a s s i f i c a t i o n of Aluminas Aluminas can be c l a s s i f i e d according to the temperature at which they were obtained from the hydroxide or based upon the c r y s t a l l o g r a p h i c * , 1 7 structure of the alumina Aluminas are divided into two groups according to the temperature at which they were formed. (a) Low temperature aluminas or the y - g r o u P ' A^O^.n^O i n which 0<n<0.6; obtained by dehydration at temperatures not exceeding 600°C. (b) High temperature aluminas, c a l l e d the 6-group. These are nearly anhydrous aluminas obtained at temperatures between 900°C and 1000°C. To group (a) belong p~, X~» 1- and y-aluminas a r K i t o g r 0 u p (b) belong K - , 9- and 6-aluminas. A c l a s s i f i c a t i o n based on the c r y s t a l l o g r a p h i c structure of alumina was proposed by Krischner i n 1966^. As these structures are a l l based on a more or less close-packed oxygen l a t t i c e with aluminum ions i n the octahedral and tetrahedral i n t e r s t i c e s , three seri e s could be distinguished v i z : 11 a-series with hexagonal close-packed l a t t i c e , schematically ABAB... 3-series with a l t e r n a t i n g close-packed l a t t i c e , schematically ABAC-ABAC or ABAC-CABA y- s e r i e s with cubic close-packed l a t t i c e , schematically ABCABC The only representative of the a-series i s a-alumina. The 3-series consists of a l k a l i or a l k a l i n e earth oxide containing B-alumina and, x _ and K-alumina. The Y~series can be sub-divided to a y- or low-temperature group (consists of n- and y-alumina) and a 6- or high-temperature group (6- and 6-alumina). Out of these d i f f e r e n t c r y s t a l l i n e forms of alumina only y- and 19 20 21 22 23 ri-phases are important c a t a l y t i c a l l y ' ' ' ' The c a t a l y t i c a c t i v i t y 7 8 22 23 of n-alumina usually turns out to be higher than that of Y - a l u m i n a ' ' ' The other main form of alumina, v i z . a-alumina i s considered to be most i n e r t 17 24 of a l l aluminum oxides ' , and i s used mainly as a c a t a l y s t support (inert carrier)"* because of i t s high temperature s t a b i l i t y . As mentioned above y- and Ti-aluminas consist of cubic close-packed structure of oxide ions. L i p p e n s ^ who did an extensive X-ray c r y s t a l l o -graphic study of aluminas proposed models for y- and n-aluminas. The most important s t r u c t u r a l c h a r a c t e r i s t i c of alumina i n c a t a l y s i s i s the surface and since alumina occurs i n the form of lamellae i t i s most probable that only one type of surface plane i s predominent. According to Lippens t h i s i s the ( l l l ) - p l a n e for n-alumina and the (110)- or (lOO)-plane f o r y-alumina. In p r a c t i c e the (111)- and (llO)-planes are considered to form 23 the surface layers of n- and Y - a l u m i n a s , r e s p e c t i v e l y It has been shown that the ( l l l ) - p l a n e of n-alumina and (llO)-plane of Y ~ a l u m i n a have aluminum ions arranged i n both tetrahedral and octahedral p o s i t i o n s . But there i s a higher density of aluminum ions i n tetrahedral 12 posit i o n s i n n-alumina and t h i s i s considered to give r i s e to the higher 8 23 a c i d i t y and c a t a l y t i c a c t i v i t y observed with n-alumina ' Thus both Y — ana^ n-aluminas consist of cubic close-packed oxygen l a t t i c e with the aluminum ions d i s t r i b u t e d i n octahedral and tetrahedral s i t e s and some cation s i t e s are l e f t vacant for stoichiometric reasons. Therefore they are said to have 'defect' s p i n e l structure a f t e r the mineral 8 s p i n e l (MeAl„0.) which has a s i m i l a r structure . 2 4 As mentioned above a-alumina consists of a hexagonal close-packed 3+ oxygen l a t t i c e and, unlike n- and y-forms, a l l the A l ions are located i n 8,17 octahedral s i t e s with one i n every three cation s i t e s vacant . The 3+ presence of A l ions only i n octahedral s i t e s probably leads to the low 3+ c a t a l y t i c a c t i v i t y observed i n a-alumina, since the A l ions i n octahedral s i t e s are not as a c i d i c as those i n tetrahedral s i t e s because of the higher coordination number. 1.1.3 The Surface Structure of Alumina The chemical nature of the alumina surface i s of primary importance i n i t s c a t a l y t i c and adsorptive properties. So-called ' a c t i v e - a l u m i n a ' ^ , which i s alumina used as adsorbents and c a t a l y s t s , i s not pure alumina but contains, depending upon temperature and water vapour pressure, from few tenths to about 5% water. Depending on preparative conditions, other components may be present too, e.g. a l k a l i oxide, i r o n oxide and s u l p h a t e ^ . The presence of even minute amounts of sodium oxide was found to decrease the c a t a l y t i c e f f e c t of alumina on the dehydration of propanol and b u t a n o l ^ . Dehydration of cyclohexanol over pure alumina prepared from aluminum isopropoxide produced cyclohexene and upto 60% methylcyclopentenes. But over alumina containing about 0.4% of sodium or potassium ions, cyclohexene was 13 the only product owing to the absence of strong acid s i t e s which were 14 neu t r a l i z e d by the a l k a l i metal ions . The presence of sulphate or other anions i n general i s considered to increase the ' a c i d i c nature' of alumina. It i s also known to a f f e c t the rate of dehydration and the rate of s i n t e r i n g 18 of the c a t a l y s t A c t i v e alumina adsorbs water e i t h e r as hydroxyl ions or as water molecules on the surface, depending upon the temperature. When exposed to water vapour at about room temperature y-alumina adsorbs water as undissociated molecules bonded with strong hydrogen bonds to the underlying 17 25 26 surface ' . On n-alumina which i s c a t a l y t i c a l l y more a c t i v e , Borello et a l . have shown that i n addition to molecular adsorption of water, hydroxyl ion formation on the alumina surface occurs by d i s s o c i a t i v e adsorption of water. At higher water vapour pressures more water i s adsorbed i n a multi-l a y e r p h y s i c a l adsorption but t h i s water can be removed e a s i l y at about 120°C. The strongly adsorbed water which cannot be removed at 120°C was 27 defined as 'chemisorbed' water by de Boer et a l . I t was found to be the 2 same (25 mg/100 m ) independent of the temperature of dehydration of alumina. 25 P e r i and Hannan presented i n f r a - r e d spectroscopic evidence for the occurrence of both hydroxyl groups and undissociated water molecules on the surface of y-slumina at low temperatures. During the process of drying by heating water molecules, not desorbed and removed from the surface, react to form surface hydroxyl groups. Hence, i n i t i a l l y , there i s a decrease i n the number of water molecules and an increase i n the concentration of hydroxyl groups on the surface. A l l water molecules are removed af t e r evacuation at 400°C. At higher temperatures the hydroxyl groups are gradually expelled as water but even at 800 to 1000°C and vacuum some tenth of a percent of water i s s t i l l retained i n the alumina, y-alumina heated to 650-700°C 14 showed three major bands of hydroxyl s t r e t c h i n g frequencies at 3698, 3737, and 3795 cm ^ due to ' i s o l a t e d ' (not hydrogen bonded) hydroxyl groups. Two more bands were seen under high r e s o l u t i o n at 3780 cm ^ and 3733 cm ^ 28 i n samples that were w e l l dried . For a l l alumina modifications so far 23 investigated a t o t a l of f i v e bands have been observed . This indicates the presence of i s o l a t e d hydroxyl groups i n f i v e d i f f e r e n t environments on the alumina surface. The lowest frequency band was found to be the most a c i d i c because i t exchanged most e a s i l y with deuterium between 250° and 500°C, and with butene at 200°C. In y-alumina the hydroxyl band at 3737 cm ^ was removed more r a p i d l y on drying while with n-alumina the bands at 3780 23 26 and 3700 were removed more r e a d i l y ' , but a l l three major bands c l e a r l y remained a f t e r drying at 850°C. These r e s u l t s show that the hydroxyl groups are present i n d i f f e r e n t chemical environments (these may play d i f f e r e n t roles i n c a t a l y t i c r e a c t i o n s ) . The adsorption of large molecules (e.g. CCl^) generally reduced the peak i n t e n s i t y of the hydroxyl bands and broadened and s h i f t e d them to lower frequencies. The hydroxyl groups must therefore be on the surface, rather than within the alumina l a t t i c e . The 25 number of hydroxyl groups on the surface was also determined by measuring the number of hydrogen atoms that could be exchanged with deuterium gas. The r e s u l t s showed that the surface i s 40% covered with hydroxyl groups a f t e r drying at 400°C, 15% at 600°C, 2% at 800°C and 1% or l e s s above 900°C. Hydrogen bonding appeared to ex i s t on alumina dried below 600°C. Adsorption of water vapour at room temperature on y-alumina previously dried at 800°C produced absorption bands s i m i l a r to those of the s t r e t c h i n g and bending v i b r a t i o n s of l i q u i d water. The hydroxyl band at 3795 cm ^ was replaced by a band near 3500 cm "'"but most of the i s o l a t e d hydroxyl groups apparently were not perturbed by the adsorbed water. 15 1.1.3.1 A Model for the Surface of Alumina 20 Using h i s i n f r a r e d and gravimetric data as a guide, P e r i i n 1965 proposed a model for the surface of Y _ a l u r a i n a where he considered only the (lOO)-plane on the surface. This model was l a t e r improved by Kn'dzinger 23 and Ratnasamy who considered a l l three low index planes (100), (110) and (111) of alumina. For energetic reasons only anion layers w i l l terminate a c r y s t a l l i t e and i t was shown with the a i d of Pauling's e l e c t r o s t a t i c 29 valence r u l e (which states that the net charge i n a stable i o n i c structure should be equal or nearly equal to zero) that these surface layers w i l l most favourably consist of OH groups (Fig. 2 ). The v a r i e t y of d i f f e r e n t surface hydroxyl groups w i l l be b r i e f l y considered i n the next se c t i o n . 1.1.3.1.1 Surface Hydroxyl Groups 23 As indicated above KnOzinger and Ratnasamy have shown that on the surface of y- or n-alumina there are f i v e types of hydroxyl groups; 3+ Type l a , a terminal OH group coordinated to a si n g l e tetrahedral A l i o n : OH Type l a | A l / | \ Type I l a , a bridging OH group which l i n k s a tetrahedral and an octahedral c a t i o n : H Type I l a \ | / 0 \ A l A l / | \ / | \ 16 Type l i b where the OH group l i n k s to two cations i n octahedral positions: H • 0 | Type l i b A l A l Type III where i t i s coordinated to three cations i n octahedral i n t e r s t i c e s : H \ l / \ I / A l A l / | \ v / / | \ Type III 1 A l 1 /|\ and Type lb where the OH group i s coordinated to a si n g l e cation i n an octahedral i n t e r s t i c e OH Type lb A l / l \ Table II summarizes the f i v e possible OH configurations, the coordination number of surface anions and also the net charge on the oxygen when i t ex i s t s as an oxide ion and when i t i s protonated to form a hydroxyl group. These values have been obtained as the sum of the negative charge of the anion and sum of the strengths of the e l e c t r o s t a t i c bonds (= cation charge divided by coordination number) to the anion from adjacent cations. KnBzinger and Ratnasamy making use of the net charge on OH groups assigned the s t r e t c h i n g frequencies given i n the l a s t column of Table II to the f i v e hydroxyl groups. They correspond c l o s e l y to the f i v e s t r e t c h i n g bands observed with alumina as mentioned e a r l i e r . The band of highest wavenumber (3800 cm "*") was assigned to the configuration lb which bears the most negative charge (-0.5) while the band of lowest wavenumber (3700 cm TABLE II POSSIBLE OH CONFIGURATIONS ON ALUMINA SURFACE 2 3 Configuration Coordination Numbers of Net Charge Net Charge v (OH) Surface Anion at 0 at OH cm"-'-to A1(VI); to A1(IV) .  Q a 0 D III 3 - -0.5 +0.5 3700 - 3710 l i b 2 - -1.0 0 3740 - 3745 I l a 1 1 -0.75 +0.25 3730 - 3735 l a - 1 -1.25 -0.25 3760- 3780 lb 1 - -1.5 -0.5 3785 - 3800 Large c i r c l e : oxide ion or hydroxyl group; • : Al ( V I ) ; O : Al(IV) 18 was a t t r i b u t e d to configuration III which exhibits the most p o s i t i v e charge (+0.5). The remaining three bands were assigned with decreasing wavenumber to the corresponding configurations with increasing p o s i t i v e net charge. In a d d i t i o n the net charge on the hydroxyl group should determine the r e l a t i v e a c i d i t i e s and b a s i c i t i e s of the hydroxyl groups. One would expect the OH configuration of Type III with a net p o s i t i v e of +0.5 to be the most a c i d i c . The protonic a c i d i t y of the OH groups should decrease as the net charge on them becomes more negative and t h e i r b a s i c i t y should increase at the same time. The ease of removal of the OH groups should p a r a l l e l t h e i r b a s i c i t y since the remaining net p o s i t i v e charge at the anion vacancy i s lower the higher the net negative charge of the leaving OH group. The exceptional l a b i l i t y of the l a and lb OH groups i s r e f l e c t e d 18 23 i n t h e i r ease of exchange with C 0^, even at room temperature . The oxygens of other configurations were exchanged much more slowly. 1.1.3.1.2 Surface Dehydration 23 According to KnBzinger and Ratnasamy proton a c i d i t y and the ease of removal of OH groups should govern the dehydration process, at l e a s t i n the i n i t i a l stages at low temperatures. Thus a proton from an a c i d i c OH group (e.g. Type I l a ) would combine with a neighbouring basic hydroxyl group (e.g. Type la ) to form a water molecule. Therefore, the charge defects created are as small as possible and i n i t i a l dehydration leads to the formation of weak Lewis acid s i t e s and neighboring weak basic s i t e s . In f a c t i t has been observed that on dehydration i n t e n s i t y of the I 0 0 _ - 0. \ l / \ I — • \ I / \ A l A l A l A l A l A l + H„0 / l \ / | \ / | \ / | \ / | \ / | \ 2 Type I l a Type l a 19 i n f r a r e d bands due to a c i d i c and basic hydroxyl groups decrease f a s t e r than 23 those of the c e n t r a l bands According to t h i s model dehydration would proceed without the 3+ formation of high energy multiple vacancies (adjacent A l ions) and c l u s t e r s of oxide ions. They showed that i t can proceed u n t i l about 50 to 65% dehydroxylation has occurred, depending on the c r y s t a l planes involved. By studies of dehydration of d i f f e r e n t alumina samples i t has been shown that 23 t h i s degree of dehydroxylation occurs at temperatures between 300° and 400°C (Fig 2). 1.1.4 Nature of Active S i t e s I t i s important to determine whether these weak i n d i v i d u a l Lewis acid and basic s i t e s , formed during the i n i t i a l dehydroxylation, are the c a t a l y t i c a l l y a c t i v e s i t e s . It has been very c l e a r l y demonstrated by many authors that alumina should be pretreated at elevated temperatures f o r the 23 development of c a t a l y t i c a c t i v i t y i n most reactions . For example, alumina had to be pretreated at temperatures of roughly 300° to 400°C i n vacuum for double bond isomerization and s k e l e t a l isomerization of 22 30 1-pentene . van Cauwelaert and H a l l have shown that c a t a l y s i s of ortho-para hydrogen conversion occurs on alumina pyrolyzed above 300°C and that the rate constant increased by a factor of about 10 corresponding to a steep decrease i n surface hydroxyl concentration. These r e s u l t s ' c l e a r l y demonstrate that the removal of water and/or OH groups from alumina surface 23 i s e s s e n t i a l f o r the development of the c a t a l y t i c a c t i v i t y ' Although c a t a l y t i c a c t i v i t y develops on a c t i v a t i o n at temperatures 23 1A 2 above 300° to 400°C, i t has been shown that 3.7 x 10 cm anion vacancies and c o o r d i n a t i v e l y unsaturated (cus) oxygen atoms have already 20 been formed at 300°C. Therefore, Knozinger and Ratnasamy concluded that Lewis acid and basic s i t e s produced during regular dehydroxylation can hardly be involved i n most c a t a l y t i c reactions as c a t a l y t i c s i t e s . The same conclusion was reached by comparing the number of Lewis acid and basic s i t e s 14 -2 formed during dehydration (=10 cm ) with the number of c a t a l y t i c a l l y a c t i v e 12 13 -2 s i t e s determined by s p e c i f i c poisoning experiments (10 to 10 cm ). For comparison, the l a t t i c e s i t e d e n s i t i e s i n i d e a l planes are of the order f i n 1 5 ~ 2 of 10 cm Hence, KnSzinger and Ratnasamy assumed that at temperatures between 300° to 400°C s p e c i a l s i t e configurations of low p r o b a b i l i t y are beginning to develop which possess s t r u c t u r a l and energetic properties required for an a c t i v e s i t e . These s p e c i a l configurations have been i d e n t i f i e d with defects i n the p a r t i a l l y dehydroxylated surface; that i s with multiple vacancies and c l u s t e r s of oxygen atoms i n c e r t a i n environments on the surface. They showed that such defects should be formed on further dehydration above 300° to 400°C, due to condensation of equivalent OH groups with the formation of t r i p l e t vacancies and neighboring oxide ions 23 31 (Fig. 2). It has also been postulated ' that the mobility of ions as the temperature i s increased increases the chances of forming defects but t h i s m o b i l i t y can also r e d i s t r i b u t e the high surface energy of defect s i t e s . These two opposing e f f e c t s can lead to an optimum a c t i v a t i o n temperature of alumina, for c e r t a i n reactions (e.g. isomerization of 32 alkenes ). 1.1.4.1 D i f f e r e n t Types of Active Sites From the above discussion i t i s c l e a r that the alumina surface should consist of both a c i d i c and basic s i t e s of d i f f e r e n t strengths. 21 300°-400°C Completely hydroxylated ( I l l ) - f a c e Hydroxyl group on top layer Oxide ion on lower layer Aluminum ion > 400°C (III)-Face at 50% dehydroxylation Oxide ion on top layer Neighboring oxide ions T r i p l e t vacancy i n ( I l l ) - f a c e T r i p l e t vacancy F i g . 2 (III)-Face of Alumina at D i f f e r e n t Stages 23 of Dehydroxylation 22 The a c i d i c s i t e s are the (a) Bronsted Acid s i t e s , for example 3+ Type III OH groups, and (b) Simple Lewis Acid s i t e s ( i n d i v i d u a l A l i o n s ) , both present at r e l a t i v e l y low temperatures, and (c) T r i p l e vacancies 3+ (three neighboring A l ions) present at temperatures greater than 300° to 400°C. The basic s i t e s are the (d) Basic hydroxyl groups, for example Type lb hydroxyl groups, (e) I n d i v i d u a l oxide ions, both of which are again present at r e l a t i v e l y low temperatures, and (f) Clusters of oxide ions formed at temperatures greater than 300° to 400°C. As discussed l a t e r (Section 1.2.3) i n t h i s Introduction, mutarotation of glucose i n homogeneous s o l u t i o n i s catalyzed by Bronsted acids, bases such as hydroxide ions and carboxylate ions and even Lewis acids such as 3+ 2+ 2+ 2+ A l , Zn , Cu , and Ni . Hence a l l these a c i d i c and b a s i c s i t e s are p o t e n t i a l c a t a l y t i c s i t e s for the mutarotation reacti o n . Evidence for the presence of those active s i t e s has come from 24 adsorption studies of acid or base s e n s i t i v e substances . The use of such substances as s p e c i f i c poisons of c a t a l y t i c reactions has shown that these 33 s i t e s a c t u a l l y p a r t i c i p a t e i n the surface reactions . Since the same methods can 'be used to determine the a c t i v e s i t e s f o r glucose mutarotation they w i l l be described i n the next section. 1.1.4.2 Evidence for the Presence of Active Sites 24 1.1.4.2.1 Acid Sites The presence of acid s i t e s on the alumina surface has been shown by t i t r a t i o n of the s o l i d a c i d suspended i n dry benzene with n-butylamine. The use of various i n d i c a t o r s with d i f f e r e n t pK values to detect the end 23 point has enabled the determination of the amount of acid at various a c i d strengths. The acid strengths (H q) as measured by amine t i t r a t i o n ranged from +3.3 to -5.6 for a sample of pure alumina prepared from aluminum 2 A isopropoxide and pyrolyzed between 300° and 1000°C . The amount of acid at various strengths was a function of p y r o l y s i s temperature. It increased on heating above =-200°C but decreased at high temperatures ( =:1000oC). The amount of acid s i t e s has also been determined by t i t r a t i o n with potassium hydroxide, dioxane and also by chemisorption of ammonia* trimethyl-amine and pyridine. Each of these methods gave a c i d i t y values "for alumina which apparently approximate those of s i l i c a - a l u m i n a . 32 Pines and Haag determined the number of acid s i t e s on pure alumina prepared from aluminum isopropoxide and activated at d i f f e r e n t temperatures, 13 by chemisorption of trimethylamine at 300°C. There were 1.6 x 10 s i t e s / 2 13 2 13 cm when pyrolyzed at 400°C, 2 x 10 sites/cm at 600°C, 1.4 x 10 s i t e s / 2 13 2 cm at 700°C and 0.3 x 10 sites/cm at 900°C, assuming that each adsorbed molecule corresponds to one acid s i t e . These values are i n good agreement 13 2 34 with those obtained by n-butylamine t i t r a t i o n (1.1 to 2.5 x 10 sites/cm ) ( i ) The Nature of Acid Sites The methods described above do not d i s t i n g u i s h between Lewis acid s i t e s and BrBnsted acid s i t e s . The nature of the acid s i t e s has been i n t e n s i v e l y investigated by d i f f e r e n t techniques during the l a s t two 24 decades . Hammett in d i c a t o r s (pK +6.8 to -8.2) which produced acid colour with synthetic cracking c a t a l y s t s and n a t u r a l clays gave no acid colour 35 with alumina . To test whether protonic a c i d i t y i s developed only at higher temperatures, the colour test was performed i n r e f l u x i n g xylene (b. pt. 144°C) using n e u t r a l red (pK +6.8). However, no acid colour was 24 32 produced. Hence,Pines and Haag concluded that BrBnsted acid s i t e s on alumina, i f present at a l l , are of very low acid strength. The same conclusion was reached from the f a i l u r e of alumina to undergo cation exchange 32 35 with ammonium acetate ' However, when they used a set of i n d i c a t o r s (triphenylmethane derivatives) which give colour with Lewis acids (by hydride ion abstraction) but not with BrSnsted acids, alumina pyrolyzed at 700°C showed the presence of Lewis acid s i t e s . Exposure of the c a t a l y s t to the humidity of the atmosphere before t e s t i n g i n h i b i t e d the development of colour i n the 32 alumina . The a c t i v e centres apparently were poisoned by the strong adsorption of water which i s removed by heating at high temperatures. More useful information on the nature of the acid s i t e s on alumina has come from the use of more s e n s i t i v e instrumental methods. For example, i n f r a r e d and nuclear magnetic resonance spectroscopy have been used to determine the nature of the chemisorbed bases on the alumina surface, as described below. (a) Ammonia Ammonia i s a strong Lewis base (K^ = 1.8 x 10 ^ i n water at 25°C) and i t i s small i n s i z e . By i n f r a r e d spectra of adsorbed species on alumina four 33 surface species have been i d e n t i f i e d , beside weakly held ammonia which was simply hydrogen bonded. The dominant species was coordinately held ammonia that adsorbs on Lewis acid s i t e s . A l i m i t i n g form of t h i s species v i z . +NH 3, has been observed on strongly dehydrated (>500°C) alumina. I t might represent ammonia associated with strongly p o s i t i v e charged t r i p l e vacant s i t e s . At high pretreatment temperatures (>500°C) and low hydroxyl d e n s i t i e s , NH9 groups are formed probably due to d i s s o c i a t i v e chemisorption 25 on acid-base p a i r s i t e s . On alumina containing high OH concentrations (pretreatment <400°C) protonated ammonia (NH^+) has been observed due to weak BrOnsted acid s i t e s . (b) Pyridine - 9 Pyridine i s l e s s b a s ic (K^ = 2.3 x 10 i n water at 25°C) than ammonia but i t i s s t i l l a f a i r l y hard base. The i n f r a r e d spectra of adsorbed pyridine have shown the presence of Lewis acid s i t e s on alumina. But, according to most authors, no i n t r i n s i c Bronsted s i t e s could be found by pyridine adsorption since pyridinium ion (PyH +) was not formed at a 33 detectable l e v e l due to lower b a s i c i t y of pyridine . Even when the spectra were recorded at temperatures up to 300°C, the PyH + species could not be detected i n d i c a t i n g that the protonic a c i d i t y of alumina was not apparently increased i n t h i s temperature range. Only Bremer and + 36 co-workers claim to detect PyH species on ri -alumina 37 However, Dewing et a l . detected protonated species due to weak Bronsted acid s i t e s , i n addition to coordinated species, by adsorption of 38 2 , 6 - d i t e r t i a r y - b u t y l p y r i d i n e on y-alumina. More recently, Pearson used wide l i n e nuclear magnetic resonance to study the nature of deuterated pyridine adsorbed on alumina at 0°C. He detected the protonated species on 13 -2 alumina (2.6 x 10 s i t e s cm on alumina activated at 600°C) probably because of the greater s e n s i t i v i t y of the n.m.r. method. At temperatures greater than 350°C pyridine reacted with surface OH 33 groups forming surface pyridone species with the production of hydrogen This i n d i c a t e s the presence of strongly basic OH ions held to c e r t a i n s i t e s on the alumina surface, t h e i r number being of the order of magnitude 13 -2 10 cm . A d d i t i o n a l evidence for the existence of these r e a c t i v e and 26 strongly b a s i c OH ions has come from adsorption of n i t r i l e s and ketones on alumina^ 3. (c) Butylamine -4 The i n f r a r e d spectra of n-butylamine (K^ = 4.8 x 10 i n water at 25°C) adsorbed at room temperature on alumina pretreated at 500°C showed 39 the presence of Lewis acid s i t e s . But no Bronsted s i t e s were observed even on heating to 500°C. Adsorption of pyridine on the same sample showed the presence of Lewis acid s i t e s but no Bronsted s i t e s . When n-butylamine was adsorbed at room temperature on alumina pretreated at 100°C both Brbnsted and Lewis a c i d s i t e s were observed. When pyridine was adsorbed on the same sample only Lewis acid s i t e s were observed due to the lower b a s i c i t y of pyridine. The above discussion shows that Lewis acid s i t e s are the predominant a c i d s i t e s on alumina, and also that there are weak Bronsted acid s i t e s on the strongly hydroxylated (Note: according to the model discussed i n Section 1.1.3.1. a c i d i c protons are removed during the i n i t i a l dehydration) surface. Formation of +NH 3 and NH^ probably gives evidence f o r the presence of defect s i t e s on strongly dehydrated alumina and the presence of strongly basic hydroxide ions are indicated by the formation of pyridone on the surface. 27 1.1.4.2.2 Basic Si t e s The basic s i t e s on alumina have been investigated using carbon 33 dioxide, a c e t i c a c i d and tetracyanoethylene (a) Carbon Dioxide Carbon dioxide i s a f a i r l y small molecule with a c i d i c properties and has frequently been used as a probe molecule for b a s i c surface s i t e s and as a poison i n c a t a l y t i c reactions. A f t e r heat treatment at roughly below 500°C, the alumina surface i s s t i l l strongly hydroxylated and carbon dioxide adsorption leads to the formation of a surface bicarbonate ion 33 predominantly . On formation of t h i s species, CO2 s e l e c t i v e l y reacts with the highest-frequency (3800 cm OH groups of the alumina surface. I t i s assumed that the bicarbonate ion forms on an A1-0H p a i r s i t e which was c a l l e d 40 " X - s i t e " by Fink . The number of these s i t e s varied between 1.2 and 1.8 x 13 2 10 /cm and may be i d e n t i c a l to those that convert pyridine to pyridone. A free carbonate ion has also been observed on the surface. On more extensively dehydroxylated alumina surfaces, bidentate (on A acid-base p a i r s i t e s , p o s s i b l y a - s i t e s A l Al) and unidentate carbonate 41 groups have been observed. Gregg and Ramsay showed that on ic-alumina surface heat treated at 1000°C, only 1 i n 10 oxide ions i s r e a c t i v e . Hence i t appears that only a small percentage of the surface oxide ions are located i n s u i t a b l e environments f o r carbonate formation. Further, i t has also been suggested that carbon dioxide gets adsorbed on Lewis acid s i t e s 28 on dehydroxylated alumina surfaces (b) Tetracyanoethylene (TCNE) Tetracyanoethylene has been used to detect donor s i t e s on oxide surfaces. E l e c t r o n i c and ESR spectra of the adsorbed acceptor molecules are c h a r a c t e r i s t i c of the surface anion r a d i c a l s which are assumed to be 33 formed according to Eq. (6) . The b a s i c OH ions on a hydroxyl r i c h TCNE + [D] mmt ^  [TCNE_ D +] (6) s s surface and the oxide ions on a strongly dehydroxylated surface can act as donor s i t e s . Hence as the surface i s dehydroxylated the spin concentration of the anion r a d i c a l passes through two maxima: the f i r s t i s located between 400° and 500°C ( OH donor s i t e s ) and the second (brought about by the oxide ions) i s between 600° and 700°C. These r e s u l t s i n d i c a t e that there are two types of basic s i t e s on the alumina surface. The predominant species i s determined by the temperature of dehydration of the catalyst. Thus the basic hydroxide ions predominate at temperatures le s s than 500°C (Note: according to the model discussed i n Section 1.1.3.1 ba s i c OH groups are removed during i n i t i a l dehydration), while the re a c t i v e oxide ions are more abundant at higher dehydration temperatures. 1.1.4.2.3 Evidence for the Presence of Defect Sites The existence of defect s i t e s was investigated by D e l i a Getta 29 42 et a l . using carbon monoxide as a surface probe. Carbon monoxide i s a rather s o f t base and therefore should throw l i g h t on the presence of strong Lewis acid s i t e s on the surface. When ri-alumina was dehydroxylated below 400°C, no CO adsorption was detected at 36°C. Dehydration at 400°C led to weak r e v e r s i b l e adsorption of CO. A second CO species, which i s strongly held at 36°C, f i r s t appeared on heat treatment around 500°C. Both species were described as CO molecule linked coordinatively to cus surface cations through 0-dative bonds. The s i t e density for the strongly adsorbed species was determined to be 2 x 10^ " -2 cm and i t was shown that t h i s species was attached to X-sites on the surface. The CO adsorption on Y~alumina leads also to the formation of two species, a more strongly held form and a l e s s energetic form. The c h a r a c t e r i s t i c d i f f e r e n c e between the y - and n-forms i s the f i n d i n g that 12 -2 the number of strong s i t e s was only about 6 x 10 cm on Y - a l u m i n a - Even the number of weakly adsorbing s i t e s was much lower than on n-alumina. This lower density of defect s i t e s i s probably the main reason for lower 23 c a t a l y t i c a c t i v i t y of Y - a l u m i n a 1.1.5 Evidence for the P a r t i c i p a t i o n of these A c i d i c and Basic Sites i n  C a t a l y t i c Reactions In the above discussion the presence of d i f f e r e n t types of a c i d i c and basic s i t e s on the alumina surface activated at d i f f e r e n t temperatures has been shown using s p e c i f i c surface probes. The mutarotation of glucose i s catalyzed by both acids and bases i n s o l u t i o n . Hence a l l these s i t e s are p o t e n t i a l c a t a l y t i c s i t e s for the mutarotation reaction. Some of the evidence concerning active s i t e s a v a i l a b l e from various 30 reactions previously studied on alumina w i l l be b r i e f l y discussed below. This discussion w i l l point out the advantages of using the mutarotation reaction as a probe for surface active s i t e s . 1.1.5.1 Hydrogen-Deuterium Exchange As already mentioned alumina catalyzes the e 9 u i l i b r a t i o n and also o-p-^ c o n v e r s i o n ^ at low temperatures ( a-75°C f or equi-l i b r a t i o n ) provided the oxides are dehydroxylated at temperatures higher than 325°C. This suggests that defect s i t e s of low p r o b a b i l i t y must be involved i n these reactions. 23 It was shown that the hydrogen species which led to exchange with T>2 was a species chemisorbed atomically onto the surface with a saturation 12 -2 coverage at -75°C of 9.7 x 10 hydrogen atom cm . The s i t e s responsible for t h i s type of hydrogen chemisorption should exhibit a very strong gradient of e l e c t r i c f i e l d strength. Thus the most probable c a t a l y t i c s i t e s are the defect s i t e s . The necessity of oxide ions for these reactions 33 was shown by the i n h i b i t o r y e f f e c t of carbon dioxide on these reactions 1.1.5.2 Isomerization and Exchange Reactions of Hydrocarbons Aluminas can also a c t i v a t e C-H bonds i n saturated and unsaturated hydrocarbons. They are therefore a c t i v e c a t a l y s t s for double bond and cis - t r a n s isomerization reactions and also for exchange reactions such as D^ exchange with hydrocarbons and deuterium scrambling (e.g. C^H^/C^D^ or 33 CH./CD.) . The behaviour of aluminas i n these reactions has turned out 4 4 to be extremely complex, and only a few s p e c i f i c examples w i l l be described here. Olefins and aromatic hydrocarbons such as benzene and toluene undergo 31 exchange reactions with deuterium gas around room temperature on alumina 23 activated at 500°C and higher temperatures Deuterium exchange of benzene and toluene occurs v i a h e t e r o l y t i c r i n g C-H bond cleavage, with loss of H + to the ca t a l y s t surface and the formation of a carbanion l i k e species. The exchange with a l k y l side chain 23 was much slower than with r i n g hydrogen With terminal o l e f i n s methylenic hydrogen atoms are p r e f e r e n t i a l l y exchanged. Therefore a l k e n - l - y l species have been suggested as intermediates for the exchange reactions which must be formed v i a a d i s s o c i a t i v e chemisorption step. Hence i t i s clear that h e t e r o l y t i c cleavage of unsaturated hydro-carbon C-H bonds occurs on activated alumina. As possible s i t e s one has to assume strong acid-base p a i r s i t e s which exhibit very high gradients of e l e c t r i c f i e l d strength. As expected a new OH group was formed from H + s p l i t o f f during chemisorption of benzene and o l e f i n s , g i ving r i s e to a -1 23 new OH st r e t c h i n g band at 3590 cm . Furthermore, the c h a r a c t e r i s t i c high-frequency band of the Type l a OH groups was perturbed by the chemisorbed alkenes or benzene. Precoverage of the alumina surface by alkene or benzene blocked the hydrogen chemisorption. Hence hydrogen (or deuterium) and hydrocarbon molecules are chemisorbed on the same defect X - s i t e s , with the small hydrogen or deuterium molecule being located within the multiple anion vacancies underneath the hydrocarbon molecule. Carbon dioxide has been used as a poison f o r i d e n t i f i c a t i o n of acti v e s i t e s of exchange reactions such as CH^ + B^, CH^ + CD^, exchange of o l e f i n s with deuterium and benzene with deuterium. These exchange reactions were strongly influenced by the carbon dioxide chemisorption and from the i n f r a r e d spectra of adsorbed carbon dioxide the active s i t e has 32 has been i d e n t i f i e d as a reactive oxide ion. The number of such s i t e s a c t i v e i n the exchange reactions as determined by carbon dioxide poisoning 12 -2 vary between 2.4 x 10 cm f o r the exchange of methane with and about 13 -2 33 1.4 x 10 cm for the exchange of o l e f i n s with . Hence, again, low p r o b a b i l i t y s i t e configurations were suggested as the active s i t e s . Since CO2 s t i l l adsorbs to a large extent as a surface bicarbonate species a f t e r a c t i v a t i o n at 530°C, the active s i t e for the exchange reactions was assumed a d d i t i o n a l l y to involve an adjacent hydroxyl group. This type of a c t i v e H s i t e has been schematically pictured as 0 0 43 44 Hightower and H a l l ' have shown that with alumina activated at 530°C, deuterium exchange and the intramolecular double-bond isomerization of o l e f i n s are independent processes since the s i t e s that catalyze exchange with o l e f i n s are blocked by but s i t e s active i n double-bond isomerization remain unaffected by CO2 chemisorption. On the other hand R^S and methyl mercaptan are e f f e c t i v e poisons at 25°C for s i t e s that catalyze double-bond migration and c i s / t r a n s isomerization of o l e f i n s but had no e f f e c t on exchange s i t e s . The isomerization s i t e s have a surface 13 -2 3+ 45 density of 5 x 10 cm and consists of c e r t a i n exposed A l ions 1 3 2 A 6 Deuterium exchange s i t e s are much less numerous (0.3 to 0.8 x 10 cm ) and are associated with a very small f r a c t i o n of highly energetic surface oxide ions, as mentioned e a r l i e r . These studies have led to the postulation of dual nature of s i t e s on alumina surface. The need of Lewis a c i d s i t e s for double bond isomerization has been c l e a r l y demonstrated by the retarding e f f e c t of ammonia, triethylamine 33 and p y r i d i n e , on such reactions 33 1.1.5.3 Dehydration of Alcohols The elimination of water from a l i p h a t i c alcohols on alumina i s known to proceed through two possible routes 33 , namely, monomolecular o l e f i n formation, \ C / / — c — c — c = c \ + OH and bimolecular ether formation 2 ROH The a c t i v e s i t e s i n the two reaction routes w i l l be discussed separately. 1.1.5.3.1 O l e f i n Formation 47 48 Knozinger and Deo et a l . have proposed that o l e f i n formation proceeds through hydrogen bonded adsorbed species. Although Lewis acid s i t e s had e a r l i e r been considered as the adsorption s i t e s of alcohol undergoing dehydration to o l e f i n s , i t was shown that pyridine 47 adsorbed on alumina had no e f f e c t on the rate of o l e f i n formation . The importance of hydroxyl groups was shown by the fa c t that dehydration a c t i v i t y of alumina has been shown to e x h i b i t an optimum value for a given pretreatment temperature of alumina. The necessity of basic s i t e s was 33 shown by poisoning e f f e c t of TCNE on o l e f i n formation from isopropanol Based on these r e s u l t s Knozinger proposed the mechanism given below for o l e f i n formation on alumina. That the reaction i s by mechanism was shown by the deuterium isotope e f f e c t on dehydration of 2-methyl propan-l-ol, 7 49 butan-2-ol and t-butanol ' . No isotope e f f e c t s from deuteration of the OH groups of these reactants were found. Very high g-deuterium isotope e f f e c t s were observed; the greatest e f f e c t being exhibited by primary 34 w m m m . w m m m . vfffWm/m. Adsorption T r a n s i t i o n state alcohol (3.44 at 150°C) and the smallest by the t e r t i a r y alcohol (2.42 at 150°C). The sequence suggested that the mechanism f o r dehydration of the primary alcohol possessed the highest character. A s h i f t towards character was seen on moving to the secondary and t e r t i a r y alcohols, but even so the mechanism of dehydration of the t e r t i a r y alcohol was s t i l l predominantly of the E^ type. The observed k i n e t i c isotope e f f e c t s were lower at higher temperatures, t h i s being associated with an increase i n E^ character of the elimination reactions. It was suggested that t h i s tendency may be due to the increase of Bronsted a c i d i t y of the surface hydroxyl groups on alumina at higher temperatures which would favour i o n i c contributions to the reaction mechanism. 1.1.5.3.2 Ether Formation It has been shown that only those alcohols that form detectable surface alcoholate species on alumina undergo bimolecular dehydration with 33 ether and water as reaction products . Alcoholate formation i s due to the d i s s o c i a t i v e chemisorption step of the alcohol that occurs on Al—0 p a i r s i t e . 35 33 Figueras Roca and co-workers showed that the rate of ether formation from methanol and ethanol responded very s e n s i t i v e l y to poisoning with TCNE. This proves the p a r t i c i p a t i o n of b a s i c s i t e s i n the bimolecular alcohol dehydration. P a r t i c i p a t i o n of Lewis acid s i t e s was shown by 33 poisoning with pyridine . Pyridine influenced the surface concentration of a surface ethanolate as shown by i n f r a r e d spectroscopy and the r a t e of d i e t h y l ether formation of the pyridine-poisoned alumina c a t a l y s t was found to be d i r e c t l y proportional to the number of alcoholate groups i n the surface. This indicates that only one alcoholate group p a r t i c i p a t e s per reaction step. The second molecule was assumed to be a hydrogen bonded 47 alcohol molecule, since i t has been shown , as i n o l e f i n formation, that OH groups are also necessary as a c t i v e centers for the ether formation. The above discussion gives c l e a r evidence for the active p a r t i c i p a t i o n of Lewis acid and base s i t e s i n reactions catalyzed by aluminas. The only example of possible Bronsted acid c a t a l y s i s was i n the dehydration of alcohols. The reaction mechanism changed from to E^ when the temperature was increased from 120° to 200°C, probably because of the increase of Bronsted a c i d i t y at higher temperatures. At 120°C the proton a c i d i t y was too low and t-butanol dehydrated mainly by an E^ mechanism. Direct evidence for the p a r t i c i p a t i o n of Bronsted acid s i t e s i n c a t a l y t i c reactions was obtained by John e t . a l . " ^ . They showed that alumina dried at 450° or 750°C and treated with D 20 at >180°C catalyzed double bond migration of propene at 210°C by a carbonium ion mechanism i n v o l v i n g Bronsted a c i d s i t e s . No such mechanism occurred at 25°C. Hence i t appears that although weak Bronsted acid s i t e s had been detected i n s i g i n f i c a n t 13 -2 qu a n t i t i e s (2.6 x 10 s i t e s cm on alumina activated at 600°C) at room 36 temperature using s e n s i t i v e surface probe molecules and s e n s i t i v e 38 techniques , c a t a l y t i c a c t i v i t y of such s i t e s has not been observed at 33 ordinary temperatures. According to Knozinger the l i f e time of a protonated species of a probe molecule may be very low due to the high mobility of surface protons and may contribute to low d e t e c t a b i l i t y of the protonated form of the poison. He also stated that protons that can hardly be detected d i r e c t l y by protonated probe molecules may w e l l i n i t i a t e c a t a l y t i c reactions due to t h e i r p o l a r i z i n g action during t h e i r f l u c t u a t i o n s . In other words, i t may be po s s i b l e to detect c a t a l y t i c a c t i v i t y of.those weak Bronsted acid s i t e s even at room temperature. According to John et a l . " ^ "the controversy that has h i t h e r t o centred on the presence or otherwise of Bronsted centres on alumina as investigated by spectroscopic techniques, seems to emphasize the need f o r the use of s e n s i t i v e c a t a l y t i c reactions as a d i r e c t probe to reaction mechanism, rather than i t s inference from i n d i r e c t techniques". A reaction that has been very extensively studied i n homogeneous sol u t i o n , which i s catalyzed by both Bronsted acids and bases, i s the 51 .. 52 mutarotation of glucose . Bronsted and Guggenheim i n t h e i r studies of acid and base c a t a l y s i s stated that the reaction to be investigated should be very s e n s i t i v e to H^ o"*" and OH i n order that the e f f e c t of weak acids and bases may be detectable. They selected mutarotation of glucose as the model reaction since i t f u l f i l l e d those conditions of s e n s i t i v i t y . I.2 Mutarotation of Glucose The term mutarotation r e f e r s to the change i n the o p t i c a l r o t a t i o n of a s o l u t i o n to an equilibrium value. The mutarotation of a s o l u t i o n of 53 D-glucose was discovered i n 1846 by Dubraunfaut , and since then various 37 theories were advanced to explain the mutarotation of sugars. Although glucose was at f i r s t considered to be a s i x carbon s t r a i g h t chain pentahydroxy aldehyde VII, i t f a i l e d to undergo c e r t a i n reactions 54 54 t y p i c a l of aldehydes . Colley i n 1870 and T o l l e n i n 1883 explained the lack of t y p i c a l aldehyde reactions as a r i s i n g from an a l t e r a t i o n of the aldehyde group by the formation of an inner hemiacetal type linkage. Fischer i n 1893 and von Lippmann^'''pointed out that r i n g formation would produce a new asymmetric carbon atom and thus the existence of a- and S-isomers VIII and IX of D-glucose and i t s de r i v a t i v e s was c l a r i f i e d . I n i t i a l l y the r i n g formation was considered to be of 1,4-type (furanose), 55 but i n 1926 Haworth showed that i t i s between 0-5 of glucose and i t s aldehyde group, i . e . i t i s a pyranose form. CHO H HO H H OH H OH OH CH20H a-D-Glucopyranose g-D-Glucopyranos e VII VIII IX In 1899 Lowry"^ recognized that mutarotations shown by reducing sugars are due to interconversion of r i n g isomers a - and 6-forms to the equilibrium mixture of the two forms. a - D-glucose 8-D-glucose (7) 1.2.1 K i n e t i c s of Mutarotation Urech, i n 1883, using f i r s t order equation derived by Wilhelmy for 38 following the inversion of sucrose, showed that the mutarotation of D-glucose obeys f i r s t order k i n e t i c s reasonably w e l l ~ ^ ^ a \ In 1903, Hudson"*^ ^ ' ^ showed that the v e l o c i t y of the f i r s t order r e v e r s i b l e reaction (Eq. 7) i s given by, dx — = k (a - k x (8) dt x where a i s the i n i t i a l concentration of a-D-glucose and x = concentration of B-form at time t. Upon i n t e g r a t i o n Equation (8) becomes, *1 + h • - - ( -H t \ x - x I where x = equilibrium concentration of the 8-isomer. He showed that t h i s e equation can be expressed i n terms of o p t i c a l r o t a t i o n as, 1 a ,. k^ + k2 = - In ( ° ~ g ) (9) t (a , - a ) t = where a i s the i n i t i a l r o t a t i o n , a. i s the r o t a t i o n at time t and a i s o t <* the equilibrium r o t a t i o n . Therefore, ln (a. - a ) = -(fe, + V)t + In (a - a ) (10) t <* 1 2 o <* Hence a plo t of l n (a, - a ) vs time should be a s t r a i g h t l i n e and the k i n e t i c s of mutarotation can be e a s i l y and conveniently followed by measuring the change i n o p t i c a l r o t a t i o n with time. K i n e t i c s of mutarotation of glucose has also been measured by nonpolarimetric methods such as changes i n r e f r a c t i v e index, volume, e l e c t r i c a l conductivity, pH, ca l o r i m e t r i c methods, s o l u b i l i t y methods, nmr and i n f r a r e d spectra, and g a s - l i q u i d 3 9 , 51 (a), 58,59,60 „ « . * . . . • , « . , . . chromatography . However, the measurement of o p t i c a l r o t a t i o n Is s t i l l the simplest and most s a t i s f a c t o r y means of studying the k i n e t i c s , 51(a) of the reaction It has been shown by many a u t h o r s ^ that the same value of mutarotation constant (k^ + k^) i s obtained by mutarotation of the a- and 6-anomers. Except for two reports, by Reine et a l . ^ and Cox and Natarajan^ 2 for mutarotation i n dimethylformamide, i n a l l the studies of mutarotation of a-D-glucose so f a r , with d i f f e r e n t solvents, ca t a l y s t s and at d i f f e r e n t temperatures, the k i n e t i c s has been found to obey the f i r s t order equation derived by Hudson. Because of these c h a r a c t e r i s t i c s the mutarotation of glucose "has long been a proving ground for the t e s t i n g of k i n e t i c . . . 