UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Kinetics of gasification and sulphur capture of oil sand cokes Nguyen, Quoi The 1988

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

Item Metadata

Download

Media
831-UBC_1988_A1 N48.pdf [ 19.95MB ]
Metadata
JSON: 831-1.0058890.json
JSON-LD: 831-1.0058890-ld.json
RDF/XML (Pretty): 831-1.0058890-rdf.xml
RDF/JSON: 831-1.0058890-rdf.json
Turtle: 831-1.0058890-turtle.txt
N-Triples: 831-1.0058890-rdf-ntriples.txt
Original Record: 831-1.0058890-source.json
Full Text
831-1.0058890-fulltext.txt
Citation
831-1.0058890.ris

Full Text

KINETICS OF GASIFICATION AND SULPHUR CAPTURE OF OIL SAND COKES by QUOI THE NGUYEN B.Eng. National I n s t i t u t e of Technology, VietNam, 1977 B.A.Sc Chemical Engineering, UBC, 1984 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES Chemical Engineering We accept t h i s thesis as conforming, to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1988 © Quoi The Nguyen, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of C.U&Mt'CAt- f=M&JAl f=^,'jJ& The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DF-fin/ft-M ABSTRACT: K i n e t i c s of steam g a s i f i c a t i o n of both delayed and f l u i d cokes, byproducts from thermal cracking processes of Athabasca bitumen, have been s tudied in l a b o r a t o r y - s i z e s t i r r e d and f ixed bed r e a c t o r s . The hydrogen sulphide i n the product gas was captured i n - s i t u using c a l c i n e d dolomite and limestones as acceptors . E x p e r i m e n t s were c a r r i e d out at a t m o s p h e r i c pressure and at temperatures between 8 0 0 ° C and 9 3 0 ° C . The coke p a r t i c l e s i z e ranged from 0.1 to 3.5 mm, and the steam p a r t i a l pressure was v a r i e d from 15.15 to 60.6 kPa. The carbon and sulphur convers ions were computed from the knowledge of gas c o m p o s i t i o n s and f l o w r a t e s and the g a s i f i c a t i o n k i n e t i c s of both species e s t a b l i s h e d . The e f f ec t s of sorbent type, p a r t i c l e s i z e , c a l c i n a t i o n c o n d i t i o n s , and Ca/S molar r a t i o s on the extent of sulphur capture during g a s i f i c a t i o n were examined i n separate s e r i e s of experiments. Scanning e l ec tron microscopy, surface area a n a l y s i s , and mercury porosimetry were employed to r e l a t e p h y s i c a l s t ruc ture changes i n the s o l i d s to experimental k i n e t i c da ta . The rate of g a s i f i c a t i o n f o r the delayed coke was genera l ly higher than that f o r the f l u i d coke, and both cokes were almost unreact ive to steam g a s i f i c a t i o n at temperatures below 8 0 0 ° C . Increased r e a c t i o n temperatures or reduced p a r t i c l e s i zes increased both carbon as w e l l as s u l p h u r c o n v e r s i o n . The carbon convers ion rates were found to go through maxima as the time of r e a c t i o n and ex ten t of c o n v e r s i o n i n c r e a s e d . As the reac t ion proceeded the surface area of the coke increased to a maximum of about f i v e times i t s i n i t i a l value and then f e l l o f f sharp ly . The extent of carbon conversion alone d i c ta ted the s p e c i f i c surface area i r r e s p e c t i v e of temperature, p a r t i c l e s i ze and steam p a r t i a l pressure . B o t h c a l c i n e d dolomite and c a l c i n e d limestone were found to be e f f e c t i v e i n removing sulphur from the product gas. Sorbents possessing a l a r g e r s p e c i f i c area or smal ler g r a i n s i ze had higher capaci ty to accept su lphur . At a Ca/S molar r a t i o of 2 .0 , the o v e r a l l sulphur removal was a p p r o x i m a t e l y 90% f o r the f i r s t 3 h r s and the H 2 S concentrat ion i n the produced gas was reduced to about 200 ppm from n e a r l y 1250 ppm. The r a t e of sorbent conversion from CaO to CaS decreased monotonical ly with t ime. Three a v a i l a b l e k i n e t i c models f o r g a s i f i c a t i o n - the Random C a p i l l a r y Model , the Random Pore Model and the M o d i f i e d V o l u m e t r i c Model, were tested wi th the experimental g a s i f i c a t i o n data . Although reasonable f i t s were obtained f o r X^- t r e s u l t s , the sharp drop i n rate at h igh conversion could not be adequately modelled. Rate constants were e s tab l i shed f o r the i n i t i a l stage of r e a c t i o n on ly . The G r a i n model and Continuous r e a c t i o n models were tested with the 1 e x p e r i m e n t a l s u l p h i d a t i o n r e s u l t s . The s u l p h i d a t i o n process was c o n t r o l l e d by c h e m i c a l r e a c t i o n at low s o r b e n t c o n v e r s i o n , and s u b s e q u e n t l y by d i f f u s i o n t h r o u g h the p r o d u c t l a y e r at h i g h e r convers ions . The r e a c t i o n rate constant and the e f f e c t i v e d i f f u s i v i t y were a c c o r d i n g l y e s t a b l i s h e d as func t ions of temperature. Values compared favourably with r e s u l t s of s u l p h i d a t i o n k i n e t i c s done without simultaneous g a s i f i c a t i o n reported i n the l i t e r a t u r e . i i i The r e s u l t s suggest that the g a s i f i c a t i o n process and the sulphur capture p r o c e s s , which o c c u r t o g e t h e r i n g a s i f i e r s w i t h s o r b e n t i n j e c t i o n , can be treated independently. Indexing terms: G a s i f i c a t i o n , Carbon Conversion, Sulphur Conversion, Su lphur Removal , C a l c i n e , L i m e s t o n e , D o l o m i t e , Hydrogen Sulphide , Su lph ida t ion . i v ACKNOWLEDGEMENTS: I am g r a t e f u l to my superv i sor , Prof . A . P . Watkinson, for h i s guidance and i n t e r e s t in t h i s research . I am also indebted to many others who helped during the course of t h i s work: To the Natura l Science and Engineering Research Counc i l of Canada which provided f i n a n c i a l support i n the form of research grants . To Energy, Mines and Resources Canada for t h e i r f i n a n c i a l support. To the U n i v e r s i t y of B r i t i s h Columbia which contr ibuted f i n a n c i a l support in the form of Graduate Fe l lowships . To the Department of Metals and M a t e r i a l s Engineer ing , U . B . C . , for the use of e l e c t r o n microscope. To the Department of Mining and Mineral Process Engineering for the use of the Quantasorb surface area analyzer . To D r . S. B a r t o n at the Royal M i l i t a r y Co l l ege , Kingston for measuring p o r o s i t y and pore s ize d i s t r i b u t i o n of s o l i d samples. To Stee l Bros . L t d . for donating the Cache Creek l imestone. S p e c i a l thanks are due to my wife , Dieu , and my fami ly for t h e i r encouragement, understanding and pat ience . v TABLE OF CONTENTS ABSTRACT . . . . . . . . . i i ACKNOWLEDGEMENTS v LIST OF FIGURES ix LIST OF TABLES x v i LIST OF SYMBOLS x i x INTRODUCTION AND OBJECTIVES 1 1.0 LITERATURE REVIEW 5 2.1 OIL SANDS 5 2.2 OIL SAND COKES 8 2.3 GAS EOROUS-SOLID REACTIONS 17 2.4 GASIFICATION 19 2.4.1 WATER-GAS SHIFT REACTION 22 2.4.2 EFFECTS ON GASIFICATION REACTIONS 22 2.4.3 KINETIC MODELS FOR GASIFICATION 26 2.4.4 RANDOM PORE MODEL 32 2.4 .5 RANDOM CAPILLARY MODEL 39 2.4.6 MODIFIED VOLUMETRIC MODEL 47 2.5 DESULPHURIZATION . 49 2.5.1 SORBENTS 49 2.5.2 CALCINATION 52 2.5.3 ESTIMATION OF e, p and S p 61 2.5.4 THE SULPHIDATION REACTIONS 64 2 .5 .5 RELATIVE REACTIVITIES OF SORBENTS 71 2.5.6 GRAIN MODEL 73 2.5.7 KINETIC DATA ON THE SULPHIDATION PROCESS 81 2.5 .8 CONTINUOUS FIRST-ORDER MODEL 88 2.6 POROSIMETRY - BET SURFACE MEASUREMENT - SCANNING ELECTRON MICROSCOPY 89 2.6.1 MERCURY POROSIMETRY . . 89 2.6.2 BET SURFACE AREA MEASUREMENTS 91 3.0 EXPERIMENTAL PROCEDURES AND APPARATUS 94 3.1 INTRODUCTION 94 3.2 MATERIALS, SAMPLE PREPARATION AND CHARACTERIZATION . . . 95 3.2.1 MATERIALS 95 3.2.2 SAMPLE PREPARATION AND CHARACTERIZATION . . . . . 97 3.2.3 SAMPLE CHARACTERIZATION 98 v i 3.3 EXPERIMENTAL APPARATUS 100 3.3.1 THE CALCINER 100 3.3.2 THE GASIFICATION SYSTEM 103 3 .3 .2 .1 WATER AND N 2 TRANSPORT 103 3 .3 .2 .2 THE PREHEATER 105 3 .3 .2 .3 THE GASIFIER 106 3 .3 .2 .4 THE PRODUCT GAS HANDLING FACILITY . . . . 110 3 .3 .2 .5 THE CONTROL PANEL 113 3.4 EXPERIMENTAL TECHNIQUES 114 3.4.1 CALCINATION 114 3.4.2 GASIFICATION 118 3.4.3 SURFACE AREA MEASUREMENTS AND S . E . M 121 3.4 .4 ANALYSES 123 4.0 EXPERIMENTAL RESULTS IN GASIFICATION 127 4.1 GASIFICATION OF DELAYED COKE 127 4 .1 .1 EFFECT OF MECHANICAL MIXING 127 4 .1 .2 EXPERIMENTS AT DIFFERENT REACTION TEMPERATURES . . 132 4 .1 .2 .1 EXPERIMENTAL RESULTS 132 4 .1 .2 .2 DISCUSSION 137 4.1 .3 EXPERIMENTS WITH DIFFERENT COKE PARTICLE SIZES . . 143 4 .1 .3 .1 EXPERIMENTAL RESULTS 143 4 .1 .3 .2 DISCUSSION 150 4 .1 .4 ANALYSIS OF UNREACTED & PARTIALLY REACTED COKE . . 154 4 .1 .4 .1 B . E . T . SURFACE AREA MEASUREMENT 155 4 .1 .4 .2 EXAMINATIONS USING SCANNING ELECTRON MICROSCOPE 158 4 .1 .4 .3 POROSIMETRY 158 4 .1 .4 .4 E L E M E N T A L A N A L Y S I S OF REACTED COKE SAMPLES 163 4 .1 .4 .5 DISCUSSIONS 166 4 .1 .5 REPRODUCIBILITY OF EXPERIMENTAL DATA 170 4.2 GASIFICATION OF FLUID COKE 171 4 .2 .1 EXPERIMENTS WITH DIFFERENT REACTION TEMPERATURES . 174 4 .2 .1 .1 EXPERIMENTAL RESULTS 174 4 .2 .1 .2 S . E . M . EXAMINATION OF F . C 180 4 .2 .1 .3 DISCUSSIONS 183 4 .2 .2 EFFECT OF PH20 185 4 .2 .2 .1 EXPERIMENTAL RESULTS 185 4 .2 .2 .2 DISCUSSIONS 190 4.3 COMPARISONS BETWEEN GASIFICATION OF D . C . AND F . C 192 4.3.1 COMPARISON OF CARBON CONVERSION 193 4.3.2 COMPARISON OF H 2 S EMISSIONS 195 4 .3 .3 COMPARISON OF SULPHUR CONVERSIONS 197 4 .3 .4 COMPARISON OF THE RATES OF CARBON CONVERSION . . . 197 4 .3 .5 DISCUSSION 199 4.4 SUMMARY OF GASIFICATION RESULTS 203 5.0 DESULPHURIZATION 205 5.1 EFFECT OF SORBENT ADDITION ON CARBON CONVERSION 206 5.2 EFFECT OF MIXING 209 5.3 EFFECT OF CALCINATION CONDITIONS 211 v i i 5.3.1 EXPERIMENTAL RESULTS 211 5.3.2 DISCUSSIONS 226 5.4 EFFECTS OF Ca/S RATIOS 231 5.4.1 EXPERIMENTAL RESULTS 232 5.4.2 DISCUSSION 239 5.5 EFFECTS OF SORBENT PARTICLE SIZE 241 5.5.1 EXPERIMENTAL RESULTS 241 5.5.2 DISCUSSION 248 5.6 EFFECT OF TYPES AND GEOLOGICAL ORIGINS 250 5.6.1 CALCINES OF DIFFERENT ORIGINS 250 5.6.2 CALCINES OF DIFFERENT TYPES 251 5.7 SUMMARY 258 6.0 TESTING KINETIC MODELS 260 6.1 TESTING MODELS USING GASIFICATION EXPERIMENTAL RESULTS . 260 6.1.1 RANDOM CAPILLARY MODEL (RCM) 260 6.1.2 RANDOM PORE MODEL (RPM) 268 6.1.3 MODIFIED VOLUMETRIC MODEL (MVM) . . . . . . . . . 278 6.1.4 DISCUSSION 280 6.2 TESTING SULPHUR CAPTURE MODELS 282 6.2.1 GRAIN MODEL (GM) 282 6.2.2 CONTINUOUS FIRST-ORDER MODEL (CM) 295 6.2.3 DISCUSSION 299 7.0 CONCLUSION 300 7.1 SUMMARY OF FINDINGS 300 7.2 RECOMMENDATIONS 305 REFERENCES 307 APPENDICES 323 APPENDIX A: CALIBRATION OF THE SULPHUR G . C . AND OF THE ROTAMETERS 324 APPENDIX B: SAMPLE CALCULATION FOR STEAM PARTIAL PRESSURE . . 328 APPENDIX C: TECHNIQUES OF SULPHUR REMOVAL 331 APPENDIX D: SAMPLE CALCULATIONS OF De AND K 335 APPENDIX E : PLASMA SPECTROSCOPY RESULTS OF REPRESENTATIVE SAMPLES OF SORBENTS EMPLOYED IN THIS STUDY . . . . 344 APPENDIX F : D E T A I L S OF G A S I F I C A T I O N EXPERIMENTS AT VARIOUS CONDITIONS 345 v i i i LIST OF FIGURES Page Chapter 1 2.1 P a r t i c l e s i ze d i s t r i b u t i o n of o i l sand cokes 11 2.2 D i s s o c i a t i o n pressure of c a l c i t e and magnesite 55 2.3 E q u i l i b r i u m constants for high temperature sulphur removal 68 2.4 E q u i l i b r i u m H 2 S pressure over CaO i n the range of steam p a r t i a l pressure employed in t h i s study 70 2.5 Value of n i n equation (124) 83 2.6 Rate constants vs . H 2 S concentrat ions . Data from Squires [168] and Richards [149] . . . 85 Chapter 3 3.1 Schematic view of the c a l c i n e r 101 3.2 Bowls used i n c a l c i n a t i o n of carbonate rocks . . . . . . 102 3.3 Schematic view of the apparatus employed i n t h i s study 104 3.4 Sample holder employed i n runs with d ^ g = 0.14 mm . . . 108 3.5 Engineering drawing of the g a s i f i e r and i t s l i v e flange I l l 3.6 Engineering drawing of the coo l ing sec t ion of the g a s i f i e r 112 3.7 Temperature p r o f i l e during c a l c i n a t i o n 117 Chapter 4 4.1.1 Produced gas concentrat ions in cases with and without the s t i r r e r 130 4.1.2 Carbon conversion and temperature of the sample i n runs with and without the s t i r r e r 131 ix 4.1.3 Carbon conversion for runs of d i f f e r e n t temperatures . . 133 4 .1 .4 Sulphur conversions for runs at d i f f e r e n t temperatures 135 4 .1 .5 Rate of carbon conversion for runs at d i f f e r e n t temperatures 136 4 .1 .6 Sulphur conversion as a funct ion of carbon conversion for runs at d i f f e r e n t temperatures 138 4 .1 .7 Carbon conversions a f t er 5 hours of g a s i f i c a t i o n , at var ious temperatures, determined by gas ana lys i s and by weight change 139 4 .1 .8 Sulphur conversion of runs of d i f f e r e n t temperatures 142 4 .1 .9 Carbon conversions for runs of d i f f e r e n t coke p a r t i c l e s i zes 145 4.1.10 Sulphur conversion for runs of d i f f e r e n t coke p a r t i c l e s i zes 147 4.1.11 Rate of carbon conversion for runs at d i f f e r e n t coke p a r t i c l e s izes 148 4.1.12 Sulphur conversion as a funct ion of carbon conversion for runs at var ious p a r t i c l e s ize 149 4.1.13 Carbon conversions, at var ious coke p a r t i c l e s i z e s , a f t er 5 hours of reac t ion . 151 4.1.14 Surface area as a funct ion of carbon conversion 157 4.1 .15a-g S . E . M . photos of delayed coke at var ious carbon conversions 159-162 4.1.16 Poros i ty determination: equivalent pore diameter as a funct ion of volume penetrated 164 4.1.17 Pore volume as a funct ion of carbon conversion 165 4.1.18 Gas concentrat ion for four repeated runs. d =2 mm, T = 9 3 0 ° C , P T I =0.3 atm 172 pc H 2 0 4.1.19 Carbon conversions for four repeated runs 173 4 .2 .1 Carbon conversion for runs at d i f f e r e n t temperatures, Syncrude coke 175 x 4.2.2 Sulphur conversions of runs at various temperatures, Syncrude coke 176 4.2.3 Rates of carbon conversion for runs at d i f f e r e n t temperatures, Syncrude coke 178 4.2.4 Sulphur conversion as a funct ion of carbon conversion, Syncrude coke 179 4 .2 .5a-c S . E . M . photos of f l u i d coke at var ious carbon conversions 181-182 4.2 .6 Carbon conversions for runs at var ious steam p a r t i a l pressure in the g a s i f y i n g medium 186 4.2 .7 Sulphur conversions for runs at var ious steam p a r t i a l pressures 188 4.2 .8 Rate of carbon conversion at var ious steam p a r t i a l pressures 189 4 .2 .9 Sulphur conversion as a funct ion of carbon conversion for runs at var ious steam p a r t i a l pressures 191 4.3.1 Carbon conversions of delayed and f l u i d cokes for runs at d i f f e r e n t temperatures 194 4.3^2 Hydrogen sulphide concentrat ion for delayed and f l u i d cokes at var ious temperatures 196 4.3.3 Sulphur conversions for runs with delayed and f l u i d cokes at var ious temperatures .198 4.3.4 Rates of carbon conversion for runs with delayed and f l u i d cokes at var ious temperatures 200 Chapter 5 5.1.1 E f f e c t of sorbent a d d i t i o n on carbon conversion 208 5.2.1 E f f e c t of mixing on H 2 S concentrat ions 210 5.3.1 E f f e c t of c a l c i n a t i o n time on H 2 S concentrat ions in runs using c a l c i n e d limestone 213 5.3.2 E f f e c t of c a l c i n a t i o n time on H 2 S concentrations in runs using ca l c ined limestone 214 5.3.3 Ef fec t s of c a l c i n a t i o n condi t ions on sorbent conversion 216 x i 5.3.4 Rate of sorbent conversion for runs with ca l c ines of var ious c a l c i n a t i o n condi t ions 217 5.3.5 F r a c t i o n a l H 2 S removal vs time for runs with sorbents of d i f f e r e n t c a l c i n a t i o n condi t ions . .218 5.3.6 E f f e c t s of c a l c i n a t i o n condi t ions on surface area of sorbents and on f r a c t i o n a l H 2 S removal 220 5 .3 .7a S . E . M . photos of Cache Creek l imestone, d =0.68 mm . . .221 ps 5.3.7b S . E . M . photos of Cache Creek l imestone, ca l c ined i n 40 minutes, d =0.68 mm 221 ps 5.3.7c S . E . M . photos of Cache Creek l imestone, ca l c ined i n 120 minutes, d =0.68 mm 222 ps 5.3.7d S . E . M . photo of Cache Creek l imestone, c a l c i n e d i n 300 minutes, d =0.68 mm 222 ps 5.3.7e S . E . M . photo of Texada l imestone, c a l c i n e d in 300 minutes, d =0.68 mm 223 ps 5 .3 .7 f S . E . M . photo of dolomite c a l c i n e d , i n 90 minutes, d =0.68 mm 223 ps 5 .3 .7g-h S . E . M . photo of dolomite c a l c i n e d , i n 90 minutes, d =0.68 mm 224 ps 5 . 3 . 7 i - j S . E . M . photos of reacted c a l c i n e , T = 9 3 0 ° C , d =0.68 mm 225 ps 5.3.8 P o r o s i t y determination of c a l c i n e s . Equivalent pore diameter vs . volume penetrated 227 5.3.9 F r a c t i o n a l sulphur removal v s . surface area for runs with sorbent at d i f f e r e n t c a l c i n a t i o n condi t ions 230 5.4.1 H 2 S concentrat ion for runs using sorbents of d i f f e r e n t Ca/S molar r a t i o . Runs with dolomite 233 5.4.2 H 2 S concentrat ion for runs using sorbent of d i f f e r e n t Ca/S molar r a t i o . Runs with limestone 234 5.4.3 E f f e c t s of Ca/S molar r a t i o on f r a c t i o n a l sulphur removal. Runs with dolomite 235 5.4.4 E f f e c t of Ca/S molar r a t i o s on sorbent conversion . . . .237 x i i 5.4.5 Rate of sorbent conversion for runs of d i f f e r e n t Ca/S molar r a t i o s 238 5.4.6 Weight of sulphur i n spent sorbent determined by gas and by weight ana lys i s . . 240 5.5.1 E f f e c t of sorbent p a r t i c l e s i ze on H 2 S concentrat ion i n the produced gas 242 5.5.2 F r a c t i o n a l H 2 S removal for runs with d i f f e r e n t sorbent p a r t i c l e s i zes 243 5.5.3 Sorbent conversion i n runs of d i f f e r e n t sorbent s i zes 245 5.5.4 Rate of sorbent conversion i n runs us ing c a l c i n e d of d i f f e r e n t s i zes 246 5.5 .5 E f f e c t of surface areas of ca l c ines in runs of d i f f e r e n t sorbent s i z e 247 5.6.1 E f f e c t of sorbent of d i f f e r e n t g e o l o g i c a l o r i g i n s on H 2 S concentrat ion i n the produced gas 252 5.6.2 E f f e c t of types of sorbent on H 2 S concentrat ions in the produced gas 254 5.6.3 F r a c t i o n a l sulphur removal in runs with d i f f e r e n t types of ca l c ines 255 5.6.4 Sorbent conversion vs . s p e c i f i c surface area for runs wi th d i f f e r e n t types of c a l c i n e . . . . . . . . .257 Chapter 6 6.1 F i t of carbon conversions vs . time data to RCM and MVM Runs with d i f f e r e n t temperatures 263 6.2 F i t of carbon conversion vs . time data to RCM and MVM. Runs with d i f f e r e n t coke p a r t i c l e s izes 264 6.3 Tes t ing RCM and RPM using experimental (dXc/dt)-Xc data of runs at d i f f e r e n t temperatures 265 6.4 Tes t ing RCM and MVM using experimental (dXrj/dt)-Xc data of runs with d i f f e r e n t p a r t i c l e s i ze 266 6.5 Test ing RCM using experimental Xrj-dimensionless time data 267 x i i i 6.6 Test ing RCM using experimental normalized ra te s , ( d x c / d t ) / ( d x c / d t ) X c = 0 1 2 6 9 6.7 P l o t t i n g equation (41) to determine the value of the s t r u c t u r a l parameter v|» 271 6.8 P r e d i c t i o n of surface area by RPM v i a equation (38) 273 6.9 Arrhenius- type p lo t for runs wi th delayed-coke at d i f f e r e n t temperatures, us ing RPM 275 6.10 Tes t ing RCM and RPM on the r e l a t i o n s h i p between rate and surface area 276 6.11 Test ing the G.M. in the region of r e a c t i o n c o n t r o l , using experimental data for runs with d i f f e r e n t sorbent s i zes 284 6.12 Tes t ing the G.M. i n the region of r e a c t i o n c o n t r o l , using experimental data for runs with sorbents from d i f f e r e n t c a l c i n a t i o n condi t ions .285 6.13 Test ing the G.M. in the region of product l ayer c o n t r o l . Using experimental data of runs with d i f f e r e n t sorbent s i zes . . 288 6.14 Test ing G.M. i n the region of product layer c o n t r o l , using experimental data of runs with sorbents of d i f f e r e n t c a l c i n a t i o n condi t ions .289 6.15 Probable mechanism of CaS formation of i o n i c d i f f u s i o n through the product layer 291 6.16 Pred i c t i ons by the G.M. on Xgg- t , using e a r l y stage equation (121) and l a t e r stage equation (122). Runs at d i f f e r e n t sorbent p a r t i c l e s i zes 294 6.17 Tes t ing the C M . using experimental data of runs with d i f f e r e n t sorbent s i zes 296 6.18 Tes t ing the C M . using experimental data of runs with sorbents of d i f f e r e n t c a l c i n a t i o n condi t ions 297 x i v Appendices A . l C a l i b r a t i o n curve of the gas chromatograph analyzing H 2 S 325 A.2 C a l i b r a t i o n curve for N 2 rotameter .326 A.3 C a l i b r a t i o n curve for H 2 0 rotameter 327 xv LIST OF TABLES Page Chapter 2 2.1 Some proper t i e s of Athabasca o i l - s a n d bitumen 6 2.2 Some c h a r a c t e r i s t i c s of the Suncor and Syncrude coking operations 9 2.3 Mul t i - e l ementa l ana lys i s of coke from Athabasca o i l sand bitumen 12 2.4 Suncor delayed coke 14 2.5 Syncrude f l u i d coke 15 2.6 Chemical composition of carbonate rocks employed in t h i s study 51 2.7 E f f e c t s of c a l c i n a t i o n condi t ions on r e s u l t i n g limes 59 2.8 E f f e c t s of types of carbonate rocks on r e s u l t i n g ca l c ines at 8 4 3 ° C for 1 hour 60 2.9 T y p i c a l values of l imestones and i t s r e l a t e d species 65 Chapter 3 3.1 Apparent c h a r a c t e r i s t i c s of four d iverse types of carbonate rocks employed in t h i s study 96 3.2 Size f r a c t i o n s of D . C . and Carbonate rocks employed i n t h i s study 96 3.3 Range of operating condi t ions employed in t h i s research 115 3.4 Amount of l i q u i d water in jec ted at d i f f e r e n t experimental condi t ions 124 xv i Chapter 4 4.1.1 D e s c r i p t i o n of bed appearance and mixing 129 4.1.2 Experimental condi t ions for runs at d i f f e r e n t temperatures with delayed coke 132 4.1.3 Experimental condi t ions for delayed coke runs at d i f f e r e n t d ; PT T =30.3 kPa and T = 9 3 0 ° C 144 pc H 2 0 4 .1 .4 Surface area based on i n i t i a l carbon for g a s i f i c a t i o n at d i f f e r e n t conversion l e v e l s 157 4 .1 .5 V and N i concentrat ions i n p a r t i a l l y reacted coke . . . .166 4 .1 .6 Average value of Xrjg for past and present s tudies . . . .167 4 .2 .1 Experimental condi t ions for runs at d i f f e r e n t P^ ^ . . . .185 4 .3 .1 Experimental condi t ions for runs comparing D . C . and F . C 193 4.3.2 Carbon conversion of D . C . and F . C . a f ter 5 hours of g a s i f i c a t i o n 195 4.3.3 Sulphur conversion of D . C . and F . C . a f ter 5 hours . . . .199 4 .3 .4 Values of carbon conversion where (dX/dT) a t ta ins maximum 202 Chapter 5 5.1.1 Chemical compositions of ca l c ined sorbents 208 5.3.1 C a l c i n a t i o n cond i t ion for var ious samples c a l c i n e d i n N 2 at 9 5 0 ° C 212 5.6.1 Comparisons between c a l c i n e s employed in t h i s study . . .256 Chapter 6 6.1 Values of M and N i n equation (148) for var ious experiments 261 6.2 Value of \|> and k s in equation (149), obtained i n f i t t i n g experimental data into equation (41) . . . . . . .272 6.3 Values of A and B in equation (68) and k in equation (76) 279 x v i i 6.4 Summary of sorbent characters and r e s u l t s in t e s t i n g models using sulphur capture data 287 6.5 D i f f u s i v i t i e s of reac t ing gas through product layer 290 6.6 Rate constants of the CaO + H 2 S r eac t ion determined by means of C M 298 Appendix A . l S p e c i f i c volume of steam at var ious temperature and pressure 328 F . l D e t a i l s of g a s i f i c a t i o n experiments 345 x v i i i LIST OF SYMBOLS A , B Parameters i n equation (68) Ag,Ap Externa l surface area of the i n d i v i d u a l grains and the p e l l e t s r e s p e c t i v e l y , cm^ AUJ A r e a occupied by one molecule of the adsorbed spec ies , cra^ a s t o i c h i o m e t r i c c o e f f i c i e n t s a ^ ( i = l , 2 , . . . ) Rate constants for equat ions(7) , (9) B Q , B i Moments of p r o b a b i l i t y dens i ty X.(r) i n equation (53 and 54) b S to ich iometr ic c o e f f i c i e n t b j ( i = l , 2 , . . . ) Rate constant f o r equat ions(7) , (9) C , C A Concentrat ion of gaseous reactant , mol/cm^ C^ I n i t i a l c o n c e n t r a t i o n of gaseous r e a c t a n t , o raol/cra^ C§ Concentrat ion of ac t ive s i t e s c s t o i c h i o m e t r i c constant c j ( i = l , 2 , . . . ) Rate constant of equations(7) , (9) D The most probable pore diameter, cm De E f f e c t i v e d i f f u s i o n c o e f f i c i e n t , m^/s Dp D i f f u s i v i t y through product l a y e r , m^/s d S to i ch iometr i c constant d ^ ( i = l , 2 , . . . ) Rate constant for equation (9) d p C Average coke p a r t i c l e s i z e , mm d Average sorbent p a r t i c l e s i z e , mm ps d _„ P a r t i c l e diameter of 50% removal, mm p50 E A c t i v a t i o n energy, kJ/mol x i x e^( i=l ,2 , . . . ) Rate constant for equation (9) eo Expansion Rat io Fg,Fp S h a p e f a c t o r f o r g r a i n s and t h e p e l l e t r e s p e c t i v e l y ( = 1, 2 and 3 f o r f l a t p l a t e s , c y l i n d e r s and spheres, r e s p e c t i v e l y ) f i ( i = l , 2 , . . . ) Rate constant for equation (9) f r S i z e d i s t r i b u t i o n of non-overlapped system at anytime i n equations (15-17) g ^ ( i = l , 2 , . . . ) Rate constant f o r equation (9) A gp (Xsg) C o n v e r s i o n f u n c t i o n f o r s m a l l a d e f i n e d by g equation (114) hj_(i=l, 2, . . . ) Rate constant for equation (9) I Intercept of the s t r a i g h t l i n e given in equation (142) i ^ ( i = l , 2 , . . . ) Rate constant f o r equation (9) j j ( i = l , 2 , . . . ) Rate constant for equation (9) K Reaction rate constant Kj ,K2 E q u i l i b r i u m constants K D Constant of p r o p o r t i o n a l i t y i n equation (124) k R e a c t i o n r a t e c o n s t a n t i n e q u a t i o n (102) , (cra^/mol) n"^'cm/s k' R e a c t i o n r a t e c o n s t a n t i n e q u a t i o n ( 1 0 ) , cm/min.atm. k^( i= l ,2 , . . . ) Rate constant for equation (9) k s Rate constant for surface r e a c t i o n in equation (19) k * s Rate constant k s ( P H Q / c ) = k s ( R T ) , c m 4 / m i n » m o l k Average rate constant L Length of overlapped system L c c h a r a c t e r i s t i c length as in equation (44) xx Lg T o t a l length of non-overlapped system, cm li£0 L J J at t=0, cm L 0 L value at t=0 1^ Length c h a r a c t e r i s t i c of i n i t i a l pore s tructure as in equation (45) Avogadro's number n Reaction order with respect to gas phase P Pressure, kPa Pj(j=H20,C02,CO,H2) P a r t i a l pressure of the j component P D Force to move mercury, kPa 2 P P s Surface tens ion f o r c e , kPa ,m p^ P a r t i a l pressure of the adsorbed species A Pp- (^SB) C o n v e r s i o n f u n c t i o n f o r l a r g e r a de f ined by p equations (116-119) p Q Vapour pressure of adsorbate q Increase i n pore radius due to r e a c t i o n R P a r t i c l e radius R 1 Distance coordinate i n the p e l l e t , cm R Q I n i t i a l p a r t i c l e r a d i u s , cm r Pore rad ius , cm r * , r Smallest and larges t pore r a d i i r ' Distance coordinate perpendicu lar to the moving reac t ion front i n the i n d i v i d u a l g r a i n s , cm rg g r a i n rad ius , cm r m I n s t a n t a n e o u s r a t e of r e a c t i o n ( p e r mass of carbon remaining, sec"^) in equation (7) r s Surface reac t ion rate defined by equation (42), g carbon/unite a r e a « t i m e S Reaction surface per u n i t volume x x i Slope of the s t r a i g h t l i n e given in equation (142) S u r f a c e a r e a of c y l i n d r i c a l system per u n i t volume as in equation (16), cm^ S E at t=0 S p e c i f i c surface areas, m^/g I n i t i a l S p e c i f i c Surface area based on volume, cm^/cm^ B . E . T . Surface area, m^/g Temperature time Diraensionless time Dimensionless time Time at which the rate of g a s i f i c a t i o n a t t a i n i n g maximum Time to X c =0.5 Dimensionless time defined by equation (111) Volume enc losed by reac t ion surface , per u n i t volume of space as i n equation (25), cm^ Volume enclosed by c y l i n d r i c a l system, per u n i t volume of space, cm^ I n i t i a l enclosed volume of non-overlapped pore, cm^ Volume of i n d i v i d u a l g r a i n s and the p e l l e t r e s p e c t i v e l y , cra^ V at t=0 Pore volume, cm^/g Reaction v e l o c i t y as i n equation (42) x x i i Reaction rate for species A, i n equation (101), gmole cm^/sec. V g Volume occupied by one mole of gas at standard cond i t ions , cm 3 /mol v m Volume of gas adsorbed at monolayer coverage, cm 3 /g v n Volume of gas adsorbed on u n i t mass of porous s o l i d , cm 3 /g V p Volume of pores l a r g e r than r def ined by equation (138) W Weight of carbon, g W , Weight of ash ash WD I n i t i a l weight of s o l i d , g W v Weight of v o l a t i l e , g X c Carbon conversion based on i n i t i a l mass of carbon for g a s i f i c a t i o n XrjR Conversion at time of maximum rate of conversion Xfjs Conversion at time of maximum surface area Xg Sulphur conversion Xgg Sorbent conversion Surface tens ion , N/ra E Poros i ty E C I n i t i a l poros i ty of c a l c i n e E 0 I n i t i a l value of E eg Poros i ty of completely sulphided lime c-j- T o t a l poros i ty defined by equation (51) E T O I n i t i a l value of n D i m e n s i o n l e s s d i s t a n c e c o o r d i n a t e defined by equation (110) x x i i i C D i m e n s i o n l e s s p o s i t i o n of the reac t ion front w i th in the g r a i n def ined by equation (106) \,Xo P r o b a b i l i t y dens i ty funct ion at t and at t=0 p Densi ty , g/cm^ p c Apparent dens i ty of carbon in the char Pl c(k=CaO,CaS) True dens i ty of component k, cm^/g p'l c(k=CaO,CaC03,CaS) Bulk dens i ty of component k, cm^/g p"CaO Molar dens i ty of CaO, mol/cm^ p m Molar dens i ty of the s o l i d , mol/cm^ p s D e n s i t y of c h a r e x c l u d i n g t r a n s i t i o n a l and macropore volume p t True dens i ty of coke, g/cm^ 0 Contact angle between mercury and pore wal l p P a r t i c l e s i ze parameter def ined by equation (35) A o General ize g a s - s o l i d reac t ion modulus defined by equation (112) •c K i n e t i c parameter def ined by equation (36) \|> Structure parameter defined by equation (34) $ Dimensionless concentrat ion of gaseous reactant xx iv CHAPTER 1 1 INTRODUCTION AND OBJECTIVES The two larges t Canadian o i l sand p lants - Suncor and Syncrude -produce about 3 0 x l 0 3 m^/day of synthet i c crude o i l and about 3000 tonne/day of by-product cokes. The h igh sulphur and vanadium l e v e l s i n the coke preclude many p o t e n t i a l uses, and in sp i t e of i t s h igh heating value (30MJ/kg) i t i s not f u l l y u t i l i z e d as a f u e l . The coke i s now being s t o c k p i l e d , causing environmental impacts i n the use of land f o r s u r f a c e s t o r a g e , and perhaps some concerns about contamination of surface and ground water run o f f w i t h l e a c h a t e s . R e v i s e d r e g u l a t i o n s f o r e n v i r o n m e n t a l p r o t e c t i o n r e q u i r e , among other th ings , that a t t en t ion be paid to p o t e n t i a l sulphur emissions before a s u b s t a n t i a l f r a c t i o n of the cokes can be u t i l i z e d . A process other than d i r e c t combustion that could employ o i l sand coke as a s o l i d f u e l at low sulphur emission would be a t t r a c t i v e i f i t was economical . G a s i f i c a t i o n of the coke to produce low or medium c a l o r i f i c value gas for i n - p l a n t use, or generat ion of e l e c t r i c a l power i s one such p o s s i b i l i t y . Steam g a s i f i c a t i o n of the coke produces H 2 , which i n t u r n reacts wi th sulphur in the coke s t ruc ture to form hydrogen sulphide ( H 2 S ) in the gas mixture . H 2 S must be removed from the product gas before i t i s used as synthesis gas or f o r power generat ion . I n - s i t u d e s u l p h u r i z a t i o n using low cost calcium-based sorbents to produce high temperature gas has recent ly been proposed. In t h i s process , the spent 2 sorbent could be regenerated to y i e l d H 2 S for the conventional Claus process . It i s known that the rate of g a s i f i c a t i o n depends s trongly on the p r o p e r t i e s of the carbon-bearing s o l i d i n quest ion . O i l sand cokes -the res idue from thermal cracking of Athabasca bitumen - have unique p h y s i c a l s t ruc ture and chemical composi t ions , wh ich make them more d i f f i c u l t to be g a s i f i e d than e i t h e r coals or chars . The r e a c t i v i t i e s of the cokes i n a given atmosphere depend e s s e n t i a l l y on the reac t ion temperature, r e a c t i o n time, and the s i ze of the s o l i d p a r t i c l e s . The sulphur emission rate w i l l a l so be a funct ion of these v a r i a b l e s . Surface area and poros i ty of s o l i d s are reported [10,41] to change during the course of g a s i f i c a t i o n . These changes r e f l e c t the rate of r e a c t i o n , and r e l a t e to the l e v e l of carbon convers ion . Equal ly important for the H 2 S capture are the ro l e s of surface areas of ca l c ined limestone or dolomite on the rate of s u l p h i d a t i o n and the o v e r a l l sorbent c a p a c i t y . Furimsky [64] inves t igated the g a s i f i c a t i o n r e a c t i v i t y of o i l sand cokes i n d i f f e r e n t g a s i f i c a t i o n media. However, because of h i s l i m i t e d experimental condi t ions and small number of t r i a l s , the k i n e t i c s were not a d e q u a t e l y e x p l o r e d . Watkinson et a l . , [189] c a r r i e d o u t experimental s tudies of g a s i f i c a t i o n and i n - s i t u sulphur capture of o i l sand cokes i n a p i l o t - s c a l e spout - f lud ized bed with steam and oxygen at atmospheric pressure . However, the s i ze of the u n i t and i t s throughput (ca 20 kg coke/hr) precluded c o n t r o l l e d experiments under a wide range of o p e r a t i n g condi t ions s u f f i c i e n t to e s t a b l i s h the k i n e t i c s of the process . To date there has been no fur ther systematic work reported i n 3 the l i t e r a t u r e on k i n e t i c aspects of g a s i f i c a t i o n of o i l sand cokes or of the i n - s i t u d e s u l p h u r i z a t i o n i n a g a s i f i e r using calcium-based sorbents or of the combination of the two. Thi s study was, there fore , d i rec t ed mainly to a fundamental study of the g a s i f i c a t i o n k i n e t i c s of the cokes in s t i r r e d and f i x e d bed r e a c t o r s to produce a c lean synthet ic f u e l gas, with sulphur being removed i n - s i t u by c a l c i n e s . The general object of the research i s two f o l d : 1 - To provide the necessary k i n e t i c information which w i l l permit a l o g i c a l process design of a g a s i f i c a t i o n - d e s u l p h u r i z a t i o n process for the o i l sand indus try . 2 - To acquire i n s i g h t s into the reac t ion mechanism, which may become valuable i n s o l v i n g the problems that may ar i se i n operat ion of such a process . The s p e c i f i c objec t ives of t h i s study are: 1 - To study f a c t o r s a f f e c t i n g the carbon conversion and the sulphur re lease during g a s i f i c a t i o n of both f l u i d and delayed cokes which o r i g i n a t e from the thermal cracking of Athabasca bitumen. These fac tors are r e a c t i o n temperatures between 8 0 0 ° C and 9 3 0 ° C , coke p a r t i c l e s i zes from 0.1 to 3.5 mm, reac t ion time up to 720 m i n . , and steam p a r t i a l pressure i n the range of 15.15 to 60.6 kPa. 2 - To determine the rate of carbon conversion at the var ious operat ing condi t ions and re la t e these r a t e s to the changes i n p h y s i c a l 4 s t r u c t u r e , i . e . surface area , p o r o s i t y and pore s i z e s of the reacted cokes. 3 - To evaluate the e f fec t s of sorbent p a r t i c l e s i zes (0.1 - 3.5 mm), of c a l c i n a t i o n cond i t i ons , of r eac t ion temperature ( 8 0 0 - 9 3 0 ° C ) , of sorbent types (l imestone or do lomite ) , and of molar Ca/S r a t i o s (0-2.0) on the capture of hydrogen sulphide during g a s i f i c a t i o n with f u l l y ca l c ined dolomite and l imestone. Achievement of these objec t ives w i l l e s t a b l i s h reference data for design of processes to produce h igh temperature combustible gases from o i l sand cokes with minimal sulphur content . 5 CHAPTER 2 LITERATURE REVIEW 2.1 OIL SANDS The Athabasca depos i t , which contains over 1.4x10^^- m^ of bitumen, i s the only o i l sand deposit c u r r e n t l y under commercial e x p l o i t a t i o n . The Suncor and Syncrude p lants both operate open p i t mines and use the hot water e x t r a c t i o n process for bitumen recovery [180]. In the Cold Lake p l a n t , the steam f looding i n - s i t u recovery process , which acts by thermal s t i m u l a t i o n of the heavy bitumen to increase i t s m o b i l i t y , w i l l be used [180,120]. The c h a r a c t e r i s t i c s of o i l sand are w e l l documented i n the l i t e r a t u r e [128,180], and several model s t ruc tures of o i l sands have been proposed and widely accepted [180]. Bitumen extracted from o i l sand by the hot water e x t r a c t i o n process contains approximately 5% sulphur and 15-20% asphaltenes [68]. These values are higher than those found i n conventional crude o i l s [140,182]. T y p i c a l proper t i e s of Athabasca o i l - s a n d bitumen are l i s t e d in Table 2 .1 . Bitumen r e c o v e r e d e i t h e r by s u r f a c e m i n i n g and subsequent e x t r a c t i o n from the sand through a washing process , as i n Suncor and Syncrude p l a n t s , or by thermal f looding of the deposi t accompanied by i n - s i t u r e c o v e r y , as i n the Cold Lake p l a n t , must subsequently be upgraded to produce a s y n t h e t i c crude t r a n s p o r t a b l e by e x i s t i n g p i p e l i n e s and acceptable to e x i s t i n g r e f i n e r i e s . 6 Table 2.1 SOME PROPERTIES OF ATHABASCA OIL-SAND BITUMENS [80,143] Elemental composition (wt%) Carbon 83.3 ± 0.3 Hydrogen 10.3 ± 0.1 Oxygen 1.1 ± 0.3 Ni trogen 0.4 ± 0.1 Sulphur 4.9 ± 0.4 Molecular Weight 540 - 800 S p e c i f i c g r a v i t y 1.01 H . H . V . MJ/kg 41 Conradson Carbon (wt%) 14.3 Metals (ppm): Vanadium 250 N i c k e l 90 Iron 75 Copper 5 7 Upgrading of bitumen can be accomplished by e i t h e r thermal cracking or hydrocracking . Although the l a t t e r method produces higher l i q u i d y i e l d s and higher value of byproduct cokes, thermal c r a c k i n g s t i l l r e t a i n s the advantage that less hydrogen i s required f o r upgrading as the carbon and sulphur , as w e l l as the metals n i c k e l and vanadium report to the coke product . Both e x i s t i n g p lants thus employ the former method [61,75] . K i n e t i c cons iderat ions of thermal cracking and hydrocracking of Athabasca bitumen i n terms of the o v e r a l l r e a c t i o n : Bitumen -» o i l + coke + gas were s t u d i e d by Barbour [11] and Koseoglu [113] r e s p e c t i v e l y . The nature of the cokes formed depends upon the coking process [61,75]. In the Suncor p l a n t , the primary upgrading i s accomplished by delayed coking which i s a batch operat ion . In t h i s process , bitumen i s preheated to 4 0 0 - 5 0 0 ° C [68] then charged into a coke drum. Thermal cracking of large bitumen molecules then takes place with a l l cracked products l eav ing the coke drum as vapour, while the coke which deposits as a porous mass r e t a i n s much of the sulphur and metals i n the feed. The y i e l d of coke i s about 20 wt% of the feed [119]. In the Syncrude p l a n t , continuous f l u i d coking i s employed. The coke p a r t i c l e s are f l u i d i z e d i n steam at 4 7 5 - 6 0 0 ° C . The steam also s t r i p s the d i s t i l l a t e o i l from the coke p a r t i c l e s and c a r r i e s the p r o d u c t o f f overhead as vapour which undergoes f u r t h e r process ing . Greater y i e l d s of va luable l i q u i d hydrocarbons and lower y i e l d s of coke [6] are der ived from f l u i d coking compared to delayed coking due to the be t ter thermal contact and the higher temperature in the former. The 8 coke p a r t i c l e s are cont inuously withdrawn from the cokers and fed to a second bed c a l l e d the "burner" where they are f l u i d i z e d i n a i r and some of the coke i s burned, thereby heating the remaining coke to about 6 5 0 ° C . The hot coke i s returned to the coker where i t heats the bitumen feed which i n turn vaporizes to y i e l d d i s t i l l a t e o i l and leaves more coke i n the coker. About 5% of coke produced i s burned to provide the coker ' s heat requirement; the res t i s s t o c k p i l e d [119]. T a b l e 2.2 summarizes some re levant information on the Suncor and Syncrude p l a n t s . The thermal cracking process that w i l l be used i n the Cold Lake p lant i s known as f l e x i c o k i n g . F l e x i c o k i n g i s the conventional f l u i d coking combined wi th g a s i f i c a t i o n of the coke product . 2.2 OIL SAND COKES The proper t i e s of o i l sand cokes have been the subject of numerous s tudies [80,94,180,54,55] which reveal that cokes produced from each .of the three thermal cracking processes have d i s t i n c t chemical compositions and s t r u c t u r a l p r o p e r t i e s . I n s p e c t i o n of the delayed coke by scanning e l e c t r o n microscope ( S . E . M . ) reveals the i r r e g u l a r shaped p a r t i c l e s expected from hydrau l i c c l e a r i n g of the coke from the c o k i n g drum [ 9 4 ] . There i s some disagreement on the reported values of the coke p o r o s i t y . Jack [94] found no poros i ty i n the p a r t i c l e s , while H a l l [74] and Probste in [143] argued that delayed coke has a r e l a t i v e l y high poros i ty because i t has not been subjected to high temperature in the c y c l i c burner-heater uni t s that are a feature of f l u i d cokers . Further tes ts are required to explore t h i s d iscrepancy. 9 Table 2.2 SOME CHARACTERISTICS OF THE SUNCOR AND SYNCRUDE COKING OPERATIONS. P lant SUNCOR SYNCRUDE Coking method Delayed F l u i d Coking temperature, °C 450-500 475-600 Coke product ion , tonne/day 2700 2300 Coke burned, tonne/day 2000 115 Natura l gas burned for steam, ra^s-* 0 20 Coke y i e l d , wt% of bitumen 20-25 15 Coke p r o p e r t i e s , wt% Sulphur 6.1 7.9 V o l a t i l e content 10.6 7.3 Ash 5.0 8.6 L i q u i d product ion , tonne/day 9.2 x 10 6 20 x 10 6 L i q u i d y i e l d , wt% of bitumen 70 77 Sulphur in l i q u i d , wt% 3.3 3.7 C a l o r i f i c value of coke, MJ/kg 31.9 29.8 10 F l u i d coke cons i s t s of hard granular p a r t i c l e s , formed from non-v o l a t i l e mater ia l s i n the f l u i d bed, with s i ze d i s t r i b u t i o n s i m i l a r to a f ine beach sand, i . e . 0.08-3 mm [6 ,80] . P a r t i c l e s i ze d i s t r i b u t i o n s for d e l a y e d , f l u i d and f l e x i cokes from Athabasca bitumen are shown in Figure 2 .1 . In f l u i d coking, the c y c l i c ac t i on between the coker vesse l and the heater ves se l creates dense s p h e r i c a l coke p a r t i c l e s wi th a layered , "on ion- l ike" i n t e r n a l s tructure [6 ,75 ,80 ] . Coke p a r t i c l e s der ived from t h i s f l u i d coking process maintain a smooth surface and are not porous [94]. Some authors [6 ,80,154,55] , however, r e p o r t t h a t Syncrude f l u i d coke has a s p e c i f i c area , as measured by n i trogen adsorpt ion , between 10.0 and 12.0 m 2 / g . Coke obtained from f l e x i c o k i n g has lower n i t rogen , hydrogen and sulphur contents , but higher ash and metal contents and higher s p e c i f i c surface areas than that found i n delayed and f l u i d cokes [94]. Mul t i - e l ementa l analyses of cokes from Athabasca o i l sand bitumen v i a each of the three coking processes are shown i n Table 2 .3 . The usual h igh l e v e l s of vanadium and n i c k e l are unique c h a r a c t e r i s t i c s of the o i l sand cokes [42]. Most of these metals are bound up i n porphyrin r ing ' systems [90]. The formation of complexes between V, N i s a l t s and organosulphur, which are a l l present in the o i l sand coke matr ix , i n the presence of steam r e s u l t s in the release of H 2 S i n the product gas [42]. The presence of h igh concentrat ion of V in fue l o i l was reported to cause severe vanadic corros ion i n b o i l e r s [13,47], while the combination of V, S, Na and oxygen acce lerates the corros ion process on the b o i l e r tubes [59]. 11 100 to 80 CD M CD CD > 60 -4= 4 0 jo £ o 20 0 A Fluid coke -— / ° / / ' O A Flex i coke • / 1 / A / o r / o ?/ o t A " * - D e l a y e d coke o o A F A ^ 1 1 0.01 0.1 1.0 10 Part ic le size (mm) 100 F i g . 2.1 P a r t i c l e s i z e d i s t r i b u t i o n . o f o i l sand cokes [94], by weight. Table 2.3 MULT I-ELEMENTAL ANALYSIS OF COKE FROM ATHABASCA OIL SAND BITUMEN. Concentrat ion (ppm) Element Delayed-Coker Coke(*) Delayed-coker Coke(**) ' F l u i d ' Coke(**) F l ex i coker Coke(**) S i 7526 a a a A l 4498 4720 6930 7020 As - 26 b 90 Ba 21.3 30 30 80 Ca 786.2 3950 602 1640 Cr 15.4 9.5 51 59 Co 6.60 8 10 18 Cu 6.10 4 11 20 Fe 2658 2360 2570 6990 Pb 7.73 5 20 6 Mg 380.4 730 272 750 Mn 62.6 51 49 98 Mo 44.6 80 121 220 N i 455.0 356 660 1760 P 122 102 b 188 K 564.5 2600 b 5470 Ag - 0.5 0.4 2 Na 304.20 260 370 960 Sr 17.7 18 28.8 38 T i 762 828 1160 2140 W - 20 b 40 V 1010 840 1590 8010 Zn 10.1 6.4 13 33.4 Zr ~ 30 43 70 a E l e m e n t s not detected: B, Cd, Eu, Au, Se, Te, Sn, U. S i l i c o n i s excluded i n t h i s determination ^Not determined (*)From t h i s study (**) From reference [21] 13 In a study invo lv ing combustion of Syncrude coke in a c i r c u l a t i n g f l u i d i z e d bed r e a c t o r , Hamer et a l . [76] found no evidence of corros ion on a surface which could be a t t r i b u t e d to the vanadium content i n the coke; and almost a l l vanadium reported in the r e s i d u a l ash. On the other hand, Luthra [121] claims that during combustion the vanadium i s re leased in to the gas phase as oxide and hydroxide. Ult imate and proximate analyses of the coke samples used in t h i s research and s i m i l a r analyses made by previous workers [119,7,61,74] are presented i n Table 2.4 and Table 2 .5 . O i l sand cokes have r e l a t i v e l y low v o l a t i l e content as a r e s u l t of high temperature thermal treatment. Of the two, Suncor Coke i s known to be more reac t ive than Syncrude coke due to i t s higher v o l a t i l e and lower ash content . The f l u i d coking process was designed to reduce the sulphur l e v e l i n the d i s t i l l e d o i l , and consequently to produce higher sulphur-content coke than comes from the delayed coking process . Ash, heavy metals and sulphur contents are major parameters that can a f fec t de tr imenta l ly the value of byproduct cokes. Conventional petroleum cokes have lower metal , s i l i c a and often have lower sulphur content than o i l sand cokes. As such, they are pre ferred mater ia l s for product ion of anode grade carbon, and metals contents r u l e out t h e i r use as m e t a l l u r g i c a l coke. F i n a l l y , t h e i r low s u r f a c e area r u l e s out adsorbent a p p l i c a t i o n s [6] . Another obvious use of byproduct coke i s as a f u e l . P o t e n t i a l means of u t i l i z i n g coke as a f u e l are combustion and g a s i f i c a t i o n . Table 2.4 SUNCOR DELAYED COKE PROXIMATE ANALYSIS: I, Dry bas i s : This Study Anthony et a l . [7] Lewis [119] Ash 3.09 4.42 4.31 V o l a t i l e matter 12.16 12.95 10.31 Fixed Carbon 84.75 _82._63 85.38 100.00 100.00 100.00 ULTIMATE ANALYSIS: Dry bas i s : This Study Anthony et a l . [7] Lewis [119] Carbon Hydrogen Ni trogen Chlor ine Sulphur Ash Oxygen ( d i f f . ) 84.90 3.93 32 01 02 09 73 100.00 83.15 3.54 1.42 5.65 4.42 1.82 100.00 75.28 3.30 1.01 5.32 5.51 1.38 100.00 Table 2.5 SYNCRUDE FLUID COKE PROXIMATE ANALYSIS: %, Dry b a s i s : This Study H a l l et a l . [74] F r i e d r i c h [61] Ash 8.30 8.7 6.79 V o l a t i l e matter 7.16 7.3 7.54 Fixed Carbon 84.54 84.0 85.67 100.00 100.00 100.00 ULTIMATE ANALYSIS: Dry b a s i s : This Study H a l l et a l . [74] F r i e d r i c h [61] Carbon Hydrogen Ni trogen Chlor ine Sulphur Ash Oxygen ( d i f f . ) 79.49 1.56 1.69 0.70 6.96 8.30 1.30 100.00 80.5 1.6 1.5 6.0 6.0 4.40 100.00 81.32 70 24 03 79 1.92 100.00 16 There have been some concerns that the r e l a t i v e l y low v o l a t i l e content of the cokes, e s p e c i a l l y Syncrude f l u i d coke, may prevent t h e i r use as a f u e l without supplementary fue l and a gr ind ing step to support d i r e c t combust ion [ 6 ] . Such concerns are not s u p p o r t e d by the experience of Suncor [119] in f i r i n g delayed coke i n b o i l e r s , with some c a r r y o v e r of unburned carbon, to ra i se steam, nor CANMET [61,7] i n burning s u c c e s s f u l l y Syncrude coke, i n a p i l o t - s c a l e a i r - c i r c u l a t i n g f l u d i z e d bed, and Suncor coke i n a p i l o t - p l a n t atmospheric f l u d i z e d bed combustor. Combustion of s u l p h u r - b e a r i n g cokes as a f u e l may be l i m i t e d , however, by cons tra in t s on sulphur d iox ide emissions [68,7] . Consequently, even though the o i l sand coke i s an a t t r a c t i v e b o i l e r f u e l ( c a l o r i f i c value of 30 MJ/kg) , about 3000 tonnes per day of the coke are being s t o c k p i l e d . Thus, there recent ly has been a growing in t ere s t i n d e s u l p h u r i z a t i o n of the coke p a r t i c l e s p r i o r to combustion. The sulphur in o i l sand (and i n petroleum) cokes i s almost e n t i r e l y o r g a n i c i n nature [6,61,69,80,182,189] . The sulphur i s bound as a v a r i e t y o f a l i p h a t i c and a r o m a t i c o r g a n o s u l p h u r c o m p o u n d s [42,64,169,44] . Mercaptans, sulphides [182] and thiophenes [182,42,43] are considered as being representat ive of the sulphur species occurr ing i n the bitumens of Athabasca o i l sand. Desu lphur iza t ion of o i l sand cokes b a s i c a l l y involves the b r e a k i n g of c a r b o n - s u l p h u r bonds and removal of the s u l p h u r from the coke matr ix [74]. C h a r a c t e r i s t i c approaches for these processes are h y d r o d e s u l p h u r i z a t i o n of cokes p a r t i c l e s [75] and impregnation of the cokes with s u i t a b l e a l k a l i n e reagents [69] . Optimum temperature for hydrodesulphur izat ion of delayed coke i s i n the range 7 5 0 - 8 0 0 ° C and for the f l u i d coke i s about 890-17 9 0 0 ° C , f or p a r t i c l e s i ze of -325 mesh [75]. The extent of sulphur removal at t e m p e r a t u r e s below opt ima, and the rate of removal at temperatures above these ranges are reduced. The c o l l a p s e and the obs t ruc t ion of the i n t e r n a l pore s tructure account f o r the reduct ion of sulphur removal rate at high temperature (above 7 0 0 ° C ) [75]. George et a l . [69] postulated that progress ive incorporat ion of sulphur atoms into s table polyaromatic thiophene r ings may i n h i b i t sulphur removal a f t e r a prolonged r e a c t i o n t ime. It becomes very obvious that i f a g a s i f i c a t i o n process i s to be proposed to u t i l i z e the o i l sand cokes as a f u e l wi th the sulphur to be removed i n - s i t u , i t must involve severa l g a s - s o l i d r e a c t i o n s . The k i n e t i c s of these react ions w i l l be introduced in the fo l lowing s e c t i o n . 2.3 GAS POROUS-SOLID REACTIONS For a l l porous s o l i d s , the r a t i o n a l design of a p r a c t i c a l reactor requires an understanding of fundamental r eac t ion k i n e t i c s , the i n t e r n a l s t ruc ture and the process of g a s - s o l i d contact . D i f f u s i o n and reac t ion of gaseous reactants w i t h i n porous p a r t i c l e s t end to r e c e i v e much a t t en t ion because of a wide v a r i e t y of a p p l i c a t i o n s i n the chemical process i n d u s t r i e s . The steps involved in these react ions w i t h i n the p a r t i c l e are the transport of reac t ing gas into the i n t e r i o r of the porous s t r u c t u r e , chemica l r e a c t i o n on the pore s u r f a c e s and the transport of the products from the p a r t i c l e surface from the reac t ion s i t e s . Because of the continuous evo lut ion of the porous s t ruc ture and the d i f f u s i o n a l l i m i t a t i o n , the n o n c a t a l y t i c reac t ion i s very d i f f i c u l t to model. Under these cond i t i ons , the i n t e r n a l s t ruc ture i s consumed at 18 d i f f e r e n t ra t e s , r e s u l t i n g i n phys i ca l and transport proper t i e s which change with conversion as we l l as with p o s i t i o n w i th in the p a r t i c l e [170,10,112,73]. Externa l t r a n s f e r steps may also be important. The s t r u c t u r a l change i n n o n c a t a l y t i c reac t ion i s a func t ion of the nature of the reac t ion products . Thus, i f a l l products formed are gaseous, the p a r t i c l e s are cont inuously consumed by r e a c t i o n u n t i l the poros i ty reaches a c r i t i c a l value where d i s i n t e g r a t i o n of the s o l i d occurs , as i n the case of carbon g a s i f i c a t i o n [49,66] . On the other hand, when the r e a c t i o n products are s o l i d s , the s t r u c t u r a l changes depend on the molar volume r a t i o of the s o l i d products to that of s o l i d reactant . Depending on the value of the expansion r a t i o parameter, e Q , two main cases a r i s e . When e Q i s l e s s than 1, the system i s charac ter i zed by a gradual increase in p o r o s i t y and pore s i z e [148], However, depending on the value of e Q between 0 and 1, the porous s tructure may become very d i f f e r e n t . At e Q •• 0, s i g n i f i c a n t s t r u c t u r a l changes occur , represent ing those of a g a s i f i c a t i o n r e a c t i o n where the large p o r o s i t y increases may cause the p a r t i c l e to c o l l a p s e . For e Q -* 1, the s o l i d reactant i s converted into a s o l i d product without causing not iceable change i n the p h y s i c a l s t ructure i n the p a r t i c l e . When e Q i s l arger than 1, the system i s charac ter i zed by a continuous bu i ld -up i n the s o l i d product which r e s u l t s i n pore p lugging . T y p i c a l of t h i s s i t u a t i o n are the su lphat ion of CaO (e o=3.10) and s u l p h i d a t i o n of CaO (e Q =1.64). Unl ike the case of e 0 < l , a complete conversion of reactant s o l i d might not be poss ib le i n the case of s o l i d b u i l d u p i n pores [81,18]. 