6 3 theories It has been shown that sugars such as D-glucose which obey Equation (9) contains a- and B-pyranose forms i n s o l u t i o n with only a small proportion of other species. Such mutarotations are designated simple mutarotations"^ and can also be expressed by Equation (11), the exponential form of Equation (9): [ a ] t = Ae"*'* + c (11) where A i s the d i f f e r e n c e between the i n i t i a l and f i n a l r o t a t i o n s , c i s the equilibrium r o t a t i o n and Zc' = (k^ + ) of Equation (9). The mutarotation of sugars containing s u b s t a n t i a l proportions of the furanose forms i n addition to the a- and B-pyranose forms require an equation having two exponential terms. Such mutarotations are c a l l e d complex mutarotations. 40 1.2.2 Forms of Glucose i n Solution 1.2.2.1 C y c l i c Forms of Glucose As mentioned e a r l i e r glucose forms intramolecular c y c l i c hemiacetal structures by the reaction of the carbonyl group with neighboring C-5 hydroxyl group. The solutions and c r y s t a l s of glucose consist p r i m a r i l y of only the anomers of D-glucopyranose. This has been established by X-ray analysis of c r y s t a l s , uv and i n f r a r e d measurements of freeze-dried samples and by nmr of e q u i l i b r a t e d s o l u t i o n s 5 It has been established that i n aqueous s o l u t i o n glucopyranose occurs i n the chair conformation and that the chair conformation with the OH groups attached to C-2, C-3, C-4 and C^OH on C-5 are i n equatorial p o s i t i o n s , 51(a) 64 i s more stable ' . The other chair conformation with those groups i n a x i a l p o s i t i o n s i s l e s s stable because of i n t e r a c t i o n s among 1,3 a x i a l substituents on the same side of pyranose r i n g . The anomeric hydroxyl group on C-l can be e i t h e r a x i a l (a-form VIII) or equatorial (6-form IX). The equatorial p o s i t i o n i s favored by s t e r i c e f f e c t s and the a x i a l p o s i t i o n i s favored by e l e c t r o s t a t i c repulsion between the C+ 0 dipole due to anomeric hydroxyl group and the dipole of the unshared electron pairs on the r i n g oxygen (anomeric e f f e c t ) 5 1 ^ T h e p o s i t i o n of equilibrium i s changed by the d i e l e c t r i c constant of the solvent. In solvents of high d i e l e c t r i c constant, e.g. water (e = 78.5), the dipole-dipole i n t e r a c t i o n i s lowered and the B-form i s favored (a : Bratio = 36 : 64). In DMSO, a highly polar solvent (E = 49), the equilibrium composition i s 65 almost the same . According to Kabayama and Patterson, the aldopyranose having an equatorial OH group f i t s i n t o the tridymite structure of water, whereas the isomer having an a x i a l anomeric OH does n o t 5 " ^ a \ Thus the 41 coordination of the equatorial OH group with solvent also tends to counteract the anomeric e f f e c t . The anomeric e f f e c t i s enhanced i n l e s s polar solvents and thus i n methanol (e = 32.6) a- and g-D-glucopyranoses 64 have equal s t a b i l i t i e s Solvent can also a f f e c t the pyranoid to furanoid r i n g r a t i o . Since bulky groups can be accomodated better on pyranoid than on furanoid r i n g s , reducing the s i z e of r i n g substituents favours furanose forms. Thus i n aprotic solvents l i k e DMSO and dimethylformamide, hydroxyl groups of sugars are less w e l l solvated than i n water and are consequently less bulky. Therefore the mose 64,66 64 proportion of five-membered compounds increases . Although the furanose form has not been detected ( i . e . <1%) i n aqueous solutions of glucose about 2.3% has been detected*^ i n dimethylformamide at 20°C and about 3% 68 i n r e f l u x i n g pyridine , by gas chromatographic methods. 1.2.2.2 A c y c l i c Forms Because of rapid r e v e r s i b l e reactions, d i r e c t chemical methods for measuring the concentration of the a c y c l i c form of a sugar i n s o l u t i o n are not generally s a t i s f a c t o r y . Determination of the concentration of the a c y c l i c form by a p h y s i c a l means, which does not a l t e r any equilibrium i n s o l u t i o n , i s more s a t i s f a c t o r y . Measurement of uv absorption c l e a r l y shows that under normal conditions the concentration of the free carbonyl form must be low"*^ a\ Infrared spectra of solvent-free e q u i l i b r a t e d glucose showed f a i n t absorption bands that can be a t t r i b u t e d to the presence of traces of the free aldehyde sugar together with absorptions c h a r a c t e r i s t i c of a- and B-pyranose forms. The amount of reducible form present i n solutions of several 69 aldohexoses was measured by Los, Simpson and Wiesner by polarographic 42 me thods f o r a three component equilibrium (Eq. 12). The proportion of the h k~2 a Zm~—. Y « B (12) * - l *2 reducible form (y) averaged 0.0026% of the t o t a l D-glucose present, at 25°C i n phosphate b u f f e r of pH 6.9. The rate constants i n 0.0153 M buffer are -3 -1 -3 -1 -1 ^ = 9.8 x 10 min , ^ = 3.2 x 10 min , k_^ - 139 min and k_^ = 78 min \ The small proportion of the 11 y". form i s a r e f l e c t i o n of the extreme r a p i d i t y of the r i n g closure reactions (k ^ and k ^ ) • The absence of a sub s t a n t i a l proportion of the aldehydrol i n aqueous s o l u t i o n shows that the r i n g forms are far more stable than the aldehydrol. Formation of alde-hydrol requires introduction of a hydroxyl group from the solvent. The 18 observation that i n 0, one oxygen atom of D-glucose i s exchanged only 70 18 18 slowly and D-glucose-1- 0 exchanges i t s 0 with water at a rate lower 71 than that of the mutarotation reaction , shows that solvation of the a c y c l i c form i s slower than r i n g interconversion. 1.2.2.3 Ionic Forms Reducing sugars are amphoteric. In aqueous s o l u t i o n they y i e l d cations by addition of a proton, and anions by removal of a proton. The basic properties a r i s e i n large measure from the n u c l e o p h i l i c character of the r i n g 0-atom; the a c i d i c properties are a t t r i b u t a b l e to the OH groups e s p e c i a l l y the anomeric OH group since the anomeric carbon i s attached to r i n g oxygen as w e l l . The pK^ for d i s s o c i a t i o n of the anomeric OH group on D-glucose at 25°C i s 12.1. From t h i s i t can be shown that i n a molar s o l u t i o n of D-glucose at 25°C there are 8.0 umoles of D-glucose anion/ l i t r e at pH 7 5 1 ( a ) . 43 1.2.3 C a t a l y s i s of Mutarotation by Acids and Bases Mutarotation of glucose occurs i n pure water and i s catalyzed by acids and bases"*"*" The acc e l e r a t i o n by acids was f i r s t reported by 72 Erdmann i n 1880 and i t s c a t a l y s i s both by acids and bases was described by U r e c h 7 3 i n 1882. 74 75 Lowry and co-workers ' studied the mutarotation of tetra-O-methyl-a-D-glucopyranose and found that the rate of reaction i s low i n dry pyridine or i n dry c r e s o l but high i n a mixture of the two solvents or i n either solvent when moist. Lowry and S m i t h " ^ ^ concluded that the mutarotation requires an acid c a t a l y s t and a base c a t a l y s t and that amphoteric solvents are complete c a t a l y s t s f o r the process whereas a p r o t i c solvents are not. Thus i n DMSO mutarotation of glucose proceeds extremely 65 slowly, i f at a l l , i n the absence of acids and bases Although early workers a t t r i b u t e d c a t a l y t i c a c t i v i t y to acids and bases, l a t e r work revealed that c a t a l y s i s of mutarotation i s not the exclusive property of H + ions and OH ions. For example, Lowry and co-workers^^ ^  ^ ' 7 ^ showed that molecules of undissociated acids (e.g. a c e t i c a c i d ) , cations of weak bases (e.g. ammonium ion) and anions of weak acids (e.g. acetate ion) have c a t a l y t i c properties. Much the same concept was 52 76 developed by Bronsted and Guggenheim ' and came to be known as general acid and base c a t a l y s i s . I t was found that the rate of mutarotation of a sugar i n the presence of a mixture of several c a t a l y s t s may be represented by an equation of the type, <*1 + V = \0[*2°] + V.[HV + \ [ Bn ] ( 1 3 ) where the symbols i n brackets represent the concentrations ( a c t i v i t i e s ) of the c a t a l y s t s and the c o e f f i c i e n t s k^ and k^ represent the c a t a l y t i c 44 a c t i v i t i e s of the acid and base c a t a l y s t s , r e s p e c t i v e l y . This kind of rate equation a r i s e s from a stepwise process where the acid c a t a l y s t and the base c a t a l y s t act separately. I f they act together, both the acid and the base c a t a l y s t s take part i n the t r a n s i t i o n state with the addition of a proton at one point i n the molecule and with the elimination of a proton at another point. T h e o r e t i c a l l y , the rate constant for a concerted reaction i n v o l v i n g several acid-base combinations y i e l d an equation of the f o r m 5 1 ^ k = E X Zc„ fc [HA. ] [B ] (14) n. D i n J n j n J where the observed v e l o c i t y depends on the product of the v e l o c i t i e s of acid and base c a t a l y s t s . Numerous workers have examined the rate constants for the muta-ro t a t i o n of D-glucose i n the presence of ac e t i c ac i d and sodium acetate i n an attempt to ascer t a i n whether Equation (13) or (14) applies. Although the two equations d i f f e r widely, i t i s not easy to d i s t i n g u i s h experimentally between them 5 1^^_ i t i s probable that the concerted mechanism i s of no s i g n i f i c a n c e i n the acid and base catalyzed mutarotation of sugars i n aqueous so l u t i o n . In solvents of low d i e l e c t r i c constants the formation of i o n i c intermediates becomes les s favoured,and the concerted mechanism may apply (as discussed i n Section 1.2.5.1.3). Some c a t a l y t i c constants for the mutarotation of D-glucose i n d i f f e r e n t solvents are given i n Table I I I . Comparison of r e s u l t s f or benzoate c a t a l y s i s i n d i f f e r e n t solvents shows that there i s almost no rate increase on going from water to DMSO or DMF as solvent. Thus for c a t a l y s i s by benzoate the c a t a l y t i c constant i n DMSO i s approximately equal to that i n water (allowing for small temperature TABLE I I I 4 -1 -1 CATALYTIC CONSTANTS (10 k _ LITRE MOLE SEC ) FOR MUTAROTATION OF GLUCOSE OR TETRAMETHYLGLUCOSE cat Catalyst Water Solvent DMSO DMF Dioxan Benzene Hydronium Ion Ace t i c Acid Ammonium Ion Benzoate Ion Acetate Ion Hydroxide Ion 2-Hydroxypyridine Mutarotase <v.s77 1 0 0 ( 2 5 ° C ) 0 . 9 1 ( 1 8 ° C ) 0 . 2 0 ( 2 0 ° C ) 5 . 8 ( 1 8 ° C ) 1 0 . 2 ( 1 8 ° C ) < > ^ 6 5 7 5 ( b ) 0r,.52 0r,,52 U 78 1 4 5 x 1 0 ( 2 5 ° C ) o.s79 1 4 0 ( 2 5 ° C ) 1 0 . 2 ( 2 5 ° C ) 6 2 8 3 . 7 ( 2 5 ° C ) 1 . 4 7 ( 3 0 ° C ) 1 6 2 5 x 1 0 4 ( 2 1 . 5°Z)V* b 0 . 8 8 ( 3 0 ° C ) 7 9 O«N 6 2 4 8 ( 3 0 ° C ) 7 9 7 0 0 ( 2 5 ° C ) 7 7 , - 4 - 1 Rate constant for c a t a l y s i s by pure water at 2 5 C i s 4 . 1 x 1 0 sec Concentration of tetramethylglucose = 0 . 0 9 1 M Concentration of glucose = 0 . 1 1 M 46 difference) and that i n DMF i s approximately eight times large. This i s i n sharp contrast to studies i n v o l v i n g anionic bases and nucleophiles i n dipolar a p r o t i c solvents where rate increases of approximately 10 f o l d 62 have been commonly observed . This unusual r e s u l t could be due to the fact that the proton trans f e r i n water from the base to the glucose molecule could occur v i a hydrogen bonded water molecule, there being no need for su b s t a n t i a l desolvation of the base. The unusually high a c t i v i t y of 2-hydroxypyridine i n benzene w i l l be discussed i n Section 1.2.5.1.3. 1.2.3.1 Lewis Acids as Catalysts There are a few reports of Lewis acid c a t a l y s i s of the glucose 82 mutarotation. According to Broser and Ruecker A l C l ^ and ZnCl^ i n a c e t i c acid/sodium acetate b u f f e r pH 3.7, showed c a t a l y t i c a c t i v i t y due to t h e i r a ction as Lewis acids. MgC^ showed minor c a t a l y t i c e f f e c t and NaCl had 83 no e f f e c t at a l l . According to Mitzner and Behrenwald mutarotation of 2+ sugars i s catalyzed by Ni ion due to the formation of a sugar-metal ion 84 complex. Bobbio et a l . have reported that the mutarotation of glucose 2+ i s catalyzed by Cu due to formation of a complex i n which C-l of the sugar i s involved. 1.2.3.2 C a t a l y t i c E f f e c t of Sa l t s O r d i n a r i l y , s a l t s of strong acids and strong bases have l i t t l e 52 influence on the rate of mutarotation i n water . Thus mutarotation of glucose i s retarded s l i g h t l y by sodium chloride or l i t h i u m c h l o r i d e " ^ ^ 5 ^ Under some conditions however, large e f f e c t s are observed. Thus Eastham 85 and co-workers found that the mutarotation of tetramethylglucose i n 47 p y r i d i n e or i n nitromethane i s slow, but that i t i s enhanced by a f a c t o r of ten on addition of 2 mM l i t h i u m perchlorate. I t has been suggested that the s a l t s t a b i l i z e s the t r a n s i t i o n state by formation of an ion-pair 86 - + complex. Further, Rony et a l . have shown that ion p a i r s such as PhO 'NR^  are e f f e c t i v e general base ca t a l y s t s for mutarotation of tetramethylglucose i n benzene. 1.2.4 Isotope E f f e c t s i n Mutarotation R e a c t i o n s 5 1 ' ^ Deuterium isotope e f f e c t s , k^/k^, a r i s e from a combination of k i n e t i c and solvent isotope e f f e c t s . K i n e t i c isotope e f f e c t s are caused by differences i n the energy required for a l t e r a t i o n of the normal and i s o t o p i c bond i n the corresponding t r a n s i t i o n states; solvent isotope e f f e c t s can e x i s t when the i s o t o p i c compound i s used both as a reactant and as a solvent. The observed isotope e f f e c t s may have values smaller than, equal to, or greater than unity. The differences i n the rates of mutarotation of sugars i n water and i n deuterium oxide have provided valuable means for studying the mechanisms of mutarotation reactions. 1.2.4.1 K i n e t i c Isotope E f f e c t A k i n e t i c isotope e f f e c t i s large when the bond j o i n i n g the i s o t o p i c atom to the substrate i s formed or i s broken i n the rate-determining step. In general, the stronger the a l t e r e d bond, the greater the isotope 88 e f f e c t . O r d i n a r i l y , t h e heavier isotope gives the lower re a c t i o n rate , and, hence,values of k^/k^ greater than unity are designated as normal isotope e f f e c t s and values l e s s than unity as inverse isotope e f f e c t s . But i f the proton t r a n s f e r occurs a f t e r the r a t e - c o n t r o l l i n g step i t w i l l have no primary k i n e t i c consequence. I f proton addition occurs 48 before the r a t e - c o n t r o l l i n g step and no proton t r a n s f e r occurs i n the t r a n s i t i o n state ( i . e . no k i n e t i c isotope e f f e c t ) , the reaction i s subject to hydronium ion c a t a l y s i s and an inverse isotope e f f e c t (due to solvent) i s observed. 1.2.4.2 Solvent Isotope E f f e c t Change of solvent from H^ O to causes changes i n the degree of i o n i z a t i o n and s o l v a t i o n both of the reactants and the c a t a l y s t . The reactions that involve a rapid pre-equilibrium, i n which proton t r a n s f e r from the c a t a l y s t to the substrate (HS) occurs before rate-determining step, w i l l have a higher concentration of the intermediate conjugate acid i n D^O than i n H^ O because DgO + i s a stronger acid than H^o"*". Therefore, whether the observed isotope e f f e c t i s normal or inverse w i l l depend on the nature of rate-determining step. If the rate-determining step does not involve a proton t r a n s f e r , an inverse isotope e f f e c t w i l l be observed due to the higher concentration of protonated substrate (H$D) i n D^ O and the reaction i s subject to s p e c i f i c a c i d c a t a l y s i s . HS + HO ^ ^ HS~H + H 0 3 2 HS + D 30 M HSD + D 20 + HSH • products + HSD • products However, a proton (or deuteron) t r a n s f e r i n the rate-determining step w i l l give r i s e to a normal k i n e t i c isotope e f f e c t f or that step. Hence the observed isotope e f f e c t can be normal i f the k i n e t i c isotope e f f e c t of the 89 rate-determining step i s large enough . Such mechanisms are subject to 49 general acid c a t a l y s i s . The r e s u l t s of deuterium isotope e f f e c t studies with glucose muta-ro t a t i o n w i l l be discussed i n the next section where the possible mechanisms of mutarotation are considered. 1.2.5 The Mechanism of Mutarotation of Glucose There are at l e a s t three mechanisms possible for the mutarotation f i 58,65 of glucose (1) Mutarotation can proceed through the free aldehyde form (y-glucose): • 0 H OH a-Glucose y-Glucose 0 H B-Glucose OH Acids can catalyze the reaction by protonation of the r i n g oxygen; thereby the ease with which C-0 bond can be broken i s increased. Bases can catalyze by removing the proton from the anomeric hydroxyl group. 90 (2) According to Christansen , the reaction proceeds as a r e s u l t of the removal of the C - l hydrogen, although i t i s d i f f i c u l t to understand why C-l hydrogen should i o n i z e over hydrogen on the anomeric OH group, H + H OH OH OH 50 Acid c a t a l y s i s could presumably r e s u l t from the p r i o r protonation of the neighboring oxygen. (3) Another pathway i s the d i r e c t displacement of anomeric OH by a water molecule or hydroxide ion. I t can occur v i a carbonium ion formation i n the case of a c i d c a t a l y s i s . H OH, + H A l l the evidence a v a i l a b l e from mechanistic studies i n s o l u t i o n supports Mechanism (1). The best evidence for t h i s i s the observation 18 that glucose-1- 0 undergoes oxygen exchange with water more than 30 times 58,71 slower than the mutarotation reaction" I f the r e a c t i o n involved elimination and addition of a OH group at the anomeric carbon atom 18 18 (Mechanism (3)), the rate of exchange of 0 i n D-glucose-1- 0 should be equal to the rate of mutarotation, and according to Mechanism (2), the anomeric hydroxyl group should not exchange at a l l . Further, i f Mechanism (3) occurs by S^2 type displacement of anomeric OH group by OH ions, then there should not be any deuterium isotope e f f e c t . While, acid catalyzed carbonium ion type mechanism should give r i s e to an inverse isotope e f f e c t due to solvent as discussed above. However, normal deuterium isotope e f f e c t s have been observed for glucose mutarotation under a l l conditions studied f 51(b) so f a r 18 The small proportion of 0 exchanged may occur by r e v e r s i b l e formation of a gem-diol of the intermediate a c y c l i c sugar. The existence 51 of the free aldehyde form has been confirmed by p h y s i c a l and chemical methods as mentioned i n Section 1.2.2.2. The observation that exchange of 18 0 i s r e l a t i v e l y slow indicates that the nascent carbonyl group of the a c y c l i c intermediate reacts faster, intramolecularly, with neighboring hydroxyl groups than with the hydroxyl groups of the solvent. To explain the r a p i d i t y of the r i n g closure reaction and the slow exchange of anomeric 91 oxygen with solvent, I s b e l l et a l . postulated that the c y c l i c sugar passes through a so-called "pseudo-acyclic" intermediate during mutarotation. 1.2.5.1 S p e c i f i c Mechanisms The a c y c l i c intermediate postulated i n Mechanism (1) can be formed by a c i d or base c a t a l y s i s or by pure water ac t i n g as a c a t a l y s t . Thus i t can be formed by several s p e c i f i c mechanisms which are discussed below. 1.2.5.1.1 C a t a l y s i s by Acids 0 I f a l l rate constants are of the same order t h i s mechanism leads to complicated k i n e t i c s . But f i r s t order k i n e t i c s has been observed f o r glucose 52 mutarotation as mentioned e a r l i e r . I f the f i r s t step i s rapid and r i n g opening reaction i s r a t e -determining, then the rate i s proportional to [glucose] [H^O] . Therefore the reaction i s subject to s p e c i f i c acid c a t a l y s i s and should show an inverse isotope e f f e c t due to higher concentration of deuteronated glucose i n D 20. On the other hand, i f the conjugate base A takes part i n the r a t e -determining step by removal of the proton on anomeric OH group (Equation 16), the rate i s proportional to [glucose][HA] and the reaction i s subject to general acid c a t a l y s i s . Further, the second step would show a normal k i n e t i c isotope e f f e c t and thus the o v e r a l l reaction can have a normal isotope e f f e c t . Hence the mechanism shown i n Equation (16) i s a possible route for the mutarotation c a t a l y s i s by acids. For c a t a l y s i s by strong + acids HA = H^ O and A = H^O. For strong a c i d c a t a l y s i s i n water, deuterium 92 isotope e f f e c t observed i s about 1.37 at 25°C while f o r weak acids (e.g. a c e t i c acid) i t i s around 2 . 6 5 1 ^ a t 25°C. The higher isotope e f f e c t with the weak acid i s due to the formation of a stronger H-A bond + (compared with weaker H-OH^bond) i n the t r a n s i t i o n state. Another possible mechanism for acid c a t a l y s i s i s shown i n Equation (17), where the r i n g opening occurs during H-A bond rupture. This mechanism also leads to general acid c a t a l y s i s , f i r s t order k i n e t i c s and a normal 53 17) isotope e f f e c t . But Capon and Walker''"' have shown that for such a mechanism, the rate constant k for the reverse of rate-determining step would be R higher than the d i f f u s i o n controlled l i m i t f o r a strong a c i d , weak acid or even for the water catalyzed mutarotation of glucose. Hence the mechanism shown i n Equation (16) i s the preferred mechanism for acid catalyzed mutarotation. However, they also showed that f o r the water catalyzed reaction by Mechanism (16) the rate constant for the rate-determining step would be above the rate constant for d i f f u s i o n . The mechanism i n Equation (16) (as w e l l as Equation 17) i s therefore not possible for water i t s e l f . Electron withdrawing substituents at 6-position ( i . e . 6-substituted 6-deoxy glucoses) decreased the rate of mutarotation by oxonium ions and by 93 water . The p value of -2.87 f o r acid c a t a l y s i s i s consistent with a mechanism i n which proton t r a n s f e r to the ri n g oxygen atom i s further advanced than breaking of the r i n g C ( l ) - 0 bond. The p value f o r water c a t a l y s i s was -2.92. In a ddition, 2-substituted 2-deoxy-D-glucoses gave r e s u l t s consistent with Mechanism i n Equation (16). Thus 2-deoxy-D-glucose showed a higher rate, and conjugate acid of 2-amino-2-deoxy-D-glucose showed a lower rate for mutarotation than D-glucose i t s e l f . 1.2.5.1.2 Base C a t a l y s i s Again, two mechanisms, s i m i l a r to those considered f or acid 19) Both the mechanisms would show f i r s t order k i n e t i c s , general base c a t a l y s i s and normal isotope e f f e c t s . The deuterium isotope e f f e c t s observed f o r base c a t a l y s i s are generally higher than that observed for strong acid c a t a l y s i s because of the stronger H-B bond formed. The observed k„/1 values for acetate, p y r i d i n e and hydroxide (without c o r r e c t i n g for 94 95 87 c a t a l y s i s by glucosate anion) are 2.38 , 2.86 and 1.83 respectively. Capon and Walker showed that f o r mechanism i n Equation (19) the rate constant for the reverse of the r a t e - c o n t r o l l i n g step would be close to or K greater than the d i f f u s i o n c o n t r o l l e d l i m i t , while for mechanism i n Equation (18) the rate constant for the slow step i s w e l l within the d i f f u s i o n c o n t r o l l e d l i m i t . Hence mechanism i n Equation (18) i s favoured for base c a t a l y s i s . 55 In contrast to acid c a t a l y s i s , e l e c t r o n withdrawing substituents at the 6-position increased the rate of mutarotation catalyzed by bases. The p values for c a t a l y s i s by T r i s and hydroxide ions are +3.97 and +6.22 resp e c t i v e l y 1.2.5.1.3 Concerted Mechanism The mechanisms discussed above for acid and base c a t a l y s i s of mutarotation i n aqueous s o l u t i o n involve i o n i c intermediates which can be e a s i l y s t a b i l i z e d by polar water molecules. But i n solvents of low d i e l e c t r i c constant the formation of free ions becomes less l i k e l y , s o that a concerted mechanism may be favoured even for reactions which do not follow i t i n water or s i m i l a r solvents. In fact, the concerted mechanism was f i r s t 96 put forward by Lowry on the basis of observations on the mutarotation of tetramethylglucose i n media of low d i e l e c t r i c constant. As mentioned e a r l i e r (Section 1.2.3), t h i s r eaction was very slow i n dry pyridine (possessing no acid properties) or i n dry c r e s o l (possessing hardly any basic p r o p e r t i e s ) , but was rapid i n a mixture of the two solvents or i n either solvent when moist, suggesting that both an acid and a base must take part i n the reaction. 97 Swain and Brown studied the same reaction i n d i l u t e benzene (e = 2.3) solutions of amine and phenol. They found that the reaction was k i n e t i c a l l y of the 3rd order, the v e l o c i t y being proportional to the product of the concentrations of phenol, amine and tetramethylglucose as 98 expected from Equation (14). But l a t e r Pocker pointed out that the rate proportional to the product of concentrations of amine and phenol could be _ + a t t r i b u t e d to base c a t a l y s i s by phenoxide ion i n an ion pair such as PhO -NH^ R. 86 - + This i s supported by the demonstration that ion-pairs such as PhO -NR^ are e f f e c t i v e basic ca t a l y s t s for mutarotation of tetramethylglucose i n 56 benzene, although they contain no a c i d i c group. 80 Swain and Brown also showed that 2-hydroxypyridine i s a powerful s p e c i f i c c a t a l y s t f or the mutarotation reaction; at a concentration of 0.001 M i t i s 7000 times as e f f e c t i v e as a mixture of 0.001 M pyridine and 0.001 M phenol, though i t i s only one ten-thousandth as strong a base as pyridine and one-hundredth as strong an acid as phenol. Since, further, the v e l o c i t y i s proportional to the f i r s t power of the concentration of 2-hydroxy-pyridine, i t i s c l e a r that the operation of the concerted mechanism where both the proton transfers occur simultaneously (Equation 20) i s f a c i l i t a t e d by the presence of an a c i d i c and a basic group i n the same cat a l y s t molecule. On the other hand, 3- or 4- hydroxypyridines where the nitrogen and hydroxyl groups are too far apart, do not show b i f u n c t i o n a l c a t a l y s i s , 80 and consequently they are poor c a t a l y s t s H Since two 0-H bonds are broken i n the t r a n s i t i o n state one would expect a deuterium isotope e f f e c t higher than that observed for bases. In fact,the observed isotope e f f e c t f o r 2-hydroxypyridine c a t a l y s i s of the 99 mutarotation of tetramethylglucose i n benzene i s 3.5 It has been shown that b i f u n c t i o n a l c a t a l y s t s are e f f e c t i v e i n non-polar solvents only when they can i n t e r a c t with the substrate without 100,101,102 , . . i . formation of high energy d i p o l a r intermediates ; this implies that they can e x i s t i n two tautomeric forms of comparable energy, and hence 57 are also described as tautomeric c a t a l y s t s . Other examples of b i f u n c t i o n a l c a t a l y s t s i n non-polar media are carboxylic acids, pentane-2,4-dione, and . 102 pyrazole The above cat a l y s t s act as e f f e c t i v e b i f u n c t i o n a l c a t a l y s t s i n benzene because of the absence of e f f e c t i v e a c i d i c and basic groups on the solvent. No such b i f u n c t i o n a l c a t a l y s i s has been observed for these _ 79,80,103 . _ ..... n . „ „ cat a l y s t s i n water since i t can s t a b i l i z e any polar intermediates formed and also can act as an acid or base. Thus 2-hydroxypyridine does not show b i f u n c t i o n a l c a t a l y s i s i n water and the observed deuterium isotope 99 e f f e c t i s 2.3 (compared with feH/^ = 3.5 i n benzene, 2.91 i n DMS0-D20 -3 mixture with D 20 concentration = 22.2 mol dm and 2.53 i n DMS0-D20 -3 79 mixture with D 20 concentration = 44.4 mol dm ) There are other factors that determine whether c a t a l y s i s w i l l occur by a stepwise or a concerted mechanism. For example, Fiandanese and 104 Naso have shown that benzamidine,a p o t e n t i a l b i f u n c t i o n a l c a t a l y s t , i s a powerful c a t a l y s t for tetramethylglucose mutarotation i n benzene. They also showed that c a t a l y t i c a c t i v i t y i s not due to a b i f u n c t i o n a l action, but due to i t s a c t i v i t y as a base. They concluded that the presence of both a c i d i c and basic centres s u i t a b l y located i n a tautomeric system does not represent a s u f f i c i e n t condition for a b i f u n c t i o n a l intervention i n the mutarotation process. With benzamidine,the high strength and r e a c t i v i t y of the basic centres make i r r e l e v a n t the advantage occurring from i t s favourable p o s i t i o n r e l a t i v e to the a c i d i c function. 1.2.5.1.3.1 C a t a l y s i s by Water It was mentioned i n Section 1.2.5.1.1 that although the substituent effects for acid c a t a l y s i s and water c a t a l y s i s are the same, the 58 rate constant for the rate-determining step, for water c a t a l y s i s by the mechanism i n Equation (16), would be above the rate constant f o r d i f f u s i o n . Also,the diff e r e n c e i n the substituent e f f e c t s between water and base catalyzed mutarotation of the 6-substituted 6-deoxy glucoses makes i t un l i k e l y that these reactions follow the same mechanism. Ad d i t i o n a l evidence against the conventional stepwise mechanism has come from the observation that some of the necessary f r a c t i o n a t i o n factors f o r mutarotation of a-D-tetramethylglucose i n mixed H2O-D2O solvents were d i f f i c u l t to i n t e r p r e t when the stepwise mechanism was c o n s i d e r e d 1 ^ 5 . However,a c y c l i c concerted mechanism i n v o l v i n g two or three water molecules led to p l a u s i b l e f r a c t i o n a t i o n f a c t o r s . D i r e c t evidence for p a r t i c i p a t i o n of several water molecules i n the t r a n s i t i o n state has come from the e f f e c t of solvent composition, i n mixtures of water and an organic solvent, on the mutarotation reaction. 65 For example, Ballash and Robertson found that i n DMSO-water mixtures, the order with respect to water i s about three. But for acid c a t a l y s i s under s i m i l a r conditions the order with respect to water was zero. Chin and Huang 1^ 5 studied the mutarotation of a-D-tetramethylglucose i n aqueous dioxan and DMSO solutions. They found that f or both solvents the order with respect to water was about two. For pyridine catalyzed reaction the order with respect to water was one, suggesting that one molecule of water has been replaced by a molecule of base. Hence the following concerted mechanism invo l v i n g two water molecules, one acting as an acid and the other as a base, has been postulated (see page 59). It i s important to note that a c y c l i c t r a n s i t i o n state i n v o l v i n g a b i f u n c t i o n a l c a t a l y s t can catalyze a reaction i n a concerted ( i . e . correspondence between the proton transfers within times of <10 ^ sec) 59 \ H / H \ 0 / ' H have calculated the a c t i v a t i o n energies f o r proton transfers by concerted and stepwise mechanisms f o r the r e v e r s i b l e hydration of 1,3-dichloroacetone. Their r e s u l t s i n d i c a t e that the a c t i v a t i o n energy for the concerted mechanism i s about twice as great for the stepwise process. Hence they concluded that the l a t t e r would be preferred. This conclusion was supported by the fa c t that the observed deuterium isotope e f f e c t for the uncatalyzed reaction i s 2.7. This i s low compared with the abnormally high isotope e f f e c t expected for the concerted process. The same argument may apply for the mutarotation reaction. For example, although the observed deuterium isotope e f f e c t f o r water c a t a l y s i s i s about 3 . 8 5 1 ^ a t 25°C, the isotope e f f e c t expected 99 for a concerted process, according to c a l c u l a t i o n s by Schowen , i s 12. Hence Chin and Huang1'"'"' have suggested that water catalyzed mutarotation occurs by an intimate stepwise mechanism rather than a concerted process. 1.2.6 Heterogeneous C a t a l y s i s of the Glucose Mutarotation A l l the r e s u l t s discussed above r e f e r to studies of glucose mutarotation c a r r i e d out using homogeneous c a t a l y s t s . On the other hand, there are only a few reports of heterogeneous c a t a l y s i s of the glucose 108 mutarotation by s o l i d c a t a l y s t s . Thus Tanabe et a l . have shown that s i l i c a - a l u m i n a , n i c k e l sulphate, and s o l i d phosphoric acid catalyze the 60 mutarotation of tetramethyl-a-D-glucopyranose i n benzene s o l u t i o n , with the observed a c t i v i t y depending on the acid properties of the c a t a l y s t . Silica-alumina which has the highest acid strength showed the greatest c a t a l y t i c a c t i v i t y . They suggested that s i l i c a - a l u m i n a and n i c k e l sulphate may act as b i f u n c t i o n a l c a t a l y s t s since they both have a c i d i c and basic s i t e s . N i c k e l sulphate heat treated at 250°C showed higher c a t a l y t i c a c t i v i t y than that heated at 350°C, i n d i c a t i n g that the c a t a l y t i c a l l y a c t i v e s i t e s are the Bronsted type but not the Lewis type. 109 According to Rustamov and Tyrina the mutarotation of glucose i n water was accelerated by n i c k e l prepared by reduction of Ni2(0H)2C0 3 i n hydrogen at 300°C. The a c t i v a t i o n energies f o r the forward and reverse reactions were decreased from 13.65 to 7.32 and 14.63 to 9.38 Kcal./mole respectively. Heating n i c k e l in vacuo to remove the adsorbed hydrogen did not a f f e c t i t s a c t i v i t y . Rustamov and Usmanov 1 1^ have shown that the mutarotation of glucose i n water i s not catalyzed by anion exchange resins AN-2F, EDE-10P and PE-9. According to Schmid and B a u e r 1 1 1 mutarotation of a-D-glucose i s catalyzed by f i n e l y divided copper but the a c t i v a t i o n energy for hetero-geneous copper c a t a l y s i s (19 Kcal./mole) agreed, w i t h i n experimental error, with that for homogeneous c a t a l y s i s by water. Hence they concluded that heterogeneous copper c a t a l y s i s i s by water molecules adsorbed on the copper surface. Hence i t i s clear that there i s only a l i t t l e work done on heterogeneous c a t a l y s i s of glucose mutarotation and they are also incomplete. Because of the lack of any systematic study of the k i n e t i c s and mechanisms of the c a t a l y s i s of glucose mutarotation by s o l i d c a t a l y s t s , i t was decided to study the e f f e c t of aluminum oxide on the above reaction. It should, 61 for the f i r s t time, provide information on the k i n e t i c s and mechanism of an alumina catalyzed simple organic reaction i n s o l u t i o n . The r e s u l t s obtained can be compared with the r e s u l t s already a v a i l a b l e on the mechanism of homogeneous c a t a l y s i s , to determine s i m i l a r i t i e s and differences i n the mechanisms. The same approach, to understand the action of s o l i d acid/base c a t a l y s t s , has been used i n the study of alcohol dehydration on s o l i d surfaces based on advances made i n understanding i t s mechanism i n homogeneous systems^. Since the mutarotation reaction i s more s e n s i t i v e to Bronsted acids than any of the reactions that have been studied on alumina, i t could be a s e n s i t i v e probe for the presence of c a t a l y t i c a l l y a c t i v e Bronsted acid s i t e s on the surface. Deuterium isotope e f f e c t studies can provide evidence for the presence of general or s p e c i f i c acid-base c a t a l y s i s , c y c l i c or a c y c l i c mechanism and b i f u n c t i o n a l c a t a l y s i s . Presence of b i f u n c t i o n a l c a t a l y s i s would i n d i c a t e the presence of a c i d i c and basic s i t e s s u i t a b l y located on the surface. The solvent chosen was dimethyl sulfoxide. I t i s a highly polar (e = 49) a p r o t i c solvent which dissolves glucose e a s i l y and does not catalyze the mutarotation^"*. Hence c a t a l y s i s can be e a s i l y arrested by f i l t e r i n g the s l u r r y . The non-aqueous solvent w i l l also enable the study of the e f f e c t of water on the surface reaction. Commercially a v a i l a b l e Woelm alumina neutral for t h i n layer chromatography (hence f o r t h referred to as alumina neutral) was chosen as the c a t a l y s t , s i n c e i t i s widely a v a i l a b l e and i s s i m i l a r to the alumina 12 used by Posner (Woelm 200 neutral chromatographic alumina) i n h i s studies of alumina catalyzed organic reactions. The p h y s i c a l c h a r a c t e r i s t i c s of alumina n e u t r a l and other aluminas 62 prepared by dehydration of alumina neutral are given i n the next section. This thesis i s then developed by a general study of the k i n e t i c s of glucose mutarotation by alumina. It i s followed by a study of glucose adsorption onto the surface, and then, by in v e s t i g a t i o n s into the nature the c a t a l y t i c a l l y a c t i v e s i t e s . F i n a l l y , the mechanism of the surface catalyzed reaction i s discussed. CHARACTERIZATION OF ALUMINAS 64 II CHARACTERIZATION OF ALUMINAS Alumina neutral and other aluminas prepared by p y r o l y s i s of alumina neutral at d i f f e r e n t temperatures (Section XII) were characterized by determining surface areas, pore s i z e and p a r t i c l e s i z e d i s t r i b u t i o n s , c r y s t a l l i n e structure and trace impurities present. The theory and experimental methods are described i n Appendix A. Hence only the r e s u l t s of these determinations are given i n t h i s s e c t i o n . II.1 Surface Areas 112 The surface areas were determined by the multipoint BET method using nitrogen gas as adsorbate. The r e s u l t s given i n Table IV show that s p e c i f i c surface areas have decreased with increase of p y r o l y s i s temperature. The extent of s i n t e r i n g ( i . e . adhesion of the p a r t i c l e s of a s o l i d to form 113 aggregates) has increased markedly above the Tammann temperature (^870°C) and aluminas pyrolyzed at 1100°C and 1250°C possess low s p e c i f i c surface 18,32 areas II.2 Pore Size D i s t r i b u t i o n s The adsorption and desorption isotherms of alumina neutral and alumina pyrolyzed at 800°C showed the presence of both c a p i l l a r y A A u <- • • 113,114,115,116 . . _ condensation and h y s t e r i s i s . These aluminas are therefore porous and the pore s i z e d i s t r i b u t i o n s , calculated using the desorption isotherms, are given i n F i g . 3. This fi g u r e shows that most of the pore o volume i n alumina neutral i s i n pores of r a d i u s ^ 2 3 A while most of the o pores have r a d i i between 18 and 27 A. With alumina pyrolyzed at 800 C o most of the pore volume i s i n pores of radius ^ 28 A while most of the pores 65 TABLE IV BET SURFACE AREAS Sample Preparation Alumina Neutral Dehydrated at 150 ± 5°C 0.01 mm pressure, 2 days Dehydrated at 600 ± 50°C 4 hours under dry N 2 Dehydrated at 800 ± 50°C 4 hours under dry N 2 Dehydrated at 1100 ± 50°C 3 hours under dry N 2 Dehydrated at 1250 ± 50°C 6 hours under dry N 0 BET Surface Area m /g 140 ± 5 130 ± 5 116 ± 4 100 ± 4 14 ± 1 6.2 ± 0.1 o have r a d i i between 22 and 31 A. Further, the shapes of the hysteresis . ^ i • - . ,114,115 loops can be taken to i n d i c a t e that the pores are c y l i n d r i c a l On the other hand, the adsorption and desorption isotherms of alumina pyrolyzed at 1250°C showed no c a p i l l a r y condensation or h y s t e r e s i s . This indicates the absence of a porous structure and i s due to s i n t e r i n g e i • . u 113,114,115,116 of alumina p a r t i c l e s at high temperatures . 66 F i g . 3 Pore Size D i s t r i b u t i o n s of (a") Alumina Neutral and (b) Alumina Pyrolyzed at 800°C f o r 4 hours 67 II.3 P a r t i c l e Size D i s t r i b u t i o n s II.3.1 P a r t i c l e Size by use of an Electrozone Celloscope The p a r t i c l e s i z e d i s t r i b u t i o n s of alumina neutral and sintered (1250°C) alumina were determined using an Electrozone Celloscope and are given i n F i g . 4. The c h a r a c t e r i s t i c s of the pl o t s are shown i n Table V. They show that the mean p a r t i c l e diameters are 7.92 microns for alumina neutral and 7.39 microns for 1250°C alumina. The approximately 7% decrease i n the p a r t i c l e s i z e on p y r o l y s i s at 1250°C i s due to s i n t e r i n g of p a r t i c l e s . Further, >99% of the p a r t i c l e s have diameters between 2.8 and 30 microns. TABLE V CHARACTERISTICS OF THE PARTICLE SIZE DISTRIBUTION PLOTS P a r t i c l e Diameter (microns) Log mean Mode Median Alumina Neutral 7.92 8.49 7.74 1250 °C Alumina 7.39 6.59 7.22 II.3.2 P a r t i c l e Size by use of an Electron Microscope The alumina p a r t i c l e s were also observed with a Scanning Electron Microscope. As described i n other r e p o r t s ^ 1 3 ' , p a r t i c l e s of sintered alumina (Fig. 6) show smoother and rounder edges than those of alumina neutral (Fig. 5). Further, the p a r t i c l e s shown i n F i g s . 5 and 6 have average diameters ranging from 9 to 12 microns which f a l l w i t h i n the range 68 P a r t i c l e Diameter (centimicrons) F i g . 4 P a r t i c l e Size D i s t r i b u t i o n s of (a) Alumina Neutral and (b) Alumina Sintered at 1250°C for 6 hours Fig. 5 Electron Micrograph of Alumina Neutral for TLC (x 4000) Fig. 6 Electron Micrograph of Sintered C1250°C) Alumina (x 8000) 70 of diameters determined using Electrozone Celloscope. 11.4 C r y s t a l l i n e Structure of Aluminas X-ray powder photographs showed that alumina neutral i s mainly y-alumina with some <- (and X-) alumina. On the other hand, alumina sintered at 1250°C contains only a-alumina^ 7. 11.5 Trace Impurities Present Trace metals present i n the alumina samples were determined using an Inductively Coupled Argon Plasma Spectrograph (by Can Test L t d . ) . The r e s u l t s showed that there i s 0.2% trace metals (by weight) i n alumina neutral, the main impurities being Ca (0.11%), Fe (0.03%), Na (0.02%), and Mg (0.02%). PRELIMINARY KINETIC STUDIES 72 III PRELIMINARY KINETIC STUDIES The k i n e t i c s of the heterogeneous c a t a l y s i s was c a r r i e d out by s t i r r i n g a 0.05 M so l u t i o n of a-D-glucose i n dimethyl sulfoxide with alumina (26.7 mg/ml) at 25.0°C as described i n the Experimental Section. A t y p i c a l 3 f i r s t order p l o t of l n 10 (a^ - a m) versus time (mins) , where a^ = o p t i c a l r o t a t i o n at time t and a^ = o p t i c a l r o t a t i o n when equilibrium i s f i n a l l y reached, for the c a t a l y s i s of mutarotation of a-D-glucose by alumina neutral i s shown i n F i g . 7a, and that f o r 8-D-glucose i n F i g . 7b. The r e p r o d u c i b i l i t y of the r e s u l t s was excellent (see Section XI). There are two c h a r a c t e r i s t i c features i n these p l o t s . F i r s t , with both a- and 6-D-glucose, there i s an i n i t i a l rapid decrease i n o p t i c a l r o t a t i o n . In the case of 6-D-glucose the o p t i c a l r o t a t i o n decreases i n i t i a l l y , although the actual mutarotation of t3-D-glucose which leads to the formation of a-D-glucose causes an increase i n o p t i c a l r o t a t i o n . Secondly, unlike the w e l l known l i n e a r f i r s t order p l o t s obtained f o r homogeneous c a t a l y s i s of glucose mutarotation i n s o l u t i o n 5 1 , the f i r s t order plots are curved throughout the runs. This continued curvature i n the f i r s t order plots means either the surface reaction i s not f i r s t order i n glucose or the ca t a l y s t i s undergoing progressive deactivation during the reaction. For both these pl o t s i n F i g . 7 the experimental i n f i n i t y value of the o p t i c a l r o t a t i o n (a r o) was used. It was the same i n both cases (+0.124° at X = 365 nm) and was l e s s than the i n f i n i t y value observed (+0.144°) f or homogeneous c a t a l y s i s (see Experimental Section). This difference i n the i n f i n i t y values could be due to adsorption of glucose onto alumina, which might also give r i s e to the i n i t i a l rapid decrease i n o p t i c a l r o t a t i o n i n both p l o t s , or because there are side reactions occurring on the surface of alumina which might also be responsible f o r the non-linear f i r s t order p l o t s . 73 F i g . 7 F i r s t Order K i n e t i c Plots f or Mutarotation of (a) a-D-Glucose and (b) g-D-Glucose by Alumina Neutral 74 Therefore an analysis of the reaction products i n s o l u t i o n was c a r r i e d out to determine whether any side products are formed during the mutarotation catalyzed by alumina. PRODUCT ANALYSIS 76 IV PRODUCT ANALYSIS The products of the mutarotation reaction were determined by glc analysis of t r i m e t h y l s i l y l ethers of glucose, prepared according to the 68 procedure of Sweeley et a l . , as described i n the Experimental Section. Two main peaks and a small peak were observed. The three peaks A, B and C had r e l a t i v e retention times 0.82, 1.00 and 1.35 which accounted for 2.2%, 40.3% and 57.5%, re s p e c t i v e l y , of the t o t a l area. The two main peaks B and C were i d e n t i f i e d as penta-O-trimethylsilyl-a- and -B-D-gluco-pyranose, re s p e c t i v e l y , by comparisons of t h e i r retention times with those of the t r i m e t h y l s i l y l ethers of pure a- and B-D-glucopyranose. These r e s u l t s agree w e l l with those of Hveding et a l . ^ 7 who c a r r i e d out the mutarotation (homogeneous c a t a l y s i s by solvent) i n dimethylformamide. They i d e n t i f i e d a small peak with r e l a t i v e retention time 0.82 (area about 2.3% when the re a c t i o n was c a r r i e d out at 20°C), as the t r i m e t h y l s i l y l ethers of a- and B-D-glucofuranose. These r e s u l t s show that the c a t a l y s i s of the mutarotation of a-D-glucose by alumina produces g-D-glucose and that side reactions are n e g l i g i b l e . Therefore the cause of the almost 14% discrepancy between the i n f i n i t y values must be the adsorption of glucose onto alumina (see adsorption studies i n the next s e c t i o n ) . ADSORPTION OF GLUCOSE ON ALUMINA SURFACE PART I PRELIMINARY STUDIES 78 V ADSORPTION OF GLUCOSE ON ALUMINA SURFACE PART I PRELIMINARY STUDIES A heterogeneous c a t a l y t i c reaction i n s o l u t i o n consists of f i v e categories of elementary s t e p s 1 (Equation (21)); ( i ) d i f f u s i o n of react a n t ( s ) , S, from s o l u t i o n to the surface of the c a t a l y s t , ( i i ) adsorption of the reactant(s) on the surface, ( i i i ) r e action on the surface forming the product(s), P, (iv) desorption of product(s) from the surface, and (v) d i f f u s i o n of product(s) from the surface to the bulk s o l u t i o n . d i f f u s i o n adsorption surface desorption d i f f u s i o n s • S + C ^ ^ SC ^ PC ^ * P + C • P reaction (21) Therefore, before a surface catalyzed reaction can occur the reactant must get adsorbed on the c a t a l y s t surface. Once the surface reaction has occurred the product formed must be able to desorb i n t o the solu t i o n . Hence r e v e r s i b l e adsorption of glucose on alumina i s an e s s e n t i a l p r e - r e q u i s i t e for mutarotation c a t a l y s i s by alumina. To study the adsorption of glucose on alumina surface, a s o l u t i o n of a- and g-D-glucose, of equilibrium composition (henceforth referred to as the a,3 mixture), i n DMSO was prepared as described i n the Experimental Section. Sixty ml 0.05 M solutions of a,B mixture were s t i r r e d with d i f f e r e n t weights of alumina and the change i n o p t i c a l r o t a t i o n with time was followed u n t i l there was no furth e r change. Since the solutions are completely e q u i l i b r a t e d any decrease i n o p t i c a l r o t a t i o n must be due to adsorption of glucose on the alumina surface. In a l l cases a decrease i n o p t i c a l r o t a t i o n was observed. The greater the weight of alumina used 79 the greater was the amount adsorbed. A graph of change i n o p t i c a l r o t a t i o n (a - a ), which i s proportional to the amount of glucose o eq f t - & adsorbed, where a Q i s the i n i t i a l o p t i c a l r o t a t i o n of the s o l u t i o n of a,g mixture and a the equilibrium o p t i c a l r o t a t i o n of the s l u r r y , versus the weight of alumina used i s given i n F i g . 8. A p l o t of 1/a versus the eq weight of alumina used i s shown i n F i g . 9. I t i s clear that the p l o t i n F i g . 9 i s l i n e a r when the weight of alumina i s low but curves upwards when the weight of alumina i s increased, while the p l o t i n F i g . 8 i s curved throughout the range of concentrations of alumina used. Further experiments showed that the upward curvature i n F i g . 9 can be reproduced whether the s l u r r y i s s t i r r e d slowly or r a p i d l y , i n d i c a t i n g that the increase i n adsorption observed at higher concentrations of alumina i s not due to break-up of p a r t i c l e s due to a rapid rate of s t i r r i n g . V.1 T h e o r e t i c a l Study of Adsorption on Reversible and I r r e v e r s i b l e S i t e s The following model systems were analyzed to investigate the type of adsorption which gives r i s e to the p l o t s observed i n Figs. 8 and 9. V . l . l Only Reversible Adsorption If there are only r e v e r s i b l e adsorption s i t e s , at equilibrium we get, G eq (22) where c7 i s glucose i n s o l u t i o n , C designates the free adsorption s i t e s on c a t a l y s t and GC i s the glucose-catalyst complex. Therefore equilibrium constant K = iGCl l i t r e mole -1 (23) 0 5 10 15 20 25 Weight of Alumina (g) Fig. 8. Plot of Change i n Optical Rotation versus Weight of Alumina for Adsorption of Glucose on Alumina Neutral Weight of Alumina (g) F i g . 9 Plot of (Equilibrium O p t i c a l Rotation) versus Weight of Alumina for Adsorption 00 h-1 of Glucose on Alumina Neutral 82 when a l l concentrations are expressed i n mole l i t r e 1 and, [G ] = IG ] - [GC} (24) eq o where [<? ] = i n i t i a l concentration of glucose i n s o l u t i o n , o Substituting [ C C ] = K ]_G i n Equation (24) we get, [G ] = [ G ] - X [<? ] [ C ] eq o eq Therefore [ff ] = e q 1 + X [ C ] Rearranging the above equation we get, 1 " K id + 1 eq o o Therefore a graph of r r} -. , or of — — which i s proportional to K 7 a ^ eq eq - T g — r , versus [ C ] would be a s t r a i g h t l i n e at a l l c a t a l y s t concentrations. eq At high c a t a l y s t concentrations the f r a c t i o n of the adsorption s i t e s used to form GC complex would be small and the concentrations of a l l the s i t e s on the c a t a l y s t [ C ] ^ [ C ] . Hence a graph of — — versus [ C ] should be l i n e a r ' "eq ° 1 at high values of [C ]. For example, the t h e o r e t i c a l plot of J-T—r versus o 16- J eq [CQ] for a system where [G ] = 1 and K = 0.01 i s given i n F i g . 10. The values of [GC] (and hence [G ] = XG 1 - r£C"l) for d i f f e r e n t values of C were deter-eq o J 1 J o [GC] mined by so l v i n g Equation 0.01 = ] _ [GC])' O D t a i n e d from Equations o (23) and (24). I t i s cle a r , unlike the experimental p l o t i n F i g . 9, that the t h e o r e t i c a l p l o t for r e v e r s i b l e adsorption i s completely s t r a i g h t , even at high concentrations of the c a t a l y s t , where about 80% adsorption i s observed. Note that the weight of c a t a l y s t used i s proportional to the t o t a l number of s i t e s C . 84 V.1.2 Only I r r e v e r s i b l e Adsorption I f a f r a c t i o n k of the t o t a l s i t e s C q gives r i s e to i r r e v e r s i b l e adsorption, then as long as there i s glucose remaining i n s o l u t i o n , the amount of glucose adsorbed i s proportional to the amount of c a t a l y s t , i . e . [GC] = k[C ] and, o [Gea ] " f f f o ] "  k[°o ] (25> Therefore a p l o t of [G ] versus [C ] would be l i n e a r and a graph of eq o —=• versus C w i l l be a rectangular hyperbola. A p l o t of versus eqJ ° *> eq-" CQ for a system with only i r r e v e r s i b l e adsorption where k = 0.01 and [G Q] = 1 i s given i n F i g . 11(a). I t i s clear that i r r e v e r s i b l e adsorption gives r i s e to an upward curvature s i m i l a r to that experimentally observed i n F i g . 9. Since alumina catalyzes the mutarotation of glucose, adsorption on the c a t a l y t i c s i t e s must be r e v e r s i b l e . However, from F i g . 9, there appears to be i r r e v e r s i b l e s i t e s as w e l l . Therefore, as reported i n the following section, the shape of the p l o t when a c a t a l y s t contains both r e v e r s i b l e and i r r e v e r s i b l e s i t e s was studied next. V.1.3 Both Reversible and I r r e v e r s i b l e Adsorption Sites Present At a l l concentrations of the ca t a l y s t a l l the i r r e v e r s i b l e adsorption s i t e s would form GC complex as long as there i s glucose remaining i n s o l u t i o n . Hence the c a l c u l a t i o n f o r G can be s i m p l i f i e d by considering f i r s t the adsorption on i r r e v e r s i b l e s i t e s (using Equation (25)) and then adsorption on r e v e r s i b l e s i t e s . When a l l the i r r e v e r s i b l e s i t e s are occupied the glucose concentration i n s o l u t i o n i s [c7] = [G ] - klC ]. The o o amount of glucose adsorbed on r e v e r s i b l e adsorption s i t e s can be calculated using the procedure described i n Section V . l . l with [&] replacing ]. The t h e o r e t i c a l p l o t of r 1 -. versus [ C ] f o r a system where I Cr J O L eq J 0 25 50 _ l 75 100 [C ] (mole.litre ) o Fi g . 11 Theoretical Plots of l/[£ ] versus [C ]> the Concentration of Adsorption eq o Sites of a So l i d Catalyst Containing (a) O n l y I r r e v e r s i b l e Adsorption S i t e s , oo (b) Both I r r e v e r s i b l e and Reversible Adsorption S i t e s , and (c) O n l y Reversible Adsorption Sites 86 k = 0.01, K = 0.01 and [t? ] = 1 i s given i n F i g . 11 (b), and i t i s clear that i t i s similar to the experimental plot i n Fig. 9. A further refinement on p o s s i b i l i t i e s i s when there are two possible reversible adsorption sit e s as presented i n the next section. V.1.4 Both Weak and Strong Reversible Adsorption Sites Present When two types of reversibly adsorbing sit e s are present we get the following equations for the two e q u i l i b r i a . To simplify calculations the concentrations of the two sites are considered to be equal. fe] 1 ~ C[c ] - [*])([<?] - Dr] - Ly]) o o [y] * 2 CD7 o] - L y ] ) ( [ G o ] - fe] - [y]) where fe] = concentration of glucose-catalyst complex on si t e s of Type 1, [y] = concentration of glucose-catalyst complex on si t e s of Type 2, and \C ] = i n i t i a l concentration of si t e s of Type 1 = i n i t i a l concentration of o sit e s of Type 2. The two equations can be solved for fe] and [y], using a computer, and the equilibrium concentration of glucose t^ eq] = t£Q] ~ fe] ~ Cy] c a n be calculated. The theoretical plot for a system where = 1, ~ 0.01 and [G ] = 1 i s shown i n Fig. 12. I t i s lin e a r with a slope greater than that observed for reversible adsorption i n Section V . l . l . For comparison, plots for systems with K = 0.01, ] = 1, and K = 1, [ff ] = 1 are also included i n Fig. 12. 100 0 25 50 75 100 [C ] (moles.litre ) Fig. 12 Theoretical Plots of HTn , versus [C ], the Concentration of Adsorption eq Sites of a Solid Catalyst Containing (a) Both Weak and Strong Reversible oo Adsorption S i t e s , (b) O n l y Strong Adsorption S i t e s , and (c) O n l y Weak Reversible Adsorption Sites 88 Since,as given above, s t r a i g h t l i n e s don't f i t the experimental observations i t should be concluded that alumina neutral contains both r e v e r s i b l e and i r r e v e r s i b l e adsorption s i t e s . V.2 Determination of the Number of I r r e v e r s i b l e Adsorption Sites per  Gram of the Catalyst From Equations (25) and (23) i t i s c l e a r that as the c a t a l y s t concentration i s increased at constant i n i t i a l glucose concentration [^0]» the i r r e v e r s i b l e s i t e s w i l l i n c r e a s i n g l y adsorb glucose and hence the equilibrium glucose concentration w i l l decrease. Therefore at high alumina concentrations the i r r e v e r s i b l e s i t e s w i l l gain glucose from the r e v e r s i b l e s i t e s . F i n a l l y a stage w i l l be reached when a l l the glucose molecules are occupying i r r e v e r s i b l e s i t e s and there i s none i n s o l u t i o n . At the stage when no glucose can be detected i n s o l u t i o n ( i . e . when the DMSO so l u t i o n shows zero o p t i c a l rotation) there cannot be any glucose i n r e v e r s i b l e s i t e s . Hence the number of i r r e v e r s i b l e s i t e s should be equal to the number of glucose molecules i n the o r i g i n a l s o l u t i o n . To determine the number of i r r e v e r s i b l e adsorption s i t e s the experiment described at the beginning of Section V was continued using higher proportions of the c a t a l y s t . In order to make f i l t r a t i o n easier a 60 ml 0.01 M a,g mixture was used instead of 0.05 M s o l u t i o n so that lower concentrations of the c a t a l y s t could be used. The o p t i c a l rotations were measured using a 1 dm path length c e l l i n order to obtain higher s e n s i t i v i t y . The r e s u l t s are given i n F i g . 13, which shows that with 60 ml 0.01 M glucose s o l u t i o n zero o p t i c a l r o t a t i o n i s reached when between 8.0 and 9.0 g of alumina i s present. The most probable range i s between 8.25 and 8.75 g alumina. 8 9 4 6 8 1 0 1 2 Weight of Alumina Neutral (g) Fig. 13 Plot of Equilibrium Optical Rotation versus Weight of Alumina for the Determination of the Number of Irreversible Adsorption Sites on Alumina Neutral 90 Hence the number of i r r e v e r s i b l e s i t e s on 8.5 ± 0.25 g of alumina neutral i s equal to the number of moles of glucose i n 60 ml 0.01 M -4 glucose s o l u t i o n which i s 6.0 x 10 mole. Therefore the number of -4 i r r e v e r s i b l e adsorption s i t e s i s (0.70 ± 0.02) x 10 mole/g. V.3 Rate of Adsorption The rate of adsorption of D-glucose onto alumina surface was studied by s t i r r i n g a 60 ml 0.05 M s o l u t i o n of the a,3 mixture with a pre-weighed sample of alumina, as described i n the Experimental Section. The rate of change of o p t i c a l r o t a t i o n with time was followed u n t i l there was no further change. The procedure was repeated with d i f f e r e n t weights of alumina. The r e s u l t s given i n F i g . 14, although not accurate enough to obtain i n d i v i d u a l rate constants, show that adsorption i s complete i n 15 to 20 minutes. Comparison of these r e s u l t s with a t y p i c a l f i r s t order p l o t for a-D-glucose mutarotation (Fig. 15) shows that the i n i t i a l rapid decrease i n o p t i c a l r o t a t i o n stops at about the same time the adsorption of glucose onto alumina surface i s complete. In addition,the decrease i n o p t i c a l r o t a t i o n when the a,8 mixture was s t i r r e d with alumina was equal to the diff e r e n c e between the t h e o r e t i c a l and experimental i n f i n i t y values observed i n a k i n e t i c run, when the same weight of alumina was used. Hence i t i s clear that the i n i t i a l rapid decrease i n o p t i c a l r o t a t i o n observed i n the k i n e t i c runs i s due to the adsorption of glucose on alumina surface and the rest of the p l o t represents the actual c a t a l y t i c reaction. Therefore i n a l l non-l i n e a r f i r s t order pl o t s described i n t h i s t h e s i s , the rate constants were determined by measuring the slope at t = 25 min when the adsorption i s complete. Time (min) F i g . 14 Rate of Adsorption of Glucose on Alumina Neutral Time (min) Fi g . 15 F i r s t Order K i n e t i c Plot ( l e f t and • ) for a-D-Glucose Mutarotation by Alumina Neutral and Rate of Adsorption of Glucose (right and X ) on Alumina Neutral DIFFUSION IN HETEROGENEOUS CATALYSIS 94 VI DIFFUSION IN HETEROGENEOUS CATALYSIS As mentioned in Section V, before adsorption and surface reaction can take place, the substrate molecules must diffuse from the solution to the adsorption sites on the surface of the catalyst and the desorbed product must diffuse from the surface to the bulk solution. diffusion adsorption surface desorption diffusion S • S+C ^ * SC ^ ^ PC ^ » C+P • P reaction (21) For a heterogeneous catalytic system involving a porous catalyst like alumina (Section II.2), diffusion may be divided to external diffusion of the substrate from solution to the surface of the catalyst particle and internal diffusion from the surface of the catalyst particle through the pores to the adsorption site inside the pore"'". If the catalyst is not rapidly agitated in solution a concentration gradient w i l l be set up between the bulk solution and the catalyst surface,if the rates of adsorption, surface reaction,and desorption are faster than that of diffusion, and in such a case, the rate of diffusion can be rate limiting. The concentration gradient set up outside the catalyst particles can be eliminated by rapidly s t i r r i n g the slurry, which makes the solution homogeneous since the substrate i s quickly brought to the surface and the desorbed product i s quickly removed from the surface. If the rate of a reaction i s diffusion controlled,then the rate would increase when the rate of st i r r i n g is increased (and hence the concentration gradient is decreased) and f i n a l l y reaches a plateau value when the rate of the reaction i s independent of the diffusion of substrate to the surface"'". In order to agitate the slurry efficiently, different s t i r r e r s , viz. 95 magnetic s t i r r i n g bar, overhead Corning V i b r a s t i r , overhead Fischer Dyna-Mix connected to screw type and dispersion s t i r r e r s , were employed. The disper-sion s t i r r e r was found to be the most e f f i c i e n t and the observed rate constant (determined at t = 25 min a f t e r the adsorption i s complete), increased with the increase i n rate of s t i r r i n g and reached a maximum plateau value (Fig. 16). This optimum rate of s t i r r i n g was used i n a l l the k i n e t i c and adsorption studies described i n t h i s t h e s i s . Although external d i f f u s i o n can be eliminated by rapid a g i t a t i o n of the s l u r r y , the rate of a reaction i n a porous catalyst l i k e alumina neutral might s t i l l be c o n t r o l l e d by i n t e r n a l d i f f u s i o n through the pores, since the volume of l i q u i d i n s i d e the pores i s not mixed by the mechanical s t i r r e r present outside the pores. To test whether i n t e r n a l d i f f u s i o n i s rate l i m i t i n g , the method suggested by Boudart and Burwell' 1 was used. It consists of breaking down the c a t a l y s t p a r t i c l e s by grinding to decrease the p a r t i c l e s i z e and hence decrease the pore length. If i n t e r n a l d i f f u s i o n i s rate l i m i t i n g i n the unground c a t a l y s t p a r t i c l e s , then the observed rate constant should be higher with the ground sample. About 10 g of alumina neutral was ground w e l l with a p e s t l e and mortar for 15 mins. The p a r t i c l e s i z e d i s t r i b u t i o n s of the ground and unground samples were determined with the Electrozone Celloscope. The alumina used was from a second b o t t l e of "alumina n e u t r a l f or t i c " and showed two maxima at approximately 11.5 microns and 24 microns (Fig. 17(a)). Comparison of the p a r t i c l e s i z e d i s t r i b u t i o n curves i n F i g . 17 shows that, by grinding, the s i z e of the peak corresponding to p a r t i c l e s of larger diameter has decreased r e l a t i v e to the other peak i n d i c a t i n g that the larger p a r t i c l e s have been broken down to smaller ones. In a d d i t i o n s t h e [a-D-Glucose] = 0.05 M [Alumina] = 13.3 mg.ml 25 50 75 100 Relative Rate of S t i r r i n g (x 38 = rpm) F i g . 16 The E f f e c t of Rate of S t i r r i n g on the Observed Rate Constant 97 (a) Unground Sample — — — — CM W CI P a r t i c l e Diameter (centimicrons) (b) Ground Sample fl^mw — — r > ( D « j i D <j m n eo oj to o — — — — N N O O f l P a r t i c l e Diameter (centimicrons) F i g . 17 P a r t i c l e Size D i s t r i b u t i o n s of (a) Unground and (b) Ground Samples of Alumina Neutral 98 log mean diameter has decreased from 11.2 microns to 10.4 microns, i.e. a 7% decrease in log mean diameter. The cumulative percentages greater than stated diameters given in Table VI also clearly show that the particle size has decreased on grinding. However, kinetic runs were superimposable and therefore no change i n the rate constants was observed when equal weights of unground and ground samples were used with 60 ml 0.05 M a-D-glucose solution in DMSO. The rate of catalysis by alumina neutral i s therefore not controlled by internal diffusion. External diffusion can be eliminated by rapid s t i r r i n g as described above. TABLE VI CUMULATIVE PERCENTAGES GREATER THAN STATED DIAMETERS FOR GROUND AND UNGROUND SAMPLES OF ALUMINA NEUTRAL Diameter (microns) Cumulative Percentage Greater than the Stated Diameter Ground Unground 1 100 100 2.5 99.9 100 5 90.4 93.1 7.5 73.4 77.3 10 55.3 59.8 15 27.2 33.0 20 12.2 19.4 25 1.7 4.6 30 1 2 99 Since d i f f u s i o n i s not rate l i m i t i n g when the heterogeneous c a t a l y t i c system i s s t i r r e d as described above, the observed rate constant should be a function of the rate constants f o r adsorption, desorption, both forward and reverse surface reactions, and the concentrations of c a t a l y s t , substrate and product. The rate equation f o r the surface catalyzed reaction, when d i f f u s i o n i s not rate l i m i t i n g , i s derived i n the next section. RATE EQUATION FOR THE HETEROGENEOUS CATALYTIC SYSTEM 101 VII RATE EQUATION FOR THE HETEROGENEOUS CATALYTIC SYSTEM A simple heterogeneous c a t a l y t i c reaction, which i s not d i f f u s i o n controlled, can be represented by the following equation, h k3 * 2 S + C ,« » SC , * PC ~i—C + P (26) k2 -fej where S, P and C represent substrate, product and catalyst, r e s p e c t i v e l y , and SC and PC are the substrate-catalyst and product-catalyst complexes. kr, and k, are the rate constants f o r the forward and reverse surface c a t a l y t i c reactions. For adsorption i n v o l v i n g hydrogen bonding and dipole-dipole i n t e r a c t i o n s between the hydroxyl groups of glucose and the polar alumina surface, rate constants for adsorption (k^) and desorption (k^) of a-D-glucose (substrate S) and B-D-glucose (product P) would be almost the same since the only d i f f e r e n c e between the two molecules i s the configuration at the anomeric carbon. The rate equation for the reaction i n Equation 26 i s derived by assuming that a heterogeneous c a t a l y t i c system, consisting of a powdered s o l i d c a t a l y s t dispersed evenly i n a s o l u t i o n of a substrate and product, behaves s i m i l a r to a homogeneous c a t a l y t i c system. Then, the rate of the reaction, when the c a t a l y s t C and the complexes SC and PC are present i n steady state, i s given by , dp (k k k^s - k j f e ^ p ) ^ dt k22 + kj<k + k2k3 + k1(k2 + k^ + k,)s + k^k2 + k3 + k^)P where O Q i s the t o t a l c a t a l y s t concentration, and s and p are the substrate and product concentrations at time t. 1 0 2 If s + p = s the concentration of the substrate in solution when r o the adsorption i s complete ( ^  i n i t i a l concentration of substrate i f the amount adsorbed is small) then, dp dt fei fe.fe-c? s 1 2 3 o V + KK + k0k„ * k.(fe„ + fe~ + fe.) s 2 2 4 2 3 1 2 3 4 o + fe0 + k„k, + fe.fe, + fe, (fe. + fe„ + fe.) s z 2 4 2 3 1 2 3 4 o where X = X3s + Xhp k]k2k3CQ 3 K2 + feA + KK + K (fe, + k + fe ) S 2 24 2 3 1 2 3 4 o and X are constants, k kji.o 1 2 4 o 4 fe22+fe2fe4+fe2fe3+fe1(fe2+fe3+ fe4)ao dp Therefore — = X. (s - p) - X.p (27) dt 3 ° 4 Equation (27) is similar to Equation (8) for homogeneous catalysis discussed in the Introduction and becomes upon integration t l n ( V - v ) X3 + X^ = ^ ln I ) (28) P e " P 1 0 3 where p = equilibrium concentration of the product. Equation (28) can be expressed i n terms of o p t i c a l r o t a t i o n as, , (a - a ) i o °° J„ + X. = - l n — 3 4 * where a i s the i n i t i a l r o t a t i o n , a, i s the r o t a t i o n at time t, and a i s O t ' co the equilibrium r o t a t i o n . l n (a. - a ) = -(X. + X,)t + l n (a - a ) (29) t °° 3 4 o 0 0 Hence the system should show f i r s t order k i n e t i c s with observed rate constant k , - Z 0 + X. obs 3 4 \ \ % (fc3 + fc4) ( 3 Q ) k i 2 + % + ¥ 3 + V W V So A surface reaction involves breakage and formation of chemical bonds, while adsorption and desorption involve formation and breakage of di p o l e -dipole i n t e r a c t i o n s and hydrogen bonds. Hence the rate constants for adsorption and desorption should be greater than those for the surface reaction, i . e . k^, k^ » k^, k^,and the above equation s i m p l i f i e s to kAk. + k.)c , 1 3 4 O fr>i\ k ( 3 1 ) o b s k + k q K2 + V o _ J 4 _ ^ ( 3 2 ) 104 where K = k^/k^ i s the equilibrium constant for adsorption of glucopyranose on alumina. From the above discussion i t i s clear that the observed rate constant for a heterogeneous c a t a l y t i c system involves the rate constants for adsorption and desorption steps (k^ and k"), and the t o t a l substrate concen-t r a t i o n ( S q ) i n addition to the rate constants for the c a t a l y t i c (surface) reaction (k^ and k^) and the c a t a l y s t concentration ( G Q ) - This i s a c h a r a c t e r i s t i c d i f f e r e n c e between a simple homogeneous c a t a l y t i c system and a heterogeneous or any other system where there i s i n i t i a l complex formation 1 80 (binding) between substrate and c a t a l y s t ' . Hence any i n h i b i t o r y , activatory, or deuterium isotope effect, observed on the rate constant during mechanistic studies can r e s u l t from the corresponding e f f e c t s on the c a t a l y t i c r eaction or on the adsorption desorption process or both. From Equation (32) i t i s also c l e a r that the c a t a l y t i c constant f o r the surface reaction (k^ + k^) can be determined from the observed rate constant using equilibrium constant K for adsorption of glucopyranose on alumina. The r e l a t i o n between the equilibrium constant K = k^/k^ and the experimentally determined equilibrium constant i s discussed i n the next sec t i o n . V I I . l Relation between Equilibrium Constant K f o r Adsorption of Glucopyranose  on Alumina and the Observed Equilibrium Constant In Equation (26) f o r the c a t a l y t i c r eaction two surface complexes were considered, v i z . SC and PC the a-D-glucopyranose adsorbed on alumina and B-D-glucopyranose adsorbed on alumina, and they were considered to undergo d i r e c t interconversion on the surface. As mentioned i n the Intro-duction the homogeneous c a t a l y s i s of mutarotation occurs v i a the a c y c l i c 105 intermediate whose concentration i n water i s only about 0.003%. Hence i t i s l i k e l y that the heterogeneous process also occurs v i a an a c y c l i c (or possibly even carbonium ion) intermediate (J) which may be s t a b i l i z e d by the surface. In order to accomodate the IC complex Equation (26) should be modified to, h k 2 S + C - - SC , ~» IC ^ T ^ " PC T~** P + C (26') k2 h In the adsorption studies mentioned i n t h i s thesis the amount adsorbed was measured by determining the i n i t i a l and f i n a l o p t i c a l rotations of solutions of a, 3 mixture i n DMSO. Hence the amount l o s t from the s o l u t i o n should be present as SC, PC,and IC complexes, and the observed equilibrium constant K i s equal to . .{SC + PC + IC~\ ( 2 3') IS + P] [C] which was given simply as [g-^] [g] ^n Equation (23) with GC = SC + PC + IC. From t h i s discussion i t i s c l e a r that the actual adsorption process, given as a one step process i n Equation (22), occurs v i a two consecutive steps at least f or a small percentage of glucose molecules. C + S + P T " ^ + PC < IC (22') In s o l u t i o n the concentration of a c y c l i c form i s very small but i t i s possible that the a c y c l i c form may be p a r t i a l l y s t a b i l i z e d by the surface. However, most probably, i t i s s t i l l much less than the concentration of 106 pyranose forms on the surface. Thus Equation (23').simplifies to, Hence the observed equilibrium constant should be very close to that f o r the adsorption of the pyranose forms from s o l u t i o n . It has been shown i n t h i s section that glucose mutarotation according to the mechanism i n Equation (26) should show f i r s t order k i n e t i c s (Equation (29)). However, as described i n Section I I I , the f i r s t order p l o t s f o r glucose mutarotation by alumina show curvature throughout the k i n e t i c runs. The cause of t h i s curvature i n the f i r s t order p l o t s i s investigated i n the next section. [SC + PC + XC~\ ISC + PC] k1 K = \.S + P] [ C ] IS + P] [c7] k2 VIII STUDY OF THE CURVATURE IN FIRST ORDER PLOTS 108 V I I I STUDY OF THE CURVATURE IN FIRST ORDER PLOTS As mentioned i n S e c t i on I I I , the upward curvature i n f i r s t order p l o t s cou ld be a r e s u l t of p rog re s s i ve d e a c t i v a t i o n of alumina du r ing c a t a l y s i s or due to the f a c t that the c a t a l y t i c r e a c t i o n i s not f i r s t order i n a-D-g lucose. The muta ro ta t i on of a-D-glucose i n s o l u t i o n i s a w e l l known f i r s t order r e a c t i o n 5 1 . There are a few examples where 2 , 3 , 4 , 6 - t e t r a - O - a c e t y l - D -88 97 119 g lucose , 2 ,3 ,4 ,6 - te t ra -O-methy l -D -g lucose and D-glucose ( i n each case both a - and 8-anomers) s e l f - c a t a l y z e d the muta ro ta t i on of the r e s p e c t i v e a-D-g lucose d e r i v a t i v e s i n p y r i d i n e . But, even i n such cases, the r e a c t i o n was f i r s t - o r d e r w i t h re spect to the corresponding a-D-glucose 61 67 d e r i v a t i v e . Hveding et a l . ' have repor ted that the f i r s t - o r d e r p l o t s f o r muta ro ta t i on of a-D-glucose i n N,N-dimethylformamide, and i n N,N-d imethy l -formamide - water mixtures where the mole f r a c t i o n of water was l e s s than about 0.7, showed s i g n i f i c a n t d e v i a t i o n from a s t r a i g h t l i n e . They s t a ted tha t the muta ro ta t i on under such cond i t i on s i s complex i n analogy w i t h that found f o r D-ga lactose i n water . But f o r d imethy l s u l f o x i d e and d imethy l s u l f o x i d e - water m i x tu re s , Robertson et a l . found that the muta ro ta t i on i s f i r s t - o r d e r i n g lucose. Hence i t i s more l i k e l y that alumina i s undergoing p rog re s s i ve d e a c t i v a t i o n dur ing c a t a l y s i s . The d e a c t i v a t i o n of the c a t a l y s t du r i ng c a t a l y s i s can a r i s e from s t rong ad so rp t i on on the a c t i v e s i t e s by the s ub s t r a te i t s e l f (or even 21 by the s o l v e n t ) s r e f e r r e d to as s e l f - p o i s o n i n g or cok ing , or i t may be due to the f o rmat i on of a s i de product which competes w i t h the sub s t r a te 118 f o r the a c t i v e s i t e s . The l a t t e r process i s q u i t e u n l i k e l y s i nce the s ide products detected by g l c a n a l y s i s ( i n Sec t i on IV) are n e g l i g i b l e and 109 also, the small percentage of a- and 8-D-furanosides detected towards the end of a k i n e t i c run are known to be formed at the beginning of the r e a c t i o n ^ . To determine whether the curvature i n the k i n e t i c p l o t s i s due to deactivation of the c a t a l y s t , the c a t a l y t i c a c t i v i t y of a sample of alumina which had been previously used to catalyze mutarotation was compared with that of a fresh sample of alumina under s i m i l a r conditions as described below. VIII.1 Tests f o r Deactivation of the Catalyst The c a t a l y t i c a c t i v i t y of a fresh sample of alumina neutral was determined by s t i r r i n g a 60 ml sol u t i o n which i s 0.05 M i n a-D-glucose and 3 0.05 M i n a,3 mixture with 1.6 g of the c a t a l y s t . The pl o t of l n 10 (a^-a^) versus time gave the "standard curve" shown i n F i g . 18. It was found that the mutarotation of 60 ml 0.05 M so l u t i o n of a-D-glucose (contains 0.54 g a-D-glucose) by 1.6 g alumina i s complete i n about 6 hours. To test f o r deactivation of alumina during t h i s period, a sol u t i o n of 0.54 g a-D-glucose i n 50 ml DMSO was s t i r r e d for 6 hours with 1.6 g alumina. A f t e r 6 hours, a 10 ml s o l u t i o n containing 0.54 g a-D-glucose i n DMSO was added to the e q u i l i b r a t e d s o l u t i o n and the change i n o p t i c a l r o t a t i o n with time was followed. The r e s u l t s given i n F i g . 18 show that the observed rate constant has decreased by about 50% a f t e r an i n i t i a l 6.5 hours of reaction. As mentioned e a r l i e r t h i s d eactivation can a r i s e from strong adsorption of glucose and/or solvent on the active s i t e . To determine whether DMSO poisons the c a t a l y s t , 1.6 g of alumina was s t i r r e d with 50 ml DMSO f o r 20 hours. A s o l u t i o n of 0.54 g a-D-glucose and 0.54 g a,3 mixture i n 10 ml DMSO was added and the change i n o p t i c a l r o t a t i o n with time was followed. The r e s u l t s given i n F i g . 18 show that alumina has "Standard curve" Deactivation by DMSO for 20 hrs Deactivation during mutarotation for 6 hrs Deactivation during mutarotation for 22 hrs Deactivation during mutarotation for 70 hrs-[a-D-Glucose] = 0.05 M [a,6 Mixture] = 0.05 M [Alumina] = 26.7 mg.ml 50 100 150 200 Time (min) 250 300 350 400 Fig. 18 Deactivation of Alumina Neutral by DMSO and by Glucose During Catalysis of Mutarotation I l l been deactivated s l i g h t l y by DMSO during the 20 hours, but t h i s deactivation i s n e g l i g i b l e compared to deactivation during 6 hours of c a t a l y t i c reaction with glucose. Moreover,alumina that has been used for 6 hours of c a t a l y s i s underwent very l i t t l e d eactivation over the next 5 hours, and gave a l i n e a r f i r s t order pl o t over almost two h a l f - l i v e s , while alumina treated with DMSO gave r i s e to a "standard" non-linear p l o t . To determine whether there i s any further deactivation during mutarotation, two more experiments were c a r r i e d out. In one, 10 ml a-D-glucose s o l u t i o n was added a f t e r 22 hours of c a t a l y s i s and i n the other, a f t e r 70 hours of c a t a l y s i s . Results i n F i g . 18 show that alumina has undergone further deactivation although at a much slower rate, 55% af t e r 22 hours and 70% a f t e r 70 hours. The r e p r o d u c i b i l i t y of the experiment c a r r i e d out a f t e r 22 hours of c a t a l y s i s was found to be very good. I t i s clear from F i g . 18 that alumina deactivated f or 22 hours has given r i s e to a l i n e a r f i r s t order pl o t f o r more than two h a l f - l i v e s . Hence the c a t a l y s i s of the mutarotation of a-D-glucose by alumina i s f i r s t order i n a-D-glucose, and the curvature of the f i r s t order pl o t s i s due to progressive deactivation of the c a t a l y s t . VIII.2 Cause of Deactivation Since i t has been shown that the a c t i v i t y of the ca t a l y s t decreases during c a t a l y s i s , i t i s of i n t e r e s t to determine the possible cause of deactivation. I t has been mentioned e a r l i e r that side products may act as a c a t a l y s t poison. Alumina i s a w e l l known c a t a l y s t f o r dehydration of 14 alcohols , and there are many reactions catalyzed by alumina, where removal 23 of water from the surface increases the a c t i v i t y of the ca t a l y s t Therefore we investigated the e f f e c t of water added to DMSO and onto alumina 112 on i t s c a t a l y t i c a c t i v i t y . F i f t y ml of DMSO containing 0.054 ml (= 0.006 mole = the number of moles of D-glucose used f o r the deactivation studies i n Section VIII.1) of d i s t i l l e d water was s t i r r e d with 1.6 g of alumina for one hour. A 10 ml sol u t i o n of 0.54 g a-D-glucose and 0.54 g a,3 mixture i n DMSO was added and the change i n the o p t i c a l r o t a t i o n with time was followed. Comparison of r e s u l t s , given i n F i g . 19, with the "standard curve" shows that alumina has been only very s l i g h t l y deactivated compared with deactivation by glucose during 6 hours of c a t a l y s i s . Since the deactivation can be due to DMSO, the experiment was repeated a f t e r p r e t r e a t i n g the alumina with DMSO containing water f o r 22 hours. Results i n F i g . 19 show that the alumina has been deactivated further, s l i g h t l y more than deactivation by DMSO during the same period. Since water added i n t o DMSO had very l i t t l e e f f e c t on the a c t i v i t y , we investigated the e f f e c t of water added d i r e c t l y onto alumina on i t s a c t i v i t y . D i s t i l l e d water (0.054 ml) was added onto 1.6 g of alumina held i n a 100 ml 3-neck f l a s k . The f l a s k was stoppered and the alumina was mixed slowly, by ro t a t i o n of the f l a s k , for 60 minutes. A 60 ml so l u t i o n of 0.54 g a-D-glucose and 0.54 g a,3 mixture i n DMSO was added and the change i n o p t i c a l r o t a t i o n with time was followed. The r e s u l t s given i n F i g . 19 show that alumina has been only s l i g h t l y deactivated by d i r e c t contact with water and that the extent of deactivation i s almost the same as the deactivation caused by water i n DMSO over the same period. These r e s u l t s show conclusively that the deactivation of alumina during mutarotation c a t a l y s i s i s not due to water produced by dehydration of glucose by alumina. As mentioned e a r l i e r , the deactivation of the ca t a l y s t could a r i s e from the strong adsorption, on the act i v e s i t e s , of the 5.5 4.5 3.5 2.5" [a-D-Glucose] [a,3 Mixture] [Alumina] [Water] 0.05 M 0.05 M 26.7 mg.ml 0.1 M Deactivation during mutarotation for 6 hrs • "Standard Curve" X Deactivation by DMSO for 20 hrs O Deactivation by DMSO containing water for 1 hr • Deactivation by DMSO containing water for 22 hrs D Deactivation by water for 1 hr 50 100 150 200 Time (min) 250 300 350 400 Fig. 19 The E f f e c t of Water on the C a t a l y t i c A c t i v i t y of Alumina Neutral 114 reactant i t s e l f or due to the formation of a side product which competes with the substrate for the active s i t e s . That i s , the i n h i b i t i n g substance could be present on the surface or i n s o l u t i o n . However, the r e s u l t s of product analysis (in Section IV) showed only a small percentage of side-products. To confirm that deactivation of alumina i s due to strong adsorption of substrate molecules onto the act i v e s i t e s and not due to side products having an i n h i b i t o r y e f f e c t , which might not have been detected by g l c , 120 the following experiment, s i m i l a r to that described by Dirkx and Baan , . was performed. Two g of alumina was s t i r r e d with a solu t i o n of 0.54 g a-D-glucose i n 50 ml DMSO for 24 hours. The s l u r r y was f i l t e r e d , and the alumina was washed with 20 ml DMSO as described i n the Experimental Section. The "used" alumina was dried i n vacuum at room temperature for 24 hours. Thus 70 ml of "used" glucose (equilibrated) s o l u t i o n and about 2 g of "used" alumina were obtained. Five g of alumina was treated with 100 ml of DMSO for 24 hours. The s l u r r y was f i l t e r e d as above and the alumina was dried i n vacuum at room temperature f o r 24 hours. Thus about 5 g of "fr e s h " alumina (DMSO has very l i t t l e e f f e c t on the a c t i v i t y of alumina) was obtained. The co n t r o l run was ca r r i e d out by s t i r r i n g 1.6 g " f r e s h " alumina with a so l u t i o n of 0.54 g of a-D-glucose and 0.54 g of a,(3 mixture i n 80 ml DMSO. The f i r s t order k i n e t i c p l o t obtained i s given i n F i g . 20 and w i l l be used as a co n t r o l run to determine whether "used" alumina has been deactivated and whether "used" glucose s o l u t i o n causes deactivation of fresh alumina. The f i r s t order k i n e t i c p l o t when 1.6 g "f r e s h " alumina was s t i r r e d with, 70 ml "used" glucose s o l u t i o n and 0.54 g of a-D-glucose i n 10 ml DMSO 5.0 s i a 4.5 4.0 3.5 3.0 [a-D-Glucose] [a,B Mixture] [Alumina] = 0.05 M = 0.05 M = 26.7 mg.ml # Alumina dri e d at room temperature and " f r e s h " glucose so l u t i o n X "Fresh" alumina, "used" glucose solution and a-D-glucose • "Fresh" alumina and "fresh" glucose s o l u t i o n O "Used" alumina and "fresh" glucose s o l u t i o n 30 60 90 120 Time (min) 150 180 210 Fig. 20 Tests for the Presence of Inhibitory Products i n Solution and on the Surface of the Catalyst i—1 i—1 Ul 116 i s a l so shown i n F i g . 20. The a c t i v i t y of the alumina i s about the same as that of the cont r o l and hence,"used" glucose s o l u t i o n does not contain any i n h i b i t o r s . However,the k i n e t i c p l o t when 1.6 g "used" alumina i s s t i r r e d with a sol u t i o n of 0.54 g a-D-glucose and 0.54 g a , 8 mixture i n 80 ml DMSO shows that the a c t i v i t y of the alumina has decreased 42% during 24 hours of c a t a l y s i s . In a d d i t i o n , i t gave r i s e to a l i n e a r f i r s t order p l o t as i n Section VIII.1. For comparison the f i r s t order p l o t obtained when 1.6 g alumina which had not been pre-treated with DMSO but evacuated at room temperature for 24 hours, was s t i r r e d with 0.54 g a-D-glucose and 0.54 g a , 8 mixture i s given i n F i g . 20. I t s a c t i v i t y i s higher than that of alumina which had been pre-treated with DMSO and dried i n vacuum. The decrease i n a c t i v i t y of the " f r e s h " alumina could be due to removal of water on the ca t a l y s t surface (discussed l a t e r i n Section XIII) by DMSO, or due to the presence of some DMSO on alumina which would decrease the percentage of alumina i n a given weight of the sample. The r e s u l t s discussed i n t h i s section have shown that the glucose mutarotation by alumina i s a f i r s t order reaction and the curvature i n f i r s t order p l o t s i s due to progressive deactivation of the ca t a l y s t caused by strong adsorption of glucose on act i v e s i t e s . Hence the rate constant obtained at t = 25 min should be the actu a l i n i t i a l rate constant a f t e r surface adsorption i s complete but before much ca t a l y s t deactivation has occurred. Therefore, the observed rate constant should be equal to the f i r s t order rate constant derived i n Section VII (Equation (32)). 1 1 7 (k. + k.) e k = — 3 i _ a 02) ° b S 1/K + 8 o According to t h i s equation, & ^ should be d i r e c t l y proportional to the t o t a l c a t a l y s t concentration a , while a p l o t of 1/k versus J o obs substrate concentration s should be l i n e a r . These r e l a t i o n s h i p s are o tested i n the next two sections. THE EFFECT OF CATALYST CONCENTRATION ON THE OBSERVED RATE CONSTANT 119 XI THE EFFECT OF CATALYST CONCENTRATION ON THE OBSERVED RATE CONSTANT It was shown i n Section VII, that the observed rate constant f o r the c a t a l y t i c r eaction i s given by, (fc, + fc,) o k = _ J * ° (32) o b s 1/K + s o and hence, i t i s d i r e c t l y proportional to the t o t a l catalyst concentration C q when the substrate concentration S q i s constant. As suggested i n Section VIII, the rate constant obtained at t = 25 mins gives the a c t u a l rate constant f o r the f i r s t order reaction, and i t should be d i r e c t l y proportional to the weight of alumina used. F i r s t order p l o t s were obtained by s t i r r i n g 60 ml 0.05 M a-D-glucose solutions with d i f f e r e n t weights of alumina. The rate constants at t = 25 mins were obtained and are plo t t e d against the weight of ca t a l y s t i n F i g . 21. Results c l e a r l y show that at constant substrate concentration s , the rate J o constant fc i s d i r e c t l y proportional to the weight of alumina used. o b s (fc 3 + k) Further, the slope i s equal to i / j f + g — ' a n c ^ c a n b e u s e c ^ t o determine the o c a t a l y t i c constant for the surface reaction (fc^ + fc^) when the equilibrium constant f o r adsorption K and the number of r e v e r s i b l e adsorption s i t e s on a gram of the c a t a l y s t are determined from the adsorption isotherm i n Section XV. THE EFFECT OF SUBSTRATE CONCENTRATION ON THE OBSERVED RATE CONSTANT 122 X THE EFFECT OF SUBSTRATE CONCENTRATION ON THE OBSERVED RATE CONSTANT According to Equation (32), at constant c a t a l y s t concentration C Q , the observed rate constant should decrease with the increase of glucose concentration s . A graph of k , versus concentration of glucose should be a curve while a graph of Ilk , versus glucose concentration should be 6 ^ obs l i n e a r with slope = ( ^ +\)co a n d ^ " " P * = K(k3 +\)OQ (fc, + k.) a v _ 3 4 o Kobs ~ (32) l/K + s o l/k = l— + — ^ (33) o b s + k,)a (k„ + k.)c 3 4 o 3 4 o The rate constants were obtained as before using 1.6 g of c a t a l y s t and d i f f e r e n t concentrations of glucose. The rate constants at high glucose concentrations (0.4 and 0.8 M) were corrected f o r slow homogeneous c a t a l y s i s . The r e s u l t s given i n Figs. 22 and 23 agree w e l l with the predictions of Equations (32) and (33). Again, from the slope of F i g . 23, the c a t a l y t i c constant (k^ + k^) can be determined when the number of adsorption s i t e s i s determined from the adsorption isotherm. However, the intercept i s too close to zero to be of any use. 123 0 0.2 0.4 0.6 0.8 1.0 Concentration of a-D-Glucose (M) Fig. 22 The Effect of a-D-Glucose Concentration on the Observed Rate Constant REPRODUCIBILITY OF THE RESULTS AND COMPARISON OF CATALYTIC ACTIVITIES OF DIFFERENT ALUMINAS 126 XI REPRODUCIBILITY OF THE RESULTS AND COMPARISON OF CATALYTIC ACTIVITIES OF DIFFERENT ALUMINAS As mentioned i n the Introduction (Section 1.1.3) small amounts of impurities can change the a c t i v i t y of the c a t a l y s t . Small amounts of water adsorbed on the alumina, when exposed to the atmosphere or from the solvent during a k i n e t i c run, can change i t s surface a c t i v e s i t e s , s p e c i a l l y 121 Lewis acid s i t e s . In fact, r e p r o d u c i b i l i t y of the r e s u l t s i s one of 122 the main disadvantages of many heterogeneous c a t a l y t i c systems when compared with homogeneous systems. XI.1 E f f e c t of Drying and D i s t i l l i n g DMSO on the C a t a l y t i c A c t i v i t y To check the e f f e c t of any impurities present i n DMSO on the c a t a l y t i c a c t i v i t y , c a t a l y s i s was car r i e d out i n u n d i s t i l l e d DMSO, i n DMSO that has been dried and d i s t i l l e d as described i n the Experimental Section, and also with dried and d i s t i l l e d DMSO which has been treated with neutral alumina a c t i v i t y super I (a highly dehydrated form of aluminum oxide, see Section XI.3). The r e s u l t s i n F i g . 24 show that there i s very l i t t l e d i f f e r e n c e i n a c t i v i t y between the three runs. However, i n a l l k i n e t i c r e s u l t s described i n this t h e s i s , DMSO which has been dried and d i s t i l l e d over C a ^ and stored, over n e u t r a l alumina a c t i v i t y super I, under nitrogen was used because of the need f o r anhydrous DMSO for c e r t a i n experiments. XI.2 R e p r o d u c i b i l i t y of Results with Alumina Neutral f o r TLC The r e p r o d u c i b i l i t y of k i n e t i c s with the same batch of alumina n e u t r a l was checked and the following r e s u l t s were obtained. The f i r s t order p l o t s obtained w i t h i n two consecutive days were superimposable. The r e p r o d u c i b i l i t y 5.5 O Cat a l y t i c a c t i v i t y with u n d i s t i l l e d DMSO X C a t a l y t i c a c t i v i t y with dried and d i s t i l l e d 50 100 150 200 250 300 Time (min) Fig. 24 The E f f e c t of Drying and D i s t i l l i n g of DMSO on the C a t a l y t i c A c t i v i t y of Alumina Neutral 128 a f t e r about 3 weeks was s t i l l very good as shown i n F i g . 