19 This thes i s deals with react ions where both e Q <l and e Q >l . The case where e Q •* 0, involves the g a s i f i c a t i o n of the o i l sand cokes i n which gaseous p r o d u c t s are formed , pores e n l a r g e d and p a r t i c l e s u l t i m a t e l y d i s i n t e g r a t e . The case where e Q >l , involves the s u l p h i d a t i o n r e a c t i o n (CaO + H 2 S CaS + H 2 O ) i n which the s o l i d product CaS i s formed, pores i n the CaO become p lugged and i n c o m p l e t e s o r b e n t conversion i s to be expected. 2.4 GASIFICATION As descr ibed i n Sect ion 2.2, hydrogen may be used to desulphurize o i l sand coke p a r t i c l e s . Steam g a s i f i c a t i o n produces H 2 which i n turn r e a c t s w i t h s u l p h u r i n the coke s tructure to form H 2 S i n the gas mixture . In 1972, Syncrude [6] reported the f i r s t economic and design study of s team/air g a s i f i c a t i o n of i t s f l u i d coke to produce low value f u e l gas. This opt ion d i d not m a t e r i a l i z e because of the lack of a v a i l a b l e technology and a low economic r e t u r n . The s tudy a l s o showed a r e l a t i v e l y large p o r t i o n of the c a p i t a l investment was s l a t e d for the d e s u l p h u r i z a t i o n of the produced gases. Any technology i n g a s i f i c a t i o n that could u t i l i z e the sulphurous o i l sand cokes to produce a c lean f u e l gas at low cost would benef i t the o i l sand indus try . One of the p o t e n t i a l options i s the process of atmospheric pressure steam g a s i f i c a t i o n , with i n - s i t u desu lphur iza t ion employing inexpensive limestone or dolomite. Furimsky's [64] i n v e s t i g a t i o n of the r e a c t i v i t y of o i l sand cokes i n bench-scale atmospheric f ixed and f l u i d i z e d beds revealed that the 2 0 coke r e a c t i v i t y was s e n s i t i v e to both reac t ion temperature and type of g a s i f y i n g medium. Carbon molar conversion in 1 hour increased from 31.7 to 44.5% i n response to a change in temperature from 8 3 0 ° C to 1 0 1 0 ° C . The g a s i f i c a t i o n r e a c t i v i t y was found to improve wi th the presence of oxygen i n s t eam-gas i f i ca t ion of cokes. The study, however, was not extensive enough to cover many other important operat ing condi t ions such as p a r t i c l e s i zes and longer operating times. Watkinson et a l . [189] have s u c c e s s f u l l y g a s i f i e d o i l sand cokes in a p i l o t - s c a l e s p o u t - f l u i d bed, i n steam and oxygen atmospheres, with sulphur being captured i n - s i t u by limestone and dolomite . At 9 5 0 ° C , the t y p i c a l produced gas composition was 35% H 2 , 25% CO and 40% C O 2 , which y i e l d s an average heating value of 7.6 MJ/ra^. An average of 85% carbon conversion was achieved. Moreover, when dolomite was employed as bed m a t e r i a l , over 95% sulphur reduct ion i n the gas phase was recorded compared to the case when no sorbent was used. The g a s i f i c a t i o n process of coal and char has been extens ive ly s tudied [91 ,125 ,143 ,191 ,38 ,98 ,189 ] . Steam g a s i f i c a t i o n of carbon a c t u a l l y produces an e q u i l i b r i u m mixture of CO, H 2 , C H 4 , H 2 O and C O 2 [115], and involves severa l poss ib le react ions i n c l u d i n g : G a s i f i c a t i o n reac t ions : A H 1000K (kJ/mol) R e l a t i v e r e a c t i o n ra te , 1000K, 0.1 atm [43] C(s) + H 2 0 ( v ) - C 0 ( g ) + H 2 ( g ) +135.9 3 (1) C(s) + C 0 2 ( g ) - 2 C 0 ( g ) +172.5 1 (2) C(s) + 2 H 2 ( g ) - C H 4 ( g ) -89.5 (3) 2C( S ) + 2 H 2 0 ( v ) - C 0 2 ( g ) + C H 4 ( g ) + 11.6 very slow (4) 21 Gas s h i f t r e a c t i o n : A H 1000K (kJ/mol) R e l a t i v e r e a c t i o n r a t e , 1000K, 0.1 atm [143] c a t . C 0 ( g ) + H 2 0 ( g ) - C 0 2 ( g ) + H 2 ( g ) -34.7 slow, unless cata lyzed (5) P o l l u t a n t formation r e a c t i o n : _M*298K_ H2(g) + S ( S ) •» H 2 S ( g ) -65.0 (6) In p r a c t i c e , the carbonaceous compounds such as c o a l s , chars and o i l sand cokes t h a t undergo g a s i f i c a t i o n are u s u a l l y not a pure substance but a mixture of organic matter and inorganic mineral matter [107]. When the o i l sand coke conta in ing v o l a t i l e matter i s i n i t i a l l y subjected to an e levated temperature, a s er i e s of complex p h y s i c a l and c h e m i c a l changes o c c u r i n i t s s t r u c t u r e , accompanied by thermal p y r o l y s i s react ions which r e s u l t i n d e v o l a t i l i z a t i o n of c e r t a i n coke components. The mechanism of p y r o l y s i s i s not yet f u l l y e s t a b l i s h e d . I t i s , however, known that the rate of p y r o l y s i s and the l e v e l and composition of v o l a t i l e products from a given sample depend upon many f a c t o r s such as temperature, pressure , p a r t i c l e s i z e , heating r a t e , the h i g h e s t t empera ture t h a t the s o l i d had been t r e a t e d at and the environment i n which the thermal p y r o l y s i s takes place [49,50] . The d e v o l a t i l i z a t i o n can be considered very rap id at temperatures greater than 1000K [49,174,50,104]. In a d d i t i o n to thermal p y r o l y s i s , the formation of methane occurs . The amount of carbon converted to methane i n t h i s per iod increases s i g n i f i c a n t l y with the a d d i t i o n of H 2 . A f t e r the d e v o l a t i l i z a t i o n and methane formation per iod i s ended, the r e a c t i o n between the gaseous r e a c t a n t and the r e m a i n i n g char occurs at a r e l a t i v e l y s lower r a t e than t h a t i n the d e v o l a t i l i z a t i o n p e r i o d 22 [125,96,174,104]. A l l g a s i f i c a t i o n react ions are slow i n the absence of c a t a l y s t , although rates can be increased up to a po int by increas ing pressure . However, at high pressure the rate of carbon g a s i f i c a t i o n by steam tends to zeroth order [187]; f u r t h e r , e q u i l i b r i u m conversion i s reduced with increas ing pressure . The rate of carbon g a s i f i c a t i o n i s a s t r o n g f u n c t i o n of t emperature [187] , but i n c r e a s i n g temperature s e v e r e l y l i m i t s the e q u i l i b r i u m p r o d u c t i o n of C H 4 , e s p e c i a l l y at temperatures above 1000K [189]. 2.4.1 WATER-GAS SHIFT REACTION S y n t h e t i c gas ( H 2 + CO) can be s h i f t e d with steam to obta in hydrogen v i a reac t ion (5) [125]. The water gas s h i f t r e a c t i o n i s the summary of a large number of r e l a t i v e l y simple r e a c t i o n s , which take place s imultaneously; the number of intermediate r e a c t i o n steps can be as many as 12 [103]. Young [20] reported that the mineral matter i n some coal chars has been found to cata lyze the water-gas s h i f t r eac t ion e f f e c t i v e l y , that the reac t ion r a p i d l y reaches e q u i l i b r i u m at 6 5 0 ° C for a contact time on the order of 1 second, and that the extent of the s h i f t r e a c t i o n i s a funct ion of temperature. Even though e q u i l i b r i u m favors the water s h i f t r eac t ion to produce H 2 and C O 2 , equation (5) , an appropriate c a t a l y s t i s required to provide adequate ac t ive s i t e s . MgO i s recognized as a good c a t a l y s t for the water gas r e a c t i o n [185]. 2.4.2 EFFECTS ON GASIFICATION REACTIONS The c a t a l y t i c rate enhancement by a l k a l i n e ear th metal (Mg, Ca, Ba, S r ) compounds i n the H 2 0 - C g a s i f i c a t i o n to promote the C O - s h i f t 23 reac t ion has been observed [107,144]. An increase of 450 times i n the C - C O 2 g a s i f i c a t i o n rate at 893K i n the presence of 100 ppm N i was reported . Nevertheless , the e f f ec t of increas ing concentrat ion of N i l e v e l s o f f towards a sa turat ion value at around 1000 ppm. For Fe t h i s concentrat ion was 350 ppm [124]. There are c o n f l i c t i n g reports on the r e l a t i v e a c t i v i t i e s of metals as c a t a l y s t s i n g a s i f i c a t i o n r e a c t i o n s . Marsh and Ada ir [124] found the c a t a l y t i c e f f e c t on g a s i f i c a t i o n of carbon i n decreasing order are N i , Co, Cu, Ag, Fe and C a . On the other hand, Walker et a l . [188] reported the order to be Fe , Co and N i for g a s i f i c a t i o n o f g r a p h i t e by carbon d i o x i d e . S i m i l a r s t u d i e s [143,152,185] have e s tab l i shed t h a t the a d d i t i o n of c a l c i u m based sorbents in to the g a s i f i e r improves the performance of the integrated g a s i f I c a t i o n - d e s u l p h u r i z a t i o n p r o c e s s e s . However, to i n c r e a s e the c a t a l y t i c e f fec t on carbon convers ion, MgO and CaO should be in t imate ly d ispersed i n the coke [96,189,144,115]. Walker [185,186] d e s c r i b e d calc ium as an e f f e c t i v e c a t a l y s t not only for the g a s i f i c a t i o n r e a c t i o n , but a lso f o r the w a t e r - g a s s h i f t r e a c t i o n . C a l c i u m may r e q u i r e chemical ly ac t ive groups to become act ive [115]. The e f f ec t s of the composition of gas i fy ing atmosphere on the gaseous product , on the r e a c t i v i t y and on the carbon conversion are we l l documented [ 6 4 , 1 2 7 , 2 0 0 , 1 0 7 , 1 6 3 , 1 8 9 ] . The i n h i b i t i n g e f f e c t s of r e d u c t i v e agents such as hydrogen and carbon monoxide have been inves t igated ex tens ive ly . Petroleum coke was g a s i f i e d i n steam in the presence of added H 2 at about 9 2 7 ° C [163]. At a feed gas mol r a t i o 0.1<H2/H20<0.5 the added H 2 i n h i b i t s the g a s i f i c a t i o n reac t ion p r i m a r i l y by the oxygen exchange mechanism, w i t h some s l o w l y r e v e r s i b l e 24 c o n t r i b u t i o n from competing hydrogen chemisorption [163]. The study of K a i r a i t i s [100] showed that at about 1200K, the i n h i b i t i n g e f fec t of H 2 , and the r e l a t i v e r e a c t i o n rate decreased i n inverse proport ion to the H 2 concentrat ion . However, a d d i t i o n a l increases of H 2 to br ing the molar r a t i o H 2 / H 2 O > 0 . 5 , i n the presence of constant water, increased the product ion of CO d r a s t i c a l l y , while C 0 2 product ion decreased. The net o v e r a l l g a s i f i c a t i o n rate therefore i n c r e a s e d . The change i n the g a s i f i c a t i o n r a t e can be e x p l a i n e d as f o l l o w s . With the H 2 / H 2 0 r a t i o >0.5, the inorganic impur i t i e s i n the coke were presumably reduced to t h e i r ac t ive state and so cata lyzed the CO product ion , whereas at H 2 / H 2 0 < 0 . 5 the coke impur i t i e s l o s t t h e i r c a t a l y t i c a c t i v i t y due to o x i d a t i o n of the m e t a l s [124] . Walker and h i s coworkers [185] in terpre ted the increase i n product ion of CO as due to the d i s s o c i a t i v e adsorpt ion of H 2 0 . Other inves t iga tors [107,163,144] have made s i m i l a r observations of the c a t a l y s i s of carbon g a s i f i c a t i o n by a l k a l i and t r a n s i t i o n s metals which are found i n petroleum cokes. Steam g a s i f i c a t i o n of sulphurous cokes produces hydrogen sulphide which i s both tox i c and a p o l l u t a n t . The e f fec t of H 2 S , up to a H 2 S concentrat ion of 400 ppm, on the g a s i f i c a t i o n reac t ion was measured by Kasaoka et a l . [106,109] who r e c e n t l y suggested that the presence of H 2 S i n the g a s i f i e r has l i t t l e e f fec t on the rate of g a s i f i c a t i o n . The o v e r a l l c h e m i c a l or i n t r i n s i c r e a c t i v i t y of carbonaceous p a r t i c l e s depends on t h e i r nature, the concentrat ion of ac t ive s i t e s on the carbon s u r f a c e [ 5 4 , 3 4 ] , the a c c e s s i b i l i t y of ac t ive s i t e s to reac t ive gas, and the c a t a l y s i s by m i n e r a l m a t t e r [ 5 4 , 2 0 1 ] . The a c c e s s i b i l i t y of ac t ive s i t e s can p o s s i b l y be determined by using oxygen 25 s o r p t i o n . F a i r b r i d g e et a l . [54] employed the oxygen chemisorpt ion method to obta in quant i ta t ive information on the nature of ox idat ion s i t e s of o i l sand coke surfaces; the information i s thence r e l a t e d to the g a s i f i c a t i o n r e a c t i v i t y . Studies , us ing graphite as the carbon-bearing s o l i d , ind icated that the react ions invo lv ing non-atomic species occur with the edge-wise carbon. In a carbonaceous aromatic matrix such as coa l and coke, the a c c e s s i b i l i t y of unsaturated chemical bonds and the dense concentrat ion of inorganic impur i t i e s at c r y s t a l l i t e edges cause the edge carbon atoms to be more r e a c t i v e than t h e i r b a s a l counterparts [185,181]. While the r e a c t i v i t y of a carbonaceous surface depends upon i t s inhomogeneity, the g a s i f i c a t i o n process causes the s t r u c t u r e of s o l i d and the c o n c e n t r a t i o n of i m p u r i t i e s to change markedly, which a l t e r s the surface concentrat ion of ac t ive s i t e s [127]. The r e a c t i v i t y during g a s i f i c a t i o n i s therefore not constant [107]. However, the rate of g a s i f i c a t i o n and the r e a c t i v i t y of coke appears to be more s t rong ly d i c ta t ed by chemical proper t i e s of the surface of the cokes, which c o n t r o l the surface react ions between the gaseous reactants and the s o l i d surface , than the composition and the amount of mineral matter [104]. Working with coal derived cokes, Walker et a l . [186] found that the presence of nat ive c a t a l y s t s and the possession of a h igh concentrat ion of carbon ac t ive s i t e s are necessary but not s u f f i c i e n t to achieve high r e a c t i v i t y . The act ive s i t e s , located mainly i n the micropores , must be we l l connected to the e x t e r i o r surface of the p a r t i c l e s by t r a n s i t i o n a l or raacropores (feeder pores) through which the reactant gas can r a p i d l y d i f f u s e ; otherwise the ac t ive s i t e s w i l l not be a v a i l a b l e [186]. 26 2.4.3 KINETIC MODELS FOR GASIFICATION The most common o x i d i z i n g agents employed i n g a s i f i c a t i o n are C O 2 and steam. The inf luence of steam concentrat ion , as the g a s i f y i n g medium, on the r e a c t i v i t y has been widely s tudied [191,38,104,127]. In a g a s i f i e r , the presence i n the product gas of H 2 , C O 2 , CO and C H 4 can be s i g n i f i c a n t w i t h r e s p e c t to the i n h i b i t i o n of the H20-Carbon r e a c t i o n , the Boudouard r e a c t i o n , and the h y d r o g a s i f i c a t i o n r e a c t i o n . Using d i f f e r e n t sets of elementary r e a c t i o n steps , many authors have w r i t t e n about the mechanism of the simple g a s i f i c a t i o n react ions and come up w i t h a s i n g l e L a n g m u i r - H i n s h e l w o o d - t y p e r a t e f o r steam g a s i f i c a t i o n at atmospheric pressure [127] r = (7) 1 + V H 2 0 + C 1 P H 2 ( where: d X C -2/3 r m - <8> f o r s p h e r i c a l p a r t i c l e s . Xq i s carbon conversion def ined i n equation (11). Shaw [155] combined some elementary r e a c t i o n steps proposed by others and added some new ones to der ive an equation which expressed the reac t ion k i n e t i c s in steam and C O 2 g a s i f i c a t i o n as: 27 a 2 F C 0 2 + b 2 E H 2 0 + C 2 P 2 C 0 2 + d 2 P H 2 0 E C 0 2 + ^ H . O + f 2 P C 0 2 P H 2 + 8 2 P H 2 0 P H 2 r (9) 1 + h 2 P C 0 2 + + j 2 P C 0 + k 2 P H 2 This equation has not been tes ted by experimental data [127], perhaps because of i t s complexity. Assuming the r e a c t i o n between char and steam proceeds uni formly throughout the i n t e r i o r surfaces of the p a r t i c l e s [93], the rate of disappearance of carbon per u n i t mass due to r e a c t i o n with steam for k i n e t i c c o n t r o l condi t ions [38,193] can be expressed as: d X c / dt = b k , p H 2 0 ( i - X c ) (10) Here, represents the conversion of f ixed carbon i n the sample and i s defined as: X = (W - W - W) / (W - W - W u ) (11) C o v o v ash WQ i s the i n i t i a l weight of s o l i d and W, W v and W a s n are the weights of f ixed carbon, v o l a t i l e s and ash present i n the s o l i d , r e s p e c t i v e l y , given i n the proximate a n a l y s i s . The rate constant k' i s f or r e a c t i o n occurr ing uniformly throughout the i n t e r i o r of the s o l i d p a r t i c l e s and b i s the s t o i c h i o m e t r i c c o e f f i c i e n t f o r the r e a c t i o n : a A ( g ) + b B ( s ) •• Product (12) In equation (10) the r a t e of r e a c t i o n i s assumed to be l i n e a r l y propor t iona l to the steam p a r t i a l pressure . The r a t e , dX^/dt , can be obtained from the observed change i n weight with time dW/dt as fo l lows: 28 dXc/dt - - ( d W / d t ) / ( W 0 - W v - W a s h ) (13) The amount of v o l a t i l e matter i n the p a r t i c l e i s i d e n t i c a l w i th the amount released during the d e v o l a t i l i z a t i o n process [81]. In a separate study Kasaoka et a l . [104] suggested that the rate of g a s i f i c a t i o n was p r o p o r t i o n a l to the p a r t i a l pressure of steam to the power of between 0.43 and 0.67. I n v e s t i g a t o r s [2 ,49] who s t u d i e d the k i n e t i c s of g a s i f i c a t i o n observed that the shapes of the ra te -convers ion c u r v e s f o r v a r i o u s samples d i f f e r e d s i g n i f i c a n t l y . The v a r i e t y of the rate -convers ion curves i s due to the fac t that d i f f e r e n t carbonaceous samples v a r y g r e a t l y from one another with respect to t h e i r pore s t r u c t u r e s , pore surface area and the changes of such porous c h a r a c t e r i s t i c s with carbon conversion and temperature. Walker et a l . [187] postulated that the ranges of temperature at which reac t ion occurs determine the rate c o n t r o l l i n g mechanism. At low temperature, the r e a c t i o n rate i s very slow; the r e a c t i o n i s therefore c h e m i c a l l y c o n t r o l l e d . I f the r e a c t i o n t e m p e r a t u r e i s i n t h e intermediate range, the o v e r a l l rate i s l i m i t e d by the rate of d i f f u s i o n of the gaseous reactants through the pore s t ruc ture of the carbon. When the reac t ion temperature i s h igh , the rate of r e a c t i o n becomes very r a p i d ; and the rate of steam consumption i s more than what can be s u p p l i e d by the bulk gas phase. The mass transport of H 2 O to the external surface of the p a r t i c l e therefore d i c t a t e s the r e a c t i o n . In the l a s t case the rate i s independent of the carbon r e a c t i v i t y . 29 Recent inves t iga t ions [3,104, 107] of g a s i f i c a t i o n k i n e t i c s of carbon ind ica ted that the i n t r a p a r t i c l e d i f f u s i o n res i s tance could be ignored, at l eas t up to the temperature of 1273K f o r p a r t i c l e s i n the range of 0.6 mm diameter. Extensive inves t iga t ions have been c a r r i e d out on the modelling of the dynamic change i n the rate as a funct ion of conversion f o r non-c a t a l y t i c r e a c t i o n between s o l i d and gases [2 ,49] , Consequently a number of s t r u c t u r a l models for gas s o l i d r e a c t i o n have been proposed. T y p i c a l examples are the Shrinking Core Model [118], the Continuous Model proposed by Ishida and Wen [93], and the Gra in Model by Szekely and Evans [176,177,165]. The Continuous Model (CM) assumes that the reactant gas enters and reacts throughout the p a r t i c l e at a l l t ime, most l i k e l y at d i f f e r e n t r a t e s at d i f f e r e n t l o c a t i o n w i t h i n the p a r t i c l e . Thus, the s o l i d reactant i s converted continuously and p r o g r e s s i v e l y t h r o u g h o u t the p a r t i c l e . With t h i s model, f i r s t - o r d e r k i n e t i c s with respect to calc ium oxide have been widely used to descr ibe the r e a c t i o n between c a l c i n e d d o l o m i t e and c a l c i n e d l i m e s t o n e w i t h h y d r o g e n s u l p h i d e gas [152,183,101,137,131,149]. The more commonly used G r a i n Model (GM) has been s u c c e s s f u l l y appl ied to n i c k e l oxide p e l l e t s wi th hydrogen [178], to the s u l p h i d a t i o n reac t ion of ca lc ium oxide [27], to the su lphat ion of lime [81] as we l l as to several other g a s - s o l i d systems. T h i s model, however, p r e d i c t s a m o n o t o n i c a l l y d e c r e a s i n g s u r f a c e a r e a d u r i n g g a s i f i c a t i o n , while c o a l , chars and cokes often show a maximum i n the surface area with conversion [86,2 ,171,3 ,104] . The GM does not account f o r such changes i n porous s tructure during the course of r e a c t i o n . A 30 v a r i a t i o n of the s p h e r i c a l g r a i n model has a lso been proposed by Park and L e v e n s p i e l [134] whose c r a c k i n g core concept does al low for s igmoidal conversion time curves with the use of an e x t r a f i t t i n g parameter. In s i m i l a r work, Kasaoka et a l . [104,105] have used the representat ion of a volumetr ic or continuous model, and put forward an e m p i r i c a l mathematical express ion, c a l l e d the Modif ied Volumetric model (MVM), which i s capable of p r e d i c t i n g the g a s i f i c a t i o n rates for which a s i g m o i d a l c h a r a c t e r w i l l be present . Another e m p i r i c a l model was introduced by Simons et a l . [159] and Simons [160] us ing the ideas of b r a n c h i n g and combin ing of pores w i t h a r e s t r i c t i v e pore s i z e d i s t r i b u t i o n , which i s not always obeyed in g a s i f i c a t i o n of carbonaceous s o l i d s . The above mentioned models have s tressed the d e s c r i p t i o n of the progress of the conversion of the s o l i d s , and have not pa id a t t e n t i o n to changes i n porous s t r u c t u r e . Peterson [140] was the f i r s t to take account of changes i n the pore s t r u c t u r e . The developing surface was modeled as a sum of c y l i n d r i c a l pore contr ibut ions reduced by the sum of the e f f ec t s of the random i n t e r s e c t i o n s [138]. This model, assuming the pores to be of uniform s i z e , was commonly used f o r the ana lys i s of g a s i f i c a t i o n r e a c t i v i t y , although other inves t iga tors [22] reported ly employed i t for react ions with a s o l i d product . T h i s comparatively simple model, however, was not su i tab l e for coa l char and cokes whose pore s i ze d i s t r i b u t i o n s are bimodal, often even t r i m o d a l , and cannot be represented by a uniform pore s i ze model. More r e c e n t l y Bhat ia and Perlmutter [17] and Gavalas [66] have r e s p e c t i v e l y proposed a Random Pore Model (RPM) and a Random C a p i l l a r y Model (RCM) in which the pores 31 are considered to have a r b i t r a r y s i ze d i s t r i b u t i o n s , and the models therefore can p r e d i c t surface maximum with convers ion . In the RPM, the pore s i ze d i s t r i b u t i o n i s represented by a s t r u c t u r a l parameter which i s assumed to be a p o s i t i v e constant and can be estimated from the i n i t i a l surface area , poros i ty and pore length . This model has been tes ted for coals and chars by other workers [2,171,104]. Su and Perlmutter [171] reported that the RPM has succes s fu l l y descr ibed the e f f e c t of pore s t ruc ture on char ox ida t ion; while Kasaoka et a l . [104] had d i f f i c u l t i e s i n applying the model to the k i n e t i c s of steam and CO2 g a s i f i c a t i o n of coa l chars . T h e i r reported values for the s t r u c t u r a l parameter of the same char v a r i e d f i v e f o l d and i n some cases, gave negative va lues , which are inadmiss ib le . The RCM i s a mathematical model employing s t a t i s t i c a l manipulat ions and i s developed from the concepts of pore i n t e r s e c t i o n and overlap volume which induce the changes i n pore volume and surface area with extent of r eac t ion i n many carbonaceous s o l i d s . The RCM has been s u c c e s s f u l l y tes ted with experimental g a s i f i c a t i o n data of Majahan et a l . [92] and Dutta and Wen [49]. Recently a d d i t i o n a l models have been proposed by Lee et a l . [116] and B a l l a l et a l . [10]. These models take in to account the major p h y s i c a l fac tors which inf luence the g a s i f i c a t i o n ra te : the changing magnitudes of surface area, p o r o s i t y , a c t i v a t i o n energy and e f f e c t i v e d i f f u s i v i t y during g a s i f i c a t i o n . They, however, c o n s i s t of numerous adjustable parameters that are not measurable and therefore cannot be r e a d i l y a p p l i e d . From the d i s c u s s i o n , the RPM of Bhatia and Perlmutter [17] and the RCM of Gavalas [66] appear to be the most a p p r o p r i a t e models f o r 32 carbonaceous s o l i d s , such as o i l sand cokes, because of the a t t r i b u t e s mentioned above. In the f o l l o w i n g s e c t i o n , the RPM for which the experimental information on i n i t i a l porous s tructure i s e s s e n t i a l i s reported . This i s fol lowed by a sec t ion i n which the RCM, being der ived by mainly using mathematical and s t a t i s t i c a l a p p l i c a t i o n s , i s d iscussed i n d e t a i l . The MVM, a pure ly empir i ca l model, w i l l be then descr ibed . The GM and the CM employed f o r the s u l p h i d a t i o n r e a c t i o n i s d e t a i l e d i n Sections 2.5.6 and 2.5.8 r e s p e c t i v e l y . 2 .4.4 RANDOM PORE MODEL A v a r i e t y of s tudies on g a s i f i c a t i o n [86,50,116] have shown that the rate of r eac t ion v a r i e s with conversion and shows a maximum rate at an intermediate l e v e l of convers ion. The s t r u c t u r a l proper t i e s of the s o l i d reactant a lso change cont inuously as r e a c t i o n proceeds, and t h i s e v o l u t i o n of pore s t r u c t u r e s t r o n g l y a f f ec t s r e a c t i o n k i n e t i c s by changing the surface area a v a i l a b l e f o r r e a c t i o n . Results from these s tudies ind ica ted that for c e r t a i n pore c h a r a c t e r i s t i c s , the i n i t i a l increase i n the g a s i f i c a t i o n r a t e i s a t t r i b u t e d to the growth of r e a c t i o n surfaces from the i n i t i a l pores that are the passages by which the gas has access to the react ing s o l i d . However, t h i s e f f ec t i s overshadowed at higher conversion by the i n t e r s e c t i o n s of the growing pores , r e s u l t i n g eventual ly in net decreases i n t o t a l surface area which i n turn acts to decrease the reac t ion r a t e . The pore s tructure of s o l i d reactant may be charac ter i zed by i t s pore s i ze d i s t r i b u t i o n , pore volume and s p e c i f i c surface area. The pore 33 s i ze d i s t r i b u t i o n and pore volume can be obtained exper imenta l ly by mercury porosimetry or by adsorpt ion-desorpt ion techniques. The pore volume can also be es t imated by u s i n g the m e r c u r y - h e l i u m d e n s i t y measurement. Other proper t i e s such as the surface area can be d i r e c t l y measured by the BET method [170], To account for the e f fec t of changes i n pore s t r u c t u r e , the RPM was proposed, on the bas is of i n i t i a l pore s tructure informat ion, by Bhat ia and P e r l m u t t e r [ 1 7 ] . The RPM a l l o w s f o r an a r b i t r a r y pore s ize d i s t r i b u t i o n i n the react ing s o l i d . Consider the isothermal chemical reac t ion of s o l i d p a r t i c l e B with reactant gas A according to the s to i ch iometr i c r e a c t i o n (12). a A ( g ) + b B ( s ) •* c P ( g ) + d Q ( s ) (14) The r e a c t i o n i s i n i t i a t e d on the surface of pores i n the s o l i d B . As the r e a c t i o n proceeds the product layer Q i s formed around each pore which separates the growing pore surface from gaseous reactant A w i t h i n the pores . To react wi th the s o l i d B, the reactant A now has to d i f fuse through the product layer Q to the reac t ion surface where c h e m i c a l change takes p lace . I t i s assumed that the s o l i d B has no s i g n i f i c a n t c losed pore volume; and that the r e a c t i o n occurs i n the k i n e t i c regime i n which a l l d i f f u s i o n a l res i s tances are assumed n e g l i g i b l e . As the r e a c t i o n cont inues , each pore i n the s o l i d p a r t i c l e B has assoc iated with i t a growing r e a c t i o n surface which i n i t i a l l y was the inner surface of the pore. The neighboring growing surfaces w i l l eventual ly in t er sec t one another as the s o l i d separat ing them i s consumed by reac t ion and replaced by product Q. The actual reac t ion surface of the s o l i d B at 34 any time i s formed by the random overlapping of a set of c y l i n d r i c a l surfaces of s i ze d i s t r i b u t i o n f ( r ) where f ( r ) d r i s the t o t a l length of the c y l i n d r i c a l surfaces per un i t volume of space having r a d i i between r and r+dr. The t o t a l l e n g t h , t o t a l s u r f a c e and t o t a l enclosed volume of nonoverlapped c y l i n d r i c a l system are defined as: L E = | f ( r ) dr (15) CO 4 S E = 2-rr I r f ( r ) dr (16) OO r f ( r ) dr (17) Assuming that no c y l i n d e r s are created or destroyed, the balance over the s i ze d i s t r i b u t i o n of the growing c y l i n d e r s r e s u l t s i n : 6f § _ at + 3r f * E dt (18) If the r e a c t i o n rate on the surface i s p r o p o r t i o n a l to the surface area thus: dr , „n XT = k C dt s (19) 35 Because each element of the ac tua l surface must belong to the non-o v e r l a p p e d p o r t i o n of some c y l i n d e r . Combining equation (18) and equation (19) y i e l d s 9f , _n 3f . „ „ . aT = - k s C o7 ( 2 0 ) Using equation (15) together with f(0) = f ( ° ° ) = 0, the i n t e g r a t i o n of equation (20) wi th respect to r y i e l d s : D L E "dE = 0 ( 2 1 ) Employing equation (15) and (16), m u l t i p l y i n g equation (20) by 2nr, and in tegra t ing wi th respect to r , we may write IT " ^ s ^ E ( 2 2 ) S i m i l a r l y , m u l t i p l y i n g equation (20) by irr^, i n t e g r a t i n g wi th respect to r and s u b s t i t u t i n g equation (17), we obtain an equation for the t o t a l volume: dV^ E . „n dt - k s C S E ( 2 2 ) S o l v i n g e q u a t i o n s (21) and (22) y i e l d s the evo lut ion of the non-overlapped surface SJJ and volume Vg as conversion proceeds: 36 S E - | S * + 4 « L _ ( V _ - V p ) (23) i f E o E o E E o and V = V + S (k C"t) + TTL (k C"t)* (24) The increment in the volume enclosed by ac tua l (or nonoverlapped) system i s o n l y a f r a c t i o n of the growth i n the nonoverlapped c y l i n d r i c a l system. This f r a c t i o n i s (1 -V) , the f r a c t i o n a l volume space occupied by the unreacted s o l i d B dV = ( l - V ) d V E (25) when V E -» 0, V -» 0. Integrat ion of equation (25) gives V = 1 - exp( -V E ) (26) Consider the increase in volume of the overlapped system by an amount dV i n time dt : dV = S ( k s C n ) dt (27) Combining equations (22), (25) and (27) r e s u l t s i n S = S E (1 -V) (28) S i m i l a r l y , L = L E ( 1 - V ) (29) 37 For a s p h e r i c a l p a r t i c l e in the regime of reac t ion c o n t r o l 1-V 1 -k C n t s (30) and the i n i t i a l c o n d i t i o n Vo = e. (31) The c o m b i n a t i o n of e q u a t i o n (23 ) , ( 24 ) , ( 2 6 ) , (28-31) y i e l d s a dimensionless form 1 - \|iLn 1-Xr. (1-x/o)- (32) and X c = 1 - ( l - t / o r exp [-c (1 + I|>T:/4)] (33) where \|i = ^-Hto_LL^.o l _ a s t r u c t u r a l parameter (34) o = RoSo 1 - E O a p a r t i c l e s ize parameter (35) and k C n S t s o 1 - E O = a k i n e t i c parameter (36) I t has been shown that where the external surface area i s n e g l i g i b l e compared to i n t e r n a l pore surface , o = a > . In t h i s case 38 X c - 1 - exp [-z (1 + *t /4) ] (37) and d X c dt S So = (1-: 1-\|> L n ( l - X c ) (38) The maximum i n r e a c t i o n surface a r i s e s from the two opposing e f f e c t s , the growth of r e a c t i o n surface associated with the pores and the loss of the surfaces as they progres s ive ly co l lapse by over lapping . The l a t t e r e f f e c t dominates the o v e r a l l r e s u l t s when the v a l u e of \J> i s s m a l l enough. The convers ion, X^s* when the surface area a t ta ins a maximum can be obtained from the d i f f e r e n t i a t i o n of equation (32) y e i l d i n g The value of v|> i n the RPM can be determined by independent methods. \|> can be obtained from d i r e c t measurements on the i n i t i a l pore s t ruc ture of the s o l i d . A l t e r n a t i v e l y , i f \|> can be determined from experimental k i n e t i c data using the l i n e a r form of equation (38) [171]: X C S = 1 " exp [(l-^)/2>l>] (39) Thi s maximum e x i s t s over 2 < \|» < <*> and reveals that 0 < X C S < 0.393 (40) 39 Thus from the RPM one can obta in the i n t r i n s i c reac t ion rate k s C n from the in tercept at XQ=0 and the pore s tructure parameter \|> from the r a t i o of s lope to i n t e r c e p t . One important feature of the RPM i s that i t p r e d i c t s the reac t ion surface area at any given conversion as a funct ion of i n i t i a l pore s t ruc ture parameters. 2 .4 .5 RANDOM CAPILLARY MODEL The porous p a r t i c l e s are descr ibed by a random c a p i l l a r y model p r e d i c t i n g the frequency of pore in tersec t ions and the evo lu t ion of pore volume and surface area during r e a c t i o n by a s ing l e dens i ty func t ion , r e l a t e d but not i d e n t i c a l to the customary pore s i ze d i s t r i b u t i o n . The development of the RCM has been f u l l y descr ibed elsewhere [69]. For our i n t e r e s t , only the a p p l i c a t i o n of the model to char g a s i f i c a t i o n at chemical ly c o n t r o l l e d rates i s mentioned here. In t h i s model, the porous s tructure i s assumed to c o n s i s t of i n f i n i t e l y long, s t r a i g h t , c y l i n d r i c a l c a p i l l a r i e s or pores with r a d i i i n a range r^<r<r*, and the axes of the c a p i l l a r i e s are located c o m p l e t e l y randomly . Gan et a l . [65] sugges ted the broad s i z e d i s t r i b u t i o n of pores i n c o a l , chars and cokes can be d i v i d e d into three r a n g e s , the m i c r o p o r e s ( r < 6 x l 0 - ^ um), the t r a n s i t i o n a l pores ( 6 x l 0 - ^ < r < 1 .5xl0 _ 2 um) and the macropores ( 1 .5x l0 - 2 < r < 1.5 um). The d i s t i n c t i o n between the three ranges i s a r b i t r a r y , however. For i n s t a n c e , the d i v i s i o n between micropores and t r a n s i t i o n a l pores i s based on the change from very slow a c t i v a t e d d i f f u s i o n to Knudsen d i f f u s i o n . However, the r a t e of d i f f u s i o n i s a func t ion of many 40 parameters, such as the type, shape and p o l a r i t y of the d i f f u s i n g m o l e c u l e s . On the other hand, the d i s t i n c t i o n between t r a n s i t i o n a l pores and macropores depends upon the porosimetry techniques used for these two pore s i ze ranges. The term chemical c o n t r o l i s only r e l a t i v e and i n d i c a t i v e . It i s assumed that the c o n t r i b u t i o n of the micropores to the o v e r a l l reac t ion rate per u n i t of t r a n s i t i o n a l pore and macropores area i s independent on carbon convers ion. Since the ac t iva ted d i f f u s i o n i n micropores i s very slow, the reac t ion occurs mostly i n a small region near the i n t e r s e c t i o n between the t r a n s i t i o n a l pores or the macropores and the micropores . As the r e a c t i o n proceeds, the surface of these former pores i s consumed, and exposes more micropore area i n such a way that the r e a c t i o n rate due to the micropores, per un i t area of the l a r g e r pores , i s constant . It i s therefore assumed that while the d i f f u s i o n i n the micropores i s very slow, the d i f f u s i o n i n the t r a n s i t i o n a l and macropores i s very r a p i d . Hence a l l the surface area of pores in the t r a n s i t i o n a l and macropores are exposed to a uniform concentrat ion of the reactant gases. These are condi t ions for chemical ly c o n t r o l l e d ra tes . The v e l o c i t y with which a pore surface element recedes owing to r e a c t i o n i s : v ( C , C T) - - ~ r (C,C ,T) (42) s p s s s where r s ( C , C s , T ) i s the r e a c t i o n r a t e , i n grams of c a r b o n / u n i t area*time. r s i s a funct ion of the concentrat ion of reactant gas (C) , temp. (T) and concentrat ion C s of act ive s i t e s which re la t e s to the 41 e f f e c t i v e mineral c a t a l y s t s . The rate i s expressed per u n i t area of pores inc lud ing macro, t r a n s i t i o n a l and micropores, and p s = P c / ( l - c T ) ( 4 3 ) i s the dens i ty i n the p o r t i o n of the volume excluding t r a n s i t i o n a l pores and macropores and i s a constant . p c i s the apparent (mercury) dens i ty of carbon in the char. The a d d i t i o n a l assumption i s that C s does not vary during the course of g a s i f i c a t i o n . As a r e s u l t of the above a s s u m p t i o n , the quant i ty v depends on T , C and the p a r t i c u l a r r e a c t i o n but not on carbon convers ion, X^. Using the c h a r a c t e r i s t i c length L c in h i s mathematical argument, Gavalas [69] der ived the f u n c t i o n a l r e l a t i o n s h i p that expresses the pore surface area, the pore volume and the conversion as a u n i v e r s a l funct ion of t v / L ~ . As f o r carbon conversion ( 4 4 ) From equation ( 4 4 ) a number of important conclus ions fo l low: 1 - For a given char, conversion-t ime curves corresponding to d i f f e r e n t react ing gases, pressure and temperatures can be reduced to a s i n g l e master curve by us ing the dimensionless time t ' = t v / L c . S ince v i s u s u a l l y unknown, i t i s c o n v e n i e n t to d e f i n e a dimensionless time at a g iven convers ion, X c = 0 . 5 f o r instance . Let t g . 5 be the time at X c = 0 . 5 . From equation ( 4 4 ) 42 tQ. 5 = 0.5 (45) where l o . 5 i - s a c o n s t a n t c h a r a c t e r i s t i c of the i n i t i a l pore s t r u c t u r e . Although In. 5 a i *d v are unknown, t g ^ can be measured exper imental ly . With the dimensionless t" = t / t Q . 5 » w e ob ta in : Xc = X 0.5 c . t" (46) 2 -A l l curves again can be reduced to a s ing le master curve i f p l o t t e d using appropriate dimensionless t ime c o r r e s p o n d i n g to a f i x e d carbon conversion l e v e l . The rate of r eac t ion per u n i t i n i t i a l weight can be w r i t t e n dX^ dt v tv Lc -f~ G(X C ) ^c (47) where the funct ion G i s inf luenced by the char c h a r a c t e r i s t i c s , but i s independent of react ing gas and temperature. Equation (47) impl ies that dXfj/dt vs Xq curves corresponding to d i f f e r e n t react ing gases, temperature, e t c . , can be a l l reduced to a s ing le curve by appropriate s c a l i n g of dXfj/dt. Also i f dXfj/dt a t ta ins a maximum at a c o n v e r s i o n X ^ R , t h i s c o n v e r s i o n i s independent of temperature, react ing gas e t c . , but dependent on the char on quest ion. Equation (48) a lso ind icates that the maximum rate (dXc /dt )^ i s propor t iona l to the reac t ion v e l o c i t y v; therefore i t too can be used to determine the experimental a c t i v a t i o n energy and reac t ion order . 43 Let q ( t ) be denoted as the increase in the radius of any c a p i l l a r y in the time from 0 to t ; which i s : r ( t ) - r 0 + q( t ) (49) The p r o b a b i l i t y dens i ty funct ion at time t i s given i n terms of the i n i t i a l p r o b a b i l i t y dens i ty funct ion Xo: K ( r , q ) = \ o ( r - q ) (50) and the t o t a l pore volume i s r* c T (q ) - 1 - exp [-2nj (RQ+q) X o ( r Q ) d r Q ] (51) which may be wr i t t en as: 1 " c T (q ) 1 - c = exp [ -2TI ( B Q q 2 + 2B x q)] (52) To Where the z e r o t h moment B Q i s the t o t a l number of c a p i l l a r y axes i n t e r s e c t i n g a u n i t surface area. Bo " X ( r Q ) d r Q (53) o* and the f i r s t moment B^ i s equal to the product of B Q and the mean pore radius 44 C° I r 0 \ ( r 0 ) d r 0 Bj = \ r 0 \ ( r 0 ) d r 0 (54) When q increases by a length dq, the pore volume increases by d e T = S(q) dq (55) where S(q) i s the t o t a l surface area . Thus d E T S(q) = = 4ir [1 - e (q) ] [B Q q + B ^ (56) Pore volume and surface area are now a funct ion of a s ing l e v a r i a b l e q. The conversion X^ i s defined as the amount of carbon reacted d iv ided by the i n i t i a l amount of carbon; which i s : Xc(q) = 1 - (1- - 9 - ) 3 exp [ - 2 n ( B 0 q 2 + 2 B i q ) ] (57a) where R Q i s the i n i t i a l p a r t i c l e rad ius . In the chemical c o n t r o l reg ion , i t can be shown that q<<RQ. The est imat ion from the use of the number dens i ty of pore mouths on any surface , X. shows that f o r small E O : X = eo/2-rrr0 (57b) so that: 45 R«-2TC C O 1/2 (58) Ignoring the terms due to small change in p a r t i c l e r a d i u s , the bas ic expression i s der ived: X c ( t ) = 1 - exp [ -2 (B Q q 2 + 2 B i q ) ] (59) when the pore surface r e a c t i v i t y , the temperature and the concentrat ion of react ing as remain constant , then: q = vt (60) v i s a constant given by equation (42). Equation (59) now become: d X c = 4n ( B Q v 2 t + Bxv) exp [ - 2 T i ( B 0 v 2 t 2 + 2 B 1 v t ] ) dt (61) The r e l a t i o n between r e a c t i o n rate and conversion can be obtained by e l i m i n a t i n g t between (60) and (61) to y i e l d : dX 2 — - 4 « < 1 - X c > [ ( B , v ) 2 + B o Y . ^ ^ j 1/2 The reac t ion rate i s a maximum when d 2 Xfj/dt 2 =0, that i s vt = B 0 11 (62) (63) 4 6 The maximum value of dX(Vdt and the corresponding conversion X ^ R can be r e a d i l y obtained by s u b s t i t u t i n g t h i s value of vt into equation ( 5 9 ) and ( 6 2 ) to g ive : D XC ( 1 / 2 B l 2 U L = V ( 4 T T B 0 ) 1 / Z exp [ - ( 1 / 2 - 2 * - r ^ - )] ( 6 4 ) D T > M ° B o and »! X C R = 1 - exp [-(1/2-2H ^ ) ] ( 6 5 ) If B Q <4nBi 2 , the reac t ion rate decreases monotonical ly wi th convers ion. 2 As B 0 >> 4-nB^ and s ince B O > 0 X C R < 1 - e 1 / 2 - 0 . 3 9 3 ( 6 6 ) With the assumptions made of chemical c o n t r o l and no inf luence of carbon conversion on v , experimental values of X Q above 0 . 3 9 3 suggest d e v i a t i o n from the random c a p i l l a r y model. In equation ( 6 2 ) , B Q v 2 and Bjv can be determined us ing g a s i f i c a t i o n k i n e t i c s but not B 0 , B j , and v i n d i v i d u a l l y . B 0 v 2 and B^v can be computed simply by rewr i t ing equations ( 5 9 ) and ( 6 2 ) i n a l i n e a r form. The R C M i s not s u i t a b l e for high poros i ty s o l i d s , and i s not recommended to be appl ied beyond a c e r t a i n upper l i m i t such as X ^ = 0 . 7 . It was reported [ 5 0 ] that at X Q about 0 . 8 , char p a r t i c l e s d i s i n t e g r a t e into fragments. 47 2.4.6 MODIFIED VOLUMETRIC MODEL There i s no c l e a r t h e o r e t i c a l development for the MVM of Kasaoka et a l . [104]; i t i s a pure ly empir i ca l model . However, i t has been employed extens ive ly [105,106,109] for the g a s i f i c a t i o n reac t ion of many coals imported into Japan and many other chars . The c a r b o n convers ion XQ i s defined by equation (11), and the g a s i f i c a t i o n rate i s given by: d X c / d t = k ( X c ) (1 -X C ) (67) The e x p e r i m e n t a l d a t a from i s o t h e r m a l g a s i f i c a t i o n tes t s i s normally shown as a curve of carbon conversion vs g a s i f i c a t i o n time t . The r e l a t i o n between carbon conversion and r e a c t i o n time i s expressed in the MVM as: X c = l - e x p ( - A t B ) (68) where A and B are constants whence k ( X c ) = A ( 1 / B ) B [ - L n ( l - X C ) ] ( B - 1 ) / B (69A) The constant A i s considered to be more c l o s e l y re la t ed to the i n t r i n s i c c h e m i c a l r e a c t i v i t y of the char, while B i s considered to be more c l o s e l y r e l a t e d to i t s p h y s i c a l propert ie s [104]. The l i n e a r i z e d form of equation (68) i s given by: M = Ln A + BN (69B) where: M = Ln [ - L n ( l - X c ) ] (70) 48 and: N = Ln( t ) (71) The experimental data can be subs t i tu ted into t h i s l i n e a r i z e d form, and the values of the parameters A and B can therefore be determined. A simple ana lys i s of equation (68) reveals that when 0<B<1 there i s no s i g m o i d a i b e h a v i o r i n the X^j-t c u r v e , the g a s i f i c a t i o n r a t e decreasing with extent of r e a c t i o n . I f B=l, equation (68) becomes Xc = 1 - exp ( -At) (72) Thi s i s the conversion-t ime expression of the continuous model ( C M ) , that w i l l be discussed in sec t ion 2 .5 .8 , with A being the rate constant of the f i r s t order r e a c t i o n . In the case B>1, a s igmoidai r e l a t i o n is pred ic ted and the values of t and XQ corresponding to the point of maximum g a s i f i c a t i o n rate on the X(]-t curve are obtained by equations: t M ~ [ ( B - 1 ) / A B ] 1 / B (73) and X C R - 1 - exp [ - ( B - l ) / B ] (74) Equation (68) allows the rate of g a s i f i c a t i o n to a t t a i n i t s maximum at any value of X ^ R between 0 and 1. This d i f f e r s from p r e d i c t i o n of the P R M and the R C M , which Is that : 0 < X C R < 0 . 3 9 3 . The rate of g a s i f i c a t i o n 49 dXfj/dt can be determined when the two constants A, B are obtained from equation (69B). d X c / d t = ( l - X c ) * A B t ( B 1 J (75) As an i n d i c a t i o n of g a s i f i c a t i o n r e a c t i v i t y , the average rate constant k i s def ined by equation (76): 1 0.99 k = j k ( X c ) d X c -I k ( X c ) d X c (76) •^ 0 *Voi 2.5 DESULPHURIZATION The removal of sulphur compounds from a gaseous stream can be c a r r i e d out i n a number of ways, which are discussed i n Appendix C. However, the technique of using calcium-based sorbents such as limestone and dolomite to capture sulphur i n - s i t u at high temperature has been emphasized and proven success fu l i n recent years . 2.5.1 SORBENTS The ready a v a i l a b i l i t y of l imestone and dolomite at low cost favours t h e i r use as sorbents i n the process for sulphur removal from the gas [81]. The common sorbent for capturing sulphur i s CaO which i s introduced into the combustors and g a s i f i e r s as l imestone, CaC03 and dolomite CaMg(C03)2-Limestone, which i s a sedimentary rock i s formed mainly through the depos i t i on of calcareous mater ia l [79], i s composed mainly of calc ium 50 carbonate, wi th magnesium carbonate as a common c o n s t i t u e n t . Therefore the term limestone i s frequent ly used to descr ibe minerals conta in ing not only ca lc ium carbonate but a l so u s u a l l y includes stones with vary ing contents of magnesium carbonate. Magnesium subs t i tu tes f o r ca lc ium as a r e s u l t of transformat ion of limestones by magnesium-rich s o l u t i o n s . The amount of magnesium i n limestone can vary from trace amounts up to the composition of dolomite . I d e a l l y i n dolomite the molar r a t i o of ca lc ium to magnesium i s u n i t y . In n a t u r a l l y occurr ing dolomites t h i s r a t i o v a r i e s between 0.8 and 1.2 [84,101,136,30]. S i l i c a i s probably the most undes irable impurity i n limestone and dolomite as i t appears to fuse and g lass over the c a l c i n e , CaO, and thus reduce the e f f ec t iveness of the p a r t i c l e i n r e a c t i n g wi th sulphur. The quartz occurs as rough, c l e a r g r a i n s , mainly 10 to 20 pm i n diameter, that are widely scat tered throughout the p a r t i c l e , and i s completely i n e r t i n the d e s u l p h u r i z a t i o n r e a c t i o n . Other major impur i t i e s commonly found i n l i m e s t o n e are a lumina and i r o n . Harvey [84] observed m i c r o s c o p i c a l l y that p y r i t e (FeS2) and hematite (Fe203) occur as t i n y granules , about 1 to 10 urn i n d i a m e t e r , e v e n l y d i s p e r s e d b e t w e e n c r y s t a l l i t e s o f d o l o m i t e . A d d i t i o n a l l y there w i l l be a host of t race elements. The compositions of se lec ted Canadian limestone and dolomite samples used i n t h i s study are shown i n Table 2 .6 . More g e n e r a l l y , the c h e m i c a l c o m p o s i t i o n i s o n l y of minor importance i n sulphur s o r p t i o n , whereas a number of experiments have demonstrated that wide v a r i a t i o n i n r e a c t i v i t y may be re la t ed to the d i f f erences i n the phys i ca l proper t i e s of the carbonate rocks [ 4 6 ] . Most l imestone and dolomite p a r t i c l e s are polygranular i n that about 5 51 Table 2.6 CHEMICAL COMPOSITION OF CARBONATE ROCKS EMPLOYED IN THIS STUDY, MASS %. Samples CaC03 MgC03 Fe203 A I 2 O 3 Si02 Na20 K 2 0 S A l b e r t a Syncrude 91.4 2.08 1.10 1.24 4.11 0.03 0.03 0.01 B r i t i s h Columbia Texada 95.7 1.50 0.27 0.65 1.80 0.03 0.04 0.01 Cache Creek 95.60 1.62 0.22 0.50 1.95 0.03 0.04 0.01 Washington State K e t t l e F a l l s 56.28 42.18 0.11 0.17 1.22 0.01 0.02 0.01 52 to 20 i n d i v i d u a l gra ins or c r y s t a l l i t e s are f i r m l y in ter locked to form a s i n g l e p a r t i c l e . The s i ze of each g r a i n making up the p a r t i c l e v a r i e s from 1 um to the maximum s ize of the p a r t i c l e , approximately 63 um; however, the g r a i n s i ze d i s t r i b u t i o n ind ica tes that the average diameter i s about 20 um [28,84] . Pores i n d o l o m i t e p a r t i c l e s are mainly i n the range of 20 um, compared to 0.5-2 um i n limestone [28]. These pores occur along the i n t e r l o c k i n g boundaries of the g r a i n s . The p o r o s i t y of the carbonate rocks genera l ly v a r i e s between 0.3 and 12% [28,46] . I n i t i a l pores in the na tura l carbonate rocks p e r s i s t through the c a l c i n a t i o n stage; and the carbonate rocks already porous p r i o r to t h e i r c a l c i n a t i o n a r e , g e n e r a l l y , be t t er su i ted for the sulphur removal process than common, dense limestone eg. marble [83]. The pores al low the reactant gases such as S O 2 and H 2 S to d i f fuse in to r e a c t i o n surface , thus promoting chemical r e a c t i o n [84,110,83,79]. The ca lc ium carbonate i n limestone occurs p r i n c i p a l l y i n the form of c a l c i t e and l ess frequent ly as a r a g o n i t e . A r a g o n i t e p o s s e s s e s an orthorhombic l a t t i c e whereas c a l c i t e and magnetite have a rhombohedral c r y s t a l s t ruc ture [28,84,202,46]. The rhombohedral form of many of the o r i g i n a l limestone c r y s t a l l i t e s i s re ta ined during the r e a c t i o n [46]. 2 .5.2 CALCINATION Calcium carbonate, CaC03 and dolomite , CaMg(C03)2 and i r o n oxides (FeO, Fe203) are c a p a b l e of removing hydrogen sulphide from a gas stream. However, under g a s i f i c a t i o n temperatures which are u s u a l l y above 8 0 0 ° C , the use of i ron oxides i s r e s t r i c t e d due to the formation 53 of l i q u i d oxysulphide and the r e a c t i o n of the oxides at the operat ing temperature [183]. Limestone and dolomite react with hydrogen sulphide only when they have been c a l c i n e d . Thi s i s a k i n e t i c ra ther than a thermodynamic r e s t r i c t i o n [110]. Abel [1] reported that limestone showed l i t t l e a b s o r p t i o n c a p a c i t y f o r H2S u n t i l heated above i t s c a l c i n a t i o n temperature, while Ruth [151] observed that r e a c t i o n of p u l v e r i z e d limestone with H2S i s very slow at 8 0 0 ° C . Calcium carbonate or c a l c i t e decomposes upon heating to form CaO + CO2. The thermal decomposition occurs f i r s t at the outer surface of the p a r t i c l e and then advances inwards [70]. The bas ic c a l c i n a t i o n reac t ion i s endothermic with a heat of r e a c t i o n of about 169.7 kJ/mol [191]. CaC0 3( s) _ C a O ( s ) + C0 2( g) A H 1173K = 169.7 kJ/mol C a l c i n a t i o n occurs only when the carbonate d i s s o c i a t i o n pressure at the given temperature exceeds the p a r t i a l pressure of carbon dioxide i n the surrounding atmosphere, and the rate of decomposition of e i t h e r calc ium or magnesium carbonate i s p r o p o r t i o n a l to the e q u i l i b r i u m CO2 pressure at the temperature [30]. The values of the e q u i l i b r i u m composition pressure of calc ium carbonate can be estimated by using the equation proposed by H i l l s [70]: log P, CO, 2 -8547 T + 7.260 (78) where T i s the c a l c i n a t i o n temperature i n K and P i s e q u i l i b r i u m p a r t i a l pressure of CO2 in atmospheres. 54 Abel [1] observed that the presence of 8.4% C O 2 i n the c a l c i n a t i o n atmosphere caused an i n c r e a s e of about 1 1 0 ° C i n the t emperature n e c e s s a r y to a c h i e v e r e a c t i o n of a l imestone with H 2 S gas. His explanat ion for t h i s decrease i n r e a c t i v i t y was that the reac t ion does not occur u n t i l CaO i s present . The p r e s e n c e o f C O 2 i n the s u r r o u n d i n g atmosphere h i n d e r s c a l c i n a t i o n preventing CaO formation u n t i l the d i s s o c i a t i o n pressure of the c a r b o n a t e r o c k exceeds the p a r t i a l p r e s s u r e of C O 2 i n the surrounding atmosphere. The pore s t ruc tures of the r e s u l t i n g ca l c ines are a l so af fected by a C O 2 atmosphere. The s p e c i f i c area and the c r y s t a l l i t e s i z e of the ca lc ium oxide are r e s p e c t i v e l y cons iderably reduced and increased i n the presence of carbon diox ide i n the surrounding atmosphere [84]. The reduct ion i n surface area i s due more to the enhancement of s i n t e r i n g by C O 2 t h a n t o a d e c r e a s e i n t h e r a t e o f n u c l e a t i o n i n t h e r e c r y s t a l l i z a t i o n of the c a l c i u m o x i d e . The e q u i l i b r i u m p a r t i a l pressure of C O 2 over c a l c i t e was measured by H i l l s [89], and that over magnetite was c a l c u l a t e d by Coutur ier [46] as shown i n Figure 2 .2 . Abel [1] extended t h i s type of p l o t to include the e q u i l i b r i u m p a r t i a l pressure of C O 2 over c a l c i t e f o r cases of t o t a l pressure higher than 101 kPa. The o p e r a t i n g c o n d i t i o n s imposed f o r c a l c i n a t i o n of dolomite, C a M g ( C 0 3 ) 2 are c r i t i c a l l y important i n d i c t a t i n g the type of f i n a l products . When dolomite i s decomposed at 750-800 oC in a i r , with an a c t i v a t i o n of 188.1 k J / m o l , the product i s the f u l l y - c a l c i n e d dolomite, according to: 1 0 0 0 i i i r 1 I r I 0 0 J I o M g C 0 3 a M g O ( s ) - f C 0 2 ( g ) / 4—1 / A / A / / A A I / • / / • / / / / / • / / C a C 0 3 ( s ) r C a O ( s ) + C 0 2 ( g p / I I I I I I 2 0 0 4 0 0 6 0 0 8 0 0 Temperature (°C) 1 0 0 0 F i g . 