25. A f t e r one year the a c t i v i t y of alumina decreased only by 4%. A d i f f e r e n t batch of alumina neutral showed an a c t i v i t y which i s only 20% lower than that of the f i r s t batch. These r e s u l t s show that one can obtain excellent r e p r o d u c i b i l i t y with Woelm alumina neutral and that d i f f e r e n t batches show a c t i v i t i e s which are of the same order. XI.3 Comparison of C a t a l y t i c A c t i v i t i e s of D i f f e r e n t Neutral Aluminas It i s of i n t e r e s t to compare a c t i v i t i e s of aluminas for column chromatography with those of the t i c aluminas used i n t h i s project. Hence, the f i r s t order pl o t s f o r alumina n e u t r a l a c t i v i t y super I (used by 12 Posner i n h i s studies of alumina promoted organic reactions) and alumina n e u t r a l a c t i v i t y I were determined and are given i n F i g . 25. The a c t i v i t i e s of alumina neutral a c t i v i t y super I and a c t i v i t y I are only 18% and 22%, res p e c t i v e l y , o f the f i r s t batch of alumina neutral, and the percentage of glucose adsorbed under s i m i l a r conditions (0.8 g c a t a l y s t used with 60 ml 0. M glucose solution) was about 8% compared with 10% adsorption with alumina neutral. However, according to the manufacturer (Woelm Pharma) the BET 2 surface area of alumina neutral a c t i v i t y super I i s 200 m /g compared with 2 140 m /g f o r alumina neutral. The nature of the exposed surface and the pore s i z e d i s t r i b u t i o n may be responsible f o r the decrease i n glucose adsorption by alumina a c t i v i t y super I with higher surface area. On the other hand a-alumina (manufactured by A l f a Products; contains 2 90% klyO^ and 9% ^ 0 ; surface area 320 m /g) showed high a c t i v i t y while Puratronic aluminum oxide, (distributed by A l f a Products; 99.999% metals * C a t a l y t i c a c t i v i t i e s of a c i d i c , b a s ic and neutral aluminas for t i c w i l l be compared i n Section XII 5.5 o i H C 3.5V 3.0V: [a-D-Glucose] = 0.05 M [Alumina] = 13.3 mg.ml -1 O and x change in a c t i v i t y of alumina neutral ( f i r s t batch) for t i c i n 3 weeks Change i n a c t i v i t y a f t e r 1 year A c t i v i t y of the second batch of alumina neutral f or t i c a Alumina a c t i v i t y super I « Alumina a c t i v i t y I * Alumina a c t i v i t y V 50 100 150 200 Time (min) 250 300 Fig. 25 Reproducibility of the Kinetics with Alumina Neutral and Comparison of the Cat a l y t i c A c t i v i t i e s of Different Chromatographic Aluminas (Neutral) VO 130 F i g . 26 Comparison of the C a t a l y t i c A c t i v i t i e s of Some Non-Chromatographic Aluminas 131 basis) had very low a c t i v i t y (Fig. 26). A sample of pure alumina prepared by hydrolysis of aluminum isopropoxide (as described i n the Experimental Section) showed high a c t i v i t y , probably because of high surface area (percentage adsorption 28% compared with 14% adsorption by alumina neutral under s i m i l a r conditions) and the presence of highly active c a t a l y t i c s i t e s on the surface The low a c t i v i t i e s observed with Puratronic aluminum oxide, alumina n e u t r a l a c t i v i t y super I and alumina neutral a c t i v i t y I may be due to the highly dehydroxylated nature of the surface . This i s supported by the f a c t that the a c t i v i t y of alumina a c t i v i t y V prepared by adding 19% water to alumina a c t i v i t y super I i s about 2.5 times the a c t i v i t y of the l a t t e r (Fig. 25), and also by the studies of the e f f e c t of heat on the c a t a l y t i c a c t i v i t y of t i c aluminas described i n the next section. * According to the manufacturer the water l o s t by alumina a c t i v i t y super I on i g n i t i o n at 1000°C i s 1% by weight. EFFECT OF DEHYDRATION ON THE CATALYTIC ACTIVITY OF ALUMINA 133 XII EFFECT OF DEHYDRATION ON THE CATALYTIC ACTIVITY OF ALUMINA It was mentioned i n the Introduction (Section 1.1.3) that on heating a sample of alumina which has water molecules adsorbed on the surface, f i r s t , some of the water molecules are desorbed and some react with the surface forming OH groups. As the sample i s heated further, a c i d i c and basic hydroxyl groups eliminate water to form Lewis acid and b a s i c s i t e s . When the sample i s heated above 300° to 400°C, defect s i t e s , which are c a t a l y t i c a l l y more ac t i v e , are formed (Section 1.1.4). As the temperature i s increased further, migration of ions tend to decrease the strength of the a c i d i c and basic s i t e s and at temperatures >1100°C a-alumina, which i s normally considered to be c a t a l y t i c a l l y i n a c t i v e , i s formed. Hence the change i n a c t i v i t y per unit surface area as the sample i s heated can give information on the nature of the act i v e s i t e s . For example, i f only Lewis acid s i t e s or/and basic oxide ions are c a t a l y t i c a l l y a c t i v e , an increase i n surface a c t i v i t y per unit area w i l l be observed, while there should be a decrease i n a c t i v i t y i f only Bronsted acid s i t e s or/and basic hydroxyl ions are ac t i v e . On the other hand, because d i f f e r e n t groups can have d i f f e r e n t s p e c i f i c a c t i v i t i e s , i f several d i f f e r e n t types of surface groups are active i n the mutarotation i t would be d i f f i c u l t to predict the change i n a c t i v i t y caused by heating. To determine the e f f e c t of dehydration, alumina neutral was heated under d i f f e r e n t conditions and the loss i n weight was determined. Conditions used to dehydrate alumina, the loss i n weight and the surface areas of the r e s u l t i n g samples (from Section II.1) are tabulated i n the f i r s t three columns of Table VII. A c t i v i t i e s of the samples were determined as described 134 TABLE VII EFFECT OF DEHYDRATION ON CATALYTIC ACTIVITY Dehydration Percentage Surface Rate Constant Rate Constant Percentage Conditions Loss i n Area m2/g - 10^ s e c - 1 g - 1 per U n i t Area Glucose Weight 10 6 s e c - 1 m~2 Adsorbed 1 None 140 1.9 ± 0.1 1.4 14 2 Room Temp (24°C), 0. 01 mm press 2.8 over P2°5 4 days 144' 1.4 ± 0.2 1.0 14 3 150 ± 5°C 0.01 mm press 5.7 2 days 130 0.9 ± 0.1 0.7 14 4 600 ± 50°C under dry N^ 4 hours 6.2 116 0.5 ± 0.05 0.43 13 5 800 ± 50°C under dry ^  6.2 4 hours 100 0.7 ± 0.1 0.7 10 6 1100 ± 50°C under dry N 2 7.2 3 hours 14 0.54 ± 0.02 3.9 3.5 7 1100 ± 50°C under dry N^ 6 hours 1.53 ± 0.05 3.5 8 1250 ± 50°C under dry N, 6 hours 1 7.9 6.2 2.20 ± 0.05 36 F i r s t order rate constant calculated f o r 60 ml 0.05 M a-D-glucose s o l u t i o n with 1.6 g alumina using natural logarithms. Errors were estimated from the maximum and minimum slopes. Estimated from surface area of alumina neutral and i t s loss of weight on evacuation. 135 e a r l i e r using 1.6 g samples of the c a t a l y s t and 60 ml 0.05 M solutions of a-D-glucose i n DMSO (Fig. 27). The f i n a l o p t i c a l rotations (a r o) for samples with low a c t i v i t y were determined by adding a few drops of 0.1 N NaOH or n-butylamine, or by s t i r r i n g the c a t a l y s t with an e q u i l i b r a t e d glucose solu t i o n . The a c t i v i t i e s per gram and a c t i v i t i e s per unit area of the c a t a l y s t ( i n the 60 ml 0.05 M glucose s o l u t i o n ) , and the percentage of glucose adsorbed are given i n columns 4, 5, and 6 of Table VII. The change i n a c t i v i t y per unit area with temperature of dehydration i s plotted i n F i g . 28. Results show that the c a t a l y t i c a c t i v i t y decreases as the sample i s heated upto 600°C. However,the main c h a r a c t e r i s t i c s of the curved f i r s t order p l o t s ( i n i t i a l adsorption followed by deactivation of the c a t a l y s t ) have been retained. At 1100°C, when alumina has undergone extensive s i n t e r i n g and change i n c r y s t a l structure to form a-alumina, the c a t a l y s t shows a complete change i n i t s c a t a l y t i c behaviour. With t h i s sintered (1100°) alumina the amount of glucose adsorbed has decreased as the surface area decreased and yet the c a t a l y s t showed a c t i v a t i o n rather than deactivation. In addition,the f i r s t order pl o t s were l i n e a r over three h a l f - l i v e s and the c a t a l y t i c a c t i v i t y increased further on heating to 1250°C. This increase i n a c t i v i t y of the c a t a l y s t with decrease i n surface area causes a rapid increase i n s p e c i f i c a c t i v i t y ( a c t i v i t y per unit area) as shown i n Fi g . 28 . Such high a c t i v i t y for a-alumina (compared with y-alumina) i s v i r t u a l l y unknown. As mentioned i n the Introduction (Section 1.1.2), a-alumina i s considered to be c a t a l y t i c a l l y less active than y-alumina and has shown a c t i v i t y towards only a few reactions. R e p r o d u c i b i l i t y of the r e s u l t s with alumina dehydrated at room temperature, 150°C, 600°C and 800°C was very good. Samples, when stored 136 Time (min) F i g . 27 The E f f e c t of Dehydration Temperature of Alumina Neutral on Its C a t a l y t i c A c t i v i t y 137 F i g . 23 The E f f e c t of Dehydration Temperature of Alumina Neutral on Its C a t a l y t i c A c t i v i t y per Unit Area 138 under anhydrous conditions, did not show any change i n a c t i v i t y over several weeks. By dehydrating alumina neutral under the same conditions new batches of dehydrated alumina with the same a c t i v i t y could be prepared. However, the r e p r o d u c i b i l i t y of k i n e t i c s with alumina dehydrated at 1100°C was not good. A c t i v i t y of a s i n g l e batch decreased with time even when stored under anhydrous conditions. A c t i v i t y of d i f f e r e n t batches prepared under the same conditions varied as much as 20%. Mixing a batch i n a m i n i m i l l f o r several hours did not improve the r e s u l t s . A c t i v i t y of alumina dehydrated at 1250°C showed much bett e r r e p r o d u c i b i l i t y . New batches with almost the same a c t i v i t y could be prepared but i t also decreased i n a c t i v i t y , over several days, when stored under anhydrous conditions. R e p r o d u c i b i l i t y of k i n e t i c runs with alumina sintered at 1250°C was very good when ca r r i e d out withi n a few hours of each other. Hence alumina dehydrated f o r 6 hours at 1250°C was used, where necessary, i n the mechanistic studies described i n thi s t h e s i s . Absence of deactivation i n alumina heated to 1100°C and higher temperatures appears, at f i r s t , to be rel a t e d to the lack of porosity i n the sintered alumina (see Section II.2 and Appendix A.2). However, i t i s u n l i k e l y that the presence of pores i n low temperature aluminas (dehydrated at 800°C or less ) i s responsible f o r deactivation by blocking of the pores by substrate molecules. The dimensions of the pores observed i n standard o alumina neutral (most between r = 18 and 27 A) and the pores i n aluminas P heated to 800 C (most between r^= 22 and 31 A) are both much larger than the dimensions of a glucose molecule (radius of pyranose form ^ 4.5 . It i s more l i k e l y that the deactivation i s caused by the presence of strongly adsorbing s i t e s on the surface which were observed on alumina neutral during adsorption studies described i n Section V, and w i l l be referred to again i n 139 Section XIV on Adsorption Isotherms. The presence of c a t a l y t i c s i t e s which can be converted to i n a c t i v e or les s active sites,when glucose molecules are allowed to i n t e r a c t with low temperature aluminas f o r a s u f f i c i e n t l y long time,is probably due to the presence of reac t i v e f u n c t i o n a l groups at the c a t a l y t i c s i t e s causing the glucose molecules to undergo slow but i r r e v e r s i b l e reaction (e.g. ether formation) with the surface. X I I . l Cause of High C a t a l y t i c A c t i v i t y of ct-Alumina Examination of F i g . 28 shows that the s p e c i f i c a c t i v i t y of alumina has increased r a p i d l y when the p y r o l y s i s temperature was increased over about 113 800°C. The Tammann temperature f o r alumina i s about 870°C. Above the Tammann temperature i o n i c d i f f u s i o n (volume d i f f u s i o n ) begins to occur at an appreciable rate, causing a rapid increase i n the p l a s t i c i t y of the s o l i d . This r e s u l t s i n a marked acc e l e r a t i o n of the rate of s i n t e r i n g as observed i n Table VII. The volume d i f f u s i o n above the Tammann temperature can also r e s u l t i n a change i n the composition of the c r y s t a l surface depending on the impurities present i n alumina. As mentioned i n Section II.5, there are about 0.2% of c a t i o n i c impurities i n alumina neutral used i n these studies and hence, a r e d i s t r i b u t i o n of ions may be responsible f o r the high c a t a l y t i c a c t i v i t y of the a-alumina produced by p y r o l y s i s . To check whether t h i s increase i n a c t i v i t y towards mutarotation on p y r o l y s i s at high temperature i s c h a r a c t e r i s t i c of a l l neutral aluminas, the same experiments were performed with a second batch of Woelm alumina ne u t r a l , purchased from ICN Pharmaceuticals. This alumina was progressively deactivated, rather than activated, on further heating above 800°C ( F i g . 29). — However, since a-alumina produced at 1250°C has a low surface area per unit weight the s p e c i f i c a c t i v i t y of the sintered alumina may be higher than that of the o r i g i n a l alumina sample. 140 The same behaviour was observed with alumina neutral f o r t i c (aluminim oxide 150 neutral Type T) purchased from BDH Chemicals. Examination of the composition of samples,given i n the Experimental Section,and the pH's of the 10% s l u r r i e s i n d i s t i l l e d water, given i n Table VIII, does not i n d i c a t e the cause of the increase i n a c t i v i t y observed with the f i r s t batch of alumina neutral. TABLE VIII pH's OF 10% SLURRIES IN WATER Alumina Sample Neutral Woelm f i r s t batch O r i g i n a l Sample Sintered (1250°C >6 hrs) Sample 7.5 10.1 Neutral Woelm second batch 7.5 9.3 Neutral BDH 7.6 A c i d i c Woelm 4.5 9.2 Basic Woelm 9.8 10.3 Therefore,the c a t a l y t i c a c t i v i t i e s of both a c i d i c and basic aluminas (Woelm) f o r tic,purchased from ICN Pharmaceuticals, were examined to determine the cause of t h i s d i f f e r e n c e i n behaviour. As shown i n F i g . 29 the a c t i v i t y of alumina basic i s close to that of the f i r s t batch of alumina ne u t r a l , while that of alumina a c i d i c i s only about 30% of the a c t i v i t y of 141 5.2* 60 ml 0.05 M a-D-Glucose / 1.6 g Alumina neutral (second batch) dehyd. at 1250°C 4.7 4.2 s i a o tH Pi 3.7 1.6 g Acidic alumina 1.6 g A c i d i c alumina dehyd. at 1250°C 3.2 1.6 g Alumina\ 0.16 g Alumina basic dehyd. \ basic dehyd. at at 1250°C I 1250°C 11 1.6 g Alumina neutral (second batch) 1.6 g Alumina neutral ( f i r s t batch) 1.6 g Alumina basic 1.6 g Alumina neutral ( f i r s t batch) dehyd. at 1250°C 2.71 30 60 90 120 Time (min) 150 180 F i g . 29 Comparison of the C a t a l y t i c A c t i v i t i e s of A c i d i c , Basic, and Neutral Aluminas (for TLC) and the E f f e c t of Dehydration at 1250°C for '6 hrs on Their C a t a l y t i c A c t i v i t i e s 142 alumina n e u t r a l . Both a c i d i c and basic aluminas underwent deactivation during c a t a l y s i s and adsorbed 13% and 16% of the glucose i n s o l u t i o n (1.6 g c a t a l y s t used with 60 ml 0.05 M glucose s o l u t i o n ) . When they were sintered at 1250°C, the a c i d i c and basic aluminas showed a marked di f f e r e n c e i n c a t a l y t i c a c t i v i t y . Alumina a c i d i c showed only a s l i g h t increase (^10%) i n a c t i v i t y on p y r o l y s i s at 1250°C and i t d i d not undergo deactivation and adsorbed only 3% glucose l i k e other sintered aluminas. However, alumina basic became highly a c t i v e although i t also adsorbed only 3% glucose from s o l u t i o n . In f a c t i t s a c t i v i t y was too high to measure accurately at normal concentrations (26.7 mg/ml) of aluminum oxide (Fig. 29). When k i n e t i c s was c a r r i e d out using 2.7 mg/ml of alumina, the c a t a l y s t showed an increase i n a c t i v i t y with time and the i n i t i a l rate constant for the same weight of sintered c a t a l y s t was about 30 times that f o r alumina basic for t i c . I t i s i n t e r e s t i n g to note that pH's of s l u r r i e s of a l l aluminas are higher a f t e r s i n t e r i n g at 1250°C. This indicates an increase i n surface b a s i c i t y on p y r o l y s i s but does not explain why some aluminas become activated while others become i n a c t i v e on s i n t e r i n g . The increase i n a c t i v i t y observed with alumina basic when pyrolysed at 1250°C suggests that the somewhat s i m i l a r behaviour observed with the f i r s t batch of alumina neutral f or t i c may be due to the presence of r e s i d u a l basic character i n that batch of neutral alumina. Hence i t seemed possible to prepare a sample of alumina which would behave s i m i l a r to the f i r s t batch of alumina neutral by t r e a t i n g the second batch of alumina ne u t r a l with a base, or by p a r t i a l n e u t r a l i z a t i o n of alumina basic with an a c i d . Preparation of a batch of alumina which behaves s i m i l a r to the f i r s t batch of alumina neutral by t r e a t i n g alumina basic with an acid i s described i n Appendix B. 143 The above r e s u l t s and discussion show that the f i r s t batch of alumina neutral used i n t h i s work i s not a t y p i c a l alumina neutral but i t appears to possess some basic character which i s brought out by p y r o l y s i s . In any case, as described above, highly a c t i v e sintered alumina can be prepared from alumina basic. I t was pointed out i n the Introduction (Section 1.2.3, Table III) that bases are v e r y ' e f f i c i e n t c a t a l y s t s of the mutarotation reaction. The fac t that high a c t i v i t y i s observed with sintered alumina prepared from basic alumina or p a r t i a l l y n eutralized basic alumina but not with sintered alumina prepared from neutral or a c i d i c alumina suggests that the high a c t i v i t y per unit area of sintered basic aluminas may be due to predominance of basic s i t e s at temperatures above the Tammann temperature. It would increase the c a t a l y t i c constant for surface reaction (k^ + k^) i n Equation (32) (this w i l l be discussed further i n Section XVIII.1). The observed rate constant would also be affected by changes i n Equilibrium constant f o r adsorption, as w i l l be discussed i n Section XIV, where the adsorption isotherms are presented. XII.2 Some Observations on the Nature of Active Sites It was pointed out at the beginning of t h i s section that dehydration causes, i n i t i a l l y , the loss of water adsorbed on the surface and l a t e r , the dehydroxylation of a c i d i c and basic hydroxyl groups. From F i g . 28 i t i s c l e a r that i n i t i a l dehydration has caused a decrease i n the s p e c i f i c a c t i v i t y of a l l the aluminas towards glucose mutarotation. This e f f e c t of mild thermal treatment i s unlike other reactions where c a t a l y t i c a c t i v i t y i s observed only on heating above 300 to 400°C (giving r i s e to defect s i t e s ) . Hence i t i s clear that defect s i t e s (consisting of clusters of vacancies and neighboring oxide i o n s ) , even i f they possess c a t a l y t i c a c t i v i t y 144 towards mutarotation, do not possess a monopoly over the c a t a l y t i c a c t i v i t y as has been observed with many other reactions. Further, the presence of c a t a l y t i c a c t i v i t y on aluminas containing a large amount of adsorbed water molecules i n d i c a t e that Lewis acid s i t e s are not e s s e n t i a l for c a t a l y t i c a c t i v i t y (this w i l l be discussed further i n Section XVIII.2.1). These r e s u l t s emphasize the f a c t that glucose mutarotation d i f f e r s from a l l the reactions previously studied on alumina as mentioned i n the Intro-duction. This d i f f e r e n c e i n behavior should necessarily be r e l a t e d to the high s e n s i t i v i t y of the mutarotation reaction to a c i d i c and basic s i t e s on the surface. It was shown e a r l i e r i n t h i s section (Table VII) that evacuation of alumina at room temperature, causing the loss of l o o s e l y bound water molecules without any change i n the nature of the surface, has decreased the c a t a l y t i c a c t i v i t y per unit area. Hence i t i s c l e a r that water plays an a c t i v e role i n the c a t a l y t i c process. The e f f e c t of water on the c a t a l y t i c a c t i v i t y w i l l be investigated i n the following section. XIII EFFECT OF WATER ON THE CATALYTIC ACTIVITY 146 XIII EFFECT OF WATER ON THE CATALYTIC ACTIVITY XIII.1 E f f e c t of Water on Alumina Evacuated at Room Temperature It was pointed out i n the previous section that the removal of adsorbed water by evacuation at room temperature under P2°5 ^ a s c a u s e d a decrease i n the c a t a l y t i c a c t i v i t y . To check whether the decrease i n a c t i v i t y i s r e v e r s i b l e , 3% water ( ^ l o s s i n weight) was added to a sample of evacuated alumina held i n an erlenmeyer f l a s k , stoppered, mixed w e l l and was allowed to stand overnight at room temperature. K i n e t i c s was carried out as described before using 0.8 g of the hydrated sample (Fig. 30). I t i s clear that the addition of water, l o s t from alumina by mild heating has increased i t s a c t i v i t y to the o r i g i n a l value. Hence,addition and loss of water on y-alumina at room temperature i s a r e v e r s i b l e process. XIII.2 E f f e c t of Water on Alumina Dried at 150°C Dehydration at higher temperatures causes some water molecules to react with the surface and surface hydroxyl groups combine eliminating water as described i n Section 1.1.3. Further, surface areas decrease (see Section II.1) and the c r y s t a l structure begins to change (Sections I.1.1 and II.4). Hence the change should be i r r e v e r s i b l e when rehydration i s attempted at room temperature. Experiments were conducted by adding water as described above, to alumina dried at 150°C, f o r 3 days. Results given i n F i g . 31 show that the a c t i v i t y i s increased by addition of water, but the a c t i v i t y reaches only about 56% of the o r i g i n a l value ( i . e . the a c t i v i t y of alumina neutral) by the addition of 5% water which was o r i g i n a l l y l o s t from the alumina. Hence the hydrated 150°C alumina i s s t i l l less a c t i v e than standard alumina 147 5.2, 8 I O rH C 100 200 300 Time (min) F i g . 30 The E f f e c t of Water on Alumina Dried at Room Temperature 148 5 .21 ta-D-Glucose] = 0.05 M [Alumina] = 13.3 mg.ml -1 4.7 s i £5 4.2 3.71 3.21 • A c t i v i t y of alumina neutral • A c t i v i t y a f t e r dehyd. at 150°C •I A c t i v i t y a f t e r adding 1% water to the d r i e d sample O A c t i v i t y a f t e r adding 3% water X A c t i v i t y a f t e r adding 5% water O A c t i v i t y after_adding 7% water 100 200 300 Time (min) F i g . 31 The E f f e c t of Water on Alumina Dried at 150°C 149 n e u t r a l . Addition of excess water (7%) appears to keep the i n i t i a l a c t i v i t y constant, But slows down the rate of d e a c t i v a t i o n ( F i g . 31). XIII.2.1 Cause of the Increase i n A c t i v i t y by Water There may be many causes for the increase i n a c t i v i t y by water added onto dehydrated aluminas. I t may be due to an independent c a t a l y t i c process by water adsorbed on the surface. Water might also act as a c a t a l y s t promoter and increase the a c t i v i t y of c a t a l y t i c s i t e s ( i . e . i n c r e -ase (k^ + k^)) or/and a c t i v a t e c a t a l y t i c a l l y i n a c t i v e s i t e s ( i . e . increase e ). I t might e a s i l y form hydrogen bonds with glucose and thus, might increase the amount of glucose adsorbed on the surface ( i . e . increase equilibrium constant K). To determine how water increases the c a t a l y t i c a c t i v i t y , experiments were ca r r i e d out by adding water to a known weight of alumina dehydrated at 150°C (for 2 days). Water was added to the alumina in the usual reaction f l a s k , the f l a s k was then stoppered and rotated 2.5 hours to mix the s o l i d with the water. The rate of mutarotation was followed a f t e r addition of 60 ml 0.05 M a-D-glucose s o l u t i o n to the r e a c t i o n f l a s k . The experiment was repeated by adding d i f f e r e n t amounts of water to the same weight of alumina (to keep the surface area constant). The r e s u l t s are given i n F i g . 32 where the i n i t i a l c a t a l y t i c a c t i v i t y per unit weight of c a t a l y s t i s plotted against the percentage (by weight) of water added. I t shows that the c a t a l y t i c a c t i v i t y has increased l i n e a r l y with the increase of amount of water u n t i l about 3% by weight of water has been added. Further addition of water increased the rate only very slowly; the r a t i o of the slopes i n the two regions being about 200 : 1. 150 F i g . 32 Relation of the Observed Rate Constant to the Amount of Water Added to Alumina Dehydrated at 150°C 151 It was mentioned at the beginning of t h i s section that the increase i n a c t i v i t y by water can a r i s e from the increase of the c a t a l y t i c constant (k^ + k j ) , the concentration of c a t a l y t i c a l l y a c t i v e s i t e s c , and the equilibrium constant for adsorption K, and also from an independent c a t a l y t i c process by water. As described i n Section VII, the observed rate constant i s related to K. c , and (fe„ + fe,) by Equation (32). o 3 4 ^ obs (fe + fe.) c 3 4 o 1/K + s (32) 1/K + s o/ L (fe_ + fe.) a 3 4 o (34) According to Equation (34), the observed rate constant i s a product of two fa c t o r s ; one determined by the equilibrium constant f o r adsorption and the substrate concentration which i s kept constant, and the other i s the product of the c a t a l y t i c constant f o r the surface reaction and the concen-t r a t i o n of the c a t a l y t i c a l l y a c t i v e s i t e s . Hence, the r e s u l t s can be analyzed i n terms of the e f f e c t of water on the two fa c t o r s . XIII.2.1.1 E f f e c t of Water on the Amount of Glucose Adsorbed When the e f f e c t of water on the rate of reaction was studied (Section XIII.2), no change (within experimental error of about 5%) i n the amount of glucose adsorbed was detected (Table IX), using a 0.1 dm path length c e l l , even though a 60% increase i n the rate constant was observed. This means that there i s no change i n the equilibrium constant f o r 152 TABLE IX EFFECT OF WATER ON EQUILIBRIUM OPTICAL ROTATION Percentage (by weight) Water Added onto 150°C Alumina O p t i c a l r o t a t i o n (degrees) at t = 00 0.124/0.125 0.124 0.124 0.124 0.124 0.124 15 0.124 adsorption and hence, the f a c t o r ( . *~ ) should be independent of the X/ K. T S o concentration of water on the surface. Therefore the observed increase i n the rate constant with addition of water onto alumina should be due to an increase i n the factor (k^ + o r due to an independent c a t a l y t i c process by water on the surface, as discussed below. 153 XIII.2.1.2. E f f e c t of Water on the Factor (/c_ + k.)c and the Expression 3 4 o *-for the Observed Rate Constant i n the Presence of an Indepen-dent C a t a l y t i c Process due to Water Using Equations (13) and (14), the factor (k^ + k^c^ i n Equation (32), can be expressed as a sum of several terms, ( k 3 + kL)cQ = kA[A] + kB[B] + k^B[A.B] (35) where k., k , and k. „ are the c a t a l y t i c constants for acid, base, and A D A .tl b i f u n c t i o n a l c a t a l y s i s , r e s p e c t i v e l y . [A] and [B] are the concentrations of a c i d i c and basic s i t e s , r e s p e c t i v e l y , and [A.B] i s the concentration of acid/base p a i r s i t e s capable of b i f u n c t i o n a l c a t a l y s i s of glucose mutarotation. I f there i s an independent c a t a l y t i c process by water (see Section 1.2.5.1.3.1), then the observed rate constant i s given by, H 2 ° n k f = k , + Ic, n[H,0] (36) obs obs 1^0 2 where k , i s the observed rate constant i n the absence of water, obs On the other hand, i f water acts as an acid or a base favouring b i f u n c t i o n a l mechanism, then, (fc 3 + kl)aQ = kA[A] + kg[B] + kA B[A.B] + k'[A or B.H 20] (37) In such a case, the presence of water on the alumina surface w i l l change the c a t a l y t i c constant for the surface react i o n . 154 Another p o s s i b i l i t y i n a surface catalyzed acid/base reaction i s that water can act as a medium to tr a n s f e r protons or hydroxide ions from a c i d i c or b asic s i t e s on the surface to the adsorption s i t e s . Thus, adsorption s i t e s which are not c a t a l y t i c a l l y a c t i v e w i l l become e f f e c t i v e mutarotation c a t a l y t i c s i t e s on the addi t i o n of water. In such a case, Equation (35) would s t i l l apply with higher values for the concentration of a c i d i c and basic s i t e s . For equations of type (36) the order (n) with respect to water has been found to be 2 or 3 from studies i n non-aqueous solvents as mentioned i n the Introduction (Section 1.2.5.1.3.1). The observed l i n e a r increase i n observed rate constant with increase i n concentration of water shows that f o r the surface reaction n = 1 and hence,independent c a t a l y s i s by water i s not possi b l e . This conclusion w i l l be supported by deuterium isotope e f f e c t studies discussed l a t e r i n Section XIX. Hence water should increase the observed rate constant either by acting according to Equation (37) as one of the p a r t i c i p a n t s of a b i f u n c t i o n a l system or by acting as a medium for proton or hydroxide ion transfer. Both of these mechanisms would show a l i n e a r increase i n the observed rate constant with increase i n the concentration of water. Again, the deuterium isotope e f f e c t studies discussed i n Section XIX w i l l be used to d i s t i n g u i s h between the two p o s s i b i l i t i e s . XIII.3 E f f e c t of Water on the A c t i v i t y of Alumina Dehydrated at  800°C and 1250°C Alumina dehydrated at 800°C was very s e n s i t i v e to water and i t s a c t i v i t y increased more than 300% (see Section XIX) on treatment with water. Thus, hydration produced a ca t a l y s t which i s more active than the 1 5 5 o r i g i n a l alumina neutral. However, alumina dehydrated at 1250°C gave irreproducible results on treatment with water. The a c t i v i t y of the catalyst always increased on treatment with water (freshly d i s t i l l e d ) but the increase i n a c t i v i t y appeared to depend on contact time as w e l l as other unknown factors. The results discussed i n the previous sections have shown that there are two forms of alumina with characteristic properties towards glucose mutarotation. One i s y-alumina ( porous ) with low a c t i v i t y per unit area which undergoes deactivation during mutarotation and the other i s low surface area a-alumina (non-porous) with high a c t i v i t y per unit area which shows constant a c t i v i t y during a k i n e t i c run. This difference i n behaviour should at least partly be determined by the surface area available for glucose adsorption and the strength of adsorption. These factors are investigated i n the next section where the glucose adsorption isotherms are presented. ADSORPTION OF GLUCOSE ON ALUMINA SURFACE PART II ADSORPTION ISOTHERMS 157 XIV ADSORPTION OF GLUCOSE ON ALUMINA SURFACE PART II ADSORPTION ISOTHERMS The f i r s t t h e o r e t i c a l equation which describes the r e l a t i o n s h i p between the amount of gas adsorbed and the equilibrium pressure of the gas 123 at constant temperature was advanced by Langmuir . The basic assumption of the theory was that adsorption was l i m i t e d to formation of a unimolecular layer. Langmuir theory can also be applied to adsorption of non-electrolytes from s o l u t i o n , on the assumption that adsorption i s e s s e n t i a l l y confined 1X3 12 A to a monolayer next to the surface ' . The assumption i s v a l i d i f the solute molecules i n t e r a c t with the surface and not with each other so that bulk s o l u t i o n e x i s t s above the monolayer of solute on the surface. In Section V . l . l the adsorption of glucose onto the alumina surface was represented by the equation, G + C •« * GC (22) This i s an o v e r s i m p l i f i c a t i o n because the solvent molecules also get adsorbed on the surface and hence,the adsorption process i s best considered as a competition between the solute and solvent molecules (Equation (38)). This leads to the formation of an i d e a l two dimensional s o l u t i o n of solute 113,124 and solvent molecules on the surface *1 G + DC m * cr. + D (38) v • K2 where D i s the solvent DMSO. 158 r .... . , , „, kl IGC] [D] Equilibrium constant A = -^j = [£] [DC] For a d i l u t e s o l u t i o n [D] i s a constant and therefore, K% k, [GC] 1 K (39) [D] k'2[D] [G][DC] where K i s the observed equilibrium constant discussed i n Sections V . l . l and VII.1. Hence i t follows that the f i r s t order rate constant fc mentioned i n those sections i s r e a l l y a pseudo f i r s t order rate constant, for i t i s equal to the product of a second order rate constant and a concentration term which i s a constant. From Equation (39) i t also follows that, the greater the strength of adsorption the greater i s the value of [GC] for a given concentration of glucose and greater i s the magnitude of the equilibrium constant. I f [GC] i s the maximum concentration of adsorbed glucose at a constant c a t a l y s t concentration and at constant temperature then, [GC] = [GC] + [DC] [DC] = [GC]m - [GC] and from Equation (39), [GC] = K[G][DC] = K[G]([GC]m - [GC]) I ^ J - = K[G] [GC] . . m m ( l - ^ - ) \ [GC]J 159 a n d 1G£L_ m J0G]_ ( 4 0 ) [GC]M 1 + K[G] which i s the f a m i l i a r form of a Langmuir equation for adsorption of glucose 113 12 A onto alumina ' . Hence a p l o t of the moles of glucose adsorbed per gram of the c a t a l y s t versus the equilibrium concentration of glucose should show a l i n e a r increase at low glucose concentrations, with a slope proportional to K, and should show a constant maximum adsorption at high glucose concentrations. Furthermore, the smaller the value of K (weak adsorption) the higher i s the concentration of glucose at which the plateau i s reached. Equation (40) can be transformed to, [GC] + K[GC] = K[GC] (41) m [G] r GC] If K i s a constant then a graph of -|-^— versus [GC] would be l i n e a r with slope = -K and intercept = K[GC]^. Hence,if [GC]^ i s known then,Z can be determined from the intercept as w e l l , whether or not K i s a constant depends on the energy of the s i t e s and on the l a t e r a l i n t e r a c t i o n s of adsorb glucose molecules. In the i d e a l case a l l adsorption s i t e s would be e n e r g e t i c a l l y homogeneous and adsorbate molecules would not i n t e r a c t with each other on the surface. Under such conditions a l i n e a r r e l a t i o n s h i p would be observed f o r a l l values of 0, the f r a c t i o n of surface occupied by the solute. But most surfaces are e n e r g e t i c a l l y heterogeneous and 123 there i s l a t e r a l i n t e r a c t i o n of adsorbate molecules on the surface However, a l i n e a r r e l a t i o n s h i p i s exhibited by many systems over c e r t a i n values of 6 when the two factors compensate each other and tend to keep 160 K constant. XIV.l Adsorption Isotherm f o r Alumina Neutral The procedure for obtaining the adsorption isotherm i s described i n d e t a i l i n the Experimental Section. Hence only a b r i e f account i s given here. An e q u i l i b r a t e d s o l u t i o n of a, 8 mixture i n DMSO (60 ml) was s t i r r e d with 1.6 g alumina neutral at 25°C u n t i l the o p t i c a l r o t a t i o n of the f i l t e r e d s l u r r y was e s s e n t i a l l y constant. The number of moles of glucose adsorbed onto the alumina and the equilibrium concentration of glucose were calculated from the change i n o p t i c a l r o t a t i o n of the s o l u t i o n and the f i n a l o p t i c a l rotation,respectively, using the c a l i b r a t i o n curve described i n the Experimental Section. The experiment was repeated with solutions of d i f f e r e n t concentrations and the r e s u l t s are given i n Table X. The adsorp-t i o n isotherm i n F i g . 33 was obtained by p l o t t i n g the amount of glucose adsorbed per gram c a t a l y s t against the equilibrium concentration of glucose. XIV.2 Adsorption Isotherm f o r Sintered Alumina To obtain the adsorption isotherm f o r alumina heated at 1250°C for 6 hours the same procedure was adopted,but with 3.2 g samples of alumina. The weight of c a t a l y s t used was doubled because of lower adsorption with sintered aluminas. The r e s u l t s are given i n Table XI and the adsorption isotherm i n F i g . 33. The adsorption isotherms show that the amount of glucose adsorbed onto alumina increases with increase of concentration of glucose i n s o l u t i o n TABLE X DATA FOR THE ADSORPTION ISOTHERM FOR GLUCOSE ONTO ALUMINA NEUTRAL AT 25.0°C I n i t i a l Concentration O p t i c a l Rotation (Degrees) of Glucose (M) I n i t i a l F i n a l Equilibrium Change i n Moles (x 10 ) Glucose Concentration O p t i c a l Rotation Adsorbed per Gram (M) (Degrees) Catalyst  0.0025 0.077 0.008 0.00029 0.070 0.91 0.005 0.146 0.049 0.0017 0.100 1.29 0.01 0.291 0.171 0.0059 0.120 1.55 0.025 0.720 0.571 0.0198 0.150 1.94 0.05 1.454 1.274 0.044 0.180 2.33 0.10 (a) 2.910 2.700 0.094 0.210 2.70 (b) 2.908 2.695 0.213 (c) 2.938 2.733 0.205 0.25 7.280 7.250 0.25 0.230 2.98 0.50 14.450 14.220 0.50 0.230 2.98 1.0 (a) 28.780 28.560 1.0 0.220 2.90 (b) 28.788 28.562 0.226 1.75 50.340 50.110 1.75 0.230 2.98 ON F i g . 33 Isotherms for Adsorption of Glucose on (a) Alumina Neutral and (b) Alumina Sintered at 1250°C for 6 hours. TABLE XI DATA FOR THE ADSORPTION ISOTHERM FOR GLUCOSE (AT 25.0°C) ONTO ALUMINA HEATED AT 1250°C FOR 6 HOURS I n i t i a l Concentration of Glucose (M) 0.0005 0.001 0.005 0.01 0.24 0.05 0.10 0.24 0.5 1.0 2.02 Op t i c a l Rotation (Degrees) Equilibrium Concentration I n i t i a l F i n a l 0.0.15 0.027 0.139 0.275 0.696 1.400 2.768 7.000 13.902 27.716 0.006 0.015 0.122 0.255 0.673 1.372 2.732 6.951 13.842 27.644 58.420 (a) 58.360 (M) 0.00021 0.00052 0.0042 0.0088 0.0232 0.0473 0.0943 0.240 0.488 0.954 2.01 Change i n Opt i c a l Rotation (Degrees) 0.009 0.012 0.017 0.020 0.023 0.028 0.036 0.049 0.060 0.072 0.07 ± 0.01 4 Moles (x 10 ) Glucose Adsorbed per Gram of Catalyst  0.06 0.08 0.11 0.13 0.15 0.18 0.23 0.32 0.39 0.46 0.45 ± 0.07 (b) 58.340 164 u n t i l a pleateau i s reached. This increase should be due to r e v e r s i b l e adsorption of glucose onto the alumina surface. The presence of r e v e r s i b l e adsorption s i t e s , discussed e a r l i e r i n Section V, can be confirmed by re l e a s i n g the adsorbed glucose into the s o l u t i o n . These tests f o r r e v e r s i b i l i t y of adsorption are described i n the Experimental Section. Further, from the plateau's of the adsorption isotherms, surface areas of the alumina samples can be estimated as described i n the next section. XIV.3 Maximum Amount of Glucose Adsorbed and the Surface Areas of Samples From the adsorption isotherms the maximum amount of glucose - 4 -4 adsorbed on alumina neutral i s 3.0 x 10 mole/g,and 0.45 x 10 mole/g for alumina heated at 1250°C f o r 6 hours. Since both isotherms show Langmuir type adsorption with saturation of adsorption at high concentration of glucose, the plateaus should correspond to completion of monolayer of glucose on the alumina surface. Hence the maximum amounts of glucose adsorbed can be used to estimate the area of the surface a v a i l a b l e f o r glucose adsorption. Using atomic models the average van der Waal's radius o O f the D-glucopyranose molecule was found to be about 4 A. The glucose molecule i s most l i k e l y adsorbed with the plane of the pyranose r i n g p a r a l l e l to the surface, since i t allows the maximum i n t e r a c t i o n of the hydroxyl groups with the surface. Hence the area occupied by a glucose o 2 - 4 molecule adsorbed on the alumina surface i s about 50 A -, and 3.0 x 10 2 mole of glucose molecules would occupy 90 m . On a gram of sintered -4 2 alumina 0.45 x 10 mole glucose molecules should occupy 13 m • It i s clear that the area occupied by glucose molecules i s only 62% of the area covered by a monolayer of nitrogen adsorbed on the same weight of alumina n e u t r a l . One po s s i b l e reason f o r the d i f f e r e n c e i s the presence 165 o of pores less than 4 A in radius into which only the nitrogen molecules (average van der Waal's radius^2 A ) can enter. But i t i s not likely that much of the surface area of the sample is present in very small pores, since i t can be shown that >75% of the surface area available for nitrogen o adsorption is present in pores with radii >13 A. As mentioned before, most o o of the pores in alumina neutral have radii between 18 A and 27 A. Therefore i t seems more likely that there are distinct sites on the Y-alumina surface for adsorption of glucose, which are determined by the surface structure of the catalyst. In other words,glucose molecules cannot occupy a l l the surface area available for nitrogen molecules and a relatively large fraction of y-alumina surface is l e f t vacant. The "effective area" of a glucose molecule adsorbed on the y-alumina surface would seem to be o2 about 80 A . With the assumptions given above in the calculation of surface areas, the surface area apparently occupied by glucose molecules adsorbed on sintered alumina is higher than the area determined by adsorption of 2 nitrogen (viz. 6.2 m /g; see Section II. 1 ) . Since the adsorption isotherms do not indicate the presence of multilayer adsorption, the higher area observed may indicate presence of an acyclic intermediate on the surface or of glucopyranose molecules adsorbed with plane of the ring not parallel to the surface. Of course, such details cannot be determined within the many assumptions involved in the area/g calculation; however, i t is interesting that area determinations are at least similar with use of nitrogen and of glucose. XIV.4 Strength of Adsorption of Glucose As mentioned in Section XIV, the greater the strength of adsorption the lower is the value of [G] at which plateau is reached. Comparison of 166 adsorption isotherms given i n F i g . 33 shows that on alumina neutral the monolayer i s complete when [G]^0.25 M while a glucose concentration of 1 M i s needed to complete the monolayer on sintered alumina. This indicates that at l e a s t part of the adsorption s i t e s on sintered alumina are weaker than the weakest s i t e s on alumina neutral. The strengths of adsorption of glucose by the two aluminas can be compared by determining the equilibrium constants f o r adsorption using Equation (41). XIV.4.1 Determination of Equilibrium constant K f o r Adsorption of Glucose  on Alumina Neutral It was mentioned i n Section XIV that the equilibrium constant K for adsorption of glucose onto alumina can be determined by p l o t t i n g 1 versus [GC], Since the Langmuir adsorption applies to r e v e r s i b l e adsorption, the concentration of glucose adsorbed r e v e r s i b l y was determined by subtracting the amount adsorbed on i r r e v e r s i b l e s i t e s (determined i n Section V.2) from the t o t a l amount of glucose on the surface (Table XII). The values of and [GC] are also shown i n the same Table and they are plotted i n F i g . 34. This p l o t shows two l i n e a r regions*; one with a high K, and the other with low K (plotted again F i g . 35). This indicates the presence of two types of s i t e s with d i f f e r e n t strengths of adsorption each giving r i s e to a l i n e a r p l o t . Equations that explain such behaviour can be derived as follows. If adsorption on s i t e s of Type 1 occurs independent of adsorption on s i t e s of Type 2, then Equation (41) can be applied to each type. The data do not f i t Temkin ( i . e . l i n e a r , [GC] versus ln[G] p l o t ) or Freundlich ( i . e . l i n e a r , lnJGC] versus l n [ c 7 ] p l o t ) i s o t h e r m s 1 1 3 , ^ . TABLE XII DATA FOR THE LANGMUIR PLOT OF [GC]/[G] VERSUS [GC] MOLE LITRE - 1 FOR ADSORPTION OF GLUCOSE ON ALUMINA NEUTRAL AT 25.0°C Equilibrium Glucose Concentration M Moles (x 10 ) Glucose Adsorbed per Gram Catalyst Moles (x 10 ) Glucose Adsorbed on Reversible Sites per Gram Catalyst  [GC] x 10 -1 mole l i t r e " (for 1 g dispersed i n 60 ml) [GC] [G] 0.00029 0.91 0.21 3.5 ± 0.3 1.1 ± 0.1 0.0017 1.29 0.59 9.9 ± 0.3 0.58 ± 0.02 0.0059 1.55 0.85 14.2 ± 0.3 0.24 ± 0.01 0.0198 1.94 1.24 20.7 ± 0.4 0.104 ± 0.002 0.044 2.33 1.63 27.2 ± 0.4 0.061 ± 0.001 0.094 2.70 2.00 33.3 ± 0.5 0.037 ± 0.0004 0.25 2.98 2.28 38.0 ± 0.5 0.0152 ± 0.0002 168 [GC] (x 10 m o l e . l i t r e ) F i g . 34 The Langmuir Plot for Adsorption of Glucose on Alumina Neutral 169 35 The Langmuir Plot for Adsorption of Glucose on Alumina Neutral at High Concentrations of Glucose 170 [GC]. i + KAGC], = KAGC] (42) [G] 1 i i ml [GC] and + KAGC]- = KAGC]m (43) [G] 2 2 2 where, and ^ a r e t n e equilibrium constants for adsorption on s i t e s of Types 1 and 2 r e s p e c t i v e l y . [GC]^ and [GC]^ are the concentrations of glucose adsorbed on s i t e s of Types 1 and 2,respectively, when the equilibrium concentration of glucose i s [G], and [GC] and [GC] are the maximum concentrations of glucose on s i t e s of Types 1 and 2 r e s p e c t i v e l y . Adding Equations (42) and (43) we get, [GC], + [GC]. ± - + K, [GC], + KAGC]. = K [GC] + KAGC] (44) [G] 1 2 If Z, » K~, and [GC] and [GC] are of the same order of magnitude, 1 2 mj_ rri2 then, [GC]^ » [GC] ^ a t l ° w concentrations of glucose [ff], and Equation (44) reduces to, [GC] - + KAGC]1 = K1[GC]m (42) [ff] * "1 [GC] + [GC] [GC] Hence a p l o t of versus [GC] + [GC] ^ [GC],, [G] [G] at low concentrations of glucose should be l i n e a r with slope = and intercept = KAGC]^ . Therefore the strength of adsorption on s i t e s of Type 1 and the number of s i t e s of Type 1 can be determined. When the glucose concentration i s high, [GC]^ = [GC]^ and Equation 171 (44) becomes [GC] + [GC]2 + K 1 [ G C \ + K 2 [ G C ] 2 = K 1 [ G C \ + K2[GC]m2 [GC]m + [GC]2 1 + KAGC]. = K [GC] (45) [G] 2 Adding A"_[GC] to both sides of Equation (45) we get, 2 [GC]m + [GC]2 - + K2([GC]2 + [GC]m ) = K2([GClm + lGC]m ) (46) [G] ' "1 ' 1 2 m 2 Hence a graph of versus (.[GC] + [GC] ) should be a [G] 1 [Cc?]^ + [c?c7] c7] s t r a i g h t l i n e with slope = -Z„ and intercept = KA[GC] + [GC] ). ^ ^ ^2 ^ 1 Therefore the strength of adsorption on s i t e s of Type 2 and the t o t a l number of s i t e s can be determined. Using t h i s treatment the experimental p l o t s i n F i g s . 34 and 35 w i l l be analyzed below. 2 From the region showing strong adsorption the slope = K = (8.2 ±1.2)10 l i t r e mole 1 and the concentration of s i t e s responsible for strong adsorption i s i n t e r c e P t _ (17 + 2)10 ^ mole l i t r e - " * " . slope From the slope of F i g . 35, K = 44 ± 4 l i t r e mole \ and using the intercept and concentration of a l l r e v e r s i b l e s i t e s (from Table XII) K2 = 47 ± 4 l i t r e mole "*". Hence Kn , > = 46 ± 3 l i t r e mole "*". Further, 2 (average) the concentration of weak adsorption s i t e s = concentration of a l l r e v e r s i b l e -4 adsorption s i t e s - concentration of strong adsorption s i t e s = (21 ± 2)10 172 mole l i t r e ^. Note that according to Equation (46) the concentration of a l l r e v e r s i b l e adsorption s i t e s = [GC] + [GC] , on one gram of ca t a l y s t dispersed i n 0.06 l i t r e s o l u t i o n , i s given by " L n ^ ^ ^ t - of the p l o t i n -4 -1 F i g . 35 and equals (41 ± 5)10 mole l i t r e . Within experimental error t h i s agrees with the experimentally observed maximum r e v e r s i b l e adsorption = (38 ± 0.5)10~ 4 mole l i t r e 1 (Table XII), and v a r i f i e s the a p p l i c a b i l i t y of the Equation (46) to the observed Langmuir p l o t . Therefore, the r e s u l t s from adsorption studies have given evidence for three types of adsorption s i t e s on alumina n e u t r a l . There are 0.7 x 10~ 4 mole (cs:23%) i r r e v e r s i b l e adsorption s i t e s , 1.0 x 10 4 mole (=03%) -4 strong r e v e r s i b l e adsorption s i t e s and 1.3 x 10 mole (^43%) weak reversib adsorption s i t e s on a gram of alumina n e u t r a l . Out of these three types of adsorption s i t e s , i r r e v e r s i b l e s i t e s cannot possess any c a t a l y t i c a c t i v i t y while both types of r e v e r s i b l e adsorption s i t e s have the p o t e n t i a l to be c a t a l y t i c s i t e s f o r the muta-ro t a t i o n reaction. Their c a t a l y t i c a c t i v i t y i s determined i n Section XV. XIV.4.2 Determination of the Equilibrium Constant for Adsorption of  Glucose onto Alumina Sintered at 1250°C Similar to the case of glucose on standard alumina neutral as described i n Section XIV.4.1, the equilibrium constant for adsorption of glucose onto sintered alumina can also be determined from the Langmuir r GC\ p l o t of r „•} versus [GC] . Data for the c l o t are given i n Table XITI \.G\ •k Since 1 g of c a t a l y s t i s found i n 0.06 l i t r e the amount of strong -4 -4 adsorption s i t e s on 1 g c a t a l y s t i s (17 x 10 ) ~K 0.06 = 1.0 x 10 mole. 173 TABLE XIII DATA FOR THE LANGMUIR PLOT OF Ic7C]/[c7] VERSUS [GC] MOLE LI T R E - 1 FOR ADSORPTION OF GLUCOSE (AT 25.0°C) ON ALUMINA SINTERED AT 1250°C Equilibrium Concentration of Glucose M Moles (x 10 ) Glu cose Adsorbed per Gram Catalyst [GC] ^ 10* mole l i t r e (for 1 g dispersed i n 60 ml) [GC] [G] 0.00021 0.6 1.0 0.48 0.00052 0.8 1.3 0.26 0.0042 1.1 1.8 0.044 0.0088 1.3 2.2 0.025 0.0232 1.5 2.5 0.011 0.0473 1.8 3.0 0.0063 0.0943 2.3 3.8 0.0041 0.240 3.2 5.3 0.0022 0.488 3.9 6.5 0.0013 0.954 4.6 7.7 0.0008 174 and the p l o t i n F i g . 36. Note that the amount of i r r e v e r s i b l e s i t e s ( i f any) was not determined and hence the q u a n t i t a t i v e data obtained from the pl o t may contain a small error. The p l o t i n F i g . 36 i s very s i m i l a r to that observed for alumina neutral which, as shown i n Section XIV.4.1, has two types of r e v e r s i b l e adsorption s i t e s . The strong adsorption s i t e s on sintered alumina have an equilibrium constant of about 6000 l i t r e mole 1 and there are 1.2 x 10 moles of strong adsorption s i t e s on a gram of c a t a l y s t . The equilibrium constant for adsorption on weak s i t e s i s 13 l i t r e mole 1 and there are 3.4 x 10 ^ moles of weak adsorption s i t e s on a gram of sintered alumina. The r e s u l t s discussed i n t h i s section have shown that both alumina neutral and alumina sintered at 1250°C contain two types of r e v e r s i b l e adsorption s i t e s . The c a t a l y t i c a c t i v i t i e s of both strong and weak re v e r s i b l e adsorption s i t e s on alumina neutral are determined i n the next section. 175 F i g . 36 The Langmuir Plot f o r Adsorption of Glucose on Alumina Sintered at 1250°C for 6 hours CATALYTIC ACTIVITY OF REVERSIBLE ADSORPTION SITES AND THE ACTIVE SITE DENSITY OF ALUMINA NEUTRAL 177 XV CATALYTIC ACTIVITY OF REVERSIBLE ADSORPTION SITES AND THE ACTIVE SITE DENSITY OF ALUMINA NEUTRAL For a r e v e r s i b l e c a t a l y t i c process given by equation k 1 S + C SC - PC n P + c (26) k 1 the i n i t i a l rate of the reaction (after adsorption i s complete) i s given by, Hence the i n i t i a l rate of the reaction i s d i r e c t l y proportional to the concentration of a-D-glucose, [SC], on the surface. The amount of substrate-catalyst complex formed for a given a-D-glucose concentration i s given by the adsorption isotherm. Therefore,if a l l r e v e r s i b l e adsorption s i t e s are equally active,then a p l o t of i n i t i a l rate of reaction versus concentration of glucose i n s o l u t i o n , and a p l o t of moles of glucose r e v e r s i b l y adsorbed versus the concentration of glucose i n s o l u t i o n , should be superimposable. To determine the i n i t i a l rate of the reaction at d i f f e r e n t glucose concentrations,plots of o p t i c a l r o t a t i o n versus time were made using the data from Section X. The slope at t = 25 mins gave the rate of the reaction i n degrees sec ^ (rates at high glucose concentrations were corrected for slow homogeneous r e a c t i o n ) . Using o p t i c a l rotations of equimolar solutions of a-D-glucose and B-D-glucose i t can be shown that the o p t i c a l r o t a t i o n t measured at 365 nm with a 0.1 dm path length c e l l 5 d e c r e a s e s by 1° when 0.012 mole a-D-glucose i n 60 ml DMSO i s converted to 0.012 mole g-D-glucose. Hence the number of moles r e a c t i n g per second i n the 60 ml s o l u t i o n i s d[P] dt 178 given by, slope (degrees sec "*") x 0.012 (mole degree "*") and are recorded i n Table XIV. The concentration of a-D—glucose i n s o l u t i o n when adsorption i s complete was determined from the adsorption isotherm. The p l o t s of i n i t i a l rate versus concentration of a-D-glucose and the adsorption isotherm for r e v e r s i b l e adsorption (from Table XII) are shown i n F i g . 3 7. They show that when adsorption i s about 37% complete the i n i t i a l rate i s only 7% of i t s maximum value. This indicates that s i t e s responsible for i n i t i a l adsorption (strong adsorption s i t e s ) are not as act i v e as weak adsorption s i t e s . In order to compare the a c t i v i t i e s of the two types of s i t e s , t h e i r adsorption isotherms were constructed using the equilibrium constants (K^ and X ) and the maximum concentrations of glucose adsorbed on the two s i t e s , [GC] ^  and [ G C ^ m ' a r e shown i n F i g . 38 together with the p l o t of i n i t i a l rate versus concentration of glucose. The values for i n i t i a l rate have been multipled by 0.0303 to coincide the maxima of the i n i t i a l rate and the adsorption isotherm due to weak adsorption s i t e s . From F i g . 38 i t i s clear that there i s no r e l a t i o n s h i p between i n i t i a l rate of the reaction and the adsorption isotherm due to strong adsorption s i t e s . For example,when adsorption on those s i t e s i s 75% complete the i n i t i a l rate i s only 7% of i t s maximum value. Hence the strong adsorption s i t e s are e i t h e r c a t a l y t i c a l l y i n a c t i v e or t h e i r a c t i v i t y i s r e l a t i v e l y very slow. The F i g . 38 also shows that within experimental error the adsorption isotherm due to weak s i t e s and the i n i t i a l rate p l o t are superimposable. Note that the i n i t i a l rates observed at low a-D-glucose concentrations are less than what i s expected from the adsorption isotherm. This i s because the assumption that the concentration of product (3-D-glucose i s n e g l i g i b l e would hold w e l l at high a-D-glucose concentrations but not at low TABLE XIV DATA FOR VARIATION OF INITIAL RATE WITH EQUILIBRIUM CONCENTRATION OF a-D-GLUCOSE AT CONSTANT CATALYST CONCENTRATION (26.7 MG/ML) I n i t i a l Rate I n i t i a l Concentration Concentration of (observed) i n 60 ml solu t i o n (scaled) of a-D-Glucose a-D-Glucose a f t e r i n degrees^ mole sec"-*- x 10^ x 0.303 g M " Adsorption M sec"-*- x 10 mole sec x 10 0.01 0.006 2.5 3.0 0.09 0.02 0.015 10.3 12.4 0.38 0.03 0.025 15.5 18.6 0.57 0.05 0.044 22.2 26.6 0.81 0.10 0.094 29.8 35.8 1.09 0.20 0.20 33.0 39.6 1.20 0.40 0.40 35.2 42.2 1.28 0.80 0.80 35.0 42.0 1.27 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Concentration of Glucose (M) F i g . 37 Adsorption Isotherm for Reversible Adsorption of Glucose on Alumina Neutral ( l e f t and • ) and Plot of I n i t i a l Rate on Alumina Neutral Versus Concentration of Glucose (right and • ) Amount of Glucose (x 10 moles) Adsorbed per Gram of Q 1 Alumina Neutral, and I n i t i a l Rate (x 0.303 x 10 mole.sec ) on 1.6 g of Catalyst i n 60 ml a-D-Glucose Solution o o I—1 I—' rt co p. rt tu i-t M O 3 181 182 concentrations. At low a-D-glucose concentrations the r e l a t i v e amount of 0-D-glucose formed at t = 25 mins can be appreciable and hence,the reverse reaction also would occur at the same time. Therefore the observed i n i t i a l rate would be slower than what i s expected when [B-D-glucose] =;0. The above r e s u l t s lead to the conclusion that the c a t a l y t i c a c t i v i t y of alumina i s due to the weak adsorption s i t e s and the strong adsorption s i t e s have e i t h e r no or n e g l i g i b l e c a t a l y t i c a c t i v i t y . However,it i s impossible to prove that a l l the weak adsorption s i t e s are c a t a l y t i c a l l y a c t i v e , because the same r e s u l t (superimposable adsorption isotherm and i n i t i a l rate p l o t ) can be produced even when a f r a c t i o n of the weak adsorption s i t e s are c a t a l y t i c a l l y a c t i v e . That i s , two types of s i t e s on the alumina surface with the same values f o r K but only one being c a t a l y t i c a l l y a ctive w i l l also produce the same r e s u l t . It may be argued that the p r o b a b i l i t y of existence of two types of s i t e s having d i f f e r e n t chemical f u n c t i o n a l groups but with the same strength of i n t e r a c t i o n (related to heat of adsorption) with the reactant molecules i s very small. V a l i d i t y of such reasoning can be v e r i f i e d only by determining the a c t i v e s i t e density by other more r e l i a b l e methods (see below). However, i t i s safe to conclude at t h i s point that the number of weak adsorption s i t e s gives the upper l i m i t of the number of a c t i v e s i t e s on the alumina surface. This upper l i m i t f o r the number of c a t a l y t i c s i t e s on alumina -4 19 neutral i s therefore (1.3 ± 0.1)10 mole/gram = (7.8 ± 0.8)10 s i t e s / 13 2 gram = (5.4 ± 0.6)10 sites/cm . XV.1 Comparison with Other Methods of Determining Active Site Density The method used above to determine the number of active s i t e s per gram can be applied to other c a t a l y t i c systems when there i s an experimentally 183 observed r e l a t i o n between the adsorption isotherm and the p l o t of i n i t i a l rate versus substrate concentration. If there i s no observable r e l a t i o n , one might be able to divide the observed adsorption isotherm i n t o component isotherms and compare these with the i n i t i a l rate p l o t . In the case of glucose on alumina t h i s i s possible only because the i n d i v i d u a l isotherms are widely d i f f e r e n t with d i f f e r e n t equilibrium constants. In other cases t h i s method w i l l not be a p p l i c a b l e . The most d i r e c t method f o r determining the number of active s i t e s , 33 125 which was most s u c c e s s f u l l y applied by Kokes ' , involves using the reactant i t s e l f as a probe molecule (just as i n the method described i n t h i s t h e s i s ) . They were able to show that the surface compounds observed by i n f r a r e d spectroscopy are intermediates of the c a t a l y t i c reaction and not only their precursors. The number of intermediates could be determined from the adsorption isotherm and t h e i r formation was followed by i n f r a r e d spectroscopy. This i s an i d e a l method which i s l i m i t e d to systems where there i s an appreciable concentration of surface intermediates formed which can be detected without any interference by other adsorbed species. Another rather d i r e c t approach involves d e t a i l e d analysis of absolute rates for the estimation of the active s i t e density from pre-126 exponential factors . From the t r a n s i t i o n state theory i t has been shown that rate constant of a surface reaction, assuming the rate determining step to be unimolecular change of adsorbed reactant, i s V = A exp (-E/RT) = C (kT/h) exp (AS/i?) exp(-E/RT) (47) where V i s the rate constant i n molecules reacting per unit surface area per second when the surface i s f u l l y covered, C& i s the concentration of 184 c a t a l y t i c a l l y a c t i v e s i t e s , k and h are Boltzmann and Planck constants respec-t i v e l y , and AS i s the entropy of a c t i v a t i o n . I f AS i s known or i f i t i s assumed to be approximately zero, because both reactant and activated complex are adsorbed species, then the s i t e density can be determined when experimental values of V and a c t i v a t i o n energy E are known. The v a l i d i t y of the t r a n s i t i o n state method has been demonstrated by agreement (within an order of magnitude) of the s i t e density obtained using Equation (47) with that obtained by other quite r e l i a b l e means'*^. An error i s introduced when E i s determined from an Arrhenius p l o t since t h i s method neglects the occurrence of T i n the pre-exponential f a c t o r . I f AS i s not known and i s considered to be zero, because both reactant and activated complex are adsorbed species, again an error w i l l be introduced i n the value of C estimated. 33 F i n a l l y there i s a more i n d i r e c t method c a l l e d s p e c i f i c poisoning of a c a t a l y s t surface which i s very commonly used by c a t a l y t i c chemists. A probe molecule other than the reactant molecule i s preadsorbed as a poison and i t s e f f e c t on the c a t a l y t i c a c t i v i t y i s studied. From the number of poisoning molecules necessary to b r i n g the a c t i v i t y to zero, the upper l i m i t of the a c t i v e s i t e density i s determined. It i s assumed that the reaction of the i n h i b i t o r with the surface i s i r r e v e r s i b l e , that i t combines only with c a t a l y t i c a l l y a c t i v e sites,and that there i s one i n h i b i t o r molecule per a c t i v e s i t e . Because of these assumptions the upper l i m i t of a c t i v e s i t e density i s obtained. For example, by poisoning the 32 acid s i t e s on y-alumina by a l k a l i , Pines and Haag showed that the upper l i m i t of the t o t a l number of s i t e s ( a l l capable of dehydrating 1-butanol) and the number of more highly reactive s i t e s (also capable of isomerizing 13 13 2 cyclohexene) are 10 x 10 and 0.8 x 10 sites/cm r e s p e c t i v e l y . Rosynek 185 46 and Hightower have shown by the poisoning e f f e c t of CO^j that the upper l i m i t of s i t e s capable of exchanging hydrogen atoms on o l e f i n s i s 3-8 x 12 2 10 /cm . From the above discussion i t i s c l e a r that the determination of a c t i v e s i t e s by r e l a t i n g the i n i t i a l rate to the adsorption isotherm i s b e t t e r than the poisoning method at determining the upper l i m i t . The adsorption isotherm versus rate method i s able to eliminate s i t e s that are not c a t a l y t i c a l l y a c t i v e but which may s t i l l possess acid and base f u n c t i o n a l groups capable of combining with a reactive i n h i b i t o r molecule. The r e s u l t s discussed i n t h i s section have shown that the weak adsorption s i t e s on alumina neutral possess c a t a l y t i c a c t i v i t y while the strong adsorption s i t e s are e i t h e r c a t a l y t i c a l l y i n a c t i v e or t h e i r a c t i v i t y i s r e l a t i v e l y very low.* The possible reasons for the i n a c t i v i t y or very low a c t i v i t y of strong adsorption s i t e s on alumina neutral are discussed below. XV.2 C a t a l y t i c I n a c t i v i t y of Strong Adsorption Sites The observed rate constant k , (Equation 32) mav be divided into obs two terms, one due to weak adsorption s i t e s and the other due to strong The i n i t i a l rates at several concentrations of glucose were not determined for sintered alumina and hence,it i s not known whether both types of r e v e r s i b l e adsorption s i t e s or only one type i s c a t a l y t i c a l l y a c t i v e . However, the observation that there are two d i s t i n c t types of r e v e r s i b l e adsorption s i t e s on sintered alumina, as w e l l as on alumina neutral, suggests that the method for determining the number of a c t i v e s i t e s on a c a t a l y s t described e a r l i e r may be applicable here. 186 adsorption s i t e s . k m <*3 + = ^ 3 + k j f ' ° o + (*3 + V'V The c a t a l y t i c i n a c t i v i t y of r e v e r s i b l e adsorption s i t e s (whose concentration i s appreciable) can be due to two reasons. E i t h e r the term (k^ + k^) i s zero or very small, or the term i s very large. However,the r e s u l t s i n Section XIV.4.1 showed that the term l/K^ for strong adsorption s i t e s i s smaller than the term 1/K for weak adsorption s i t e s and th i s f actor a c t u a l l y helps to increase the a c t i v i t y . Therefore the reason for the presence of c a t a l y t i c a c t i v i t y on weak adsorption s i t e s and the absence of a c t i v i t y on strong adsorption s i t e s i s that the term (k + k^)W i s r e l a t i v e l y high for weak s i t e s while the term (k^ + k^) — 0 for strong adsorption s i t e s . That i s , the strong adsorption s i t e s are i n a c t i v e because the c a t a l y t i c constant for the surface reaction (k^ + k ) i s very small (probably because the a c i d i c and/or basic groups necessary f o r c a t a l y s i s are either absent or not i n the proper p o s i t i o n f o r c a t a l y s i s to occur) and not because the adsorption i s unfavorable. The c a t a l y t i c constant f o r surface reaction on weak adsorption s i t e s (k + k ~) and other k i n e t i c parameters of the surface reaction are determined i n the next se c t i o n . 187 XVI KINETIC PARAMETERS OF THE CATALYTIC SYSTEM 188 XVI KINETIC PARAMETERS OF THE CATALYTIC SYSTEM The equilibrium constant f o r adsorption on the weak s i t e s on alumina neutral and the upper l i m i t of the c a t a l y t i c s i t e density can be used to determine various k i n e t i c parameters such as the c a t a l y t i c constant f o r the surface reaction i n Equation 26, the turnover number of a c a t a l y t i c s i t e , and the c a t a l y t i c constant for the o v e r a l l reaction. k l k3 k2 S + C SC PC ^ T * * P + C (26) k2 h  ki XVI.1 Determination of the C a t a l y t i c Constant f o r the Surface  Reaction (k^ + k^) (a) From the p l o t of k , versus concentration of c a t a l y s t i n obs F i g . 21 (Section IX) the experimentally obtained slope = 1.14 x 10 ^ sec -1 -, . g l i t r e . From the act i v e s i t e density determination, one gram of c a t a l y s t i n -4 one l i t r e has 1.3 x 10 mole c a t a l y t i c s i t e s . 1.14 x 10 ^ _ _ n o -1 , -1 . . Slope = ^ r — = 0.088 sec mole l i t r e 1.3 x 10 i k 3 + h ) o o Since * , = the slope of the plo t i n F i g . 21, = o b s 1/K+ s o k3 + kA -1 -1 = 0.088 sec mole l i t r e . 1/K + s o From the Langmuir p l o t i n F i g . 35 the equilibrium constant for the 189 catalytic sites K = 46 l i t r e mole Therefore 1/K = 0.022 mole l i t r e 1 and the total substrate concentration s = 0.050 mole l i t r e \ o (fe3 + V -1 -1 = 0.088 sec mole l i t r e 0.072 mole l i t r e - 1 and + fe.) « 6.3 x 10 3 sec" 1 3 4 6 x 10 sec (b) The catalytic constant can also be determined from the slope of the plot of , 1 versus concentration of glucose (s ) given in Fig. 23 /c . o obs (Section X). Again, since _ ° + L fcv ( k „ + k.)o K(k. + k,)on obs 3 4 o 3 4 ° Slope = -(fe, + k.)o 3 4 o -4 where CQ the catalyst concentration = 26.7 mg/ml = 35 x 10 mole catalytic s i t e s / l i t r e . 4 -1 = 7.8 x 10 sec l i t r e mole , (k_ + k,)c (fe. + 7c.)35 x 10"4 3 4 o 3 4 from the plot i n Fig. 23. (k. + fe.) = 3.7 x 10~3 sec" 1 3 4 l n-3 -1 — 4 x 10 sec 190 The two values for (k^ + k ) obtained by the two d i f f e r e n t p l o t s -3 -1 are quite close to each other and gives (k. + k.) = 5 x 10 sec M ° 3 4 average The term (k„ + k.) i s the c a t a l y t i c constant f o r interconversion 3 4 of a- and B-D-glucopyranose on the two dimensional homogeneous medium offered by the alumina surface (Equation 26). Hence i t may be c a l l e d the c a t a l y t i c constant for "homogeneous reacti o n " on alumina surface, i n comparison with the c a t a l y t i c constant f o r homogeneous reaction i n s o l u t i o n . The c a t a l y t i c constant f o r homogeneous c a t a l y s i s by water at 25.0°C i s 4 x -4 -1 10 sec (see Table III) and hence,it follows that t h i s alumina surface o f f e r s a better medium for glucose mutarotation than water. XVI.2 C a t a l y t i c Constant for Heterogeneous C a t a l y s i s of Glucose Mutarotation From Section XVI.1 the observed rate constant when the c a t a l y s t concentration i s 1 mole l i t r e 1 ( i . e . the c a t a l y t i c constant k ) i s 0.088 J cat sec 1 mole l i t r e and may be compared with other c a t a l y t i c constants given i n Table I II for homogeneous c a t a l y s i s of glucose mutarotation. It i s i n t e r e s t i n g to note that i t i s about 100 times more active than a weak base l i k e acetate i n water, about 9 times more active than strong acids i n water, and a l i t t l e more active than 2-hydroxypyridine i n benzene. However, i t i s important to note that the observed high c a t a l y t i c constant f o r alumina i s a r e s u l t not only of the medium ( i . e . k^ + k^) but also of r e l a t i v e l y strong substrate-catalyst binding ( i . e . K^) at low substrate concentrations (concentration of glucose = 0.05 M). In s p i t e of these favorable conditions, alumina neutral i s 2 x 10^ times less a c t i v e than nature's c a t a l y s t , enzyme mutarotase (concentration of glucose = 0.11 M). 191 XVI.3 Turnover Number of a C a t a l y t i c S i t e The turnover number (also c a l l e d c a t a l y t i c center a c t i v i t y ) i s defined as the number of molecules of substrate reacting per second per acti v e s i t e . Using the upper l i m i t of active s i t e density and the i n i t i a l rate when the surface i s completely covered,the minimum value f o r the turnover number can be calculated as follows. I n i t i a l rate when the surface of 1.6 g alumina i s completely covered = 0.42 x 10 ^ moles sec 1 (from Table XIV). = 2.5 x 1 0 1 7 molecules sec 1 .*. Rate on one gram of ca t a l y s t = 1.6 x 10^ molecules sec 1 gram .*. Number of molecules reacting on one s i t e i n one second -3 -1 -1 = 2 x 10 molecules s i t e second It may be compared with turnover numbers for some other surface catalyzed reactions at 25°C, N^^, (calculated using a c t i v a t i o n energies) given i n Table XV. I t i s cl e a r that alumina i s remarkably e f f i c i e n t i n cata l y z i n g the mutarotation reaction compared to most other reactions that have been studied so f a r . The turnover number for o-p conversion i s higher than that for glucose mutarotation, but i t occurs only on activated aluminas. The high c a t a l y t i c a c t i v i t y of standard alumina n e u t r a l , which has not been activated, towards glucose mutarotation should be related to the high s e n s i t i v i t y of the reaction towards weak a c i d i c and basic s i t e s on alumina. This confirms the advantage of using mutarotation reaction as a probe for weak a c i d i c and basic s i t e s on the alumina surface. The r e s u l t s discussed i n the l a s t two sections have shown that only a part (maximum^40%) of the adsorption s i t e s on alumina neutral possess 192 TABLE XV TURNOVER NUMBERS FOR SOME SURFACE CATALYZED REACTIONS 126(a) Reaction Catalyst Turnover Number N^^ molecules s i t e - ! s e c - ! Cyclohexanol dehydration n-Propanol dehydration Ethanol dehydration HC02H decomposition Cyclohexane dehydrogenation tert-Butylbenzene cracking A 1 2 ° 3 A 1 2 ° 3 S i 0 2 - A l 2 0 3 A 1 2 ° 3 P t - A l 2 0 3 tert-Butylbenzene cracking S i 0 2 - A l 2 0 3 > cogelled S i 0 2 - A l 2 0 3 (prepared by d i a l y s i s ) 9 x 10 2 x 10 6 x 10 8 x 10 7 x 10 6 x 10 16 -7 -6 -10 -4 -5 -4 o-p H 2 conversion A 1 2 0 3 70' Turnover number at -196°C 193 c a t a l y t i c a c t i v i t y . I r r e v e r s i b l e adsorption s i t e s cannot give r i s e to c a t a l y t i c a c t i v i t y even i f they contain a c i d i c and basic functional groups, and r e v e r s i b l e adsorption on s i t e s where the a c i d i c and/or basic f u n c t i o n a l groups necessary for c a t a l y s i s are e i t h e r absent or not i n proper p o s i t i o n f o r c a t a l y s i s to occur w i l l not cause mutarotation. Hence les s than 40% of the adsorption s i t e s show r e v e r s i b l e adsorption and also possess appropriate functional groups (a c i d i c and/or basic s i t e s ) which catalyze mutarotation. The nature of adsorption and the nature of acid/base functional groups present at the active s i t e s are investigated i n the next two sections. XVII NATURE OF ADSORPTION ON ACTIVE SITES 195 XVII NATURE OF ADSORPTION ON ACTIVE SITES The adsorption of glucose on alumina should occur by the i n t e r a c t i o n of the polar (hydroxyl) groups on glucose molecule with the polar groups 3~f" 2 —  (Al , 0 , OH, OH ) on the surface. It was observed i n Section VII.2 that excess water added onto alumina neutral had no e f f e c t on the rate constant f o r mutarotation. This shows that the glucose molecule i s undergoing a strong s p e c i f i c adsorption on the alumina surface. To understand the s p e c i f i c i t y of adsorption by c a t a l y t i c s i t e s , t h e i n h i b i t o r y e f f e c t of simple organic molecules with d i f f e r e n t functional groups was studied. (a) Benzene Benzene i s a simple organic molecule with a IT electron cloud ( a weak base) which can i n t e r a c t with electron d e f i c i e n t centres on the surface. Benzene i s known to exchange i t s hydrogen with D^,at 25°C,on activated alumina. However,when a 60 ml s o l u t i o n , 0.05 M i n a-D-glucose and 0.05 M i n benzene,was s t i r r e d with 1.6 g alumina neutral no e f f e c t on the rate was observed. (b) Naphthalene When the above experiment was performed with naphthalene (which has a more reactive ir-electron cloud), instead of benzene, again, no change i n a c t i v i t y was observed. These r e s u l t s i n d i c a t e that glucose molecules on active s i t e s are probably not adsorbed on strong Lewis acid s i t e s . This r e s u l t i s not s u r p r i s i n g since any strong Lewis acid s i t e s on the surface should already be occupied by water molecules present on alumina n e u t r a l . 196 (c) Methanol Since water had no i n h i b i t o r y e f f e c t on the reaction (as mentioned above) the e f f e c t of the simplest alcohol methanol was studied. When a 60 ml so l u t i o n , 0.03 M i n a-D-glucose and 0.15 M i n methanol, was s t i r r e d with 1.6 g alumina neutral no change i n rate was observed. Increasing the [MeOH] concentration of methanol to 0.4 M (so that 13) had very [glucose] l i t t l e e f f e c t on the rate. These experiments show that simple hydroxy compounds cannot compete with glucose f o r i t s active s i t e s even when used i n excess. Hence the e f f e c t of the following polyhydroxy compounds on the rate of mutarotation was studied. (d) Methyl a-D-Glucoside When a 60 ml 0.05 M s o l u t i o n of methyl a-D-glucoside was s t i r r e d with 1.6 g alumina ne u t r a l , about 8.5% adsorption (complete i n 20-30 mins) was observed at 25°C. However, there was no hydrolysis of the glucoside, even when the temperature was raised to 100°C. This suggests that the mutarotation of glucose on alumina does not occur v i a a carbonium ion intermediate. The decrease i n adsorption compared to a-D-glucose (adsorption 14% under s i m i l a r conditions) should at le a s t p a r t l y be due to the presence of the methyl group which a f f e c t s the i n t e r a c t i o n of the surface with one side of the pyranose r i n g . I t may be expected that t h i s polyhydroxy compound should be adsorbed on almost every s i t e a v a i l a b l e for adsorption of glucose. Not s u r p r i s i n g l y , i t was found to be an e f f e c t i v e i n h i b i t o r of glucose mutarotation. For example,a 60 ml so l u t i o n 0.05 M i n a-D-glucose and 0.01 M i n the glucoside decreased the a c t i v i t y (compared with the a c t i v i t y i n the absence of 197 i n h i b i t o r ) of 1.6 g alumina neutral by 17%. When the concentration of the glucoside was increased to 0.05 M the c a t a l y t i c a c t i v i t y dropped to 28% of the i n i t i a l value. (e) i - I n o s i t o l i - I n o s i t o l i s one of the nine isomers of hexahydroxycyclohexane and i s s t r u c t u r a l l y s i m i l a r to D-glucose. Both have the same molecular formula i - I n o s i t o l X a-D-Glucose VIII and comparison of structures VIII and X shows that carbons 1, 2, 3 and 4 of a-D-glucose and 1, 2, 3 and 4 of i - i n o s i t o l have the same configuration. Further, i n o s i t o l has one hydroxyl group more than glucose, and does not possess a methyl group (as i n methyl glucoside) which r e s t r i c t s adsorption. Hence i t i s expected to be a very e f f i c i e n t i n h i b i t o r of glucose mutarotation. In fact,when a 60 ml s o l u t i o n 0.05 M i n a-D-glucose and 0.05 M i n i - i n o s i t o l , was s t i r r e d with 1.6 g alumina neutral the c a t a l y t i c a c t i v i t y was decreased about 67%. I t also decreased the percentage glucose adsorbed on the surface from 14% to 6%. The r e s u l t s discussed above show that there are r e l a t i v e l y s p e c i f i c s i t e s for adsorption of glucose on alumina surfaces. These s i t e s adsorb polyhydroxy compounds strongly, possibly through the cooperative placement of several hydroxyl groups. Simple monohydroxy groups cannot compete for those s i t e s even when used i n excess. 198 A possible intermediate of the mutarotation reaction (as described i n the Introduction) i s the a c y c l i c form of glucose which i s a polyhydroxy aldehyde VII (see Section 1.2). The e f f i c i e n c y of mutarotation w i l l be determined by the s t a b i l i z a t i o n of th i s species (more p a r t i c u l a r l y the t r a n s i t i o n state leading to i t ) by the ac t i v e s i t e s on the alumina surface. To determine whether the c a t a l y t i c s i t e s can i n t e r a c t with an aldehyde the following i n h i b i t o r s were investigated. (f) Hexanal Hexanal i s s i m i l a r to the a c y c l i c intermediate from glucose except f o r the absence of hydroxyl groups. I t has only one polar group unlike the polyhydroxy compounds used before and i t s i n t e r a c t i o n with the ca t a l y s t depends on the presence of s i t e s that can i n t e r a c t with an aldehyde functional group containing e l e c t r o p o s i t i v e carbon and the electron r i c h oxygen. Hence i t i s a probe for s i t e s that can s t a b i l i z e the a c y c l i c form of a-D-glucose. When a 60 ml solution,0.05 M i n a-D-glucose and 0.01 M i n hexanal, was s t i r r e d with 1.6 g alumina neutral a 17% decrease i n a c t i v i t y was observed. When the concentration of hexanal was increased to 0.05 M, c a t a l y t i c a c t i v i t y decreased by 26% of the o r i g i n a l value, and the percentage glucose adsorbed on alumina decreased from 14% to 12%. When the concentration of hexanal was increased further to 0.25 M the c a t a l y t i c a c t i v i t y decreased by 95% and the amount of glucose adsorbed on alumina was 6%. These r e s u l t s show that hexanal i s as e f f e c t i v e as methyl a-D-glucoside i n i n h i b i t i n g glucose mutarotation on alumina and also show that the c a t a l y t i c a l l y a c t i v e s i t e s on alumina can i n t e r a c t with an aldehyde functional group. However, any s p e c i f i c i n t e r a c t i o n of the hydroxyl groups of glucose adsorbed on the alumina surface with the aldehyde fu n c t i o n a l group of hexanal 199 may also contribute to the i n h i b i t o r y e f f e c t of hexanal. (g) DL-Glyceraldehyde Glyceraldehyde i s a dihydroxy aldehyde and should be a more e f f e c t i v e i n h i b i t o r than hexanal since i t has two hydroxyl groups which also can i n t e r a c t with the surface. Further, i t has s t r u c t u r a l s i m i l a r i t i e s to the a c y c l i c intermediate that may be formed during mutarotation. When a 60 ml solution, 0.05 M i n a-D-glucose and 0.05 M i n DL-glycer-aldehyde, was s t i r r e d with 1.6 g alumina neutral the a c t i v i t y wa's about 50% less than that observed i n the absence of the i n h i b i t o r . The percentage glucose adsorbed also decreased from 13% to 6%. Glyceraldehyde i s therefore l i t t l e less e f f e c t i v e than i n o s i t o l as an i n h i b i t o r . A l l the r e s u l t s obtained with the i n h i b i t o r s are tabulated i n Table XVI. Comparison of these r e s u l t s leads to the following conclusions. (i ) Equimolar amounts of i n o s i t o l and glucose can decrease the rate by 67% and the amount of glucose adsorbed by about the same percentage (^60%). Hence i n o s i t o l i s adsorbed on alumina more e f f e c t i v e l y than glucose i t s e l f . This i s not s u r p r i s i n g since i t has one hydroxyl group more than glucose. ( i i ) On the other hand,glyceraldehyde and glucose are able to compete equally w e l l f o r adsorption and active s i t e s of alumina. Again,the percentage decrease i n a c t i v i t y and adsorption are almost equal. This indicates that i n o s i t o l and glyceraldehyde have no preference f o r adsorption on i n a c t i v e over c a t a l y t i c a l l y active s i t e s on the surface. ( i i i ) However,the more i n t e r e s t i n g i n h i b i t o r i s hexanal. I t i s le s s e f f i c i e n t than i n o s i t o l and glyceraldehyde mole to mole, since i t has only one polar group. But, the r e s u l t s i n Table XVI show, that the percentage decrease i n a c t i v i t y i s 200 TABLE XVI EFFECT OF INHIBITORS ON THE CATALYTIC ACTIVITY AND ON THE AMOUNT OF GLUCOSE ADSORBED Inhi b i t o r Benzene Naphthalene Methanol Concentration of I n h i b i t o r M 0.05 0.05 (i) 0.15 ( i i ) 0.40 Methyl a-D-Glucoside (i) 0.01 ( i i ) 0.05 i - I n o s i t o l Hexanal ( i ) ( i i ) ( i i i ) DL-Glyceraldehyde 0.05 0.01 0.05 0.25 0.05 Concentration of Glucose M 0.05 0.05 0.03 0.03 0.05 0.05 0.05 0.05 0.05 0.05 0.05 Percentage I n h i b i t i o n 0 0 ^ 0 17 28 67 17 26 95 50 Percentage Decrease i n Glucose Adsorbed 0 0 0 0 60 5 14 60 54 201 always higher than the percentage decrease i n adsorption. This indicates that hexanal i s a more s e l e c t i v e adsorbent and prefers c a t a l y t i c a l l y a c t i v e s i t e s to other adsorption s i t e s . Hence i t may be concluded that c a t a l y t i c a l l y a c t i v e s i t e s are s p e c i f i c f o r adsorption of polyhydroxy compounds and can, i n addition, i n t e r a c t with an aldehyde group better than the other adsorption s i t e s . XVII.1 E f f e c t on C a t a l y t i c A c t i v i t y of Pretreatment of Alumina with  Inhi b i t o r s I t was observed during the study of the curvature i n the f i r s t order plo t s (Section VIII), that pretreatment of alumina with glucose deactivated the c a t a l y s t and the deactivated c a t a l y s t then gave r i s e to l i n e a r f i r s t order p l o t s . The r e s u l t s showed that the curvature arose from the progressive deactivation of the c a t a l y s t by glucose. I t was concluded that deactivation was caused by i r r e v e r s i b l e adsorption of glucose, for example by ether formation, on the a c t i v e s i t e s . The i n h i b i t o r s mentioned above, with d i f f e r e n t f u n c t i o n a l groups which adsorb on the active s i t e s , can be used i n a study of the fu n c t i o n a l groups involved i n the deactivation process. Experiments s i m i l a r to those used i n the study of curvature i n f i r s t order p l o t s were c a r r i e d out by p r e t r e a t i n g alumina neutral with i n h i b i t o r s instead of glucose. (a) Methyl a-D-Glucoside As mentioned above,methyl cx-D-glucoside i s not hydrolyzed by alumina and hence,it has only hydroxyl groups that can combine with functional groups on the surface. F i f t y ml DMSO containing 0.582 g glucoside was s t i r r e d with 1.6 g alumina neutral. A f t e r s i x hours a 10 ml so l u t i o n of 202 0.54 g a-D-glucose i n DMSO was added (so that the r e s u l t i n g 60 ml s o l u t i o n i s 0.05 M i n a-D-glucose and 0.05 M i n methyl a-D-glucoside) and the change i n o p t i c a l r o t a t i o n with time was followed. The r e s u l t s given i n F i g . 39 show that pretreatment with methyl glucoside for s i x hours has deactivated the c a t a l y s t by 35% (compared with about 50% by glucose) and the f i r s t order p l o t obtained i s l i n e a r . Pretreatment for 24 hours did not deactivate the c a t a l y s t any further. (Note that with glucose the deactivation increased with pretreatment time although at a very slow rate F i g . 18). These r e s u l t s show that methyl glucoside i s able to deactivate active s i t e s which r a p i d l y get deactivated during mutarotation but i t i s much less e f f e c t i v e than glucose i t s e l f . (b) Hexanal The same experiments were performed with hexanal. I t was found to be less e f f e c t i v e than even methyl glucoside i n deactivating the c a t a l y s t . There was about 26% deactivation a f t e r pretreatment for 6 hours and about 33% a f t e r 24 hours pretreatment. However, the f i r s t order p l o t s s t i l l showed curvature (Fig. 40) i n d i c a t i n g that at l e a s t some of the r a p i d l y deactivating s i t e s have not been deactivated by hexanal. These r e s u l t s show that deactivation of the c a t a l y s t can be caused by compounds with hydroxyl groups as w e l l as with an aldehyde group. However, hydroxyl groups most e f f e c t i v e l y cause the i n i t i a l rapid deactivation observed with glucose and any further deactivation due to hydroxyl groups i s n e g l i g i b l e . (Similar behavior was shown by methanol. When 1.6 g alumina dried at 150°C was pretreated d i r e c t l y with 1.1 ml (0.03 mole) methanol for 150 mins, a c a t a l y s t with a c t i v i t y the same as the i n i t i a l a c t i v i t y of the o r i g i n a l c a t a l y s t was obtained. However, the methanol treated c a t a l y s t did 203 a i a o c 100 200 300 Time (min) F i g . 39 The E f f e c t of Pretreatment of Alumina Neutral with Methyl a-D-Glucoside on I t s C a t a l y t i c A c t i v i t y 204 5.2, s i s o rH c [a-D-Glucose] = 0.05 M [Alumina] = 26.7 mg.ml -1 • Catalytic a c t i v i t y in the absence of hexanal O Catalytic a c t i v i t y i n the presence of hexanal (0.05 M) Catalytic a c t i v i t y after pretreat-ment of alumina neutral with hexanal for 6 hrs Catalytic a c t i v i t y after pretreat-ment of alumina neutral with hexanal for 24 hrs 3.7 3.2 2.7! 100 200 300 Time (min) Fig. 40 The Effect of Pretreatment of Alumina Neutral with Hexanal on I t s Catalytic A c t i v i t y 205 not get progressively deactivated during a run). Some further deactivation possibly may be caused by aldehyde fu n c t i o n a l groups formed during the mutarotation reacti o n . In summary, the r e s u l t s discussed i n t h i s section have shown that the c a t a l y t i c a l l y a c t i v e s i t e s are s p e c i f i c for adsorption of polyhydroxy compounds. About 35% of the a c t i v i t y can be l o s t by progressive deactivation of the cat a l y s t by reaction of hydroxyl groups with surface f u n c t i o n a l groups. The active s i t e s can also i n t e r a c t with aldehyde groups and aldehyde groups may be involved i n reactions that cause progressive c a t a l y s t deactivation at a much slower rate. Hence,the adsorption and s t a b i l i z a t i o n of f u n c t i o n a l groups which are e s s e n t i a l f o r surface c a t a l y s i s of mutarotation also lead to side reactions which cause permanent deactivation of some of the a c t i v e s i t e s of alumina n e u t r a l . These c a t a l y t i c s i t e s , i n addition to possessing adsorption s i t e s for glucose and any intermediate formed, should also possess acid/base functional groups to catalyze mutarotation. XVIII NATURE OF ACID/BASE FUNCTIONAL GROUPS ON ACTIVE SITES 207 XVIII NATURE OF ACID/BASE FUNCTIONAL GROUPS ON ACTIVE SITES It was mentioned i n the Introduction that the alumina surface contains 3+ 6+ • _ 6-many a c i d i c and basic f u n c t i o n a l groups ( v i z . A l , O-H, 0 , HO and the defect s i t e s ) which are a l l p o t e n t i a l c a t a l y s t s for glucose mutarotation. I t was also mentioned that the use of acid and base s e n s i t i v e substances as i n h i b i t o r s has given evidence for the p a r t i c i p a t i o n of these s i t e s i n c a t a l y t i c reactions. The upper l i m i t of a c t i v e s i t e density had also been obtained using those i n h i b i t o r s (see also Section XV.1). To study the nature of s i t e s on alumina involved i n glucose mutarotation the same i n h i b i t o r s were used as described below. XVIII.1 Test for Basic Sites - I n h i b i t i o n by Carbon Dioxide As mentioned i n the Introduction, CO^ reacts with basic f u n c t i o n a l groups (0 and OH) on the alumina surface. To test f o r the e f f e c t of CO^ on alumina n e u t r a l , i t was f i r s t evacuated at room temperature (for 2 days) to remove excess water adsorbed on the surface. Dry C0 2 was passed through a portion of the evacuated sample packed i n a glass column for 24 hours. The co n t r o l sample was prepared by passing dry nitrogen through another portion of the evacuated sample for 24 hours. The a c t i v i t i e s were determined by s t i r r i n g 1.6 g of each sample with 60 ml 0.05 M a-D-glucose s o l u t i o n . The r e s u l t s showed that alumina neutral has been deactivated 10% by treatment with C0 2 (Table XVII). The procedure was repeated with aluminas pyrolyzed at 800°C and at 1250°C. However,with these samples the. controls were not treated with dry nitrogen since they are already free of adsorbed water molecules. The r e s u l t s showed that alumina pyrolyzed at 800°C has been deactivated 27% and 208 alumina pyrolyzed at 1250°C has been deactivated 85% by CO (Table XVII), TABLE XVII EFFECT OF C0 2 ON DIFFERENT ALUMINAS Alumina Percentage Deactivation Alumina Neutral 10 ± 5 Pyrolyzed at 800°C 27 ± 7 Pyrolyzed at 1250°C 85 ± 5 Errors were estimated from maximum and minimum slopes The p a r t i a l i n h i b i t i o n of c a t a l y t i c a c t i v i t y of dehydrated aluminas by C0 2 should be due to the formation of carbonate ions (which are much less a c t i v e than oxide ions) on the surface. Hence,it should be possible to decompose the carbonate ions,by heating the CO^ treated sintered alumina to a low temperature, and get back the o r i g i n a l a c t i v i t y . I t was done by heating the C0 2 treated sintered sample, at 650°C f o r 3 hrs, which i s a "mild" heat treatment compared with the o r i g i n a l dehydration at 1250°C for 6 hrs. The r e s u l t s i n F i g . 41 show that the deactivated sample has regained i t s o r i g i n a l a c t i v i t y a f t e r "mild" heat treatment. However, attempts to detect by micro-analysis an increase of carbon on sintered alumina a f t e r i t has been treated with C0 0, were not successful probably 209 5.3r [a-D-Glucose] = 0.05 M [Alumina] = 26.7 mg.ml" 4 .8' 4.3 8 I O 3.a 3.31 O C a t a l y t i c a c t i v i t y of alumina sintered at 1250°C C a t a l y t i c a c t i v i t y of alumina sintered at 1250°C and treated with carbon dioxide C a t a l y t i c a c t i v i t y of alumina sintered at 1250°C, treated with carbon dioxide, and heated at 650°C for 3 hrs. 2.81 25 50 75 Time (min) 100 125 F i g . 