2.2 D i s s o c i a t i o n pressure of c a l c i t e and magnesite. [46] - O - [89] 56 CaC0 3 'MgC0 3 -» [CaOMgO] + 2C0 2 (79) whereas i f the thermal decomposition i s conducted under a s u f f i c i e n t p a r t i a l pressure of C 0 2 to prevent d i s s o c i a t i o n of CaC03, the product i s h a l f - c a l c i n e d dolomite, [CaC03MgO] [1,151,202]. In one atmosphere of C 0 2 , the decomposition of dolomite ceases at the h a l f - b u r n t stage for a l l temperatures l e ss than 8 5 0 ° C , and the product so obtained contains l i t t l e , 0.1-0.2%, calc ium oxide [202] CaC0 3 «MgC0 3 •• [CaC0 3 «MgO] + C 0 2 (80) Each of the s o l i d products der ived from dolomite are denoted by chemical formula wi th brackets as a reminder that the chemical symbols do not stand f o r the chemical spec ies , but an in terming l ing of m i c r o s c o p i c c r y s t a l s of magnesium oxide and the ca lc ium compounds. Harvey [84] having done extensive work in c h a r a c t e r i z i n g dolomite and i t s h a l f - b u r n t stage postulated that i n each c r y s t a l l i n e g r a i n of dolomite the p o s i t i v e ions of magnesium and calc ium a l t ernate between layers of negative ions of carbonate i n a sequence of C 0 3 - M g - C 0 3 - C a - C 0 3 -M g - C 0 3 - C a - and so on, along the carbon C-ax i s of the c r y s t a l h a b i t s . H a l f - c a l c i n a t i o n breaks the bonds between Mg and C 0 3 leaves MgO and C a C 0 3 c r y s t a l l i t e s homogeneously mixed, almost i n an atomic s ca l e . D u r i n g the escape of C 0 2 bound with M g 2 + i n the o r i g i n a l dolomite c r y s t a l , new pores of t y p i c a l l y less than 1 um in diameter but several times that i n length were formed. They are interconnected throughout the p a r t i c l e with narrow j u n c t i o n s . Although MgO c r y s t a l l i t e s do not p a r t i c i p a t e i n the sulphur removal r eac t ions , t h e i r presence contr ibutes 57 to the open of g r a i n s tructure and to the maintenance o f the h i g h p o r o s i t y of o r i g i n a l pores al lowing h igher rate of r e a c t i o n , f u l l e r sorbent u t i l i z a t i o n and advantageous sorbent regenerat ion . The dimensions of the magnesia c r y s t a l s formed i n a c a l c i n a t i o n are inf luenced by the c a l c i n a t i o n c o n d i t i o n s . X - r a y d i f f r a c t i o n patterns of h a l f - c a l c i n e d dolomite prepared at 7 5 0 ° C and 7 8 7 ° C and 1 atmosphere p r e s s u r e o f C 0 2 show t h a t a l l d o l o m i t e m i n e r a l c r y s t a l l i t e s were transformed i n t o c a l c i t e (CaCC^) and p e r i c l a s e (MgO) w i t h s i z e of 1 um and 0.1 um r e s p e c t i v e l y [84]. In another study, Young [202] reported that i n an atmosphere of C 0 2 magnesia c r y s t a l s are smal ler (-150 A) than the c a l c i t e c r y s t a l s (~400 A) up to 7 0 0 ° C . Both products have p a r t i c l e s i z e s about 400 A near 9 0 0 ° C . I f , on the other hand, the dolomite i s decomposed i n vacuo, at temperatures around 1 0 0 0 ° C the c r y s t a l s s i z e i s l a r g e r than 1400 A. The MgO c r y s t a l l i t e s are so w e l l d i spersed i n the c a l c i n e d dolomite that t h e i r presence can be found i n every sample of the c a l c i n e s along the c a l c i t e [84] . Besides the carbon dioxide composition i n the surroundings , other operat ing cond i t ions such as c a l c i n a t i o n temperature and t ime, and the heat ing ra te have pronounced e f f ec t s on the r e a c t i v i t y as w e l l as on the p h y s i c a l s t r u c t u r e of the produced lime [27,83,129,132,117,79] . The c r y s t a l s t r u c t u r e of the so f t -burnt lime produced i n the temperature range between 8 0 0 - 9 5 0 ° C i s very porous [1 ,117,79,46] , possess ing a h igh s p e c i f i c area [1 ,27 ,46] , small g r a i n s i ze [70] of CaO, and smal l pore s i z e d i s t r i b u t i o n . With an equal re t en t ion t ime, h igher temperatures of burning produces a hard-burnt lime with cons iderably lower p o r o s i t y and surface area , and thus lower r e a c t i v i t y [183]. This e f f e c t i s more 58 s i g n i f i c a n t at shorter re tent ion time [129]. F i n a l l y , lime produced at temperature above 1 5 5 0 ° C i s v i r t u a l l y unreact ive , and i s c a l l e d dead-burnt lime [46]. It i s known that the pore volume and the s p e c i f i c area of ca l c ines decreases wi th increas ing c a l c i n a t i o n time [1 ,27,142] . Borgwardt [27] observed t h a t the surface area of micros ize lime was s i g n i f i c a n t l y reduced a f t e r 20 min. of s i n t e r i n g immediately fo l lowing the i n i t i a l c a l c i n a t i o n stage at 9 5 0 ° C . A l s o , Abel [1] noted that when l ime, having been i n i t i a l l y porous and high i n surface area , i s fur ther heated, the c r y s t a l s t ruc ture becomes more dense and thus reduced i n both poros i ty and surface area. Genera l ly , when the c a l c i n a t i o n t e m p e r a t u r e i s brought to 1 2 8 2 - 1 4 7 6 ° C , the lime s i n t e r s . Th i s leads to the p a r t i c l e s s h r i n k i n g which increases the bulk dens i ty , and lowers the p o r o s i t y , the pore surface area and the capac i ty to capture sulphur [30]. Boynton [30], studying the c a l c i n a t i o n of Ontario limestones and dolomites under n i t rogen atmospheres up to a temperature of 1320'C and a re tent ion time of 474 m i n . , reported a p a r t i c l e shrinkage up to 44.2% and an increase of bulk dens i ty to 2.76 g/cra^, from the normal value of 1.47 g/cra^. Studies [129,132] show that poros i ty of the c a l c i n e s i s determined p r i n c i p a l l y by the rate of heating before and during c a l c i n a t i o n , while the heating condi t ions subsequent to c a l c i n a t i o n e . g . the maximum temperature or the t o t a l re tent ion time have l e s s e r e f f e c t . Ca lc ines thermal ly decomposed in i n c r e a s i n g l y higher heating rates p r i o r to and d u r i n g c a l c i n a t i o n seem to be l e ss porous and thus, l ess r e a c t i v e . Murray et a l . [129] suggested that an optimum re tent ion time which gives highest poros i ty and r e a c t i v i t y can be found for each c a l c i n a t i o n . 59 The e f fec t s of a l l the mentioned above condi t ions of c a l c i n a t i o n on surface area, poros i ty and on pore s i ze d i s t r i b u t i o n of the ca l c ine were studied extens ive ly by many i n v e s t i g a t o r s , and the publ ished r e s u l t s are enormous. Two sets of r e s u l t s , one from a commercial c a l c i n e r and the other from labora tory - sca l e c a l c i n e r , were chosen and reported here. Table 2.7 shows the e f fec t s of c a l c i n a t i o n condi t ions on the r e s u l t i n g lime whi l e , Table 2.8 reveals the inf luences of types of c a r b o n a t e rocks . S i m i l a r r e s u l t s were a lso reported by Dogu [48] for c a l c i n a t i o n temperatures between 750 and 9 5 0 ° C . The l e v e l of c a l c i n a t i o n can be q u a n t i t a t i v e l y determined by de f in ing the percent c a l c i n a t i o n as fol lows [191]: i • m r . r 1 L O I (Product) , , . percent c a l c i n a t i o n = 100 [1 - rr——r=—TT ] (81) LOI ( r e e d ) Table 2.7 EFFECTS OF CALCINATION CONDITIONS ON RESULTING LIMES [129] Calcination Maximum Retention Time to Temperature Time mln. Reach of Bed, C 900 C, mln. Ignition CO2 Surface Porosity Bulk Loss percent Area, Sq percent Specific percent meters Gravity per g. 19-1 1010 474 182 1.7 0.2 11.3 46.6 1.78 19-8 930 67 29 2.9 2.1 17.1 48.4 1.72 21-1 1100 474 163 1.9 0.1 8.15 46.0 1.80 21-11 963 46 19 1.1 1.1 8.80 38.4 2.05 23-1 1214 474 146 1.1 0.1 8.4 45.7 1.81 23-11 1196 47 15 0.7 0.6 4.54 25.8 2.47 25-1 1294 474 136 0.3 0.1 4.28 43.5 1.88 25-11 1283 48 14 0.2 0.1 2.35 19.8 2.67 60 Table 2 . 8 EFFECTS OF TYPES OF CARBONATE ROCKS ON RESULTING CALCINES AT 8 4 3 ° C FOR 1 hr [130]. Sample Surface Pore Most- Bulk Porosity Area, volume Probable density % m2/gram cm^/gram pore diameter grams/cm^ Limestone 1211 24.6 0.32 1214 14.7 0.38 1217 20.6 0.34 1213 13.6 0.34 Dolomites 1216 30.1 0.40 1218 34.2 0.39 1212 26.4 0.42 1215 20.2 0.41 1219 26.1 0.33 0.090 1.61 51.6 0.17 1.47 55.8 0.18 1.56 53.1 0.18 1.56 53.0 0.099 1.45 58.0 0.096 1.47 57.3 0.14 1.41 59.3 0.087 1.42 58.2 0.C85 1.61 53.1 where LOI i s the weight loss during c a l c i n a t i o n [ 1 , 1 9 1 , 1 2 9 ] . The t h e o r e t i c a l LOI i n the c a l c i n a t i o n of c a l c i n a t i o n of ca lc ium carbonate, CaC0 3 i s 44% M 100 [1 - C a ° ] = 44% CaC0 3 The p r o p e r t i e s of the l ime p r o d u c t are a lso affected by the handl ing and exposure condi t ions to which the c a l c i n e i s s u b j e c t e d during coo l ing and storage [129]. Test a lso showed that the surface area to be s table over a per iod of about 6 weeks s t o r a g e at room temperature i n an a i r t i g h t g lass conta iner [27]. At room temperature, 61 l ime, reacts q u i c k l y with water i n the ambient atmosphere to form ca lc ium hydroxide according to the hydrat ion reac t ion [157]: CaO( s ) + H 2 0 ( 1 | V ) Ca(OH) 2 AH = -66.5 kJ/mol (82) E q u i l i b r i u m p a r t i a l pressure of water over MgO and CaO as a funct ion of temperature i s g iven by Coutur ier [46]. The g a s i f i e r employed i n t h i s study i s operated at temperatures between 800 and 9 3 0 ° C , and with a p a r t i a l pressure of steam i n the range of 15 to 60 kPa. In t h i s range of operat ing condi t ions there w i l l be no formation of Ca(0H) 2 from CaO and H 2 0 . 2 .5.3 ESTIMATION OF e, p and S p In recent years , considerable a t t ent ion has been given to the use of c a l c i n e s as sorbents because of t h e i r high r e a c t i v i t y toward sulphur removal react ions [148,62,164,48], The r e a c t i v i t y and capac i ty of l ime, which depend upon the phys i ca l and chemical proper t i e s of the limestone used, and on the c a l c i n a t i o n cond i t i ons , are c r i t i c a l l y important to the d e s u l p h u r i z a t i o n processes [83,79,48] . However, the chemical propert ie s of the limestone were found experimental ly to have only minor inf luence on the sorpt ion of sulphur [83,129,148], while besides the geo log ica l o r i g i n and the nature of the limestone i t s e l f , p h y s i c a l proper t i e s such as p o r o s i t y , pore s i ze d i s t r i b u t i o n , pore volume, surface area and g r a i n s i z e of l i m e s t o n e s and t h e i r ca l c ines s trong ly a f f ec t the rate of r e a c t i o n and the f i n a l sorbent conversion [18,28,81,126,83,56,117,36] . Assuming there i s no shrinkage in p a r t i c l e s during c a l c i n a t i o n and the carbonate rocks are non porous, using the He ( true) dens i t i e s of 62 CaC0 3 (p=2.72 g / cm 3 ) , CaO (p=3.34 g / cm 3 , molar volume 16.89 cm 3/mole) and CaS (p=2.61 g/cm 3 ) one can simply compute the t h e o r e t i c a l i n i t i a l p o r o s i t y of a l ime, which i s ~54% us ing : ( M V )C a C 0 , - ( M V ) C a O E c a f c — ( 8 3 > v ' C a C 0 3 and i t s completely sulphided s ta te , which i s ~24% us ing: ( M V ) C a S From these va lues , 100% conversion of CaO to CaS i s a t t a i n a b l e , which i s an important feature of the CaO + H 2 S r e a c t i o n ; whereas the complete conversion of CaO to CaS04 seems to be a remote p o s s i b i l i t y due to the s i g n i f i c a n t l y l a r g e r molar volume (52 cm 3 /mole) of the r e a c t i o n product ca lc ium sulphate . Thus, the c o r r e l a t i o n es tab l i shed by Falkenberry [56] reveals that the s p e c i f i c volume plays a very important r o l e i n sorbent c a p a c i t y . The i n i t i a l pore s i ze d i s t r i b u t i o n of the s o l i d CaO i s found to inf luence both the r e a c t i v i t y and the maximum CaO convers ion; smal ler pores i n p a r t i c l e s acce lerate the r e a c t i o n rate i n i t i a l l y . The r a p i d formation of product s o l i d may cause small pore c losure or plugging near the pore mouths r e s u l t i n g i n e v e n t u a l r e d u c t i o n i n reac t ion rate [161,162,41,117,48]. Therefore the i n t r a p a r t i c l e vo id space and the pore s i ze d i s t r i b u t i o n can be a l i m i t i n g f a c t o r [27,82]. Pores in lime p a r t i c l e s are mostly macrosize i . e . >500 A [35]. These e f fec t s w i l l be 63 mentioned again i n Sect ion 2 .5 .6 , when d i scuss ing the g r a i n model for the s u l p h i d a t i o n r e a c t i o n . Since the rate of the g a s - s o l i d reac t ion i s determined p r i m a r i l y by the t o t a l a v a i l a b l e reac t ion surface , the c a l c i n e s pos se s s ing l a r g e i n t e r n a l surface area are most r e a c t i v e . The ro l e which the s p e c i f i c surface area plays i n the CaO + H 2 S and CaO + S0 2 r e a c t i o n systems were q u a n t i t a t i v e l y e l u c i d a t e d ; the r e a c t i v i t y increases wi th the BET surface area to the power from 1 [170] to more than 2 [27,26] . When the pores are assumed to be n o n - i n t e r s e c t i n g c y l i n d r i c a l c a p i l l a r i e s , then the surface area of pores, Sp, i s g iven by [35]: 4 Vp V ( 8 5 ) Where Vp and D are the pore volume (cm^/g) and the most probable pore diameter r e s p e c t i v e l y . Both v a l u e s can be o b t a i n e d from mercury p o r o s i m e t r y . S i m i l a r l y , when the l ime i s assumed to cons i s t of unconsol idated equal s p h e r i c a l gra ins of radius r g , i t s s p e c i f i c surface area can be estimated from [7]: S = (86) P VcaO where PCaO ^ s t n e helium or true dens i ty of the c a l c i n e . Equations (85) and (86) p r e d i c t values of surface areas of lime c o n s i s t e n t l y d i f f e r e n t from that measured by the BET procedure from s o r p t i o n isotherms of N 2 at 77K [35,70,129,130] due to the s i m p l i f i e d assumptions i n g r a i n or pore s i z e . They should therefore be used as rough estimates only . 64 Another important parameter with respect to the r e a c t i v i t y of c a l c i n e s i s the p o r o s i t y . P o r o s i t y of a c a l c i n e i s g iven by [129]: P CaO P CaO / 0 , . cr Z (87) C p CaO where PCaO a n < * P'CaO a r e *-he t r u e (He) dens i ty and the bulk (Mercury displacement) dens i ty of the c a l c i n e . ' The bulk dens i ty w i l l be: M„ n  CaO , P CaO " " M — P C a C 0 3 ( 8 8 ) CaC0 3 •* and the i n t e r n a l v o i d volume V j can be estimated from: e C V l - — I (89) HCaO T a b l e 2.9 summarizes some t y p i c a l dens i ty values of l ime, carbonate rocks and t h e i r sulphided s ta te . 2 .5 .4 THE SULPHIDATION REACTIONS Processes for h igh temperature removal of H 2 S , from f u e l gas, which i s a product of the g a s i f i c a t i o n of sulphurous s o l i d s or from other r e d u c i n g gases u s i n g calcium-based sorbents have been described by Squires et a l . [168,151], Keairns et a l . [110] and Borgwardt [ 2 7 ] . These processes are not only capable of removing the small amounts of H 2 S present i n the f u e l gas, but the schemes a l so al low for regeneration 65 Table 2.9 TYPICAL PROPERTIES OF LIMESTONES AND ITS RELATED SPECIES. Species Dens i ty , g/cm^ Molecular Molar Volume Apparent (Hg) approx. True (He) Weight g/gmole cm^/gmoli CaO 1.51 3.32 56.08 16.89 CaS 2.61 72.08 27.62 MgO 40.39 11.05 CaO•MgO 1.47 96.39 36.2 CaS•MgO 112.39 27.4 CaC03«>MgO 140.39 16.2 CaC0 3 .MgC0 3 2.65 184.39 CaC0 3 2.69 2.71 100.08 36.90 CaS04 2.62 136.08 51.94 66 of the s o l i d to release H2S and concentrate i t s u f f i c i e n t l y to be used i n a c o n v e n t i o n a l C l a u s p r o c e s s , wh ich converts H2S to elemental su lphur , or to be used i n product ion of H 2 S04. Squires et a l . [35,130] c a l l f o r the use of these p r o c e s s e s as an i n t e g r a l p a r t of the g a s i f i c a t i o n system in order to u t i l i z e the t o t a l thermal e f f i c i e n c y of the opera t ion . Ca lc ined dolomite , h a l f - c a l c i n e d dolomite and c a l c i n e d limestone are candidate acceptors for the adsorpt ion of H2S according to react ions [131]: React ions: A H 2 9 8 K » ^ J / m ° l CaO + H 2 S -> CaS + H 20 -65.2 (90) [CaO^MgO] + H 2 S -» [CaS^MgO] + H 20 -66.88 (91) [CaC0 3«MgO] + H 2 S -• [CaS^MgO] + H 20 + C0 2 91.96 (92) By vary ing the amount of H2S, H 20 (steam) and CO2, the r e a c t i o n (92) can be made to go e i t h e r d i r e c t i o n . When the reac t ion goes from l e f t to r i g h t i t i s an absorpt ion stage; from r i g h t to l e f t i t i s a regenerat ion stage. I t s h o u l d be noted t h a t the e q u i l i b r i u m does not favor the formation of magnesium sulphide from magnesium o x i d e and hydrogen sulphide at the temperature of in t ere s t i n t h i s study [152,168,183]. Although magnesia does not p a r t i c i p a t e i n the r e a c t i o n , i t s w e l l -d i s t r i b u t e d presence [84] in c a l c i n e d dolomite i s use fu l to maintain the s o l i d sorbent porous during s u l p h i d a t i o n and regenerat ion processes . To confirm that MgO in c a l c i n e d dolomite does not contr ibute to the H2S 67 s o r p t i o n , a tes t was made [1] us ing pure MgO in powdered form as the sorbent . I t absorbed v i r t u a l l y no H2S. Since MgO i s e s s e n t i a l l y chemical ly i n e r t i n the su lph ida t ion and regenerat ion reac t ions , and also has n e g l i g i b l e s o l u b i l i t y i n CaO and in CaS, the MgO w i l l not be present i n the e q u i l i b r i u m c o n s i d e r a t i o n s below. The s tate of e q u i l i b r i u m for the removal of hydrogen sulphide , as i n (90) and (91) i s favored over a wide range of t empera ture of p r a c t i c a l i n t e r e s t , independent of t o t a l pressure , and i s described by Squires [168,151] for u n i t a c t i v i t y of the s o l i d s as: [ H 2 ° ] 3519 2 lo*io r^sT= T - ° - 2 6 8 < 9 3 > where the brackets s i g n i f y mole f r a c t i o n and T i s i n K. Turkdogan [183] independently determined the e q u i l i b r i u m f o r the same set of reac t ion but expressed i t i n terms of the e q u i l i b r i u m constant . 3420 l o g 1 0 Kj = ^ - 0.190 (94) V where: K = — (95) H 2 S Here i s p a r t i a l pressure of the species i n d i c a t e d . E q u i l i b r i u m c o n s t a n t s f o r h i g h t empera ture s u l p h u r r e m o v a l expressed by equation (94) are given g r a p h i c a l l y i n Figure 2 .3 . The e q u i l i b r i u m H 2 S pressure over CaO, as a funct ion of temperature for 68 F i g . 2.3 Equilibrium constants for high temperature sulphur removal. 69 var ious steam p a r t i a l pressure ranges used in t h i s present study, i s determined using equation (93) and shown i n Figure 2.4. The high value of the e q u i l i b r i u m c o n s t a n t f o r the c a l c i u m o x i d e r e a c t i o n a t temperatures app l i cab le to g a s i f i c a t i o n j u s t i f i e s i n t e r e s t in i t f or removal of H 2 S . In a d d i t i o n , the rate of r eac t ion between CaO and H 2 S i s favorable for u t i l i z a t i o n of ca lc ium oxide i n the desu lphur iza t ion process . E q u i l i b r i u m for r e a c t i o n (92) i s favorable for removal of H 2 S only at high temperature, above 800"C. Thi s e q u i l i b r i u m can be obtained by coupl ing the e q u i l i b r i u m decomposition pressure of CaC03, proposed by H i l l s et a l . [89,88] wi th equation (93) [ H 2° ] [ CV P _ , „ 5,280.5  1 O B 1 0 [H^S] 7 ' 2 5 3 - J - T - ( 9 6 ) where t o t a l pressure P i s i n atm. The regenerat ion of c a l c i n e sorbent makes both environmental and economic sense. F i r s t , d i r e c t d i s p o s a l or u t i l i z a t i o n of the d r y , p a r t i a l l y converted sulphur acceptor i s not considered to be an opt ion that i s genera l ly a v a i l a b l e [110]. Second, to lower the c o s t of d e s u l p h u r i z a t i o n of gases, the c a l c i n e s or any other H 2 S sorbent must be regenerated and recyc led to recover sulphur from the l eav ing gas [183]. Both Squires [151,168] and Turkdogan [183] have discussed regenerat ion s teps . In steam g a s i f i c a t i o n of sulphurous s o l i d s with sulphur being captured i n - s i t u by c a l c i n e s , the water-gas s h i f t and the s u l p h i d a t i o n r e a c t i o n occur s imultaneously . The water s h i f t r eac t ion ( reac t ion 5) i s F i g . 2.4 E q u i l i b r i u m R^S pressure over CaO i n the range of steam p a r t i a l pressure employed i n t h i s study. 7 1 favored by e q u i l i b r i u m to produce C O 2 + H 2 i n the same temperature range as that of the su lph ida t ion react ions (react ions 90, 91, 92) shown in F igure 2 .3 . For the combined system with hydrogen sulphide and carbon monoxide gases encountering ac t ive CaO, i f e q u i l i b r i u m concentrat ions are a c h i e v e d , S c h r e i b e r [152] showed t h a t H 2 S concentrat ions are t y p i c a l l y l e ss with the water gas s h i f t r e a c t i o n than without i t by a f a c t o r of 3. T h e o r e t i c a l l y , the water vapour formed by the s u l p h i d a t i o n react ions i . e . r e a c t i o n : (90), (91), (92) and the carbon monoxide present i n a t y p i c a l g a s i f i e r product gas combine to produce H 2 and C O 2 . As equation (95) shows, removing the water vapour by the w a t e r - g a s r e a c t i o n suppl ies a s tress to enhance the removal of H 2 S by s h i f t i n g r e a c t i o n (90), (91) and (92) to the r i g h t [152,137]. An a p p r o p r i a t e c a t a l y s t i s required to provide adequate ac t ive s i t e s for the water-gas s h i f t r e a c t i o n . MgO i s recognized as a good c a t a l y s t f o r the g a s - s h i f t r e a c t i o n [185,137,153]. F u l l y c a l c i n e d dolomite , can supply CaO for the desu lphur iza t ion r e a c t i o n and MgO as a c a t a l y s t f o r the water gas r e a c t i o n . The r e l a t i v e r e a c t i v i t i e s of CaO, [CaO»MgO] and [CaC03«MgO] are considered i n the next s e c t i o n . 2 .5 .5 RELATIVE REACTIVITIES OF SORBENTS There are c o n f l i c t i n g reports on the r e l a t i v e r e a c t i v i t i e s of l ime, h a l f - c a l c i n e d dolomite and f u l l y - c a l c i n e d dolomite . Studying the e f fec t of types of c a l c i n e s on the s u l p h i d a t i o n process , Abel [1] reported that f u l l y - c a l c i n e d limestone exh ib i t s higher 72 a b s o r p t i o n c a p a c i t y than a h a l f - c a l c i n e d d o l o m i t e . S i m i l a r l y , Falkenberry [56] and P e t r i e [141] found that , at an equivalent Ca/S r a t i o , l imestones are seen to e f fec t a greater reduct ion in sulphur emissions than dolomites . On the other hand, Borgwardt [24], Narayen [131] and Turkdogan [183] observed that a l though c h e m i c a l l y i n e r t , magnesia maintains open poros i ty i n the sorbents dur ing s u l p h i d a t i o n . For t h i s reason the rate with f u l l y - c a l c i n e d dolomite i s f a s t e r than that with burnt l ime. Between the above two groups, Harvey [83] found t h a t a r o c k c o n t a i n i n g about 19% by w e i g h t MgO d i s p l a y e d an a p p r o x i m a t e l y median r e a c t i v i t y w i t h i n a group of 10 commercial l imestones. The r e l a t i v e r e a c t i v i t i e s of f u l l y and h a l f - c a l c i n e d dolomite were a l so examined by a number of authors whose r e s u l t s are i n disagreement. Harvey [84] and Ruth et a l . [151] concluded that the h a l f - c a l c i n e d dolomite was super ior to o ther m a t e r i a l s i n c a p a c i t y and r a t e of absorpt ion of sulphur from d i l u t e H 2 S gases at the des i red temperature and at 101 kPa according to r e a c t i o n (92). The r e s u l t s from a study of Keairns et a l . [110] supported t h i s c la im; h a l f - c a l c i n e d dolomite was shown to improve sulphur removal at 8 1 6 ° C over that recorded at 7 6 0 ° C ; when temperature was increased to 8 1 7 ° C , however, there was a d r a s t i c loss of desu lphur iza t ion ac t ion as the dolomite decomposed to the f u l l y -c a l c i n e d s ta te . In c o n t r a s t to the above, e x p e r i m e n t a l r e s u l t s o b t a i n e d by Turkdogan et a l . [183] ind icated that for powdered samples, the rate of r e a c t i o n w i t h c a l c i n e d dolomite i s about twice that with p a r t i a l l y c a l c i n e d dolomite owing p a r t l y to the lack of p o r o s i t y of the l a t t e r . 73 The r e l a t i v e rates are a lso markedly affected by the d i f f erence in heat of r e a c t i o n s . The s u l p h i d a t i o n of f u l l y - c a l c i n e d dolomite i . e . r eac t ion (91) i s exothermic with AH=-66.8 k J / m o l w h i l e r e a c t i o n ( 9 2 ) , the s u l p h i d a t i o n of p a r t i a l l y c a l c i n e d dolomite, i s s t rong ly endothermic wi th AH=83.6 k J / m o l . Consequently, the rate of r eac t ion (92) w i l l be r e t a r d e d because of lower rate of gas d i f f u s i o n i n the less porous p a r t i c l e s and, more important ly , the lower heat t r a n s f e r to compensate for the heat of r e a c t i o n . Therefore , dolomite must be f u l l y c a l c i n e d p r i o r to i t s use in H 2 S removal. While e q u i l i b r i u m constants for both g a s i f i c a t i o n - d e s u l p h u r i z a t i o n p r o c e s s e s >1, t h e e f f i c i e n c y o f t h e c o m b i n e d g a s i f i c a t i o n -desu lphur iza t ion i s u l t i m a t e l y c o n t r o l l e d by the o v e r a l l rate of gas-s o l i d react ions [152]. Models for the g a s i f i c a t i o n process have been descr ibed i n Sections 2 .4 .4 to 2 . 4 .6 . In the next two sect ions the Gra in Model (GM) and the F i r s t - O r d e r K i n e t i c or Continuous Mode (CM), which w i l l be used to model the CaO + H 2 S r e a c t i o n , are descr ibed . 2 .5.6 GRAIN MODEL E a r l y m a t h e m a t i c a l models to s i m u l a t e n o n - c a t a l y t i c g a s - s o l i d react ions invo lv ing s o l i d products have included the sharp in ter face shr ink ing core models discussed i n d e t a i l by Szekely et a l . [175]. The r e a c t i o n of H 2 S i n porous lime represents a complex coupl ing of chemical r e a c t i o n w i t h the mass t r a n s p o r t t h r o u g h a t i m e - v a r y i n g porous s t r u c t u r e . The ro l e of the CaS deposits on the s u l p h i d a t i o n process have been the subject of several i n v e s t i g a t i o n s [27, 1 5 8 , 9 6 , 6 0 , 3 3 ] . 74 Working wi th preca lc ined limestones and dolomites P e l l et a l . [136] and Turkdogan [183] found that the r e a c t i o n was f i r s t - o r d e r with respect to H 2 S , and the rate decreased r a p i d l y as the s u l p h i d a t i o n continued. More advanced transport theor ies have been developed to describe d i f f u s i o n - l i m i t e d p r o c e s s i n s u l p h u r c l e a n u p . The G r a i n Model , presented by Szekely and Evans [140,176,166], and the pore model [176] were the f i r s t models to incorporate s t r u c t u r a l p r o p e r t i e s of the s o l i d as parameters of the models instead of lumping them in to an empir i ca l r e a c t i o n r a t e c o n s t a n t . W h i l e the G r a i n Model has advanced our understanding of g a s - s o l i d reac t ions , i t has neglected the changes in porous s t ruc ture wi th extent of r eac t ion [67,40]. Hartman and Coughlin [81] extended the GM to account for the changes i n p o r o s i t y d u r i n g r e a c t i o n and to be able to p r e d i c t the maximum a t ta inab le convers ion. The drawback of t h i s model i s that i t t rea t s the s o l i d p a r t i c l e as h a v i n g an average porous s tructure [40]. P i g f o r d and S l i g e r [142] appl ied the GM with the assumption that the reac t ion rate i s governed by e i t h e r d i f f u s i o n t h r o u g h the pores or d i f f u s i o n through the s o l i d product . Th i s model, however, i g n o r e s the importance of k i n e t i c l i m i t a t i o n s . Georgakis et a l . [67] developed an e m p i r i c a l model to account f o r the changes i n g r a i n s i zes by noting that the l o s s of p o r o s i t y must be due to increases i n gra in s i ze wi th r e a c t i o n t ime. The poor p r e d i c t i o n of an e v e r - i n c r e a s i n g pore s u r f a c e i n s p i t e of a decrease of pore volume prevents t h i s model from being widely used. Simons et a l . [161,162] a lso presented a model implying a concept of the plugging of the smal lest pores , which may be r a t e - c o n t r o l l i n g . Lee et a l . [117] and Chrostowski and Georgakis [41] concentrated on a s ing le 75 pore ins tead , and developed pore plugging models in which the pore mouth c losure due to the accumulation of reac t ion product taking place with extent of r eac t ion can be a n a l y t i c a l l y d e t e r m i n e d . These mode l s , n e v e r t h e l e s s , are s e m i - e m p i r i c a l and posses s s e v e r a l a d j u s t a b l e parameters. Ramachaudran and Smith [145] and Kocaefe et a l . [112] extended the GM to fur ther include the combined e f fec t s of s i n t e r i n g and dens i ty changes as reac t ion proceeds. The computational d i f f i c u l t i e s encountered i n these models p r o h i b i t t h e i r use in the modeling of H 2 S sorpt ion i n CaO. Deta i l ed d e s c r i p t i o n s of pore growth and combination have been r e a l i z e d by Christman and Edgar [40]. F i n a l l y , Bhat ia and Perlmutter [18,19,17,21] and Gavalas [66] have developed RPM and RCM, r e s p e c t i v e l y , which account f o r i n t e r s e c t i o n s among pores as wel l as changes i n pore s t r u c t u r e . Bhat ia [20] a l so suggested that these models are most s u i t a b l e for coals and cokes. From the foregoing c r i t i c a l d i s c u s s i o n , the GM i s s t i l l considered to be one of the most su i tab l e models f o r the C a O - H 2 S or the C a 0 - S 0 2 g a s - s o l i d systems; and many a u t h o r s [27, 81 , 84 , 82 , 26 , 142 ] have s u c c e s s f u l l y appl ied the GM to t h e i r desu lphur iza t ion using ca l c ines as sorbents . In t h i s present study the GM i s employed to model the H 2 S + CaO r e a c t i o n because the model expresses the behavior of monotonical ly decreasing rate and surface area during s u l p h i d a t i o n [20]. A l s o , i t s assumption of uniform i n d i v i d u a l gra ins was approximately met according to observations of the ca l c ines using SEM. Szekely and coworkers [176-179] have proposed the GM model in which the f r a c t i o n r e a c t e d and t ime c o u l d be re la t ed through a s i n g l e , dimensionless express ion, which i s v a l i d regardless of the geometries of 76 the p e l l e t and the i n d i v i d u a l g r a i n s , provided that s t r u c t u r a l change on reac t ion i s n e g l i g i b l e . According to t h i s theory, a porous p e l l e t of volume Vp and of s u p e r f i c i a l area Ap, i s made up of nonporous i n d i v i d u a l g r a i n s , separated by pores through which the react ing gases d i f f u s e , having a volume and surface area Vg and Ag r e s p e c t i v e l y . Both the macroscopic s o l i d p e l l e t and the grains may have the shape of a sphere, a c y l i n d e r or an i n f i n i t e s l a b . I t i s assumed that the reactant gas d i f fuse s through the product l ayer surrounding i n d i v i d u a l grains and reacts at the in ter face between the product layer and the unreacted core . The s o l i d wi th in the unreacted core or a g r a i n i s assumed to be s t a t i c p h y s i c a l l y and c h e m i c a l l y . F u r t h e r m o r e , the model assumes i s o t h e r m a l b e h a v i o r , t h e v a l i d i t y o f the p s e u d o - s t e a d y - s t a t e a p p r o x i m a t i o n and e q u i l i m o l a r c o u n t e r d i f f u s i o n . The f o l l o w i n g a d d i t i o n a l assumptions are made: 1. The res i s tance due to external mass t r a n s f e r i s n e g l i g i b l e . 2. The r e a c t i o n between gas and s o l i d i s of f i r s t - o r d e r wi th respect to reactant and i s i r r e v e r s i b l e . 3. The unreacted p o r t i o n of the grains w i l l be s i m i l a r to the shape of the o r i g i n a l g r a i n . Consider an i r r e v e r s i b l e g a s - s o l i d r e a c t i o n according to the scheme of equation (14). Within the framework of these assumptions the problem may be s tated by combining an expression for the conservat ion of the gaseous reactant with a mass balance for the react ing s o l i d . Thus, DeV 2 C. - v = 0 (101) 77 where De i s the e f f e c t i v e d i f f u s i o n c o e f f i c i e n t w i t h i n the porous medium, i s the reactant concentrat ion and v^ i s the l o c a l rate of consumption of A. The l o c a l rate of reac t ion at a s o l i d surface may be expressed as: p n T ^ r " — k C A ( 1 0 2 ) where p m i s the molar d e n s i t y of the s o l i d , r ' i s the distance coordinate perpendicular to the moving r e a c t i o n s u r f a c e , b i s the s t o i c h i o m e t r i c f a c t o r , and k i s the reac t ion rate constant . Equations (101) and (102) may be expressed i n dimensionless forms of the governing equations as fo l lows: *2 "2 n F -1 V 5 - 2 F g F p O fie 8 = 0 (103) If* = " a" (104) C A where: $ = — — (105) ° A . e = ( 1 0 6 ) *8V8 Fg and Fp are shape fac tors for g r a i n and p a r t i c l e r e s p e c t i v e l y (=1, 2 A and 3 f o r f l a t p l a t e s , c y l i n d e r s and sphere, r e s p e c t i v e l y ) and a i s the genera l i zed g a s - s o l i d reac t ion modulus. Equation (103) expresses the conservat ion of the gaseous reactant and Eq (104) descr ibes the reac t ion 78 of i n d i v i d u a l g r a i n s . The boundary condi t ions for equations (103) and (104) are: £ = 1 at t* = 0 (107) 5 = 1 at n, = 1 (108) and d3> — = 0 at n = 0 (109) where: n - jfir (no) with R i s the dimensional lengths coordinate w i t h i n p e l l e t , and the dimensionless time t* wr i t t en as fo l lows: bkC" Ao aPm F g V g (111) Equations (107), (108) and (109) express tha t , i n i t i a l l y , a l l the grains are u n r e a c t e d , t h a t the c o n c e n t r a t i o n of the gaseous reactant i s maintained at a constant value at the outside of the p a r t i c l e , and the existence of symmetry. In genera l , the system of equations (101)-(111) has to be solved n u m e r i c a l l y . The g e n e r a l i z a t i o n g a s - s o l i d reac t ion modulus appearing in equation (103) i s defined as fo l lows: V p a = —c-A p \ ( i - o k cj;1 F , 2De F g V g (112) 79 where c i s the p o r o s i t y of the s o l i d p a r t i c l e . o represents , in p h y s i c a l terms, the r a t i o of the a b i l i t y the system to react chemical ly to the a b i l i t y of the system to transport the gaseous reactants in to the p e l l e t . The l i m i t i n g behavior of the system may provide the most convenient form in which the r e s u l t s may be presented. 1 - Chemical r eac t ion c o n t r o l When the rate of chemical r e a c t i o n i s much slower than that of A d i f f u s i o n w i t h i n the p a r t i c l e , i . e . a-*0, the reactant concentrat ion i s uniform throughout the p a r t i c l e and equal to t h a t at the e x t e r n a l surface (5=1). Under these cond i t i ons , the o v e r a l l conversion Xgg i s g iven by the r e l a t i o n s h i p [165]: t* = g F g ( X S B ) (113) A where the conversion funct ion gpQ for small a i s def ined as: g F g ( X S B > - 1 " ( 1 - X S B ) 1 / F g < 1 1 4 > For s p h e r i c a l gra ins : 8 F g ( X S B ) " 1 " ( 1 - X S B ) 1 / 3 ( 1 1 5 ) Equation (114) corresponds to that used by Levenspie l [187] der ived on the assumpt ions t h a t the s o l i d r e a c t a n t i s non-porous, the s o l i d 80 reac t ion product i s porous and the process i s c o n t r o l l e d by chemical r eac t ion at the s o l i d reactant-product i n t e r f a c e . For the case of t h i s present study, these assumptions are appl ied to the i n d i v i d u a l gra ins - not to the p a r t i c l e . 2 - Produc t - l ayer d i f f u s i o n c o n t r o l When the rate of chemical r eac t ion i s much f a s t e r than that of A d i f f u s i o n i . e . a -» » , the concentrat ion of the gaseous reactant becomes zero at the in ter face between the completely reac ted and u n r e a c t e d zones. The fo l lowing asymptotic s o l u t i o n w i l l apply [165]. Fp SB (116) where: P F p < X X B > " SB X SB + ( 1 - X S B > L n ( 1 - X S B > 1 - 3 < 1 - X S B ) 2 / 3 + 2 " - X S B > f o r Fp = 1 for Fp = 2 for Fp = 3 (117) (118) (119) Equation (119) corresponds to the case of the s p h e r i c a l "shr ink ing-core -raodel" c o n t r o l l e d by d i f f u s i o n through the product layer [187]. A In the intermediate region (0.3 < a < 3.0) where both chemical k i n e t i c s and product layer d i f f u s i o n are s i g n i f i c a n t , the f o l l o w i n g approximate r e l a t i o n s h i p has been proposed [178,179]. fc* = * F g X SB> + ; 2 p F g ( X S B ) (120) 81 Borgwardt [27] succes s fu l l y appl ied the GM presented here to model the s u l p h i d a t i o n reac t ion between calc ium oxide and hydrogen su lphide . The asymptotic Xgg-t r e s u l t appl ied for s p h e r i c a l non-porous grains of ca l c ines f o r the case of chemical reac t ion c o n t r o l i s w r i t t e n as: t %*r- r g [ 1 - ( 1 - X S B ) 1 / 3 ] (121) and f o r the case of reac t ion product l i m i t i n g P" t - -6§S!r r e 2 [ 1 - 3 ( 1 " X S B ) 2 / 3 + ^ S B ' 1 ( 1 2 2 ) where t i s the r e a c t i o n time, p"caO * s t n e true (He) molar dens i ty of CaO, and C i s the gas phase concentrat ion . 2 .5.7 KINETIC DATA ON THE SULPHIDATION PROCESS The r e a c t i o n of hydrogen sulphide w i th in the porous p a r t i c l e of lime i s a complex process [81]. E a r l i e r s tudies with s p h e r i c a l samples of 5-8 mm i n diameter, by Squires et a l . [168] had shown that the CaO + H2S r e a c t i o n occurs homogeneously throughout the s o l i d and i s a f i r s t order r e a c t i o n wi th respect to unreacted CaO, provided that the p a r t i c l e i s approximately < 5 mm i n diameter and at temperatures below 9 0 0 ° C . Thus: d[CaQ] dt - K [CaO] (123) 82 where [CaO] i s the concentrat ion of unreacted calc ium o x i d e i n the stone, t i s time and K i s the reac t ion rate constant dependent upon temperature and presumably a l so dependent upon gas composit ion. I t can be w r i t t e n as: K = Ko [ H 2 S ] n (124) where Ko i s the constant of p r o p o r t i o n a l i t y (dependent on temperature), [ H 2 S ] i s concentrat ion of hydrogen sulphide in mole percent and n i s the power. Express ion (123) i s v a l i d only when the calc ium oxide throughout the stone i s r e a d i l y access ib le to the reac t ing gas [168]. The degree of inf luence of [ H 2 S ] on the s u l p h i d a t i o n r e a c t i o n rate was examined by severa l authors [101,152,158]. Values of n at var ious temperatures were determined by Richards [168] as in F igure 2 .5 . At an H 2 S concentrat ion up to 5% volume at 7 0 0 ° C and for the p a r t i c l e s i z e of 0.389 mm, Schre iber [152] determined the rate constant f o r reac t ion of H 2 S wi th lime as: K ( s _ 1 ) = 1.50 x 1 0 - 4 [ H 2 S ] 1 - 0 (125) and wi th c a l c i n e d dolomite as: K ( s _ 1 ) = 1.66 x 10~ 4 [ K 2 S ] 0 . 8 5 (126) whereas Karaath [101] reported a value of K f o r reac t ion with ca l c ined dolomite: K ( s _ 1 ) = 1.43 x 10~ 4 [ H 2 S ] 1 - 0 6 (127) 83 F i g . 2.5 Value of n i n equation (124) [149] 84 In genera l , the rate of change of [CaO] increases with an increas ing temperature or an increas ing of hydrogen sulphide concentrat ion [168] as shown in Figure 2 .6 . Schreiber [152] conducted experiments with simulated low c a l o r i f i c value gas to focus on the e f fec t of carbon monoxide on the rate of s u l p h i d a t i o n . No s i g n i f i c a n t e f fec t on CO concentrat ion up to 35% i volume was observed. Also hydrogen concentrat ion from 10 to 40% volume has no e f fec t on the reac t ion r a t e . This l a t t e r f i n d i n g d i f f e r s from the r e s u l t s of Borgwardt [27] and P e l l [137] who noted a strong to moderate reduct ion i n rate with the presence of H 2 , r e s p e c t i v e l y . An increase in steam concentrat ion i n the reactor improves the rate of H 2 S with c a l c i n e s [137]. A p lo t of (l-Xgjj) vs t f or reac t ion of H 2 S with c a l c i n e d dolomite , r e a c t i o n (90) at temperature above 9 0 0 ° C revealed a change i n slope of the curves as observed by Richards [168], Narayen [131], Squires [168] and P e l l [137]. The reac t ion shows a high i n i t i a l rate which diminishes to a constant and slow f i n a l r a t e , which i s apparently c o n t r o l l e d by product layer d i f f u s i o n . Turkdogan [183] reported that there i s a marked decrease i n the slope of the curve of the Arrhenius p l o t with increas ing temperature which i s i n d i c a t i v e of the f a m i l i a r g a s - d i f f u s i o n e f f e c t . Bhat ia [19,27] s i m i l a r l y concluded that the reac t ion i s i n i t i a l l y rap id and c o n t r o l l e d by chemical r e a c t i o n , and goes through a sudden t r a n s i t i o n to a much slower regime c o n t r o l l e d by product layer d i f f u s i o n , Borgwardt [27] confirmed that the d i f f u s i o n a l res i s tance developed in the i n t e r i o r 85 -I ! O I O i to - 3 — 10 - 4 ! 0 - 5 I 0 1 Squires 1,2,3 1 Richards 4 , 5 , 6 I 9 0 0 °C 2 / 8 0 0 ° C 3 / • 7 0 0 ° C 9 0 0 ° C 4 -5 -—' 8 0 0 ° C - 7 0 0 ° C 6 1 10 -2 10 -1 10 0 10 Volume H 2 S (%>) F i g . 2.6 Rate constants v s . H^S concen t r a t i ons . Data from Squi res [168] and Richards [149] . 86 of the p a r t i c l e s becomes l i m i t i n g only a f t er the conversion reaches a value of 20% or h igher . The conclus ion of la te - t ime d i f f u s i o n a l contro l was fur ther r e i n f o r c e d wi th Freund's [60] report that the d i f f u s i o n in H 2 S + CaO r e a c t i o n may be in the Knudsen regime with an e f fec t iveness f a c t o r of 0.003, and Squires [168] f i n d i n g t h a t the v a l u e of the observed d i f f u s i v i t y of hydrogen sulphide i n sulphided c a l c i n e d dolomite (at 8 5 0 ° C ) was equal to 6% of that in open space. Borgwardt [27] obtained an a c t i v a t i o n energy of 129.60 kJ/mol for the s u l p h i d a t i o n p r o c e s s . T h i s v a l u e l i e s between the v a l u e of 96.14 kJ/gmol reported by Squires [168] and the value of 188.10 kJ/mol g iven by Freund [60]. These r e l a t i v e l y high values are supposed to belong to the k i n e t i c c o n t r o l regime. However, the strong inf luence of surface area on the react ions and the sudden change i n the Xgg-t curves imply that the rate i s most probably l i m i t e d by a d i f f u s i o n process [27]. The d i f f u s i v i t y through the product l a y e r , Dp, cannot be e a s i l y measured but must be estimated from k i n e t i c r e s u l t s . Thus the values obtained depend on the choice of models used to analyze the data and hence e s t i m a t e s v e r y w i d e l y . For the S02~CaO system, Hartman and Coughl in [81] and Georgakis [67], employed GM to f i t t h e i r k i n e t i c data and r e p o r t e d v a l u e s of p r o d u c t l a y e r d i f f u s i o n c o e f f i c i e n t of 6xl0~13 m 2 / s and 2.5 x 1 0 - * 0 m 2 / s , r e s p e c t i v e l y , whereas Bhat ia et a l . [10], us ing the RPM, found a value of 2.5 x 10~H m 2 / s . Simons and Garman [161] proposed an express ion, r e l a t i n g Dp to su lphat ion reac t ion temperature, which y i e l d s approximately the same r e s u l t s for a l l models 87 Dp = 5 x 10" 6 e - 1 4 , 0 0 0 / T ( 1 2 8 ) where temperature T i s i n K, and Dp i n m 2 / s . The lowest d i f f u s i v i t y in CaO was recorded by Bhat ia et a l . [19] f o r a system of C 0 2 - C a O r e a c t i o n ; the values ranged from l O - * ® to 1 0 " 2 2 m 2 / s . The slow product layer d i f f u s i o n phenomena i s best explained by e i t h e r the migrat ion of ions through a s o l i d [9,51] or by the a c t i v a t e d -s tate d i f f u s i o n model [19,27]: d i f f u s i o n of S - toward the product layer / CaO in ter face and counterd i f fus ion of 0 - to the product layer / gas sur face . The surface reac t ion i s thus postulated as: H 2 S + 0" •* S - + H 2 0 (129) A low d i f f u s i o n c o e f f i c i e n t might a lso be a s s o c i a t e d w i t h the thermal ly induced l a t t i c e defect i n the s o l i d [14]. Shewmon [156] and Arn ikar [9] used i r r a d i a t i o n t e c h n i q u e s to determine the migrat ion of ions i n s o l i d s . The d i f f u s i o n c o e f f i c i e n t of a l k a l i earth C a 2 + ions in to a s ing le c r y s t a l l i t e of c a l c i t e at 3 0 0 ° C was measured at 4 .9x l0 - *5 m 2 / s [9]; and the d i f f u s i o n c o e f f i c i e n t of the sodium ion was found to be between l O - * 7 and 1 0 - * 2 m 2 / s . S i m i l a r l y , Edstrom [51] a lso experimental ly measured the migrat ion of i ron ions t h r o u g h hemat i t e at a d i f f u s i v i t y of 9 x l 0 - ^ 4 m 2 / s . Moreover, the d i f f u s i v i t y through CaS product l ayer in the CaO + H 2 S reac t ion was estimated by Borgwardt [27] at a value of 3 x l 0 - ^ 2 m 2 / s . 88 2.5.8 CONTINUOUS FIRST-ORDER MODEL The cont inuous model (CM) of f i r s t - o r d e r reac t ion rate def ined above by equations (123) and (124) with n=l was employed by many authors [60,101,152,137-149] to model the reac t ion of hydrogen sulphide with c a l c i n e d dolomite and ca l c ined limestone i n the absence of g a s i f i c a t i o n . It was found that the model was s u f f i c i e n t for many cases except at high temperatures where the rate i s presumably c o n t r o l l e d by produc t - l ayer d i f f u s i o n . The i n t e n t i o n of the in t roduc t ion of t h i s model in the present study i s to permit comparison between the reac t ion rate obtained by previous work and the ear ly -s tage rate a r i s i n g from t h i s i n v e s t i g a t i o n . Part of t h i s simple model was mentioned i n sec t ion 2 .5 .7 . The CM f i r s t -order reac t ion rate was developed with dependency on the concentrat ion of unreacted s o l i d eg. calc ium oxide. The reac t ion i s assumed to take place homogeneously throughout the s o l i d between hydrogen sulphide and CaO according to equation (123). Integrat ion of equation (123) y i e l d s : [CaO] t (130) 0 0 or: Ln[CaO] [CaO] t=0 - K t (131) In terras of conversion: [CaO] [CaO] t=0 = 1 - X SB (132) Thus: Ln ( l - X O T J ) = - K t (133) 89 The reac t ion rate constant K i s dependent on react ion temperature, and the inf luence of reactant gas on the rate was described i n equation (124). Equation (133) impl ies that , i n the k i n e t i c c o n t r o l l i n g regime, the logari thm of the sorbent conversion i s a l i n e a r funct ion of re tent ion t ime. As- f r e q u e n t l y ment ioned , the p h y s i c a l pore s t ruc ture p lays an important r o l e i n the reac t ion of c a l c i n e d samples w i th H2S. In the next s e c t i o n , three of the many techniques of c h a r a c t e r i z a t i o n of porous s t ruc tures are b r i e f l y descr ibed; extensive coverage of these top ic s i s g iven elsewhere [175,71,12]. 2.6 POROSIMETRY - BET SURFACE MEASUREMENT 2.6.1 MERCURY POROSIMETRY Mercury penetrat ion i s a d i r e c t approach to f i n d pore volume and pore s i ze d i s t r i b u t i o n of pores w i t h i n the diameter range of 0.03 um and 200 um. Consider a uniform c y l i n d e r pore of radius r , which in ter sec t s the surface of a porous p a r t i c l e . If the p a r t i c l e i s placed i n a container that i s f i r s t evacuated, and then f i l l e d with mercury, a non-wetting l i q u i d , the force required to move the mercury into that c y l i n d r i c a l pore, Pp, must equal or exceed the surface tens ion force , Ps, that i s : 90 Pp ^ " P s (134) where P p = n r 2 P - (135) P s = 27ir"tfcos 6 (136) Here P i s the pressure app l i ed , ~6 the surface tens ion and 9 i s the contact angle between mercury and the pore w a l l . At 2 5 ° C , has an approximate value of 0.473 N/m. Equate (135) and (136): P = - 2 ^ cos 6 / r (137) Using t h i s equation, a pore s i ze d i s t r i b u t i o n can be c a l c u l a t e d from a g r a p h of the volume of mercury intruded into the p a r t i c l e against pressure [175,71,5] . In t h i s method, on increas ing pressure , the large pores are f i l l e d f i r s t . Since the volume of a l l pores l a r g e r than r i s g iven by: v 00 r 2 L ( r ) dr (138) where L ( r ) i s the pore length d i s t r i b u t i o n with r a d i i r . Then - n r 2 L ( r ) = v ( r ) (139) U t i l i z i n g equation (137), the volume d i s t r i b u t i o n funct ion becomes: V > - - - -f- <»») 91 when the r ight -hand s ide of equation (140) i s p l o t t e d as a funct ion of r , the r e s u l t i n g d i s t r i b u t i o n curve gives the volume of pores which have a given r a d i u s . Even though the experimental measurement i s f a i r l y s imple, the i n t e r p r e t a t i o n of the r e s u l t s i s sometimes d i f f i c u l t . The contact angle 9 may depend on the s o l i d mater ia l s and on impur i t i e s on the surface or i n the mercury. It i s u s u a l l y assumed to be 1 3 0 ° [5] or 1 4 0 ° [71,181] by many workers. The e r r o r introduced by t h i s assumption i s near ly always small compared to other e r r o r s involved i n i n t e r p r e t a t i o n . The large e r r o r s occur i n s o l i d s with bot t l e -neck l i k e pore system where large pores are f i l l e d only at high pressures that correspond to the diameter of the neck; these pores are thus erroneously assigned as small pores . In such bot t l e -neck l i k e pores, the pores w i l l not completely empty as the mercury pressure i s lowered causing the r e t r a c t i o n curve to d i f f e r from the i n t r u s i o n curve in a mercury volume versus pressure p lo t [175,95]. Mercury i n t r u s i o n porosimetry r e s u l t s should be used i n conjunct ion wi th other measurements, and considered as proximate data . 2.6.2 BET SURFACE AREA MEASUREMENTS The c o n c e p t s of the Langmuir t h e o r y have been extended to m u l t i l a y e r p h y s i c a l adsorpt ion by Brunauer et a l . [31] whose theory can be appl ied to a l l f i ve types of adsorpt ion isotherm [181]. The BET isotherm takes the form 92 EA / v n < P o - P A ) = 1 ( a - l ) p A (141) + a v mPo where v„ i s the volume ( e x p r e s s e d at standard cond i t ions ) of gas adsorbed on u n i t mass of porous s o l i d , v m the volume of gas adsorbed at monolayer coverage, p A the p a r t i a l pressure of the adsorbed species i n the gas phase, p 0 the vapour pressure of the adsorbate, and a constant , c h a r a c t e r i s t i c of the system. The BET isotherm i s often used for the experimental determinat ion of the s p e c i f i c surface area of a porous s o l i d . T h i s i s done by c a r r y i n g out a s e r i e s of adsorpt ion measurements at var ious p a r t i a l pressures and then p l o t t i n g PA/(Po~PA) v n against p&/p0. In accordance with equation (141) a s t r a i g h t l i n e should be obtained. v m may be c a l c u l a t e d from the s lope S* and in tercept I because I f the area occupied by one molecule of the adsorbed spec ies , A J Q , i s known, the surface area per u n i t mass Sg may be c a l c u l a t e d from: where N A i s Avogadro's number and Vg the volume occupied by one mole of gas at standard c o n d i t i o n s . The value of AJ,J f o r l i q u i d N 2 at 77K was determined [35] to be 16.2A. Among the assumptions made i n d e r i v i n g the Langmuir isotherm, the most questionable are those regarding the energy of adsorpt ion and the i n t e r a c t i o n s among the adsorbed spec ies . Real surfaces are always nonuniform e n e r g e t i c a l l y . On such a surface , the energy of adsorpt ion tends to decrease with coverage because s i t e s with v m = (I + S ' ) - l (142) Sg - ( v m N ^ / v g ) ^ (143) 93 h i g h e r energy of a d s o r p t i o n are o c c u p i e d f i r s t . However , t h e i n t e r a c t i o n between the adsorbate molecules tends to cause the energy of adsorpt ion to increase with coverage. In many cases, these two opposing e f f e c t s c a n c e l each o t h e r out approximate ly . The success of the Langmuir isotherm i n descr ib ing the adsorpt ion on many r e a l surfaces may be a t t r i b u t e d to these compensating f a c t o r s . 94 CHAPTER 3 EXPERIMENTAL PROCEDURES AND APPARATUS 3.1 INTRODUCTION In t h i s chapter , the experimental procedures which ou t l ine each opera t iona l step employed in t h i s present work are discussed in d e t a i l . The f i r s t part of t h i s chapter deals with m a t e r i a l s , sample preparat ions and c h a r a c t e r i z a t i o n ; t h i s i s f o l l o w e d by the d e s c r i p t i o n of the e x p e r i m e n t a l a p p a r a t u s . F i n a l l y , the experimental techniques are presented. The sulphurous o i l sand delayed and f l u i d cokes were g a s i f i e d using steam to produce f u e l gases accompanied by hydrogen su lph ide . C a l c i n e s , p r o d u c t s of thermal decomposition of l imestones and dolomites , were introduced in to the reactor to capture sulphur i n - s i t u . U s i n g the procedures described i n t h i s chapter , the carbon and sulphur conversions of the s o l i d , the reduct ion i n H 2 S l e v e l s i n the off -gases and the conversion of sorbents were q u a n t i t a t i v e l y r e c o r d e d . The p h y s i c a l s t ruc ture of the s o l i d s at var ious operating condi t ions were examined by means of porosimetry, scanning e l ec t ron microscopy and s u r f a c e area determinat ion . The l a b o r a t o r y - s c a l e s t i r r e d bed was used for a l l g a s i f i c a t i o n exper iment s i . e . g a s i f i c a t i o n and t h e c o m b i n e d g a s i f i c a t i o n -desu lphur iza t ion experiments. The instrumentat ion and contro l s included with the g a s i f i e r apparatus enabled the o p e r a t i o n to be kept near isothermal and the reac t ion temperatures to be c o n t r o l l e d accurate ly . 95 3.2 MATERIALS, SAMPLE PREPARATION AND CHARACTERIZATION 3.2.1 MATERIALS The delayed coke ( D . C . ) and f l u i d coke ( F . C ) , byproducts of the thermal cracking processes of Athabasca bitumen, were purchased and transported from Suncor L t d . and Syncrude L t d . , r e s p e c t i v e l y , in 200 l i t r e b a r r e l s . In the as -rece ived c o n d i t i o n s , the s i ze d i s t r i b u t i o n s of D . C . and F . C . are given in Figure 2 .1 . These cokes were the p r i n c i p a l s o l i d reactants used in the g a s i f i c a t i o n s t u d i e s . C a r b o n a t e r o c k s of d i f f e r e n t types were obtained from several sources represent ing d i f f e r e n t geo log ica l o r i g i n s . The ca l c ines formed from these rocks , prepared according to the procedures mentioned i n the subsequent part of t h i s s e c t i o n , have the p o t e n t i a l to capture sulphur. The Cache Creek limestone was donated by Stee l Bros . L t d . of Cache Creek, B . C . i n 200 l i t r e b a r r e l s from t h e i r t y p i c a l sample stock of the q u a r r y . In a s i m i l a r way, the Texada limestone was a lso suppl ied d i r e c t l y from the quarry on Texada I s land , B . C . The t h i r d sample was an A l b e r t a l imestone, c o l l e c t e d i n the v i c i n i t y of the Syncrude operat ion i n Athabasca; and donated to t h i s projec t by Energy Research Laboratory of CANMET. The above samples of limestone were coarse and v a r i e d in s i z e s . Dolomite rocks , quarr ied i n the B.C.-Washington State border r e g i o n , was purchased l o c a l l y i n bags of 20 kg . Table 3.1 gives some p r o p e r t i e s of the sorbents . 96 Table 3.1 APPARENT CHARACTERISTICS OF FOUR DIVERSE TYPES OF CARBONATE ROCKS EMPLOYED IN THIS STUDY. Carbonate Rocks Calc ined Samples Geo log i ca l O r i g i n Rock or Minera l Ca/Mg molar r a t i o Size Range, mm Apparent Appearance, Co lor Color Cache Creek, B . C . Limestone 139.0 0.02-4 whi te -p ink-gray, hard stone white Texada, B . C . Limestone 115.0 0.05-3 dark gray, hard stone white Syncrude, Fort McMurray, A l b e r t a Limestone 79.3 0.12-4 cement- l ike gray, sof t texture l i g h t brown K e t t l e F a l l s , Washington State Dolomite 1.10 0.68-1.3 White, hard stone white Table 3.2 SIZE FRACTIONS OF D . C . AND CARBONATE ROCKS EMPLOYED IN THIS STUDY Size# Average, mm Smaller than, mm Larger than, mm 1 2 3 4 5 6 3.50 2.00 1.30 1.00 0.68 0.14 4.75 2.36 1.40 1.18 0.85 0.20 2. 1. 1. 0. 0. 0. 36 72 18 85 50 075 97 3.2.2 SAMPLE PREPARATION AND CHARACTERIZATION Each of the D . C . and c a r b o n a t e r o c k s were s e p a r a t e l y d r i e d overnight at room temperature, then crushed and s ieved to obta in s i ze f r a c t i o n s needed for the experimental work. To reduce the s ize of D . C . and carbonate rocks , a Baun(^) jaw crusher , equipped w i t h a 2 HP AC motor , was u s e d . B e f o r e each operat ion , the minimum jaw opening was set to the required s i z e , which u s u a l l y was approximately equal to the s i ze f r a c t i o n most needed, by means of a r o t a t i n g adjus tor . The large p a r t i c l e s were then fed into the crusher from the top at a rate of about 100 cc s o l i d / m i n . , and the crushed products were discharged into a s t e e l t ray located beneath the machine. These smal ler p a r t i c l e s were i n turn fed to the Gibson( R ) S iev ing machine. The s i ev ing machine operates by v i b r a t o r y motion and i n batch mode. Each batch, about 2 l i t r e s of the product , prev ious ly withdrawn from the jaw crusher , was spread onto the top rectangular (50x100 cm) screen. A f t e r a r e t e n t i o n time of about 20 minutes, the needed s i ze f r a c t i o n s were c o l l e c t e d from bottom t r a y s . The s i x s i ze f r a c t i o n s of in t ere s t to t h i s study are shown i n Table 3 .2 . This sequence of c r u s h i n g and s i e v i n g was repeated u n t i l a s u f f i c i e n t amount of s o l i d i n the s i ze f r a c t i o n s of i n t e r e s t was obtained to c a r r y out the large number of experiments planned. Due to the f a i r l y u n i f o r m s i z e d i s t r i b u t i o n of the F . C , no crushing was required; however, i t was s i e v e d to o b t a i n the s i z e f r a c t i o n s i m i l a r to t h a t of D . C . (between 90 and 150 pm) and to e l iminate some bulk fore ign i m p u r i t i e s . 98 As mentioned in Sect ion 2 .5 .2 , l imestones and dolomites react with H 2 S only when c a l c i n e d [110,151]; thus they must be c a l c i n e d p r i o r to t h e i r use in H 2 S removal [183] [ sect ion 2 . 5 . 5 ] . Borgwardt [27,28] also e s tab l i shed that l i t t l e inf luence on the c a l c i n e s t ruc ture would be expec ted i f the c a l c i n a t i o n temperature i s higher than subsequent s u l p h i d a t i o n temperature. I f the su lph idat ion temperature i s higher than the p r e v i o u s c a l c i n a t i o n temperature, however, then the pore s t r u c t u r e of the u n s u l p h i d e d l ime can be a l t e r e d by the h i g h e r temperature of s u l p h i d a t i o n . To avoid such u n c o n t r o l l a b l e s t r u c t u r a l changes and to e s t a b l i s h r e l i a b l e k i n e t i c data on the s u l p h i d a t i o n p r o c e s s , a l l s o r b e n t s u t i l i z e d i n t h i s s tudy were prepared at a temperature of 9 5 0 ° C , which was marginal ly higher than t h a t of the s u l p h i d a t i o n . The parameters i n the c a l c i n a t i o n process include re t en t ion t ime, p a r t i c l e s ize and geo log ica l o r i g i n . C a l c i n a t i o n was c a r r i e d out i n an e l e c t r i c a l l y heated and e l e c t r o n i c a l l y c o n t r o l l e d oven, descr ibed i n sec t ion 3 .3 . The r e s u l t i n g porous c a l c i n e s were c h a r a c t e r i z e d by a c o m b i n a t i o n of means i n c l u d i n g p o r o s i m e t r y , e lectronmicroscopy and surface area determinat ion, in conjunct ion with major component and trace element analyses. 3.2.3 SAMPLE CHARACTERIZATION Representat ive samples of the s ized D . C . and F . C . were sent to Commercial Tes t ing and Engineering Co. of Vancouver for analyses . The r e s u l t s of proximate and ult imate analyses of D . C . and F . C . were given in Table 2.4 and Table 2.5 of sec t ion 2.2. Trace metals analyses of the D . C . were made by Cantest , Vancouver, and the r e s u l t s were compiled in 99 Table 2 .3 . The r e s u l t s of analyses for samples employed in the present research f a l l w i th in the range reported i n the l i t e r a t u r e for the same m a t e r i a l s . Plasma spectroscopy was employed by Cantest to determine both major components and trace elements of representat ive samples of sorbents . The r e s u l t s of these analyses are shown i n Appendix E . L . O . I , was determined g r a v i m e t r i c a l l y a f t er heating the sample at 9 5 0 ° C for one hour. I t i s e v i d e n t t h a t a s i d e from the o r i g i n and nature of the carbonate rock i t s e l f , the s t r u c t u r a l proper t i e s such as p o r o s i t y , pore s i z e d i s t r i b u t i o n , pore volume, surface area and g r a i n s i ze of the c a l c i n e formed by thermal decomposition of limestone and dolomite are s trong ly d i c t a t e d by the operating condi t ions of the c a l c i n a t i o n . These s t r u c t u r a l proper t i e s i n turn af fect the c a p a c i t y and r a t e of the c a l c i n e wi th H 2 S [1 ,27 ,81 ,48] . In l i g h t of these f i n d i n g s , the samples of s o l i d s employed i n t h i s work were charac ter i sed by determining the p o r o s i t i e s , and pore s i ze d i s t r i b u t i o n s us ing the mercury penetrat ion technique, the surface area us ing BET isotherm a d s o r p t i o n - d e s o r p t i o n w i t h N 2 at 77K, and the g r a i n s i z e s u s i n g the scanning e l ec tron microscope. The p r i n c i p l e s of these methods were p r e v i o u s l y presented; the ac tua l experimental procedures are included i n s ec t i on 3.4. 100 3.3 EXPERIMENTAL APPARATUS 3.3.1 THE CALCINER S o r b e n t s f o r t h e r e m o v a l o f t h e s u l p h u r r e l e a s e d d u r i n g g a s i f i c a t i o n of o i l sand cokes were prepared in a Thermolyne(^), type 1800 laboratory muffle furnace. The samples were charged into h igh-temperature ceramic bowls, and thermal ly decomposed in a N 2 atmosphere. A schematic view of the furnace i s shown in F i g u r e 3 . 1 . The f u r n a c e i s b a s i c a l l y a thermal ly i n s u l a t e d , and e l e c t r i c a l l y heated chamber whose dimensions were 24.1 cm h igh , 25.0 cm wide and 55.9 cm deep. I ts e l e c t r i c a l s p e c i f i c a t i o n s are 240 volts /3phase/30.2ampere/ 12,600 watts . The furnace maximum operating temperature i s 1 2 6 0 ° C . To measure the chamber temperature, a P t / P t 13% Rh thermocouple i s located at o n e - t h i r d from the top and at the center l i n e of the back of the chamber. Both the c e i l i n g and the f l o o r are covered by e l e c t r i c a l heat ing surfaces which cons i s t of iron-chromium-aluminum a l l o y s elements wires embedded and supported by s p e c i a l r e f r a c t o r y form. At the beginning of each c a l c i n a t i o n , l i m e s t o n e and d o l o m i t e p a r t i c l e s were loaded in to 6.4 cm ID ceramic bowls, covering on top of the N 2 s p i r a l d i s t r i b u t o r . In operat ion , n i trogen gas can then be purged from high pressure tanks through a loop of 316 s t a i n l e s s s t e e l tubes of 1.6 mm i n diameter. The sec t ion of the loop ins ide the ceramic sample bowls was a f l a t closed-end s p i r a l with openings from which N 2 was released to form a n i trogen c a l c i n a t i o n atmosphere. This set up i s i l l u s t r a t e d i n Figure 3.2. Nitrogen supply loop C a s e Soft insulation T / c protection tube \ — T h e r m o c o u p l e Carbonate sample C e r a m i c bowl Heat ing element F i g . 3.1 Schematic view of the c a l c i n e r i—1 o 102 F i g . 3.2 Bowls used i n c a l c i n a t i o n of carbonate rocks. 103 The c a l c i n a t i o n was done i n b a t c h e s ; each b a t c h was f o r one p a r t i c u l a r set of c a l c i n a t i o n cond i t ions . The c a l c i n a t i o n operating v a r i a b l e s were type of carbonate rocks , c a l c i n a t i o n time and p a r t i c l e s i z e , while temperature was kept constant at 9 5 0 ° C , s l i g h t l y above the highest g a s i f i c a t i o n and su lph ida t ion temperature. 3.3.2 THE GASIFICATION SYSTEM Because of the d i f f e r e n t cokes and d i f f e r e n t sorbents which were to be used, i t seemed expedient to employ a reac t ion system that would be able to handle a wide range of p a r t i c l e s i z e s , without feeding problems. A minimum reactor s i ze of about 700 cm^ was needed to operate with q u a n t i t i e s of samples between 100 g and 500 g, and a lso to provide r e l i a b l e r e s u l t s g iven the heterogeneity of the coke and the sorbent. Because the g a s i f i c a t i o n p r o c e s s i s endothermic some method of minimizing temperature gradients i n the bed of s o l i d s was d e s i r a b l e . A s t i r r e d bed was therefore designed and i n s t a l l e d . F i g u r e 3.3 i l l u s t r a t e s s c h e m a t i c a l l y the o v e r a l l experimental arrangement. The system cons i s t s of 5 major components: water and N 2 transport f a c i l i t y , a preheater, a g a s i f i e r , the product gas handling f a c i l i t y and a c o n t r o l panel . Each of these components are descr ibed below. 3 .3 .2 .1 WATER AND N? TRANSPORT Steam f o r the steam g a s i f i c a t i o n r e a c t i o n was generated from d i s t i l l e d water which was i n i t i a l l y stored i n a 12 l i t r e 30 cm diameter p l e x i g l a s s c y l i n d r i c a l container which had a hole in the l i d so that the T / c • a s -z = = : P u m p f i P N. Gas M o t o r [0 H 2 S c y l i n d e r C o o l i n g s e c t i o n React ion, , c h a m b e r G a s i f i e r — - - S l i p T / c S t i r r e r r i n g \/ /; / / / / / s / / // / / / j\ a t> o onoooeoaoooooo at » o o o o & 6 e '£> a o o o a o l t0 to « o o o o ooo o S' oi \ /////////////// s\ PQ T / c a . - 4 0 Y C y c l o n e P Q 4 A b s o r b e r 7 • P T / c V D u s t i c o l l e c t o r Samp l ing po r t C o n d e n s e r C o n i c a l s e c t i o n •a T / c P r e h e a t e r F i g . 3.3 Schematic view of the apparatus employed i n th i s study. o 105 water was at atmospheric pressure . Water was pumped by a M a s t e r f l e x ( R ) pump system through a rotameter to the preheater where i t was vapourized and superheated. The water f lowrate was accurate ly c o n t r o l l e d with a needle valve at f lowrates betwe en 0.06 and 2.9 l i t r e / m i n . A Porter (R) rotameter, wi th a s t a i n l e s s s t e e l b a l l , was prev ious ly c a l i b r a t e d for water and u t i l i z e d to measured water f lowrates . Ni trogen , serving as a heat c a r r i e r medium and as an i n e r t t i e component, was suppl ied by Union Carbide i n k s i ze tanks at 13.8 MPa. A two s tage r e g u l a t o r was employed to br ing the N 2 pressure down to 180 kPa before the gas was metered by a Brooks( R ) rotameter which had a s t a i n l e s s s t e e l b a l l , and was prev ious ly c a l i b r a t e d wi th N 2 . A needle va lve was used to c o n t r o l the n i trogen flow. The c a l i b r a t i o n curves for rotameters are f i l e d i n Appendix A. 3 .3 .2 .2 THE PREHEATER Both the preheater and the g a s i f i e r were constructed of 96.5 cm long , 76 mm, schedule 40 pipe of 316 s t a i n l e s s s t e e l having an ins ide diameter of 7.8 cm and 0.3 cm wal l th ickness . Each i s e l e c t r i c a l l y heated by two f u l l v e s t i b u l e , 2.54 cm t h i c k , 70 cm l o n g , s e m i -c y l i n d r i c a l Watlow( R ) ceramic f i b e r heaters of 4.8 kW/2 phase/240 v o l t s ; and insu la ted by 2.54 cm t h i c k ceramic f i b e r b lanket . The power input to the preheater heaters and for the g a s i f i e r headers were e l e c t r i c a l l y c o n t r o l l e d by the preheater PID and the g a s i f i e r PID c o n t r o l l e r s , r e s p e c t i v e l y . The preheater was completely packed with 1.27 cm ceramic Raschig r ings to r e t a i n the heat and to increase heat t r a n s f e r to the gas [139]. 106 To increase the surface area for vapour i za t ion , the l i q u i d water was forced through a s t a i n l e s s s t ee l d i s t r i b u t o r which was placed i n s i d e , on the top of the packing and along the whole length of the preheater . The drop le t s of water landed on a large surface of the packing where they evaporated at a temperature of 700°C to superheated steam. At the same time, the N 2 was heated in t h i s preheater, and the mixture of steam and n i trogen then passed through a loop of 0.95 cm ID s t a i n l e s s s t e e l tubing to two opposite ports located i n the c o n i c a l s ec t i on of the bottom of the g a s i f i e r . The two d i f f e r e n t p o r t s f o r the heated gases were designed to lower the chance of plugging of these i n l e t s by f a l l i n g s o l i d p a r t i c l e s from anywhere in the top sec t ion of the g a s i f i e r . The pressure i n the preheater was indicated by a Bourdon pressure gauge whereas the temperature near the e x i t of the p r e h e a t e r was measured by a thermocouple inserted from the hot end of the preheater . The preheater was mounted h o r i z o n t a l l y 30 cm above the laboratory f l o o r on a s tab le s t e e l frame, and supported at i t s two ends which were not covered by the s e m i - c y l i n d r i c a l heaters . A l l thermocouples i n s t a l l e d i n t h i s experimental apparatus were Chromel-Alumel, sheathed in Inconel , Type K, 4 .8 mm OD of v a r i o u s lengths; a l l were connected to a ro tary se l ec tor switch and to a E L P H ^ ^ cent igrade d i g i t a l temperature i n d i c a t o r mounted on the c o n t r o l pane l . 3 .3 .2 .3 THE GASIFIER The g a s i f i e r was the p r i n c i p a l component of the experimental system. I t i s e l e c t r i c a l l y heated i n an i d e n t i c a l manner as the preheater , d iscussed above. 107 The lower sec t ion of the g a s i f i e r was packed wi th 1.27 cm Raschig ceramic r ings to r e t a i n heat. In t h i s s e c t i o n , the steam-N2 stream was fur ther heated to the des ired reac t ion temperature (800-930*0. The temperature of t h i s ceramic packed bed at the end c lose to the reac t ion chamber was measured by means of a 45.7 cm long thermocouple f i t t e d t h r o u g h the bottom of the c o n i c a l s e c t i o n . The r e a c t i o n chamber occupied near ly the e n t i r e i n t e r n a l volume of the upper s ec t i on of the g a s i f i e r , and i t was held by flanges at the j o i n t between the upper and the coo l ing sec t ions . The r e a c t i o n chamber was 40.6 cm long, 6.35 cm schedule 40 pipe of 316 s t a i n l e s s s t e e l having an ins ide diameter of 6.3 cm. The bottom of t h i s chamber was a s t a i n l e s s s t e e l a l l o y gas d i s t r i b u t o r , made from Dynapore ( R ) perforated p lates of 90 um openings and 1.65 ram t h i c k . In the l a t e r phase of t h i s study, the "reaction chamber was modif ied to increase surface for g a s - s o l i d contact and adapted f o r the f l u i d coke whose p a r t i c l e s i zes were mostly i n the range between 75 and 200 um, and D . C . of the same s i ze range. The modi f i ca t ion involved truncat ing the e x i s t i n g r e a c t i o n chamber 25 cm from the bottom and r e p l a c i n g i t by a small-opening mesh sample ho lder . This holder cons i s ted of two hollow concentr ic c y l i n d r i c a l frames of d i f f e r e n t diameter, 1.67 and 0.70 cm, as shown i n Figure 3.4. The outer surface of each c y l i n d e r was covered by s t a i n l e s s s t e e l of 49 um standard mesh openings which enabled the reac t ing gas to pass through, yet were small enough to r e t a i n the s o l i d w i t h i n . At the beginning of each t e s t , the s t a i n l e s s s t e e l mesh surrounding each c y l i n d e r was replaced by a new one, and the 109 s o l i d reactants were then charged into the "ring" volume between these two mesh-covered c y l i n d e r s . The upper-most s ec t ion was e s s e n t i a l l y a water-cooled sec t ion which kept the s t i r r e r and i t s gland-nut from thermal expansion. The g land-nut packed the asbestos wire packing inwards, keeping the produced gas from leaking out to the atmosphere and the s t i r r e r shaft v e r t i c a l l y a l i g n e d . Cool ing water at a pressure of 156 kPa entered t h i s sec t ion from the lower i n l e t and warm water l e f t by the higher o u t l e t of the same s i ze of 9.5 mm i n diameter. Thermocouples were placed to measure both i n l e t and o u t l e t water temperature. The t o t a l length of t h i s coo l ing s ec t i on was 9.53 cm. I t was fabr i ca ted from the same s t a i n l e s s s t e e l pipe as for the preheater and the g a s i f i e r . A l s o , the s t r u c t u r a l support for the s t i r r e r was welded onto t h i s s e c t i o n . A t h i c k - w a l l e d , 316 s t a i n l e s s s t e e l tube, 9.5 mm OD, 6.2 mm ID was employed as the s t i r r e r shaf t . Temperature of the r e a c t i o n zone was measured, at 2.5 cm from the bottom of the sample ho lder , by means of an Omega(R) s l i p - r i n g assembly model SR-2 equipped with a thermocouple 81.3 cm long, in ser ted ins ide the hollow shaft . The s l i p - r i n g thermocouple and the thermocouple located at the preheater e x i t were r e s p e c t i v e l y wired to the g a s i f i e r PID and the preheater PID c o n t r o l l e r s , which automat ica l ly monitored and c o n t r o l l e d the temperatures i n the reac t ion zone and i n the preheater independently. The power suppl ied to the g a s i f i e r h e a t e r s was dependent upon the d i f f erence between the set temperature on the g a s i f i e r PID c o n t r o l l e r and the ac tua l temperature read by the s l i p - r i n g thermocouple . The preheater PID c o n t r o l l e r operated s i m i l a r l y . 110 Both the preheater and the g a s i f i e r PID c o n t r o l l e r s were made by Thermo E l e c t r i c ( R ) , with d i g i t a l set point and d i g i t a l d i s p l a y , Series 900. The maximum operating range was 1370 oC. The s t i r r e r was d r i v e n by a G K H ( r ) 1/20 HP high torque motor through a sprocket and chain system. The speed of the s t i r r e r was v a r i e d using the motor c o n t r o l l e r u n i t . When the f lange connecting the coo l ing and the upper sect ions was undone, the coo l ing sec t ion (and the s t i r r e r attached to i t ) could be taken apart as a s ing le u n i t , as could be the reac t ion chamber. Between any two flanges a high-temperature gasket made from P i l o t J o i n t i n g ( R ) , 2 mm t h i c k sheet was p laced . The gasket mater ia l was asbestos, wi th graphite coat ing on e i t h e r s ide and s t e e l w ire mesh r e i n f o r c i n g embedded i n the centre . A frame made from 2.5 cm square s t e e l tube was f a b r i c a t e d to support the g a s i f i e r , the motor for the s t i r r e r , the pump system for water, and the rotameters. The g a s i f i e r and the product gas handling f a c i l i t y were jo ined by a s p e c i f i c a l l y designed l i v e f langes , i l l u s t r a t e d i n Figure 3 .5 . Figure 3.6 shows the sketch of the coo l ing sec t ion plus the s t i r r i n g mechanism. 3 .3 .2 .4 THE PRODUCT GAS HANDLING FACILITY The produced gas m i x t u r e passed from the g a s i f i e r t h r o u g h a s t a i n l e s s s t e e l cyclone and subsequently to a 25 cm long, 8.90 cm ID p l e x i g l a s s g lass wool f i l t e r . Most of the s o l i d s were caught by these two dev ices . The gas was then allowed to cool in a s t a i n l e s s s t ee l condenser, where i t s water vapour became l i q u i d . Before being vented, I l l I *• *8 1\ J 4 — W1TERWL--3IGS5 H 8 9 *] .WlflTYORDCRERI F i g . 3.5 Engineering drawing of the g a s i f i e r and i t s l i v e flange. 112 HOLES 7.6Z 9.53 0=1.27 N.P.T , NfPPLE «S=U7 HKT/ NippLe .COOLING SOON. "MATERIAL 31* 5 S MEASURES IN c m QUANTITY ORDEKEP: I F i g . 3.6 Engineering drawing of the cooling section of the g a s i f i e r . 113 the dry gas leav ing the condenser passed through an externa l calc ium oxide bed absorber, where most of i t s hydrogen sulphide was s tr ipped away. Samples of product gas mixture were taken from ports upstream and downstream from the condenser. The c y c l o n e was 5.04 cm i n d i a m e t e r and capable of removing p a r t i c l e s down to a dp5Q=30 urn. The dust c o l l e c t o r which i s attached at the bottom of the cyclone was emptied a f t er each run, and the c o l l e c t e d dust was weighed. The f i n e r dust trapped i n the g lass wool was removed by hand. The condenser operated using c i t y water which ran on the outer s h e l l , whereas the condensate was c o l l e c t e d i n the tube s ide , and removed from the bottom through a b a l l v a l v e . 3 .3 .2 .5 THE CONTROL PANEL The major features of the c o n t r o l panel were the power supply s w i t c h e s and i n d i c a t o r l i g h t s , the g a s i f i e r and the p r e h e a t e r c o n t r o l l e r , the temperature i n d i c a t o r , the r o t a r y s e l e c t o r switch, the speed c o n t r o l l e r for the s t i r r e r , a mercury and a water manometer. 1. The o n / o f f switches and i n d i c a t o r l i g h t s corresponding to each e l e c t r i c a l device were used as a v i s u a l safety f e a t u r e i n the experimental equipment. 2. The c o n t r o l l e r s had a propor t iona l current of e i t h e r 4 or 20 mA, an adjustable p r o p o r t i o n a l band between 1 and 20% of set p o i n t , and an a c c u r a c y i n se t p o i n t of ± 0 . 5 % w i t h a r e s o l u t i o n of 1 ° C . Temperature i n the reac t ion zone in the g a s i f i e r or at the ex i t of the preheater was d isplayed on the c o n t r o l l e r or a l t e r n a t e l y on the temperature i n d i c a t o r by using the ro tary switch. In most cases, 114 the temperatures given by those two i n d i c a t o r s for the same spot were w i th in ±1°C of each other . 3. The temperatures across the apparatus were convenient ly taken by 8 thermocouples which a l l connected to the ro tary s e l e c t o r switch and the d i g i t a l temperature i n d i c a t o r , mentioned in part 2 above. 4. In a s i m i l a r manner, 6 pressure taps were placed in d i f f e r e n t parts of the apparatus and l inked d i r e c t l y to the manometers by copper tub ing . The pressures were ind ica ted by e i t h e r a mercury or a water manometer, using a three-way b a l l sw i t ch mounted on the pane l . 5. The speed of the s t i r r e r could be v a r i e d from the c o n t r o l l e r knob. However, the s t i r r e r was o p e r a t e d at an e x p e r i m e n t a l l y p r e -determined optimum value of 28 rpm. 3.4 EXPERIMENTAL TECHNIQUES The e s s e n t i a l e x p e r i m e n t a l work i n t h i s r e s e a r c h i n v o l v e s c a l c i n a t i o n of carbonate rocks , g a s i f i c a t i o n of o i l sand coke w i t h steam, and e x a m i n a t i o n of the s o l i d s t r u c t u r a l parameters . The experimental techniques are discussed below. The range of operating condi t ions employed are given i n Table 3 .3 . 3.4.1 CALCINATION Sorbents for the sulphur released during g a s i f i c a t i o n of o i l sand cokes were prepared approximately i so thermal ly in a c a l c i n e r , prev ious ly d e t a i l e d i n Sect ion 3 .3 .1 , under a n i trogen atmosphere and at 9 5 0 ° C . 115 Table 3.3 RANGE OF OPERATING CONDITIONS EMPLOYED IN THIS RESEARCH Condit ions Range Reaction temperatures, °C 800-930 Reaction time, min 0-720 G a s i f i e r pressure , atm 1.0 P i n the g a s i f y i n g gas, atm 0.15-0.60 2 Average s o l i d p a r t i c l e s i z e , mm Cokes 0.1-3.5 Sorbents 0 .1-3.5 C a l c i n a t i o n temperature, °C 950 C a l c i n a t i o n t ime, min 40-300 C a l c i n a t i o n atmosphere N 2 S p e c i f i c surface area of sorbents , m 2 g~l 2-20 Types of carbonate rocks 4 116 As seen in Table 3.3, the c a l c i n a t i o n t empera ture ( 9 5 0 ° C ) i s s l i g h t l y greater than the highest g a s i f i c a t i o n temperature ( 9 3 0 ° C ) used. The h i g h e r c a l c i n a t i o n t emperature was d e s i g n e d to p r e v e n t any u n c o n t r o l l e d changes in the sorbent, from i t s i n i t i a l c a l c i n e d form, from occurr ing during the su lph ida t ion r e a c t i o n , as p r e v i o u s l y described i n Sect ions 2.5 .5 and 3 .3 .1 . To study the e f fec t s on sorbent s tructure parameters (eg. surface areas, p o r o s i t y , g r a i n s ize and pore s i ze d i s t r i b u t i o n ) of c a l c i n a t i o n t imes, g e o l o g i c a l o r i g i n s , types of carbonate rocks and p a r t i c l e s i ze s , four d i f f e r e n t samples of carbonate rocks and var ious r e t e n t i o n times were employed . The e x p e r i m e n t a l steps were performed as fo l lows: I n i t i a l l y four ceramic bowls were arranged ins ide the c a l c i n e r . This was f o l l o w e d by the l o c a t i o n of the s t a i n l e s s s t e e l s p i r a l N 2 d i s t r i b u t o r s at the bottom, ins ide these bowls. These d i s t r i b u t o r s were then f i t t e d to n i trogen supply l i n e s , o r i g i n a t i n g from N 2 tanks . Next, raw samples of dolomite and limestone were weighed (approximately 180 gm each) and loaded i n t o separate bowls, on top of the d i s t r i b u t o r s . Ni trogen was then allowed to flow to these d i s t r i b u t o r s at a rate of 1.4 1/min, measured by a rotameter on the main l i n e from the tanks, and the c a l c i n a t i o n temperature was set and locked at 9 5 0 ° C on the c o n t r o l l e r attached to the c a l c i n e r . A f t e r manually checking to make sure that N 2 was d e l i v e r e d to each bowl, the s ec t i ona l doors of the furnace were c losed; and the c a l c i n a t i o n s t a r t e d . The c a l c i n a t i o n temperatures were recorded as a funct ion of time. A t y p i c a l temperature p r o f i l e i s presented in Figure 3 .7 . From the s t a r t u n t i l the temperature reached 9 5 0 ° C , the heating rate was about 9 . 2 5 ° C / m i n . The c a l c i n a t i o n time was 117 T E M P E R A T U R E PROFILE IN CALCINATION EXPERIMENTAL DATA 1200 o « W PH w O > — • 2 o <J o 900 600 300 0 • 1 • • • 4 • • • LEGEND • = CALCINATION TEMP. 0 60 120 180 TIME, MIN. 240 300 F i g . 3.7 Temperature p r o f i l e during c a l c i n a t i o n . 118 defined as the time elapsed from the time at which temperature f i r s t reached the preset temperature of 9 5 0 ° C . When the c a l c i n a t i o n time had been completed, the power to the furnace was shut o f f , the doors were opened wide, and the c a l c i n e d samples were allowed to cool i n N 2 , ins ide the c a l c i n e r , f or about 20 to 30 minutes, to a furnace temperature of about 1 2 0 ° C . The n i trogen was then turned o f f and the supply l i n e s were disengaged. The ca l c ined products were t r a n s f e r r e d , while s t i l l hot , into l a b e l l e d ground-neck g lass b o t t l e s . F i n a l l y the c a l c i n e ins ide each b o t t l e was mixed to obta in homogeneity by gent ly r o t a t i n g the bot t l e at d i f f e r e n t p o s i t i o n s . S m a l l e r samples were extracted from the bulk for chemical analyses , S . E . M . , surface area de terminat ion and pore s i z e measurement; the remainder was set aside for the experimental s u l p h i d a t i o n r e a c t i o n s . The sulphur removal experiments employing a given c a l c i n e d sample were c a r r i e d out w i th in 5 weeks from the time the carbonate rock had been c a l c i n e d . It was postulated that the c a l c i n e s are s table i n the a i r - t i g h t conta iner , at room temperature for a per iod of more than 6 weeks [27]. 3.4.2 GASIFICATION In the t e s t s where c a l c i n e was needed the amount of c a l c i n e was computed according to the procedure in Sect ion 3 . 4 . 4 . 2 , f or the required Ca/S molar r a t i o . Ca lc ine and the o i l sand coke were weighed separate ly and added into a 500 ml Erlenmeyer F lask which was then slowly rotated u n t i l the coke and the ca l c ine were v i s i b l y wel l -mixed. The s o l i d mixture was then ready to be loaded into the sample h o l d e r . In a 119 t y p i c a l run at a molar C a / S - 2 . 0 , 150 g coke and about 32.5 g ca l c ined l imestone, c o n s i s t i n g of approximately 96 wt% of CaO, were used. The temperature of the preheater was preset at 7 0 0 ° C for a l l t e s t s , and that of the r e a c t i o n zone i n the g a s i f i e r was preset to the des ired value using the thumbwell on the c o n t r o l l e r . N 2 was passed through the system at a rate of 7.5 L/rain, at 2 0 ° C . The s t i r r e r was turned on at 26 rpm; and the h e a t e r s of the preheater and the g a s i f i e r were then switched on. The r e a c t i o n zone reached the preset temperature i n about 60 minutes. When the temperature of the sample had reached the set temperature, 0.97 to 8.27 mL/min of l i q u i d water was in jec ted into the preheater by means of a micropump to form a steam at a p a r t i a l pressure of 15.15 to 60.6 kPa, r e s p e c t i v e l y . The gas product leaving the reactor was sampled every 10 minutes f o r the f i r s t hour, and every 20 minutes for the rest of the experiment by us ing Press Lock Precision^ 1 *) sampling syr inges . The syringes of gas were analysed immediately a f t er the samples were taken. At the same t ime, temperatures and pressures at var ious points across the apparatus were noted. A n a l y s i s for H 2 , N 2 , 0 2 , CH4, CO and C 0 2 was done at 100°C on a Hewlett-Packard^ 1 *), Ser ies 5710A gas chromatograph which was equipped wi th a 51 cm long, 0.32 cm - diameter molecular s ieve and a 390 cm long, 0.32 cm - diameter Poropak Q column. The Poropak Q column, a coarse p a c k i n g , separated C 0 2 from the gas mixture; whereas, the molecular s ieve column, a f i n e r packing, analysed for CO, CH4, N 2 and 0 2 . Argon ( U . H . P . grade) , purchased from L i n d e ( R ) , was the c a r r i e r gas. 120 Hydrogen sulphide and other suphur-containing species such as S 0 2 , COS, C S 2 and mercaptans were analysed using a V a r i a n t ) , s er i e s 6500 G . C . equipped with a 80 cm long carbopack B . H . T . 100 column and a flame photometric detec tor . H 2 and A i r ( L i n d e ( R ) , U . H . P . grade) were used to set up the flame and Ar ( L i n d e ( R ) , U . H . P . ) was used as a c a r r i e r gas. The r e l i a b i l i t y of the gas chromatographs were c a r e f u l l y checked by c a l i b r a t i o n u s i n g s tandard gases, whose concentrat ions were i n the ranges of that of the produced gas and were p r e d e t e r m i n e d by the s u p p l i e r . The c a l i b r a t i o n s were c a r r i e d out r i g h t before each run and sometimes repeated during the run . The detect ion l i m i t s f o r these gases are ~50 vppm. No other sulphurous gas besides H 2 S was detected i n the produced gas. The c a r b o n and s u l p h u r c o n v e r s i o n s were computed from the c o m p o s i t i o n and f l o w r a t e of the gas p r o d u c t , u s i n g N 2 as a t i e substance. The experimental time v a r i e d from 0 to 12 hours (0 to 720 minutes); however, most runs were about 6 hours. When the r e a c t i o n time had e lapsed, the power suppl ies to the heaters , to the s t i r r e r and to the water pump were turned o f f , the flow of n i trogen was lowered to about 2 1/min from about 7.5 1/min, and the system was cooled overnight to room temperature i n an atmosphere of n i t rogen . The f lange holding the c o o l i n g sec t ion and the upper sec t ion of the g a s i f i e r was then undone, and the sample h o l d e r removed. The r e a c t e d s o l i d s a m p l e was subsequently t rans ferred from the sample holder to a l a b e l l e d g lass j a r to s tore f o r use in chemical composition analyses, S . E . M . , porosimetry, surface area determinations, and for future reference. 121 3.4.3 SURFACE AREA MEASUREMENTS AND S . E . M . The surface area of a l l reacted cokes and sorbents were determined by the B . E . T . method, described i n Sect ion 2 .6 .2 , using the n i trogen a d s o r p t i o n i so therms at 77K and employing a Quantasorb( R ) surface measurement instrument. To obta in the r e l i a b l e surface area, which represents the sum of a l l pore surfaces , a c a r e f u l heating procedure for the removal of water from the pores , e s p e c i a l l y from small ones, was fo l lowed. O i l sand coke samples were spread on P e t r i d i s c s and d r i e d in an oven at 110°C for 16 hours (overnight ) . Two glass c e l l s with known weights were f i l l e d with the r e s u l t i n g d r i e d coke; the f i r s t was used i n the adsorb-desorb c y c l e , the second was placed in the degassing p o s i t i o n to be f lushed with N 2 f o r a minimum of 3 to 4 hours . On the Quantasorb( R ) N 2 and H 2 (the c a r r i e r gas) was turned on to 136 kPa, and the detector was set at a current of 150 mA. Before taking ac tua l readings , the f i r s t sample went through the adsorb-desorb cyc le at l east 6 t imes. Th i s was an experimental ly found v a l u e which gave consis tent measured surface area r e s u l t s . In the adsorpt ion stage, the c e l l containing the sample was immersed in l i q u i d n i t r o g e n u n t i l e q u i l i b r i u m . The desorpt ion s tar ted when the l i q u i d n i t rogen was removed; and the sample was allowed to re turn to ambient t e m p e r a t u r e . The adsorbed and desorbed volumes were reported in e l e c t r o n i c counts. These counts were then compared to the c a l i b r a t e d value derived from the i n j e c t i o n of a known volume of N 2 , withdrawn from the N 2 loop, into the detector a f t er each adsorb-desorb cyc l e ; thus, the 122 adsorbed and desorbed volumes were determined. The ambient pressure and temperature were a lso recorded. The s p e c i f i c area of the s o l i d sample was determined from values of the slope and intercept of a s t r a i g h t l i n e constructed according to equation (141), whose v a r i a b l e i s the p a r t i a l pressure of N 2 i n the gas phase, P^. Therefore , at l eas t 5 values of adsorbate f lowrates were employed for each sample to generate f i v e corresponding P^ va lues , and thus 5 data points on the l i n e a r p l o t . The c o r r e l a t i o n c o e f f i c i e n t of these l i n e a r p lo t s were in the range from 0.960 to 0.999. The time spent to measure the s p e c i f i c a r e a f o r one s o l i d s a m p l e was approximately 3 to 4 hours. When a l l the readings needed to compute the s p e c i f i c area for the sample i n the f i r s t c e l l had been completed, the degassed sample i n the second c e l l was i n turn used i n the adsorb-desorb c y c l e . The contents of the f i r s t c e l l were replaced by a new sample which was then attached to the degassed p o s i t i o n to be f lushed with N 2 . The operat ion was repeated u n t i l the surface areas of a l l samples were determined. The surface areas of c a l c i n e d samples were determined, employing the i d e n t i c a l procedure as d e s c r i b e d above f o r the coke samples . Several samples of known surface areas were re-determined as a check for the r e l i a b i l i t y of the Quantasorb( R ) machine. The repeated values of the surface area for those samples were i n agreement with the stated values w i th in ±6%. A l s o , four samples, two cokes and two sorbents , were done i n d u p l i c a t e , w i t h ±3 to ±7% d i f f e r e n c e between t h e two measurements of the same sample. 123 Scanning e l e c t r o n microscopy was employed to examine the s t r u c t u r a l changes of o i l sand cokes at var ious stages of carbon convers ion , of c a l c i n e s from d i f f e r e n t o r i g i n s and at d i f f e r e n t c a l c i n a t i o n c o n d i t i o n s . Each sample was d r i e d i n an oven at 1 1 0 ° C for 3 hours p r i o r to the examination. The sample was then attached to an aluminum base, and coated wi th a gold l ayer of about 100A t h i c k . The examinations were conducted i n a H i t a c h i ( R ) S-570, i n vacuo of 10""^  mbar, and at a voltage of 20 k V . These e x a m i n a t i o n s i n v o l v e d v i s u a l o b s e r v a t i o n a t magni f icat ions from 50X to 3000X and taking of photographs of t y p i c a l areas for future comparisons. 3 .4 .4 ANALYSES PARTIAL PRESSURE OF STEAM IN THE GASIFYING GAS, P R Q The steam p a r t i a l pressure i n the g a s i f y i n g medium was computed based on a constant volumetr ic flow rate of the g a s i f y i n g gas i n a l l t e s t s . Table 3.4 summarizes the amount of water in jec ted at d i f f e r e n t operat ing c o n d i t i o n s , and the sample c a l c u l a t i o n for P i s g iven i n H 2 ° Appendix B. THE MOLAR Ca/S RATIOS In t h i s study, Ca/S molar r a t i o s of 0 .5 , 1.0, 1.5 and 2.0 were employed. These r a t i o s were estimated from the ul t imate a n a l y s i s of the coke and the c o m p o s i t i o n of the sorbent under c o n s i d e r a t i o n . 124 Table 3.4 AMOUNT OF LIQUID WATER INJECTED AT DIFFERENT EXPERIMENTAL CONDITIONS Water Injected \ T e m p , °C P H 2 0 ' a t X ^ 850 880 900 930 V (1) mL/min 1.27 1.19 1.17 1.14 0.15 y ( * » (v) cm^/s 105.69 105.69 105.69 105.69 \° (1) mL/min 2.46 2.39 2.34 2.28 0.30 cm^/s V (v) 211.38 211.38 211.38 211.38 (1) mL/min 4.92 4.78 4.69 4.57 0.60 V (v) cra^/s 422.75 422.75 422.75 422.75 (*) vo lumetr ic vapour flows c a l c u l a t e d . 1 2 5 CARBON, SULPHUR AND SORBENT CONVERSIONS The carbon and sulphur conversions were c a l c u l a t e d by knowing the molar feed rate of carbon and sulphur in the cokes, and the f lowrate and composition of the gaseous product . The conversions from the time steam was in jec ted are designated as carbon i n g a s i f i c a t i o n , denoted as X Q and X 5 f or carbon and sulphur, r e s p e c t i v e l y . The carbon re leased i n the p y r o l y s i s stage i s thus not inc luded i n the convers ion. The amounts of carbon converted during the p y r o l y s i s per iod for runs of var ious operat ing condi t ions i s given in Table F . l i n the Appendix. The sorbent conversion, denoted as Xgg, was estimated from the reduct ion of hydrogen sulphide i n the product gas i n runs w i t h the a d d i t i o n of sorbents . Thus the base case of H2S re lease i n the absence of sorbents was f i r s t e s t a b l i s h e d . The conversion was then double-checked by analyzing the spent sorbents f o r sulphur, using a Leco (R) s o l i d sulphur analyzer . SURFACE AREA Measurement of surface area was conducted using B . E . T . sorpt ion isotherms at 77K with N 2 . The s p e c i f i c areas measured, Sgg^, for fresh c a l c i n e d samples were employed without any c o r r e c t i o n . However, the £>BET for p a r t i a l l y converted coke samples i s considered to be the sum of the surface area of reac t ive p o r t i o n of coke and that of ash in the coke. It i s known that the surface area of the ash i s much smal ler than that of carbon. Since the carbon conversion i n g a s i f i c a t i o n , Xfj, i s evaluated on the bas is of i n i t i a l weight of carbon a v a i l a b l e for the g a s i f i c a t i o n process , the surface area of any l e v e l of conversion, S p , 126 was e v a l u a t e d a l s o based on the i n i t i a l weight o f c a r b o n f o r g a s i f i c a t i o n as fo l lows: SP • [ ( 1 -V + ( I - T ^ > ] S B E T ( 1 4 4 ) where a i s the weight f r a c t i o n of ash based upon i n i t i a l carbon f o r g a s i f i c a t i o n . Th i s method was s u c c e s s f u l l y appl ied by p r i o r authors [3,2,198] for t h e i r reacted coa l char samples. 127 CHAPTER 4 EXPERIMENTAL RESULTS IN GASIFICATION The exper imental work of t h i s r e s e a r c h i n v o l v e s two phases : g a s i f i c a t i o n a n d t h e c o m b i n e d g a s i f i c a t i o n w i t h i n - s i t u d e s u l p h u r i z a t i o n . In the f i r s t phase, the g a s i f i c a t i o n of delayed ( D . C . ) and f l u i d cokes ( F . C . ) were c a r r i e d out i n the absence of c a l c i n e d sorbents , to e s t a b l i s h k i n e t i c data for the g a s i f i c a t i o n process . These data were then u t i l i z e d as the bases for the second phase, whose r e s u l t s are shown i n Chapter 5. 4.1 GASIFICATION OF DELAYED COKE The g a s i f i c a t i o n of D . C . was performed under d i f f e r e n t experimental o p e r a t i n g c o n d i t i o n s to s tudy the e f f e c t s of m i x i n g , r e a c t i o n temperatures and the coke p a r t i c l e s i z e s . The e f f ec t of steam p a r t i a l pressure , P , i n the gas i fy ing gas i s shown i n Sect ion 4 .2 . In t h i s H 2 ° s ec t ion i t i s kept constant at a value of P =30.3 kPa. A l s o , the H 2 ° r e p r o d u c i b i l i t y of the experimental data i s examined at the end of t h i s s e c t i o n . 4 .1.1 EFFECT OF MECHANICAL MIXING The m e c h a n i c a l s t i r r e r rotated w i th in the r e a c t i o n chamber to provide mixing of the s o l i d s . To aid in the s e l e c t i o n of a su i tab l e m i x e r , m i x i n g t e s t s were c a r r i e d out in a p l e x i g l a s s c y l i n d e r of 128 approximately the same d iameter as t h a t of the r e a c t i o n chamber. M a t e r i a l s used i n these tes ts were delayed coke and green p l a s t i c p a r t i c l e s i n roughly the same s i ze (ca . 1.0 mm), making up a bed depth of 5 cm. Parameters examined were types of s t i r r e r and r o t a t i o n a l speed. The r e s u l t s of experimental v i s u a l observat ion are reported i n T a b l e 4 . 1 . 1 . The B type s t i r r e r was s e l e c t e d and used i n the g a s i f i c a t i o n tes ts at a constant speed of ca . 26 rpm. The e f fec t s of mixing on carbon conversion in g a s i f i c a t i o n , Xfj, and on the temperature i n the reac t ion chamber were i n v e s t i g a t e d . The c o n c e n t r a t i o n s of gaseous species produced i n the g a s i f i c a t i o n as a func t ion of time for cases with and without s t i r r i n g are shown i n F igure 4 . 1 . 1 , while the corresponding carbon conversions computed from these concentrat ions and the temperature i n the reac t ion chamber during the experiments are presented in Figure 4 .1 .2 . These p l o t s reveal that s t i r r i n g has l i t t l e inf luence on concentrat ions of the product gases CO, C02> H 2 and CH4, and on the value of X^ at any t ime. The carbon conversion a f t er 5 hours increased to 36.5% i n the run with the s t i r r e r from 34% i n the run without the s t i r r e r , i . e . the increase was ~7%, which i s w i t h i n the bounds of e r r o r of ±8%, reported i n Sect ion 4 . 1 . 5 . The small d i f f erence may be due to the shallow bed employed i n t h i s research . The s t i r r e r , however, improved the s t a b i l i t y and perhaps the un i formi ty of the temperature in the reac t ion chamber around i t s preset value over the course of the t e s t s ; while the temperature f luc tuated cons iderably in the case where the s t i r r e r was o f f . Results of s i m i l a r experiments, i n v e s t i g a t i n g the e f fec t of s t i r r i n g on the r e l e a s e of 129 Table 4 .1 .1 DESCRIPTION OF BED APPEARANCE AND QUALITY OF MIXING .cm I TYPE A STIRRER jjb' _*j Time, min 2 3 4 5 2 3 4 5 RPM 10 No No Lumps/ Channel Lumps/ Channel No Poor Poor Fair 20 Lumps Lumps Good/ Channel Good/ Channel F a i r Fa ir Good Good 30 Good/ Channel Good/ Channel Good/ Channel Good/ Channel Good Good Good Good 40<*> Good/ Channel Good/ Channel Good/ Channel Good/ Channel Good Good Good Good TYPE B STIRRER K -H =3 =1 (*) The shaft of the s t i r r e r vibrates. c o 0. 2 6 2 4 2 2 2 0 I 8 16 14 12 10 8 6 4 2 0 4 2 0 3 A—I 1 1 1 H-, U C 0 2 o - F i x e d b e d a - S t i r r e d b e d P H Q = 0 . 3 a t m , r p m = 2 6 T = 9 0 0 ° C 2 3 4 R e a c t i o n t i m e ( h o u r s ) F i g . 4.1.1 Produced gas concentrations i n cases with ( S t i r r e r B) and without the S t i r r e r . o 131 TESTING THE EFFECTS OF MIXING CARBON CONV. AND TEMP. PROFILE 100 O t—4 t—i in > iz; o o PQ 80 60 40 20 0 * 0 LEGEND • =Rl)N #16T=930C FIXED 0=RUN #20T=930C STIRRED D : Q C < 1 • j • < D t D C < > < D C C C D o o U l CC 3 cc UJ a UJ 60 120 180 240 REACTION TIME.MIN. 300 F i g . 4.1.2 Carbon conversion and temperature of the sample i n runs with ( © ) and without ( D ) the s t i r r e r . D e l a y e d c o k e ; d = 2 mm; p 1 T , = 3 0 . 3 k P a J pc H^O 132 hydrogen sulphide are presented in Chapter 5. The s t i r r e r was therefore re ta ined for the experimental program. 4 .1 .2 EXPERIMENTS AT DIFFERENT REACTION TEMPERATURES 4 .1 .2 .1 EXPERIMENTAL RESULTS To study the reac t ion temperature e f fec t on the rate of carbon c o n v e r s i o n and on the conversion of s o l i d sulphur to H 2 S , the coke p a r t i c l e s i z e , d and the steam p a r t i a l pressure , P were kept c H 2 Q unchanged at 2 mm 2nd 0.3 atm, r e s p e c t i v e l y . The temperatures were set at 800, 850, 880, 900 and 9 3 0 ° C . The operat ing condi t ions f o r t h i s set of experiments are given in Table 4 .1 .2 . Table 4.1.2 E X P E R I M E N T A L CONDITIONS FOR RUNS AT D I F F E R E N T TEMPERATURES WITH DELAYED COKE Run# Temp. °C d , mm G a s i f i c a t i o n P r i a time, min 2 P 18 800 2 300 0.3 22 850 ?. 300 0.3 24 880 2 300 0.3 20 900 2 300 0.3 27 930 2 300 0.3 F i g u r e 4 . 1 . 3 i s a Xfj-t p l o t f o r runs at d i f f e r e n t reac t ion temperatures. The s o l i d curves are l e a s t - s q u a r e f i t t e d p o l y n o m i a l f u n c t i o n s of degree 3. From t h i s p l o t , i t i s observed that the conversion-t ime curves for runs at d i f f e r e n t t e m p e r a t u r e s were not l i n e a r but exh ib i t i n f l e c t i o n points at values of X^ between 15 and 20%. A l s o , de layed coke shows i t s r e l a t i v e l y low r e a c t i v i t y i n steam 133 CARBON CONV. VS. REACTION TIME DURING GASIFICATION STEP 50 o 4 0 b OQ 30 o > o o PQ <i o 20 10 0 0 LEGEND • = RUN #27 T=930C Dp=2mm 0=RI)N #24 T=880C Dp=2mm • =RUN #22 T=850C Dp=2mm • =RUN #20T=900C Dp=2mm V=RUN #18 T=B00C Dp=2mm - J ? — i 60 120 180 240 REACTION TIME,MIN. 300 F i g . 4.1.3 Carbon conversions f or runs of d i f f e r e n t temperatures. Delayed coke; p., _ = 30.3 kPa 134 g a s i f i c a t i o n at 800*0 or l e s s . With the same g a s i f i c a t i o n time of 5 hours , carbon conversions of -43% and 2% were obtained f o r r e a c t i o n temperatures of 930 o C and 8 0 0 ° C , r e s p e c t i v e l y . The carbon conversions at the end of 5 hours of g a s i f i c a t i o n are h i g h l y s e n s i t i v e to temperature over the the range of condi t ions used i n t h i s i n v e s t i g a t i o n . O i l sand coke i s high i n sulphur content . Upon steam g a s i f i c a t i o n the sulphur species i n the coke matrix i s p a r t l y converted i n t o hydrogen sulphide by reac t ing wi th H2 produced i n the g a s i f i c a t i o n r e a c t i o n s . Values of the sulphur convers ions , Xg as a func t ion of time for runs at d i f f e r e n t reac t ion temperatures, are p l o t t e d i n F igure 4 .1 .4 . As with the r e s u l t s of carbon convers ion , sulphur conversions a f t e r 5 hours of g a s i f i c a t i o n increase wi th an increase i n r e a c t i o n temperature. The value of Xg increases from -15% to -25% corresponding to an increase i n temperature from 850 to 9 3 0 ° C . As ment ioned prev ious ly , the X^- t curves were s igmoidal having i n f l e c t i o n p o i n t s around Xrj~0 .18 , wh ich i s the carbon conversion corresponding to the rate maxima, X T J R . F igure 4 .1 .5 presents the rate o f c a r b o n c o n v e r s i o n , d X ^ / d t , v e r s u s c a r b o n c o n v e r s i o n , XQ, for d i f f e r e n t t e m p e r a t u r e s between 850 and 9 3 0 ° C . The c u r v e s were constructed by taking the f i r s t d e r i v a t i v e wi th respect to time of the best f i t t e d polynomial funct ion f o r the X ^ - t data . The low rate at 8 5 0 ° C provides fur ther evidence that the coke i s r e l a t i v e l y unreact ive w i th steam at low temperature. The rates of carbon convers ion wi th dp c=2 mm and P =30.3 kPa, i n m i n - * , are 17.0x10"^ and 5 . 5 x l 0 - ^ for runs at H 2 ° 9 3 0 ° C and 8 5 0 ° C r e s p e c t i v e l y . 135 SULFUR CONV. VS. REACTION TIME EXPERIMENTAL DATA 30 LEGEND 0 60 120 180 240 300 REACTION TIME,MIN. F i g . 4.1.4 Sulphur conversions for runs at d i f f e r e n t temperatures. Delayed coke; p = 30.3 kPa 136 CARBON CONVERSION VS. dX c /dt FROM EXPERIMENTAL DATA 20 16 S 12 o X o 8 i o y LEGEND • =RUN #27 T=930C Dp=2mm 0= RUN #24 T=880C Dp=2mm •= RUN if 22 T=850C Dp=2mm • = RUN #20 T=900C Dp=2mm 0 10 20 30 40 CARBON C0NV. IN GASIFICATIONS 50 F i g . 4.1.5 Rate of carbon conversion for runs at d i f f e r e n t temperatures. Delayed coke; = 3 Q 3 ^ 137 The carbon and sulphur contents i n the D . C . are approximately 83.4 and 6.0% by weight, r e s p e c t i v e l y ; thus the sulphur to carbon molar r a t i o i s about 0.027. I t i s of i n t e r e s t to f i n d out the amount of sulphur g a s i f i e d per mole of base-carbon f o r each set of operat ing c o n d i t i o n s . With t h i s informat ion , the sulphur emission in the gas phase may be p r e d i c t e d . P l o t s which express sulphur convers ion , Xg as a func t ion of carbon conversion at d i f f e r e n t o p e r a t i n g c o n d i t i o n s , such as t h a t presented in Figure 4 .1 .6 were constructed . In t h i s Xg-X^ graph, the curves corresponding to each r e a c t i o n temperature change t h e i r slopes when the g a s i f i c a t i o n begins to increase r a p i d l y . 4 .1 .2 .2 DISCUSSION F i g u r e 4 . 1 . 3 shows t h a t D . C . was n e a r l y unreac t ive i n steam g a s i f i c a t i o n at temperatures of 8 0 0 ° C . This low r e a c t i v i t y of o i l sand coke toward g a s i f i c a t i o n was a l so observed by severa l other authors [119,64,189]. The strong e f f e c t of temperature i s again d i sp layed in F igure 4 . 1 . 7 . In t h i s f i g u r e , the carbon conversions a f ter 5 hours of g a s i f i c a t i o n change r a p i d l y wi th a change in r e a c t i o n temperature. At 8 0 0 ° C the f i n a l conversion was 2.4%; when the r e a c t i o n temperature was r a i s e d by 50°C to 850 o C, the carbon c o n v e r s i o n was s i g n i f i c a n t l y improved to 15.7%. Yet , a smal ler increase in temperature to 8 8 0 ° C from 8 5 0 ° C near ly doubled the value of X < ^ to 30.2%. Further increases in r e a c t i o n temperature s t i l l improved the value of X Q , but at a smaller r a t e . Figure 4 .1 .7 expresses the carbon c o n v e r s i o n s i n 5 -hour of g a s i f i c a t i o n as a funct ion of temperatures, and cons i s t s of two curves; one was obtained by gas ana lys i s and the other by weight changes. The 138 SULFUR CONV. VS. CARBON CONV. EXPERIMENTAL DATA 30 o » — • 04 > o o • J (72 LEGEND • =RUN #27 T=930C Dp=2mm 0=RUN #24 T=880C Dp=2mm •= RUN #22 T=850C Dp=2mm = RUN #20T=900C Dp=2mm 0 10 20 30 40 CARBON CONVERSIONS F i g . 4 . 1 . 6 Sulphur conve r s ion as a f u n c t i o n of carbon convers ion f o r runs at d i f f e r e n t temperatures, Delayed coke; p = 30.3 kPa 139 CARBON CONV. VS. REACTION TEMP. BY GAS ANALYSIS AND WT. CHANGE 60 LEGEND 0=BY GAS ANALYSIS • = BY CHANGE IN WT. O 40 K W > o o o PQ 20 K o 0 700 800 900 1000 REACTION TEMPERATURE, C F i g . 4.1.7 Carbon conversions, at various temperatures, aft e r 5 hours of reaction. Delayed coke; p = 30.3 kPa 140 carbon conversion obtained by weight change i s c o n s i s t e n t l y higher than that obtained by gas a n a l y s i s . The d i f f erence may be due to the c a r r y -over of f i n e s , or re lease of v o l a t i l e matter such as hydrocarbons, N 2 and c h l o r i n e that are not accounted for by the gas ana lysers . The present experimental r e s u l t s suggest that the o i l sand coke has a very low r e a c t i v i t y when compared to c o a l char . For s i m i l a r operat ing c o n d i t i o n s , bituminous coa l reaches 45% carbon conversion i n 60 minutes [38], and thus more severe g a s i f i c a t i o n c o n d i t i o n s , i . e . h i g h e r temperature, longer reac t ion time and smal ler p a r t i c l e s i z e , must be appl ied to achieve higher values of Xfj f o r cokes. The Xfj-t curves f o r D . C . possess i n f l e c t i o n po ints which lead to maxima i n the carbon conversion r a t e , dXfj/dt when p l o t t e d against carbon c o n v e r s i o n as i n F i g u r e 4 . 1 . 5 . T h i s f i g u r e r e v e a l s t h a t t h e g a s i f i c a t i o n s t a r t e d at a r e l a t i v e l y low r a t e , and then increased r a p i d l y presumably as the small pore systems are being developed. It a t t a i n e d a maximum r a t e which depends on t e m p e r a t u r e , at carbon conversions which are i n the range of 0.15 to 0.20 at 880 and 9 3 0 ° C , r e s p e c t i v e l y . The maximum rate at 9 3 0 ° C i s 1 7 . 