41 The E f f e c t of Carbon Dioxide on the C a t a l y t i c A c t i v i t y of Alumina Sintered at 1250°C for 6 hours 210 because of the low s e n s i t i v i t y of the technique. The r e s u l t s of i n h i b i t i o n by C0 2 given i n Table XVII show that the f r a c t i o n of a c t i v i t y due to basic s i t e s has increased with increase of dehydration of the alumina surface. According to the model of the alumina surface discussed i n the Introduction,basic hydroxide ions are removed and oxide ions are formed as the surface i s dehydrated. This suggests that the basic c a t a l y t i c s i t e s on the alumina surface should be oxide ions, whether they are i n d i v i d u a l oxide ions or defect s i t e s cannot be determined without further' t e s t s . From the deactivating e f f e c t s of carbon dioxide i t i s also c l e a r that the high a c t i v i t y per unit area for sintered alumina (reported i n Section XII) i s because the active s i t e s are predominantly basic s i t e s . The observed increase i n a c t i v i t y per unit area as alumina was pyrolyzed at temperatures >600°C,appears to be due to a large increase i n the number of b a s i c s i t e s . Since excess CO^ has not been able to deactivate any of the aluminas completely, i t appears that there are c a t a l y t i c s i t e s on alumina which are not poisoned by CO^. Since the s i z e of the C0 2 molecule i s smaller than that of the glucopyranose molecule, the i n h i b i t o r (CO,,) should be able to reach a l l active s i t e s accessible to the substrate (glucose). Hence the s i t e s which are not poisoned by CO^ may be a c i d i c s i t e s or, l e s s l i k e l y , they are weak basic s i t e s which are too weak to react with C0 2-XVIII.2 Tests for A c i d i c Sites on Alumina Neutral XVIII.2.1 E f f e c t of Pyridine As discussed i n the Introduction, pyridine reacts with Lewis acid 211 s i t e s and r e l a t i v e l y strong BrBnsted acid s i t e s . Also,the s i z e of the pyridine molecule i s comparable to that of the glucopyranose molecule and hence, a l l the ac t i v e s i t e s should be s t e r i c a l l y a c cessible to pyridine. Tests of possible i n h i b i t i o n by pyridine was c a r r i e d out by s t i r r i n g 1.6 g alumina neutral with 10 ml DMSO containing 0.04 ml (0.5 mM) pyrid i n e . A f t e r 15 mins,a s o l u t i o n of 0.54 g a-D-glucose i n 50 ml DMSO was added / ..I. ^ x . . r moles glucose 6 . , . _ _. (so that the ratxo of —:: 2 r-r. = — ) and the mutarotation was moles pyridine 1 followed as usual. The r e s u l t s showed that the i n h i b i t o r y e f f e c t of pyridine was n e g l i g i b l e (^2-3%). The s l i g h t v a r i a t i o n i n rate was within experi-mental error. Homogeneous c a t a l y s i s by pyridine not adsorbed on the alumina was also shown to be absent. The procedure was repeated with 0.24 ml (3 mM) of pyridine so that the r a t i o m o j - e s glucose _ ^ Again,negligible (3-4%) i n h i b i t o r y moles pyridine ' ef f e c t was observed. Further, there was almost no change i n the amount of glucose adsorbed and homogeneous c a t a l y s i s was, again, n e g l i g i b l e . Since pyr i d i n e reacts with Lewis acid s i t e s and r e l a t i v e l y strong Bronsted acid s i t e s , such s i t e s cannot be involved i n the c a t a l y t i c mutarotation of aluminas. In order to study the e f f e c t of d i r e c t treatment of alumina by a p o t e n t i a l i n h i b i t o r , 0.24 ml (3 mM) pyridine was added onto 1.6 g alumina, the f l a s k was sealed and the sample was mixed f o r 15 mins. The i n h i b i t o r y e f f e c t of d i r e c t treatment was again found to be neg l i g i b l e (s:4%) and there was almost no change i n the amount of glucose adsorbed. Since a l l the above experiments showed n e g l i g i b l e e f f e c t on the c a t a l y t i c a c t i v i t y , the e f f e c t of prolonged treatment with pyridine was investigated by s t i r r i n g 1.6 g alumina neutral with 10 ml DMSO containing 0.24 ml pyridine f o r 24 hrs. In t h i s case only about a 10% decrease i n a c t i v i t y and about a 5% decrease i n glucose adsorption was observed (Note 212 that about 90% of a c t i v i t y of alumina neutral i s due to a c i d i c s i t e s ) . These r e s u l t s show that Lewis acid s i t e s on alumina n e u t r a l are not c a t a l y t i c a l l y active and any c a t a l y t i c a l l y active Bronsted acid s i t e s are too weak to react with p y r i d i n e . Therefore,the i n h i b i t o r y e f f e c t of a s t i l l stronger base was investigated next. XVIII.2.2 E f f e c t of n-Butylamine As mentioned i n the Introduction, n-butylamine i s a stronger base than pyridine and should be s t e r i c a l l y able to reach a l l the s i t e s on an alumina surface which are active towards glucose mutarotation. Tests of the i n h i b i t i o n by n-butylamine was ca r r i e d out by s t i r r i n g -4 1.6 g alumina neutral with 10 ml DMSO containing 0.03 ml (3 x 10 mole) n-butylamine. A f t e r 150 mins, 50 ml DMSO containing 0.54 g a-D-glucose f . moles glucose 10 v J J J j i • • j (so that ; °—.-——— : = — ) was added and k i n e t i c s was followed. moles n-butylamme 1 The samples removed from the reaction f l a s k showed a rapid decrease i n o p t i c a l r o t a t i o n with time, while the f i l t e r e d s l u r r i e s also continued to show a slow decrease i n o p t i c a l r o t a t i o n ( i . e . i n the absence of alumina). Hence there was concurrent homogeneous and heterogeneous c a t a l y s i s , as i s shown i n the f i r s t order plots given i n F i g . 42. The pl o t (a) was obtained by measuring the o p t i c a l rotations of samples as soon as they were f i l t e r e d and hence,is the r e s u l t of simultaneous homogeneous and heterogeneous -4 -1 c a t a l y s i s . The slope at t = 25 mins equal to 3.83 x 10 sec , i s the o v e r a l l (homogeneous + heterogeneous) rate constant. The f i r s t order pl o t s (b) and (c) for homogeneous c a t a l y s i s were obtained by following the change i n o p t i c a l r o t a t i o n of f i l t e r e d s l u r r i e s with time, and are l i n e a r and p a r a l l e l as expected. From plo t s (b) and (c) the average rate constant for -4 -1 homogeneous c a t a l y s i s by the butylamine i n so l u t i o n i s 0.28 x 10 sec 213 60 ml 0.05 M a-D-Glucose 0 25 50 75 100 125 150 Time (min) F i g . 42 The E f f e c t of n-Butylamine on the C a t a l y t i c A c t i v i t y of Alumina Neutral 214 -4 -1 Therefore the rate constant f o r heterogeneous c a t a l y s i s i s 3.55 x 10 sec -4 -1 ^3.6 x 10 sec . Under s i m i l a r conditions,but i n the absence of any -4 -1 additive,1.6 g alumina showed an i n i t i a l rate constant of 2.5 x 10 sec Hence,n-butylamine has increased the c a t a l y t i c a c t i v i t y of alumina neutral by about 44%. It was mentioned i n the Introduction that n-butylamine reacts with Lewis acid s i t e s and also Brbnsted acid s i t e s . I t was shown above by the e f f e c t of pyridine that Lewis acid s i t e s are not active i n glucose muta-r o t a t i o n . Further, the reaction of n-butylamine with a Lewis acid s i t e w i l l produce an i n a c t i v e s i t e and hence should decrease the c a t a l y t i c a c t i v i t y . Hence,the increase i n a c t i v i t y by n-butylamine should a r i s e from i t s reaction with Bronsted acid s i t e s and the c a t a l y t i c a c t i v i t y of the r e s u l t i n g s i t e . The reaction of n-butylamine with a Bronsted acid s i t e w i l l produce a more acti v e b asic s i t e and therefore, as observed, there should be an increase i n the c a t a l y t i c a c t i v i t y . H A l A l / | \ \ / / | \ / | \ + NH, CH, CH, CH, CH, A l A l / | \ \ / / | \ / | \ (+) NH, I ' CH 2 CH, I 1 CH, I ' CH, A c i d i c (Type III) hydroxyl group Hence, t h i s increase i n a c t i v i t y on treatment with n-butylamine shows that there are c a t a l y t i c a l l y a c t i v e Bronsted acid s i t e s which are too weak 215 to react with pyridine. They are, however, reactive with n-butylamine and become more c a t a l y t i c a l l y a c t i v e . I t was mentioned i n the Introduction (page 20) that Pearson using a s e n s i t i v e n.m.r. method showed the presence of protonated species when pyridine was adsorbed on alumina at 0°C. Further, as mentioned i n the Introduction (page 35), the c a t a l y t i c a c t i v i t y of Bronsted acid s i t e s on alumina had been detected only at r e l a t i v e l y high temperatures (^200°C). With use of glucose i t has been possible to detect, at 25°C, the c a t a l y t i c a c t i v i t y of Bronsted acid s i t e s that are too weak to react with pyridine, and hence not detectable by the n.m.r. method of Pearson. This appears to be the only example of a reaction catalyzed by weak Bronsted acid s i t e s on alumina at 25°C. Therefore i t follows that glucose mutarotation i s more s e n s i t i v e to Bronsted acid s i t e s than any of the other reactions that have been studied on alumina so f a r . Further, glucose mutarotation i s a more s e n s i t i v e probe for weak Bronsted acid s i t e s on alumina (or any other s o l i d c a t a l y s t ) than any of the spectroscopic methods used so far. It i s very l i k e l y that the weak Bronsted acid s i t e s on alumina neutral are involved i n the 90% of i t s a c t i v i t y which i s not affected by CO^- I t was observed i n Section XVI.2 that the c a t a l y t i c constant f o r alumina neutral i s about 10 times greater than that for strong Bronsted acids i n water or DMSO ( i . e . homogeneous systems). Generally, the c a t a l y t i c a c t i v i t y of weak acids ( l i k e a c e t i c acid) i n water i s only about one hundredth that f o r strong Bronsted acids. Hence,it seems e s p e c i a l l y notable, that the a c t i v i t y of weak Bronsted acid s i t e s on the alumina surface i s greater than that of even strong Bronsted acids i n water. The high c a t a l y t i c a c t i v i t y of weak s i t e s when present on the alumina surface may be due to two f a c t o r s , (i ) Strong s p e c i f i c adsorption of glucose molecules on the alumina surface with the a c i d i c group i n the correct o r i e n t a t i o n to protonate the r i n g 216 oxygen of glucose and/or ( i i ) S t a b i l i z a t i o n of the t r a n s i t i o n state leading to the a c y c l i c form by the fun c t i o n a l groups on the alumina surface. It was observed e a r l i e r that a c t i v e s i t e s on alumina can s t a b i l i z e an aldehyde fu n c t i o n a l group while, as mentioned i n the Introduction, there appears to be very l i t t l e i n t e r a c t i o n (e.g. no oxygen atom exchange) between solvent (water) and the a c y c l i c form of glucose during mutarotation. In these respects the alumina surface resembles enzymes i n i t s c a t a l y t i c properties, whatever the reason for i t may be, i t may be concluded that the alumina surface o f f e r s a better medium f o r glucose mutarotation than pure water. XVIII.2.2.1 Determination of the Amount of n-Butylamine on the Surface It was observed i n F i g . 42 that there was also homogeneous c a t a l y s i s of glucose mutarotation i n the heterogeneous system when alumina was treated with n-butylamine. The slope of the f i r s t order pl o t f o r homogeneous c a t a l y s i s should be proportional to the concentration of n-butylamine i n s o l u t i o n since c a t a l y s i s by solvent i s n e g l i g i b l e . A comparison of the rate constant for a homogeneous reaction catalyzed by n-butylamine (without previous addition of alumina) with the rate constant f o r a homogeneous component of a reaction system with alumina present would give a measure of the n-butylamine concentration i n s o l u t i o n . This i n turn gives the n-butylamine adsorbed on the surface of alumina. The rate constant f o r homogeneous c a t a l y s i s by n-butylamine, i n the absence of alumina, was determined using a 60 ml s o l u t i o n 0.05 M i n a-D-glucose and 0.005 M i n n-butylamine (Fig. 42 plo t (d)). I t was observed that f i l t r a t i o n of the s o l u t i o n had no e f f e c t on the rate of c a t a l y s i s . This shows that the rate constant f o r homogeneous c a t a l y s i s determined 217 (from p l o t s (b) and (c)) using f i l t e r e d s l u r r i e s i s equal to the rate constant f o r homogeneous c a t a l y s i s i n the s l u r r y . The r a t i o , rate constant f o r homogeneous c a t a l y s i s i n the presence of alumina _ rate constant for homogeneous c a t a l y s i s i n the absence of alumina -4 -1 0.28 x 10 sec 1 c ., 1 c ,^ , fc .. , T _ 1 = . Therefore, only of the n-butylamine added 0.93 x 10 sec 3.3 3 - 3 2.3 -4 -4 i s i n s o l u t i o n , and x 3 x l 0 = 2 . 2 x 1 0 mole n-butylamine has reacted with the surface. This may be used to determine the number of acid 2 sites/cm on the surface of alumina neutral as shown below. The amount of n-butylamine that has reacted with 1.6 g alumina neutral -4 19 equals 2.2 x 10 mole = 13 x 10 molecules. Therefore the number of molecules that has reacted with a unit area ,, 2. . , , - 13 x 1 0 1 9 molecules r 0 ,^13 , . -2 (1 cm ) of the sample equals r j = 5.8 x 10 molecules cm 140 x 1.6 x 10 cm and may be considered as the active s i t e density on alumina neutral (Note that only about 10% of the a c t i v i t y of alumina neutral i s due to basic s i t e s and s i n c e , i n general,the a c t i v i t y of a basic s i t e i s several orders of magnitude greater than that of an acid s i t e , the number of acid s i t e s on alumina neutral ^ the number of a c t i v e s i t e s ) . However, i t i s not possible to state whether i t gives the upper l i m i t of c a t a l y t i c s i t e density since there may be c a t a l y t i c a l l y active Bronsted acid s i t e s which are too weak to react with n-butylamine. However, i t i s important to note that the s i t e density determined using n-butylamine equals, within experimental error, the s i t e density determined using adsorption isotherms and the i n i t i a l 13 -2 rate data ( = 5.4 x 10 s i t e s cm ) i n Section XV. XVIII.2.3 Tetramethylammonium Hydroxide Since there may be c a t a l y t i c a l l y active Bronsted acid s i t e s which are too weak to react with n-butylamine, the e f f e c t of a s t i l l stronger base, N(Me) OH, was investigated. Solutions of the hydroxide i n DMSO were 218 prepared and standardized by t i t r a t i n g against HCl as described i n the Experimental Section. + The e f f e c t of N(Me)^ OH was determined by t r e a t i n g alumina neutral with 10 ml of DMSO containing a known amount of the base. Af t e r 150 mins, 50 ml DMSO containing 0.54 g a-D-glucose was added and the rate of muta-rotation was followed. The experiment was repeated with increasing amounts of the base. For i n i t i a l experiments 1.6 g of the ca t a l y s t was used and the rate constant for heterogeneous c a t a l y s i s increased r a p i d l y with increase i n the concentration of base. Therefore, as the amount of base was increased the weight of the c a t a l y s t used was progressively decreased (e.g. 0.32 g, 0.08 g) so that the heterogeneous c a t a l y s i s could be followed conveniently and the presence of any homogeneous c a t a l y s i s could also be detected. The r e p r o d u c i b i l i t y of the r e s u l t s was good and independent of the pretreatment time at r e l a t i v e l y high base concentrations. With increasing base the homogeneous c a t a l y s i s was f i r s t detected ( in the f i l t e r e d s l u r r y ) when 0.08 g alumina neutral was treated with 2.0 x 10 ^  mole base (Fig. 43, p l o t s (b) and (c)). No homogeneous c a t a l y s i s was observed when 0.08 g alumina was treated with 1.5 x 10 "* mole base and much more rapid homogeneous c a t a l y s i s was observed when the same weight of the cata l y s t was treated with 2.5 x 10 ^  mole base. The rate constant f o r homogeneous c a t a l y s i s by 0.5 x 10 "* mole base i n the absence of alumina was determined using 60 ml -5 + sol u t i o n containing 0.54 g a-D-glucose and 0.5 x 10 mole N(Me)^ OH (Fig. 43, plo t (d)). The r e s u l t s were analyzed, as i n the case of n-butyl-amine treatment, and the amount of base that reacted with 0.08 g alumina neutral was determined to be ~ 2.0 x 10 ~* mole. The amount of base that reacts with the surface was also determined by back t i t r a t i o n s as described i n the Experimental Section. The amount 219 F i g . 43 The E f f e c t of Tetramethylammonium Hydroxide on the C a t a l y t i c A c t i v i t y of Alumina Neutral 220 -4 of base that reacts with 1.6 g alumina neutral was found to be 5.0 x 10 mole and i t was independent of the contact time and also the amount of excess base used. Therefore, the average amount of base that reacts with the surface of -4 1.6 g alumina neutral equals (4.5 ± 0.5) 10 mole and may be used to determine the upper l i m i t of a c i d i c a c t i v e s i t e density (csactive s i t e density) on the surface. The upper l i m i t of active s i t e density of alumina neutral using the data from S(Me)^ OH treatment turns out to be (1.2 ± 0.1) 14 2 10 sites/cm , which i s about double the upper l i m i t obtained using n-butyl-amine or adsorption isotherms and i n i t i a l rate data. For comparison, the 23 15 reported t o t a l l a t t i c e s i t e density on alumina surface range from 0.9 x 10 15 -2 to 1.5 x 10 s i t e s cm . However,it i s not s u r p r i s i n g that the treatment of alumina with OH ions has given a higher value for the number of acid s i t e s , since i t should react with a l l strong and weak, Bronsted and Lewis acid s i t e s on the surface whether they are c a t a l y t i c a l l y a c t i v e or not. I t i s i n t e r e s t i n g to note that the c a t a l y t i c a c t i v i t y of alumina ne u t r a l has increased by a fa c t o r of 10^ on treatment with S(Me)^ OH . Note that h a l f as much n-butylamine reacted with the surface and increased the c a t a l y t i c a c t i v i t y only 44%. This d i f f e r e n c e i n the increase i n a c t i v i t y can be due to many fa c t o r s . For example, OH ions should have converted a l l weak Brbnsted acid s i t e s (which are too weak to react with n-butylamine and too weak to catalyze mutarotation) to strong conjugate b a s i c s i t e s with very high a c t i v i t y . Thus, new act i v e s i t e s with very high a c t i v i t y could have been formed. Further, the two types of ion p a i r s o formed on the surface by the two bases, v i z . ^  ^ ( M e ) ^ and^O^$H 3-(CH 2) 3"CH 3 could have d i f f e r e n t a c t i v i t i e s towards mutarotation due to the differences i n the accompanying cations. For example, s t e r i c e f f e c t s of the four carbon 221 chain on n-butylamine and the decrease i n the b a s i c i t y of the anion by hydrogen bonding to ^ ^ - j - c a n decrease the c a t a l y t i c a c t i v i t y of the ion pai r s formed from n-butylamine. XVIII.2.A Deactivation of Bronsted Acid S i t e s - E f f e c t of CH N 2 During the studies described above on the e f f e c t of bases on c a t a l y t i c a c t i v i t y , evidence has been gathered for the presence of c a t a l y t i c a c t i v i t y due to weak Bronsted acid s i t e s . Since the re a c t i o n of a Bronsted acid s i t e with a base produces the conjugate base with higher c a t a l y t i c a c t i v i t y towards mutarotation, the c a t a l y s t produced by treatment with a r e l a t i v e l y strong base always had a higher a c t i v i t y than the o r i g i n a l c a t a l y s t . On the other hand, to deactivate the Brbnsted acid s i t e s , (without converting them to basic s i t e s ) methylation by t r e a t i n g alumina with diazomethane i n ether so l u t i o n was attempted. However, the diazomethane decomposed on the alumina surface and s u r p r i s i n g l y producing a c a t a l y s t with s l i g h t l y higher a c t i v i t y than the c o n t r o l samples. This may be due to breaking up of the surface by diazomethane decomposition; however, t h i s alumina sample was not investigated further. XVIII.3 Deactivation of Bronsted Acid and Basic Sites - E f f e c t of ( C H 3 ) 2 S i C l 2 Dichloromethylsilane should react with both hydroxyl groups and the basic s i t e s on the alumina surface producing a new surface without any 132 a c i d i c or basic groups (Equations 48, 49, 50) S i l y l a t i o n of the surface of alumina neutral was c a r r i e d out as described i n the Experimental Section. The s i l y l a t e d aluminum oxide had no c a t a l y t i c a c t i v i t y but adsorbed about the same amount of glucose over several hours. C H ^ / C H 3 / S 1 \ C H 3 / C H 3 + / \ OH OH 0 0 CH 3 / C H 3 CH 3 / C H 3 S i S S i 777kr/, Cl c i • Cl \ + 0 OH CH3OH C H 3 / C H 3 ^ S i CH 0 \ 3 0 777/777, CH 3 / C H 3 C l / S - C l C \ / H 3 CH30H CH 3 CH 3 X S i I ^ OCH 0 Cl 223 In summary, the s p e c i f i c poisoning of the ca t a l y s t surface has given evidence for the presence of basic (probably oxide ions) and weak Bronsted acid s i t e s on the surface. I t was mentioned i n the Introduction that acids and bases show d i f f e r e n t deuterium isotope e f f e c t s f o r mutarotation i n s o l u t i o n . Further, the study of deuterium isotope e f f e c t has given information on the mechanism (for example, s p e c i f i c or general acid/base c a t a l y s i s , b i f u n c t i o n a l c a t a l y s i s ) i n s o l u t i o n . Therefore, deuterium isotope e f f e c t for glucose mutarotation by alumina was investigated i n the next section. DEUTERIUM ISOTOPE EFFECT ON GLUCOSE MUTAROTATION BY ALUMINA 225 XIX DEUTERIUM ISOTOPE EFFECT ON GLUCOSE MUTAROTATION BY ALUMINA XIX.l Deuterium Isotope E f f e c t for Catal y s i s by 800°C Alumina Alumina dehydrated at 800°C was chosen f o r de t a i l e d studies of glucose mutarotation because i t has advantages over other aluminas. For example, i n comparison to alumina sintered at 1250°C, i t gives be t t e r r e p r o d u c i b i l i t y even i n the presence of water (see Sections XII and XIII.3). Relative to alumina neutral i t also has less hydroxyl groups on the surface and hence,it i s easy to swamp out the hydroxyl protons with D^O. The deuterium isotope e f f e c t for c a t a l y s i s by 800° alumina was studied both i n the presence of water (or D^O) and under anhydrous conditions as described below. XIX.1.1 Deuterium Isotope E f f e c t under Anhydrous Conditions Deuterated 800° alumina and a control sample of 800° alumina were prepared as described i n the Experimental Section. Sixty ml 0.05 M sol u t i o n of glucose-O-D ( i . e . O-deuterated a-D-glucose) i n DMSO was s t i r r e d with 0.8 g of deuterated 800° alumina and the k i n e t i c s was followed as usual. However, the f i r s t measurement of o p t i c a l r o t a t i o n was made only a f t e r 25 mins to allow s u f f i c i e n t time for lumpy p a r t i c l e s of alumina (see Experimental Section) i n the sample to break down. The k i n e t i c s was also followed f o r mutarotation of 60 ml 0.05 M so l u t i o n of glucose-O-H (standard a-D-glucose) i n DMSO by 0.8 g of the control sample of 800° alumina. The two f i r s t order p l o t s are shown i n F i g . 44, and i t i s clear that O-deuteration has decreased the observed rate constant. The deuterium isotope e f f e c t , from the slopes between t = 30 and 40 mins, i s % / 7 = 1.4 ± 0.1. 226 [a-D-Glucose] = 0.05 M [Alumina] = 13.3 mg.ml 0 50 100 150 200 Time (min) F i g . 44 Deuterium Isotope E f f e c t with 800°C Alumina Under Anhydrous Conditions 227 XIX.1.2 Deuterium Isotope E f f e c t i n the Presence of Water/Deuterium Oxide To study the deuterium isotope e f f e c t i n the presence of water, hydrated 800° alumina was made by adding 7% water to a sample of 800° alumina and 7% D^O was added to another sample of 800° alumina (Note that there was >6% loss i n weight on heating alumina neutral to 800°C). The samples were then mixed w e l l i n a m i n i m i l l . To swamp out the r a p i d l y exchangeable hydroxyl protons on glucose i t was decided to add 10 times as many deuterons (as added D^O). Since 60 ml 0.05 M glucose s o l u t i o n contains 0.003 mole glucose, there are 0.015 mole of exchangeable protons i n the s o l u t i o n . There-fore 0.15 mole deuterons are required and they are present i n 1.35 ml D^O. To 60 ml 0.05 M glucose s o l u t i o n i n DMSO, 1.35 ml D 20 was added and was s t i r r e d with 0.8 g of 800° alumina treated with D 20 (as described above). The k i n e t i c s was followed as usual and the f i r s t order p l o t i s given i n F i g . 45. The control run was c a r r i e d out by adding 1.35 ml water to 60 ml 0.05 M glucose s o l u t i o n and s t i r r i n g with 0.8 g of 800° alumina treated with water. In both the runs no homogeneous c a t a l y s i s was detected i n the f i l t e r e d s l u r r i e s (Note that the mole f r a c t i o n water = 0.09 and homogeneous c a t a l y s i s 6 5 due to water i s observed only when mole f r a c t i o n of water i s >0.7 ). The two plots i n F i g . 45 again show that ^H^k^ i-n t n e presence of H 20/D 20 i s 1.4 ± 0.1. To check the r e p r o d u c i b i l i t y of the re s u l t s , t h e experiment was repeated with 2.0 ml H^ O or D^O ( i . e . with 15 times the exchangeable protons i n glucose) instead of 1.35 ml. The r e s u l t s i n F i g . 45 again show that fcpA = 1.5 ± 0.1. Further, no homogeneous c a t a l y s i s was observed i n the f i l t e r e d s l u r r i e s . The experiment was again repeated, t h i s time adding a l l the D 20 or H o0 to the DMSO. Since 0.8 g of hydrated 800° alumina contains 7% water 228 0 50 100 150 200 Time (min) F i g . 45 Deuterium Isotope E f f e c t with 800°C Alumina i n the Presence of H 20/D 20 229 or D 20 (^0.05 g water or D 20) , 0.75 g (= 0.80 - 0.05) anhydrous 800° alumina was used and 1.40 ml (= 1.35 + 0.05) D 20 or water was added to the 60 ml 0.05 M glucose s o l u t i o n . The r e s u l t s i n F i g . 45 again show that k H/ 7 = 1.4 ± 0.1. KT) These r e s u l t s show that the same normal isotope e f f e c t has been observed for glucose mutarotation by 800° alumina i n the presence and i n the absence of water. It i s also c l e a r from F i g . 45 that addition of 7% water to 800° alumina has increased i t s a c t i v i t y ^340%. In fact the a c t i v i t y of water treated 800° alumina i s greater than that of the o r i g i n a l alumina n e u t r a l . This should probably be due to the presence of a greater percentage of b a s i c s i t e s on the 800° alumina. XIX.1.3 Interpretation of the Observed Isotope E f f e c t Before these r e s u l t s can be interpreted and compared with the known values for the isotope e f f e c t by homogeneous c a t a l y s t s , i t i s important to determine the i n d i v i d u a l steps i n the c a t a l y t i c reaction that give r i s e to the observed isotope e f f e c t . As mentioned i n the Introduction, the observed rate constant for homogeneous c a t a l y s i s (Equation 51) i s k , = (k. + k~) a obs 1 2 S + C « » P + C (51) where c i s the c a t a l y s t concentration, and k^ and fe are the rate constants for forward and reverse reactions. Hence the observed deuterium isotope e f f e c t , 230 H. « H. (fe1 + k2 ) ^ » ~% V~ (52) « 2 + V> and i s the r e s u l t of the isotope e f f e c t on rate constants fe^ and k^- I t was shown i n Section VII that the observed rate constant f o r c a t a l y s i s by alumina (Equation 26) fe^ fe3 &2 S + C —** SC S — » PC -j—** P + C (26) *2 k4 *1 fe.fe. (fe, + fe.)c 1 2 3 4 o i s obs fe2 + fe2fe4 + fe2fe3 + kAk2 + fe3 + fe4)so (30) where c = t o t a l c a t a l y s t concentration and s = t o t a l substrate concentration, o o Since fe^ and fe2> the rate constants for adsorption and desorption are expected to be greater than fe3 and fe^, the rate constants f o r forward and reverse surface reactions (see Section VII), the Equation 30 s i m p l i f i e s to ... (fe, + k )o Hence,the observed deuterium isotope e f f e c t r e f l e c t s the deuterium isotope e f f e c t on a l l four steps (both, adsorption and desorption steps and surface c a t a l y t i c steps) unlike the homogeneous reactio n . However, i t i s obvious that for comparison of the mechanisms of homogeneous and heterogeneous reactions one should compare the isotope e f f e c t s for the steps that involve transformation of substrate to product ( i . e . (fe3 + fe^) for the 231 heterogeneous process i n Equation 26 with (k^ + k^) for the homogeneous reaction i n Equation 51). In order to carry out t h i s comparison i t i s convenient to consider f i r s t , t h e s i m p l i f i e d Equation 31 which contains the equilibrium constant for adsorption on c a t a l y t i c s i t e s . According to Equation 31,the observed deuterium isotope e f f e c t i s the r e s u l t of isotope e f f e c t on ( i ) the c a t a l y t i c constant for the surface reaction (k^ + k^) and ( i i ) on the equilibrium constant for adsorption on c a t a l y t i c s i t e s (k,/k ). The observed f i n a l 1 2 o p t i c a l rotations were s l i g h t l y higher when D^O was used instead of water (e.g. 0.131° when 1.35 ml water was used and 0.132° when 1.35 ml D^O was used, and 0.130° when 1.40 ml water was used and 0.131° when 1.40 ml D^O was used) but the d i f f e r e n c e i s not s i g n i f i c a n t because the p r e c i s i o n of the measurements i s ±0.001°. Further, more accurate measurement of o p t i c a l r o t a t i o n , for example,with a 1 dm path length c e l l at 25.0°C may not be very useful because we need to determine the isotope e f f e c t on adsorption at active s i t e s and not the isotope e f f e c t on t o t a l adsorption. From the above discussion i t i s c l e a r that i t i s d i f f i c u l t to determine whether there i s an isotope e f f e c t on the equilibrium constant for adsorption on c a t a l y t i c s i t e s . A more d i r e c t approach to the i n t e r p r e t a t i o n of observed isotope e f f e c t i s to determine whether there i s any isotope e f f e c t on the rate constants k^ and k^ by studying the adsorption/desorption process i n the absence of the c a t a l y t i c process. If there i s no isotope e f f e c t on the rate constants k^ and , then the observed isotope e f f e c t should be due to isotope e f f e c t on the c a t a l y t i c constant f o r the surface reaction (k^ + k^) and can be d i r e c t l y compared with the isotope e f f e c t observed for homogeneous reactions. On the other hand, i f there i s an isotope e f f e c t on adsorption and desorption, then the i n t e r p r e t a t i o n of isotope 232 e f f e c t r e s u l t s would be d i f f i c u l t . XIX.1.3.1 Deuterium Isotope E f f e c t on the Rate of Adsorption of Methyl  a-D-Glucoside The study of isotope e f f e c t on the rate of adsorption was car r i e d out using methyl ot-D-glucoside. This compound has e s s e n t i a l l y the same set of hydroxyl groups for adsorption on alumina surface and i s very s i m i l a r to glucose adsorption as represented by the f i r s t step i n Equation 26. Methyl glucoside deactivated alumina neutral towards glucose muta-r o t a t i o n (Section XVII) and hence,it should get adsorbed on the c a t a l y t i c a l l y a c t i v e s i t e s . Further, i t does not undergo mutarotation and therefore any complications due to a surface reaction and adsorption of intermediates i s eliminated (see below). A 100 ml so l u t i o n , 0.05 M i n methyl a-D-glucoside was made by di s s o l v i n g 0.97 g of the glucoside i n DMSO, adding 2.3 ml H^ O (so that , exchangeable protons i n H2O _ n o \ j , . ^ 1 r a t i o of : fi—— c : : rj- —13) and making up to the mark exchangeable protons i n glucoside with DMSO. Sixty ml of the so l u t i o n was s t i r r e d with 1.6 g 800° alumina and the change i n o p t i c a l r o t a t i o n with time was measured ( F i g . 46) as described i n the Experimental Section. The procedure was repeated using a s o l u t i o n containing 2.3 ml D^O instead of water. The r e s u l t s i n F i g . 46 c l e a r l y show that there i s no deuterium isotope e f f e c t on the rate of adsorption of methyl-a-D-glucoside. The rate of adsorption of the methyl glucoside from DMSO so l u t i o n , i n the presence of water, i s given by H H Rate^ H Q ^ = ka [Methyl glucoside - OH] [Catalyst] - k^ [Methyl glucoside - OH-Catalyst Complex] (53) 4.88 4.68 o Rate of adsorption i n the presence of H^O O Rate of adsorption i n the presence of D^ O 4.78 [Methyl a-D-glucoside] = 0.05 M [Alumina] = 26.7 mg.ml -1 4.58 75 Time (min) 100 125 150 Fig. 46 Deuterium Isotope E f f e c t on the Rate of Adsorption of Methyl a-D-Glucoside on 800°C Alumina 234 and the rate i n the presence of D^O i s , Rate^ D Q ^ = k^ [Methyl glucoside - OD] [Catalyst] - k^ iMethyl glucoside - OD - Catalyst Complex] (54) where k i s the rate constant f o r adsorption of methyl glucoside onto 3. alumina surface and k^ i s the rate constant for desorption. Since the two pl o t s i n F i g . 46 coincide at a l l times i t follows that Equation 53 equals Equation 54 at a l l concentrations of glucoside, catalyst,and glucoside-H D H D c a t a l y s t complex. Hence k = k and k. = k and there i s no deuterium r a a d d isotope e f f e c t on adsorption and desorption of glucoside on alumina surface. Since adsorption of methyl glucoside and adsorption of glucose onto alumina surface should both involve hydroxyl groups i t may be concluded that there should not be any deuterium isotope e f f e c t f or adsorption of glucose H D H D ( i . e . k^ should equal k^ and k^ should equal k ) . XIX.1.3.2 Deuterium Isotope E f f e c t on the Rate of Adsorption of Glucose It was mentioned e a r l i e r , that methyl ct-D-glucoside was chosen to study the deuterium isotope e f f e c t on adsorption by hydroxyl groups since i t does not undergo mutarotation on an alumina surface and hence, any complications due to surface reactions and adsorption of intermediates do not occur. At this stage i t i s of i n t e r e s t to determine how such a d d i t i o n a l factors would a f f e c t deuterium isotope e f f e c t on adsorption by studying the isotope e f f e c t on the adsorption of glucose onto alumina as described below. A 100 ml of 0.05 M e q u i l i b r a t e d glucose s o l u t i o n was prepared using 20 ml 0.25 M e q u i l i b r a t e d glucose s o l u t i o n , 2.7 ml ^ 0 (so that the exchangeable protons i n water _ . , „ s r a t i o n—. , . „ : z —12) and making upto the mark with exchangeable protons m glucose & r 235 DMSO. Sixty ml of the s o l u t i o n was s t i r r e d with 1.6 g 800 alumina and the rate of change of o p t i c a l r o t a t i o n with time was followed as before (Fig. 47). Another 100 ml, 0.05 M e q u i l i b r a t e d glucose s o l u t i o n was prepared using 2.7 ml D^ O (instead of water) and the rate of adsorption was again measured. The r e s u l t s i n F i g . 47 show that there i s a small normal deuterium isotope e f f e c t for adsorption of glucose onto alumina. Similar normal isotope e f f e c t was also observed when the rate of adsorption was measured with an e q u i l i b r a t e d 0.25 M s o l u t i o n . This observation of a normal isotope e f f e c t f or glucose adsorption may be due to an a d d i t i o n a l surface reaction undergone by glucose as explained below. I t was mentioned i n Section V I I . l that the catalyst-glucose complexes on alumina surface should consist not only of glucopyranose molecules adsorbed v i a hydroxyl groups but also a small amount of adsorbed intermediates (e.g. the a c y c l i c form). Hence the adsorption process was represented by kl _ k3 Glucopyranose + Catalyst m m Glucopyranose- m Intermediate-k2 Catalyst kS Catalyst Complex Complex T h u s , i n i t i a l rapid adsorption leads to formation of a glucopyranose-catalyst complex adsorbed v i a hydroxyl groups ( s i m i l a r to methyl glucoside-catalyst complex i n Equation 53 or 54). However,slow conversion of glucopyranose molecules to the intermediate species by step 3 w i l l decrease the concen-t r a t i o n of glucopyranose-catalyst complex. To maintain the equilibrium of the f i r s t step, more glucose w i l l get adsorbed onto the surface at a rate equal to the rate of the surface reaction by step 3. Hence any deuterium isotope e f f e c t for the rate constant k^ w i l l be r e f l e c t e d i n the rate of adsorption of glucose from s o l u t i o n . Hence a normal isotope e f f e c t for 1.40 • Rate of adsorption i n the presence of H^O O Rate of adsorption i n the presence of D^ O [Glucose] = 0.05 M [Alumina] = 26.7 mg.ml -1 25 50 75 100 Time (min) 125 150 175 47 Deuterium Isotope Effect on the Rate of Adsorption of Glucose on 800°C Alumina 237 k 1 w i l l r e s u l t i n a normal isotope e f f e c t f or adsorption of glucose, and vice versa. This explains the observed normal isotope e f f e c t f or glucose adsorption although deuterium s u b s t i t u t i o n d i d not a f f e c t the rate of adsorption of methyl glucoside onto alumina. From the above studies i t i s reasonable to conclude that for H D H D adsorption of glucopyranose onto alumina, k^ = k^ and k^ = k^ . This conclusion i s supported by the fact that no deuterium isotope e f f e c t has been observed for dehydration of alcohols on alumina when O-deuterated 49 alcohols were used . Note that adsorption of alcohols onto alumina also occurs v i a hydroxyl groups. Since there i s no isotope e f f e c t on the rate constants for adsorption and desorption of glucose i t follows from Equation 30 that the observed kobs k3E + kk isotope e f f e c t f o r glucose mutarotation — — = — — and hence i s k I k" + k, obs 3 4 the r e s u l t of deuterium isotope e f f e c t on rate constants k^ and k^ for forward and reverse surface reactions ( c f . Equation 52 obtained for homogeneous re a c t i o n ) . Hence the observed isotope e f f e c t should be d i r e c t l y comparable to those obtained for mutarotation i n s o l u t i o n . XIX.1.3.3 Mechanistic Conclusions From the discussion of deuterium isotope e f f e c t on glucose muta-ro t a t i o n given i n the Introduction (Section 1.2.4), the observed normal isotope e f f e c t of about 1.4 indicates that there i s general acid/base c a t a l y s i s by a c t i v e s i t e s on alumina surface. Further, the observed isotope e f f e c t l i e s between those observed for bases (2.4 for acetate, 2.9 for pyridine etc.) and strong acids 0<YJy - 1.37) i n s o l u t i o n and rules out any b i f u n c t i o n a l concerted c a t a l y s i s by 800° alumina. I t was mentioned i n the Introduction that b i f u n c t i o n a l c a t a l y s i s i s observed i n non-polar media 238 which cannot s t a b i l i z e an i o n i c t r a n s i t i o n state. The polar alumina surface should be able to s t a b i l i z e any i o n i c t r a n s i t i o n states formed during mutarotation and hence,catalysis by alumina surface occurs by a stepwise mechanism. As suggested by experiments with methyl glucoside (Section XVII (d)), the normal isotope e f f e c t also rules out carbonium ion mechanism for alumina c a t a l y s i s of glucose mutarotation. The s i m i l a r i t y of the observed isotope e f f e c t to those observed f o r homogeneous c a t a l y s i s also suggests that surface c a t a l y s i s also occurs by the same mechanism (via an a c y c l i c intermediate). The a b i l i t y of the c a t a l y t i c s i t e s to s t a b i l i z e an aldehyde fu n c t i o n a l group, and hence the t r a n s i t i o n state leading to i t , was observed i n Section XVII. XIX.1.3.3.1 E f f e c t of Water on C a t a l y t i c A c t i v i t y As described e a r l i e r i n t h i s section,the same deuterium isotope e f f e c t (within experimental error) was observed i n the presence and i n the absence of water while the a c t i v i t y of 800° alumina has increased about 340% on the addition of 7% water onto the surface (Note that the a c t i v i t y of hydrated 800° alumina i s greater than that of the o r i g i n a l alumina neutral probably because of the higher percentage of basic s i t e s ) . I t was mentioned i n the Introduction that the deuterium isotope e f f e c t f o r c a t a l y s i s by water i s about 3.9 and i s due to concerted mechanism invo l v i n g two or three water molecules i n the t r a n s i t i o n state. Hence, i f the increase i n a c t i v i t y of 800° alumina on treatment with water i s due to a separate mechanism invo l v i n g several water molecules, then there should be an increase i n the observed deuterium isotope e f f e c t . The absence of an independent mechanism i n v o l v i n g several water molecules was also indicated by the observed l i n e a r r e l a t i o n s h i p between the a c t i v i t y of water-treated 150° alumina and the amount of water added (see Section XII.2.1). 239 Further, any b i f u n c t i o n a l c a t a l y s i s i n v o l v i n g a water molecule and an a c i d i c or basic s i t e on the surface also should increase the isotope e f f e c t observed with hydrated alumina. Therefore,the increase i n a c t i v i t y observed when 800° alumina was treated with H^ O or D^O should be due to an increase i n the number of a c t i v e s i t e s by water molecules acting as a medium to transfer protons from a c i d i c s i t e s or to basic s i t e s from the adsorbed glucose molecules. In other words, on hydrated alumina,many p o t e n t i a l c a t a l y t i c s i t e s are not a c t i v e because the a c i d i c and basic f u n c t i o n a l groups on the surface and the adsorbed glucose molecules are not i n proper o r i e n t a t i o n f or proton trans f e r to take place. Adsorption of water onto the surface bridges the gap and helps to t r a n s f e r protons between functional groups and glucose molecules (Note, however, as observed i n Section XV, that the surface of a hydrated alumina, l i k e alumina neutral, s t i l l contains r e v e r s i b l e adsorption s i t e s which are not made active by adsorbed water). It was mentioned i n Section XIII.2.1.1 that treatment with water did not increase the amount of glucose adsorbed on 150° alumina. A small (~10%) increase i n the amount of glucose adsorbed was observed on t r e a t i n g 800° alumina with water. Howeve^the contribution of the increased glucose adsorption to about 340% increase i n a c t i v i t y probably i s n e g l i g i b l e . XIX.2 Deuterium Isotope E f f e c t f o r C a t a l y s i s by Sintered Alumina The deuterium isotope e f f e c t with alumina sintered at 1250°C was obtained only under anhydrous conditions, since as mentioned i n Section XIII.3, the r e p r o d u c i b i l i t y of the r e s u l t s i n the presence of water i s not good. Deuterated sintered alumina was not prepared since alumina dried at >900°C 25 contains <1% of the surface covered with hydroxyl groups . Since reproducible 240 r e s u l t s with alumina sintered at 1250°C can be obtained only when the k i n e t i c runs are ca r r i e d out within a short time (see Section X I I ) , a l l the necessary experiments f o r isotope e f f e c t determinations were c a r r i e d out within a few hours as described below. A batch (10 g) of alumina sintered at 1250°C for 6 hrs was prepared and was mixed well i n a m i n i m i l l . Sixty ml 0.05 M sol u t i o n of glucose-O-H i n DMSO was s t i r r e d with 1.6 g sintered alumina and the k i n e t i c s was followed as usual. The procedure was repeated about an hour a f t e r completing the f i r s t run and the same re s u l t s were obtained. A t h i r d k i n e t i c run was car r i e d out immediately afterwards with 60 ml 0.05 M glucose-O-D i n DMSO and 1.6 g sintered alumina and an isotope e f f e c t of 1.92 ± 0.06 was obtained (Fig. 48). Another k i n e t i c run, using 60 ml 0.05 M glucose-O-H and 1.6 g sintered alumina, was ca r r i e d out soon afterwards and gave a plo t super-imposable with the f i r s t two plot s and showed that the ca t a l y s t had not deactivated over the 10 hour period. XIX.3 Deuterium Isotope E f f e c t with Alumina Neutral A sample of c a t a l y s t free of excess water adsorbed on i t was prepared by evacuating 3 g of alumina neutral for 24 hours. It was then mixed w e l l . . .,, „ i „ , exchangeable protons i n water _ . m a mmi m i l l . Two ml water ( r a t i o : ° — — • : = — 15) exchangeable protons i n glucose was added to 60 ml 0.05 M so l u t i o n of a-D-glucose i n DMSO and was s t i r r e d with 0.8 g of the prepared sample. The k i n e t i c s was followed as usual. The procedure was repeated with 2 ml D^O instead of water. In both cases no homogeneous c a t a l y s i s was observed i n the f i l t e r e d s l u r r i e s . From the f i r s t order p l o t s the deuterium isotope e f f e c t was determined to be 1.3 ± 0.1. 241 F i g . 48 Deuterium Isotope E f f e c t with Alumina Sintered at 1250°C for 6 hours 242 XIX.4 Relationship between the Observed Deuterium Isotope E f f e c t and the  Percentage of A c t i v i t y due to Basic S i t e s . Examination of the values for the deuterium isotope e f f e c t observed with d i f f e r e n t alumina samples shows (Table XVIII) that the isotope e f f e c t increases with the increase of percentage a c t i v i t y due to basic s i t e s as determined by the e f f e c t of (X^ on a c t i v i t y . A p l o t of deuterium isotope e f f e c t versus percentage deactivation by CO^ i n F i g . 49 shows that there i s almost l i n e a r c o r r e l a t i o n between the two. TABLE XVIII DATA FOR THE RELATIONSHIP BETWEEN THE OBSERVED DEUTERIUM ISOTOPE EFFECT AND PERCENTAGE OF ACTIVITY DUE TO BASIC SITES Sample of Alumina Deuterium Isotope E f f e c t k^/j. Percentage Deactivation by Carbon Dioxide  Alumina Neutral 1.3 ± 0.1 10 + 5 800°C Alumina 1.43 ± 0.05 27 ± 7 1250°C Alumina 1.92 ± 0 . 0 6 85 ± 5 To test t h i s r e l a t i o n s h i p a new sample of alumina which gets deacti-vated 53 ± 7% by CO^ was prepared by heating a d i f f e r e n t batch of alumina neutral at 800°C for 4 hrs. According to the c o r r e l a t i o n i n F i g . 49 i t should show a deuterium isotope e f f e c t of about 1.65. The isotope e f f e c t observed using E^O/B^O was 1.7 ± 0.1. Hence i t follows that the percentage of c a t a l y t i c a c t i v i t y towards mutarotation by the basic s i t e s on alumina 0 25 50 75 100 Percentage Deactivation by Carbon Dioxide Fi g . 49 Correlation of the Observed Deuterium Isotope E f f e c t with the Percentage Deactivation by Carbon Dioxide 244 can be predicted from the deuterium isotope e f f e c t studies described e a r l i e r i n t h i s section. From the c o r r e l a t i o n diagram i t i s also c l e a r that the isotope e f f e c t due to a sample of alumina which undergoes complete deactivation by CO^ would be about 2.1 while the isotope e f f e c t when CO^ has no e f f e c t on a c t i v i t y i s about 1.2. Hence the c o r r e l a t i o n diagram indicates that the two types of s i t e s observed during the study of the e f f e c t of i n h i b i t o r s i n Section XVIII show d i f f e r e n t isotope e f f e c t s ; 2.1 due to the basic s i t e s and 1.2 due to the a c i d i c s i t e s . It was mentioned e a r l i e r i n t h i s section that the observed deuterium isotope e f f e c t indicates the absence of b i f u n c t i o n a l c a t a l y s i s by a c i d i c and basic s i t e s on the alumina surface. The absence of a maximum i n the c o r r e l a t i o n diagram when both a c i d i c and basic s i t e s are present on the alumina surface also indicates absence of b i f u n c t i o n a l c a t a l y s i s . Hence i t follows that the a c i d i c and basic s i t e s act independently by a consecutive mechanism i n catalyzing the mutarotation of glucose molecules adsorbed on the surface. Therefore the c o r r e l a t i o n diagram should r e f l e c t the independent behaviour of the two types of s i t e s . XIX.4.1 Derivation of a Relationship Between the Observed Isotope E f f e c t  and the Percentage A c t i v i t y due to Basic Sites A t h e o r e t i c a l r e l a t i o n s h i p between the observed isotope e f f e c t and the f r a c t i o n of the a c t i v i t y due to basic s i t e s , when the a c i d i c and basic s i t e s act independently,can be derived as follows. 245 Let A = rate constant ( i n the presence of OH) when a l l s i t e s are a c i d i c s i t e s B = rate constant (in the presence of OH) when a l l s i t e s are basic s i t e s = f r a c t i o n of s i t e s that are a c i d i c f „ = f r a c t i o n of s i t e s that are basic D Since the isotope e f f e c t s due to a c i d i c and basic s i t e s are 1.2 and 2.1, respectively, the rate constant due to a c i d i c s i t e s i n the presence of -OD i s Aj^ ^ and due to basic s i t e s i s Bl^ ^ . Therefore,the observed rate constants are and k = J - f + _B_ 1.2 A 2.1 B i f the a c i d i c and basic s i t e s act independently. Hence the observed deuterium isotope e f f e c t i s given by w = —A (55) D _i_ r + -L- f •'A J B 1.2 A 2.1 * and the f r a c t i o n of a c t i v i t y due to basic s i t e s i s given by B ?B F. = C56) Using Equations 55 and 56, a t h e o r e t i c a l c o r r e l a t i o n diagram can be 246 constructed as described below. A r b i t r a r y values f o r rate constants A and B (e.g. A = 1 and B = 1 or A = 1 and B - 100) are substituted i n Equations 55 arid 56. For d i f f e r e n t values of F^, corresponding values of f^ and are calculated from Equation 56 (Note that f A + f g ~ 1)• These values f o r and / g are then substituted i n Equation 55 and the corresponding values f o r the isotope e f f e c t are determined. I t can be shown that the values of ^ H / ^ obtained for d i f f e r e n t values of F^ are independent of the numbers used f o r rate constants A and B. The t h e o r e t i c a l and experimental plots are shown i n F i g . 