0 x l 0 _ 4 m i n - * , which i s three times higher than that for 8 5 0 ° C . I t i s seen t h a t , with the trend of the rate curves shown in t h i s dX^/dt vs XQ p l o t , a convers ion of h i g h e r t h a n 80% may be not p o s s i b l e to a c h i e v e u n l e s s h i g h e r temperatures, longer times or smal ler p a r t i c l e s or a combination of the three i s a p p l i e d . It i s widely recognized [39] that the existence of i n f l e c t i o n point (d 2 Xc /d t 2 =0) depends mainly upon the operat ing c o n d i t i o n s , and i s the r e s u l t of a s er i e s of changes i n c h e m i c a l c o m p o s i t i o n , i n porous 1 4 1 s t r u c t u r e of the s o l i d and i n c a t a l y t i c e f f e c t s . A number of i n v e s t i g a t o r s [ 1 7 1 , 6 6 , 2 , 3 , 5 0 , 2 0 1 , 3 2 ] , working mostly with coa l chars , have a l so shown maxima i n dX^/dt vs XQ p l o t s ; and reported values of X Q R i n the range of 0 . 1 - 0 . 4 . Th i s w i l l fur ther be discussed i n Sect ion 4 . 3 . 5 . I n t h e range of o p e r a t i n g c o n d i t i o n s c o v e r e d , the s u l p h u r convers ion i s near ly a l i n e a r funct ion of temperature at any r e a c t i o n time as demonstrated in Figure 4 . 1 . 8 . This unusual behavior of sulphur convers ion i n the D . C . may be due to i t s i n s e n s i t i v i t y to temperature or to the small range of temperature employed i n t h i s present study. F igure 4 . 1 . 6 impl ies that the r e a c t i v i t y of the sulphur i n D . C . toward steam g a s i f i c a t i o n was higher than that of carbon at temperatures o f 8 5 0 ° C or lower. The explanat ion f o r t h i s higher rate of sulphur re lease at low temperature could be, besides the u n r e a c t i v i t y of carbon in the coke at low temperature, that the breaking of the weak bonds between the sulphur atoms on the surface or w i t h i n the aromatic sheets a n d the coke matr ix i s eas ier than the breaking of C - C bonds. Above 8 5 0 ° C bo th the r a t e s of carbon and sulphur conversions increased , presumably as a r e s u l t of the formation of small pore systems i n the coke s o l i d , but the increase i n the carbon conversion was more d r a s t i c than that of sulphur convers ion. This leads to the decrease i n the slope of the curves i n t h i s f i g u r e . Above X Q = 1 0 % , the r a t i o of sulphur c o n v e r s i o n to carbon convers ion, i . e . d X g / d X c i s about 0 . 9 1 , which represents a sulphur to carbon molar r a t i o in the produced gas of - 0 . 0 2 5 f o r the run at 8 5 0 ° C , and 0 . 5 5 f or a sulphur to carbon molar r a t i o in the produced gas of 0 . 0 1 5 for runs with temperatures above 8 5 0 ° C . The 142 O > o (7) SULPHUR CONV. VS. REACTION TEMP. AT VARIOUS REACTION TIME 30 24 18 12 0 LEGEND 0= REACTION TIME=2HRS • = REACTION TIME=3HRS • = REACTION TIME=4HRS • = REACTION TIME=5HRS 800 900 1000 REACTION T E M P E R A T U R E , C F i g . 4.1.8 Sulphur conversion of delayed coke i n runs at d i f f e r e n t temperatures. d p c = 2 m m ' PH o0 0.3 atm 143 formation of a common curve i n the Xg vs XQ p l o t reveals tha t , w i th the exception of only the run at 8 5 0 ° C , the rate r a t i o s dXg/dX^ were the same at a value of about 0.55. Th i s represents a sulphur to carbon molar r a t i o i n the produced gas of 0.015, which i s only h a l f of that of the unreacted D . C . Thus the assumption of sulphur atoms being evenly d i s t r i b u t e d throughout the coke p a r t i c l e becomes quest ionable by the present f i n d i n g s . The dXg/dX^ value for the t e s t c a r r i e d out at 850"C was approximately twice that f o r t e s t s at 8 8 0 ° C or h igher , and the molar r a t i o of sulphur to carbon g a s i f i e d f o r runs at 8 8 0 ° C or higher was only one-hal f of that of the f resh delayed coke. 4 .1 .3 EXPERIMENTS WITH DIFFERENT COKE PARTICLE SIZES 4 .1 .3 .1 EXPERIMENTAL RESULTS In t h i s set of experiments, the inf luences of coke p a r t i c l e s i z e , d p c , on carbon and sulphur conversion and on the rate are examined. To study the sole e f f e c t of d p c , a large range of coke p a r t i c l e s i zes between 0.1 mm and 3 .5 mm was employed; w h i l e o t h e r o p e r a t i n g condi t ions were kept unchanged. The s i ze f r a c t i o n s of s o l i d coke were prepared as i n the procedure descr ibed i n Sect ion 3 .2 .2 . Temperature i n the reac tor was kept at 9 3 0 ° C , the highest temperature that the m a t e r i a l used to f a b r i c a t e the reac tor (Type 316 s t a i n l e s s s t e e l ) could stand, and the p a r t i a l pressure of steam i n the gas i fy ing medium was f i x e d at 0.3 atm and the s t i r r e r speed at 26 rpm. The s i ze f r a c t i o n s of D . C . of average diameters of 0.14, 0.68, 1.00, 2.00 and 3.50 mm were u t i l i z e d . The g a s i f i c a t i o n time for a l l t e s t s was 300 minutes, except Run 86. The 144 operat ing condi t ions of Run 86, with dp c=0.68 mm, were i d e n t i c a l to that of Run 35 except that a longer g a s i f i c a t i o n time of 660 minutes was u s e d . T h i s experiment was designed to t e s t the r e p r o d u c i b i l i t y of experimental data (as w i l l be fur ther shown in Sect ion 4 . 1 . 5 ) , and to examine the rate and carbon conversion over an extended g a s i f i c a t i o n t ime. The operating condi t ions for t h i s set of experiments are provided i n Table 4 .1 .3 . Table 4 .1 .3 E X P E R I M E N T A L CONDITIONS FOR DELAYED COKE RUNS AT  DIFFERENT d p r ; P =30.3 kPa AND T=930''C Run# Range of dPc» G a s i f i c a t i o n p a r t i c l e s i z e , mm time, min mm 92 -0.20+0.075 0.14 300 35 -0.85+0.50 0.68 ,300 86 -0.85+0.50 0.68 660 33 -1.18+0.85 1.0 300 27 -2.36+1.72 2.0 300 30 -4.75+2.36 3.5 300 The experimental r e s u l t s expressing the e f fec t s of coke p a r t i c l e s i ze on carbon conversion are i l l u s t r a t e d i n an Xr^-t p l o t , Figure 4 .1 .9 , in which the s o l i d l i n e s are l e a s t squares f i t t e d p o l y n o m i a l s of degree 3. It i s seen that , at any given time, the carbon conversion increases with decreasing p a r t i c l e s i z e . The carbon conversions a f ter 5 hours of g a s i f i c a t i o n were 28.8% and 69.1% for runs with dp c =3.5 mm and 0.1 mm, r e s p e c t i v e l y . Thus, the extent of conversion was improved s i g n i f i c a n t l y with the use of smal ler coke p a r t i c l e s i z e s . Run 35 and Run 86, performed with the same p a r t i c l e s i ze of 0.68 mm, have good 145 CARBON CONVERSION VS. REACTION TIME  DURING GASIFICATION PROCESS 80 I : : : : : : : : : : 1 0 120 240 360 480 600 REACTION TIME,MIN. F i g . 4.1.9 Carbon conversions for runs of d i f f e r e n t coke p a r t i c l e s i z e s . Delayed coke; p = 30.3 kPa 146 agreement i n the X^- t r e s u l t s up to the end of Run 35, at which the v a l u e of Xrj was 53.4% for Run 35 and 51.8% f o r Run 86. A carbon conversion of 73% was achieved for a g a s i f i c a t i o n time of 660 min. in Run 86. As i n the case of varying reac t ion temperature, i n f l e c t i o n points were observed i n the Xrj-t curves at values of Xrj between 20% to 25% for runs of d i f f e r e n t coke p a r t i c l e s i z e s . The low carbon conversion for the case of dp c =3.5 mm may be r e l a t e d to the low surface area and small p o r o s i t y of the p a r t i a l l y converted coke i n the f i r s t 3 hours of r e a c t i o n , and p a r t l y to the l a r g e r res i s tance to d i f f u s i o n of reac t ing gas w i th in the p a r t i c l e i n t e r i o r . The e f f ec t of p a r t i c l e s i z e on sulphur conversions i s demonstrated i n Figure 4 . 1 . 1 0 . At the same r e a c t i o n t empera ture the s u l p h u r conversion rate increases with decreasing p a r t i c l e s i z e . The values of carbon conversion rates for f i v e d i f f e r e n t p a r t i c l e s i z e s are p l o t t e d in Figure 4.1.11 as a func t ion of X £ . As pred ic ted from the Xc~t c u r v e s , each of the curves reaches a maximum, then decreases g r a d u a l l y . As expected, the smal ler p a r t i c l e s have higher maximum rates of g a s i f i c a t i o n . The values of ( d X c / d t ) j j were l . l x l O - ^ and 2.7xl0~3 m i n - * for d p c of 3.5 mm and 0.14 mm, r e s p e c t i v e l y . The c a r b o n c o n v e r s i o n s at which r e a c t i o n rates a t t a i n maxima, X ^ R , l i e between 15% and 26%, with an average of 20%. Examination of the curves represent ing Run 35 and Run 86 i n Figure 4.1.11 shows that the agreement i n rate between the two i s w i th in 10%. Sulphur conversions for tes t s with var ious coke p a r t i c l e s izes are presented in Figure 4.1.12 as a funct ion of carbon convers ions . In t h i s 147 SULFUR CONV. VS. REACTION TIME EXPERIMENTAL DATA 6^ o C O 04 > o o 04 C O 60 120 180 240 REACTION TIME,MIN. 300 F i g . 4.1.10 Sulphur conversion f or runs of d i f f e r e n t coke p a r t i c l e s i z e s . Delayed coke; p u n = 30.3 kPa rl2>-' 148 CARBON CONVERSION VS. dX c /dt FROM EXPERIMENTAL DATA i o X # o •4 20 40 60 CARBON CONV. IN GASIFICATION,% F i g . 4.1.11 Rate of carbon conversion for runs at d i f f e r e n t coke p a r t i c l e s i z e s . Delayed coke; p = 30.3 kPa 149 SULFUR CONV. VS. CARBON CONV. EXPERIMENTAL DATA 60 O » — t C O P4 W > o fe • J £>. C O LEGEND • =RUN #27 T=930C Dp=2.0mm <D=RUN #30T=930C Dp=3.5mm • =RUN #33 T=930C Dp=1.0mm • =RUN #35T=930C Dp=.68mm V=RUN #92T=930C Dp=.14mm 20 40 60 CARBON C O N V E R S I O N S F i g . 4.1.12 Sulphur conversion as a function of carbon conversion for runs at various p a r t i c l e sizes. Delayed coke; p H 20 30.3 kPa 150 f i g u r e , each of the f i v e Xfj-t curves has a reduct ion i n slope at a carbon conversion of about 10%, implying t h a t the r a t e s of carbon conversion increased more d r a s t i c a l l y than that of sulphur convers ion. I t a l so seen tha t , at a temperature of 9 3 0 ° C , a l l Xg-Xfj curves form a s i n g l e curve having a slope of dxs/dXc~0.58, equivalent to a sulphur to carbon molar r a t i o i n the produced gas of 0.016, regardless of p a r t i c l e s i z e i n the range 0.14-3.5 mm. 4 .1 .3 .2 DISCUSSION Results of experiments t e s t i n g the r e p r o d u c i b i l i t y of the present experimental data are shown i n t h i s s e c t i o n , and again i n Sect ion 4 .1 .5 . In Figure 4 . 1 . 9 , Run 86 i s the repeat of Run 35 but was kept f o r a longer g a s i f i c a t i o n time. The r e s u l t s obtained are c o n s i s t e n t l y c lo se , and are w i t h i n ±3% with respect to carbon conversion and ±10% with respect to the rate f o r the f i v e hours. Th i s good agreement confirms the v a l i d i t y of the experimental r e s u l t s of the present study. From these two t e s t s , i t was a l so found that the rate of carbon conversion was q u i t e low at l o n g e r r e a c t i o n t i m e s . To improve the carbon convers ion from 53.3% ( i . e . end of Run 35) to 73.3% ( i . e . end of Run 86) takes more than twice the time required to achieve the f i r s t 53.3%. The e f f e c t of c o k e - p a r t i c l e s i ze on carbon conversion and on the conversion rate were examined. Carbon conversions at the end of 5 hours of g a s i f i c a t i o n , computed by gas ana lys i s and by weight changes, are p l o t t e d i n Figure 4.1.13 as a funct ion of r e c i p r o c a l of diameter. The value der ived from the weight change method i s , on the average, 6% higher than that ca l cu la ted by the gas ana lys i s method. Fines c a r r i e d 151 CARBON CONV. VS. COKE DIAMETER BY GAS ANALYSIS AND WT. CHANGE 60 LEGEND 0= BY GAS ANALYSIS • = BY CHANGE IN WT. 8 RECIPROCAL OF DIAMETER, m m - l 10 F i g . 4.1.13 Carbon conversions a f t e r 5 hours of g a s i f i c a t i o n , with various coke p a r t i c l e s i z e s , determined by gas analysis and by weight change. Delayed coke; T = 930 C; p u = 30.3 kPa H 2 ° 152 over, and v o l a t i l e s p e c i e s r e l e a s e d but not d e t e c t e d by the gas chromatograph are bel ieved to account for the d i screpancy . The curves i n t h i s p l o t have very high slopes at small values of l / d p c , i n d i c a t i n g a strong inverse e f fec t of Xrj for runs employing r e l a t i v e l y large coke s i z e s . The slope of t h i s Xc vs l / d p c curves l e v e l o f f at a value of l / d p c l a r g e r than -2 mm -*, corresponding to a dp c <0.5 mm, implying that the p a r t i c l e s i z e e f f ec t on carbon conversion becomes l e s s s i g n i f i c a n t for s u f f i c i e n t l y small coke s i z e . Decreasing coke s ize increases both carbon and sulphur convers ions . The higher hydrogen sulphide emissions which accompany more complete carbon conversion adversely a f fec t the o v e r a l l g a s i f i c a t i o n process with respect to environmental p o l l u t i o n , and require be t ter q u a l i t y as wel l as higher throughput of sorbent to remove the ex tra amount of sulphur produced. The experimental r e s u l t s of t h i s present study agree w e l l with f ind ings obtained under compatible operat ing cond i t ions p u b l i s h e d by o t h e r a u t h o r s i n the l i t e r a t u r e . Furimsky [64] reported a carbon convers ion of 11.2% for steam g a s i f i c a t i o n at delayed o i l sand coke of dpc=0.37mm at 9 3 0 ° C for one hour. This value f a l l s between 9.26% and 14.29% determined in t h i s research for p a r t i c l e s i zes of 0.68 mm and 0.14 mm, r e s p e c t i v e l y , at i d e n t i c a l c o n d i t i o n s . S i m i l a r l y , good agreement was a l so observed in the comparison of sulphur conversions to that of publ ished va lues . F i g u r e 4 . 1 . 1 1 provides fur ther evidence of the conversion rate a t t a i n i n g a maximum with extent of g a s i f i c a t i o n and of the e f fec t s of p a r t i c l e s i zes on t h i s ra te . The formation of small pores and the 153 increase of s p e c i f i c area w i l l be shown to account for the maximum i n rate of carbon convers ion. The average value of XfjR, where the maximum rate occurs , i s about 20%, which i s s l i g h t l y higher than that determined f o r the case examining the e f fec t of temperatures (Sect ion 4 . 1 . 2 . 3 ) . It i s observed that there are no i n t e r s e c t i o n s between the dXfj/dt vs XQ curves , and the curves for smal ler p a r t i c l e s i zes are located above the ones f o r l a r g e r coke s i z e s , d u r i n g the e n t i r e range of carbon convers ion . T h i s warrants a h igh conversion rate for a smal ler d p c . A l s o i n t h i s f i g u r e , the c u r v e s r e p r e s e n t i n g the h i g h e r carbon conversion ra te s , i . e . smal ler p a r t i c l e s i z e s , have l e s s e r curvature; the curves may eventual ly become s t r a i g h t l i n e s for s u f f i c i e n t l y small coke p a r t i c l e s i z e s . Th i s reduct ion in the curvature of the dXfj/dt curves demonstrates t h a t the r a t e of carbon c o n v e r s i o n f o r runs employing large p a r t i c l e s i ze i s more sharply reduced a f t e r passing i t s maximum value; on the other hand, the rate for runs with small p a r t i c l e s i z e i s re ta ined at that high rate for a wider range of f r a c t i o n a l convers ion . In the Xg vs Xfj p l o t , F i g u r e 4 .1 .12, the curves represent ing experiments at f i v e d i f f e r e n t p a r t i c l e s i zes co inc ide to form a s ing le curve . At an Xfj value of about 10%, the slope of t h i s combined curve i s reduced from about 0.70 to a dXg/dXfj of 0.58, equivalent to a sulphur to carbon molar r a t i o i n the produced gas of 0.016. A s i m i l a r value of 0 .015 was de termined f o r runs examining the e f f e c t of r e a c t i o n temperature ( sec t ion 4 . 1 . 2 . 4 ) . The r e d u c t i o n i n the s l o p e of the combined curve to dXg/dXfj=0.58 may be re la ted to the formation of small pore systems i n the coke matr ix . The S . E . M . photos showing the porous 154 s tructure of the p a r t i a l l y r e a c t e d coke samples are p r e s e n t e d i n sec t ion 4 . 1 . 4 . Thus, above Xc=10%, approximately 0.0156 mole sulphur in the D . C . was g a s i f i e d per mole of carbon. This r a t i o i s lower than the S/C molar r a t i o i n the f resh D . C , of 0 .0270. Consequently, above 10% carbon convers ion, the sulphur c o n t e n t of the r e m a i n i n g p a r t i a l l y r e a c t e d coke may increase as the carbon conversion becomes h igher . Several authors [74,135] have noted t h i s tendency remaining in the coke matrix of the sulphur i n the D . C . toward i t s removal from the coke p a r t i c l e s . They postulated that the p r o g r e s s i v e i n c o r p o r a t i o n of s u l p h u r atoms i n t o s t a b l e polyaromatic thiophene r ings may i n h i b i t sulphur removal from the s o l i d a f t e r a prolonged r e a c t i o n time or at high temperature. 4 . 1 . 4 ANALYSIS OF UNREACTED & PARTIALLY REACTED COKE The r e a c t i v i t y of carbonaceous s o l i d s and the maximum i n the rate of conversion are presumably re la t ed to changes i n surface area , i n pore s ize d i s t r i b u t i o n s and in pore volume. To examine the r e l a t i o n between these changes and the respect ive k i n e t i c r e s u l t s , representat ive samples of u n r e a c t e d and p a r t i a l l y reacted cokes were subjected to a s er i e s of analyses , inc lud ing B . E . T . s u r f a c e a r e a measurement, s c a n n i n g e l e c t r o n m i c r o s c o p y , mercury porosimetry and elemental a n a l y s i s . 155 4 .1 .4 .1 B . E . T . SURFACE AREA MEASUREMENT The surface area of coke samples was determined by the B . E . T . i s o t h e r m s method, d e s c r i b e d i n Sect ion 2.6.2 and according to the procedures given i n Sect ion 3 . 4 . 3 . 1 . Since o i l sand coke i s a mixture of carbon and other impur i t i e s and s ince the coke samples were g a s i f i e d to d i f f e r e n t l e v e l s of convers ion, which was based on the i n i t i a l carbon for g a s i f i c a t i o n , the measured B . E . T . surface areas were thus converted to per u n i t mass of i n i t i a l carbon, as d e t a i l e d i n Sect ion 3 . 4 . 4 . 4 . The surface areas of f i f t e e n samples of reacted coke were measured. These samples were the bed residues of experiments under 15 d i f f e r e n t experimental condi t ions such as r e a c t i o n t ime, r e a c t i o n temperature and p a r t i c l e s i z e . Table 4 .1 .4 s p e c i f i e s the condi t ions from which the p a r t i a l l y reacted coke sample were der ived; i t a lso includes a value g e n e r a t e d by F u r i m s k y [64] and the c o r r e s p o n d i n g e x p e r i m e n t a l c o n d i t i o n s . Figure 4.1.14 i l l u s t r a t e s the v a r i a t i o n of s p e c i f i c areas of the D . C . with f r a c t i o n a l conversion of carbon. Thi s f igure reveals that , at the beginning of the g a s i f i c a t i o n , the s p e c i f i c area of the coke sample i s r e l a t i v e l y low; i t increases sharply as the value of increases i n the range of 10% to 20%, a t ta ins a maximum of a value of Xfjs ° f about 27%, and gradual ly decreases as the r e a c t i o n proceeds to higher carbon conversions . The coke s tar ted at an i n i t i a l surface area of approximately 8 m 2 / g . This va lue , however, increases f i v e - f o l d to a surface area maximum of 45 m 2 / g ; before dropping to a va lue , even lower than i t s i n i t i a l s p e c i f i c surface area, of 6.77 m 2 / g at X Q = 7 3 . 3 % . The 156 Table 4.1.4 SURFACE AREA BASED ON INITIAL CARBON FOR GASIFICATION Al DIFFERENT CONVERSION LEVELS Run* Reaction d p c » Reaction Carbon conversion S p e c i f i c surface Temp., mm Time, at end of g a s i f . area (*), ra^/g °C min p e r i o d , % Xp 84&85 930 0.68 0 0 8.17 81 930 0.68 45 6.62 24.40 79 930 0.68 90 15.70 40.16 80 930 0.68 160 31.01 45.20 83 930 0.68 240 45.02 31.28 35 930 0.68 300 53.42 28.57 86 930 0.68 660 73.33 6.77 22 850 2.00 300 13.92 32.61 24 880 2.00 300 29.09 47.96 20 900 2.00 300 36.70 45.09 27 930 2.00 300 41.72 36.99 33 930 1.00 300 47.30 30.43 90 880 0.14 300 41.79 40.86 91 900 0.14 300 53.28 26.64 92 930 0.14 300 69.07 13.59 [51] 930 0.37 60 5.94 31.25 (*) based on i n i t i a l weight of carbon at beginning of g a s i f i c a t i o n . one value of the s p e c i f i c area of D . C . reported i n the l i t e r a t u r e [64] i s a l so p l o t t e d in t h i s f igure for reference . An important observat ion derived from Figure 4.1.14 i s that a l l data points represent ing the surface area of a l l D . C . samples f a l l along a common curve . I t may be concluded that regardless of the operating condi t ions such as temperature, r eac t ion time or coke p a r t i c l e s i z e , the s p e c i f i c s u r f a c e a r e a depends s o l e l y upon the e x t e n t of carbon convers ion , Xrj based on the i n i t i a l carbon for g a s i f i c a t i o n . 157 CARBON CONV. VS. SURFACE AREA EXPERIMENTAL MEASUREMENT 50 < o O U Jol) CM 40 30 <d W 20 w o (75 10 ( 0 O. LEGEND 0= EXPERIMENTAL MEASUREMENT = DATA FROM FURIMSKY O 20 40 60 CARBON CONV. IN GASIF.,% 80 F i g . 4.1.14 Surface area as a function of carbon conversion. Delayed coke; p u _ = 30.3 kPa 1 5 8 4 . 1 . 4 . 2 EXAMINATIONS USING SCANNING ELECTRON MICROSCOPE The porous s t ruc ture of samples of D . C . at var ious l e v e l s of carbon c o n v e r s i o n were examined i n a S . E . M . at i n t e r m e d i a t e to h i g h magn i f i ca t ion ; and photos of t y p i c a l areas were taken. The specimens employed i n the S . E . M . study cover a wide range of conversion between 0 and 7 3 % , namely 0 , 7 % , 1 8 % , 3 1 % , 5 7 % and 7 3 % . Most of the S . E . M . photos o f t y p i c a l a r e a s i n t h e s e s p e c i m e n s were t a k e n at the same m a g n i f i c a t i o n . Magni f i cat ions of up to X 4 0 0 0 were required where f ine d e t a i l s needed to be photographed. These photos are shown i n Figure 4 . 1 . 1 5 a to g. The v i s u a l observat ion of the unreacted coke at magni f icat ions X 7 0 and X 4 0 0 0 reveals that i t has l i t t l e or no p o r o s i t y . However, systems of pores were seen on samples with carbon conversions of ~ 7 % and h igher . At a carbon conversion of 1 8 % , small pore systems were found to appear on the e n t i r e surface , occas iona l ly with l a r g e r cracks . The pores were g r a d u a l l y enlarged with the extent of g a s i f i c a t i o n from an average of 7 pm across i n a sample with 3 1 % conversion to approximately 1 4 urn in sample of 5 7 % convers ion , and eventual ly to wide open as i n the case for sample at X Q - 7 3 % . Many pieces with much smal ler s i z e s and fragments were found in the specimens of coke at conversion - 7 3 % . I t i s evident that when a large f r a c t i o n of carbon in the coke was converted, the p a r t i c l e d i s i n t e g r a t e d into fragments of smal ler s i z e s . 4 . 1 . 4 . 3 POROSIMETRY M e r c u r y p o r o s i m e t r y was employed to de termine the pore s ize d i s t r i b u t i o n and the pore volume of many samples of p a r t i a l l y reacted F i g . 4.1.15 c P 1 g - 4.1.15 d S.E.M. photos of delayed coke at va r i o u s carbon conversions. o S U N C O R 0 . 5 ? C £6KV £ . £ 8 K 1 3 . 6 u m 0 0 1 4 6 8 £0KV X250 F i g . 4.1.15 e F i g . 4.1.15 f S.E.M. photos of delayed coke at various carbon conversions, F i g . 4.1.15 g S.E.M. photos of delayed coke at v a r i o u s carbon conversions. i—1 163 coke. The p r i n c i p l e of t h i s a n a l y t i c a l technique was dea l t with i n Sect ion 2 .6 .1 . F ive delayed coke samples whose carbon conversions range from 0 to approximately 73% were examined. The experimental r e s u l t s of the porosiraetry measurements are shown in Figure 4 .1 .16 . This i s a t y p i c a l p l o t r e l a t i n g the volume of mercury penetrated to the equivalent pore diameter or the appl ied pressure . From t h i s p l o t , the t o t a l volume as we l l as the volume occupied by a pore s i ze range can be determined. Volumes of mercury penetrated, in c c / g , were genera l ly higher for samples with higher carbon convers ion. Volumes of about 0.09 and 0.39 c c / g were r e c o r d e d f o r u n r e a c t e d coke and for the p a r t i a l l y converted coke having a carbon conversion of 73%, r e s p e c t i v e l y . Figure 4 . 1 . 1 7 i l l u s t r a t e s the pore volume based on the i n i t i a l carbon for g a s i f i c a t i o n as a funct ion of XQ. I t appears from t h i s p l o t that the s p e c i f i c pore volume i s highest (-0.25 c c / g ) , at a carbon conversion XC~35%. 4 .1 .4 .4 ELEMENTAL ANALYSIS OF REACTED COKE SAMPLES As prev ious ly mentioned i n the l i t e r a t u r e r e v i e w , d u r i n g the t h e r m a l c r a c k i n g of the bitumen most of the impur i t i e s and metals reported to the r e s i d u a l coke. O i l sand cokes thus contain unusual ly high contents of heavy metals , e s p e c i a l l y n i c k e l and vanadium, as shown in Table 2 .3 . In the pure form, these metals have high commercial va lues . In an attempt to c o r r e l a t e the metal contents of the p a r t i a l l y 0 0 Equiva lent pore diameter [ fJLm) 10 |.o 0.10 0 ,0 _ 0.4 "en N. o ^ 0,3 c o 2 0.2 OJ c °- 0.1 0 o 0 0 0 0 0 0 0 0 0 F i g . 4 .1 .16 . Absolu te pressure ( ps i ) P o r o s i t y de te rmina t ion : Equ iva l en t pore diameter as a f u n c t i o n o f . v o l penet ra ted . Q: X c=73%; P: Xc=57%; 0: X =31%; N: X =18%; M: X =0% ume ON -c-165 CARBON CONV. VS. PORE VOLUME O o H O Q W < u PQ •W o > H o OH EXPERIMENTAL MEASUREMENT 0.3 0.0 20 40 60 CARBON C O N V E R S I O N S F i g . 4.1.17 Pore volume as a function of carbon conversion. Delayed coke; pTJ _ = 30.3 kPa H 2 ° 166 reacted coke with the extent of carbon g a s i f i e d , four samples w i t h carbon conversions of 0, 42%, 56% and 73% were analyzed for elemental components. The average p a r t i c l e s ize for these samples was 0.68 mm, except the X =^42% sample which had a dp c=2 mm. The concentrat ions of V and N i of the four samples are reported i n Table 4 . 1 . 5 . Table 4 .1 .5 V AND N i CONCENTRATIONS IN PARTIALLY REACTED COKE X Q I % V, ppm N i , ppm mgV mgNi kgC kgC 0 1010 455 1010 455 42 1670 748 2879 1290 56 2800 1270 6364 4023 74 2030 2580 7808 9923 It i s seen that the metal concentrat ions increase with increas ing carbon conversion i n g a s i f i c a t i o n . 4 .1 .4 .5 DISCUSSIONS S p e c i f i c area, poros i ty and p h y s i c a l s t r u c t u r e are very important for gas-porous s o l i d reac t ions . From Table 4 .1 .4 and Figure 4 .1 .14, i t i s observed that , regardless of operat ing cond i t i ons , a l l values of the s p e c i f i c surface areas of the p a r t i a l l y converted coke samples form a s ing le curve, which a t ta ins a maximum at an intermediate carbon conversion of about 27%. The sole dependence of s p e c i f i c area based on the i n i t i a l carbon f o r g a s i f i c a t i o n was reported by Wu et a l . [198]. Furthermore, A d s h i r i [2] suggested that s p e c i f i c surface area can be expressed by a quadrat ic funct ion of convers ion . The s o l i d curve represents the present experimental values 167 of s p e c i f i c area vs carbon convers ion, constructed in Figure 4.1.14, having a parabo l i c shape s i m i l a r to that described by these previous workers. The maximum in surface area was e x p e r i m e n t a l l y observed [38,2,86,171,104] and t h e o r e t i c a l l y postulated [17,10,66,22] by a number of authors . The extent of the g a s i f i c a t i o n where the maximum surface area occurred, Xfjg f or past s tudies i s given in Table 4.1.6. Table 4.1.6: Average value of Xps f or past s tudies on coa l char Bhatla Adshiri Su & K u e t a l . Blacks Hashimoto Battel Nlcke Kasaoka This Study Perlmutter et a l . [17,22] [2,3] [171] [198] [23] [86] [10] [196] [104] (D.C.) Xcs 0.04 0.25-0.4 0.3 0.46 0.37 0.2-0.40 0-0.40 0.37 0.30 0.27 The ac tua l porous s tructure of a few coke samples has been examined us ing S . E . M . The photos in Figure 4.1.15 a-g revea l that l i t t l e or no p o r o s i t y i s i n i t i a l l y evident i n the D . C , even at h igh magni f i ca t ion . T h i s o b s e r v a t i o n i s r e i n f o r c e d by the mercury porosimetry r e s u l t s , F igure 4 .1 .16 , which show a very low pore volume of 0 .09 c c /g for the Xfj=0% sample. This present f i n d i n g i s fur ther supported by data of previous work [94]. It i s be l ieved that the maximum carbon conversion r a t e , dXfj /dt at a Xfj approximately, 20% was caused by the combined e f f ec t s of h igh r e a c t i v i t y due to newly developed small pores at Xfj~18% and the increas ing newly formed surface area p r i o r to reaching to i t s maximum at X<j~27%. This present suggestion i s based on the combined experimental r e s u l t s of the S . E . M . examination (Figure 4 . 1 . 1 5 . c ) , the s p e c i f i c area measurements (Figure 4 .1 .14) , and the mercury porosimetry determination (Figure 4.1.16) which indicate tha t pore s i z e i n the 168 region of 0.03 to 0.003 um were predominant in the Xc=16% sample. A fur ther increase in carbon conversion from the Xc~20%, where the rate a t ta ins a maximum, may s t a r t an acce l era t ing enlargement in pore s izes because t h e . r e a c t i o n rate was at i t s highest l e v e l . Th i s i s fol lowed by an immediate decrease i n r e a c t i o n rate in sp i t e of a small increase in surface area to i t s maximum va lue . The surface area at any time may be the net r e s u l t of two competing processes: the growth of the r e a c t i o n surface area associated with the pores and the loss of these e x i s t i n g surfaces as they progres s ive ly enlarge or co l lapse by i n t e r s e c t i o n . The formation of the new surface area i s thought to be due to the crea t ion of new small pore systems. The maximum i n surface area at X^~27% is presumably a r e s u l t of the dominant e f f e c t of the l a t t e r process i n the e a r l y stage of g a s i f i c a t i o n [172,10,170,171]. The ro l e of the surface area upon the rate of carbon conversion was inves t igated extens ive ly in many previous s tudies [171,86,2,3,49,38,58, 198]. However, some disagreement ex i s t s among the r e s u l t s reported by these i n v e s t i g a t o r s . Some claimed that reac t ion occurs at ac t ive s i t e s character ized by surface area , making the rate dependent on the surface area . As a r e s u l t , both d X c / d t and Sp go through maxima at the same value of carbon conversion or the r e l a t i o n s h i p between dX^/dt and S i s l i n e a r [171,49,58,198]. By c o n t r a s t , Dutta & Wen [38] and Fot t et a l . [58] postulated that the rate of steam-carbon g a s i f i c a t i o n has l i t t l e r e l a t i o n with the surface area but s trongly re la t e s to poros i ty of the s o l i d . 169 The S . E . M . observat ion of coke p a r t i c l e s at a conversion of 31% revealed that the pore system was formed by a mixture of small and large pores of near ly even p r o p o r t i o n s . N e v e r t h e l e s s , when the carbon conversion improved to 57%, i t i s evident that most of the pores were r e l a t i v e l y l a r g e compared to those of the Xfj~31% sample . The enlargement of the pores may be the r e s u l t of pore i n t e r s e c t i o n s and over lapping which are i n turn caused by the consumption by reac t ion of the carbon w a l l s , which prev ious ly separated these pores [ 5 8 ] . At Xfj = 57%, the p a r t i c l e s a l s o became s m a l l e r , but not s i g n i f i c a n t l y smal l er . F i n a l l y , more smal ler pieces and fragments were found i n the specimen of Xfj=73%. The fragments and small pieces were be l ieved to come from the d i s i n t e g r a t i o n of the p a r t i c l e , presumably occurr ing when the outer-most carbon wal l s separat ing the pores and the outer surface of the p a r t i c l e were consumed by g a s i f i c a t i o n r e a c t i o n s . The S . E . M . photo i n Figure 4 .1 .15 .g provides v i s u a l evidence of t h i s dynamic change in porous s t r u c t u r e . Working i n CO2 and 0 2 g a s i f i c a t i o n of coa l and char, several authors [66,49,50] have found that the char p a r t i c l e s d i s i n t e g r a t e into fragments at Xfj; about 0 . 8 . Furthermore, at t h i s stage of carbon conversion the poros i ty of the s o l i d i s r e l a t i v e l y h igh which makes many a n a l y t i c a l g a s - s o l i d models such as the CPM become i n a p p l i c a b l e [66]. A scanning e l ec t ron micrograph at magni f i cat ion X2200 of the Xfj=57% specimen, F igure 4 . 1 .15 . f , c l e a r l y shows the o r i e n t e d c o n c e n t r i c a l alignment of the f l aky carbon layers along the pore surfaces . Th i s i s i n complete agreement with the r e s u l t s reported by previous workers 170 [58,102], who found that carbon layers in the v i c i n i t y of pores were a l igned c o n c e n t r i c a l l y around pores. Mercury penetrat ion i s a simple technique to approximate the pore s i ze d i s t r i b u t i o n of s o l i d s . The r e s u l t s of poros i ty determinat ion of coke samples at var ious l e v e l s of conversions , Figure 4 .1 .16 , show that D . C . p r a c t i c a l l y has very l i t t l e poros i ty ; and that the X^~16% sample possesses e s s e n t i a l l y small pores , between 0.003 to 0.03 pm, though having small p o r o s i t y . Pore s i zes are increased to the range of 0.09-100 pm for increas ing carbon conversion as seen i n curves 0, P and Q. I t i s noted that a l l pores i n the region of 0.0025 pm to 100 pm were determined, corresponding to appl ied pressures between 13.6 kPa and 4 .1*10 5 kPa. It i s seen from Table 4 .1 .5 that V and Ni concentrat ions genera l ly increase wi th increas ing carbon convers ion. The r e s u l t s from mass balances on V and N i f o r a l l these samples imply that a major part of the o r i g i n a l V and N i i n the D . C . reported i n the remaining unreacted coke f r a c t i o n , as the carbon i n the coke was g a s i f i e d away. At least 50% of the o r i g i n a l V and almost a l l N i remained i n the coke. These a n a l y s e s were not done i n d u p l i c a t e , and the r e s u l t s s h o u l d be considered p r e l i m i n a r y . Further s tudies are required to confirm t h i s observat ion . 4 .1 .5 REPRODUCIBILITY OF EXPERIMENTAL DATA To t e s t the c r e d i b i l i t y of the experimental r e s u l t s obtained in t h i s present study, many tes t s were performed in d u p l i c a t e , s i m i l a r to the ones reported in Sect ion 4 .1 .3 . In t h i s sec t ion the v a r i a t i o n in 171 experimental d a t a c o l l e c t e d i n r e p e a t e d runs f o r a g i v e n set of operat ing condi t ions are examined. The experimental r e s u l t s of Runs 28, 29, 32, the repeats of Run 27, prev ious ly shown in Sect ion 4 . 1 . 2 . 1 , are presented here. Gas samples, taken at i n t e r v a l s of 10 min i n the f i r s t hour and 20 min for the remaining time of the experiments, were analysed f o r H 2 , CO, C O 2 , C H 4 and N 2 . C H 4 was produced only i n the f i r s t 40 minutes of the experiment , and the concentrat ion of N 2 was the balance i n the gas mixture . Therefore , only H 2 , CO and C O 2 concentrat ion trends i n the leav ing gas are presented in Figure 4 .1 .18. I t i s seen from t h i s f igure that the concentrat ions of the gases for the four repeated runs are in good agreement and w i t h i n ±8%. S i m i l a r r e p r o d u c i b i l i t y of the experimental r e s u l t s i s observed i n the X(;-t p l o t , F igure 4 .1 .19 , for these four runs. 4.2 GASIFICATION OF FLUID COKE F l u i d coke, F . C . , produced by Syncrude L t d . was steam g a s i f i e d at d i f f e r e n t operat ing condi t ions such as temperature, and steam p a r t i a l pressure i n the gas i fy ing medium. F l u i d coke possesses a narrow p a r t i c l e s i ze range between 40 um and 300 um with p a r t i c l e s mainly in the s ize range of 75 to 200 um. The coke was screened to e l iminate f ines and to obta in p a r t i c l e s in t h i s t y p i c a l s i z e range. The experiments were c a r r i e d out in a f ixed bed manner and according to the procedure d e t a i l e d in Sect ions 3 .3 .2 .3 and 3 .4 .2 , r e s p e c t i v e l y . ZLl 173 CARBON CONV. VS. REACTION TIME DURING GASIFICATION STEP 50 REACTION TIME,MIN. F i g . 4.1.19 Carbon conversions for four repeated runs. Delayed coke; p = 30.3 kPa 174 4 .2 .1 EXPERIMENTS WITH DIFFERENT REACTION TEMPERATURES 4 .2 .1 .1 EXPERIMENTAL RESULTS In t h i s set of experiments, r eac t ion temperature was v a r i e d between 8 8 0 ° C and 9 3 0 ° C , namely at 8 8 0 ° C , 9 0 0 ° C and 9 3 0 ° C ; while steam p a r t i a l pressure was kept at a constant l e v e l at 30.3 kPa. The e f f ec t s of temperature on carbon convers ion, on sulphur conversion and on the rate of carbon conversion are discussed here. The X ^ - t curves for runs at d i f f e r e n t temperatures with F . C . are shown i n Figure 4 . 2 . 1 . It i s observed that the Xfj-t curves are not l i n e a r and have i n f l e c t i o n points at XQ values ranging from about 12% at temperature of 8 8 0 ° C to 19% at 9 0 0 ° C , and 21% at 9 3 0 ° C . As i n the case of D . C , the c a r b o n c o n v e r s i o n s of F . C increase wi th increas ing temperatures. At temperature of 8 8 0 ° C , the carbon conversion a f t er 5 hours of g a s i f i c a t i o n was 27.4%. The value of XQ improved to 43.1% for a r e a c t i o n temperature of 9 0 0 ° C and to 54.8% at 9 3 0 ° C I t has been es tab l i shed that F . C . has higher sulphur content than delayed coke. C o n s e q u e n t l y , when each was steam g a s i f i e d at an i d e n t i c a l operat ing c o n d i t i o n , the emissions of hydrogen sulphide in the produced gas were higher for F . C . than for D . C . The sulphur conversions f o r runs at d i f f e r e n t temperatures are p lo t t ed against r e a c t i o n time in F i g u r e 4 . 2 . 2 . T h i s f i g u r e shows that the sulphur convers ion, Xg, increases wi th increas ing temperature. A f t e r 5 hours of g a s i f i c a t i o n , the sulphur conversion in the F . C . was 33.4%, 43.1% and 51.4% for tes t s c a r r i e d out at temperatures of 8 8 0 ° C , 900°C and 9 3 0 ° C , r e s p e c t i v e l y . 175 CARBON CONV. VS. REACTION TIME DURING GASIFICATION STEP 80 O lc CO < 60 £ ; 40 > o o m K <3 O 20 0 LEGEND • =RUN #87 SYNCR. T=930C + =RUN #88 SYNCR. T=900C V=RUN #89 SYNCR. T=880C 0 60 120 180 240 REACTION TIME,MIN. 300 F i g . 4.2.1 Carbon conversion for runs at d i f f e r e n t temperatures. F l u i d coke; p = 30.3 kPa 176 SULFUR CONV. VS. REACTION TIME EXPERIMENTAL DATA 60 40 20 0 LEGEND _=RUN #87 SYNCR. T=930C V=RUN #88 SYNCR. T=900C + =RUN #89 SYNCR. T=880C 0 60 120 180 REACTION TIME,MIN. 240 300 F i g . 4.2.2 Sulphur conversions of runs at various temperatures. F l u i d coke; p = 30.3 kPa 177 These values are margina l ly higher than those obtained for D . C . under the same operating cond i t ions . The rates of carbon conversion obtained by taking the slopes of the X(j-t curves for runs at various temperatures i n Figure 4.2.1 are p l o t t e d against carbon conversion as shown i n Figure 4 .2 .3 . The X^- t curves i n Figure 4.2.1 are s igmoidal having i n f l e c t i o n points between 10% to 16% of carbon convers ions , which lead to maxima in the rate curves , dXfj/dt vs Xfj. I t i s evident that at 8 8 0 ° C the F . C . i s r e l a t i v e l y unreact ive and the r e a c t i o n rate i s favored by higher temperature. The value of the conversion rate i s near ly doubled when the temperature increased to 9 3 0 ° C from 8 8 0 ° C . The values of d X c / d t for runs at 8 8 0 ° C , 9 0 0 ° C and 9 3 0 ° C r e s p e c t i v e l y a r e 1 . 0 4 x l 0 ~ 3 m i n - 1 , 1 . 6 1 x l 0 ~ 3 m i n - 1 and 2 .10xl0~ 3 m i n - 1 . The u l t imate ana lys i s of the f l u i d coke employed i n t h i s study shows a carbon content of 79.49 weight percent and a sulphur content of 6.96 weight %; in other words, the sulphur to carbon molar r a t i o i s about 0.033. To compare how the s o l i d sulphur i n the F . C . g a s i f i e d with r e s p e c t to the s o l i d c a r b o n , the su lphur conversions for runs at d i f f e r e n t temperatures are expressed as a funct ion of carbon convers ion , and are shown in Figure 4 .2 .4 . In t h i s f i g u r e , the dXg/dX^ r a t i o s are s l i g h t l y higher for runs with low temperature. For the 8 8 0 ° C run, t h i s v a l u e i s 1.15, represent ing a sulphur to carbon molar r a t i o i n the produced gase of 0 .038; f o r the 9 0 0 ° C r u n , t h i s v a l u e i s 1.0 corresponding to a sulphur to carbon molar r a t i o i n the gas phase of 0.033; and for the 9 3 0 ° C run, the dX s /dX c =0.92 g iv ing a S /C molar r a t i o 178 CARBON CONVERSION VS. dX c /dt FROM EXPERIMENTAL DATA 30 LEGEND • =RUN #87 SYNCRUDE T=930C 0= RUN #88 SYNCRUDE T=900C • =RUN #89 SYNCRUDE T=880C 20 40 60 CARBON CONV. IN GASIFICATIONS F i g . 4.2.3 Rates of carbon conversion for runs at d i f f e r e n t temperatures. F l u i d coke: P = 30.3 kPa 179 SULPHUR CONV. VS. CARBON CONV. EXPERIMENTAL DATA 60 LEGEND • =RUN #87 T=930C SYNCRUDE 0=RUN #88 T=900C SYNCRUDE •=RUN #89 T=880C SYNCRUDE 40 20 0 0 20 40 CARBON C O N V E R S I O N S 60 F i g . 4.2.4 Sulphur conversion as a function of carbon conversion. F l u i d coke; p 0 = 30.3 kPa 180 i n the gas phase of 0 .030. These sulphur to carbon molar r a t i o s are very s i m i l a r to that of the u l t imate ana lys i s of the o r i g i n a l F . C . 4 .2 .1 .2 S . E . M . EXAMINATION OF F . C . Scanning e l e c t r o n microscopy was employed to examine the s t r u c t u r a l c h a r a c t e r i s t i c s of F . C . and i t s p a r t i a l l y converted coke samples. The e x a m i n a t i o n i n v o l v e d v i s u a l o b s e r v a t i o n and taking photographs of t y p i c a l areas for future reference . The coke specimens employed cover a range of carbon conversion from 0% to 55%. For purposes of comparison, the magni f i cat ion was kept constant at X170, except where d e t a i l e d study was needed. The S . E . M . photographs of these samples are presented in Figures 4 . 2 . 5 . a, b and c. It i s seen in Figure 4 .2 .5 . a that the f l u i d coke p a r t i c l e s are near ly s p h e r i c a l i n shape, have a r e l a t i v e l y uniform s i ze d i s t r i b u t i o n , and have almost no p o r o s i t y . At a carbon conversion of about 30%, a major i ty of these s p h e r i c a l p a r t i c l e s cracked into halves r i g h t at the center , exposing the c r o s s - s e c t i o n a l view of t h e i r i n t e r n a l s t r u c t u r e . The concentr ic carbon l a y e r s , which packs the i n t e r n a l volume of the p a r t i c l e s i n a g r a p h i c a l l y s i m i l a r manner to that o f t h e i n t e r n a l s t r u c t u r e o f an o n i o n , are c l e a r l y seen i n Figure 4 . 2 . 5 . b . At t h i s Xc~30% the coke p a r t i c l e diameters are near ly unchanged from the i n i t i a l va lue , and i t i s evident that the p a r t i c l e s possess some p o r o s i t y . At higher carbon convers ion , e .g . X Q = 5 5 % , most of F . C . p a r t i c l e s become s e m i - s p h e r i c a l "coke s h e l l s " w i t h t h e i r i n t e r n a l carbon layers g a s i f i e d away. V i s u a l observat ion showed an increas ing number of fragments and s m a l l e r p a r t i c l e s , presumably generated from the breakage of these "coke s h e l l s " . F igure 4 . 2 . 5 - c , S.E.M. photos of f l u i d coke at v a r i o u s carbon conversions. 0 9 4 8 4 5 20KV 5@uni F i g . 4.2.5 c photos of f l u i d coke at various carbon conversions. 183 taken from a magni f icat ion of X600, shows i n d e t a i l the i n t e r i o r of such "coke s h e l l s " . 4 .2 .1 .3 DISCUSSIONS F l u i d coke i s the residue from the continuous thermal cracking process i n Syncrude o i l sand p lant to produce synthet ic crude. In t h i s p r o c e s s the bi tumen i s heated to 4 7 5 - 6 0 0 ° C to remove most of the v o l a t i l e products . The F . C . i s therefore r e l a t i v e l y unreac t ive , and contains high ash and sulphur, and low v o l a t i l e matter. The examination of the p h y s i c a l s t ruc ture of the F . C . and i t s p a r t i a l l y g a s i f i e d samples u s i n g S.E.M. r e v e a l s v e r y i n t e r e s t i n g r e s u l t s . As expected, the unreacted coke p a r t i c l e s are s p h e r i c a l and r e l a t i v e l y u n i f o r m i n s i z e , c o n t a i n no p o r o s i t y and have smooth surfaces . However, at a carbon conversion of Xc~30%, a large p o r t i o n of these s p h e r i c a l coke p a r t i c l e s cracked into ha lves , exposing c l e a r l y the o n i o n - l i k e i n t e r n a l s t ructure of the coke. The mechanism by which the coke p a r t i c l e s crack into halves i s not known; i t i s presumably caused by thermal expansion of impur i t i e s in the ash or the escape of the v o l a t i l e s from the inner coke l a y e r s . There i s some p o r o s i t y observed at t h i s convers ion, and the pore volume i s the empty space between two c o n c e n t r i c a l l a y e r s of c a r b o n of d i f f e r e n t d i a m e t e r s . T h i s i s d i s s i m i l a r to the ordinary concept of a pore which involves a pore length and a pore diameter. F igure 4 . 2 . 5 . b c l e a r l y reveals that the i n t e r n a l volume of the coke p a r t i c l e s i s occupied by carbon layers in an o n i o n - l i k e arrangement. As the g a s i f i c a t i o n continued to higher carbon convers ion , these inner layers of carbon wi th in the s p h e r i c a l halves are 184 gradua l ly consumed. At an Xfj~55% these s p h e r i c a l halves become hollow s h e l l s of coke conta in ing perhaps some i m p u r i t i e s , as seen i n Figure 4 . 2 . 5.C The present observat ion i s i n l i n e with publ ished r e s u l t s which speculated that F . C . p a r t i c l e s have smooth surface and are s p h e r i c a l with layered o n i o n - l i k e i n t e r n a l s tructure r e s u l t i n g from the c y c l i c ac t ion of the f l u i d i z e d bed cokers [74,94]. As presented e a r l i e r , the sulphur to carbon molar r a t i o of F . C . i s about 0.033. Thi s r a t i o i s almost consis tent to what was found i n the produced gas during the extent of g a s i f i c a t i o n . Thus, the carbon and sulphur i n the F . C . p a r t i c l e s are g a s i f i e d at the same molar r a t i o as i n t h e i r i n i t i a l composit ion. In F i g u r e 4 . 2 . 3 , the r a t e curves f o r exper iment s at h igher temperature are always above and have less curvature than the ones for lower temperatures. Th i s means tha t , at any carbon convers ion , the rate o f . c a r b o n conversion increases with increas ing temperatures, and that the r a t e s of c a r b o n c o n v e r s i o n at low temperatures decrease more abrupt ly than at higher temperatures. The rate of carbon conversion of F . C . i s lower when compared to the rate of carbon conversion of D . C . for the same experimental c o n d i t i o n s . This i s discussed below i n Sect ion 4 .3 . This i s due to the higher and more uniform temperature i n f l u i d i z e d bed coking to produce F . C . which leads to lower v o l a t i l e content. 185 4.2.2 EFFECT OF P„ n H 2 0 4 .2 .2 .1 EXPERIMENTAL RESULTS In a l l experiments presented i n the foregoing sec t ions , the steam p a r t i a l pressure in the gas i fy ing b las t has been kept constant at 30.3 kPa. At t h i s p a r t i a l pressure , there has always been excess steam at a l l t imes. The average steam conversion was estimated at about 25%. In t h i s s ec t i on , the steam p a r t i a l pressure i s v a r i e d while other parameters such as temperature and p a r t i c l e s i ze are f i x e d . The steam p a r t i a l pressures employed were 15.15, 30.3 and 60.6 kPa. The operat ing condi t ions f o r the experiments discussed i n t h i s s ec t i on are provided in Table 4 . 2 . 1 . Table 4 . 2 . 1 ; EXPERIMENTAL CONDITIONS FOR RUNS AT DIFFERENT P„ „ H 2 0 Run # Temperature, °C dp, mm G a s i f i c a t i o n t ime, min P H 2 0 « atm Type of coke 87 930 0.14 300 0.30 F . C . 94 930 0.14 300 0.60 F . C . 95 930 0.14 300 0.15 F . C . The experimental carbon conversion-t ime curves f o r runs with F . C . at d i f f e r e n t steam p a r t i a l pressure are shown in Figure 4 . 2 . 6 . The Xfj-t curves are not l i n e a r and have i n f l e c t i o n p o i n t s . The i n f l e c t i o n po ints occur at X^ between 18% and 21%, except for the P^ Q=0.15 atm run which i s at 12.5%. It i s seen that the inf luences of steam p a r t i a l pressure in the g a s i f y i n g medium on carbon conversion ranged from moderate to 186 CARBON CONV. VS. REACTION TIME DURING GASIFICATION STEP LEGEND • =RUN #87 SYNCRUDE PH20=0.30 0=RUN #94 SYNCRUDE PH2O=0.60 •=RUN #95 SYNCRUDE PH20=0.15 60 120 180 REACTION TIME,MIN. 240 300 F i g . 4.2.6 Carbon conversions f o r runs at various steam p a r t i a l pressure i n the gasify i n g medium. F l u i d coke; T = 930°C 187 weak for experiments at approximately P <0.3 atm and P -.^0.3 atm, H 2 ° H 2 ° r e s p e c t i v e l y . This carbon conversion a f t er 5 hours of g a s i f i c a t i o n was 30.4% for the P =0.15 atm run; 54.8% for the P =0.3 atm run; and H 2 ° H 2 ° 61.1% f o r the P -.=0.60 atm run. . 2 ° As f o r the case of carbon convers ion, the sulphur conversion i s moderately af fected by the steam p a r t i a l pressure i n the i n l e t gas i n the run wi th P <30.3 kPa, and weakly af fected i n the run with P„ _ H 2 ° H 2 ° above 30.3 kPa. The sulphur conversion a f t er 5 hours of g a s i f i c a t i o n ranged from 35.9% f o r the P„ n =0.15 atm run to 54.1% f o r the P„ _=0.6 H 2 ° H 2 ° atm run . The sulphur conversion for the P ,-.=0.3 atm run was 51.4% as H 2 ° given i n Sect ion 4 . 2 . 1 . 2 . Again these values are s l i g h t l y higher than those obtained for the D . C . i n s i m i l a r operating c o n d i t i o n s . The Xg vs t p lo t for runs at var ious P i s given i n Figure 4 . 2 . 7 . H 2 ° The rate curves (dXfj/dt vs Xfj) f or experiments at d i f f e r e n t P H 2 ° are given i n Figure 4 .2 .8 . These rate curves were obtained by taking the slope of the Xfj-t curves at d i f f e r e n t extents of carbon convers ion . The values of Xfj where the rate curves a t t a i n maxima i n the dXfj/dt vs Xfj p l o t correspond to the i n f l e c t i o n points mentioned i n Sect ion 4 . 2 . 2 . 2 . The rate curve for the P =0.15 atm run i s d i s t a n t l y below the curve H 2 ° f o r the P -.=0.3 atm r u n . T h i s i m p l i e s that the rate of carbon H 2 ° conversion f o r the P =0.15 atm g a s i f i c a t i o n i s slow compared to that H 2 0 of higher steam p a r t i a l p r e s s u r e r u n s . The n u m e r i c a l v a l u e s of ( d X c / d t ) M are 1 .28xl0~ 3 , 2 .12xl0~ 3 m i n - 1 and 2 . 3 0 x l 0 - 3 m i n - 1 for runs with P„ =0.15, 0.30 and 0.60 atm, r e s p e c t i v e l y . H 2 ° As p r e v i o u s l y r e p o r t e d , the s u l p h u r to carbon molar r a t i o c a l c u l a t e d using r e s u l t s of u l t imate ana lys i s for the F . C . i s 0.033. 188 SULFUR CONV. VS. REACTION TIME EXPERIMENTAL DATA 120 180 REACTION TIME,MIN. 240 300 F i g . 4 .2 .7 Sulphur convers ions of f l u i d coke f o r runs a t v a r i o u s steam p a r t i a l p ressu re . T = 930°C 189 CARBON CONVERSION VS. dXr /dt FROM EXPERIMENTAL DATA F i g . 4.2.8 Rate of carbon conversion at various steam p a r t i a l pressure. F l u i d coke; T = 930°C 1 9 0 The amount of sulphur in the coke p a r t i c l e s which was g a s i f i e d per mole of carbon in runs of d i f f e r e n t P^ ^ can be examined from the Xg vs XQ p l o t , Figure 4 .2 .9 . The values of dXg/dXfj were almost the same for both the P n=30.3 and the P _=60.6 kPa. runs, being 0.92. Th i s dXg/dXr was H 2 ° H 2 ° s l i g h t l y h igher , at 1.07, for the P =0.15 atm run. In genera l , the H 2 ° sulphur to carbon molar r a t i o s i n the produced gas for a l l three runs are approximately between 0.03 and 0.035, which i s very c lose to the r a t i o der ived from the ul t imate analys i s of the i n i t i a l F . C . of 0.033. Therefore , i t may be assumed that carbon and sulphur i n the F . C . were g a s i f i e d i n the same molar r a t i o as i n the f resh F . C . 4 .2 .2 .2 DISCUSSIONS I t was seen that the rate of carbon or sulphur conversion was moderately af fected by the steam p a r t i a l pressure i n the g a s i f y i n g medium for runs of small P . Nevertheless , the inf luence of steam 2 p a r t i a l pressure became much weaker at a steam p a r t i a l pressure of about 0.3 atm and above. At a l l t imes, the steam was i n excess. The s a t u r a t i o n of the e f fec t s of steam p a r t i a l pressure on the rate of carbon conversion at high P can be e a s i l y i d e n t i f i e d by the rate H 2 ° c u r v e s , F i g u r e 4 .2 .8 . In t h i s f i g u r e , when the value of P was H 2 ° doubled from 0.15 to 0.30 atm, the conversion rate increases sharply almost by a f a c t o r of 2. However, when the value of P was doubled H 2 ° again from 0.30 to 0.60 atm, the increase i n the carbon conversion rate was marg ina l . The g a s i f i c a t i o n rate may be assumed to be propor t iona l to the steam p a r t i a l pressure to the power n as fo l lows: 191 SULFUR CONV. VS. CARBON CONV. EXPERIMENTAL DATA 60 I • : : • : : : 1 LEGEND • =RUN #87 PH20=0.30 SYNCRUDE 0=RUN #94 PH20=0.60 SYNCRUDE • =RUN #95 PH20=0.15 SYNCRUDE 20 40 60 CARBON C O N V E R S I O N S 80 F i g . 4.2.9 Sulphur conversion as a function of carbon conversion for runs at various steam p a r t i a l pressures. F l u i d coke; T = 930 C 1 9 2 (dXC/dt) o<P n ( 1 4 5 )  H 2 ° An average value of n of 0.43 was given i n the l i t e r a t u r e for steam g a s i f i c a t i o n [104] for steam p a r t i a l pressures from 10.1 to 101.0 kPa. Although only three points are a v a i l a b l e i t appears that at p a r t i a l pressures of 0.3 atm and below the rate i s roughly p r o p o r t i o n a l to the steam p a r t i a l pressure ( n ~ l ) , but f o r the higher pressures the rate v a r i e s as n~0.4. S i m i l a r to what was observed i n Sect ion 4 . 2 . 1 , studying the e f fec t s of temperature, the carbon to sulphur molar r a t i o was near ly equal to that der ived from the r e s u l t s of u l t imate ana lys i s of the f re sh F . C . for the e n t i r e extent of g a s i f i c a t i o n . This suggests that the sulphur i s evenly d i s t r i b u t e d i n the F . C . matr ix . 4.3 COMPARISONS BETWEEN GASIFICATION OF D . C . AND F . C . Both delayed and f l u i d cokes are b y p r o d u c t s from the t h e r m a l crack ing process of o i l sand bitumen. The d i f f erences between the two cokes with respect to g a s i f i c a t i o n may be due to the d i f f erences i n the two coking methods - delayed and f l u i d i z e d - which are descr ibed in Sect ion 2 .1 . The f l u i d coking process takes place at h igher temperature and w i t h b e t t e r t h e r m a l c o n t a c t than that f o r the delayed coking process; there fore , i t gives higher l i q u i d y i e l d . As a consequence, the f l u i d coke i s very h igh i n ash, heavy metal and sulphur , while at the same time being low in v o l a t i l e s as g iven i n Table 2.4. The sulphur to carbon molar r a t i o s in the fresh coke are 0.027 and 0.033 for D . C . and F . C , r e s p e c t i v e l y . 193 The D . C . was crushed and sieved to obta in the same s i ze f r a c t i o n as of F . C , and both were g a s i f i e d at the same experimental condi t ions given in Table 4 . 3 . 1 . Table 4.3.1 EXPERIMENTAL CONDITIONS FOR RUNS COMPARING D . C . AND F . C . Run# Temperature, °C d p c , g a s i f i c a t i o n P ^ Q , atm Type of Coke mm time, min 2 87 930 0.14 300 0.30 F . C . 88 900 0.14 300 0.30 F . C 89 880 0.14 300 0.30 F . C . 90 880 0.14 300 0.30 D . C . 91 900 0.14 300 0.30 D . C 92 930 0.14 300 0.30 D . C . In t h i s s ec t i on , the carbon convers ion, the g a s i f i c a t i o n rate and the H2S e m i s s i o n l e v e l s of D . C . and F . C . are compared at each g a s i f i c a t i o n temperature. 4 .3.1 COMPARISON OF CARBON CONVERSION The Xfj-t curves generated from runs of d i f f e r e n t temperatures of both D . C . and F . C are constructed against reac t ion time and shown in Figure 4 . 3 . 1 . Part of Figure 4.3.1 i s F igure 4 . 2 . 1 , showing the e f fec t s of temperature on the carbon conversion of F . C In F igure 4 .3 .1 , the carbon conversion of D . C i s higher than that of F . C . for the same operat ing cond i t i ons , during the e n t i r e extent of g a s i f i c a t i o n . This may be re la t ed to the o r i g i n a l compositions of the cokes, in which F . C . has a lower v o l a t i l e but higher ash content. The carbon conversions of 194 CARBON CONV. VS. REACTION TIME DURING GASIFICATION STEP F i g . 4.3.1 Carbon conversions of delayed and f l u i d cokes for runs at d i f f e r e n t temperatures. 195 the two cokes at var ious temperatures, a f t e r 5 hours of g a s i f i c a t i o n , are presented in Table 4 .3 .2 . Table 4.3.2 CARBON CONVERSIONS OF D . C . AND F . C . AFTER 5 HRS OF  GASIFICATION Temperature, °C Carbon Conversion, %, Xp D . C . F . C . 930 69.07 48.63 900 53.28 39.95 880 41.79 31.02 It i s seen tha t , carbon conversions f o r both D . C . and F . C . increase wi th increas ing temperature; however, the carbon conversion of the D . C . i s more s e n s i t i v e to a change i n g a s i f i c a t i o n temperature than that of F . C . Carbon conversion of D . C . increases from 53.3% to 69.1%, or by 29.6%, corresponding to an increase i n temperature from 9 0 0 ° C to 9 3 0 ° C . The same increase i n temperature induces only a 21.7% increase i n Xfj for the case of F . C . 4 .3 .2 COMPARISON OF H?S EMISSIONS In t h i s present study, the only sulphur species detected from the product gas was hydrogen su lphide , H 2 S . During each run, the l e v e l of H 2 S i n the gas was measured in time i n t e r v a l s of 10 and 20 minutes. The concentrat ions of H 2 S , f or the s i x experiments given in Table 4 . 3 . 1 , are presented in Figure 4.3.2 as a funct ion of t ime. This f igure reveals that at the same g a s i f i c a t i o n temperature, the l e v e l s of H 2 S i n runs with F . C . are c o n s i s t e n t l y higher than that produced by runs with D . C . This may be re la ted to the higher i n i t i a l 196 H2S CONC. VS. REACTION TIME COMPARE SYNCRUDE WITH SUNCOR 1600 OH > ryf 1200 <; o 2; o u o CV? 800 400 LEGEND = RUN #87 SYNCRUDE T=930C 0=RUN #88SYNCRUDE T=900C • =RUN #89 SYNCRUDET=880C ~ = RUN #90 SUNCOR T=880C • =RUN #91 SUNCOR T=900C ~ = RUN #92 SUNCOR T=930C 120 240 REACTION TIME,MIN. 360 F i g . 4.3.2 Hydrogen sulphide concentration for delayed and f l u i d cokes at various temperatures. PH 20 = 3 ° - 3 k P a 197 sulphur content of the F . C . It i s a lso seen that , for each of the cokes, the H2S concentrat ions were higher for increas ing temperature. The average H2S concentrat ions for F . C . i s approximately 1300 ppm and 950 ppm for temperatures of 9 3 0 ° C and 8 8 0 ° C , whereas the l e v e l s of H2S were approximately 1200 ppm and 850 ppm, r e s p e c t i v e l y for the D . C . I t should be noted that the r e a c t i o n time in Figure 4.3.2 includes the heat-up time of 60 minutes and the g a s i f i c a t i o n time of 300 min as seen i n other f i g u r e s . During t h i s heat-up time, the H2S l e v e l s sharply increased and reduced as the v o l a t i l e s were q u i c k l y released from coke p a r t i c l e s . 4 .3.3 COMPARISON OF SULPHUR CONVERSIONS The sulphur conversions with D . C . and w i t h F . C , at the same temperature, are compared in Figure 4 . 3 . 3 . Un l ike the case of carbon convers ion , i n which carbon conversions of runs employed D . C . are always higher than those using F . C , sulphur conversions of runs employing F . C . are found to be c o n s i s t e n t l y higher than f o r runs with D . C However, the d i f f erence in sulphur conversions between the two cokes, for the same temperature, i s very smal l . Sulphur convers ions , a f t er 5 hrs of g a s i f i c a t i o n , of runs using f l u i d and delayed cokes are given i n Table 4 . 3 .3 . 4 .3 .4 COMPARISON OF THE RATES OF CARBON CONVERSION The carbon conversion of runs employing D . C , at any t i m e , i s higher than that of runs employing F . C , provided that the two runs were 198 SULFUR CONV. VS. REACTION TIME EXPERIMENTAL DATA 60 LEGEND • = RUN #92 SUNCOR T=930C • =RUN #87 SYNCR. T=930C 0=RUN #91 SUNCOR T=900C V=RUN #88 SYNCR. T=900C • = RUN #90 SUNCOR T=880C + = RUN #89 SYNCR. T=880C 120 180 REACTION TIME,MIN. 300 F i g . 4.3.3 Sulphur conversions for runs with delayed and f l u i d cokes at various temperatures. P H Q = 30.3 kPa 199 c a r r i e d out at the same cond i t i ons . The higher r e a c t i v i t y of D . C . over that of F . C . i s a l so r e f l e c t e d in the rates of carbon convers ion . Table 4 .3 .3 SULPHUR CONVERSION OF D . C . AND F . C . AFTER 5 HRS Temperature, °C Sulphur convers ion, Xg, % D . C . F . C . 930 48.63 51.41 900 39.95 43.05 880 31.02 33.36 The rates of carbon conversion of both D . C . and F . C . are shown in Figure 4 .3 .4 at var ious extents of g a s i f i c a t i o n . In t h i s f i g u r e , the r a t e c u r v e s f o r experiments wi th D . C . are always above the curves represent ing experiments with F . C , at the same temperature. It i s a l so not iced that the rate curves for the D . C . are broader i . e . have l ess curvature than those represent ing F . C . Thi s ind icates that the carbon conversion rate for the F . C dropped sharply a f ter the rate maximum, while those for the D . C . continued to be high in the same p e r i o d . The highest values of dXp/dt for runs with D . C and F . C at 9 3 0 ° C are 2 . 5 5 x l 0 - 3 m i n - * and 2 .12xl0~ 3 rain-*, r e s p e c t i v e l y . 4 .3 .5 DISCUSSION The experimental r e s u l t s ind ica te that the r e a c t i v i t y of F . C i s lower than that of D . C . This perhaps was caused by the more severe thermal cracking condi t ions in the f l u i d coking process , in which the coke p a r t i c l e s undergo a c y c l i c ac t ion between the coker and the burner at a temperature higher than that employed in delayed coking which leads 200 CARBON CONVERSION VS. dX c /dt FROM EXPERIMENTAL DATA J 20 o X! * o 20 40 60 CARBON CONV. IN GASIFICATIONS F i g . 4 . 3 . 4 Rates of carbon conversion for runs with delayed and f l u i d cokes at various temperatures. P H 2 O = 3 0 • 3 K P A 201 to a more e f f i c i e n t removal of the v o l a t i l e s from the coke i n the f l u i d i z e d bed coker. Consequently, the F . C . contains a higher amount of sulphur and other impuri t ies in i t s matr ix than does the D . C . The higher sulphur in the F . C . i s p a r t l y respons ib le for the higher H2S emissions i n the produced gas, as demonstrated i n Figure 4 . 3 . 2 . The h igher sulphur content i n the F . C , compared to that of D . C , combined wi th the higher sulphur emissions produced i n the gas lead to the approximately the same degree of sulphur conversions for both of the cokes as shown i n Figure 4 . 3 . 3 . It i s observed that the Xrj-t curves f o r runs c a r r i e d out with D . C . and wi th F . C . possess i n f l e c t i o n points which correspond to maxima in (dX^/dt) curves . The values of X ^ R at which the rates reach maxima range from 1 1 % to 2 6 % . These values are i n good agreement with both the e x p e r i m e n t a l d a t a and t h e o r e t i c a l p r e d i c t i o n s p u b l i s h e d i n the l i t e r a t u r e . In the g a s i f i c a t i o n of chars and c o a l s , many workers [ 2 , 3 , 5 0 , 4 9 , 2 0 1 ] have prev ious ly reported maxima i n the rate of g a s i f i c a t i o n . The t y p i c a l values of X C R reported are 0 . 1 - 0 . 4 by Dutta & Wen [ 5 0 ] and 0 . 2 -0 . 6 by Young et a l . [ 2 0 1 ] . Other authors [ 1 7 1 , 1 4 0 , 6 6 , 1 1 6 ] have employed models to s u c c e s s f u l l y p r e d i c t the value of X ^ R . T y p i c a l l y , t h e i r pred ic ted values of X Q R are between 0 and 0 . 5 0 . Table 4 . 3 . 4 summarizes values of X ^ R determined i n t h i s study and the values from the l i t e r a t u r e . 202 Table 4 .3 .4 VALUES OF CARBON CONVERSION WHERE ( d X r / d T ) ATTAINS MAXIMUM Authors This Suetal . Gavalas Peterson Dutta et al. Adshlrl Lee et al. Young et al. Study [171] [66] [140] [49,50] [2,3] [116] [201] 0.11- 0.2- 0-0.4 0-0.5 0.1-0.4 0.3 0.2 0.2-0.6 0.26 0.35 The s u l p h u r to carbon molar r a t i o of f r e s h D . C . and F . C . r e s p e c t i v e l y are 0.027 and 0.033, while the examination of a l l X§ vs X^ f igures f o r the D . C . and for the F . C . ( i . e . Figure 4 . 1 . 6 , 4 .1 .12 , 4.2.4 and 4 .2 .9) reveals that the average sulphur to carbon molar r a t i o s in the produced gas are approximately 0.015 i n the g a s i f i c a t i o n of D . C , and 0.034 i n the g a s i f i c a t i o n of F . C . These d i s s i m i l a r v a l u e s demonstrate tha t , during the course of steam g a s i f i c a t i o n , the sulphur i n the F . C . i s g a s i f i e d i n the same sulphur/carbon r a t i o as that i n the o r i g i n a l f r e s h coke. By contras t , the sulphur in the D . C i s g a s i f i e d at a l e s s e r su lphur/carbon molar r a t i o than found i n the i n i t i a l D . C . Two poss ib l e explanations are proposed. It may be that the carbon in the D . C i s more reac t ive than the carbon i n the F . C . matr ix , given the same r e a c t i v i t y of the sulphur i n the two cokes. The d i f f erence in r e a c t i v i t i e s could be the re su l t s of two d i f f e r e n t coking condi t ions and of the d i f f e r e n c e i n chemical compositions of the two cokes. A l s o , c y c l i c act ions i n the f l u i d coking process form the coke p a r t i c l e s layer by layer and thus, perhaps d i s t r i b u t e the sulphur atoms evenly in the F . C . s t r u c t u r e , which may account for the s i m i l a r i t y in the sulphur to carbon molar r a t i o s between the fresh F . C . and the produced gas. 203 4.4 SUMMARY OF GASIFICATION RESULTS S i g n i f i c a n t f indings that have been obtained by t h i s study i n the g a s i f i c a t i o n of delayed and f l u i d cokes at var ious operat ing condi t ions are summarized below: 1. Carbon c o n v e r s i o n , s u l p h u r c o n v e r s i o n and the rate of carbon convers ion increase with increas ing temperature and with decreasing coke p a r t i c l e s i z e s . 2. The X ^ - t curves possess i n f l e c t i o n points which r e s u l t in maxima in the rate of carbon convers ion. 3. R e g a r d l e s s o f o p e r a t i n g c o n d i t i o n s s u c h as g a s i f i c a t i o n temperatures, coke p a r t i c l e s i zes and reac t ion t imes, the surface areas based on i n i t i a l carbon for g a s i f i c a t i o n depend s o l e l y upon the extent of g a s i f i c a t i o n . The surface area may be expressed by a quadrat ic func t ion of convers ion. 4. a) SEM e x a m i n a t i o n of the D . C . specimens r e v e a l s that the unreacted coke i s non-porous and v a r i e s i n shape that small pore systems d e v e l o p e d at XQ approximately 7% which then e n l a r g e d and o v e r l a p p e d as the ex ten t of g a s i f i c a t i o n progressed, and that the D . C . p a r t i c l e s d i s i n t e g r a t e d at XQ above 75%. b) S . E . M . examination of the F . C . specimens ind ica te s that the f l u i d coke p a r t i c l e s are s p h e r i c a l , non-porous and have "onion" l i k e i n t e r n a l s tructure and that the p a r t i c l e s crack in to halves at a XQ approximately 30%, exposing carbon layers 204 which are g a s i f i e d away leaving some impur i t i e s ins ide the emptied s h e l l s . A change i n porous s tructure changes the surface area for r e a c t i o n . Both the porous s t ruc ture and the surface area of the coke d i c t a t e the rate of carbon convers ion. For a g iven experimental c o n d i t i o n , both the carbon conversion and the rate of g a s i f i c a t i o n of D . C . are higher than that of F . C ; while the sulphur emissions i n runs with D . C . are lower than that with F . C . The sulphur to carbon molar r a t i o f o r the produced gas i n runs with F . C . i s near ly equal to the S/C molar r a t i o i n the i n i t i a l f resh c o k e , whereas the S /C molar r a t i o f o r the gas wi th D . C . i s s u b s t a n t i a l l y below that found in the i n i t i a l coke. The values of carbon conversion where the surface area and the rate a t t a i n maxima are i n agreement wi th the postulated values in the l i t e r a t u r e . Maximum g a s i f i c a t i o n rates occur at lower carbon conversions than the maximum surface area . 205 CHAPTER 5 DESULPHURIZATION G a s i f i c a t i o n with steam i s one of the o p t i o n s to u t i l i z e the s t o c k p i l e d o i l sand cokes. This step re leases sulphur in to the product gas i n the form of H2S. Both limestone and dolomite have the power to accept sulphur from H2S. However, they react with hydrogen sulphide only i n the form of c a l c i n e s . F u l l y c a l c i n e d limestone and dolomite were used to capture sulphur i n - s i t u according to reac t ion (89) and (90), r e s p e c t i v e l y : CaO + H 2 S -• CaS + H 2 0 [89] [CaO^MgO] + H 2 S -» [CaS«MgO] + H 2 0 [90] In t h i s present study, the i n - s i t u removal of sulphur takes place at high temperature to render the f u e l products low i n sulphur and, at the same time, r e t a i n the sens ib le heat of the gas. A l l carbonate rock samples were c a l c i n e d in an atmosphere of N2, at 9 5 0 ° C and f o r var ious c a l c i n a t i o n t imes. The c a l c i n a t i o n temperature of 9 5 0 ° C , which i s s l i g h t l y higher than the g a s i f i c a t i o n temperature, was chosen to minimize the p o s s i b i l i t y of c a l c i n e s changing t h e i r porous s t ruc ture due to the high temperatures of g a s i f i c a t i o n . The experiments described in t h i s chapter were a l l conducted at the g a s i f i c a t i o n temperature of 930"C and with a steam p a r t i a l pressure of 30.3 kPa. 206 The amount of H2S removed from the gaseous product depends on many v a r i a b l e s : the quant i ty of sorbent used i . e . the Ca/S r a t i o , the type of c a l c i n e , the sorbent p a r t i c l e s i z e , and the condi t ions under which the carbonate rock was c a l c i n e d . The e f fec t of these v a r i a b l e s on the q u a l i t y of s u l p h u r removal and on the r e a c t i v i t y of sorbent were i n v e s t i g a t e d . To e s t a b l i s h the r e l a t i o n s h i p between the sorbent q u a l i t y and i t s porous s t r u c t u r e , representat ive specimens of the c a l c i n e d samples were taken for chemical composition analyses, f or surface area measurement, f o r S . E . M . examination and f o r p o r o s i t y d e t e r m i n a t i o n a f t e r each c a l c i n a t i o n . The r e s u l t s of the chemical composition ana lys i s for a l l samples employed i n t h i s study are reported i n Table 5 . 1 . 1 . 5.1 EFFECT OF SORBENT ADDITION ON CARBON CONVERSION The e f f e c t of the presence of c a l c i n e d l i m e s t o n e on t h e g a s i f i c a t i o n of o i l sand coke was tes ted; and the g a s i f i c a t i o n r e s u l t s from runs wi th sorbent (Run 40) and without sorbent (Run 47) at the same operat ing condi t ions are compared i n Figure 5 .1 .1 . C A L l , the sorbent used i n Run 40 was thermally decomposed from Cache Creek limestone in a c a l c i n a t i o n time of 40 minutes. The molar Ca/S r a t i o was 2 .0 . It i s seen t h a t c a r b o n c o n v e r s i o n was not improved nor reduced as a consequence of the addi t ion of ca l c ined sorbent to remove hydrogen sulphide from the produced gas. Calcium ion was postulated to promote not only the water -gas - sh i f t r e a c t i o n but a lso the g a s i f i c a t i o n react ions [185,107,144,186,188] and hence cause increases i n the carbon convers ion . However, the c a t a l y t i c 207 Table 5.1.1 CHEMICAL COMPOSITIONS OF CALCINED SORBENTS Major Components, wt% Sample ID CAL1 (*) (*) CALA CALB ONI QN3 DL1 DL2 D FM HL Silica SI02 L L 0.60 1.80 0.90 1.93 1.98 2.27 2.60 2.10 8.16 23.3 Alumina AI2O3 L L 0.50 0.65 0.45 0.12 0.50 0.17 0.40 0.20 2.10 6.22 Iron F e ^ 0.075 L 0.18 0.27 0.20 0.21 0.22 0.21 0.21 0.25 0.79 2.40 Calcium CaO 60.4 56.2 55.7 94.1 95.8 57.2 94.8 57.0 54.5 54.5 87.0 63.0 Magnesium MgO 0.34 0.29 0.35 0.79 0.57 38.7 1.45 39.9 40.1 41.0 0.79 1.69 Sodium Na^ L L 0.02 0.03 0.01 0.01 0.03 0.01 0.01 0.01 0.02 0.05 Potassium K2O 0.01 L 0.02 0.04 0.01 0.03 0.04 0.03 0.03 0.36 0.36 1.32 L.O.I. 39.1 43.6 43.1 1.48 1.96 0.74 0.25 L 2.12 1.76 1.57 1.13 Trace Components, ppm Barium Ba 40 L L 90 120 L L L 60 20 107 200 Manganese Mn 77 62 96 150 140 200 110 210 275 240 800 480 Phosphorous P2O5 Strontium Sr 175 158 830 420 300 71 75 77 90 70 284 440 Titanium Tl 82 65 177 300 205 L L L 200 100 600 1500 (*) = Cache Creek Limestone L ° Below detection IImit CALA, CALB, QN3: From Cache Creek IImestone QN1, DL1, DL2, D: From dolomite FM: From Fort MacMurray Iimestone HL: From Have lock limestone ) 208 CARBON CONV. VS. REACTION TIME DURING GASIFICATION PROCESS 60 0 60 120 180 240 300 REACTION TIME,MIN. F i g . 5.1.1 E f f e c t of sorbent addition on carbon conversion. Delayed coke; p u = 30.3 kPa; H 2 ° T = 930°C; d = 0.68 mm 209 enhancement of the calc ium species present i n the added sorbent d id not m a t e r i a l i z e in t h i s work. This f ind ing agrees with the suggestions that CaO must be f i n e l y dispersed in the coke to show the c a t a l y t i c e f fec t s on carbon conversion [96,189,65], rather than simply being mechanical ly mixed wi th the coke. Run 47 i s the repeat of Run 35 and Run 86 ( longer g a s i f i c a t i o n -time) presented i n Sect ion 4 .1 .3 . The g a s i f i c a t i o n r e s u l t s of the three runs are i n agreement to w i th in ±7%. The good r e p r o d u c i b i l i t y of Run 47 has s i g n i f i c a n t importance because these repeated runs were used as a reference i n the desu lphur iza t ion process . 5.2 EFFECT OF MIXING In Sect ion 4 . 1 . 1 , the e f fec t of mechanical mixing on g a s i f i c a t i o n process was i n v e s t i g a t e d . In the combined i n - s i t u g a s i f i c a t i o n -d e s u l p h u r i z a t i o n process , the sorbent and the coke p a r t i c l e s have been w e l l mixed p r i o r to t h e i r in troduct ion into the r e a c t i o n chamber, and then a mechanical s t i r r e r was employed to continue mixing them during the e n t i r e experiment. A quest ion of i n t e r e s t i s what would the c o n c e n t r a t i o n of H2S i n the t r e a t e d gas be i f the c a l c i n e s were s t a t i c a l l y located as a layer on top of the o i l sand coke. To study the e f f ec t of mixing of c a l c i n e s , c a l c i n e d Cache Creek l imestone, QN3, thermal ly decomposed for 120 min, was employed i n the experiments at Ca/S molar r a t i o s of 0.5, 1.0 and 2 .0 . The r e s u l t s of these experiments are i l l u s t r a t e d i n Figure 5 .2 .1 . It i s seen that the c a l c i n e in the s t a t i c layer more r e a d i l y accepted hydrogen sulphide at f i r s t , but i t s r e a c t i v i t y soon became retarded to values lower than that 210 H2S CONC. VS. REACTION TIME FOR RUNS TESTING MIXING EFFECTS 1200 800 LEGEND • = NO STIRRER Ca/S=0.5 0= WITH STIRRER Ca/S=0.5 •= NO STIRRER Ca/S=1.0 • = WITH STIRRER Ca/S=1.0 • = NO STIRRER Ca/S=2.0 ©= WITH STIRRER Ca/S=2.0 400 0 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.2.1 Ef f e c t of mixing on H^S concentrations. Runs with calcined limestone. 211 of runs employing mechanical s t i r r i n g for the remaining per iod of the reac t ion time. The o v e r a l l e f f i c i e n c y i n removing sulphur from the gas stream i s higher f o r the runs wi th the s t i r r e r o n . M o r e o v e r , as observed i n Sect ion 4 . 1 . 1 , the temperature measured ins ide the reac t ion chamber was more s tab le and c l o s e r to the preset value when the s t i r r e r was employed. Therefore , a l l experiments were hereaf ter conducted with the s t i r r e r on. 5.3 EFFECT OF CALCINATION CONDITIONS 5.3.1 EXPERIMENTAL RESULTS I t i s w e l l e s tab l i shed that the c a l c i n a t i o n condi t ions such as the p a r t i a l pressure of CO2 in the surrounding atmosphere, the heating rate and the c a l c i n a t i o n time have strong inf luences on the porous s tructure of the c a l c i n e d samples . Th i s porous s t r u c t u r e , i n t u r n , l a r g e l y d i c t a t e s the r e a c t i v i t y and the capac i ty of the c a l c i n e as a sulphur s o r b e n t . In t h i s s t u d y , a c a l c i n a t i o n atmosphere of n i trogen was employed and an unchanged heat ing rate of 9 . 2 5 ° C / m i n was appl ied to reach a c a l c i n a t i o n temperature of 9 5 0 ° C . To study the e f fec t s of the c a l c i n a t i o n condi t ions on the l e v e l of H2S emiss ion, i . e . on f r a c t i o n a l H2S removal from the gas phase, on the sorbent conversion and on the rate at which sorbent i s converted, var ious c a l c i n a t i o n t imes were t r i e d . D e t a i l s of samples used to inves t igate the e f fec t of c a l c i n a t i o n condi t ions are given i n Table 5 .3 .1 . Table 5.3.1 CALCINATION CONDITION FOR VARIOUS SAMPLES CALCINED IN N 9 AT 9 5 0 ° C Sample ID Run* Type of carbonate rocks Size range, Average s i ze mm d p s , mm C a l c i n a t i o n t ime, min. CAL1 QN3 CALB DL1 QN1 40 69 54 76 71 limestone limestone limestone dolomite dolomite -.850+.500 -.850+.500 -.850+.500 -.850+.500 -.850+.500 0.68 0.68 0.68 0.68 0.68 40 120 300 90 120 F igure 5.3.1 shows the concentrat ions of H2S i n the produced gas during g a s i f i c a t i o n f o r runs w i t h s o r b e n t s produced at d i f f e r e n t c a l c i n a t i o n t imes. The Ca/S molar r a t i o was kept at 2.0 f o r a l l three runs. The curve f o r the run without sorbent, Run 47, was a l so p l o t t e d so that the reduct ion i n the H2S l e v e l can be g r a p h i c a l l y compared. The average concentrat ion of H2S i n Run 47 i s approximately 1100 ppm. The l e v e l of H2S i n the p r o d u c t gas i s highest f o r the run conducted wi th sorbent of 40 min. c a l c i n a t i o n ; the average l e v e l i s about 600 ppm. The H2S l eve l s were lower, to the average of 475 ppm and 375 ppm, f o r runs employing sorbent prepared i n 300 min. and i n 120 m i n . , r e s p e c t i v e l y . The e f f e c t of c a l c i n a t i o n time on the r e a c t i v i t y of sorbent with H2S was a l so found i n runs wi th c a l c i n e d dolomite . In Figure 5 .3 .2 , dolomite c a l c i n e d for 120 min was seen to be more e f f e c t i v e i n capturing sulphur than the sample ca l c ined for 90 min. In genera l , H2S in the product gas was completely captured by the employed sorbents during the per iod of the s u l p h i d a t i o n process , up to 213 H2S CONC.VS. REACTION TIME FOR RUNS AT DIFF. CALCINATION 1600 n 1200 800 400 0 LEGEND •= RUN #47 NO SORBENT ADDED 0= RUN #40 LIME. 1= 40 MIN. •=RUN #69 LIME. t= 120 MIN. ~=RUN #54 LIME. t= 300 MIN. 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.3.1 Eff e c t of c a l c i n a t i o n time on H^S concentrations i n runs using calcined limestone. 214 H2S CONC. VS. REACTION TIME FOR RUNS AT DIFF. CALCINATION 1600 > GO C 1200 o I—I E-2 : w o o o cv? w 800 400 LEGEND • =RUN#47 NO SORBENT ADDED 0=RUN #76 DOLOMITE fc=90 MIN. •=RUN #71 DOLOMITE t=120M!N. 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.3.2 Ef f e c t of c a l c i n a t i o n time on H^S concentrations i n runs using calcined limestone. 215 100 min for runs with ca lc ined limestone and up to 140 rain for runs with c a l c i n e d dolomite . The lowest detectable H2S concentrat ion was 50 ppm. The s u l p h i d a t i o n react ion converts CaO in the c a l c i n e d sorbent into c a l c i u m s u l p h i d e , CaS . The u t i l i z a t i o n of sorbent or the sorbent convers ion , XgB» f ° r ca l c ines prepared i n d i f f e r e n t c a l c i n a t i o n times i s given i n Figure 5 .3 .3 . The curves have a l i n e a r s ec t ion i n the e a r l y s tages of c o n v e r s i o n , and then f a l l o f f as the conversions become higher . In t h i s f i gure the curves representing Run 54 and Run 69 are n o n - s i g m o i d a l and s i m i l a r in shape, e s p e c i a l l y i n the e a r l y stage; whereas, the curve for Run 40, which employed sorbent c a l c i n e d for 40 min, has an i n f l e c t i o n point at an Xgjj ° f ~5% and i s d i s t i n c t l y below the other two curves . The non-s igmoidal behavior of curves for Runs 54 and 69 r e s u l t s in rates corresponding to these runs decreasing monotonical ly wi th t ime. By c o n t r a s t , the rate curve f o r Run 40 a t ta ins a maximum and then drops sharply as i s demonstrated in Figure 5 .3 .4 . The u l t imate purpose of c a l c i n e add i t ion i s to produce a c lean fue l gas by removing as much hydrogen sulphide as p o s s i b l e . The f r a c t i o n a l removal of H2S from the produced gas , based on the l e v e l i n the corresponding run with no sorbent present , Run 47, i s def ined as fol lows mole H 2 S In the gas mole H 2 S In the gas r , . . . , u . In run without sorbent ~ In run with sorbent Fractional removal of H o S = , u . , ; ,.. . . . (146) L mole H 2 S in the gas In run without sorbent The f r a c t i o n a l removals of H2S as a funct ion of reac t ion time for runs employing sorbents ca lc ined at d i f f e r e n t durat ions are given in F i g u r e 5 . 3 . 5 . In t h i s f i g u r e , f o r a molar r a t i o C a / S > 0 . 5 , the 216 SORBENT CONV. VS. REACTION TIME DATA FOR DIFF. CALCINATION 20 CO cti O 15 CO K W 10 0 60 120 180 240 TOTAL REACTION TIME,MIN. 300 360 F i g . 5 .3 .3 E f f e c t s of c a l c i n a t i o n conditions on sorbent conversion. 217 SORBENT CONVERSION VS. d X » / d t DATA FOR DIFF. CALCINATION 8 I m w X * o 0 LEGEND • =RUN #40 LIM. t=40 MIN. 0=RUN #54 LIM. 1=300 MIN •=RUN #89 LIM. t=120 MIN O • ; • 0 5 10 15 SORBENT CONVERSION CaO TO CaS,% 20 F i g . 5 .3 .4 Rate of sorbent convers ion for runs w i t h c a l c i n e s of v a r i o u s c a l c i n a t i o n c o n d i t i o n s . 218 FRACTIONAL REMOVAL OF H2S SORBENT OF DIFF. CALC. COND. 0.8 0.6 0.4 0 60 120 180 240 TOTAL REACTION TIME, MIN. 300 360 F i g . 5 . 3 . 5 F r a c t i o n a l l^S removal vs. time for runs with sorbents of d i f f e r e n t c a l c i n a t i o n conditions. 219 f r a c t i o n a l removal was 1.0 for the f i r s t hour. As the su lph ida t ion r e a c t i o n proceeded, the sorbent was converted and the f r a c t i o n a l removal dropped qui te r a p i d l y . A f t e r 6 hours of r e a c t i o n , the l e v e l s of H2S were reduced, on average, by 72%, 64% and 60% for runs ..with c a l c i n e d l i m e s t o n e t h e r m a l l y decomposed i n 120 m i n . , 300 min. and 40 m i n . , r e s p e c t i v e l y . The i m p l i c a t i o n i s that the 120 min. c a l c i n e d sample has higher capac i ty to capture sulphur than that of the 300 min. or 40 min. c a l c i n e d sample. The r e l a t i o n s h i p s between c a l c i n a t i o n t ime, the f r a c t i o n a l removal of H2S from the gas phase and the corresponding B . E . T . surface area, measured u s i n g N2 at 77K, of the c a l c i n e d sorbents are shown in Figure 5 .3 .6 . In t h i s f i g u r e , both the surface area of the sorbent and the f r a c t i o n a l H2S removal are highest for Run 69 which employed sorbent c a l c i n e d i n 120 rain. The S . E . M . photographs of the f r e sh limestone and a l l c a l c i n e d samples employed i n t h i s study are presented in Figure 5 . 3 . 7 . a - j . It i s evident that the f resh limestone has very l i t t l e poros i ty and surface area; t h i s may i n h i b i t the s u l p h i d a t i o n from taking p lace . When the l i m e s t o n e was heated f o r 40 rain, at 9 5 0 ° C i n n i t r o g e n , incomplete decomposition occurred (Figure 5 . 3 . 7 . b ) ; the sample s t ruc ture was seen to cons i s t of large blocks of CaC03 and some c a l c i n e d g r a i n s . The sample was completely ca l c ined when the c a l c i n a t i o n time was extended to 120 min. At t h i s stage, the p a r t i c l e becomes very porous and cons i s t s of p r a c t i c a l l y s p h e r i c a l grains having an average diameter of 0.75 um, and a t t a c h e d to each o t h e r i n a s er i e s of 2 to 3 grains (Figure 5.2.7.C.). Further heating the sample to a t o t a l c a l c i n a t i o n time of 220 B E T s u r f a c e area of ca l c i nes (m /g ) CO <u u co <D O cd 14-4 u D • W .-I > o B co <u C H o c o c o o CO e g cO C O CO U C M-i •H O C i H . O CO O 13 C iw co CO 4J O Q) co ,a .•+-1 vi m o W co C O - H (s jqg) S D 6 paonpojd U J O J J psAoaiaj s H 1° U O J J O D J J ozz 0 9 6 1 4 5 20KV X 5 ! 0 0 K ' ' 6 ! B u m F i g . 5.3.7 a S.E.M. photo of Cache Creek limestone, d = 0.68 mm P s F i g . 5.3.7 b S.E.M. photo of Cache Creek limestone, c a l c i n e d i n 40 minutes, d = 0.68 mm 0 6 0 9 5 4 S6KV X 8 ! 0 0 K * 'Oum F i g . 5.3.7 c S.E.M. photo of Cache Creek limestone, c a l c i n e d i n 120 minutes, d = 0.68 mm 0 0 0 9 5 8 £0KV X 8 . 9 0 K 3 . S u a F i g . 5.3.7 d S.E.M. photo of Cache Creek limestone, c a l c i n e d i n 300 minutes, d =0.68 mm 0 0 9 3 6 3 £0KV X 8 . 0 0 K 3 . S u m F i g . 5.3.7 e S.E.M. photo of Texada limestone, c a l c i n e d i n 300 minutes, d^ = 0.68 mm 0 0 1 4 6 1 £0KV XS\00K" " 3 ! 8 u m F i g . 5.3.7 f S.E.M. photo of dolomite c a l c i n e d , i n 90 minutes, d = 0.68 mm 6 0 1 4 6 2 20KV X 3 . 0 0 K 1 0 . 0 u m F i g . 5.3.7 g S.E.M. photo of dolomite, c a l c i n e d i n 90 minutes, d = 0.68 mm 0 0 0 1 4 9 20KV X 3 . 8 8 K 1 8 . S u m F i g . 5.3.7 h S.E.M. photo of dolomite, c a l c i n e d i n 90 minutes, d = 1.3 mm ho K5 F i g . 5.3.7 i S.E.M. photo of Fort MacMurray limestone, c a l c i n e d i n 120 min, d = 0.68 mm P s 9 0 0 3 6 2 20KV X 3 . 0 0 K 10.Gum F i g . 5.3.7 j S.E.M. photo of reacted c a l c i n e , T = 930°C, d = 0.68 mm NO 226 300 m i n . , r e s u l t s in the g r a i n d i a m e t e r s expanding to an average diameter of 1.25 um. The p a r t i c l e was s t i l l porous (Figure 5 . 3 . 7 . d . ) . The porous c h a r a c t e r i s t i c s of the three c a l c i n e d samples, g iven i n Table 5 .3 .1 , were determined using mercury porosimetry. The r e s u l t s from p o r o s i t y determinat ions , F igure 5 .3 .8 , show that although a l l three samples have near ly the same poros i ty of approximately 0.41 ml/g or 62%, the pore s i ze d i s t r i b u t i o n curve for sample CAL1, wi th a c a l c i n a t i o n time of 40 min, i s d i s t i n c t l y d i f f e r e n t from those of CALB, with a c a l c i n a t i o n time of 300 min and of QN3, with c a l c i n a t i o n time of 120 min, which are near ly i d e n t i c a l . Most of the pores i n CALB and QN3 sorbents are confined to diameters between 0.10 and 0.15 um, whi l e , the sample CAL1 has no narrow pore s i ze d i s t r i b u t i o n . Instead, i t s pore diameters are evenly d i s t r i b u t e d i n a large range from 0.04 um to 10 um, as seen i n curve F of Figure 5 .3 .8 . 5.3.2 DISCUSSIONS In the f u l l y c a l c i n e d s ta te , the amount of CaO i n the Cache Creek sample i s about 95% or h igher , and the s o l i d i s comprised of small g r a i n s . The combination of chemical a n a l y s i s , Table 5 .1 .1 , and the r e s u l t s of S . E . M . examination, F igure 5 . 3 . 7 . b , confirm that the sorbent CAL1 was underca lc ined . From the experimental r e s u l t s presented, i t i s evident that sorbent CAL1 was not very a c t i v e i n the s u l p h i d a t i o n r e a c t i o n , and thus not e f f e c t i v e i n removing sulphur from the produced gas. Therefore , the underca lc ina t ion of the sorbent CAL1 c o u l d be p r i n c i p a l l y responsible for the low r e a c t i v i t y wi th H2S. In Figure 5.3.4 the curve corresponding to the rate of sorbent conversion f o r Run Equivalent pore d iameter {JULm) 100 10 1.0 0.10 0.01 Absolute pressure ( ps i ) F i g . 5.3.8 Poros i ty determination of c a l c i n e s . A(DL1): d - .68 mm, t - 90 m i n . , S =15.2 m 2 / g . B(DL2) : d =1.0 mm, t=90min., S =10.36 m 2 / g . r ° p _ PS p ' ° C(D): d =1.3 mm, t = 90 min.,- S =5.06 m / g . D(QN3): d =.68 mm, t = 120 m i n . , S =5.02 m 2 / g . • O P^ P n E(CALB): d =.68 mm, t = 90 m i n . , S =4.40 ni /g . F(CAL1): d =,68 mm, t = 40 m i n , , S =3.38 m 2 / g . * IT P S P K> 228 40, which employed CAL1, i s unique, and r e f l e c t s the mechanism of t h i s under c a l c i n e d sorbent as fo l lows: at the g a s i f i c a t i o n temperature, the uncalc ined p o r t i o n i n the p a r t i c l e s of the CAL1 sorbent continued to be c a l c i n e d c r e a t i n g more surface area and pores for the su lph ida t ion r e a c t i o n , thus increas ing the sorbent conversion r a t e . When the rate of consumption of the surface area and of pore plugging due to reac t ion i s f a s t e r than that created for c a l c i n a t i o n , the rate of sorbent conversion drops , r e s u l t i n g i n a rate maximum. The c a l c i n a t i o n i s known to advance inwards from the outer surface [70]; and the newly created pores and surface areas i n CAL1 sorbent were q u i c k l y covered by s o l i d product , F igure 5 . 3 . 7 . J , causing the sharp decrease i n the rate of conversion a f t e r the maximum. Unl ike the CAL1 sorbent, the CALB and QN3 samples were prev ious ly f u l l y c a l c i n e d . Thus t h e i r r e a d i l y a v a i l a b l e pore and surface areas were gradua l ly consumed by r e a c t i o n , leading to rates which decrease raonotonically wi th t ime. The i n i t i a l a v a i l a b i l i t y of the r e a c t i o n surface of both CALB and QN3 sorbents r e s u l t s i n t h e i r having near ly the same i n i t i a l rate of sorbent convers ion, which i s about twice that of the CAL1 sorbent . When comparing CALB and QN3 c a l c i n e s i n terms of f r a c t i o n a l removal of H2S from the gas phase, and of sorbent convers ions , i t i s found that the QN3 i s more e f f e c t i v e than the CALB sorbent . In the l i g h t of the S . E . M . e x a m i n a t i o n and the s u r f a c e area measurements, i t may be concluded that s i n t e r i n g of sample CALB at h i g h t emperature f o r a p r o l o n g e d c a l c i n a t i o n t ime i n c r e a s e d i t s g r a i n s i z e , reducing the surface area for the s u l p h i d a t i o n r e a c t i o n , and thus decreas ing i t s 229 r e a c t i v i t y and capac i ty as a sulphur sorbent. The changes in porous s t r u c t u r e of CALB, from ca l c ine QN3, due to s i n t e r i n g were detected by b o t h s c a n n i n g e l e c t r o n m i c r o s c o p y and by B . E . T . s u r f a c e a r e a measurement; i n t e r e s t i n g l y , however, they c o u l d not be c l e a r l y i d e n t i f i e d by mercury porosimetry. This may be one of many weaknesses of the mercury penetrat ion technique, as discussed in Sect ion 2 .6 .2 . In F igure 5 .3 .5 , the i n i t i a l s ec t ion i n which the f r a c t i o n a l H2S removal i s 1.0 probably r e f l e c t s a rate of sulphur capture which i s c o n t r o l l e d by the rate of H2S re lease from the coke. Fol lowing t h i s , the f r a c t i o n a l removal drops i n i t i a l l y r a p i d l y , and then as sorbent becomes more saturated with the s o l i d product , l e s ser amounts of H2S react with CaO per u n i t time. The two curves represent ing f r a c t i o n a l H2S removal a f t er 6 hours of r e a c t i o n vs time and the s p e c i f i c area vs time are almost i d e n t i c a l i n shape when p l o t t e d i n the same graph, Figure 5 .3 .6 , which suggests that the f r a c t i o n a l sulphur removal from the product gas i s near ly l i n e a r l y r e l a t e d to the s p e c i f i c area of sorbent. Th i s near l i n e a r r e l a t i o n s h i p holds f o r the e n t i r e extent of r e a c t i o n , as demonstrated i n F i g u r e 5 .3 .9 . A c o m p a r i s o n of the l e v e l s of H2S emissions i n Run 69, using c a l c i n e d limestone QN3, in Figure 5.3.1 to that of Run 71, f or ca l c ined dolomite QN1, i n Figure 5 .3 .2 , ind ica tes that dolomite i s more e f f e c t i v e i n removing sulphur than limestone for the same Ca/S r a t i o . Further d i s c u s s i o n of the e f fec t of the type of carbonate rock i s given in Sect ion 5.7. 230 FRACT. H2S REMOVAL VS. AREA AT VARIOUS REACTION TIME 0.8 0.6 0.4 0.2 3.5 4.5 5.5 SPECIFIC SURFACE A R E A , m / g F i g . 5.3.9 F r a c t i o n a l sulphur removal vs. surface area for runs with sorbent at d i f f e r e n t c a l c i n a t i o n conditions. 231 The l i n e a r sec t ion at low conversion followed by a f a l l i n g rate s e c t i o n at h i g h c o n v e r s i o n i n the Xgg- t curves , Figure 5 .3 .3 , i s i n d i c a t i v e of the f a m i l i a r reac t ion c o n t r o l l i n g i n the e a r l y stage of convers ion going to a much slower regime c o n t r o l l e d by d i f f u s i o n at h igher convers ion . More extensive study in to these stages i s provided in the fo l lowing chapter . The experimental r e s u l t s obtained in t h i s present study are in bas ic agreement wi th what i s publ ished i n the l i t e r a t u r e . Previous s tudies postulated that limestone i s low i n p o r o s i t y and in surface area and thus not reac t ive with hydrogen sulphide unless i t has been f u l l y c a l c i n e d [ 6 8 , 6 9 ] . However, when the c a l c i n a t i o n continues f o r an extended time, the r e s u l t i n g c a l c i n e becomes s i n t e r e d , which accounts for the increase i n g r a i n s i ze and for the reduct ion i n surface area tha t , i n t u r n , adversely a f fec ts the r e a c t i v i t y of the c a l c i n e d sorbent [1 ,27,142] . 5 . 4 EFFECTS OF Ca/S RATIOS For a given throughput of o i l sand coke, i . e . a g iven quant i ty of s o l i d su lphur , the amount of sulphur removed by the c a l c i n e d sorbent depends upon the quant i ty of the added sorbent, i . e . the Ca/S r a t i o . The higher the Ca/S r a t i o , the more sulphur i s removed from the product gas. Neverthe less , the excess of sorbent over what needed to obta in a c lean f u e l gas w i l l increase both the operat ing and c a p i t a l costs of the p lant to handle t h i s excessive amount of sorbent. 232 5.4.1 EXPERIMENTAL RESULTS T h i s s e c t i o n p r e s e n t s the e x p e r i m e n t a l r e s u l t s of runs to i n v e s t i g a t e the e f f ec t of the Ca/S r a t i o s on the d e s u l p h u r i z a t i o n p r o c e s s . In these runs both c a l c i n e d dolomite , QN1 and c a l c i n e d l imestone, QN3 were employed. Sample QNl was generated from dolomite i n a c a l c i n a t i o n time of 120 min; i t has an average p a r t i c l e s i z e of 0.68 mm. The s p e c i f i c a t i o n of sample QN3 i s given i n Table 5 .3 .1 . For dolomite , the e f fec t of Ca/S r a t i o on the l e v e l of H2S emission i s shown i n Figure 5 .4 .1 . It i s apparent from t h i s f igure that the l e v e l of H2S decreases with increas ing ca l c ium-to - su lphur r a t i o s and t h a t i n the f i r s t three hours of r e a c t i o n the l e v e l of H2S i n the produced gas was reduced s u b s t a n t i a l l y , f o r a l l Ca/S r a t i o s , from over 1100 ppm to less than 400 ppm. As the g a s i f i c a t i o n time was prolonged, t h e s o r b e n t i n the C a / S = 0.5 run was d e p l e t e d c a u s i n g the H2S concentrat ion i n the gas to s i g n i f i c a n t l y increase . The concentrat ions of H2S i n runs with Ca/S=0.9, 1.2 and 2.0 , at the end of 6 hours, a lso increase but to a l e s ser degree. An H2S concentrat ion of 600 ppm was measured at the end of 6 hours of r eac t ion f o r a Ca/S=2.0. I t i s important to note that the gas leav ing the g a s i f i e r was e s s e n t i a l l y s u l p h u r - f r e e f o r the f i r s t 140 min when a Ca/S=2.0 was u t i l i z e d . This su lphur - f ree per iod was shorter f o r smal ler Ca/S r a t i o s . S i m i l a r r e s u l t s were observed for runs us ing c a l c i n e d l imestone, at Ca/S=0.5 and 2.0 (Figure 5 .4 .2 ) . The e f f ec t of the Ca/S r a t i o on the f r a c t i o n a l removal of sulphur was s tudied and the experimental r e s u l t s f o r var ious r e a c t i o n times are presented i n Figure 5 .4 .3 . The f r a c t i o n a l H2S removal increases with 233 H2S CONC. VS. REACTION TIME  FOR RUNS AT DIFF. Ca/S RATIOS 1600 a, > G O o (—1 E-O o o CO w 1200 800 400 0 LEGEND •=RUN #47 NO SORBENT 0=RUN #72 DOLO. Ca/S=0.5 •=RUN #75 DOLO. Ca/S=0.9 ~ = RUN #70 DOLO. Ca/S=1.2 = RUN #71 DOLO. Ca/S=2.0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.4.1 H^S concentrations for runs using sorbent of d i f f e r e n t Ca/S molar r a t i o . Runs with dolomite. 234 H2S CONC. VS. REACTION TIME  FOR RUNS AT DIFF. Ca/S RATIOS 1600 n 1200 800 400 0 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.4.2 H^S concentration for runs using sorbent of d i f f e r e n t Ca/S molar r a t i o . Runs with limestone. 235 OVERALL FRACTIONAL H2S REMOVAL EXPERIMENTAL DATA 0 * ; : ; : : : : ; : 1 0 0.5 1 1.5 2 2.5 MOLAR C a / S RATIOS F i g . 5.4.3 E f f e c t s of Ca/S molar r a t i o on f r a c t i o n a l sulphur removal. Runs with dolomite. -increas ing Ca/S r a t i o or with decreasing re t en t ion t ime. This r e f l e c t s e a r l i e r suggestions, i n which the l e v e l s of H2S i n the t rea ted gas increase due to the dep le t ion of sorbent for long r e t e n t i o n times or too small Ca/S r a t i o . For a Ca/S=2.0, c lose to 80% of H2S produced by the steam g a s i f i c a t i o n of o i l sand coke was captured at the end of 6 hours. However, only 50% of the produced H2S was captured i n the run wi th a Ca/S=0.5 over the same per iod of t ime. I t i s seen that in the f i r s t hour of r e a c t i o n , the f r a c t i o n a l removal of H2S from the gas stream was 1.0 f o r a l l Ca/S r a t i o s . Sorbent conversions of the samples employed in runs of var ious Ca/S r a t i o s are p l o t t e d against r e a c t i o n time in Figure 5 .4 .4 . T h i s f i gure revea l s that the Xgg-t r e l a t i o n s h i p i s l i n e a r i n the f i r s t stage of convers ion and that the sorbent conversion increases wi th decreasing Ca/S r a t i o . For the run with Ca/S=2.0 the sorbent convers ion a f t e r 6 hours of s u l p h i d a t i o n was less than 20%; while the conversion was about 46% i n the sorbent f o r the run at Ca/S=0.5. The r a t e c u r v e s , i . e . (dXs]$/ d t ) v s Xgjj c u r v e s , of s o r b e n t convers ion are shown i n Figure 5 .4 .5 . It can be seen from t h i s f igure that both the i n i t i a l rate and the u l t imate sorbent conversion decrease wi th increas ing Ca/S r a t i o s . The i n i t i a l rate of sorbent conversion f o r the run wi th Ca/S=0.5 was three times higher than that for run with Ca/S=2.0. The u l t imate conversion i s the extension of the rate curve u n t i l dXg B /dt=0. To c o u n t e r c h e c k the va lues of sorbent convers ions , the weight percent of sulphur i n the spent sorbent was determined by two d i f f e r e n t methods: by convert ing the sulphur conversion in to a mass bas is i . e . by 237 SORBENT CONV. VS. REACTION TIME  EXPERIMENTAL DATA AT DIFF. Ca/S G O 4 0 C D o 20 0 LEGEND • =RUN #72 DOL. Ca/S=0.5 0=RUN #75 DOL. Ca/S=0.9 •= RUN #70 DOL. Ca/S=1.2 • =RUN #7J DOL. Ca/S=2.0 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5 . 4 . 4 E f f e c t of Ca/S molar r a t i o s on sorbent conversion. 238 SORBENT CONVERSION VS. dX S B /dt EXPERIMENTAL DATA AT DIFF. Ca/S 30 20 10 0 LEGEND • =RUN #72 DOL. Ca/S=0.5 0=RUN #75 DOL. Ca/S=0.9 • =RUN #70 DOL. Ca/S=1.2 ~=RUN #71 DOL. Ca/S=2.0 0 20 40 SORBENT CONVERSION CaO TO CaS,% F i g . 5 .4 .5 Rate of sorbent conversion for runs at di f f e r e n t Ca/S molar r a t i o s . 239 c a l c u l a t i o n from the cumulative gas a n a l y s i s , and by f i r s t separat ing the spent sorbent from the bed residue and then d i r e c t l y measuring the sulphur content in the reacted sorbent, us ing a s o l i d sulphur analyzer . The r e s u l t s obtained from the two methods of determination are given in F igure 5 .4 .6 . The amounts of sulphur measured by the sulphur analyzer are somewhat lower than that determined us ing the sulphur conversion b a s i s , with a maximum discrepancy of 21%. 5.4.2 DISCUSSION F r a c t i o n a l removal of H2S appears to f i r s t be h igh ly s e n s i t i v e to a change i n the ca lc ium-to - su lphur molar r a t i o at low Ca/S values (<1.0), and then be s l i g h t l y i n f l u e n c e d by t h i s r a t i o at h i g h e r v a l u e s (Figure 5 .4 .3 ) . Using t h i s f i g u r e , several operat ing combinations of Ca/S and sorbent re t en t ion time can be obtained for a given f r a c t i o n a l H2S removal. For example, a des ired f r a c t i o n a l removal of sulphur from the gas of 80% can be achieved when operating with Ca/S=0.6 for 2 hours or w i t h C a / S = 1.20 f o r 4 h o u r s , p r o v i d e d the carbon conversion i s s u f f i c i e n t at the end of these re tent ion t imes. A d d i t i o n a l curves for smal ler re t en t ion time i n t e r v a l s can be constructed f o r t h i s p l o t from the data of Figure 5 .4 .1 . The v a l u e s of C a / S r a t i o s and t h e i r corresponding f r a c t i o n a l sulphur removal presented i n t h i s research are w i th in the range of publ i shed r e s u l t s [110,64], Sorbent conversion i n the su lph ida t ion of CaO i s normally very low. Sorbent conversion reached -17% at the end of 6 hours , i n the run u t i l i z i n g a Ca/S r a t i o of 2.0 and achieving a f r a c t i o n a l sulphur removal 240 WEIGHT OF S IN SPENT SORBENT  FOR RUNS AT DIFF. Ca/S RATIOS 15 I : : : : : : : : : F i g . 5.4.6 Weight of sulphur i n spent sorbent determined by gas and by weight analyses. 241 of approximately 90%, Figure 5 .4 .3 . Moreover, i t i s determined from Figure 5.4.5 that the reac t ion between CaO p a r t i c l e s and H2S to form CaS i s i n i t i a l l y very fas t but r a p i d l y slows down. At Ca/S=0.5 and a f t e r 6 hours of s u l p h i d a t i o n , the c a l c i n e has become p r a c t i c a l l y unreac t ive , i . e . ( d X g j j / d t ) - » 0 , while less than 47% of the o r i g i n a l CaO has been converted to CaS. 5.5 EFFECTS OF SORBENT PARTICLE SIZE 5.5.1 EXPERIMENTAL RESULTS The r o l e of sorbent p a r t i c l e s i z e i n the i n - s i t u desu lphur iza t ion of o i l sand coke was inves t iga ted . In t h i s s e c t i o n , dolomite w i t h average p a r t i c l e s i zes of 1.30 mm, 1.0 mm and 0.68 mm was c a l c i n e d for 90 minutes at 9 5 0 ° C and u t i l i z e d , at a Ca/S molar r a t i o of 2 .0 , to capture sulphur from H2S i n the produced gas. The e f f e c t of sorbent p a r t i c l e s i z e on the H2S emissions i s shown i n Figure 5 . 5 . 1 . The H2S concentrat ion i n the leaving gas decreases wi th decreas ing p a r t i c l e s i z e . A f t e r s i x hours, the H2S l e v e l i n the treated gas f o r the run employing a sorbent p a r t i c l e s i z e of 1.30 mm was not s u b s t a n t i a l l y d i f f e r e n t from those of runs without sorbent a d d i t i o n ; wi th an average concentrat ion of about 950 ppm compared to 1100 ppm without sorbent . However, the H2S average concentrat ion was reduced to about 340 ppm i n the run with 0.68 mm sorbent. The f r a c t i o n of sulphur removed from the produced gas was very s e n s i t i v e to the s i ze of ca l c ine employed (Figure 5 .5 .2 ) . For the same s u l p h i d a t i o n time of 6 hours, a sulphur removal of 75% was obtained in a 242 H2S CONC. VS. REACTION TIME FOR RUNS AT DIFF. CALCINE Dp 1600 o 1200 LEGEND • =RUN #47 NO SORBENT •=RUN #76 Dp=-68 mm • = RUN #77 Dp=1.0 mm • = RUN #78 Dp=1.3 mm 800 400 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.5.1 E f f e c t of sorbent p a r t i c l e s i z e on H^S concentration i n the produced gas. 243 FRACTIONAL REMOVAL OF H2S SORBENT OF DIFFERENT SIZE LEGEND 0=RUN #76 Dp=.68mm 0=RUN #77 Dp = 1.0mm • =RUN #78Dp = 1.3mm 0 1 : : ; ; : : : : ; ' 0 60 120 180 240 300 360 TOTAL REACTION TIME, MIN. F i g . 5.5.2 F r a c t i o n a l H^S removal for runs with d i f f e r e n t sorbent p a r t i c l e s i z e s . 244 run using c a l c i n e d dolomite having an average s ize of 0.68 mm, while only 28% removal was achieved i n the run with 1.30 mm sorbent . The shapes of these curves were discussed i n Sect ion 5 .3 .2 , where i t was p o s t u l a t e d that the o v e r a l l process was c o n t r o l l e d i n i t i a l l y by H2S re lease , fol lowed by a reac t ion regime with a f a l l i n g rate due to a d i f f u s i o n a l e f f e c t . Sorbent conversions were found to increase with decreasing p a r t i c l e I s i ze as shown i n Figure 5 .5 .3 . The o v e r a l l convers ions , a f t e r 6 hours of s u l p h i d a t i o n , were 6.2%, 9.7% and 16.2% for runs employing 1.30 mm, 1.0 mm and 0.68 mm c a l c i n e d dolomite, r e s p e c t i v e l y . The curves i n t h i s graph were i n i t i a l l y l i n e a r , possessing h igh slopes which p r o g r e s s i v e l y decreased as the r e a c t i o n proceeded. The rates at which CaO i s converted to CaS, obtained by taking the s l o p e of the X g g - t curve at v a r i o u s p o i n t s , are c o n s t r u c t e d i n Figure 5 .5 .4 . In t h i s f i g u r e , the i n i t i a l conversion rate of 0.68 mm s o r b e n t i s s l i g h t l y h i g h e r than the values for 1.0 mm and 1.3 mm sorbents , which are near ly the same. The B . E . T . surface areas of the three c a l c i n e s of d i f f e r e n t s i zes ind ica ted that the surface area was l a r g e r for smal ler c a l c i n e p a r t i c l e s i z e . C a l c i n e DLI (dp s=0.68 mm), DL2 (dp s =1.0 mm) and D (dp s=1.3 mm) have B . E . T . s u r f a c e s of 15.18 m 2 / g , 10.38 m 2 / g and 5.06 m 2 / g , r e s p e c t i v e l y . The inf luence of these surface areas on the conversion of sorbent i n the s u l p h i d a t i o n reac t ion i s shown i n Figure 5 .5 .5 . Th i s f i gure emphasizes the strong ro l e of surface on sorbent c o n v e r s i o n , e s p e c i a l l y at high values of surface area, at any extent of r e a c t i o n . A f t e r 6 hours of r e a c t i o n , the sorbent conversion for the runs using the 245 cd o o o •—I 00 > o o E-1 o 00 SORBENT CONV. VS. REACTION TIME DATA FOR DIFF. SORBENT Dp 20 15 10 LEGEND • =RUN #76 DOL. Dp=0.68mm 0=RUN #77 D0L. Dp=1.00mm •=RUN #78 DOL. Dp=1.30mm 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5 .5 .3 Sorbent convers ion i n runs of d i f f e r e n t sorbent s i z e s . 246 SORBENT CONVERSION VS. dX.. /dt.  DATA FOR DIFF. CALCINE Dp 0 5 10 15 SORBENT CONVERSION CaO TO CaS,% F i g . 5 .5 .4 Rate of sorbent convers ion i n runs u s ing c a l c i n e of d i f f e r e n t s i z e s . 247 SORBENT CONV. VS. SURFACE AREA AT VARIOUS REACTION TIME 20 LEGEND 0=3 HRS OF REACTION • =4 HRS OF REACTION V=5 HRS OF REACTION • =6 HRS OF REACTION 15 O •—1 m w > o 10 o E - 1 P3 ° 5 in 0 0 5 10 15 20 SPECIFIC SURFACE A R E A , m 2 / g F i g . 5.5.5 E f f e c t of surface areas of calcines of d i f f e r e n t sorbent s i z e s . i n runs 248 c a l c i n e possessing a s p e c i f i c area of 15.18 m^/g was 16.2%; whi le , i t was 6.2% f o r the run with the c a l c i n e having only 5.06 m^/g. Mercury p o r o s i t y determinations were done by the three c a l c i n e d dolomite samples DL1, DL2 and D. The r e s u l t s of these measurements are included in F igure 5 .3 .8 . Pores i n a l l three c a l c i n e s have a very narrow d i s t r i b u t i o n , and the most probable pore s i ze increases with increas ing p a r t i c l e s i z e . The most probable pore s i zes for D L l , DL2 and D r e s p e c t i v e l y are 0.03 pm, 0.04 um and 0.05 um, while t h e i r pore volumes are almost the same, 0.41 cnrVg or ~64% p o r o s i t y . 5.5.2 DISCUSSION From the present experimental r e s u l t s , i t i s determined that the s u l p h i d a t i o n r e a c t i o n i s g r e a t l y affected by the sorbent p a r t i c l e s i z e s . Even though the sorbent D (dp s =1.3 mm) was c a l c i n e d under the same condi t ions as the DLl (dp s=0.68 mm) sample, i t exh ib i t ed poor r e a c t i v i t y wi th H2S and a very poor sorbent u t i l i z a t i o n , r e s u l t i n g i n a high l e v e l of hydrogen sulphide i n the treated gas (F igure 5 . 5 . 1 ) . T h i s low q u a l i t y of sorbent D i s c l e a r l y demonstrated in i t s X g B - t curve, Figure 5 .5 .3 , where the l i n e a r sec t ion i n the e a r l y stage of convers ion, which belongs to the r e a c t i o n c o n t r o l l e d regime, was very short and fol lowed by a near ly l e v e l s ec t ion of the curve. Furthermore, the very low slopes i n the X s B - t curves f o r runs with dp s =1.3 mm and 1.0 mm imply that the r e a c t i o n i s near ly ha l ted at very low sorbent conversion due to strong d i f f u s i o n a l r e s i s tances . The d i f f u s i o n a l res i s tances may cons i s t of both pore d i f f u s i o n and d i f f u s i o n through CaS p r o d u c t l a y e r s . The former r e s i s t a n c e was 249 experimental ly observed by severa l authors [60]; however, the l a t t e r i s known to p l a y a dominant r o l e i n the l a t t e r stage of convers ion, e s p e c i a l l y i n the CaO+H2S and Ca0+S02 r eac t ions [27,26,21] . The conversion at which the s o - c a l l e d "second stage" begins was found to be very incons i s t ent among past s tudies and var i ed with the experimental c o n d i t i o n s , s t a r t i n g from as e a r l y as 5% of convers ion. I t was determined that the sulphided product, CaS, covered the r e a c t i o n surface , bridged the grains and p lugged the pores i n the reacted sorbents (Figure 5 . 