50. I t i s clear that they are superimposable w i t h i n experimental error (error bars for the t h e o r e t i c a l p l o t are not shown i n F i g . 50). This proves that the two s i t e s are acting independently and that there i s no b i f u n c t i o n a l c a t a l y s i s . Percentage of A c t i v i t y due to Basic Sites F i g . 50 Comparison of the Experimental and Theoretical Correlation Plots SUMMARY 249 XX SUMMARY This thesis describes the k i n e t i c s and mechanism of a simple organic reaction i n s o l u t i o n catalyzed by a s o l i d held i n suspension. The reaction studied i s glucose mutarotation and the solvent i s dimethyl sulfoxide. The ca t a l y s t used i s Woelm alumina neutral ( f or t h i n layer chromatography), a 2 porous s o l i d with a BET surface area of 140 m /g (X-ray crystallography showed that i t i s the y -form). A l l reactions were c a r r i e d out at 25.0°C. As "standard conditions" the k i n e t i c s of the reaction was c a r r i e d out by s t i r r i n g a 60 ml 0.05 M s o l u t i o n of a-D-glucose containing 1.6 g alumina, with an overhead s t i r r e r such that the rate of the reaction was independent of the rate of s t i r r i n g ( F ig. 16). Under such conditions, the observed rate i s not c o n t r o l l e d by 'external' d i f f u s i o n of substrate molecules from the bulk s o l u t i o n to the surface of the alumina p a r t i c l e s . A decrease of p a r t i c l e s i z e (so as to decrease the pore length) by grinding did not a f f e c t the observed rate i n d i c a t i n g that the rate of the reaction i s also not co n t r o l l e d by ' i n t e r n a l ' d i f f u s i o n of substrate molecules from the surface of alumina p a r t i c l e s to active s i t e s w i t h i n the pores. Under such conditions, the f i r s t order k i n e t i c plots obtained were curved ( F i g . 7) unlike the w e l l known l i n e a r f i r s t order plots observed for homogeneous c a t a l y s i s . There was an i n i t i a l rapid decrease i n o p t i c a l r o t a t i o n during the f i r s t 10 to 15 minutes. Using an e q u i l i b r a t e d glucose s o l u t i o n i t was shown that t h i s i n i t i a l decrease i n o p t i c a l r o t a t i o n i s due to adsorption of glucose on the alumina surface which i s complete i n about 15 minutes (Fig. 15). Analysis of reaction products, to determine whether any other (non f i r s t order) reaction i s occurring on the surface,showed the presence of 40.3% a-D-glucopyranose,57.5% g-D-glucopyranose and <1% side 250 products i n a sample removed towards the end of a k i n e t i c run. Hence the curvature i n the f i r s t order pl o t s could mean either the surface catalyzed mutarotation i s not f i r s t order i n glucose or that alumina i s undergoing progressive deactivation during c a t a l y s i s . Tests f o r deactivation of the c a t a l y s t were c a r r i e d out with alumina that had been previously used to catalyze glucose mutarotation. Studies with 'used' alumina showed that the c a t a l y s t has undergone deactivation during the previous c a t a l y t i c process. I t was also shown that solvent DMSO or water (that may be formed by dehydration of glucose) did not deactivate the c a t a l y s t . Six and h a l f hours of mutarotation decreased c a t a l y t i c a c t i v i t y about 50% and a f t e r 70 hours of c a t a l y s i s the a c t i v i t y had decreased about 70%. P a r t i a l l y deactivated c a t a l y s t s gave l i n e a r f i r s t order plots over two or three-half l i v e s . This shows that glucose mutarotation on alumina surface i s a f i r s t order process. Further studies showed that deactivation of the c a t a l y s t i s not due to i n h i b i t o r y e f f e c t of any side products i n s o l u t i o n but due to strong adsorption (possibly by ether formation) of glucose on a c t i v e s i t e s . These r e s u l t s therefore showed that the i n i t i a l decrease i n o p t i c a l r o t a t i o n i s due to surface adsorption while the curvature i n the r e s t of the p l o t i s due to progressive deactivation of the c a t a l y s t . Hence the rate constant for the c a t a l y t i c reaction was determined by measuring the slope at t = 25 mins when the adsorption i s complete but before much deactivation has occurred. Dehydration of alumina neutral at low temperatures caused a decrease i n s p e c i f i c a c t i v i t y ( a c t i v i t y per u n i t area). But as the temperature was increased above^600°C the s p e c i f i c a c t i v i t y began to increase and the 2 a-alumina formed at 1250°C (surface area = 6.2 m /g; percentage glucose 251 adsorbed by 1.6 g of a-alumina from 60 ml 0.05 M glucose s o l u t i o n = 3% compared to 14% adsorbed by alumina n e u t r a l under the same conditions) showed a s p e c i f i c a c t i v i t y which i s 26 times that of standard alumina neutral ( F i g . 28). Such high a c t i v i t y f o r a-alumina (compared with Y-alumina) i s v i r t u a l l y unknown. Further, this c a t a l y s t did not get deactivated during the c a t a l y s i s and always produced l i n e a r f i r s t order plots over three h a l f - l i v e s (Fig. 27). The observed decrease i n a c t i v i t y on mild thermal treatment i s unlike other reactions catalyzed by alumina where c a t a l y t i c a c t i v i t y i s observed only on heating the c a t a l y s t above 300 to 400°C (giving r i s e to defect s i t e s ) . This showed that defect s i t e s ( c o n s i s t i n g of c l u s t e r s of vacancies and neighboring oxide i o n s ) , even i f they possess c a t a l y t i c a c t i v i t y towards mutarotation, do not possess a monopoly over the c a t a l y t i c a c t i v i t y as they do with many other observed reactions. The observed higher c a t a l y t i c a c t i v i t y of the hydrated alumina surface towards glucose mutarotation compared with alumina dehydrated by mild thermal treatment (forming defect s i t e s ) should be due to the high s e n s i t i v i t y of the muta-ro t a t i o n reaction towards weak a c i d i c and basic s i t e s on the hydrated surface. I t was also shown that p y r o l y s i s of alumina neutral at temperatures greater than the Tammann temperature brings out i t s basic character which gives r i s e to the observed high c a t a l y t i c a c t i v i t y of a-alumina. Rehydration of the dehydrated c a t a l y s t increased i t s a c t i v i t y without increasing tire amount of glucose adsorbed. I t was shown that the increase of a c t i v i t y i s due to the increase of the number of a c t i v e s i t e s by adsorbed water possibly acting as a medium to trans f e r protons and hydroxide groups from the a c i d i c and basic s i t e s on the surface to the adsorption s i t e s . 252 Adsorption studies using e q u i l i b r a t e d solutions of glucose showed, that i n addition to the r e v e r s i b l e adsorption si t e s , that there are i r r e v e r s i b l e adsorption s i t e s on the surface of alumina neutral. The amount of i r r e v e r s i b l e -4 adsorption s i t e s was determined to be (0.70 ± 0.02) x 10 mole/g. The adsorption isotherms f o r alumina neutral and alumina sintered at 1250°C showed i n i t i a l strong adsorption and saturation of adsorption at high glucose concentrations i n d i c a t i n g that there i s only monolayer formation on the alumina surface (Fig. 33). The maximum adsorption for alumina neutral corresponds to about 60% surface coverage. This suggests that 'the surface has f a i r l y s p e c i f i c s i t e s f o r adsorption of glucose. The Langmuir p l o t obtained from the data f o r r e v e r s i b l e adsorption on alumina neutral showed two l i n e a r regions (Fig. 34). I t was shown that such a p l o t can a r i s e from the presence of two types of r e v e r s i b l e adsorption s i t e s on the surface; one showing strong adsorption (equilibrium constant for adsorption = 2 - 1 -4 (8.2+ 1.2) x 10 l i t r e mole and there are (1.0 ± 0.1) x 10 mole strong adsorption sites/g) and the other showing weak adsorption (equilibrium constant f o r adsorption = 44 ± 3 l i t r e mole ^ and there are (1.4 ± 0.2) -4 x 10 mole weak adsorption s i t e s / g ) . The comparison of a p l o t of i n i t i a l rate versus concentration of glucose with t h e o r e t i c a l adsorption isotherms fo r weak and strong adsorption s i t e s showed (Fig. 38) that only the weak adsorption s i t e s are c a t a l y t i c a l l y a c t i v e . Thus the active s i t e density on 13 2 standard alumina neutral was determined to be (5.4 ± 0.6)10 sites/cm . -3 The turnover number of a c a t a l y t i c s i t e was determined to be 2 x 10 molecules/site/sec. I t i s one of the highest turnover numbers obtained for a reaction on an alumina surface. It was also shown that the determination of the ac t i v e s i t e density by comparison of adsorption isotherms with an i n i t i a l rate p l o t i s a more 253 d i r e c t method than the standard method using i n h i b i t o r s . Further, i t can d i s t i n g u i s h s i t e s that might contain a c i d i c and basic s i t e s but which are not a c t i v e (because of i r r e v e r s i b l e adsorption or improper o r i e n t a t i o n of func t i o n a l groups), from the a c t i v e s i t e s and give a more complete picture of the surface s i t e s than other methods of determining active s i t e density. The surface catalyzed reaction can be represented by Equation 26. k l k3 k2 S + C T " * SC PC T ~ * * P + C (26) k2 kk k l I t was shown that such a system would show f i r s t order k i n e t i c s with observed rate constant (k. + k.) a k v = — = * ^ (32) o b s kjk, + s 2 1 o (where = t o t a l c a t a l y s t concentration and S q = t o t a l glucose concentration) when k^y &2 > > ^3' ^ * As predicted from Equation 32, the observed rate constant was found to be d i r e c t l y proportional to the concentration of c a t a l y s t (Fig. 21), while the p l o t of 1/^ ^ g versus concentration of substrate was l i n e a r with a small p o s i t i v e intercept (Fig. 22). Using the slopes of the two plots and the equilibrium constant f o r adsorption on c a t a l y t i c s i t e s the c a t a l y t i c constant f o r the surface reaction (k^ + Tc^ ) was -3 -1 determined to be 5 x 10 sec . Comparison with the c a t a l y t i c constant -4 -1 for homogeneous r e a c t i o n i n pure water (= 4 x 10 sec ) showed that the alumina surface o f f e r s a better medium for mutarotation than water. Comparison of the c a t a l y t i c constant f o r o v e r a l l reaction (= kQ^s per unit c a t a l y s t concentration) with the c a t a l y t i c constants f o r homogeneous 254 cat a l y s t s ( i n Table III) showed that the c a t a l y t i c s i t e s on alumina neutral are 10 times more active than strong acids i n water and a l i t t l e more ac t i v e than the b i f u n c t i o n a l c a t a l y s t 2-hydroxypyridine i n benzene. The nature of the ac t i v e s i t e s was investigated using "neutral" compounds (to determine the nature of adsorption on the ac t i v e s i t e s ) and a c i d i c and basic compounds (to determine the nature of functional groups involved i n c a t a l y s i s ) . Simple monohydroxy compounds (e.g. water and methanol) had no e f f e c t on c a t a l y t i c a c t i v i t y or the adsorption of glucose on alumina neutral even when used i n excess. However, polyhydroxy compounds (e.g. methyl a-D-glucoside and i - i n o s i t o l ) were e f f i c i e n t i n h i b i t o r s of glucose mutarotation. For example, equimolar amounts of glucose and i n o s i t o l decreased the c a t a l y t i c a c t i v i t y of alumina neutral by 67% and the amount of glucose adsorbed by 60%. This indicated that i n o s i t o l had no preference f o r c a t a l y t i c a l l y a c t i v e over i n a c t i v e s i t e s . Hexanal was used as a probe for s i t e s that can i n t e r a c t with the a c y c l i c intermediate that may be formed during mutarotation. Unlike polyhydroxy compounds, the percentage deactivation by hexanal was always greater than the percentage decrease i n adsorption (Table XVI). This indicated that the ac t i v e s i t e s on alumina can adsorb and s t a b i l i z e the a c y c l i c intermediate (more p a r t i c u l a r l y the t r a n s i t i o n state leading to i t ) better than the other adsorption s i t e s . Using methyl a-D-glucoside and hexanal i t was shown that rapid deactivation of the c a t a l y s t occurs due to i n t e r a c t i o n of hydroxyl groups (of the glucoside) with the alumina surface,while further, but slow, deactivation occurs by i n t e r a c t i o n of the aldehyde f u n c t i o n a l group (of hexanal) with a c t i v e s i t e s . Hence i t appears that i n t e r a c t i o n of the alumina surface with f u n c t i o n a l groups,which i s e s s e n t i a l for adsorption of glucose and s t a b i l i z a t i o n of t r a n s i t i o n state, also leads to permanent 255 deactivation of some of the a c t i v e s i t e s . Treatment with carbon dioxide deactivated alumina neutral as w e l l as a l l aluminas prepared by dehydrating alumina neutral. Alumina sintered at 1250°C was deactivated 85% by carbon dioxide. However,the CO^ treated sample regained the o r i g i n a l a c t i v i t y on heating to a mild (^650°C) temperature. Further, the greater the extent of dehydration of the sample the greater was the percentage deactivation (Table XVIII). Since carbon dioxide may be expected to react with basic anionic s i t e s , t h i s deactivation indicates the presence of c a t a l y t i c a l l y a c t i v e oxide ions at the a c t i v e s i t e s . Pyridine had no i n h i b i t o r y e f f e c t on alumina neutral even when used i n excess. This showed that Lewis acid s i t e s on alumina neutral are not c a t a l y t i c a l l y a c t i v e . Stronger bases such as n-butylamine and tetramethyl-ammonium hydroxide increased the c a t a l y t i c a c t i v i t y i n d i c a t i n g that weak Bronsted acid s i t e s on alumina are c a t a l y t i c a l l y active (and are converted to more act i v e anionic s i t e s by reaction with strong bases). Mutarotation of glucose by alumina i s probably the only reaction catalyzed by weak Bronsted acid s i t e s on alumina at moderate temperatures. This shows that glucose mutarotation i s a more s e n s i t i v e probe for Bronsted acid s i t e s on s o l i d s than any of the techniques used before. Approximately 90% of the c a t a l y t i c a c t i v i t y of standard alumina neutral i s due to these weak BrOnsted acid s i t e s . The observed high c a t a l y t i c a c t i v i t y of these weak acid s i t e s may be due to ( i ) strong s p e c i f i c adsorption of glucopyranose molecules at the a c t i v e s i t e s with the Bronsted a c i d s i t e on the surface at the proper p o s i t i o n to protonate the r i n g oxygen atom and/or ( i i ) s t a b i l i z a t i o n of the t r a n s i t i o n state by the polar alumina surface, as suggested by the experiments with hexanal. Alumina neutral and other aluminas prepared by dehydrating alumina 256 neutral showed normal isotope e f f e c t s which are s i m i l a r to those observed for homogeneous ca t a l y s t s i n s o l u t i o n . There was no isotope e f f e c t on the adsorption-desorption process and hence, the observed isotope e f f e c t i s due to the c a t a l y t i c r eaction on the surface. Therefore, the observed normal isotope e f f e c t s i n d i c a t e that glucose mutarotation on alumina surface i s a general acid-base catalyzed reaction, and proceeds by a consecutive mechanism, v i a the a c y l i c intermediate. The observed isotope e f f e c t increased l i n e a r l y with the increase of percentage a c t i v i t y due to basic s i t e s as determined by CC^ i n h i b i t i o n (Fig. 49). The a c i d i c s i t e s seem to show an isotope e f f e c t of ^ J J / ^ = a n c* basic s i t e s an isotope e f f e c t of fey/^ = 2.1. The observed isotope e f f e c t agreed c l o s e l y with that c a l c u l a t e d assuming that the two s i t e s are acting independently without any b i f u n c t i o n a l concerted c a t a l y s i s (Fig. 50). In conclusion, the r e s u l t s discussed i n t h i s thesis have not only established, for the f i r s t time, the rate constants for an alumina catalyzed organic reaction i n s o l u t i o n but have also shown that the main features of the surface catalyzed glucose mutarotation (for example l i n e a r or non-linear f i r s t order p l o t s , amount of glucose adsorbed, e f f e c t of water on the c a t a l y t i c a c t i v i t y , deuterium isotope e f f e c t s ) which i s highly s e n s i t i v e to weak acid and base s i t e s on the surface, may be used to characterize d i f f e r e n t aluminas. 257 XXI EXPERIMENTAL 258 XXI EXPERIMENTAL XXI.1 General Methods Optical rotations were determined on a Perkin-Elmer 241 MC Polarimeter at X = 365 nm. The measurements were made at 365 nm, instead of the usual 589 nm,because of the greater sensitivity (sensitivity increased 24 by a factor of 3) at the shorter wavelength. Specific rotations ([a]3^5) were calculated using the equation, 2^ observed rotation (degrees) 365 path length (dm) x concentration (g/ml) Gas Liquid Chromatography (glc) was carried out in a Perkin-Elmer 900 Gas Chromatograph equipped with a flame ionization detector using helium as carrier gas. Proton nuclear magnetic resonance spectra were recorded on a Varian XL-100 spectrometer by staff members of the NMR laboratory, the University of British Columbia. Samples were prepared as 10% solutions in DMSO-d, and had 1% tetramethylsilane (TMS) as an internal standard, o Electron Micrographs of alumina samples were obtained on an Elec Corporation Autoscan Scanning Electron Microscope at 20 KV by Ms. Sally Finora, University of British Columbia. pH's of slurries of aluminas in water were determined with a Radiometer pH Meter 26 equipped with glass and calomel electrodes. The melting point of methyl a-D-glucoside was determined with a Thomas Hoover Unimelt Capillary Melting Point Apparatus using an open tube capillary and i s not corrected. Elemental micro-analysis were performed by Mr. Peter Borda, University of British Columbia. Batches of dehydrated aluminas and a,8 mixtures of glucose were mixed on a Fischer Minimill. A l l glassware used i n catalytic studies were f i r s t cleaned in a 259 chromic a c i d bath. Then the glassware, s t i r r e r , syringes and Sweeny syringe f i l t e r were washed i n an u l t r a s o n i c cleaner. The equipment was then washed i n tap water, rinsed with d i s t i l l e d water followed by acetone (reagent grade) and dried i n an oven at^l20°C. A f t e r about 3 hours the equipment was cooled i n a stream of dry nitrogen. XXI.2 Materials Aluminum oxide neutral f o r t h i n layer chromatography (without any binder) manufactured by Woelm Pharma ( d i s t r i b u t e d by ICN Pharmaceuticals), was used as the c a t a l y s t throughout t h i s work. Other aluminas used to compare a c t i v i t i e s and the names of d i s t r i b u t o r s or manufacturers are given i n the text. A l l the aluminas used i n t h i s thesis do not contain any binder. Eastman (Reagent ACS) DMSO was dried and d i s t i l l e d as described below (Section XXI.3.1). Anhydrous a-D-glucose (Reagent ACS, Matheson, Coleman and B e l l ) and anhydrous $-D-glucose (ICN Pharmaceuticals) were dried in vacuo over ^-^^ at 60°C for 24 hours. A l l anhydrous compounds were stored i n a dessicator over P2^5' F r e s h l y d i s t i l l e d water was used to hydrate alumina samples and for isotope e f f e c t studies. Deuterium oxide (Gold l a b e l , Aldrich) containing 99.8 atom % deuterium was used. Methyl a-D-glucoside (BDH Chemicals) was r e c r y s t a l l i z e d from f i v e parts of methanol 127 and dried in vacuo at 60°C mp 163-164°C ( l i t . 164-165°C). Benzene, methanol, pyrid i n e , n-rhutylamine. C a l l reagent grade), naphthalene (recrys-t a l l i z e d , Eastman), i - i n o s i t o l (Sigma), DL—glyceraldehyde (Aldrich) and, hexamethyldisilazane, trimethylchlorosilane and dichlorodimethylsilane obtained from A l f a Products were used without any further p u r i f i c a t i o n . A p u r i f i e d (by means of an alumina column to remove small amount of trimer) sample of hexanal (Aldrich) was a generous g i f t of Dr. V. G u j r a l . 260 XXI.3 Preparations XXI.3.1 Preparation of Dry, D i s t i l l e d DMSO One l i t r e of Eastman Reagent ACS Grade DMSO was dried overnight over 10 g of calcium h y d r i d e ^ 5 and was refluxed at reduced pressure for one hour. I t was then d i s t i l l e d (at 38-40°C) under reduced pressure and the center f r a c t i o n ( — 800 ml) was c o l l e c t e d over 5 g of alumina neutral a c t i v i t y super I (Woelm Pharma). The vacuum was broken by introducing dry nitrogen i n t o the d i s t i l l a t i o n apparatus. The fl a s k containing d i s t i l l e d DMSO was stored under dry nitrogen. 18 32 XXI.3.2 Preparation of Alumina from Aluminum Isopropoxide ' Two hundred grams of aluminum isopropoxide (98 + %, A l f a Products) was d i s t i l l e d (atcallO°C) under reduced pressure and middle f r a c t i o n (=*150 g) was c o l l e c t e d . The l a t t e r was d i s t i l l e d again under reduced pressure and middle f r a c t i o n (ca100 g) was c o l l e c t e d . This pure ion-free aluminum isopropoxide was slowly added with s t i r r i n g to 700 mis of d i s t i l l e d water. A f t e r a l l the isopropoxide had been added the s l u r r y was s t i r r e d f o r another s i x hours. The s l u r r y was f i l t e r e d under suction and the aluminum hydroxide was washed with excess water. I t was then dried at 120°C for three days. A portion (20 g) was further dried at 600°C under nitrogen f o r four hours to give a sample of pure alumina. XXI.3.3 Preparation of Alumina Neutral A c t i v i t y V from Alumina Neutral 128 A c t i v i t y Super I F i f t e e n grams of alumina neutral a c t i v i t y super I (on the Brockmann 128 129 scale ' ) taken i n an erlenmeyer f l a s k , was treated with. 2.85 ml 261 (19% by weight of alumina) of d i s t i l l e d water. The f l a s k was stoppered, shaken w e l l and was allowed to stand at room temperature (=c24°C) for 2 hours. XXI.3.4 Preparation of a, 8 Mixture The a, 8 mixture of equilibrium composition used i n adsorption (Section V) and the deactivation (Section VIII.1) studies was prepared by mixing together a-D-glucose and 8-D-glucose i n the r a t i o 1.00 : 1.67 65 (equilibrium constant i n DMSO at 24.7°C = 1.67) . I t was then ground w e l l with a pestle and mortar, mixed f o r an hour on a m i n i m i l l and dried in vacuo at 60°C over 1*2^ 5" XXI.3.5 Preparation of a Completely E q u i l i b r a t e d Solution of a > 8 Mixture  i n DMSO For the study of the adsorption isotherms (Section XIV.1) and the isotope e f f e c t on the rate of adsorption (Section XIX.3.2), completely e q u i l i b r a t e d 1.0 M and 2.0 M solutions of a , 8 mixture i n DMSO were made as described below. In order to prepare 500 ml of 1.0 M s o l u t i o n , 33.7 g of a-D-glucose and 56.3 g of g-D-glucose were dissolved i n about 400 ml of DMSO, the so l u t i o n was transferred i n t o a 500 ml volumetric f l a s k and the volume was made up with DMSO. Addition of a drop of 0.1 N NaOH to about 20 ml of the so l u t i o n increased i t s o p t i c a l r o t a t i o n showing that the s o l u t i o n was not completely e q u i l i b r a t e d . Hence the s o l u t i o n was transferred to a 1 l i t r e erlenmeyer f l a s k containing 10 g of sintered (1250°C) alumina and 10 g of alumina n e u t r a l , and the s l u r r y was s t i r r e d with a magnetic s t i r b a r under dry nitrogen at 25.0°C. Af t e r a few (2 or 3) days the alumina was allowed 262 to s e t t l e , a sample of the s l u r r y was removed by a Pasteur pipette and f i l t e r e d through a Swinny syringe f i l t e r . The o p t i c a l r o t a t i o n of the c l e a r s o l u t i o n was determined using a 1 dm path length c e l l thermostatted at 25.0°C. The s t i r r i n g was continued and the o p t i c a l r o t a t i o n of another sample was measured a f t e r a few more days. This process of s t i r r i n g and measuring o p t i c a l rotations of f i l t e r e d s l u r r i e s was continued u n t i l no further change i n the o p t i c a l r o t a t i o n was observed (generally i n about 7 to 10 days; i n some cases complete e q u i l i b r a t i o n was achieved i n a shorter time by using 15 to 20 g of sin t e r e d alumina). The alumina was allowed to s e t t l e and the supernatant was f i l t e r e d under suction to give a c l e a r s o l u t i o n . Complete e q u i l i b r a t i o n of the s o l u t i o n was again determined by adding a drop of 0.1 N NaOH to about 20 ml of the s o l u t i o n (no change i n o p t i c a l r o t a t i o n was observed). This completely e q u i l i b r a t e d s o l u t i o n was stored under nitrogen at 25.0°C. The same procedure was adopted for the preparation of the 2.0 M e q u i l i b r a t e d s o l u t i o n . Solutions of lower concentrations were made by progressive d i l u t i o n of more concentrated solutions with DMSO. XXI.3.6 Preparation of Dehydrated Aluminas XXI.3.6.1 Dehydration at Room Temperature Dehydration of alumina neutral at room temperature (24°C) was c a r r i e d out i n a drying p i s t o l over a t 0.01 m m H g pressure. A f t e r drying the sample for 4 days CP 2*^5 w a s changed once), the vacuum i n the system was broken by introducing dry nitrogen i n t o the drying p i s t o l . The alumina sample was then mixed w e l l on a m i n i m i l l . 263 XXI.3.6.2 Dehydration at 150°C Dehydration of alumina neutral at 150°C was c a r r i e d out i n an Ace Instatherm drying apparatus (manufactured by Ace Glass Inc., Vineland, N.J.) at 0.01 mm Hg pressure. A f t e r drying the sample for two days the apparatus was allowed to cool to room temperature and the vacuum was broken by introducing dry nitrogen. The sample was then mixed w e l l on a m i n i m i l l . XXI.3.6.3 Dehydration at 600°C and Higher Temperatures Dehydration of alumina neutral at 600°C and higher temperatures was c a r r i e d out i n an E l e c t r i c Hi-Speed Furnace Type G-05-PT (power 3500 Watts, manufactured by Hevi Duty E l e c t r i c Co., Milwaukee, Wisconsin). A long quartz tube was placed i n the furnace and one end of the tube projected out through a hole i n the rear w a l l of the furnace. Alumina sample was taken i n an Alumina Combustion Boat (Grade AD-99, purchased from Coors Porcela i n Company, Golden, Colorado) and introduced into the quartz tube. Dry nitrogen (Linde, U.S.P., 99% N^) was introduced into the furnace through the end of the tube which projected out of the furnace. The furnace was switched on and was heated at maximum rate u n t i l the required temperature was reached. A f t e r the sample had been dried at the required temperature (e.g. 600°C) for a p a r t i c u l a r length of time (e.g. 4 hours), the furnace was switched o f f and was allowed to cool to room temperature overnight under nitrogen. The dried alumina sample was then mixed on a m i n i m i l l . XXI.3.7 Preparation of Carbon Dioxide Treated Alumina About 4 g of an alumina sample was packed into a chromatographic separation tube with a dis c to support column packing. By means of a gas i n l e t tube connected to the top of the separation tube, dry carbon dioxide 264 (Linde, U.S.P.,99% CO ) was passed through the alumina sample. The flow of carbon dioxide was monitored by a gas bubbler connected to the bottom of the separation tube by a v i n y l tube. A f t e r the treatment was complete (24 hours) the sample was mixed on a m i n i m i l l . The percentage carbon, determined by micro-analysis, on sintered (1250°C) alumina i s 0.05% and on CO^ treated s i n t e r e d alumina i s 0.04%. XXI.3.8 Preparation of a Standard Solution of Tetramethylammonium Hydroxide  Pentahydrate and Determination of the Number of Acid Sites on Alumina Neutral XXI.3.8.1 Preparation and Standardization of a Solution of Tetramethyl- ammonium Hydroxide Pentahydrate A known weight (0.055 g) of the hydroxide (Eastman) taken i n a 50 ml beaker was dissolved i n 10 ml of DMSO with s l i g h t warming. The small 130 amount of the undissolved material (possibly carbonate) was removed by f i l t r a t i o n , through a Buchner glass funnel with a f r i t t e d disc ( f i n e ) , i n t o an erlenmeyer f l a s k . The clear s o l u t i o n of the hydroxide was t i t r a t e d against a standard (0.0122 M) s o l u t i o n of hydrochloric acid using phenol-phthalein as i n d i c a t o r . The amount of acid needed f o r n e u t r a l i z a t i o n was 16.8 ml. The procedure was repeated with a new sample (same weight) of the hydroxide and i t required 16.0 ml of the acid f o r n e u t r a l i z a t i o n . * -L. -i !6.4 x 0.0122 Hence the average number of moles of the acid used = -4 0 055 -4 1 0 0 0 = 2.0 x 10 moles. Therefore ' = 3.0 x 10 mole of the tetramethyl-1 8 1 -4 ammonium hydroxide pentahydrate contains 2.0 x 10 mole ( i . e . 66%) of the base. During the above operations dry nitrogen was passed over the beaker, 265 the funnel and the f l a s k and they were covered with parafilm to minimize exposure of the solutions to the atmosphere. XXI.3.8.2 Determination of the Number of Acid Sites on Alumina Neutral  by Treatment with Excess Tetramethylammonium Hydroxide XXI.3.8.2.1 Amount of the Hydroxide Required From the r e s u l t s of the treatment of alumina neutral with tetramethylammonium hydroxide discussed i n Section XVIII.2.3, 0.08 g of the alumina i s neu t r a l i z e d by 2 x 10 ^  mole of the base. Hence 1.6 g of -4 alumina neutral would need 4 x 10 mole of the base for n e u t r a l i z a t i o n . -4 Further, 10 ml of 0.0122 M hydrochloric acid s o l u t i o n needs 1.2 x 10 mole of the base for n e u t r a l i z a t i o n . Therefore,if 1.6 g of alumina neutral i s -4 -4 -4 treated with 5.2 x 10 (= 4 x 10 + 1.2 x 10 ) mole of the hydroxide, 10 ml of the standard acid would be required for n e u t r a l i z a t i o n of unreacted base. -4 From the r e s u l t s of standardization experiment above, 5.2 x 10 mole of the hydroxide i s found i n ~0.16 g of tetramethylammonium hydroxide penta-hydrate. XXI.3.8.2.2 Treatment of Alumina Neutral with Excess Tetramethylammonium  Hydroxide A known weight (0.165 g) of the hydroxide was dissolved i n 10 ml of DMSO and f i l t e r e d as described above. The cl e a r s o l u t i o n was added to 1.6 g of alumina neutral taken i n an erlenmeyer f l a s k . The f l a s k was then sealed and s t i r r e d f o r 120 mins. The s l u r r y was f i l t e r e d through a Buchner glass funnel with a f r i t t e d disc (fine) and the alumina was washed with 2 ml of DMSO i n three portions. The combined f i l t r a t e was 266 t i t r a t e d against 0.0122 M hydrochloric acid s o l u t i o n using phenolphthalein as i n d i c a t o r and 8.0 ml of the acid was needed for colour change to occur. The procedure was repeated with the same weights of the hydroxide and alumina, and the f i l t r a t e required 8.2 ml of the acid f o r n e u t r a l i z a t i o n . From the r e s u l t s of the standardization experiment described above, 0.165 g of the tetramethylammonium hydroxide pentahydrate contains 6.0 x -4 10 mole of the hydroxide. The number of moles of hydroxide l e f t a f t e r . n_ , r ^ •, . 8.1 x 0.0122 . n n i n-4 tr e a t i n g i t with 1.6 g of alumina = = 0.99 x 10 mole. 1000 Therefore the number of moles of hydroxide that has reacted with 1.6 g of -4 alumina n e u t r a l s 5 x 10 mole. The experiment was repeated using a greater amount of the hydroxide (0.216 g of the hydroxide dissolved i n 20 ml of DMSO) but tr e a t i n g i t with 1.6 g of alumina neutral f o r one hour. The unreacted hydroxide needed 24.9 ml of the 0.0122 M acid for n e u t r a l i z a t i o n . Again,from the r e s u l t s of the standardization experiment, 0.216 g of -4 the hydroxide contains 7.9 x 10 mole of hydroxide ions. The number of moles of hydroxide ions l e f t a f t e r t r e a t i n g i t with 1.6 g of alumina neutral = 24.9 x 0.0122 „ . ..-4 . _ . , , , f = 3.0 x 10 mole. Therefore the number of moles of 1000 _ A hydroxide that has reacted with 1.6 g of alumina neutral = 5 x 10 , which i s the same as the r e s u l t obtained by the previous experiment. From t h i s r e s u l t the number of acid s i t e s on alumina was calculated as described i n Section XVIII.2.3. 267 XXI.3.9 Preparation of Diazomethane and Treatment of Alumina Samples  with Diazomethane XXI.3.9.1 Amount of Diazomethane Required I t was observed i n Section XII that further dehydration of a sample of alumina neutral, that has been dehydrated at room temperature, causes about 5% loss i n weight. Hence a 4 g sample can lose . 0.2 g of water (on further dehydration) or contains 0.022 mole of hydroxyl groups on the surface. I f a l l the hydroxyl groups are a c i d i c a minimum of 0.022 mole of diazomethane i s required to methylate the surface. The amount of diazomethane required may also be estimated from the number of acid groups determined by treatment of alumina neutral with tetramethylammonium hydroxide i n Section XXI.3.8.2.2 above. Since 1.6 g -4 of alumina reacted with 5 x 10 mole of the hydroxide, 4 g of the alumina -3 sample would need 1.2 x 10 mole of the base for n e u t r a l i z a t i o n . The higher number of moles of surface hydroxyl groups determined by dehydration studies should be due to the presence of basic hydroxyl groups and strongly adsorbed water molecules on the surface. These strongly adsorbed water molecules may also react with diazomethane i f the reaction i s catalyzed by Lewis acid s i t e s on the surface. Therefore, i n order to allow for decomposition of diazomethane, i t was decided to use about 3 times ( i . e . 0.07 mole) the minimum estimated by dehydration studies. XXI.3.9.2 Preparation of Diazomethane 131 The procedure outlined by de Boer and Backer was employed to prepare an alcohol-free ethereal s o l u t i o n of diazomethane. The y i e l d of diazomethane prepared by t h i s method i s 70-787^^^ . Therefore to 268 prepare 0.07 mole of diazomethane 0.1 mole (21.5 g) of Diazald* was used. About 200 ml of ethereal** diazomethane s o l u t i o n was prepared by the method of de Boer and Backer and was dried over anhydrous sodium sulphate (Reagent ACS; from Mallinckrodt) for one hour. The ethereal s o l u t i o n was then decanted i n t o two f l a s k s . About 150 ml was added i n t o one and was used to trea t alumina samples with diazomethane and the r e s t of the diazomethane was allowed to decompose at room temperature and was used to prepare the con t r o l samples. XXI.3.9.3 Treatment of Alumina Samples with Diazomethane S l u r r i e s , of alumina neutral evacuated at room temperature and alumina sintered at 1250°C, i n d i e t h y l ether (4 g of each i n 30 ml of ether) were allowed to cool i n an i c e - s a l t mixture. They were s t i r r e d with Teflon-coated magnetic s t i r b a r s and the diazomethane s o l u t i o n (prepared i n Section XXI.3.9.2 above) was added slowly with a Pasteur p i p e t t e . The s l u r r y of sintered alumina turned yellow when a few mis of the diazomethane s o l u t i o n was added while the s l u r r y of alumina neutral needed 8-10 mis to turn yellow. Decomposition of diazomethane ( i . e . evolution of bubbles) was observed on the surface of alumina n e u t r a l . More diazomethane was added to both the s l u r r i e s and they were l e f t overnight at room temperature i n a fume hood. The diazomethane s o l u t i o n was stored i n a r e f r i g e r a t o r . A f t e r about 12 hours the s l u r r y containing sintered alumina s t i l l had a l i g h t yellow colour while that of alumina neutral was c o l o u r l e s s . More diazomethane was added to both the s l u r r i e s and as the colour i n the ** Diazald (N-Methyl—N-nitroso—p—tolunesulphonamide) i s available from A l d r i c h Chemical Company. Anhydrous Et h y l Ether (Reagent ACS) from Mallinckrodt was used. 269 s l u r r y of alumina neutral disappeared more diazomethane was added. The procedure was continued for 6 hours and both the s l u r r i e s were again allowed to stand overnight. A f t e r about 12 hours the addition of diazomethane was continued u n t i l a l l the 150 mis had been added. The s l u r r y of sintered alumina was decanted and the supernatant ( t o t a l volume ^ 50 mis) was c o l l e c t e d . The alumina was washed with three 50 ml portions of ether. Ether remaining with the diazomethane treated sintered alumina was evaporated i n a water bath and the sample was dried -in vacuo at room temperature (~23°C) f o r 4 hours. The same procedure was adopted to obtain diazomethane treated alumina neutral from i t s s l u r r y . XXI.3.9.4 Preparation of the Control Samples The supernatants c o l l e c t e d from the two s l u r r i e s ( in Section XXI.3.9.3 above) were combined with the 50 ml of the diazomethane s o l u t i o n that was l e f t to decompose at room temperature. I t was used to tre a t s l u r r i e s of alumina neutral evacuated at room temperature and sintered alumina i n ether. From the two s l u r r i e s the control samples were obtained using the procedure described above. 132 XXI.3.10 Preparation of Dichlorodimethylsilane ( ( C H 3 ) 2 S i C l 2 ) Treated Alumina XXI.3.10.1 Amount of Dichlorodimethylsilane needed The minimum amount of ( C H 3 ) 2 S i C l 2 needed can be determined by two methods; (i) from the number of oxygens on the surface determined by model studies and ( i i ) from the number of hydroxyl groups determined by dehydration studies, as discussed below. 270 23 From the l i t e r a t u r e the average number of oxygens (present as oxide 15 -2 19 -2 ions and hydroxyl groups) on the surface of alumina i s 10 cm =10 m Therefore the number of moles of oxygen on the surface of 3 g of alumina 19 144 x 10 x 3 -3 neutral dehydrated at room temperature = ^ — = 7 x 10 mole. _ 3 6 x 1 0 Hence 7 x 10 mole of ( C H ^ ^ S i C ^ should completely s i l y l a t e the alumina surface. From the dehydration studies described i n Section XII, about 5% of the weight of alumina neutral dehydrated at room temperature can be l o s t as water. Therefore the number of moles of hydroxyl groups on the surface of 5 x 3 x 2 a 3 g sample i s = 0.017 mole. The higher value determined by 100 x 18 dehydration studies may be due to the presence of strongly adsorbed water molecules on the surface of the sample. Therefore the minimum amount of (CH^) S i C ^ necessary to completely s i l y l a t e 3 g of alumina i s about 0.02 mole. However, i n order to allow for side reactions wtih any moisture i n the solvent and the surface of the glass v e s s e l , 0.08 mole of ( C H ^ ^ S i C ^ was used as described below. XXI.3.10.2 Treatment of Alumina with (CHg) S i C l 2 Into 100 ml of Fischer ACS Toluene (dried and d i s t i l l e d over l i t h i u m aluminum hydride) was added 10 g (0.08 mole) of (CH^^SiCl,, and 132 3 g of alumina neutral that had been evacuated at room temperature (24°C) The s l u r r y was s t i r r e d for 20 hours under nitrogen and was f i l t e r e d using a Buchner glass funnel with a f r i t t e d disc ( f i n e ) . Dry nitrogen was passed over the funnel to minimize exposure of the alumina surface to water vapour i n the atmosphere. The alumina i n the funnel was washed three times with 10 ml portions of methanol. Another 10 ml of methanol was added and the alumina i n the funnel was broken up with a spatula. The methanol also 271 was f i l t e r e d through and a f i f t h 10 ml portion of methanol was added. The f i l t r a t i o n was continued u n t i l the sample was dry. The alumina was transferred into a drying p i s t o l and was dried in vacuo at room temperature (24°C). XXI.3.11 Preparation of Deuterated 800° Alumina and Control Sample of Alumina XXI.3.11.1 Amount of Replaceable Hydrogen on Surface of Alumina It was observed i n Section XII that about 8% of the weight of alumina neutral can be l o s t on dehydration. Hence the amount of replaceable 4 x 8 x 2 hydrogen on 4 g of alumina neutral i s — I Q O X 13 = 0-036 mole. Five ml of 5 x 1 1 x 2 D^O contains j g = 0.56 mole of deuterons. Hence the deuterons i n 5 ml of D^O should replace most of the protons on the surface of a 4 g 133 sample of the alumina . To replace almost a l l the protons on the surface with deuterons i t was decided to treat the 4 g sample with three 5 ml portions of D^O. XXI.3.11.2 Preparation of Deuterated 800° Alumina Four g of alumina neutral was mixed w e l l with 5 ml of 1)^0 i n a Buchner glass funnel with a f r i t t e d disc ( f i n e ) . The funnel was covered with a piece of rubber sheet and was allowed to stand f o r 5 mins. The D^O was removed by suction and another 5 ml portion of D^O was added to the funnel and mixed w e l l with the alumina. A f t e r 5 mins i t was f i l t e r e d and the alumina was treated with the t h i r d 5 ml portion of D 20. The procedure was repeated and most of the D^O was removed by applying suction for one hour. The alumina sample was broken up by a spatula and dried at 800°C 272 under nitrogen for 4 hours. The dried sample had lumpy p a r t i c l e s (formed by s t i c k i n g ^together of alumina p a r t i c l e s ) and was mixed w e l l on a m i n i m i l l which broke down the large lumpy p a r t i c l e s . The smaller lumpy p a r t i c l e s remaining tended to block a Pasteur pipette when samples were removed during early stages of a k i n e t i c run. However, they broke down quickly (within about 30 mins) as the s l u r r y was r a p i d l y s t i r r e d and hence,the f i r s t o p t i c a l r o t a t i o n measurement during k i n e t i c runs inv o l v i n g such aluminas was taken at t ^ 30 mins. XXI.3.11.3 Preparation of the Control Sample of 800° Alumina The same procedure was followed except that the same quantity of d i s t i l l e d water was used instead of D^O. XXI.3.12 Preparation of O-Deuterated a - D - G l u c o s e 1 3 3 , 1 3 4 ' 1 3 5 134 An i n d u s t r i a l method for preparation of anhydrous a-D-glucose was adopted to prepare O-deuterated a-D-glucose as described below. The preparation was c a r r i e d out i n a 50 ml two neck round bottom flask.with one neck connected to a dropping funnel and the other neck connected to a receiving f l a s k through a d i s t i l l a t i o n head and a condenser. Twelve grams of a-D-glucose (contains 0.33 mole of exchangeable protons) was taken i n the f l a s k and 50 ml of B^O (contains 5.5 moles of exchangeable deuterons) was added to the dropping funnel. F i f t e e n ml of the B^0 was added to the f l a s k and the glucose was dissolved by s t i r r i n g with a magnetic s t i r b a r with s l i g h t warming on a water bath. Most of the solvent was d i s t i l l e d o f f at about 40°C under reduced pressure leaving a thick syrup i n the f l a s k . The water bath was removed and another 15 ml of B^O was added from the dropping funnel and the syrup was dissolved i n D2O. The 273 133 135 DJ3 was d i s t i l l e d o f f as before and the procedure was repeated ' with a t h i r d portion (10 ml) of B^O. A f t e r d i s t i l l i n g o f f the t h i r d portion of D2O, more D^O was added from the dropping funnel u n t i l the t o t a l volume was about 20 ml (therefore the s o l u t i o n was 60% i n glucose). The dropping funnel and the d i s t i l l a t i o n head were removed and dry nitrogen was blown over the s o l u t i o n with the temperature of the water bath raised to 65°C. The s o l u t i o n was seeded with a few c r y s t a l s of a-D-glucose (from National Bureau of Standards) and growth of the c r y s t a l s began within the f i r s t day. The c r y s t a l l i z a t i o n was allowed to proceed u n t i l almost a l l the solvent had evaporated (in about 3 days). The s o l i d mass was broken up i n a glove bag under nitrogen, ground with a pestle and mortar and dried at 60°C, in vacuo, over ^2^5" This sample of O-deuterated a-D-glucose was used to seed the 60% glucose solutions obtained during the preparation of ad d i t i o n a l batches of O-deuterated a-D-glucose. 24 o The s p e c i f i c r o t a t i o n of the O-deuterated glucose sample [ a ] = 328, and may be compared with the s p e c i f i c r o t a t i o n of a-D-glucose from National Bureau of Standards = 333°. This shows that O-deuterated glucose prepared by the above procedure i s almost completely the a-form. The proton nmr spectra of an O-deuterated glucose sample ( i . e . a-glucose-OD), anhydrous a-D-glucose ( i . e . a-glucose-OH), and anhydrous g-D-glucose ( i . e . g-glucose-OH) i n DMSO-d, are shown i n Fig s . 51, 52 and 53 re s p e c t i v e l y . The absence of o the OH resonance at 6 = 4.86 i n F i g . 51 c l e a r l y shows that the f i v e hydroxyl protons present i n normal glucose (Figs. 52 and 53) have been replaced by deuterons. O-deuterated a-D-glucose could also be prepared by the above procedure but without seeding with a-D-glucose. However, c r y s t a l l i z a t i o n s t arted only a f t e r blowing nitrogen over the sample for about 2 days and F i g . 5 2 Proton MR Spectrum of a-D-Glucose in DMSO-d 6 F i g . 