3 . 7 . J ) . Such depos i t ion of CaS may force the reac t ing gas H2S to f i r s t d i f fuse through the sulphided layer in order to react wi th the unconverted sorbent. T h i s would, of course, lower the rate of sorbent convers ion . The reac t ion between CaO and H2S to form CaS i s very rap id [34], causing the p r o d u c t l a y e r to c o n t r o l the s u l p h i d a t i o n r e a c t i o n at very low sorbent convers ion. This argument i s f u r t h e r j u s t i f i e d by the experimental fac t s that the i n i t i a l rate for sorbents of a l l three s i zes 0.68 mm, 1.0 mm and 1.3 mm were near ly equal , F igure 5 .5 .4 , but, the r a t e s f o r s o r b e n t s p o s s e s s i n g lower surface areas, sorbents DL2 (10.38 m^/g) and D (5.06 m^/g), dropped sharply at very small values of sorbent convers ion. From t h i s rate p l o t , i t i s evident that the rates for runs using large p a r t i c l e s i ze sorbents approach zero , i . e . r e a c t i o n e s s e n t i a l l y ceased, at less than 10% conversion for the run with 1.0 mm d i a . ca l c ine and at 6% for the run wi th 1.3 mm d i a . p a r t i c l e s . A higher sorbent conversion of about 22% was obtained for the run with 0.68 mm ca lc ined dolomite. The porous s tructure may be responsible for the lower r e a c t i v i t i e s of the DL2 and D c a l c i n e d samples with H2S. From the r e s u l t s of the 250 mercury p e n e t r a t i o n , F i g u r e 4 . 1 . 1 6 , and the S . E . M . e x a m i n a t i o n s , F igure 5 . 3 . 7 . f , g, h and DL1, DL2 and D c a l c i n e s , i t i s apparent that c a l c i n e s with smal ler p a r t i c l e s i z e s have s m a l l e r g r a i n s i z e s and smal ler most probable pore s i z e s , which i n turn r e s u l t s i n a l arger r e a c t i o n surface area and, consequently, higher f r a c t i o n a l H 2 S removal and be t ter sorbent u t i l i z a t i o n . 5.6 EFFECT OF TYPES AND GEOLOGICAL ORIGINS Ca lc ines of d i f f e r e n t types or of d i f f e r e n t geo log i ca l o r i g i n s may behave d i f f e r e n t l y as sorbents to remove sulphur [5 ,8 ,18,27,132,142] . Experimental r e s u l t s from runs with c a l c i n e s of d i f f e r e n t types and of d i f f e r e n t o r i g i n s are presented i n t h i s s e c t i o n . 5.6.1 CALCINES OF DIFFERENT ORIGINS Samples of limestone c o l l e c t e d from four d i f f e r e n t reg ions , Cache Creek, B . C . (CALB), Texada I s land , B . C . (CALA), For t McMurray, A l b e r t a (FM) and Havelock, N . B . (HL), were c a l c i n e d for 300 min, then analyzed f o r major and trace components. The r e s u l t s of these analyses were p r e v i o u s l y shown i n Table 5 .1 .1 . CALA and CALB samples have very s i m i l a r chemical compositions; they are mainly CaO with small amounts of s i l i c a and alumina. FM and HL samples have lower content of CaO, 87.0% and 63%, r e s p e c t i v e l y , and are very high i n s i l i c a and in alumina. S . E . M . e x a m i n a t i o n of CALA, F i g u r e 5 .3 .7 .e , and CALB, Figure 5 . 3 . 7 . d , reveals that both ca l c ines are p o r o u s , and the s t r u c t u r e appears to be unconsol idated and comprised of grains of approximately the same s izes of about 1.25 um. T h e i r BET s p e c i f i c surface areas 251 measured using N 2 adsorpt ion for CALA and CALB were r e s p e c t i v e l y 4.62 m^/g and 4.40 m^/g. On the other hand, the FM c a l c i n e i s composed of lumps of smal ler g r a i n s , F igure 5 . 3 . 7 . 1 , which i n t e r l o c k t i g h t l y to each o ther , and the i m p u r i t i e s were scat tered on the surface of these lumps. The measured s p e c i f i c area of F . M . c a l c i n e was 6.22 m^/g. The H2S l e v e l s i n the treated gas, r e s u l t i n g from runs conducted w i t h c a l c i n e s of d i f f e r e n t sources , having dp s =0.68 mm, and wi th a Ca/S=2.0, are presented i n Figure 5 .6 .1 . The F . M . c a l c i n e was expected to be more e f f e c t i v e i n sulphur removal s ince i t has a h igher s p e c i f i c surface area and smal ler g r a i n s i ze than e i t h e r CALA and CALB. However, i t was observed that the l e v e l of H 2 S i n the t rea ted gas f o r the run employing FM i s j u s t s l i g h t l y lower than i n runs wi th CALA and CALB, which were near ly the same. Thi s could be the r e s u l t of an i n h i b i t o r y e f f e c t due to high l e v e l of impur i t i e s or to the rap id d e p o s i t i o n of s o l i d product onto the unique t i g h t l y i n t e r l o c k e d porous s t r u c t u r e of the F . M . c a l c i n e . 5.6.2 CALCINES OF DIFFERENT TYPES The r e a c t i v i t i e s and c a p a c i t y of l imestone and dolomite wi th hydrogen s u l p h i d e have been the s u b j e c t o f s e v e r a l p a s t s t u d i e s [1 ,141,56,131] . E a r l i e r workers, however, c a r r i e d out the s u l p h i d a t i o n r e a c t i o n independent ly by employing a stream of H 2 S gas. In t h i s research , the i n - s i t u reac t ion between H 2 S i n the gas mixture , produced by the steam g a s i f i c a t i o n of o i l sand c o k e , and e i t h e r c a l c i n e d l i m e s t o n e or d o l o m i t e i n the bed was i n v e s t i g a t e d . The combined 252 H2S CONC. VS. REACTION TIME FOR R U N S WITH DIFF. SORBENTS 1600 1200 800 400 0 0 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.6.1 E f f e c t of sorbent of d i f f e r e n t geological origins on H^S concentration i n the produced gas. 253 g a s i f i c a t i o n - d e s u l p h u r i z a t i o n experiments d e s c r i b e d i n t h i s s e c t i o n employ ing c a l c i n e d l imestone , QN3, and c a l c i n e d dolomite , QN1, the products ofthermal decomposition of Cache Creek limestone and dolomite i n 120 min, at var ious Ca/S r a t i o s . The c o n c e n t r a t i o n s of H2S i n the t r e a t e d gas for runs using d i f f e r e n t types of ca l c i nes are p l o t t e d i n Figure 5.6.2 as a funct ion of r e a c t i o n t ime. For the same Ca/S va lue , the concentrat ion of H 2 S i n the gas i s always lower i n runs u t i l i z i n g c a l c i n e d dolomite , which leads to the be t ter f r a c t i o n a l sulphur removal from the gas phase for runs with QN1. A f t e r a re tent ion time of 6 hours, o v e r a l l f r a c t i o n a l removal of 71% and -78% were achieved for runs using QN3 and QN1, r e s p e c t i v e l y , (Figure 5 .6 .3 ) . The present study shows tha t , in genera l , c a l c i n e d dolomite i s more capable of removing sulphur than c a l c i n e d l imestone, a l l e l se be ing equal . In add i t i on to the arguments s tated i n Sect ions 2.5.4 and 2.5.5 on the ro l e of MgO c r y s t a l l i t e s rendering the porous s t r u c t u r e to f a c i l i t a t e the su lph ida t ion r e a c t i o n , the d i f f erence i n capac i ty to desulphurize w i l l be re l a t ed to the open s o l i d s t ruc ture of the two c a l c i n e s . The S . E . M . examinations of QN3, F igure 5 . 3 . 7 . d , c and of QN1, Figure 5 . 3 . 7 . J reveal that the g r a i n s i ze of the c a l c i n e d dolomite was s m a l l e r than t h a t of ca l c ined limestone under the same c a l c i n a t i o n cond i t i ons , which leads to the greater surface area f o r the former. The average surface area of the c a l c i n e d dolomite at dp s=0.68 mm was about 15 ra^/g, while ca lc ined limestone possessed a smal ler surface area of approximately 5 m^/g. 254 H2S CONC. VS. REACTION TIME R U N S AT DIFF. SORBENT T Y P E S 1600 120 240 TOTAL REACTION TIME,MIN. 360 F i g . 5.6.2 E f f e c t of types of sorbent on H^S concentration. 255 OVERALL FRACTIONAL H2S REMOVAL  E X P E R I M E N T A L DATA 1 I : : : : : : : : : 1 . J 0 0.5 1 1.5 2 2.5 MOLAR Ca/S RATIOS F i g . 5.6.3 F r a c t i o n a l sulphur removal i n runs with d i f f e r e n t types of c a l c i n e s . 256 In genera l , the s p e c i f i c surface area of a c a l c i n e d dolomite sample i s much h i g h e r than that of a ca l c ined limestone of the same s i ze andca lc ina t ion c o n d i t i o n s . Comparisons between the c a l c i n e s employed i n t h i s study are given i n Table 5 .6 .1 . Table 5.6.1 COMPARISONS BETWEEN CALCINES EMPLOYED IN THIS STUDY Run# M a t e r i a l Ca lc ine Average C a l c i n a t i o n Gra in S p e c i f i c Xgj},% Code S ize , mm time, min S i ze , um area , m 2 / g  76 Dolomite DL1 0.68 77 Dolomite DL2 1.00 78 Dolomite D 1.30 90 90 90 0.6 1.6 2.5 15.18 10.38 5.06 16.25 9.70 6.19 40 Limestone* CAL1 0.68 54 Limestone* CAL2 0.68 69 Limestone* CAL3 0.68 40 300 120 1.25 0.75 3.82 12.10 4.40 14.10 5.02 15.80 71 Dolomite QN1 0.68 70 Dolomite QN1 0.68 75 Dolomite QN1 0.68 72 Dolomite QN1 0.68 120 120 120 120 16.82 16.82 16.82 16.82 17.11 25.24 30.96 45.58 55 Limestone** CALA 0.68 300 1.25 4.62 (*) Cache Creek (**) Texada For a g iven type of sorbent, i . e . c a l c i n e d dolomite or c a l c i n e d l imestone, the s p e c i f i c surface area has a strong e f fec t on the o v e r a l l sorbent convers ion . Figure 5.6.4 i l l u s t r a t e s the sorbent conversion of c a l c i n e d limestone and of ca l c ined dolomite, at the end of 6 hours of r e a c t i o n , as a funct ion of s p e c i f i c area. It i s seen tha t , w i th in a given type of sorbent, the c a l c i n e which possessed h i g h e r s p e c i f i c surface area achieved higher sorbent convers ion. 257 SORBENT CONV. VS. SURFACE AREA R U N S WITH DIFF. SORBENT T Y P E S LEGEND ffi= RUNS WITH LIMESTONE • = RUNS WITH DOLOMITE 0 5 10 15 20 SPECIFIC SURFACE A R E A , m 2 / g F i g . 5.6.4 Sorbent conversion vs. s p e c i f i c surface area for runs with d i f f e r e n t types of c a l c i n e . 258 The surface areas computed using equation (85) and equation (86) which are based on most probab le pore diameter and on g r a i n s i ze r e s p e c t i v e l y , are , on average, 6 times higher and lower, r e s p e c t i v e l y , than that obtained experimental ly by t h i s present study. Thi s may be due to s i m p l i f i e d assumptions of c a l c i n e s c o n s i s t i n g of c y l i n d r i c a l , n o n - i n t e r s e c t i n g pores and unconsol idated s p h e r i c a l gra ins made when d e r i v i n g these equations. I t i s worth not ing that for the same Ca/S r a t i o , the weight of c a l c i n e d dolomite suppl ied i s higher than that of c a l c i n e d l imestone. A higher mass loading therefore i s required i f dolomite i s chosen as a sorbent . In the present research, the average weight r a t i o of c a l c i n e d dolomite to c a l c i n e d limestone for the same molar Ca/S r a t i o was about 1.60. 5 . 7 SUMMARY S i g n i f i c a n t f ind ings that were der ived from the experimental work on the i n - s i t u d e s u l p h u r i z a t i o n of the gas produced i n the steam g a s i f i c a t i o n of o i l sand coke, using c a l c i n e d sorbents , are summarized below 1. The add i t i on of c a l c i n e d sorbent, up to a Ca/S=2.0, has l i t t l e e f f ec t on carbon conversion of o i l sand coke. 2. a) Limestone and dolomite do not react with hydrogen sulphide unless theydave been c a l c i n e d , b) C a l c i n a t i o n condit ions af fect the s t r u c t u r a l c h a r a c t e r i s t i c s such as g r a i n s i z e , pore s i ze d i s t r i b u t i o n , p o r o s i t y and 259 surface area of the r e s u l t i n g sorbents , and the performance of the sorbent i s inf luenced by these s t r u c t u r a l parameters, c) Under or o v e r - c a l c i n a t i o n of a c a r b o n a t e rock adversely a f f ec t s the r e a c t i v i t y and capac i ty of the r e s u l t i n g c a l c i n e . 3. Ca lc ined dolomite i s more e f f e c t i v e i n removing hydrogen sulphide than c a l c i n e d l imestone, at f i xed Ca/S molar r a t i o s , a l l e l se being e q u a l ; however, t h e i r i n i t i a l rates of sorbent conversion are almost the same. 4. As a Ca/S=2.0, approximately 100% sulphur removal was achieved i n the f i r s t 2 hours of r e a c t i o n . 5. C a l c i n e d sorbents of l arger p a r t i c l e s i ze possess l a r g e r gra ins and pores and thus, have lower r e a c t i o n surface area. 6. For a given type of ca l c ined sorbent, surface area has a strong e f f ec t on sorbent convers ion. The s p e c i f i c a r e a of c a l c i n e d dolomite i s approximately three times that of c a l c i n e d l imestone, for the same c a l c i n a t i o n c o n d i t i o n s . 260 CHAPTER 6 TESTING KINETIC MODELS A number of k i n e t i c models have been proposed in the l i t e r a t u r e for g a s - s o l i d reac t ions . The models se lec ted for the g a s i f i c a t i o n and the s u l p h i d a t i o n processes were descr ibed in Chapter 2. In t h i s chapter carbon conversion data are checked against the p r e d i c t i o n of three d i f f e r e n t k i n e t i c models from the l i t e r a t u r e - The Random Pore Model [18,19,17], the Random C a p i l l a r y Model [66] and the Modif ied Volumetric Model [104] , w h i l e the r e s u l t s e x p e r i m e n t a l l y obtained i n the s u l p h i d a t i o n r e a c t i o n between CaO and H2S gas are f i t t e d by the G r a i n Model [176,179] and C o n t i n u o u s F i r s t - O r d e r M o d e l [60,101,152,137,149]. 6.1 TESTING MODELS USING GASIFICATION EXPERIMENTAL RESULTS: 6.1.1 RANDOM CAPILLARY MODEL (RCM) The RCM i s a m a t h e m a t i c a l m o d e l e m p l o y i n g s t a t i s t i c a l manipulat ions . It assumes that the s o l i d possesses a broad pore s ize d i s t r i b u t i o n , and t h a t pores i n t e r s e c t and coalesce randomly. A r e l a t i o n between the rate and conversion was proposed by Gavalas [66] f o r the RCM: dX o n 2 i / o 261 2 2 o r : X C = 1 - exp [ 2 T T ( B Q V t + 2B x vt ) ] (147) which can be rearranged to y i e l d : - L n [ l / Q - X c ) ] = Mt + N t 2 (148) where: M = 2irB 0 v 2 and N = 4 n B ] V B o l Bi are moments of p r o b a b i l i t y dens i ty X.(r) i n Equation (53) and Equat ion (54) and v i s r eac t ion v e l o c i t y . Equat ion (147) impl ies that r e a c t i v i t y data can be used to determine B 0 v 2 and Bjv but not B Q , B j , v i n d i v i d u a l l y [66]. Using l ea s t squares curve f i t t i n g of experimental data to equation (148), the parameters M and N of the l i n e a r i z e d form of the RCM were evaluated. A minimum of 16 experimental data points were employed i n the f i t t i n g f o r each of the runs. The values of M and N are l i s t e d i n Table 6 .1 . Table 6.1: V a l u e s o f M and N i n E q u a t i o n (148) f o r v a r i o u s experiments. Run Temp., °C d , mm c XQ i n 5 h r s , % M = 2nB Q v 2 *103 N = 4 n B ] V *10 7 P 27 930 2 41.72 1.52 10.9 20 900 2 36.70 1.14 13.6 24 880 2 29.09 0.76 14.8 22 850 2 13.92 0.35 5.3 18 800 2 1-77 0.04 0.8 30 930 3.50 28.80 0.81 11.7 33 930 1.00 47.30 1.57 20.2 35 930 0.68 53.42 1.85 25.6 Having obtained values of Bjv and BQVl from t h i s t a b l e , the rates and the carbon conversions pred ic ted by the RCM were computed using equations (62) and (147), r e s p e c t i v e l y . The carbon conversions obtained experimental ly and pred ic ted by the 262 RCM for runs of d i f f e r e n t reac t ion temperatures and of var ious coke p a r t i c l e s i z e s are p l o t t e d as the s o l i d curves i n Figure 6.1 and Figure 6.2, r e s p e c t i v e l y , as a funct ion of r e a c t i o n t ime. I t i s evident that the RCM f i t s the conversion-t ime data adequately. S i m i l a r l y , the p r e d i c t i o n of the RCM f o r the r a t e s of carbon c o n v e r s i o n are shown in curves i n Figure 6.3 for runs at d i f f e r e n t temperature, and in Figure 6.4 for runs with d i f f e r e n t coke p a r t i c l e s i z e s . The e x p e r i m e n t a l r a t e d a t a shown as s c a t t e r e d points in F igure 6.3 and Figure 6.4 were obtained by taking the f i r s t d e r i v a t i v e wi th respect to time of the best f i t t e d polynomial of degree 3 to the experimental X ^ - t curves . Compar i sons between the exper imenta l ly obtained rates and the c a l c u l a t e d rates from RCM in Figure 6.3 and Figure 6.4 show that the c a l c u l a t e d values genera l ly e i t h e r increase or decrease monotonical ly whi le the ac tua l experimental rates a t t a i n maxima at intermediate carbon c o n v e r s i o n s . T h u s , the RCM f a i l s to p r e d i c t the most fundamental feature of the g a s i f i c a t i o n k i n e t i c s . The RCM suggests that carbon conversions corresponding to a l l runs at d i f f e r e n t temperatures, react ing gases, pressures , and p a r t i c l e s i zes become a s ing l e curve when p lo t t ed against dimensionless t ime. Carbon conversions of seven experiments at d i f f e r e n t operat ing condi t ions are p l o t t e d i n Figure 6.5 versus dimensionless time t / t . Results x c • 1 which correspond to runs at d i f f e r e n t temperatures and coke p a r t i c l e s i z e s do reduce to a common curve, as postulated by the RCM and by severa l other authors [104,116,123]. 263 CARBON CONV. VS. REACTION TIME GAVALAS AND KASAOKA MODELS F i g . 6.1 F i t of carbon conversions vs. time data to RCM ( ) and MVM ( ). Runs with d i f f e r e n t temperatures. Delayed coke; p = 30.3 kPa 264 CARBON CONV. VS. REACTION TIME GAVALAS AND KASAOKA MODELS 80 o o fe t—I CO <: o > o GJ o m 04 <j o LEGEND • =RUN #27 Dp=2.0mm T=930C 0=RUN #30 Dp=3.5mm T=930C • =RUN #33 Dp = l.lmm T=930C • =RUN #35 Dp=.68mm T=930C 60 120 180 240 REACTION TIME,MIN. 300 F i g . 6.2 F i t of carbon conversion vs. time data to RCM ( ) and MVM ( ). Runs with d i f f e r e n t coke p a r t i c l e s i z e s . Delayed coke; p H 20 30.3 kPa 265 CARBON CONVERSION VS. dXc /d t GAVALAS AND BHATIA MODELS 20 16 I S 12 X 8 0 - - - - . • • * * * • ' 1 • , • k ( 5"" < o ° — S - t r — • • • 1 o ** o o • : • O o • *«^ ^ N . . .> \ • = RUN #27 T=930C Dp=2mm 0=RUN #24 T=880C Dp=2mm • =RUN #22 T=850C Dp=2mm - = RUN #20 T=900C Dp=2mm 10 20 30 40 CARBON CONV. IN GASIFICATIONS 50 F i g . 6.3 Testing RCM ( -) and RPM ) using experimental (dX /dt) - X data of runs c c at d i f f e r e n t temperatures. Delayed coke; p = 30.3 kPa 266 CARBON CONVERSION VS. d X c /d t GAVALAS AND KASAOKA MODELS 30 LEGEND • =RUN #27 Dp=2.0mm T=930C 0=RUN #30 Dp=3.5mm T=930C • =RUN #33 Dp=l.)mm T=930C • =RUN #35 Dp=.68mm T=930C 24 18 12 0 i ; ; : • • • — 1 0 20 40 60 80 CARBON CONV. IN GASIFICATIONS F i g . 6.4 Testing RCM ( ) and MVM ( ) using experimental (dX /dt) - X data or runs c c with d i f f e r e n t p a r t i c l e s i z e . Delayed coke; p = 30.3 kPa 267 CARBON CONV. VS. DIMENSIONLESS TIME. TESTING GAVALAS MODEL 60 6^ O »—< Sc CJ f-1 40 fe •—< CO <d O > o o o PQ <3 O 20 LEGEND • =RUN #27 T=930C Dp=2.0mm 0=RUN #24 T=880C Dp=2.0mm • = RUN #22 T=850C Dp=2.0mm -=RUN #20 T=900C Dp=2.0mm = RUN #30 T=930C Dp=3.5mm <*= RUN #33 T=930C Dp=l.lmro ~= RUN #35 T=930C Dp=.68mm 1 2 3 4 DIMENSIONLESS TIME, t / t x c = ( U F i g . 6.5 Testing RCM using experimental X - dimensionless time data. Delayed coke; p H 20 30.3 kPa 268 M o r e o v e r , the RCM also impl ies that the (dXrj/dt) vs Xrj curves corresponding to d i f f e r e n t reac t ions , temperatures, p a r t i c l e s i z e s , e t c . can be reduced to a s ing le curve by appropriate s c a l i n g of (dX^/dt) by n o r m a l i z i n g , and that the conversion at which the rate a t ta ins maxima, X^R, i s d i c t a t e d s o l e l y by the s o l i d in ques t ion . The experimental n o r m a l i z e d r a t e s ( d X c / d t ) / ( d X c ; / d t ) x = Q ^ , f o r runs of d i f f e r e n t experimental condi t ions d e t a i l e d i n T a b l e 6 . 1 , are i l l u s t r a t e d i n F igure 6.6 for var ious extent of carbon conversions i n g a s i f i c a t i o n . The normalized r a t e - c a r b o n c o n v e r s i o n c u r v e s o b t a i n e d i n the present study revea l that the rates for most runs reach maxima at XfjR approximately between 15% and 26%, and the curves became a s ing l e curve only at low values of X^. 6.1.2 RANDOM PORE MODEL (RPM) L i k e the RCM the RPM, proposed by Bhat ia and Perlmutter [17,18,19] , c o n s i d e r s the changes i n the s o l i d s t r u c t u r e by o v e r l a p p i n g , i n t e r s e c t i o n and evo lut ion of pores . This model emphasizes the i n i t i a l condi t ions of the char v i a a s t r u c t u r a l parameter \|» which r e q u i r e s p h y s i c a l measurements . For the p a r t i c l e s i z e dp c >0.1 mm the rate express ion i s obtained by combining equation (36) and equation (38): dX k . c n s (1-X_) 5 T - i - c ' ^ ^ " V < 1 4 9 > o NORMALIZED RATE VS. CARBON CONV. TESTING GAVALAS MODEL 1.4 I : : : : : — • ii • re. • • o El • O o o LEGEND • = RUN #27 T=930C Dp=2.0mm 0= RUN #24 T=880C Dp=2.0mm •= RUN #22 T=850C Dp=2.0mm B= RUN #20T=900C Dp=2.0mm H= RUN #30 T=930C Dp=3.5mm <$>=RUN #33 T=930C Dp=l.lmm ffi=RUN #35T=930C Dp=.85mm 0 2 0 4 0 CARBON CONV. IN GASIFICATION^ 60 F i g . 6.6 Testing RCM using experimental normalized rate - X data, c Delayed coke; p = 30.3 kPa 270 No simple a n a l y t i c a l expression i s a v a i l a b l e from the RPM f o r the Xfj-t data . The carbon conversion can be computed numerica l ly using Equation ( 3 3 ) . No attempt was made by t h i s study to f i t XQ us ing t h i s method. The s p e c i f i c area Sp at any carbon conversion can be approximated based on i t s i n i t i a l value S D . S p = S Q ( 1 -X C ) / 1 - M J . L n ( l - X c ) ( 3 8 ) The s t r u c t u r a l parameter \J« and the i n t r i n s i c rate constant k s can be evaluated by making a l i n e a r p l o t of the equation "dX„/dt" | 2 r . _n_ _ k C S 2 r . „n„ -\ k C S C s o - + s o _1 - X C L 1 - E O _ L 1 - E O J Ln (1 -X C ) (41) Resul ts of ac tua l measurements done i n t h i s study for the coke at the b e g i n n i n g of the g a s i f i c a t i o n s tage were S Q = 99,960 cm 2 / cm 3 and e o = 7 . 5 6 x l 0 - 2 (cm 3 pore volume)/(cm 3 of i n i t i a l carbon for g a s i f i c a t i o n ) . Using these values of S Q and E q i n equation (41), k s and \|» were r e a d i l y obtained from the in tercept at XQ=Q and from the s lope , r e s p e c t i v e l y , of F igure 6.7. Note that s t r a i g h t l i n e s are evident over a l i m i t e d ranges of convers ion . The r e s u l t i n g values of k s and \|> are presented i n Table 6 .2 . In t h i s t a b l e , the values of >l> are between 9 and 16; an average value of \|> of 12 was taken for subsequent c a l c u l a t i o n s . The value \JJ= 12 was subs t i tu ted into equation (149) to compute the pred ic ted rates of conversion by the RPM; and the r e s u l t i n g rates are shown in Figure 6 . 3 as dashed c u r v e s . The v a l u e s of the r a t e s c a l c u l a t e d from RPM 271 STRUCTURAL PARAMETER FOR DELAYED COKE, BHATIA MODEL 500 400 Cv} 300 o X -4-> o ^ 200 00 O 100 0.0 0.1 • : • :. U/T : ° 1 LEGEND •= RUN #27 T=930C Dp=2mm 0=RUN #24 T=880C Dp=2mm •= RUN #20 T=900C Dp=2mm ) 0.2 L n ( l - X r ) 0.3 0.4 F i g . 6.7 Determination of ¥ and ks i n Eq. (41) or RPM. Delayed coke; p = 30.3 kPa 272 Table 6.2 VALUE OF v|> AND k g IN EQUATION (149), OBTAINED IN FITTING  EXPERIMENTAL DATA INTO EQUATION (41) Run Reaction Y - I n t e r c e p t . Slope *10 6 , k s *10 8 \|i temp., " C * 1 Q 7 , min~ 2 min~ 2 c m / m i n « a t m  27 930 13.36 14.38 3.56 10.77 20 900 10.05 8.96 3.09 8.91 24 880 4.82 7.61 2.14 15.79 22 850 1.39 0.52 1.60 12.40 ( F i g u r e 6 .3) agree w i t h the e x p e r i m e n t a l r e s u l t s o n l y at carbon conversions less than 15% and only for runs at 9 0 0 ° C or h igher . The model could not p r e d i c t the sharp drop-of f i n rates caused by d r a s t i c changes i n porous s t ruc ture and i n pore surface area during the steam g a s i f i c a t i o n of o i l sand cokes. Those s t r u c t u r a l changes are t y p i c a l in g a s - s o l i d r e a c t i o n s , e s p e c i a l l y i n g a s i f i c a t i o n of coals and chars . With the value of \|> and the i n i t i a l surface area S Q known, the pred ic ted value of the s p e c i f i c surface area Sp was c a l c u l a t e d using equation (38) and i l l u s t r a t e d as the dashed curve i n Figure 6.8. It i s quite apparent that the c a l c u l a t e d values of surface area are about f i v e times lower than the experimental r e s u l t s . Combining equation (38) and equation (149) of the RPM y i e l d s : [ ( d X c / d t ) / S p ] - [ k s C n / ( l - c 0 ) ] - constant (151) The carbon conversion rates per u n i t surface area, evaluated on the bas is of the i n i t i a l weight of carbon for g a s i f i c a t i o n , for the four runs employed d i f f e r e n t coke p a r t i c l e s izes at 9 3 0 ° C (Table 6.1) were 273 CARBON CONV. VS. SURFACE A R E A TESTING BHATIA MODEL F i g . 6.8 Testing RPM on the p r e d i c t i o n of surface area i n Eq. (38). Experimental data ( ); by RPM ( ). Delayed coke; p = 30.3 kPa 274 determined, and p lo t t ed against carbon conversion in Figure 6.10. The r e s u l t s in t h i s f igure demonstrate that the conversion rate per un i t surface area for a l l these runs does not remain constant over the e n t i r e range of convers ion. Thus, the f indings confirm that the rates of convers ion are only p a r t i a l l y af fected by the s u r f a c e a r e a of the carbon, as prev ious ly pointed out i n Chapter 4. Since the rate per u n i t surface area i s not constant, the value of k s i n equation (151) i s not a c o n s t a n t over the f u l l e x t e n t of c a r b o n c o n v e r s i o n f o r a g i v e n experimental temperature; thus the equation (151) cannot be used for determining the experimental a c t i v a t i o n energy. Instead, the i n t r i n s i c r e a c t i o n rate constants , k ' s , der ived from B h a t i a ' s RPM f o r the steam g a s i f i c a t i o n of o i l sand cokes at var ious temperatures were p l o t t e d a g a i n s t the r e c i p r o c a l of temperature i n an Arrhenius - type p l o t i n Figure 6.9. From t h i s f i g u r e , an a c t i v a t i o n energy of E=113.3 kJ/mol was obtained f o r the steam g a s i f i c a t i o n of o i l sand coke. T h i s value i s w i t h i n the range of 113 kJ/mol and 134 kJ/mol f o r steam g a s i f i c a t i o n of chars reported by A d s h i r i et a l . [3] and Chin et a l . [38], r e s p e c t i v e l y , who both employed s u r f a c e - b a s e d r e a c t i o n models to d e t e r m i n e the a c t i v a t i o n energy. The exper imenta l average value of the i n t r i n s i c rate constants f o u n d by t h i s r e s e a r c h , u s i n g e q u a t i o n (149 ) at XQ<20%, of 2 . 5 x 1 0 " ® cra/min. atm. i s about 25 times higher than the values of k s , based on pore s u r f a c e area , of about 1x10"^ cm/min. atm. reported e a r l i e r in the l i t e r a t u r e [38] for steam g a s i f i c a t i o n of a c t i v a t e d carbon produced from brown coa l char and from bituminous coa l char . A R R H E N I U S - T Y P E PLOT FOR T E M P E R A T U R E DEPENDENCE RUNS 1 1 : ; : : : : : : 1 0.80 0.82 0.84 0.86 0.88 0.90 1000./T , K _ 1 Fig. 6.9 Arrhenius - type plot for runs with delayed -coke at d i f f e r e n t temperatures, using RPM. Delayed coke; p = 30.3 kPa 276 •r-1 ! cj Q G O o X * OS I o RATE P E R UNIT SURFACE A R E A E X P E R I M E N T A L M E A S U R E M E N T 10 8 0 LEGEND = RUN #86 T=930C Dp=.68mm • = RUN #20 T=900C Dp= 2.0mm 0=RUN #33 T=930C Dp= 1.1mm " = RUN #27 T=930C Dp= 2.0mm • o-• o • • • • 0 20 40 60 CARBON CONVERSION IN GASIF.,% F i g . 6.10 Testing RCM and RPM on the r e l a t i o n s h i p between rate and surface area. Delayed coke; p = 30.3 kPa H 2 ° 277 This may be due to the 62 times higher i n i t i a l pore surface area of the l a t t e r compared to that of the former. The l e v e l of carbon conversion at which the surface area a t ta ins a maximum may be pred ic ted by the RPM: X C S = 1 - exp [ ( l - * ) / 2 * J (39) A value of XQ = 37% was obtained by s u b s t i t u t i n g the average value of \|)=12 into equation (39). The pred ic ted value of XQS i s h igher than the experimental value of about 27%. Equation (39) ind ica te s that Xfjg i s an extremely weak funct ion of \|> for v|i> 10; thus equation (39) cannot be used to compute Other authors [104] have experienced great d i f f i c u l t i e s i n determining the value of \]» us ing t h e i r experimental data , and found that \|> was not cons is tent for the same set of data . M o r e o v e r , e m p l o y i n g a v a l u e o f t r u e d e n s i t y o f coke o f p t =1.2 g/cra^, the t o t a l pore length per un i t mass of the i n i t i a l coke for g a s i f i c a t i o n was estimated using equation (34). The estimated value for the coke sample i s about 7x10^ m/g compared to 5x10*0 m / g reported f o r char ox idat ion [171]. The most s tra ight forward t e s t of the RPM may be formulated in terms of the pore s tructure parameter \J> v i a equation (41). An average of \J;=12 was obtained from the p l o t [ ( d X p / d t ) / ( l - X c ) J 2 vs - L n ( l - X r ) as shown i n Figure 6.7. This value i s c lose to the range of \|i reported by Su and Perlmutter [171] of between 7 and 14 for the char ox idat ion process . I t should be noted that the curves i n Figure 6.7 of equation (41) are l i n e a r only at carbon conversions smal ler than approximately 20%, and thus only these l i n e a r sect ions were employed to determine the value of 278 the s t r u c t u r a l parameter L i n e a r regress ion of data in Figure 6.7 g ives values of c o r r e l a t i o n c o e f f i c i e n t s between 0.96 and 0.99, which are i n d i c a t i v e of s t r a i g h t l i n e s . The n o n - l i n e a r i t y i n the [ ( d X c / d t ) / ( l - X c ) 1^  v s ~Ln(l""Xrj) r e l a t i o n s h i p was c o n s i s t e n t l y observed, not only i n t h i s present research, but a l so i n a paper publ ished by Perlmutter [171], the co-author of the RPM. Thi s problem i n determining the value of \|>, which i s c r i t i c a l l y important i n the RPM, may be a c o n t r i b u t i n g f a c t o r to the f a i l u r e of the RPM to p r e d i c t the rates at carbon conversion higher than 20% and i s r e s p o n s i b l e f o r the low p r e d i c t e d values of s p e c i f i c surface area shown i n Figure 6.8. 6.1.3 MODIFIED VOLUMETRIC MODEL (MVM) Kasaoka and Sakata [104,105] modif ied the conventional volumetric r e a c t i o n model to a simple two-constant integrated rate equation known as the MVM: X c = 1 - exp [ -At B ] (68) or i n a form s u i t a b l e for l i n e a r p l o t t i n g Y = Ln A + B Lnt (69) where Y = Ln [ - L n ( l - X c ) ] The rate pred ic ted by the MVM i s the f i r s t d e r i v a t i v e of equation (68): d X c / d t = (1-X C ) A . B t ( B - 1 ) = (1 -X C ) k ( X c ) (150) 279 The average rate constant i s given by: 1 i k = | k ( X c ) d X c 0.99 k ( X c ) d X c (76) 0.01 where k i s def ined i n equation (67). Simple l i n e a r regress ion of equation (69) us ing Xfj-t data y i e l d s the values of A and B i n the MVM. These values of parameters A and B and the average rate constant k in equation (76) f o r runs of d i f f e r e n t r e a c t i o n temperatures and at v a r i o u s p a r t i c l e s i z e s are g i v e n i n Table 6.3. Table 6.3 VALUES OF A AND B IN EQUATION (68), AND k IN EQUATION (76): Run Temp, °C d c' mm A*10 4 B C o r r e l a t i o n Coeff . f o r Eq . (69) Average ra Constant, m i n - * 27 930 2 4.61 1. 25 0.99 2 . 5 3 x l 0 - 3 20 930 2 4.12 1. 24 0.99 2 . 0 3 x l 0 - 3 24 880 2 3.27 1. 22 0.99 1 . 5 9 x l 0 - 3 22 850 2 1.99 1. 16 0.99 0 . 6 2 x l 0 - 3 18 800 2 1.75 1. 21 0.99 30 930 3.5 3.20 1. 23 0.99 1 . 5 7 x l 0 - 3 33 930 1.0 5.13 1. 26 0.99 2 . 7 2 x l 0 - 3 35 930 0.68 5.84 1. 27 0.99 3 . 1 5 x l 0 - 3 The carbon conversions der ived from the f i t t i n g of the experimental data of runs at d i f f e r e n t temperatures and with var ious coke s i zes to the MVM are shown in Figure 6.1 and Figure 6.2, r e s p e c t i v e l y , as the dotted curves . I t appears that the f i t t e d carbon conversions by the MVM are in exce l l en t agreement with the experimental data of t h i s study. 280 The rates of carbon c o n v e r s i o n were o b t a i n e d by f i t t i n g the experimental data in to equation (150) of MVM. These pred ic ted rates are presented i n Figure 6.4 as the dotted curves. Although the MVM f a i l s to pred ic t the sharp decreases i n the rates a f t e r t h e i r maxima and shows the maxima i n the rates at very small carbon convers ions , i t genera l ly f i t s the experimental rates over the lower two-th irds of the extent of carbon conversion i n each run. Using the values of the c a l c u l a t e d average rate constants k by the MVM for runs of d i f f e r e n t temperatures (Table 6 .3) , an a c t i v a t i o n energy of 179 kJ/mol was computed. 6.1.4 DISCUSSION The value of a c t i v a t i o n energy determined by the present research us ing the average rate constants , which were der ived from the MVM - a volume-based model - f or runs of d i f f e r e n t temperatures, i s near ly 1.8 times higher than the value presented by using the RPM - a surface-based model. The d i f f erence i n the l e v e l of a c t i v a t i o n energy for the same reac t ion i s presumably caused by the use of two d i f f e r e n t types of k i n e t i c models. Bowen et a l . [29] compared the unreacted-core model, invo lv ing a sharp reac t ion i n t e r f a c e , with a model i n v o l v i n g a r e a c t i o n zone of a f i n i t e th ickness , and concluded on mathematical grounds that the sharp in ter face r e a c t i o n v e l o c i t y constant i s r e a l l y the square-root of the r e a c t i o n v e l o c i t y constant i n the reac t ion zone, implying that the energy of a c t i v a t i o n ca l cu la ted from the surface-based r e a c t i o n model i s h a l f that c a l c u l a t e d from the volume-based model. A s i m i l a r 281 observat ion was reported by severa l other authors in the l i t e r a t u r e [104]. The h igh a c t i v a t i o n energy coupled with the X ^ - t curves , presented i n Figure 6.1 and Figure 6.2 which show no sharp decreases i n t h e i r s l o p e s sugges t t h a t the react ions were taking place under chemical r e a c t i o n c o n t r o l . Th i s present observat ion agrees with that postulated by Revankar [147] who suggested that the carbon-steam r e a c t i o n i s k i n e t i c a l l y c o n t r o l l e d at temperatures below 1 0 0 0 ° C . A l s o , the values of carbon conversion at which the rates reach maxima pred ic ted by the RPM and the RCM are 0 < X Q R £ 39.3%, which agrees wi th values found in t h i s study of between 15% and 26% as shown i n Figure 6.3 and Figure 6.4. In summary, three models - the RCM, the RPM and the MVM - were tes ted using the experimental k i n e t i c data obtained i n t h i s research . The p r e d i c t e d Xrj-t da ta by RCM i s i n good agreement w i t h the e x p e r i m e n t a l d a t a . When r e d u c e d t o d i m e n s i o n l e s s f o r m , t h e conversion-t ime data appear to fo l low expectations of the RCM. The f i t t e d parameters from Xrj vs t data y i e l d rate curves which e i t h e r increase or decrease with conversion but the RCM does not y i e l d rate maxima as found by experiment. The predic ted co l lapse of dimensionless rate versus conversion to a s ing l e curve occurs only at very low carbon convers ions . Thus i n several aspects the RCM i s inadequate to model the g a s i f i c a t i o n . The RPM suf fers from some of the same drawbacks. The value of vj> determined from the rate -convers ion p lo t i s f a r too low to descr ibe the s p e c i f i c surface area generation with convers ion. The experimental rate maxima are too sharp to be f i t t e d by the model with the given \J> 282 va lue . At conversions below those corresponding to the rate maximum, the model i s adequate and rate constants have been evaluated. The MVM f i t s conversion-t ime data w e l l . Rates are w e l l descr ibed , but again the sharp drop a f t e r the maximum i s not w e l l f i t t e d . The e m p i r i c a l nature of the model i s another drawback. 6.2 TESTING SULPHUR CAPTURE MODELS 6.2.1 GRAIN MODEL (GM) The g r a i n model i s one of the most s u i t a b l e models for the H2S + CaO g a s - s o l i d r e a c t i o n b e c a u s e i t e x p r e s s e s t h e b e h a v i o r o f monotonical ly decreasing rate and surface area during s u l p h i d a t i o n . The rate of r eac t ion of a reac t ing gas with a porous s o l i d p a r t i c l e i s r e l a t e d by the GM to the s i ze of the non-porous gra ins which comprise the i n t e r i o r of the s o l i d . In t h i s present case the gra ins are of CaO. As r e a c t i o n proceeds, i t i s assumed that H2S d i f fuse s through the CaS product l ayer surrounding the i n d i v i d u a l gra ins and reacts at a sharp in ter face between the product l ayer and the unreacted c a l c i n e , and that the gas d i f f u s i o n through the i n t e r g r a i n vo ids i s an i n s i g n i f i c a n t r e s i s t a n c e . The o v e r a l l rate can be l i m i t e d by chemical r e a c t i o n at the CaO-CaS i n t e r f a c e , by the d i f f u s i o n through the p r o d u c t l a y e r , or by some combination of the two. In the case of r e a c t i o n c o n t r o l l i n g , the Xgg-t response i s g iven by equation (121): 283 P" t - r g [ 1 - ( 1 - X S B ) 1 / 3 ] (121) and f o r the case of product - layer c o n t r o l l i n g by equation (122): P" t = "oDeC ^ [ 1 - 3 ( 1 - X S B ) 2 / 3 + 2 ( 1 - X S B ) 1 (122) As d iscussed i n Chapter 5, at the beginning of the s u l p h i d a t i o n , the c a l c i n e i s very porous and high i n reac t ion surface , and the system appears to be c o n t r o l l e d by chemical r e a c t i o n . However, i n the l a t e r stages of the s u l p h i d a t i o n r e a c t i o n , the s o l i d product f i l l s the pores and covers the surface of the c a l c i n e , causing the slope of the Xgg vs t curve to drop s h a r p l y . In the reg ion of chemical c o n t r o l , the reac t ion rate constant k can be computed from the slope of the t vs [ I - ( I - X S B ) * ^ ] p l o t suggested by equation (121). The r e a c t i o n rate constants for runs with sorbent of d i f f e r e n t p a r t i c l e s i z e s and for runs with sorbent from d i f f e r e n t c a l c i n a t i o n condi t ions were determined from F i g u r e s 6.11 and 6.12 r e s p e c t i v e l y . The reac t ion times shown in these p l o t s exclude the 60-minutes p y r o l y s i s stage, and the points shown begin at 30 minutes a f t er p y r o l y s i s , by which time the periods of 100% H 2 S capture have p a s s e d ( F i g u r e 5 . 5 . 2 ) , and the H 2 S g e n e r a t i o n p r o c e s s i s not c o n t r o l l i n g . In each f i g u r e , only the l i n e a r s e c t i o n at the v e r y beginning of each curve was employed for the rate constant c a l c u l a t i o n using l i n e a r regress ion a n a l y s i s . In the present study, the c r i t e r i o n f o r the s e l e c t i o n of the mentioned l i n e a r sec t ion was to obtain a c o r r e l a t i o n c o e f f i c i e n t f o r the a n a l y s i s no s m a l l e r than 0.98 by 284 TESTING T H E GRAIN MODEL IN T H E REACTION CONTROL REGION F i g . 6.11 Testing the G.M. i n the region- of reaction control, using experimental data for runs with d i f f e r e n t sorbent s i z e s . 285 TESTING T H E GRAIN MODEL IN T H E REACTION CONTROL REGION 240 H 1 2 0 U 0 . • o o • o • o-o • O • • O o • o • • LEGEND • = RUN #40 t=40 min. 0= RUN #54 t=120 min • = RUN #22 t=300 min 0 20 40 10 3 *[ l - ( l -X S B ) l / 3 ] 60 F i g . 6.12 Testing the G.M. i n the region of reaction c o n t r o l , using experimental data for runs with sorbents from d i f f e r e n t c a l c i n a t i o n conditions. 286 t r i a l - a n d - e r r o r . Resu l t ing from t h i s chosen value of the c o r r e l a t i o n c o e f f i c i e n t , the s u l p h i d a t i o n time at which the "early stage" ends was taken to be 180 minutes for a l l runs. The conversions at the end of the ear ly - s tage are given i n Table 6.4. Data for Figure 6.11 and Figure 6.12 and the sample c a l c u l a t i o n to derive these data are presented in Appendix D. The values of the rate constant for the e a r l y stage of the s u l p h i d a t i o n r e a c t i o n for runs with ca l c ined limestone and c a l c i n e d dolomite are between 1.57x10"^ m/s and 2.4x10"^ m/s. In the region of s o l i d p r o d u c t l a y e r d i f f u s i o n c o n t r o l , the e f f e c t i v e d i f f u s i v i t y De was estimated from the slope of the l i n e a r s ec t ion c o r r e s p o n d i n g to the l a t e r s tage c o n v e r s i o n i n the t vs [ l - 3 ( l - X S B ) 2 / 3 + 2 ( l - X S B ) ] p l o t of equation (122). Data and sample c a l c u l a t i o n s f o r t h i s equation are f i l e d i n Appendix D. The t e s t i n g of the product l ayer d i f f u s i o n c o n t r o l mechanism u t i l i z e d experimental data of runs wi th sorbents of d i f f e r e n t p a r t i c l e s i ze (Figure 6.13) and with sorbents from d i f f e r e n t c a l c i n a t i o n condi t ions (Figure 6.14). It i s determined from these f igures tha t , i n each run, the l i n e a r sec t ion wi th c o r r e l a t i o n c o e f f i c i e n t greater than 0.98 ex i s t s a f t e r 180 minutes of s u l p h i d a t i o n . Therefore data points i n Figure 6.13 and Figure 6.14 c o r r e s p o n d i n g to 180 minutes or longer were employed i n the l i n e a r regress ion ana lys i s to determine the e f f e c t i v e d i f f u s i o n c o e f f i c i e n t s . Results of l i n e a r regress ion analyses of equation (122) are reported i n Table 6.4. I t i s seen from t h i s table that the values of the e f f e c t i v e d i f f u s i v i t y are in the range of l x l O - * 3 m 2 / s . This d i f f u s i v i t y i s about a f a c t o r of 10® below the d i f f u s i v i t y of H2S i n pores of ca lc ined l imestone, which i s reported to be about 8x10"^ m 2 / s [81,33]. It i s TABLE 6.4 SUMMARY OF SORBENT CHARACTERISTICS AND MODEL RESULTS USING SULPHUR CAPTURE DATA Run d p a , C a l c i n a C a l c l n t t 1 o n V V X S B ' X S B at t h a R e a c t i o n c o n t r o l R e g i m e , CM nro. C o d t t l m a , m l n . m 2 / g X and o f tht E a r l y s t a g * E a r l y a t a g a C o r r a l a t 1 o n k * 1 0 5 1 ( H i t a t a g a , Y . I n t . , a l o p a , c o a f f . f o r m/t % m l n . m l n . Eq ( 1 2 1 ) 40 0 . 5 8 C A M 40 3 . 8 2 2 . 5 1 2 . 1 0 6 .1 3 0 . 1 7 3 6 0 . 9 0 . 9 9 54 0 . 6 8 CALB 300 4 . 4 0 1.25 1 4 . 1 0 8 .1 2 2 . 0 5 6 9 2 . 0 0 . 9 9 2 . 0 4 69 0 . 6 9 0N3 120 6 . 0 2 0 . 7 5 1 6 . 8 0 9 . 0 2 4 . 2 5 0 4 6 . 3 0 . 9 9 2 . 3 0 78 0 . 6 8 0L1 90 1 5 . 1 8 0 . 6 1 6 . 2 5 9 . 2 2 4 . 5 4 8 3 9 . 4 0 . 9 8 2 . 4 0 77 1.00 DL2 90 1 0 . 3 8 1.8 9 . 7 0 6 . 9 1 8 . 4 6 3 8 8 . 9 0 . 9 9 1.81 78 1.3 0 90 5 .OS 2 . 5 6 . 1 9 4 . 8 1 4 . 2 7 3 7 9 . 5 0 . 9 9 1.57 TABLE 6.4 Cont'd, on next pa TABLE 6.4 Cont'd.: SUMMARY OF SORBENT CHARACTERISTICS AND MODEL RESULTS USING SULPHUR CAPTURE DATA Run P r o d u c t l a y e r d l f f m l o n r e g i m e , GM R e a c t i o n C o n t r o l r e g i m e , CM l a t e r ( t a g * L a t e r t t a g e C o r r e l a t i o n Oe , m It E a r l y t t a g e C o r r e l a t i o n k , t e c Y . I n t . , t l o p e , c o e f f . f o r * 1 0 1 4 e l o p e , c o e f f . f o r * 1 Q 6  w i n . m i n . Eg (122) * 1 0 4 , m i n . " 1 Eg ( 1 3 3 )  40 1 1 5 . 0 4 6 , 7 3 9 0 . 9 9 4 . 1 8 0 . 9 9 6 . 9 6 64 9 3 . 0 3 6 , 6 1 5 0 . 9 9 1 6 . 4 9 S . 6 9 0 . 9 9 9 . 4 9 69 9 5 . 7 2 8 , 9 6 0 0 . 9 9 7 .61 6 . 1 7 0 . 9 9 1 0 . 2 76 9 5 . 0 2 7 , 4 7 2 0 . 9 9 2 0 . 2 4 6 . 0 3 0 . 9 9 1 0 . 0 77 4 7 . 4 7 9 , 8 6 0 0 . 9 9 1 2 . 3 8 4 . 0 4 0 . 9 9 6 . 7 3 70 - 3 9 . 4 2 8 1 , 8 0 0 0 . 9 9 8 . 6 8 2 . 2 4 0 . 9 8 3 . 7 4 288 TESTING T H E GRAIN MODEL IN THE SOLID PRODUCT CONTROL REGION F i g . 6 . 1 3 T e s t i n g t h e G.M i n t h e r e g i o n o f p r o d u c t l a y e r c o n t r o l . U s i n g e x p e r i m e n t a l d a t a o f r u n s w i t h d i f f e r e n t s o r b e n t s i z e s . 289 TESTING T H E GRAIN MODEL IN  T H E PRODUCT CONTROL REGION 360 I : : : : . » 9 : B 240 120 LEGEND • = RUN #40 t=40 rain. 0= RUN #54 t=120 min. • = RUN #69 1=300 min. 0 I : • : : • • • : : 1 0 20 40 60 80 100 F i g . 6.14 Testing G.M. i n the r e g i o n o f product layer c o n t r o l , using experimental data of runs with sorbent of d i f f e r e n t c a l c i n a t i o n conditions. 290 t y p i c a l of that for product layer d i f f u s i o n . The strong e f fec t s of surface area on the H2S + CaO reac t ion as shown i n F igure 5 .3 .9 , F igure 5.5.5 and Figure 5.6.4 imply tha t , i n the l a t e r stage of convers ion , the rate i s most probably l i m i t e d by a d i f f u s i o n process i n the product l ayer surrounding the i n d i v i d u a l grains according to equation (122). The c o n c l u s i o n i s s u p p o r t e d by t h e c l o s e f i t o f t h e t v s [ 1 - 3 ( 1 - X S B ) 2 ^ 3 + 2 ( 1 - X S B ) ] data to the l i n e a r response of equation (122) f o r extended r e t e n t i o n t imes, as shown i n Figure 6.13 and Figure 6.14. Borgwardt [27] reported a d i f f u s i v i t y of 3.2xl0~^2 m 2 / s accompanied by a h i g h a c t i v a t i o n energy of 130 k J / m o l f o r the CaO + H2S r e a c t i o n . S i m i l a r low d i f f u s i v i t y r e s u l t s are given by Bhat ia and Perlmutter [19] f o r the r e a c t i o n of CaO w i t h CO2 and by many other inves t iga tors [81,67,18,40,51,161,26] for the CaO + SG 2 r e a c t i o n , both of which are d i c t a t e d by product l ayer d i f f u s i o n . The e f f e c t i v e d i f f u s i v i t i e s determined by these s tudies at 8 5 0 ° C are l i s t e d i n Table 6 .5 . Table 6.5 DIFFUSIVITIES OF REACTING GAS THROUGH PRODUCT LAYER Author Present Borgwardt Bhatia Hartman Bhatia Georgekis Chrostowskl Simons Study [27] [19] [81] [18] [67] [41] [161] Reaction CaO+H2S CaO+H2S Ca0+C02 Ca0+S02 Ca0+S02 Ca0+S02 CaO+S02 Ca0+S02 Do, m2/s 1.2X1CT 1 3 3 . 2 x i r r 1 2 1 0 - 2 2 BxlO" 1 3 7 . 6 x i r 1 3 8X1CT 1 3 7.5X10" 1 1 4.4X1CT 1 1 to to to to 10-18 2 . 5 x i r r 1 1 2.5X1Q- 1 0 3.0X10T 1 0 The extremely low e f f e c t i v e d i f f u s i v i t i e s may be r e l a t e d to i o n i c species d i f f u s i o n . The probable mechanism of CaS formation by i o n i c d i f f u s i o n through the product layer was suggested by Borgwardt [27] in equation (129), and g r a p h i c a l l y shown i n Figure 6.15. F igure 6.15 PROBABLE MECHANISM OF CaS FORMATION OF IONIC DIFFUSION  THROUGH THE PRODUCT LAYER [7] CaO CaS 0= H 2 S S i m i l a r suggestions were a lso made [19] for the CaO + CO2 r e a c t i o n . These arguments on i o n i c d i f f u s i o n are supported by f ind ings of A r n i k a r [9] , who s tudied the i o n i c d i f f u s i o n of var ious species and reported values of i o n i c d i f f u s i v i t y through c a l c i t e c r y s t a l s between 10-15 m 2 / s and 5 x l 0 - 1 5 m 2 / s at 573K. The d i f f u s i v i t y through the product layer must be estimated from k i n e t i c da ta . Thus the value obtained depends on the choice of the model used to analyze the data , and thus estimates vary wide ly . The value determined by the present research i s w i th in t h i s large range of prev ious ly reported va lues . The e f f e c t s of s o r b e n t p a r t i c l e s i ze on the rate c o n t r o l l i n g mechanism can be observed in Figure 6.11. In t h i s p l o t , the s ec t ion of the t vs [ l - ( 1 - X g g ) ] curves for sorbent of 0.68 mm are we l l def ined and ex i s t up to a r e l a t i v e l y long extent of conversion of about Xgg=12%. This i s an i n d i c a t i o n of chemical reac t ion c o n t r o l . However, the l i n e a r sect ions of the curves for large sorbent p a r t i c l e s i zes i . e . dp S=1.0 mm 292 and e s p e c i a l l y for dp S=1.3 mm, are l e ss apparent and e x i s t to much lower e x t e n t of c o n v e r s i o n . The v a l u e s of Xgg up to which the curves corresponding to dp S=1.0 mm and dp S=1.3 mm are s t i l l l i n e a r and possess a c o r r e l a t i o n c o e f f i c i e n t no s m a l l e r than 0.98 are 6.9% and 3.9% r e s p e c t i v e l y . The observations i n t h i s s ec t ion agree w e l l w i th the r e s p e c t i v e Xgg-t curves for these large s ize c a l c i n e d samples which l e v e l o f f q u i c k l y a f t e r a short i n i t i a l l i n e a r s e c t i o n as shown i n Figure 5 .5 .2 . Furthermore, the slopes of the l i n e a r sect ions i n both Figure 6.11 and Figure 6.13 change wi th changing p a r t i c l e s i z e . The i n f l u e n c e of c a l c i n a t i o n condi t ions on the r e a c t i v i t i e s of c a l c i n e s i n the s u l p h i d a t i o n r e a c t i o n was a lso r e f l e c t e d i n the r e a c t i o n r a t e c o n s t a n t s der ived from the GM. The rate constant i n the run employing undercalc ined limestone (CAL1) i s smal ler than the rate i n runs wi th f u l l y c a l c i n e d sample (QN3) as seen i n Table 6.4. Results of p r i o r s tudies [158] on the rates of r e a c t i o n of porous c a l c i n e d l i m e s t o n e w i t h hydrogen sulphide and wi th sulphur d iox ide i n d i c a t e d that the r e a c t i o n of H2S wi th CaO proceeds almost as f a s t as that of SO2 wi th CaO; thus t h e i r rate constants can be convenient ly compared. The average value of the s u l p h i d a t i o n r e a c t i o n rate constant k at 9 3 0 ° C and f o r a p a r t i c l e s i z e of 0.68 mm, determined by t h i s study v i a the GM, i s approximately 2xl0~-> m/s f o r both c a l c i n e d limestone and c a l c i n e d dolomite sorbents as given i n Table 6.4. These values are r e l a t i v e l y c lose to the range reported by Bhat ia and Perlmutter [18] of 2 . 6 x l 0 ~ 6 m/s at 9 3 0 ° C for d p s =0.57 mm, by Wen and Ishida [195] of 3 .72xl0~ 7 m/s at 8 5 0 ° C and by Ramachandran and Smith [145] of 7 .1x l0~ 7 m/s at 8 5 0 ° C for the s u l p h i d a t i o n r e a c t i o n . 293 Since there appears to be two regions of r e a c t i o n , each c o n t r o l l e d by a d i f f e r e n t mechanism, the pred ic t i ons of the o v e r a l l process by using the e a r l y stage equation, equation (121), and by the l a t e r stage equat ion, equation (122) of G . M . , f o r runs of d i f f e r e n t p a r t i c l e s i zes are obtained by s u b s t i t u t i n g the values of k and De f o r runs employing dp S=0.68 mm and dp S=1.0 mm of Table 6.4 into equation (121) and equation (122) r e s p e c t i v e l y . The appropriate Y - i n t e r c e p t for each case where k and De were d e r i v e d ( T a b l e 6 .4) are a l s o employed to adjust the coordinates to t h e i r o r i g i n a l va lues . The r e s u l t s of these p r e d i c t i o n s are shown i n F igure 6.16. I t i s seen that for the smal ler p a r t i c l e s i ze i . e . dp S=0.68 mm, the r e a c t i o n - c o n t r o l l e d region seems to extend up to a sorbent conversion X S B of 10%, higher than that g iven i n Table 6.4; while the p r o d u c t - l a y e r - d i f f u s i o n regime becomes s h o r t e r and r e a l l y begins at a Xgg of 12%. By contras t , for the l a r g e r p a r t i c l e s i z e , i . e . dp S=1.0 mm, the r e a c t i o n - c o n t r o l l e d region holds only up to a Xgjj of about 4.6% as given i n Table 6.4; while the product l a y e r d i f f u s i o n region appears to become e f f e c t i v e at an Xgg of about 4%, much e a r l i e r than the 12% f o r the smal ler p a r t i c l e s i z e . This supports the e a r l i e r argument (Chapter 5) that the sorbent conversion versus time curves for runs of large p a r t i c l e s i ze l e v e l o f f at low conversions due to product layer d i f f u s i o n c o n t r o l . No simple a n a l y t i c a l express ion i s a v a i l a b l e f o r the combined e a r l y stage and la te stage. Numerical s o l u t i o n was sugges ted f o r the combined equation [179], as mentioned i n sec t ion 294 TESTING GRAIN MODEL IN BOTH STAGES OF REACTION 20 O •—i in PS W > O w PQ O (7) 15 10 0 0 LEGEND • =RUN #70 DOL. Dp=.68mm •=RUN #77 DOL. Dp = 1.0mm 4* 4> 4 * 4-4- _ V + i 4* A 4> Wr 4- ^ * AT • 4* • 4* 4> i S -+ * */ 4* g r f + * * 1/ 1 / 1 r / 60 120 180 240 300 TOTAL REACTION TIME, MIN. 360 F i g . 6.16 Predictions by the G.M. on X n o - t, using early stage Eq. (121) and l a t e r stage Eq. (122). Runs at d i f f e r e n t sorbent p a r t i c l e s i z e s . Early stage: . Later stage: . Experimental data: scattered points 295 6.2.2 CONTINUOUS FIRST-ORDER MODEL (CM) The CM i s one of the most frequent ly used models f o r the H2S + CaO r e a c t i o n [114,168,137,149,131]. However, i t has been found that the model was not adequate to descr ibe the l a t e r stage of the su lph ida t ion r e a c t i o n , where the product s o l i d presumably c o n t r o l s the p r o c e s s [137,131]. Thus, the use of the CM i n t h i s study i s l i m i t e d to the e a r l y stage of r e a c t i o n where the process i s assumed to be d i c t a t e d by chemical r e a c t i o n . According to the CM, the Xs jpt r e l a t i o n s h i p can be descr ibed by equation (133). Ln ( 1 - X S B ) = - K t (133) The values of the rate constant K can be determined from the slope of the l i n e a r s e c t i o n of the Ln ( l - X g a ) vs t c u r v e s . The c u r v e s represent ing runs with c a l c i n e s of d i f f e r e n t p a r t i c l e s i z e and of d i f f e r e n t c a l c i n a t i o n condi t ions are shown i n Figure 6.17 and Figure 6.18 r e s p e c t i v e l y , while data for the c o n s t r u c t i o n of these c u r v e s appear in the Appendix D. In these f i g u r e s , the c r i t e r i o n f o r the s e l e c t i o n of the l i n e a r sect ions i n the ear ly stage of conversion i s cons i s tent to that set i n sec t ion 6 .2 .1 , i n which the l i n e a r sec t ion possesses a c o r r e l a t i o n c o e f f i c i e n t of greater than 0.98. S i m i l a r to s e c t i o n 6 . 2 , the l i n e a r s e c t i o n s i n the f i r s t 180 minutes of s u l p h i d a t i o n met t h i s c r i t e r i o n and were employed i n the computation of the r e a c t i o n r a t e c o n s t a n t . The r e s u l t s obtained from regress ion ana lys i s of equation (133) using experimental data are r e p o r t e d i n 296 T H E CONTINUOUS FIRST-ORDER  MODEL, USING E X P E R I M E N T A L DATA LEGEND •=RUN #76 T=930C Dp=.68mm 0=RUN #77 T=930C Dp=1.0mm •=RUN #78 T=930C Dp=1.3mm < • - - # ) • • • -• • • • o c o o o ( • O c o n • c ] • • [ • • c - J . . . . L J . . -0 60 120 180 240 REACTION TIME, MIN. 300 F i g . 