5 3 Proton NMR Spectrum of g-D-Glucose i n DMSO-d NJ ^1 ON 277 and hence,complete c r y s t a l l i z a t i o n by th i s process took about 2 to 3 days longer. Further, the O-deuterated glucose prepared without seeding contained about 10% of g-D-glucose. XXI.4 K i n e t i c and Adsorption Methods XXI.4.1 General Procedure for Following K i n e t i c s of the Surface Catalyzed  Reaction Sixty ml of a a-D-glucose s o l u t i o n was pipetted into a 200 ml three neck round bottom f l a s k placed i n a water bath maintained at 25.0°C. One of the side necks was connected to a nitrogen i n l e t tube and the other side neck, which was generally l e f t stoppered, was used to add the c a t a l y s t and to remove samples of the s l u r r y . A dispersion s t i r r e r connected to an overhead Fischer Dyna-Mix passed through a Teflon stopper f i x e d to the ce n t r a l neck. A f t e r the s o l u t i o n i n the f l a s k had reached thermal equilibrium with the water bath ( i n 15-20 mins) the f l a s k was charged with a preweighed sample of the ca t a l y s t and immediately the s t i r r e r and timer were st a r t e d . Approximately 2 ml samples of the s l u r r y were removed p e r i o d i c a l l y with a Pasteur pipette and f i l t e r e d through a Swinny syringe f i l t e r . Whatman No. 50 f i l t e r paper was used i n a l l f i l t r a t i o n s and produced a clear f i l t r a t e ( f i r s t few drops of the f i l t r a t e which tended to be s l i g h t l y cloudy were discarded) free of any p a r t i c l e s . Whatman No. 42 f i l t e r paper could also be used to f i l t e r s l u r r i e s of sintered alumina samples. O p t i c a l rotations (at 365 nm) of the f i l t e r e d s l u r r i e s were measured using a 0.1 dm path length c e l l at room temperature (23-24°C). The i n f i n i t y 278 values of o p t i c a l rotations (a^) for c a t a l y s i s by highly a c t i v e aluminas (e.g. alumina neutral, alumina dehydrated at 150°C, sintered aluminas) or water treated c a t a l y s t s were determined a f t e r s t i r r i n g the s l u r r y overnight. For l e s s a c t i v e c a t a l y s t s (e.g. 800° alumina) the i n f i n i t y values (a^) were obtained by adding a homogeneous c a t a l y s t (e.g. few drops of n-butylamine or 0.1 N sodium hydroxide) or by s t i r r i n g the ca t a l y s t with an e q u i l i b r a t e d glucose s o l u t i o n of the same concentration. I n f i n i t y value of the o p t i c a l r o t a t i o n f o r homogeneous reaction was determined by adding a few drops of n-butylamine or 0.1 N sodium hydroxide to about 20 ml of the o r i g i n a l glucose s o l u t i o n . XXI.4.2 Rate of Adsorption of Glucose on Alumina (Section V.3) A 60 ml 0.05 M s o l u t i o n of a,g mixture i n DMSO taken i n the 200 ml f l a s k described above was allowed to come to thermal equilibrium with a water bath at 25.0°C. I t was then charged with a preweighed sample of alumina and immediately the s t i r r e r and timer were st a r t e d . Samples of the s l u r r y were removed every 2 mins, f i l t e r e d as described before and the o p t i c a l rotations were determined with a 0.1 dm path length c e l l at room temperature. In order to f i l t e r many samples i n a short time several syringes and Swinny syringe f i l t e r s were used. XXI.4.3 Determination of Adsorption Isotherms For the study of adsorption isotherms, the completely e q u i l i b r a t e d solutions of a,(3 mixtures i n DMSO, prepared according to the procedure given i n Section XXI.3.5, were used. The amount of glucose adsorbed by alumina n e u t r a l at a p a r t i c u l a r e quilibrium concentration of glucose was determined as described below. 279 Sixty ml of an e q u i l i b r a t e d glucose s o l u t i o n was s t i r r e d with a 1.6 g sample of alumina neutral under nitrogen at 25.0°C. O p t i c a l rotations of f i l t e r e d s l u r r i e s (see Section XXI.4.1) were measured p e r i o d i c a l l y , using a 1 dm path length c e l l thermostatted at 25.0°C, u n t i l there was no further decrease. Results obtained with a 0.05 M s o l u t i o n are given i n Table XIX. TABLE XIX DATA FOR ADSORPTION OF GLUCOSE BY 1.6 G OF ALUMINA NEUTRAL FROM A 60 ML 0.05 M SOLUTION OF a,B MIXTURE IN DMSO AT 25.0°C Time (mins) O p t i c a l Rotation (q°) at 25.0°C 0 1.454 60 1.285 120 1.278 200 1.274 225 1.274 300 1.273 A very slow decrease i n o p t i c a l r o t a t i o n was observed as s l u r r i e s ( s p e c i a l l y with solutions of low glucose concentrations) were continuously s t i r r e d and could be due to side reactions occurring on the surface. The e s s e n t i a l l y constant o p t i c a l r o t a t i o n a f t e r 3 to 4 hours of s t i r r i n g was taken as the equilibrium value. Further, the e f f e c t of f i l t r a t i o n on the o p t i c a l r o t a t i o n of a s o l u t i o n was found to be n e g l i g i b l e . The procedure was repeated with solutions of d i f f e r e n t glucose concentrations'. When solutions 280 of high glucose concentrations (e.g. 1.5 M, 2.0 M) were used i t was observed that a i r bubbles accumulated i n f i l t e r e d s l u r r i e s . Hence, i n order to remove the bubbles, the f i l t e r e d s l u r r i e s were allowed to stand fo r several hours before o p t i c a l rotations were determined. The i n i t i a l and f i n a l (equilibrium) o p t i c a l rotations of the solutions used are given i n Columns 2 and 3 of Table X (Section XIV.1). In order to determine the r e p r o d u c i b i l i t y of the r e s u l t s , the experiment was repeated at tvro concentrations and the r e s u l t s are included i n Table X. As shown i n F i g . 54 a p l o t of o p t i c a l r o t a t i o n versus the concentration of solutions of a,3 mixture i s l i n e a r , and passes through the o r i g i n , over the range of concentrations used i n t h i s experiment. Hence the o p t i c a l r o t a t i o n of a s o l u t i o n i s d i r e c t l y proportional to i t s concentration and the change i n o p t i c a l r o t a t i o n of a s o l u t i o n i s d i r e c t l y proportional to the -2 -1 change i n i t s concentration. The slope of the p l o t = 3.47 x 10 mole l i t r e degree \ i s the change i n concentration of a s o l u t i o n when i t s o p t i c a l r o t a t i o n changes by one degree. I f the difference between the i n i t i a l and f i n a l o p t i c a l rotations a. . . ,- a , equals x degrees then the change r i n i t i a l e q u i l . -2 -1 i n concentration of the s o l u t i o n = 3.47 x 10 x x mole l i t r e . Hence the number of moles of glucose adsorbed by a gram of c a t a l y s t dispersed i n 60 ml _2 .. . 3.47 x 10 x x x 0.060 , . . . _ . , - _ , . v s o l u t i o n i s z— and i s given i n Column 6 of Table X. l.o E q uilibrium ( f i n a l ) concentrations of the glucose solutions were determined using the equilibrium o p t i c a l rotations and the slope of the plot i n F i g . 54. The same procedure was used to determine the adsorption isotherm for alumina sintered at 1250°C f o r 6 hours. However, as discussed i n Section XIV.2, 3.2 g samples of the c a t a l y s t were used i n order to increase the s e n s i t i v i t y of measurements. Optical Rotation (degrees) Fi g . 54 Relation of the Optical Rotation to the Concentration of a,B Mixture in DMSO 282 XXI.4.4 Tests for R e v e r s i b i l i t y of Adsorption Tests f o r r e v e r s i b i l i t y of adsorption on alumina neutral were ca r r i e d out by d i l u t i n g a glucose s o l u t i o n i n equilibrium with glucose adsorbed on a sample of alumina, i n order to release some of the adsorbed glucose. (a) A 60 ml 0.05 M (concentration used i n most k i n e t i c studies) s o l u t i o n of a , 6 mixture i n DMSO was s t i r r e d with 3.2 g of alumina neutral u n t i l the o p t i c a l r o t a t i o n was constant (in about 3 hours). Samples were not removed from t h i s s l u r r y . However, the equilibrium o p t i c a l r o t a t i o n (1.116°) and concentration (0.0385 M) were determined by another experiment. Sixty ml of DMSO was added to the system and the s t i r r i n g was continued u n t i l there was no further change i n o p t i c a l r o t a t i o n ( a = 0.569°; concentration = 0.0196 M) . I f there was no glucose released on d i l u t i o n 1 116° then the o p t i c a l r o t a t i o n should be —'- = 0.558°. The increase i n 2 _ 5 o p t i c a l r o t a t i o n by 0.01° indicates that 4.5 x 10 mole of glucose has been released from the surface to the s o l u t i o n . (b) The same experiment was performed with a 0.005 M s o l u t i o n . The re s u l t s i n d i c a t e that even at th i s very low concentration there i s rev e r s i b l e adsorption because 1 x 10 5 mole of glucose was released on d i l u t i o n . XXI.4.5 Deuterium Isotope E f f e c t on the Rate of Adsorption Solutions of methyl a -D-glucoside i n DMSO containing H^ O or DJ3 were prepared as described i n Section XIX.1.3.1. Sixty ml of the so l u t i o n was taken i n a 200 ml f l a s k placed i n a water bath maintained at 25.0°C (see Section XXI.4.1). A f t e r thermal equilibrium has been reached, the fl a s k was charged with 1.6 g of 800° alumina and the rate of adsorption 283 was followed as described i n Section XXI.4.2. Several syringes and Swinny syringe f i l t e r s were used to f a c i l i t a t e f i l t r a t i o n of several samples i n a short time. Further, i n order to determine o p t i c a l rotations with high s e n s i t i v i t y a 1 dm path length c e l l thermostatted at 25.0°C was used. XXI.5 A n a l y t i c a l Methods XXI.5.1 Product Analysis Analysis of products of glucose mutarotation by alumina was c a r r i e d out by preparing t r i m e t h y l s i l y l ethers of glucose i n the f i l t e r e d s l u r r y 68 according to the procedure of Sweeley e t . a l . To 1 ml of the f i l t e r e d s l u r r y obtained near the end of a k i n e t i c run was added 0.2 ml of hexamethyldisilaze, 0.1 ml of trimethylchlorosilane and 1 ml of anhydrous pyridine (kept over KOH p e l l e t s ) . The mixture was shaken f o r 30 mins and was allowed to stand at room temperature for 6 h r s ^ 7 . It was then concentrated in vacuo, extracted into hexane and analyzed by gas chromatography using a s t a i n l e s s s t e e l column 1/8" x 6 T packed with 8% 6 7 SE 30 on AW Chromosorb W 60/80 at 215°C . Two main peaks and a small peak were observed. The three peaks A, B and C had r e l a t i v e retention times 0.82, 1.00 and 1.35 which accounted for 2.2%, 40.3% and 57.5% of the area. In order to i d e n t i f y these peaks, t r i m e t h y l s i l y l ethers of ot-D-glucopyranose and 8-D-glucopyranose were prepared and analyzed by gas chromatography as described below. Ten mg of anhydrous a-D-glucopyranose was dissolved i n 1 ml of DMSO and was s i l y l a t e d using the procedure described above. I t was analyzed by gas chromatography under the same conditions and produced a large peak with 284 retention time equal to that of peak B above ( r e l a t i v e retention time 1.00) and area=^96% of the t o t a l area, and a small peak (area 4%) with r e l a t i v e retention time 1.35. This shows that the peak B observed during product analysis i s due to a-D-glucopyranose. In addition to these two peaks , a very small peak with area=a0.5% of the t o t a l area and r e l a t i v e retention time 0.81 was also observed (see below). The same procedure was adopted to prepare the t r i m e t h y l s i l y l ether from anhydrous 3-D-glucopyranose and was analyzed by gas chromatography under the same conditions. A large peak (area 97% of the t o t a l area) with a retention time equal to peak C above ( r e l a t i v e retention time 1.35) and a small peak (area 3%) with r e l a t i v e retention time 1.00 were observed. This shows that the peak C observed during product analysis i s due to g-D-glucopyranose. In add i t i o n to these two peaks a small peak with area ca0.5% of the t o t a l area and r e l a t i v e retention time 0.82 was also observed. As mentioned i n Section IV the small peak with r e l a t i v e retention time 0.82 observed during analysis of the products and also during analysis of a- and g-D-glucopyranoses may be due to a small amount of t r i m e t h y l s i l y l ethers of a- and g-D-glucofuranoses^ 7 present i n DMSO s o l u t i o n . In addition to the three peaks A, B and C, a smaller peak with area less than 1% of the t o t a l area and r e l a t i v e retention time 1.10 was observed during the analysis of products, but i t was not i d e n t i f i e d . XXI.5.2 Analysis of Alumina Samples for Trace Metal Impurities Alumina samples were tested f or trace metal impurities by Can Test Ltd., Vancouver, B r i t i s h Columbia. According to the test report "samples were digested using a combination of acids (HF, HNO^, HC1, HCIO^). A trace metal scan was performed using an Inductively Coupled Argon Plasma 285 Spectrograph". The impurities detected i n d i f f e r e n t alumina samples obtained from Woelm Pharma and the Puratronic alumina sample are given i n Table XX. The Puratronic sample was also analyzed i n order to compare the r e s u l t s obtained by Can Test Ltd., with C e r t i f i c a t e of Analysis of the Puratronic Sample provided by Johnson Matthey Chemicals Ltd., England, where only Na (2 ppm), Ca, Mg, Si and Ag (each <1 ppm) had been detected. The reasons for the discrepancies i n the two a n a l y t i c a l test reports are not c l e a r . XXI.5.3 C r y s t a l l i n e Structure of Aluminas The c r y s t a l l i n e structures of alumina neutral and alumina sin t e r e d at 1250°C were determined from X-ray powder photographs obtained using a P h i l l i p s powder camera of 57 mm radius and n i c k e l f i l t e r e d CuK r a d i a t i o n . a The samples were ground with pestle and mortar and were sealed i n 0.5 mm OD Lindermann glass c a p i l l a r i e s . Each sample was exposed to n i c k e l f i l t e r e d CuK^ r a d i a t i o n f o r 18 hours. The density of d i f f r a c t i o n l i n e s on powder photographs was measured on a densitometer. The r e l a t i v e i n t e n s i t i e s of the peaks on the densitometer p l o t were determined by cuttin g them out and weighing. The d i f f r a c t i o n l i n e s on the photograph were measured with a t r a v e l l i n g microscope and were converted to 8 (the angle of r e f l e c t i o n ) values. The d spacings were then obtained from the 6 values using the r e l a t i o n d = 2 Sin 0 o where \ i s the wavelength of CuK a r a d i a t i o n (1.5418 A). The res u l t s for alumina neutral are given i n Table XXI together with data from the l i t e r a -t u r e 1 7 on r e l a t i v e i n t e n s i t i e s of d i f f r a c t i o n l i n e s with same d spacings and the forms of alumina giving r i s e to those l i n e s . TABLE XX TRACE METALS PRESENT IN DIFFERENT ALUMINAS Metal A l u m i n a S a m p l e Barium Calcium Copper Iron Magnesium Manganese Molybdenum Sodium Strontium T i n Titanium Vanadium Zinc Neutral Batch 1 1.5 1050 <1.5 262 216 1.8 <4 229 5.6 14.1 30.5 2.3 11.1 Neutral" Batch 1 Sintered at 1250°C, 6 hrs 1.4 1080 26.4 102 214 1.1 <4 254 8.5 5.8 140 3.5 11.5 Neutral Batch 2 3.4 1080 1.7 189 142 1.6 <4 276 11.3 16.0 36.2 1.5 <1.5 Neutral* Batch 2 Sintered at 1250 "C, 6 hrs 4.4 1180 18.1 104 137 1.2 <4 237 16.9 4.7 30.5 2.2 18.6 Basic 1.2 95 <1.5 199 2.8 2.0 8.6 1190 <0.1 <3 29.0 1.7 17.8 Basic Sintered at 1250°C, 6 hrs 1.4 138 5.6 121 <0.1 1.2 <4 1310 <0.1 6.8 27.1 • 2.6 21.0 A c i d i c 0.4 63 1.7 152 3.8 . 2.3 <4 226 0.4 14.9 104 1.4 17.7 Puratronic 0.6 32.4 <1.5 43.6 <0.1 <0.3 <4 <10 0.2 <3 7.6 <1.0 <1.5 OO Expressed i n parts per m i l l i o n by weight Aluminas for thin layer chromatography manufactured by Woelm Pharma 287 TABLE XXI X-RAY DIFFRACTION DATA FOR ALUMINA NEUTRAL d Spacing A Observed From L i t e r a t u r e 17 Relative Intensity Relative Intensity Form of Alumina 4.54 broad 11 12 2.42 38 60 2.29 33 2.12 50 30 strong K X 1.99 doublet 60 65 1.89 20 10 1.395 doublet 100 100 288 A doublet of high i n t e n s i t y at 1.40 A, a doublet at 1.98 A and a ° 17 broad band at 4.6 A are c h a r a c t e r i s t i c of y-alumina . However, a few of o o the l i n e s observed (for example at 1.89 A, 2.12 A) could have resulted from the presence of a small amount of other forms of alumina (e.g. K- and/or X-alumina) . The absence of Ti-form i s indicated by the absence of a sharp o band at 4.6 A. These r e s u l t s i n d i c a t e that alumina neutral consist mainly of the y-form with some K- (and x - ) aluminas. The presence of forms of alumina belonging to 3-series (see Section 1.1.2) may be due to the presence of a r e l a t i v e l y large amount of calcium i n alumina neutral (Table XX) . Results of X-ray d i f f r a c t i o n studies of alumina neutral sintered at 1250°C for 6 hours are given i n Table XXII. A l l the observed l i n e s were sharp and the r e s u l t s agree w e l l with the l i t e r a t u r e 1 7 values for a-alumina. This shows that Y-alumina has been converted completely to the a-form on s i n t e r i n g at 1250°C. 289 TABLE XXII X-RAY DIFFRACTION DATA FOR ALUMINA SINTERED AT 1250°C for 6 HOURS o d Spacing A Relative Intensity * 17 , Observed From l i t e r a t u r e (for a-alumina) 3.479 69 74 2.556 91 92 2.382 42 42 2.170 1.5 <1 2.087 100 100 1.964 2.0 1 1.743 44 43 1.604 89 81 1.548 3.6 3 1.515 15 4 (1.514 h 7 (1.511 A) 1.408 33 32 1.374 54 48 1.340 2.1 1.277 2.8 1.242 27 1.193 10 1.151 7 1.128 11 * A l l l i n e s were sharp 290 BIBLIOGRAPHY 291 BIBLIOGRAPHY M. Boudart and R.L. Burwell J r . i n "Investigation of Rates and Mechanisms of Reactions", Part I, 3rd E d i t i o n , Chapter XII, Techniques  of Chemistry, Vol. VI, E.S. Lewis, Ed., Wiley-Interscience, New York, 1974. H. Knozinger and P. Ratnasamy, Catal. Rev. S c i . Eng., 17 (1978), p. 32, 33. P. Sabatier, "Die Katalyze i n der Organischen Chemie", Akademische ve r l a g s g e s e l l s c h a f t , L e i p z i g , 1914. P. Sabatier and E.E. Reid, " C a t a l y s i s i n Organic Chemistry", Van Nostrand, Princeton, 1922. C.L. Thomas, " C a t a l y t i c Processes and Proven Catalysts", Academic Press, New York, 1970, p. 240-253. C.N. S a t t e r f i e l d , "Heterogeneous C a t a l y s i s i n P r a c t i c e " , McGraw-Hill, New York, 1980, p. 86-91, 253-256. C S . John and M.S. S c u r r e l l , i n " C a t a l y s i s , V o l. 1", A S p e c i a l i s t P e r i o d i c a l Report, The Chemical Society, London, 1977, Chapter 4. B.C. Gates, J.R. Katzer and G.C.A. Schuit, "Chemistry of C a t a l y t i c Processes", McGraw-Hill, New York, 1979, Chapter 3. E. Lederer and M. Lederer, "Chromatography - A Review of P r i n c i p l e s and Ap p l i c a t i o n s " , E l s e v i e r , Amsterdam, 1957, p. 61-67. V.S. Joshi, N.P. Damodaran and Sukh Dev, Tetrahedron, 24, 5817 (1968). V.S. Joshi and Sukh Dev, Tetrahedron, 33, 2955 (1977). G.H. Posner, Angew. Chem. Int. Ed. Engl., 17, 487 (1978). G.H. Posner andD.Z. Rogers, J . Am. Chem. S o c , 99, 8208 (1977). J . Manassen and H. Pines, Adv. Catal., 16, 49 (1966). 292 15. H. Knozinger, Angew. Chem. Int. Ed. Engl., ]_, 791 (1968). 16. J.E. L e f f l e r and D.W. M i l l e r , J. Am. Chem. S o c , 99_, 480 (1977). 17. B.C. Lippens and J . J . Steggerda, i n "Physical and Chemical Aspects of Adsorbents and Cata l y s t s " , B.G. Linsen, Ed., Academic Press, London, 1970, Chapter 4 . 18. S.J. Gregg and K.H. Wheatley, J . Chem. S o c , 3804 (1955). 19. D.S. Maciver, H.H. Tobin and R.T. Barth, J. Ca t a l . , 2, 485 (1963). 20. J.B. P e r i , J . Phys. Chem., 69, 220 (1965). 21. H. Knozinger, Adv. C a t a l . , _27, ( 1 9 7 6 ) , p. 204. 22. D.S. Maciver, W.H. Wilmot and J.H. Bridges, J . Ca t a l . , Z, 502 (1964). 23. H. Knbzinger and P. Ratnasamy, Catal. Rev. S c i . Eng., 17, 31 (1978). 24. K. Tanabe, " S o l i d Acids and Bases - Their C a t a l y t i c Properties", Kodansha, Tokyo, and Academic Press, New York, 1970. 25. J.B. P e r i and R.B. Hannan, J . Phys. Chem., 64, 1526 (1960). 26. E. Bo r e l l o , G. D e l i a Gatta, B. F u b i n i , C. Morterra and G. Venturello, J . Catal., 35, 1 (1974). 27. J.H. de Boer, J.M.H. Fortuin, B.C. Lippens and W.H. Meijs, J . Catal., 2, 1 (1963). 28. J.B. P e r i , J. Phys. Chem., 69, 211 (1965). 29. L. Pauling, "The Nature of the Chemical Bond", 3rd E d i t i o n , C o r n e l l Univ. Press, New York, 1960, p. 548. 30. F.H. van Cauwelaert and W.K. H a l l , Trans. Faraday S o c , 66, 454 (1970). 31. S.R. Morrison, "The Chemical Physics of Surfaces", Plenum Press, New York, 1977, p. 133. 32. H. Pines and W.O. Haag, J . Am. Chem. S o c , 82, 2471 (1960). 33. H. KnBzinger, Adv. Ca t a l . , 27, 184 (1976). 293 34. G.C. Bailey, V.C.F. Holm and D.M. Blackburn, paper presented before the Div. of Pet. Chem. of the Am. Chem. S o c , A p r i l 7-12, 1957, Miami, F l a . , V o l . 2, No. 1, p. 329. 35. H.A. Benesi, J . Am. Chem. S o c , 78, 5490 (1956). 36. H. Bremer, K.H. Steinberg and K.D.Z. Wendland, Anorg. A l l g . Chem., 366, 30 (1969). 37. J . Dewing, G.T. Monks and B. Y o u l l , J . Catal., 44, 226 (1976). 38. R.M. Pearson, J . Catal . , 46, 279 (1977). 39. T. Morimoto, J . Imai and M. Nagao, J . Phys. Chem., 78, 704 (1974). 40. (a) P. Fink, Rev. Roum. Chim., JL4, 811 (1969). (b) P. Fink, Z. Chem., 7_, 324 (1967). 41. S.J. Gregg and J.D.F. Ramsay, J . Phys. Chem., 73, 1243 (1969). 42. G. D e l i a Gatta, B. Fu b i n i , G. G h i o t t i and C. Morterra, J . C a t a l . 43, 90 (1976). 43. J.W. Hightower and W.K. H a l l , Trans. Faraday Soc., 66, 477 (1970). 44. J.W. Hightower and W.K. H a l l , J . C a t a l . , _13, 161 (1969). 45. M.P. Rosynek and F.L. Strey, J . Ca t a l . , 41, 312 (1976). 46. M.P. Rosynek and J.W. Hightower, P r o c Int. Congr. C a t a l . , 5th, Palm  Beach, F l a . , 1972, 2, 851 (1973). 47. H. Knozinger, Angew. Chem. Int. Ed. Engl., _7> (1968), p. 800-804. 48. A.V. Deo, T.T. Chuang and I.G. D a l l a Lana, J . Phys. Chem., 75, 234 (1971). 49. H. Knozinger and A. S c h e g l i l a , J . C a t a l . , ]J, 252 (1970). 50. C S . John, A. Tada and L.V.F. Kennedy, J . Chem. S o c , Faraday Trans. 1, 24, 498 (1978). 51. (a) W. Pigman and H.S. I s b e l l , Adv. Carbohydr. Chem., 23, 11 (1968). (b) H.S. I s b e l l and W. Pigman, Adv. Carbohydr. Chem., 2A, 14 (1969). 294 52. J.N. Brbnsted and E.A. Guggenheim, J . Am. Chem. Soc., 49, 2554 (1927). 53. A.P. Dubrunfaut, Compt. Rend.. 23, 38 (1846). 54. W. Pigman and D. Horton, i n "The Carbohydrates, Chemistry and Biochemistry", V o l . 1A, 2nd E d i t i o n , W. Pigman and D. Horton, Ed., Academic Press, New York, 1972, p. 33. 55. W. Charlton, W. Haworth and S. Peat, J . Chem. S o c , 89 (1926). 56. T.M. Lowry, J . Chem. S o c , 75, 211 (1899). 57. C S . Hudson, Z. Physik. Chem., 44, 487 (1903). 58. B. Capon, Chem. Reviews 69_ (1969), p. 454-456. 59. R. Bentley and N. Botlock, Anal. Biochem., 20, 312 (1967). 60. C Y . Lee, T.E. Acree and R.S. S.hallenberger, Carbohydr. Res., 9_, 356 (1969). 61. A. Reine, J.A. Hveding, 0. Kj^lb e r g and 0. Westbye, Acta Chem. Scand. B 28, 690 (1974). 62. B.G. Cox and R. Natarajan, J . Chem. S o c Perkin Trans. I I , 2021 (1977). 63. S.W. Benson, "The Foundations of Chemical K i n e t i c s " , McGraw-Hill, New York, 1960, p. 559. 64. R.J. F e r r i e r , and P.M. C o l l i n s , "Monosaccharide Chemistry", Penguin, England, 19 72, Chapter 2. 65. N.M. Ballash and E.B. Robertson, Can. J . Chem., 51, 556 (1973). 66. S.J. Angyal and V.A. P i c k l e s , Carbohydr. Res. , 4., 269 (1967). 67. J.A. Hveding, 0. KjjSlberg and A. Reine, Acta Chem. Scand., 27, 1427 (1973). 68. C C . Sweeley, R. Bentley, M. Makita and W.W. Wells, J . Am. Chem. Soc., 85, 2497 (1963). 69. J.M. Los, L.B. Simpson and K. Wiesner, J . Am. Chem. S o c , 78, 1564 (1956). 295 70. K. Goto and T. Ch i t a n i , B u l l . Chem. Soc. Japan, 16, 403 (1941). 71. D. Rittenberg and C. Graff, J . Am. Chem. S o c , 80, 3370 (1958). 72. E.D. Erdmann, Ber., _13, 218 (1880). 73. F. Urech, Ber., _15, 2130 (1882). 74. T.M. Lowry and E.M. Richards, J . Chem. S o c , 127, 1385 (1925). 75. (a) T.M. Lowry and I . J . Faulkner, J . Chem. S o c , 127, 2883 (1925). (b) T.M. Lowry and G.F. Smith, J . Chem. S o c , 2539 (1927). 76. J.N. Brbnsted, Trans. Faraday S o c , 24, 630 (1928). 77. C S . Hudson, J . Am. Chem. Soc., 29, 1571 (1907). 78. J.M. Los and L.B. Simpson, Rec. Trav. Chim., 73, 941 (1954). 79. A.M. Kjoer, H. Nielsen, P.E. S^rensen and J . Ulstrup, Acta Chem. Scand. A 34, 281 (1980). 80. C C Swain and J.F. Brown J r . , J . Am. Chem. S o c , 74, 2538 (1952). 81. R. Bentley and D.S. Bhate, J . B i o l . Chem., 235, 1219 (1960). 82. W. Broser and G. Ruecker, Z. Naturforsch, 15b, 334 (1960); CA 55: 24195. 83. R. Mitzner and E. Behrenwald, Z. Phys. Chem. (Leipzig ) , 247, 78 (1971); CA 75: 64126 b . 84. P.A. Bobbio, F.O. Bobbio, L.R. Rodrigues and F.G.R. Reyes, An. Acad. Bras. C i e n c , 50, 255 (1978); CA 90: 39134. 85. A.M. Eastham, E.L. B l a c k a l l and G.A. Latremouille, J . Am. Chem. Soc., 77, 2182 (1955). 86. P.R. Rony, W.E. McCormack and S.W. Wunderly, J . Am. Chem. S o c , 91, 4244 (1969). 87. H.S. I s b e l l and C.W.R. Wade, J . Res. Nat. Bur. Stand., 71A, 137 (1967). 88. K.B. Wiberg, Chem. Reviews, _55, 713 (1955). 89. C. C h a l l i s , F.A. Long and Y. Pocker, J . Chem. S o c , 4679 (1957). 90. J.A. Christiansen, J . C o l l o i d Interface S c i . , 22, 1 (1966). 296 91. H.S. I s b e l l , H.L. Frush, C.W.R. Wade and C E . Hunter, Carbohydr. Res., 9_, 163 (1969). 92. W.H. Hamill and V.K. La Mer, J . Chem. Phys. , 4_> 3 9 5 (1936). 93. B. Capon and R.B. Walker, J . Chem. Soc. Perkin Trans. I I , 1600 (1974). 94. J.P. Horwitz and C A . Easwaran, J . Org. Chem., 33, 3714 (1968). 95. H.H. Huang, A.N.H. Yeo and L.H.L. Chia, J . Chem. Soc. (B), 836 (1969). 96. T.M. Lowry, J . Chem. S o c , 2554 (1927). 97. C.G. Swain and J.F. Brown, J . Am. Chem. Soc., 74, 2534 (1952). 98. Y. Pocker, Chem. and Ind., 968 (1960). 99. R.L. Schowen, Progr. Phys. Org. Chem., 9_, 313 (1972). 100. P.R. Rony, J . Am. Chem. S o c , 91, 6090 (1969). 101. P.R. Rony, J . Am. Chem. S o c , 90, 2824 (1968). 102. R.P. B e l l , "Proton i n Chemistry", 2nd E d i t i o n , Chapman and H a l l , London, 1973, Chapters 8 and 9. 103. W.P. Jencks, "C a t a l y s i s i n Chemistry and Enzymology", McGraw-Hill, New York, 1969, p. 201. 104. V. Fiandanese and F. Naso, J . Chem. Soc. Perkin Trans. I I , 1047 (1977). 105. C S . Chin and H.H. Huang, J . Chem. Soc. Perkin Trans. I I , 474 (1978). 106. M. Eigen, Disc. Faraday Soc., 39, 7 (1965). 107. R.P. B e l l , J.P. M i l l i n g t o n and J.M. Pink, Proc. Roy. S o c , A, 303, 1 (1968). 108. K. Tanabe, A. Nagata and T. Takeshita, J . Res., Inst. C a t a l . , H o k k a i d o Univ., JL_5, 181 (.1967); CA 69: 44126. 109. Kh. R. Rustamov and E.G. Tyrina, Uzbeksk. Khim. Zh., No. 6, 28 (1961); CA 56: 12356. 110. Kh. R. Rustamov and R.U. Usmanov, Uzbeksk. Khim. Zh., j), No. 2, 36 (1962); CA 57: 6660. 297 111. H. Schmid and G. Bauer, Monatsh. Chem., 96, 1510 (1965); CA 64: 9805. 112. D.L. Kantro, S. Brunauer and L.E. Copeland i n "The Solid-Gas Interface", V o l . 1, E.A. Flood, Ed., Marcel Dekker, New York, 1967, Chapter 12. 113. S.J. Gregg, "The Surface Chemistry of S o l i d s " , 2nd E d i t i o n , Chapman and H a l l , 1961. 114. J.M. Thomas and W.J. Thomas, "Introduction to the P r i n c i p l e s of Heterogeneous C a t a l y s i s " , Academic Press, 1967, Chapter 4. 115. B.G. Linsen arid A. van den Heuvel i n "The Solid-Gas Interface", V ol. 2, E.A. Flood, Ed., Marcel Dekker, New York, 1967, Chapter 35. 116. D.H. Everett i n "The Solid-Gas Interface", V ol. 2, E.A. Flood, Ed., Marcel Dekker, New York, 1967, Chapter 36. 117. Z.G. Szabo and B. Jover, Proc. 5th Intern. Congr. C a t a l . , F l o r i d a , 1972, J. , 833 (1973). 118. H.R. Mahler and E.H. Cordes, " B i o l o g i c a l Chemistry", Harper and Row, New York, 1966, p. 234-236. 119. A.S. H i l l and R.S. Schallenberger, Carbohydr. Res., 11, 541 (1969). 120. J.M.H. Dirkx and H.S. van der Baan, J . Cat a l . , 67, 1 (1981). 121. H.P. Boehm, Adv. C a t a l . , 16, 179 (1966). 122. G.C. Bond, "Heterogeneous C a t a l y s i s : P r i n c i p l e s and App l i c a t i o n s " , Clarendon Press, Oxford, 19 74, p. 43. 123. S. Brunauer, L.E. Copeland and D.L. Kantro i n "The Solid-Gas Interface", Vol. 1, E.A. Flood, Ed., Marcel Dekker, New York, 1967, Chapter 3. 124. A.W. Adamson, "Physical Chemistry of Surfaces", 4th E d i t i o n , John Wiley, New York, 1982, Chapter XI. 125. (a) R.J. Kokes and A.L. Dent, Adv. C a t a l . Relat. Subj., 22, 1 (1972). (b) R.J. Kokes, Proc. 5th Intern. Congr. Cata l . , F l o r i d a , 1972, _1 A l (19 73). 298 126. (a) R.W. Maatman, C a t a l . Rev.. 8, 1 (1973). (b) R.W. Maatman, J . C a t a l . , _19_, 64 (1970). 127. B. H e l f e r i c h and W. Schafer, Org. Syn., C o l l . V o l . 1, 2nd E d i t i o n , p. 364 (1961). 128. "ICN Chromatography Products", ICN Pharmaceuticals Inc., L i f e Sciences Group, Ohio, 1977-78. 129. H. Brockmann and H. Schodder, Ber. Dtsch. Chem. Ges., 74 B , 73 (1941). 130. D. Dolman, Ph.D. Thesis, The University of B r i t i s h Columbia, Vancouver, B.C., Canada, 1966, p. 30. 131. (a) Th. J . de Boer and H.J. Backer, Rec. Trav., 73. 229 (1954). (b) M. Fieser and L. Fi e s e r , "Reagents for Organic Synthesis", John Wiley, New York, 1967, p. 191. 132. (a) C. Horvath i n "The P r a c t i c e of Gas Chromatography", L.S. E t t r e and A. Z l a t k i s , Ed., Interscience, New York, 1967, p. 191-192. (b) C. Eaborn, "Organosilicon Compounds", Butterworths, London, 1969. 133. A.F. Thomas, "Deuterium Labeling i n Organic Chemistry", Appleton-Century-Crofts, New York, 19 71, p. 7-8. 134. G.R. Dean and J.B. Go t t f r i e d , Adv. Carbohydr. Chem., _5_, (1950), p. 135-136. 135. H.S. I s b e l l and C.W.R. Wade, J . Res. Nat. Bur. Stand., 71A (1967), p. 146. 136. S. Brunauer, P.H. Emmett and E. T e l l e r , J . Am. Chem. Soc., 60, 309 (1938). 137. D.L. Kantro, S. Brunauer and L.E. Copeland i n "The Solid-Gas Interface", V o l . 1, E.A. Flood, Ed., Marcel Dekker, New York, 1967, p. 412-418, 425-426. 299 138. "I n s t r u c t i o n Manual f o r the Quanta-Sorb Sorption System", Quanta-Chrome Corporation, Greenvale, New York, 1969. 139. B.G. Linsen and A. van den Heuvel i n "The Solid-Gas Interface", Vol. 2, E.A. Flood, Ed., Marcel Dekker, New York, 1967, p. 1029-1036. 140. J.H. de Boer i n "The Structure and Properties of Porous Materials", D.H. Everett and F.S. Stone, Ed., Butterworth, Colston Papers, Vol. 10, London, 1958, p. 68. 141. R. Zsigmondy, Z. Anorg. Allgem. Chem., 71, 356 (1911). 142. (a) B.C. Lippens, B.G. Linsen and J.H. de Boer, J . C a t a l . , 3, 32 (1964). (b) J.H. de Boer, B.G. Linsen and T.J. Osind, J . C a t a l . , _4, 643 (1965). (c) R.W. Cranston and F.A. Inkley, Adv. Ca t a l . , 9_, 143 (1957). 143. S. Brunauer, L.S. Deming, W.E. Deming and E. T e l l e r , J . Am. Chem. Soc., 62, 1723 (1940). 144. R.H. Berg i n "Symposium on P a r t i c l e Size Measurement", Special Technical P u b l i c a t i o n No. 234, American Society for Testing Materials, 1958. 145. "Operation and Description of Computerized Electrozone Celloscope", P a r t i c l e Data Inc., I l l i n o i s , 1977. 300 APPENDIX A CHARACTERIZATION OF ALUMINAS A.1 Determination of BET Surface Areas A.1.1 Theory Surface areas of the aluminas were determined using the two parameter 123,136,137 BET equation 1 c - 1 — + x A I N w( 1) wo wo x m m where w^ = weight of vapour required to cover the surface by a complete monolayer, w = weight of vapour adsorbed when adsorbate r e l a t i v e pressure x = p/PQ (where p = adsorbate p a r t i a l pressure, and p Q = adsorbate saturated vapour pressure), and parameter c i s a constant f o r a great majority of 123 137 vapour adsorption isotherms from x = 0.05 to about 0.35 ' Therefore from a plo t of — versus x i n th i s range of r e l a t i v e w(- - 1) x pressures the weight of adsorbate necessary to form a complete monolayer w i s given by — ——7 , and the surface area of the sample i s m 0 J slope + intercept r 11? x (6.02 x 10 2 3) x A Yfl OS given by , where M i s the molecular weight of the adsorbate and A i s the 'cross-sectional' area of the adsorbate molecule cs 2 113 (for nitrogen A = 16.2 A per molecule ). A.1.2 Experimental Procedure To determine the quantity of vapour adsorbed (w) on a sample of alumina at d i f f e r e n t r e l a t i v e pressures (x) , a Quanta-Sorb instrument 301 (manufactured by Quanta-Chrome Corp., N.Y.) was used. Nitrogen gas was chosen as adsorbate. Using the Quanta-Sorb, nitrogen gas was adsorbed on a preweighed sample of alumina, kept at l i q u i d nitrogen temperature, from a flowing mixture of nitrogen (x = 0.15) and an i n e r t non-adsorbable c a r r i e r 137 138 gas (helium) ' . The amount of nitrogen adsorbed was monitored by measuring the change i n thermal conductivity of the gas mixture. A change i n the thermal conductivity produced a peak i n the recorder, and i t was c a l i b r a t e d with a known volume of nitrogen. By comparing the areas of the adsorption and c a l i b r a t i o n signals the weight of nitrogen adsorbed (w) at x = 0.15 was determined. The procedure was repeated at higher r e l a t i v e pressures of upto x = 0.35. From the plots of i versus x for d i f f e r e n t samples of alumina t h e i r BET surface areas were determined as described above. They are tabulated i n Table IV i n Section II.1. A.2 Determination of Pore Size D i s t r i b u t i o n s A.2.1 Theory I t has been observed that adsorption of vapours on a s o l i d most often gives r i s e to multilayer adsorption. However, i f the s o l i d i s porous, i n addition to multilayer adsorption, c a p i l l a r y condensation also takes place and adsorption—desorption isotherms often exhibit h y s t e r e s i s ^ 1 3 1 1 ^ . That i s , i n a c e r t a i n region of the isotherm, which i s determined by the range of pore r a d i i present, the adsorption isotherm does not follow the desorption isotherm. In addition, from the shape of hysteresis loop the shape of the r. J • i . , , , . ,. . 139,140 , c pores can be determined using de Boer s c l a s s i f i c a t i o n , and from the desorption isotherm the pore s i z e d i s t r i b u t i o n can be determined, using 302 , . j i j . 113,141 , _, . . . . 1 1 3 Zsigmondy s c a p i l l a r y condensation theory and K e l v i n equation , as outlined below. According to the c a p i l l a r y condensation theory, adsorbed vapour i s condensed to the ordinary l i q u i d condition i n the pores of the adsorbent. The Kelvin equation for adsorption of n i t r o g e n * 3 8 , -4.146 ? log p/pf rk = A o gives the maximum radius of the pore where c a p i l l a r y condensation of nitrogen occurs at a r e l a t i v e pressure of nitrogen = pip ' "The K e l v i n radius of the pore i s not the actual pore radius because some adsorption has taken place on the w a l l of the pore p r i o r to the occurrence of condensation i n the pore, or during desorption an adsorbed layer remains on the w a l l a f t e r evaporation has occurred. Hence the actual pore radius i s given by P fe where t i s the thickness of the adsorbed layer at the r e l a t i v e pressure plpQ' Values for t at d i f f e r e n t r e l a t i v e pressures of nitrogen have been 142 determined from studies i n v o l v i n g non-porous s o l i d s If the pores are assumed to be c y l i n d r i c a l and i f the r e l a t i v e pressure i s changed from p^V0 T O P ] / P o t n e n P o r e s between r a d i i a n d w i l l empty (p^p^, r >r^). When p 2 i s lowered to p^ the thickness of the adsorbed f i l m on previously emptied pores changes from to . As a r e s u l t , nitrogen gas gets desorbed from the surface and from the volume of 3 3 gas desorbed (^g a s c m )> corresponding volume of l i q u i d (^^q c m ) 1 S 303 determined. The r e l a t i o n between the volume of pores (V ) having r a d i i 138 139 between and r^t and i s S i v e n b ? ' » V = P V. . - AtlA l i q cm3 (57) where v and r, (in A) are the average pore and K e l v i n r a d i i i n the range ~p K o y>2 to v^, At = (t - t^) A, and ZA i s the t o t a l area of adsorbed f i l m remaining i n the previously empty pores and the pores with r a d i i between v and a f t e r evaporation out of the pores have occurred. For c y l i n d r i c a l pores, the area of the pores having r a d i i between v and i s given by 2 V Xi A 2 3 ° A = — - i q x 10 m , with V. . i n cm and r i n A. ^ p i i q P Therefore from 7, . , the volume of l i q u i d nitrogen desorbed when the lxq r e l a t i v e pressure of nitrogen i s decreased from V^^o t o Pl^Po' t* i e ^ l 1 ™ 6 of pores V with r a d i i between and can be determined using Equation (57) A.2.2 Experimental Procedure The adsorption isotherms for standard alumina neutral, alumina dried at 800°C and alumina sintered at 1250°C were determined by continuing the procedure used to obtain the BET plo t i n Appendix A . l , from p/pQ - 0.35 to 0.94. 113 138 To obtain the desorption isotherm ' for alumina neutral, pure adsorbate ( i . e . nitrogen) was allowed to flow through the sample immersed i n l i q u i d nitrogen. Then gas mixture, at the required r e l a t i v e pressure (0.94), was allowed to pass through the sample c e l l while s t i l l immersed i n the l i q u i d nitrogen. When the recorder indicated a s t r a i g h t l i n e the l i q u i d nitrogen was removed and the desorption s i g n a l was monitored. It was ca l i b r a t e d using a known volume of nitrogen and gave a point on the 304 desorption isotherm corresponding to the r e l a t i v e pressure of nitrogen = 0.94. The procedure was repeated at progressively lower r e l a t i v e pressures of nitrogen. Thus the desorption isotherm for alumina neutral was obtained. The desorption isotherms for 800° alumina and alumina sintered at 1250°C were also obtained by repeating the procedure described above. A.2.3 Results and Discussion The adsorption and desorption isotherms for alumina neutral, 800° alumina and alumina sintered at 1250°C are shown i n F i g s . 55, 56 and 57, r e s p e c t i v e l y , and i t i s c l e a r that a l l three samples show multilayer adsorption. The sintered sample shows Type II adsorption (according to the 143 c l a s s i f i c a t i o n by Brunauer, Deming, Deming and T e l l e r ), where only multilayer adsorption takes place, while both alumina neutral and 800° alumina sample show Type IV isotherm where both multilayer adsorption and c a p i l l a r y condensation occur. This shows that alumina neutral and 800° alumina are both porous while s i n t e r i n g at 1250°C has removed the porous texture of the c a t a l y s t . The presence of pores i n alumina neutral and 800° alumina has also given r i s e to hysteresis of adsorption and desorption isotherms (Figs. 55 and 56), while the absence of pores i n sintered alumina i s indicated by the observation that there i s no hysteresis of adsorption and desorption isotherms (Fig. 57). In a d d i t i o n , the shapes of hysteresis loops i n Figs. 55 and 56 may be taken to i n d i c a t e that the pores are c y l i n -. . ,139,140 drxcal Using the theory outlined above (Appendix A.2.1), the pore volumes for d i f f e r e n t i n t e r v a l s of r e l a t i v e pressure were determined. The plots of t\V ( r a t i o of pore volume i n each i n t e r v a l to the change i n pore radius) versus r (the mean pore radius for the same i n t e r v a l ) for alumina neutral 09 Ln Ln t cn O i-i T3 rt H-O 3 3 Pu a fD CD O l-i rt H-O 3 CO o rt 3" rt) >t 3 cn H O era to 3 o 3 c 3 H-3 03 3 rt) C rf i-i 0) Weight (x 10 g) of Nitrogen Adsorbed per Gram of Alumina •X3 soe F i g . 56 Adsorption ( X ) and Desorption ( O) Isotherms of Nitrogen 800°C Alumina w o CTN 3 0 7 308 and 800° alumina are shown i n F i g . 3 (Section I I . 2 ) . A.3 Determination of P a r t i c l e Size D i s t r i b u t i o n s A.3.1 Theory The p a r t i c l e s i z e d i s t r i b u t i o n s of alumina samples were determined by the e l e c t r o l y t e sensing zone method using an Electrozone Celloscope. This instrument determines the number and s i z e of p a r t i c l e s suspended i n an e l e c t r i c a l l y conductive l i q u i d by a p p l i c a t i o n of a resistance p r i n c i p l e . The p r i n c i p l e consists of f o r c i n g the suspension to flow through a small 144 aperture having an immersed electrode on each side as shown i n F i g . 58 F i g . 58 Basic Mechanism of the P r i n c i p l e of 144 Electrozone Celloscope As each p a r t i c l e passes through the aperture, i t replaces i t s own volume of e l e c t r o l y t e w i t h i n the aperture, momentarily changing the e l e c t r i c a l resistance between the electrodes. To keep the current constant, the Electrozone Celloscope produces a voltage pulse of short duration having a 309 magnitude proportional to the p a r t i c l e volume. The seri e s of pulses generated by the p a r t i c l e s i s e l e c t r i c a l l y amplified, scaled, and counted. Measurement p r e c i s i o n with 1% on diameter basis has been commonly experi-144 enced . P a r t i c l e density does not a f f e c t response but, where gross p a r t i c l e porosity e x i s t s , the pores aligned with' the aperture axis may provide a degree of e l e c t r i c a l conductivity with proportionately l e s s e r pulse height. The p a r t i c l e shape and structure have l i t t l e e f f e c t on the response. During the experimental process, a g i t a t i o n (short of a i r bubble inc l u s i o n ) and chemical dispersants are used to maintain a uniform suspension, avoid f l o c c u l a t i o n of the p a r t i c l e s , or prevent adherence to the sample container. Low p a r t i c l e concentrations (10 to 100 ppm by volume) are used to avoid the passage of two p a r t i c l e s through the o r i f i c e at the 144 same time (coincidence p a r t i c l e problem) A.3.2 Experimental Procedure Electrozone Celloscope Model 112 LTS/ADC, connected to a PDP 8/m minicomputer with 8K capacity and to a telequipment o s c i l l o s c o p e 551B and an ASR33 teletype, was used. Alumina samples were mixed for 3 hours each on a m i n i m i l l . Random spatula cuts of the samples were taken and placed i n a blender with a 10% 145 sodium hexametaphosphate (Calgon) s o l u t i o n , which was s t i r r e d vigorously for 30 sees. D i s t i l l e d water was added to reduce the phosphate concentration to 4%. To ensure representative sampling, a scoopful of the mixture was taken while s t i r r i n g r a p i d l y . The sample was then d i l u t e d with a 0.75% sodium chloride - 0.5% sodium tetrapyrophosphate e l e c t r o l y t e to proper l e v e l s ( p a r t i c l e concentration =50 ppm by volume) for t e s t i n g on the 310 Electrozone Celloscope. The blank e l e c t r o l y t e was f i r s t checked for background count (none) and then the sample was analyzed using an o r i f i c e tube with an o r i f i c e diameter = 150 microns. Plots of the number of counts (scaled) versus p a r t i c l e diameter i n centimicrons, for standard alumina neutral and alumina sintered at 1250°C are given i n F i g . 4 and the c h a r a c t e r i s t i c s of the plots i n Table V i n Section II.3.1. Determination of the c r y s t a l l i n e structure and the trace impurities present i n aluminas are described i n the Experimental Section XXI. 311 APPENDIX B PREPARATION OF A BATCH OF ALUMINA WHICH ACTIVATES ON PYROLYSIS AT HIGH TEMPERATURES, SIMILAR TO THE FIRST BATCH OF ALUMINA NEUTRAL It was observed i n Section XII that, the f i r s t batch of Woelm alumina neutral was activated r a p i d l y on p y r o l y s i s at temperatures above 800°C (Figs. 27 and 28). However, the second batch of Woelm alumina n e u t r a l and alumina neutral purchased from BDH Chemicals were progressively deactivated on further heating above 800°C (Fig. 29). I t was also observed that, Woelm alumina basic on p y r o l y s i s at 1250°C produced a highly active c a t a l y s t (Fig. 29). This suggested that the somewhat s i m i l a r behaviour observed with the f i r s t batch of alumina neutral may be due to the presence of re s i d u a l b asic character i n that batch of neutral alumina. Therefore, preparation of a batch of alumina which behaves s i m i l a r to the f i r s t batch of alumina neutral was attempted by treatment of the second batch of alumina neutral with a base and also by p a r t i a l n e u t r a l i z a t i o n of alumina basic with an acid, as described below. B.1 Treatment of the Second Batch of Alumina Neutral with a Base A s t i r r e d s l u r r y of alumina neutral (second batch) was treated with 0.1 N NH.OH u n t i l about 20% of acid s i t e s (determined i n Section XVIII.2.3) 4 were ne u t r a l i z e d . The alumina was f i l t e r e d under suction and pyrolyzed at 1250°C f or 6 hours. However, the sintered sample thus obtained was found to possess very low c a t a l y t i c a c t i v i t y . The procedure was repeated using 0.1 N NaOH as the base and again,a c a t a l y s t with very low a c t i v i t y was obtained. 312 B.2 P a r t i a l N e u t r a l i z a t i o n of Alumina Basic for TLC During the attempts to produce highly active 1250° aluminas i t was noted that,when alumina basic f or t i c was heated at 1250°C, the quartz tube used to hold the boat containing alumina v i t r i f i e d during the process. Such v i t r i f i c a t i o n was not observed when other aluminas were heated at the same temperature. Examination of compositions of aluminas given i n Experi-mental Section XXI shows that the main difference between alumina basic and other chromatographic aluminas i s the presence of a large amount of sodium i n the former. The v i t r i f i c a t i o n observed on s i n t e r i n g may be due to vapourization of sodium oxide (which sublimes at 1275°C) and/or hydroxide ( b o i l i n g point 1390°C) from the basic alumina at 1250°C. Hence removal of the excess ( a l k a l i ) metal ions from alumina basic by leaching with an acid may ultimately enable production of a le s s basic alumina which behaves l i k e the f i r s t batch of alumina n e u t r a l . Removal of metal ions was c a r r i e d out by s t i r r i n g 20 ml of 0.1 M hydrochloric acid s o l u t i o n containing 6 g of alumina basic f or 12 hours. The s l u r r y was f i l t e r e d on a Buchner glass funnel with a f r i t t e d disc (medium) and the alumina was washed with excess d i s t i l l e d water for several hours. The water coming through the funnel was very s l i g h t l y a c i d i c to litmus. The alumina sample i n the funnel was washed with more water u n t i l the water coming through was no longer a c i d i c . The alumina was allowed to dry under suction, and was then heated at 1250°C under nitrogen f or 3 hours (no v i t r i f i c a t i o n of the quartz tube was observed). The sintered sample showed high c a t a l y t i c a c t i v i t y (but le s s than that of sintered alumina basic) with l i n e a r f i r s t order pl o t s as shown i n F i g . 59. A second batch of acid washed alumina basic was prepared and was dried in vacuo for two days. I t was divided into three parts. One was 313 0 50 100 150 Time (min) Fi g . 59 C a t a l y t i c A c t i v i t i e s of D i f f e r e n t Aluminas Prepared from Acid Treated Basic Alumina 314 heated at 800°C under nitrogen f o r 4 hours, and another part was heated at 1250°C under nitrogen f o r 6 hours. F i r s t order plots obtained with the three samples are also shown i n F i g . 59, and i t i s clear that on p y r o l y s i s , t h i s acid treated alumina b a s i c behaves s i m i l a r to that of the f i r s t batch of alumina n e u t r a l . The small differences i n the observed c a t a l y t i c a c t i v i t i e s are to be expected since i t i s r e l a t e d to the amount of acid used and i t s contact time. Another batch was prepared by t i t r a t i n g a s l u r r y of basic alumina with 0.1 M hydrochloric acid u n t i l the pH was about 7.5, the pH of a s l u r r y of alumina neutral i n d i s t i l l e d water (see Table V I I I ) . I t was f i l t e r e d , washed and heated at 1250°C under nitrogen f o r 6 hours. The sintered alumina again showed a l i n e a r f i r s t order p l o t and high a c t i v i t y , although a l i t t l e less than that of the two previous batches. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0060700/manifest

Comment

Related Items