6.17 Testing the CM using experimental data of runs with d i f f e r e n t sorbent s i z e s . 297 T H E CONTINUOUS FIRST-ORDER MODEL, USING EXPERIMENTAL DATA LEGEND • =RUN #40 t=40 min. 0=RUN #54 t=120min. •=RUN #69 t=300 min. • m o • o • • o 0 o • o o * 0 60 120 180 240 REACTION TIME, MIN. 300 F i g . 6.18 Testing the CM using experimental data of runs with sorbents of d i f f e r e n t c a l c i n a t i o n conditions. 298 Table 6.4. The sorbent conversion values at the end of the ear ly step are a lso given i n t h i s t a b l e . I t i s o b s e r v e d t h a t the r a t e c o n s t a n t s f o r the su lph idat ion r e a c t i o n d e r i v e d from the CM are i n the range from 3 x l 0 - ^ to 10xl0~6 s e c - * . Numerous previous s tudies were conducted to e s t a b l i s h the rate constants f or the s u l p h i d a t i o n r e a c t i o n of d i f f e r e n t types of c a l c i n e s by means of the CM, and t h e i r reported values of K, determined from the slope of Ln ( l - X g B ) vs t are given in Table 6 .6 . Narayen [131] suggested tha t , f or the same experimental c o n d i t i o n s , the r a t e c o n s t a n t f o r c a l c i n e d d o l o m i t e i s h i g h e r t h a t t h a t of l imestone. Th i s may be true because, i n t h i s present study, even though the l i m e s t o n e was ca l c ined at the optimum c a l c i n a t i o n time and the dolomite was not , as described i n Sect ion 5 .1 , the rate constants appear to be about the same for both of the c a l c i n e s as presented i n Table 6.6. A l s o from t h i s t a b l e , the present experimental rate constants agree f a i r l y we l l wi th publ ished data . Table 6.6 RATE CONSTANTS OF THE CaO + H?S REACTION DETERMINED BY MEANS OF CM Author Present Narayen Richards Kamath Squires Schrelber Pell Study [131] [149] [101] [152] [152] [137] Conditions dps=0.68 mm dp=0.389 mm dps=0.389 mm dp=0.389 mm dp=0.389 mm dp=0.389 mm dp=0.389 mm T=930 C T=800 C T=900 C T=700 C T=900 C T=7u0 C T=800 C [HoSrO.U [H2S]=0.1% [H2S]=0.1% [rloSXUX [H2S]=0.1% [HoS]=0.1X [H2S]=0.1% K, sec" 1 1.0x1(T5 I.OXIO-5 LOxICT5 1.0x1(r5(*) 2.0x1(T5(*) 3 . 6 x i r 5 ( » ) 1.24x1(T5(*) LOxlO - 3 1x10-4(*) (*) k of c a l c i n e d dolomite. 299 6.2.3 DISCUSSION As mentioned e a r l i e r , the CM is not adequate to descr ibe the CaO + H2S r e a c t i o n a f t er the e a r l y stage where d i f f u s i o n a l res i s tance becomes dominant. The CM also f a i l s to address the v a r i a t i o n i n r e a c t i v i t i e s of c a l c i n e s of d i f f e r e n t s i zes prepared at the same c a l c i n a t i o n condi t ions such as that of t h i s present study. The d i f ference i n values of rate constants der ived from CM for runs of d i f f e r e n t p a r t i c l e s i zes reveals that the CaO + H2S r e a c t i o n departs from the CM. By c o n t r a s t , the GM describes the porous s t ruc ture of the c a l c i n e d sorbent very w e l l , p r e d i c t s a monotonic decreases i n the rate of sorbent convers ion, as observed i n t h i s research, and i s a p p l i c a b l e to both the e a r l y - s t a g e , where the process i s known to be k i n e t i c a l l y c o n t r o l l e d , and the l a t e r - s t a g e , where the product layer d i f f u s i o n contro l s as i s normally observed i n the s u l p h i d a t i o n r e a c t i o n . 300 CHAPTER 7 CONCLUSION . At the outset of t h i s research, no adequate data were a v a i l a b l e for the k i n e t i c s of the steam g a s i f i c a t i o n of o i l sand cokes, and no data were a v a i l a b l e f o r the k i n e t i c s of the combined g a s i f i c a t i o n - i n - s i t u desu lphur iza t ion using calcium-based sorbents . The primary aim of t h i s research has been f u l f i l l e d in that the k i n e t i c data f o r these processes have been o u t l i n e d and relevant k i n e t i c constants evaluated. Bench-scale s t i r r e d and f ixed bed reactors were employed to conduct the e x p e r i m e n t a l work i n the temperature range of 8 0 0 ° C - 9 3 0 < > C , at atmospheric pressure and with p a r t i c l e s izes between 0.14 mm and 3.5 mm. In the g a s i f i c a t i o n experiments, wi th steam p a r t i a l pressure from 15.15 kPa to 60.6 kPa both delayed and f l u i d cokes were examined. In the i n -s i t u d e s u l p h u r i z a t i o n process , the inf luences of c a l c i n a t i o n condi t ions and of s o r b e n t t y p e s on the s u l p h u r c a p t u r e and on the s o r b e n t c o n v e r s i o n were i n v e s t i g a t e d . K i n e t i c r e s u l t s were o b t a i n e d by measuring the gas composition and flow rates with t ime. 7.1 SUMMARY OF FINDINGS The p r i n c i p a l observat ions and conclusions r e s u l t i n g from t h i s study are l i s t e d below: 1. The o i l sand cokes were found to be r e l a t i v e l y unreact ive to steam g a s i f i c a t i o n at temperatures below 8 0 0 ° C . This i s due to the high temperature treatment in the coking processes from which the cokes 301 were produced. Increasing temperature and decreasing coke p a r t i c l e s i ze increase both carbon and sulphur conversions i n both delayed and f l u i d cokes. In our research the g a s i f i c a t i o n was done wi th and without the presence of sorbent , and i n the former case the sdphur capture was done i n - s i t u . Since there has been no c a t a l y t i c e f f ec t on the carbon conversion of the a d d i t i o n of c a l c i n e d sorbent , and s ince the su lphur capture r e s u l t s agree wi th those i n the l i t e r a t u r e c a r r i e d out i n the absence of any g a s i f i c a t i o n s t e p , t h e g a s i f i c a t i o n and the sulphur capture processes may be analyzed independently. R e g a r d l e s s of the o p e r a t i n g c o n d i t i o n s such as temperature , r e a c t i o n time and coke p a r t i c l e s i z e , the s p e c i f i c area of delayed coke depends s o l e l y upon the extent of carbon convers ion based on the i n i t i a l carbon f o r g a s i f i c a t i o n . The measured s p e c i f i c area of delayed coke at var ious extents of carbon conversion a t ta ined a maximum at a carbon conversion of approximately 27%, and decreases as the r e a c t i o n proceeds to h i g h e r c o n v e r s i o n . The maximum s p e c i f i c area , of about 46 m 2 / ( g i n i t i a l carbon f o r g a s i f i c a t i o n ) , was approximately f i v e times that of the s t a r t i n g m a t e r i a l . Scanning e l e c t r o n microscopy examination of delayed o i l sand coke revealed that the coke has l i t t l e or no p o r o s i t y . At a carbon conversion of approximately 7%, the presence of small pores was evident on the coke surface . Small pore systems were found to appear on the e n t i r e surface of the p a r t i c l e at a carbon conversion of about 18%. This was confirmed by mercury porosimetry. The 302 pores were gradua l ly enlarged as g a s i f i c a t i o n proceeded. At a carbon conversion of 70% or h igher , the pores were wide open, and, eventua l ly d i s i n t e g r a t i o n of the coke p a r t i c l e occurred . Results of S . E . M . examination of f l u i d coke ind ica ted that the coke p a r t i c l e s are s p h e r i c a l and have no p o r o s i t y . At a carbon conversion of about 30%, the coke p a r t i c l e s cracked in to halves , exposing c l e a r l y the o n i o n - l i k e i n t e r n a l s t ruc ture of the coke. L a y e r s of c a r b o n were g r a d u a l l y consumed as the g a s i f i c a t i o n continued to a higher extent of carbon convers ion, causing the s p h e r i c a l halves to become hollow s h e l l s . Carbon conversion-t ime curves corresponding to runs of d i f f e r e n t experimental condi t ions possess i n f l e c t i o n points which l e a d to maxima i n the rates of carbon convers ion. The rates of carbon conversion increase with increas ing temperature and wi th decreasing coke p a r t i c l e s i z e s . The maxima i n the rates of carbon conversion were caused by the combined e f f ec t s of high r e a c t i v i t y due to newly developed small pores and the increas ing newly formed surface area p r i o r to reaching to i t s maximum. The rates of carbon and sulphur conversion were l i n e a r l y af fected by the steam p a r t i a l pressure i n the gas i fy ing medium at P <30.3 H 2 ° kPa. The inf luence became much weaker at a steam p a r t i a l pressure of about 30.3 kPa and above. The r e a c t i v i t y of sulphur i n delayed coke toward steam g a s i f i c a t i o n was higher than that of carbon at temperature of 8 0 0 ° C or lower; t h i s could be re la ted to the C-S bonds being weaker than C-C bonds. Above 8 5 0 ° C the rates of carbon conversion in steam g a s i f i c a t i o n of 303 delayed coke increase more d r a s t i c a l l y than that of the sulphur convers ion . Resul ts from g a s i f i c a t i o n of delayed and f l u i d cokes at the same experimental operating condi t ions were compared. Carbon conversion and t h e r a t e s of carbon c o n v e r s i o n f o r d e l a y e d coke were c o n s i s t e n t l y higher than that of f l u i d coke. However, the H2S concentrat ions i n the produced gas for runs wi th delayed coke were lower than for runs with f l u i d coke, while the values of sulphur conversions of the two cokes were near ly equal . The sulphur to carbon molar r a t i o i n the produced gas of runs employing f l u i d coke was almost the same as that of the coke i t s e l f . By c o n t r a s t , the sulphur to carbon molar r a t i o i n the produced gas of runs employing delayed coke was approximately one-half of that of the f resh coke. I n - s i t u sulphur capture was invest igated using c a l c i n e s from three l imestones and one dolomite. An optimum c a l c i n a t i o n time, which gave c a l c i n e p a r t i c l e s maximum s p e c i f i c surface and smallest g r a i n s i z e , was found. For a given c a l c i n e the f r a c t i o n a l removal of H2S from the gas phase and sorbent conversion increase with increas ing surface area of the ca l c ined sample. Ninety percent H2S capture can be achieved with a Ca/S mol r a t i o of 1 and 2 f o r two and three hours r e s p e c t i v e l y with e i t h e r c a l c i n e d limestone or dolomite . The r a t e of s o r b e n t c o n v e r s i o n of f u l l y c a l c i n e d s o r b e n t s monotonical ly decreased with convers ion, while the r a t e of the sorbent conversion of an undercalc ined sorbent a t ta ined a maximum, due to the r e - c a l c i n a t i o n of the uncalc ined CaC03 i n the p a r t i c l e s . 304 The ca lc ium-to - su lphur r a t i o and sorbent p a r t i c l e s i z e are the two v a r i a b l e s having the larges t e f fec t on the H2S removal from the produced gas and on s o r b e n t c o n v e r s i o n i n the g a s i f i c a t i o n -desu lphur iza t ion process . Increasing Ca/S r a t i o or decreasing the c a l c i n e p a r t i c l e s i ze increases the f r a c t i o n a l removal of H2S from the produced gas and decreases the sorbent c o n v e r s i o n . A l s o , increas ing Ca/S r a t i o decreases the rate at which the sorbent i s converted in to CaS. The p r e d i c t i o n s of carbon conversion versus time by both the Random C a p i l l a r y and the Modif ied Volumetric models are i n good agreement with the experimental f ind ings in g a s i f i c a t i o n of o i l sand cokes. The values of i n t r i n s i c r e a c t i o n rate constants of about 2 . 5 x l 0 - ® cm/min. atm. and of a c t i v a t i o n energy of the steam g a s i f i c a t i o n of delayed coke of 113 kJ/mol were determined using the Random Pore model. The sharp drop i n r e a c t i o n rate at high conversion could not be adequately modelled. The Gra in model was tested using the data for the P^S+CaO r e a c t i o n . The reac t ion rate constant i n the e a r l y stage of conversion of about 8.5xl0~6 m/s and the d i f f u s i o n c o e f f i c i e n t i n the l a t e r stage of conversion of 6 . 7 x l 0 ~ ^ m 2 / s were determined. Both reac t ion and d i f f u s i o n mechanisms are important i n the s u l p h i d a t i o n r e a c t i o n , the combined e f fec t of the two mechanisms was thus examined. The values of the rate constant and of the d i f f u s i v i t y are i n agreement w i t h r e p o r t e d v a l u e s . The G r a i n model descr ibes the sulphur capture process using c a l c i n e d sorbents adequately. The Continuous F i r s t - o r d e r model was employed to compute the rate constants for 305 the s u l p h i d a t i o n r e a c t i o n , which were then compared with r e s u l t s from other authors . The values of the rate constants obtained by t h i s present study are w i t h i n the range of publ ished r e s u l t s . 7.2 RECOMMENDATIONS The f o l l o w i n g recommendations are of fered for f u r t h e r work and future a p p l i c a t i o n s : 1. To achieve higher carbon conversion for the steam g a s i f i c a t i o n of o i l sand coke, temperatures higher than 930 4 C and coke p a r t i c l e s i z e smal ler than 0.68 mm should be employed. To enhance fur ther the r e a c t i v i t y of o i l sand cokes, which are r e l a t i v e l y unreact ive , t h e g a s i f i c a t i o n c o u l d be conducted w i t h the a d d i t i o n of appropriate c a t a l y s t s . The use of other g a s i f y i n g media, such as 0 2 + steam should be t e s t ed . Elevated pressure g a s i f i c a t i o n should be explored . 2. The r e t e n t i o n of m e t a l l i c compounds, e s p e c i a l l y N i and V i n the reacted coke at var ious extents of g a s i f i c a t i o n , r e q u i r e s more experimental study. 3. The development of a new k i n e t i c model with a t h e o r e t i c a l base should be undertaken to descr ibe the sharp decrease i n the rate of carbon conversion at high carbon conversion which r e s u l t s from a d r a s t i c change i n porous s t r u c t u r e . 4. The r o l e of H2S and steam, and the poss ib le ro le s of CO2, CO and CH4 i n a f f e c t i n g the rate of the su lph ida t ion r e a c t i o n of CaO s h o u l d be- e x p l o r e d . The b e h a v i o r of c a l c i n e s f r o m o t h e r 306 c a l c i n a t i o n condi t ions i n the CaO + H2S r e a c t i o n could also be e s t a b l i s h e d . 5. The use of a p i l o t - s c a l e f l u i d i z e d bed g a s i f i e r i s needed to explore the o p e r a b i l i t y of the combined g a s i f i c a t i o n of o i l sand coke and the i n - s i t u desu lphur iza t ion process . 307 REFERENCES 1. A b e l , W.T. and E . P . F i s h e r , "Limestone to Remove Hydrogen Sulphide From Hot P r o d u c e r Gas" , U n i t e d S t a t e s Energy R e s e a r c h and Development A d m i n i s t r a t i o n , MERC/RI-75/3, (1976). 2. A d s c h i r i , T . , T . Kojima, T . Furusawa and D. K u n i i , "Pred ic t ion of Dynamic Change i n C 0 2 G a s i f i c a t i o n Rate of Coal Char With Increased Conversion", Ins t . Chem. Eng. Symp. Ser ies 87, pp.345-352, (1984). 3. A d s c h i r i , T . , T . S h i r a h a , K . Ogawa and T. Furusara , "Unif ied I n t e r p r e t a t i o n of C 0 2 and Steam G a s i f i c a t i o n Rates of Coal Chars by Use of Surface Area" , Proc . I n t . C o n f . C o a l S c i . , A u s t r a l i a , pp.289-292, (1985). 4. A l l a n , D . E . and W . J . M e t r a i l e r , "Advances i n F l u i d and F l e x i c o k i n g Technology", C . E . P . , pp.40-44, (1981). 5. A l l e n , T . , P a r t i c l e Size Measurement, 3rd e d . , Champmen and H a l l L t d . , London, (1981). 6. Ambrose , P . T . , "An Assessment of the P o t e n t i a l f or U t i l i z i n g Syncrude's Byproduct F l u i d Coke", Presented at the 27th Can. Chem. Eng. Conference, Ca lgary , A l b e r t a , pp.153-181, (1977). 7. Anthony , E . J . , D . L . D e s a i and F . D . F r i e d r i c h , " F l u i d i z e d Bed Combustion of Suncor Coke", CANMET, Report No. ERP/ERL 81-27(TR), August (1981). 8. Anthony , E . J . , J . R . Stephenson and A . P . de I r i b a r n e , "Calcium Sulphide Formation i n S o l i d Wastes From C i r c u l a t i n g F l u i d i z e d Bed Combustors", ERL D i v i s i o n Report ERP/ERL 87-101 (OPJ), CANMET, For presentat ion at N inth Int . Conf. on F l u i d i z e d bed c o m b u s t i o n , Boston, Massachusetts, May 3-7, (1987). 9. A r n i k a r , H . J . and B .D. Chaure, "The E f f e c t of E l e c t r i c F i e l d and Radia t ion on the Tracer D i f f u s i o n of A l k a l i E a r t h Ions Through C a l c i t e , Pros . Chem. Symposium, V o l . 2, 23-26, pp.144-149, (1969). 10. B a l l a l , G. and K. Zygoura lk i s , "Evolut ion of Pore Surface Area During Noncata ly t i c G a s - S o l i d React ions . 1-Model Development" Ind. Chem. Res . , V o l . 26, pp.911-921, (1987). 11. B a r b o u r , R . V . and J . C . P e t e r s e n . " M o l e c u l a r i n t e r a c t i o n of asphal t . Infrared Study of the Hydrogen Bonding B a s i c i t y of Asphal t" , A n a l . Chem. 46, pp.273-277, (1974). 12. B a r e r , R. and V . E . C o s s l e t t , Advances in O p t i c a l and E l e c t r o n  Microscopy, V o l . 9 , Academic Press , London, p.369, (1984). 308 13. B e l l a n , J . and S. Elghobashi , "Fuel-Composit ion E f f e c t s on High Temperature C o r r o s i o n i n I n d u s t r i a l / C o m m e r c i a l B o i l e r s and Furnaces: A Review"; J . Eng. Gas Turbines and Power; 107; pp.744-757, (1985). 14. Beniere , F . and C . R . A . Catlow, Mass Transport i n S o l i d s , Plenum Press , New York, P602, NATO SAI Series B: Phys ic s , V o l . 9 7 , (1981). 15. Bennett, H . A . and A. Meisen, "Experimental Determination of A i r and H2S E q u i l i b r i a Under Claus Furnace Condi t ions" , Can. J . of Chem. E n g . , V o l . 5 9 , pp.532-539, (1981). 16. Bennett , H . A . and A. Meisen, "Hydrogen Sulphide - A i r E q u i l i b r i a Under Claus Furnace Condit ions", Can. J . Chem. E n g . , V o l . 5 1 , Dec. pp.720-724, (1973). 17. B h a t i a , S .K. and D.D. Perlmutter , "Random Pore Model for F l u i d -S o l i d React ions: 1-Isothermal, K i n e t i c C o n t r o l " , AICHE J . , V o l . 2 6 , No.3 , pp.379-386, (1980). 18. B h a t i a , S .K. and D.D. Per lmutter , "The E f f e c t of Pore Structure of F l u i d - S o l i d React ions: A p p l i c a t i o n to the S02~lime Reaction", AICHE J . , V o l . 2 7 , No.2 , pp.226-234, (1981). 19. B h a t i a , S .K. and D.D. Perlmutter , "The E f f e c t of Product Layer on the K i n e t i c s of the C02~Lime Reaction", AICHE J . , V o l . 2 9 , N o . l , pp.79-86, (1983). 20. B h a t i a , S . K . , M o d e l l i n g the Pore Structure of C o a l , AICHE J . , V o l . 3 3 , No.10, pp.1707-1718, (1987). 21. B h a t i a , S .K. and D.D. Perlmutter , "A Random Pore Model for F l u i d -S o l i d React ions: I: D i f f u s i o n and Transport E f f e c t s " , AICHE J . , V o l . 2 7 , No.2 , pp.247-254, (1981). 22. B h a t i a , S .K. and D.D. Perlmutter , "Unif ied Treatment of S t r u c t u r a l E f f e c t s i n F l u i d - S o l i d R e a c t i o n s " , AICHE J . , 29, p p . 2 8 1 - 2 8 9 , (1983). 23. B l a k e , J . H . , G.R. Bopp, J . F . Jones, M.G. M i l l e r and W. Tambo, "Aspects of the R e a c t i v i t y of Porous Carbons with Carbon Dioxide", F u e l , 46, pp.115-119, (1967). 24. B o r g w a r d t , R . H . , " K i n e t i c s of R e a c t i o n of SO2 With Calc ined Limestone", Env. S c i . and T e c h . , V o l . 4 , N o . l , pp.59-63, (1970). 25. B o r g w a r d t , R . H . , " C a l c i n a t i o n K i n e t i c s and S u r f a c e A r e a of Dispersed Limestone P a r t i c l e s " , AICHE J . , V o l . 3 1 , N o . l , pp.103-111, (1985). 309 26. Borgwardt, R . H . and K . R . Bruce, "Effect of S p e c i f i c Surface Area on R e a c t i v i t y of CaO with S0 2 ", A I C H E J . , V o l . 3 2 , No.2 , pp.239-246, (1986) . 27. Borgwardt, R . H . , "Surface Area of Calcium Oxide and K i n e t i c s of Calcium Sulphide Formation", Environmental Process , V o l . 3 , No. 2, pp.129-135, (1984). 28. Borgwardt, R . H . and R.D. Harvey, "Propert ies of Carbonate Rocks R e l a t e d to S0 2 R e a c t i v i t y , " Environment Science and Technology, V o l . 6 , No.4, pp.350-360, (1972). 29. Bowen, J . H . and C . K . Cheng, "A Dif fuse Interface Model for F l u i d -S o l i d React ion , Chem. Eng. Science, 24, pp.1829-1831, (1969). 30. Boynton, R . S . , Chemistry and Technology of Lime and Limestone, 2nd e d . , John Wiley and Sons I n c . , New York, N . Y . (1980). 31. Braunuer S . , P . H . Eraraett and E . T e l l e r , "Adsorption of gases in multimole layers" , J . Amer. Chem. S o c , 60, pp.309-345, (1938). 32. B r o n i k o w s k i , T . , "Model S e l e c t i o n f o r Aqueous S l u r r y C o a l Desu lphur iza t ion" , F u e l , V o l . 6 3 , pp.116-120, (1984). 33. Campbell , F . R . , A.W.D. H i l l s and A. P a u l i n , "Transport Propert i e s of Porous Lime and T h e i r Influence on the Decomposition of Porous Compacts Calcium Carbonate", Chem. Eng. S c i . , V o l . 2 5 , pp.929-942, (1970). 34. Carvalho , R . J . , K i n e t i c s of the Boudouard for Low-Rank Western- Canadian Coa l s , Ph.D. t h e s i s , the U n i v e r s i t y of B r i t i s h Columbia, Canada, p.296, (1986). 35. C h a n , R . K , , K . S . M u r t h i and D. H a r r i s o n , "Thermogravimetr ic A n a l y s i s of Ontario L imes tones and D o l o m i t e s I . C a l c i n a t i o n , Surface Area , and Poros i ty" , Can. J . of Chemistry, V o l . 4 8 , pp.2972-2978, (1970). 36. Chan R . K , K . S . Murthi and D. H a r r i s o n , "Thermogravimetric Ana lys i s of Ontario Limestones and Dolomites I I . C a l c i n a t i o n , Surface Area, and Poros i ty" , Can. J . of Chemistry, V o l . 4 8 , pp.2979-2982, (1970). 37. Chang, Y . and D.D. Per lmutter , "Effect of P o r e - D i s t r i b u t e d Coke on C a t a l y s t Regeneration K i n e t i c s " , AICHE J . , V o l . 6 , No.6 , pp.940-951, (1987) . 38. C h i n , G . , S. Kimura, S. Tones and T. Otake, " G a s i f i c a t i o n of Coal Char With Steam (part 2) . Pore s tructure and r e a c t i v i t y " , Int . Chem. Eng. V o l . 2 3 , N o . l , pp.113-120, (1983). 310 39. Ch.orn.et, E . , Baldasano, J . M . and H . T . T a k i , "Kinet i c Expressions For Coal Char-Gas Reactions" Fuel 58, pp.395-400, (1979). 40. Christman, P . G . and T . F . Edgar, "Dis tr ibuted Pore-S ize Model for S u l f a t i o n of Limestone", AICHE J . , V o l . 2 9 , N o . 3 , p p . 3 8 8 - 3 9 5 , (1983). 41. Chrostowski , J .W. and C . Georgakis , "Pore Plugging Model for Gas-S o l i d Reactions", Chem. Reaction E n g i n e e r i n g , H o u s t o n , V o l . 1 9 , pp.225-237, (1978). 42. C l a r k , P . D . and J . B . Hyne, "Chemistry of Organosulphur Compound Types Occurr ing i n Heavy O i l Sands. 3-Reaction of Thiophene and tetrahydrothiophene wi th vanadyl and n i c k e l s a l t s " , F u e l , V o l . 6 3 , Dec. 1984, pp.1649-1654. 43. C l a r k , P . D . , J . B . Hyme and J . D . T y r e r , "Some C h e m i s t r y of Organosulphur Compound Types Occurring i n Heavy O i l Sands", F u e l , V o l . 6 3 , pp.125-128, (1984). 44. C l a r k , P .D . J . B . Hyne and J . D . T y r e r , "Chemistry of Organosulphur Compound Types Occurr ing i n Heavy O i l Sands I . High Temperature H y d r o l y s i s and Thermolysis of Tetrahydrothiophene i n R e l a t i o n to Steam St imula t ion Processes", F u e l , V o l . 6 2 , pp.959-962, (1983). 45. Clugston , D.M. George, A . E . , Montgomery, D . S . , Smiley, G . T . and Sawatzky, H . , "Sulfur Compounds i n O i l from the Western Canada Tar B e l t , " Am. Chem. S o c , Adv. Chem. S e r . , p. 151, (1976). 46. C o u t u r i e r , M . F . , SO? Removal in F l u i d i z e d Bed Combustors, Ph .D. Thes i s , Queen's U n i v e r s i t y , Kingston , Ontar io , Canada, (1986). 47. Demo, J . J . , "Hot Ash Corros ion of High Temperature Equipment"; M a t e r i a l s Performance, V o l . 1 9 , pp.9-15, (1980). 48. Dogu, T . , "The Importance of Pore Structure and D i f f u s i o n in the K i n e t i c s of Gas S o l i d N o n - c a t a l y t i c Reactions: React ion of Ca lc ined Limestone with SO2", Chem. Eng. J . , 21, pp.213-222, (1981). 49. Dutta , S . , C . Y . Wen and R . J . B e l t , "React iv i ty of Coal and Char. I . In Carbon Dioxide Atmosphere", Ind. Eng. Chem. Process . Des. Dev. V o l . 1 6 , N o . l , pp.20-30, (1971). 50. Dutta , S. and C . Y . Wen, "React iv i ty of Coal and Char 2. In Oxygen-Nitrogen Atmosphere", Ind. Eng. Chem. Process . Des. Dev. V o l . 1 6 , N o . l , pp.20-30, (1977). 311 51. E d s t r o m , J . O . and G. B i t s i a n e s , "Sol id State D i f f u s i o n in the Reduction of Magnetite", Journa l of Metals , Transact ions AICHE J . , V o l . 2 0 3 , pp.760-765, (1955). 52. E n g l a n d , R . S . and P a r k h i l l , S . M . , "Unleashing a Black Beauty", Compressed A i r , 83, pp.5-11 , (1978). 53. E r g u n , S. and M. M e n s t e r , "Analysis of K i n e t i c s by Method of Constant Compositions", U . S . Bur. Mines. Rept. Invest . No.5502 - 9 pages (1961). 54. F a i r b r i d g e , C , A . D . Palmer, S . H . Ng and E . Furirasky, "Surface Structure and Oxidat ion R e a c t i v i t y of O i l Sand Coke P a r t i c l e s " , F u e l , V o l . 6 6 , pp.898-901, (1987). 55. F a i r b r i d g e , C . , S . H . Ng and A . D . Palmer, " F r a c t a l A n a l y s i s of Gas Adsorpt ion on Syncrude Coke", F u e l , V o l . 6 5 , pp.1759-1762, (1986). 56. F a l k e n b e r r y , H . L . and A . V . S l a c k , " S O 2 Removal by Limestone I n j e c t i o n , Chem. Eng. Process , V o l . 6 5 , No.12, pp.61-66, (1969). 57. F e l d k i r c h n e r , H . L . and J . Huebluer, "Reaction of Coal with Steam-Hydrogen Mixtures at High Temperatures and Pressure , "Ind. Eng. Chem. Process . Des. Dev . , 4, pp.134-142, (1965). 58. F o t t , P. and P. Strake , "Effect of Pore D i f f u s i o n on R e a c t i v i t y of Lump Coke", F u e l , V o l . 6 6 , pp.1281-1289, (1987). 59. French, D . N . , "Liqu id Ash Corros ion Problems i n F o s s i l Fuel F i r e d B o i l e r s " ; Proceeding Symposium on Corros ion in F o s s i l Fuel Systems, Electrochem. Soc. V o l . 8 3 - 5 , (1983). 60. F r e u n d , H . , " I n t r i n s i c G l o b a l Rate C o n s t a n t f o r the H i g h Temperature Reaction of CaO with H 2 S 1 1 , Ind. Eng. Chem. Fund am., V o l . 2 3 , pp.338-344. (1984). 61. F r i e d r i c h , F . D . and G . K . L e e , " F l u i d i z e d Bed Combust ion of Petroleum Coke", The American Society of Mechanical Engineers , 82-JPGC-Fu-3 , (1983). 62. Fuj ima, Y. and Y. F u j i o k a , "Experimental Inves t iga t ion on Sulphur Retention With Limestone Regeneration i n F l u i d i z e d Bed Combustion", F l u i d i z a t i o n , IV In ternat iona l Conference on F l u i d i z a t i o n , Japan, pp.661-668, (1984). 63. Furimsky, E . , " G a s i f i c a t i o n Studies of Onakawana L i g n i t e " , The Can. J . Chem. E n g . , V o l . 6 2 , pp.257-265, (1984). 64. F u r i m s k y , E . , " G a s i f i c a t i o n R e a c t i v i t y of Cokes Derived From Athabasca Bitumen", Fuel Process ing Technology, V o l . 1 1 , pp.167-182, (1985). 312 65. Gan, H . , S .P . Nandi and P . L . J r . Walker, "Nature of the Poros i ty in American Coals" , F u e l , 51, pp.272-277, (1972). 66. Gavalas , G . R . , "A Random C a p i l l a r y Model with A p p l i c a t i o n to Char G a s i f i c a t i o n at Chemical ly C o n t r o l l e d Rates", AICHE J . , V o l . 2 6 , No.4, pp.577-584, (1980). 67. Georgakis , C. , C.W. Chang and J . Szekely, "A Changing Gra in Size Model F o r G a s - S o l i d R e a c t i o n s " , C h e m i c a l E n g . S c i . , V o l . 3 4 , pp.1072-1075, (1979). 68. George, Z . M . and L . G . Schneider, "Coking of Bitumen from Athabasca O i l Sands. 1. E f f e c t of Calcium Hydroxide and Other Reagents", F u e l , V o l . 6 1 , pp.169-174, (1978). 69. George, Z . M . and E . L . T o l l e f s o n , "Desulphurizat ion of a F l u i d Coke S i m i l a r to the Athabasca O i l Sand Coke", F u e l , V o l . 5 7 , pp.497-501, (1978). 70. Glasson, D . R . , "React iv i ty of Lime and Related Oxides. VII C r y s t a l Size V a r i a t i o n i n Calcium Oxide Produced from Limestone", J . A p p l . Chem., 11, pp.201-206, (1961). 71. Gregg, S . J . , The Surface Chemistry of S o l i d s , 2nd e d . , Chapman and H a l l L t d . , London, (1961). 72. G r i z z l e , P . C . , J . B . Green, V . Sanchez, E . Murgia and J . Labkowitz, Chapter IX i n Proc . 2nd In t . Conf. on Heavy Crude and Tar Sands Held i n Caracas , Venezuela, V o l . 1 7 , (1982). 73. G u l l e t , B . K . and K . R . Bruce, "Pore D i s t r i b u t i o n Changes of Calc ium-Based Sorbents Reacting wi th Sulphur Dioxide", AICHE J . , V o l . 3 3 , No.10, (1987). 74. H a l l , E . S . and E . L . T o l l e f s o n , "Upgrading of Delayed and F l u i d Cokes from O i l Sand by Desulphurizat ion", C a n . J . Chem. E n g . , V o l . 6 0 , pp.418-424, (1982). 75. H a l l , E . S . and E . L . T o l l e f s o n , "Physical Parameters that Contro l the Rates and Extents of Hydro Desulphurizat ion of O i l Sand Cokes", Presented at the 35th Can. Chem. Eng. Conference, Ca lgary , pp.296-301, (1985). 76. Hamer, C . A . , D . T . Liang and J . B . Kearns, "Combustion T r i a l s with Syncrude Coke and Minto Coal i n the Minera l Sciences Laborator ies C i r c u l a t i n g F l u i d i z e d Bed Combustor: Sulphur and Metals Capture and M e t a l l i c Corros ion" , CANMET, D i v i s i o n r e p o r t MSL 87-92 ( I R ) , (1987). 313 77. Hamer, C . A . , "Methods of Improve Limestone U t i l i z a t i o n in F l u i d i z e d Beds", D i v i s i o n report ERP/MSL84-77 (IR), CANMET, pp.1-24, (1984). 78. Hamer, C . A . , "SO2 Sorbent U t i l i z a t i o n in F l u i d i z e d Beds", D i v i s i o n report ERP/MSL83-36(OP), CANMET, pp.1-10 (1983). 79. Hamer, C A . , "SO2 Sorbent U t i l i z a t i o n i n F l u i d i z e d Beds, Combustion and Emission Abatement Parameters", D i v i s i o n report ERP/MSL 83-6 ( IR) , CANMET, pp.1-69 , (1983). 80. Har, S . H . H . , C h a r a c t e r i z a t i o n of O i l Sands F l u i d Coke, MASC T h e s i s , The U n i v e r s i t y of A l b e r t a , (1981). 81. Hartman, M. and R.W. Coughl in , "Reaction of Sulphur Dioxide with Limestone and the Grain Model", AICHE J . , V o l . 2 2 , No.3 , pp.490-497, (1976). 82. Hartman, M. and R.W. Coughline, "Reaction of Sulphur Dioxide with Limestone and the Influence of Pore Structure" , Ind. Eng. Chem. Process Des. Dev . , V o l . 1 3 , No.3 , pp.248-253, (1974). 83. Hartman, M . , J . Pata and R.W. Coughl in , Influence of P o r o s i t y of Calcium Carbonates on t h e i r R e a c t i v i t y of Sulphur Dioxide", Ind. Eng. Chem. Process Des. Dev. , V o l . 1 7 , No.4, pp.411-419, (1978). 84. Harvey, R .D . and K. George, "Behavior of Dolomite i n Absorpt ion of H2S from Fuel Gas, World Mining and Metal Technology, Proceeding the J o i n t MMIJ-AIME Meeting, 1, 163-188, ATME, New York, N . Y . , (1976). 85. Hasatan i , M . , M. Avai and M. Yuzawa, "Two-Stage F l u i d i z e d Bed Coal Combustion: Simultaneous Removal of H2S and SO2 by Porous C a l c i n e Under F u e l - R i c h Combus t ion C o n d i t i o n s " , F l u i d i z a t i o n , IV I n t e r n a t i o n a l Conference on F l u i d i z a t i o n , Japan, pp.610.1-610.8, (1984). 86. Hashimoto, K . , K. Miura , F . Yoshikawa and I . Imai, "Change i n Pore S t r u c t u r e of Carbonaceous M a t e r i a l s D u r i n g A c t i v a t i o n and A d s o r p t i o n Performance of A c t i v a t e d Carbon", Ind. Eng. Chem, Process , Des. Dev. , V o l . 1 8 , N o . l , pp.72-80, (1979). 87. Hayduk, W. and H. Pahlevanzadeh, "The S o l u b i l i t y of Sulphur dioxide and Hydrogen Sulphide i n A s s o c i a t i n g S o l v e n t s " , The C a n a d i a n J o u r n a l of Chem. E n g . , V o l . 6 5 , pp.299-307, (1987). 88. H i l l s , A . W . D . , "The Mechanism of the Thermal Decomposition of Calcium Carbonate", Chem. Eng. S c i . , V o l . 2 3 , pp.297-230, (1968). 89. H i l l s , A . W . D . , " E q u i l i b r i u m Decomposit ion Pressure of Calcium Carbonate Between 700°C and 9 0 0 ° C " , Trans . Inst . Min . M e t a l l , Sect . C , V o l . 7 6 , pp.C241-C245, (1967). 314 90. Hodgson, G.W. , E . Peak and B . L . B a k e r , " O r i g i n of Petroleum Porphyrins - P o s i t i o n of the Athabasca O i l Sands", Res. Counci l A l b e r t a , Inform. Ser. No.45. pp.75-100, (1963). 91. Hoffman, E . J . , Coal g a s i f i c a t i o n , The Energon Company Publ i sher , Maramie, Wyoming, USA, (1981). 92. I sh ida , M. and C . Y . Wen, "Comparison of Zone-Reaction Model and U n r e a c t e d - C o r e S h r i n k i n g Model i n S o l i d - G a s R e a c t i o n s - I I s o t h e r m a l A n a l y s i s " , Chem. E n g . S c i . , V o l . 2 6 , pp. 1031-1041, (1971). 93. I sh ida , M. and C . Y . Wen, "Comparison of K i n e t i c and D i f f u s i o n a l Models f o r So l id-Gas Reactions", AICHE J . , V o l . 1 4 , pp . 311-317, (1968). 94. J a c k , T . R . and E . A . S u l l i v a n , "Comparison of the Structure and Composition of Cokes From the Thermal Cracking of Athabasca O i l Sands Bitumen", F u e l , V o l . 5 8 , pp.585-588, (1979). 95. Jaycock, M . J . and G.D. P a r f i t t , Chemistry of Inter faces , John Wiley and Sons L t d . , New York, (1981). 96. J e n s e n , K . F . , "Pore D i f f u s i o n Model of Char G a s i f i c a t i o n with Simultaneous Sulphur Capture", Chem. Reaction E n g . , V o l . 2 8 , pp.335-346, (1982). 97. Johnson, J . L . , In Coal G a s i f i c a t i o n , Advances i n Chemistry Ser i e s , No. 131, Am. Chem. S o c , Washington D . C , (1974). 98. Johnson, J . L . , Fundamentals of Coal G a s i f i c a t i o n , Chemistry of Coal U t i l i z a t i o n , 2nd Supplementary Volume, Mart in A. E l l i o t t ( e d . ) , John Wiley and Sons Pub l i sh ing C o . , New York C i t y , pp. 1491-1598, (1981). 99. J o u , F . Y . , F . D . Otto and A . E . Mather, " E q u i l i b r i a of H2S and CO2 i n Triethanolamine Solut ions" , The Can. J . of Chem. E n g . , V o l . 6 3 , pp.122-125, (1985) 100. K a i r a i t i s , D . A . , J . G . Smi th and R . J . T y l e r , " R e a c t i v i t y of Petroleum Coke to Steam", Proceedings A u s t r a l i a n Conference on Coal Science , C h u r c h i l l , V i c t o r i a , pp.106-112, (1984). 101. Kamath, V . S . and T .W. P e t r i c , Rate of R e a c t i o n of Hydrogen Sulphide-Carbonyl Sulphide Mixtures wi th F u l l y Ca lc ined Dolomite, Environment Science and Technology, V o l . 1 5 , No.8, Aug.1981, pp.966-968. 102. Karaiya, K. and K. Suzuki , P r e f e r e n t i a l Alignment of Carbon Layers 315 Around Pores i n Hard Carbon and Mult i -Phase G r a p h i t i z a t i o n , Carbon, V o l . 1 3 , pp.317-320, (1975). 103. Karim, G . A . and D. Mohindra, "A K i n e t i c Inves t iga t ion of the Water Gas S h i f t React ion Inhomogeneous Systems", Journa l of the I n s t i t u t e of F u e l , V o l . 4 7 , pp.219-223, (1974). 104. Kasaoka, S . , Y . Sakata and C . Tong, "Kinet i c E v a l u a t i o n of the R e a c t i v i t y of Various Coal Chars for G a s i f i c a t i o n w i t h Carbon Dioxide in Comparison With Steam", Int . Chem. E n g . , V o l . 2 5 , pp.160-175, (1985). 105. Kasaoka, S . , Y . Sakata, S. Kayano and Y . Masuoka, "The Development of Rate E x p r e s s i o n s and the E v a l u a t i o n o f R e a c t i v i t y f o r G a s i f i c a t i o n of Various Coal Chars With Steam and Oxygen", Int . Chem. E n g . , V o l . 2 3 , No.3 , pp.472-484, (1984). 106. Kasaoka, S . , Y. Sakata and M. Shimada, "The Sulphur Tolerance of Various C a t a l y s t s f o r the G a s i f i c a t i o n of Coal Char", Int . Chem. E n g . , V o l . 9 6 , No.4, pp.705-715, (1986). 107. K a s a o k a , S. and Y . Sakata, "Research on Coal L i q u e f a c t i o n and G a s i f i c a t i o n " , Resul ts of Spec ia l Projec t Research on Energy Under G r a n t i n A i d of S c i e n t i f i c Research of M i n i s t r y of Educat ion , Science and C u l t u r e , Japan, pp.239-248, Apey 12, (1984). 108. Kasaoka, S . , Sakata, Y . and M. Shimada, F u e l , V o l . 6 6 , pp.697-701, (1987). 109. Kasaoka, S. and Y. Sakata, "A C o r r e l a t i o n of the R e a c t i v i t i e s of G a s i f i c a t i o n of Coal Chars with Propert i e s of the Parent Coals", World congress III of Chem. E n g . , Tokyo, Japan, pp.505-508, (1986). 110. K e a i r n s , D . L . and R . A . Newby, "What i s New in High Temperature Sulphur Removal Systems f o r F l u i d i z e d - B e d C o a l G a s i f i c a t i o n " , J o u r n a l of the A i r P o l l u t i o n Contro l A s s o c i a t i o n , V o l . 2 8 , No.4, pp.328-333, (1978). 111. K e n t , R . L . , B . E i s e n b e r g , " B e t t e r Data f o r Amine T r e a t i n g " , Hydrocarbon Process ing , pp.87-90, (1976). 112. K a c a e f e , D. , D. Karman, F . R . Steward, "Interpre ta t ion of the Sulphat ion Rate of CaO, MgO and ZnO with SO2 and SO3", AICHE J . , V o l . 3 3 , No.11, pp.193.5-184.3, (1987). 316 113. K o s e o g t u , R . O . and C R . P h i l l i p s , " K i n e t i c of N o n - c a t a l y t i c Hydrocracking of Athabasca Bitumen", F u e l , Volume 66, pp.741-748, (1987). 114. K u t l e r , A . A . , L . W . Z a b n s t e c h e r , "Here i s What New in Delayed Coking", The O i l and Gas J o u r n a l , pp.92-96, (1970). 115. L a n g , R . J . and R . E . N e a v e l , "Behav ior of Calc ium as a Steam G a s i f i c a t i o n C a t a l y s t " , F u e l , V o l . 6 1 , pp.620-626, (1982). 116. Lee, S . , J . C Angus, R . V . Edwards and N . C Gardner, "Non-Catalyt ic Coal Char G a s i f i c a t i o n , " AICHE J . , V o l . 3 0 , N o . 4 , p p . 5 8 3 - 5 9 3 , (1964). 117. L e e , D . C . and C . Georgakis , "A S ing le , P a r t i c l e - S i z e Model of Sulphur Retention i n F l u i d i z e d Bed Coal Combustor" , AICHE J . , V o l . 2 7 , No.3 , pp.472-481, (1981). 118. L e v e n s p i e l , 0 . , Chemical Reaction Engineer ing , John Wiley and Sons, New York, 2nd e d . , (1972). 119. Lewis , H . M . , "Experience i n Burning Petroleum Coke from Athabasca Tar Sands", 34th Canadian Chem. E n g . C o n f e r e n c e , p p . 1 5 1 - 1 5 4 , Quebec, (1984). 120. L indberg , W.R. , R.R. Thomas and R . J . Chr i s tenson , "Measurements of S p e c i f i c Heat, Thermal Conduct iv i ty and Thermal D i f f u s i v i t y of Utah t a r sands", Fuel 64, pp.80-84, (1985). 121. L u t h r a , K . L . and H . S . S p a c i l , "Impurity Deposit i n Gas Turbines From Fuels Containing Sodium and Vanadium"; J . Electrochem S o c ; 129:3, pp.649-656, (1982). 122. Lynch, D . C . and J . F . E l l i o t t , "Analysis of the Oxidat ion Reaction of CaS Meta l" . Transact ion B . , Volume 11B, pp.415-425, (1969). 123. Mahajan, O . P . , R. Yarzab and P . L . Walker J r . , " U n i f i c a t i o n of Coal G a s i f i c a t i o n R e a c t i o n Mechanisms", F u e l , V o l . 3 7 , p p . 6 4 3 - 6 4 7 , (1978). 124. Marsh, H . and R.R. A d a i r , " C a t a l y t i c G a s i f i c a t i o n of Doped Carbon -A k i n e t i c study", Carbon, V o l . 1 3 , pp.327-332, (1975). 125. Massey, L . G . , Goal G a s i f i c a t i o n , Advances in Chemistry Series 131, American Chemical Soc ie ty , Washington, D . C , (1974). 317 126. M c C l e l l a n , G . H . and S.R. Hunter, X-Ray and E l e c t r o n Microscopic  S t u d i e s of C a l c i n e d and S u l p h a t e d L i m e s t o n e s , ASTM S p e c i a l Technica l P u b l i c a t i o n 472, (1969). 127. Muhlen , H . J . , K . H . V . Heek and H. J i i n t g e n , "Kine t i c Studies of Steam G a s i f i c a t i o n of Char i n the Presence of H2, CO2 and CO", F u e l , V o l . 6 4 , pp.944-950, (1985). 128. Mossop, G . D . , "Geology of the Athabasca O i l Sands", Science, Vo l .207 , 145-152, (1980). 129. Murray, J . A . , H . C . F i s h e r and D.W. Sabean, "The E f f e c t of Time and Temperature of Burning on the Propert i e s of Quick Lime Prepared from C a l c i t e " , American Society of Tes t ing M a t e r i a l s Proceedings, V o l . 5 0 , pp.1263-1277, (1950). 130. M u r t t r i , K . S . ; D. Harr i son and R .K. Chan, "Reaction of Sulphur Dioxide wi th Ca lc ined Limestones and Dolomites", E n v i r o n . Science and Technology, V o l . 5 , No.9 , pp.776-781, (1971). 131. Narayen, H . , Rate of React ion of Hydrogen Su l f ide with Ca lc ined  Do lomi te and C a l c i n e d Pure C a l c i u m C a r b o n a t e , M . S . T h e s i s , U n i v e r s i t y of Southern I l l i n o i s at Carbondale, (1976). 132. Newby, R . A . , D.M Bachovch in , C . H . Peterson, N .D . Rohagi, N . H . U l e r i c h and D . L . K e a i r n s , "An Assessment of Advanced S u l f u r Removal System f o r E l e c t r i c U t i l i t y AFBC", Westinghouse R&D Research Centre P i t t s b u r g h , pp.1044-1059, (1980). 133. Oxley, J . H . , Reactions of Sulphur During the G a s i f i c a t i o n of C o a l , Ph.D. Thes i s , Carnegie I n s t i t u t e of Technology, (1956). 134. Park, J . Y . and 0. L e v e n s p i e l , "The Cracking Core Model f o r the Reactions of S o l i d P a r t i c l e s " , Chem. Eng. S c i . 30:2, pp.1207-1214, (1975). 135. Parmar, B . S . , "Desulphurizat ion of O i l Sand Coke", The Can. J . Chem. E n g . , V o l . 5 5 , pp.185-191, (1977). 136. P e l l , M. , R . A . Graf f and A . M . Squires , "Desulphurizat ion of Fuels With Ca lc ined Dolomite". Am. Inst , of Chem. Eng. Sulphur and S0 2 Development, pp.151-157, (1971). 137. P e l l , M. , R e a c t i o n of H y d r o g e n - S u l p h i d e w i t h F u l l y C a l c i n e d Dolomite, Ph.D. Thes i s , The C i t y U n i v e r s i t y of New York, (1971). 318 138. P e r l m u t t e r , D . D . , " M o d i f i c a t i o n of Peterson's Model of Porous So l ids" , AICHE J . , V o l . 3 1 , No.2 , pp.343-345, (1985). 139. Perry , J . H . , R . H . Perry , C . H . Chi t ton and S.D. K i r k p a t r i c k , P e r r y ' s  Chemical Engineering Handbook, 4th E d , McGraw H i l l Book Company I n c . , New York, (1963). 140. Petersen, E . E . , "Reaction of Porous S o l i d " , AICHE J . , V o l . 3 , No.4, pp.443-448, (1957). 141. P e t r i e J . G . and R . K . Dutkiewicz , "Sulphur Dioxide and Nitrogen Oxides Removal i n a F l u i d i z e d Bed Combustor", F l u i d i z a t i o n , IV I n t e r n a t i o n a l Conference on F l u i d i z a t i o n , J a p a n , pp.691-698, (1984). 142. P i g f o r d , R . L . and G. S l i g e r , "Rate of D i f f u s i o n - c o n t r o l l e d Reaction Between a Gas and a Porous S o l i d Sphere: Reaction of S0 2 with CaC03 B , Ind. Eng. Chem. Process and Des. Dev. , V o l . 1 2 , N o . l , pp .85-91, (1973). 143. Probs te in , R . , Synthet ic f u e l , Chemical Eng. Ser i e s , McGraw-Hil l Company,, (1982). 144. P u l l e n , J . R . , " C a t a l y t i c Coal G a s i f i c a t i o n " , Report No. ICTIS/TR26 IEA Coal research , London, 55 pages, (1984). 145. Ramachandran, P . A . and J . M . S m i t h , " E f f e c t of S i n t e r i n g and P o r o s i t y Changes on Rates of G a s - S o l i d Reactions" Chem. Eng. J . , 14, pp.137-146, (1977). 146. R e i s , T . , To C o k e , D e s u l p h u r i z e and C a l c i n e , H y d r o c a r b o n Process ing , pp.97-102, (1975). 147. R e v a n k a r , V . V . S . , A . N . Gokarn and L . K . Doraiswamy, "Studies i n C a t a l y t i c Steam G a s i f i c a t i o n of P e t r o l e u m Coke w i t h S p e c i a l Reference to the E f f e c t of P a r t i c l e " , Ind. Eng. Chem. Res . , V o l . 2 6 , No.5 , pp.1018-1025, (1987). 148. Reyes, S. and K . F . Jensen, "Percolat ion Concepts in Model l ing of G a s - S o l i d Reactions I I I . A p p l i c a t i o n to Sulphation of Calc ined Limestone", Chem. Eng. Science , V o l . 4 2 , No.3 , pp.565-574, (1987). 149. R i c h a r d s , W . L . , R e a c t i o n o f C a l c i n e d D o l o m i t e W i t h Low  C o n c e n t r a t i o n of H?S, M . S . Thes i s , The U n i v e r s i t y of Southern I l l i n o i s at Carbondale, (1974). 150. R o b e r t s , B . E . and P . R . T r a m a i n e , " V a p o u r - L i q u i d E q u i l i b r i u m C a l c u l a t i o n s for D i l u t e Aqueous Solut ions of C 0 2 , H 2 S , NH3 and NaOH to 3 0 0 ° C " , The Can. J . of Chem. E n g . , V o l . 6 3 , pp.294-300, (1985). 319 151. Ruth, L . A . , A . M . Squires and R .A . G r a f f , "Desulphurizat ion of Fuel W i t h H a l f - C a l c i n e d D o l o m i t e " . E n v i r o n m e n t a l S c i e n c e and Technology, V o l . 6 , No.12, pp.1009-1014, (1972). 152. Schre iber , C . H . , "Effect of Carbon Monoxide on the Reaction of H2S and CaO", Transac t ion of the ASME, Vo l .100 , pp.520-524, (1978). 153. S c h r i b e r , G . H . , M . S . T h e s i s , E f f e c t of Carbon Monoxide in a  S i m u l a t e d Coal Gas Stream on the Rate of Reaction of Hydrogen  Sulphide with Calcium Oxide, The U n i v e r s i t y of Southern I l l i n o i s at Carbondale, (1977). 154. Schumacher, M . M . , "Heavy O i l and Tar Sands Recovery and Upgrading", I n t e r n a t i o n a l Technology, N . D . C . , USA, (1982). 155. Shaw, J . T . , " T h e o r e t i c a l Work on R e a c t i o n Sequences i n the G a s i f i c a t i o n of Coke by Carbon Dioxide and by Steam i n Condit ions Remote from E q u i l i b r i u m " , F u e l , V o l . 5 6 , pp.134-136, (1977). 156. Shawmon, P . , D i f f u s i o n i n S o l i d s , McGraw-Hil l Book Company, Inc. New York, p.203, (1963). 157. Shreve, R . N . and J . A . B r i n k J r . , Chemical Process Indus tr i e s , 4th E d . , McGraw-Hi l l , New York, (1977). 158. Simons, G .A . and T.W. Rawlins, Reaction of Sulphur Dioxide and Hydrogen Sulphide with Porous Calc ined Limestone, Ind. Eng. Chem. Process . Des. Dev. , V o l . 1 9 , No.4, 585-592, (1980). 159. Simons, G . A . and M . L . F inson , "The Structure of Coal Char - I : Pore Branching", Combustion Science T e c h . , 19, pp.217-222, (1979). 160. Simons, C . A . , "The Structure of Coal Char - I I : Pore Combination", Combustion Science T e c h . , 19, pp.227-231, (1979). 161. S imons , C . A . and A . R . Garman, "Smal l Pore C l o s u r e and the Deac t iva t ion of the Limestone S u l p h a t i o n R e a c t i o n " , AICHE J . , V o l . 3 2 , No.9 , pp.149-1499, (1986). 162. S imons , C . A . , A . R . Garman and A . A . Boni , "Kinet ic Rate of SO2 Sorpt ion by CaO", AICHE J . , V o l . 3 3 , No.2, pp.211-217, (1987). 163. S m i t h , J . G . , D . A . K a i r a i t i s and R . J . T y l e r , " R e a c t i v i t y of Petroleum Coke to Steam: I n h i b i t i o n by Hydrogen", Proceeding to Int . Conf. on Coal Science , 28-31 Oct , Pergamon, Sidney, (1985). 164. Soo, S . L . and R . T . Gibbs , "Steam G a s i f i c a t i o n of I l l i n o i s Coal and Coke", Proc . Intersoc . Energy Convers. Eng. C o n f . , 14th, V o l . 1 , pp.868-873, (1979). 320 165. Sohn, H . Y . and J . S z e k e l y , "A S t r u c t u r a l Model for G a s - S o l i d Reactions With a Moving Boundary - I I I . A General Dimensionless Representat ion of the I r r e v e r s i b l e Reaction Between a Porous S o l i d and a Reactant Gas", Chem. Eng. S c i . , V o l . 2 7 , pp.763-778, (1972). 166. Sohn, H . Y . and J . S z e k e l y , "A S t r u c t u r a l Model f o r G a s - S o l i d Reaction With a Moving Boundary - I V . L a n g m u i r - H i n s h e l w o o d K i n e t i c s , Chem. Eng. S c i . V o l . 2 8 , pp.1169-1177, (1973). 167. Sohn, H . Y . and Szekely, J . , "The e f fec t of i n t r a g r a i n d i f f u s i o n on the r e a c t i o n between a porous s o l i d and a gas", Chem. Eng. S c i . , V o l . 2 9 , pp.630-634, (1974). 168. S q u i r e s , A . M . and R . A . G r a f f , D e s u l p h u r i z a t i o n of Fuels With Ca lc ined Dolomite - I . Introduct ion and F i r s t K i n e t i c Resu l t s , Chem. Eng. Progress . Symp. Ser. No. 115, 67, pp.23-34, (1971). 169. Sturm, G.P . J r . , S . E . Scheppele, R .D . Grigsby and H. Hazos, Chapter IX i n Proc . 2nd Int . Cont. on Heavy Crude and Tar Sands Held in Caravas, Venezuela, V o l . 1 , (1982). 170. S u , J . L . and D . D . P e r l m u t t e r , " E v o l u t i o n of Pore Volume D i s t r i b u t i o n During G a s i f i c a t i o n " , AICHE J . , V o l . 3 0 , No.6 , pp.967-972, (1984). 171. Su, J . L . and D.D. Perlmutter , "Effect of Pore Structure of Char Oxidat ion K i n e t i c s " , AICHE J . , V o l . 3 1 , No.6, pp.973-981> (1985). 172. Su , J . L . and D . D . P e r l m u t t e r , "Simulation of Pore Growth and Coalescence", AICHE J . , V o l . 3 0 , No.5 , pp.829-831, (1984). 173. Sundaresan, S. and N.R. Amundson, "Studies i n Char G a s i f i c a t i o n - I -A Lumped Model", Chem. Eng. S c i . , V o l . 3 4 , pp.345-354, (1979). 174. Sundaresan, A . and N.R. Amundson, "Studies i n Char G a s i f i c a t i o n IV. M u l t i - s t a g e Char G a s i f i e r , " Chem. Eng. S c i . , V o l . 3 4 , pp.463-467, (1979). 175. S z e k e l y , J . and J .W. Evans and H . Y . Sohn, G a s - S o l i d React ions , Academic Press , London, (1976). 176. S z e k e l y , J . and J . W . Evans, "A S t r u c t u r a l Model for G a s - S o l i d React ion With a Moving Boundary", Chem. Eng. S c i . , V o l . 2 5 , pp.1091-1107, (1970). 3 2 1 177. S z e k e l y , J . and J . W . Evans, "A S t r u c t u r a l Model for Gas -So l id Reactions with a Moving Boundary - I I . The E f f e c t of Gra in S ize , P o r o s i t y and Temperature on the Reaction of Porous P e l l e t s " , Chem. Eng. S c i . , V o l . 2 6 , pp.1901-1913, (1971). 178. Szekely , J . , C . I . L i n and H . Y . Sohn, "A S t r u c t u r a l Model for Gas-S o l i d Reactions with a Moving Boundary - V. An Experimental Study of the Reduction of Porous Nicke l -Oxide P e l l e t s with Hydrogen", Chem. Eng. S c i . , V o l . 2 8 , pp.1975-1989, (1973), 179. Szekely, J . and H . Y . Sohn, "Reaction Between a Porous S o l i d and a Gas", Int . of Mining and Metal lurgy T r a n s . , V o l . 8 2 , pp.C92-C100, (1973). 180. Takamura, K . , "Microscopic Structure of Athabasca O i l Sand", Can. J . Chem. E n g . , V o l . 6 0 , pp.538-545, (1982). 181. Thomas, J . M . and W . J . Thomas, Introduct ion 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 , London, (1967). 182. Tominaga, H . and T. Kunugi , "Chemistry and Technology i n Cracking and G a s i f i c a t i o n of Petroleum Residues", I n t e r n a t i o n a l Chemica l Engineer ing , V o l . 1 4 , No.4, pp.753-767, (1974). 183. Turkdogan, E . T . and R . G . Olson, "Desulphurizat ion of Hot Reducing Gases with Ca lc ined Dolomite, Iron Making and Stealmaking, No.4, pp.168-176, (1978). 184. Von F r e d e r s d o r f f , C . G . and M.A. E l l i o t , Coal G a s i f i c a t i o n , in Chemistry Coal U t i l i z a t i o n Supplementary Volume, H . H . Lowry, E d . , John Wiley and Sons, New York, pp.892-1022, (1963). 185. Walker, P . L . and S. Matsumoto, C a t a l y s i s of G a s i f i c a t i o n of C o a l -Derived Cokes and Chars , F u e l , V o l . 6 2 , pp.140-149, (1983). 186. Walker, J r . P . L . and E . J . Hippo, "Factors A f f e c t i n g R e a c t i v i t y of Coal Chars", Prep, of Papers, Am. Chem. S o c , D i v . of Fuel Chem., V o l . 2 0 , No.3 , pp.45-51 (1975). 187. Walker, P . L . J r . , F . Rusinko J r . and L . G . Aus t in Gas Reactions of  Carbon, i n advances i n c a t a l y s i s , D.D. E l e y , P.W. Selwood, and P . B . Weisz, eds. Academic Press , New York, V o l . X I , pp.133-221, (1959). 188. Walker, P . L . , M. Shelef and R . A . Anderson, Chemistry and Physics of  Carbon Ed . P . L . Walker, V o l . 4 , Adward A r n o l d , London, (1968). 189. Watkinson, A . P . , G. Cheng and D . P . C . Fung, " G a s i f i c a t i o n of O i l Sand Coke", F u e l , Accepted for P u b l i c a t i o n , (1987). 190. Watkinson, A . P . , G. Cheng and C . J . Lira, "Oxygen Steam G a s i f i c a t i o n of Coals i n a Spouted Bed", Can. J . Chem. E n g . , V o l . 6 5 , (1987). 322 191. Watkinson, A . P . and J . K . Briraacombe, "Limestone C a l c i n a t i o n in a Rotary K i l n " , M e t a l l u r g i c a l Transac t ion B . , V o l . l 3 B , pp.369-378, (1982). 192. Wen, C . Y . , C o a l C o n v e r s i o n Technology, Advanced Book Program, Addison-Wesley Publ i sh ing Company I n c . , USA, (1979). 193. Wen, C . Y . , R . C . B a i l i e , C . Y . L i n and W.S. O ' B r i e n , " G a s i f i c a t i o n of S o l i d P a r t i c l e Containing Carbon", Advan. Chem. S e r . , V o l . 1 3 1 , pp.9-28, (1974). 194. Wen, C . Y . and N . T . Wu, "An A n a l y s i s of Slow React ion in a Porous P a r t i c l e " , AICHE J . , V o l . 2 2 , No.6, pp.1012-1021, (1976). 195. Wen, C . Y . and M. I sh ida , "Reaction of Sulphur Dioxide With P a r t i c l e Containing Calcium Oxide", E n v i r o n . S c i . Technology, 7, pp.703-709, (1973). 196. Wicke, E . and K . Z . Hedden, "Signi f icance of Inner Surfaces and of D i f f u s i o n Through Pores for the Combustion and G a s i f i c a t i o n of Porous Coals", E lec trochem. , 57, pp.636-641, (1953). 197. Wieber, P .R. and J . S . Halow, "Advanced IGEE Power Systems for the United States", Energy Progress , V o l . 7 , No.2 , pp.119-125, (1987). 198. Wu, P . C . , Lower W.E. and H . C . H o t t e l , "Reaction K i n e t i c s of Carbon Dioxide With Electrode Carbon P a r t i c l e s " , F u e l , V o l . 6 7 , pp.205-214, (1988). 199. Y a g i , S. and K u n i i D . , "Producer Gas-Reaction of Coke P a r t i c l e s i n a F l u i d i z e d - b e d " , 5th I n t e r n a t i o n a l Symposium on C o m b u s t i o n , Re inhold , New York, p.231, (1955). 200. Y i h , S . M . and Sun, C . C . , "Simultaneous Absorpt ion of Hydrogen Sulphide and Carbon Dioxide into Di- isopropanolamine So lu t ion" , The C . J . of Chem. E n g . , V o l . 6 5 , pp.581-585, (1987). 201. Young, J . E . , J . F i s h e r and S . H . Wong, "Laboratory Studies of G a s i f i c a t i o n K i n e t i c s for Western Coals Under Cond i t i on Expected D u r i n g I n - s i t u Combust ion" , Soc. P e t r o l . Eng. J . , pp.740-746, (1981). 202. Young, D . A . , Decomposition of S o l i d s , V o l . 1 , T o t i o 21, E d . , F . C . Tompkins, The In ternat iona l Encyclopedia of P h y s i c a l Chemistry and Chemical Phys ic s , pp.67-72. 323 APPENDICES 324 APPENDIX A; CALIBRATION OF THE SULPHUR G . C . AND OF THE ROTAMETERS 325 CALIBRATION OF THE H2S G.C FROM JAN. TO FEB. 1987 20 15 10 0 LEGEND 0=7 JAN. 1987 • =16 JAN. 1987 = 27 JAN. 1987 = 15 JAN. 1987 V=25 JAN. 1987 i 1 l v e i 0 300 600 900 1200 1500 H2S CONCENTRATION,ppm 1800 F i g . A . l C a l i b r a t i o n curve for the Gaschromatograph analyzing H2S. 326 CALIBRATION OF THE N2 ROTAMETER F i g . A.2 C a l i b r a t i o n curve for N 2 rotameter 327 CALIBRATION OF THE H2Q ROTAMETER LEGEND 0= H20 ROTAMETER 5 10 15 ROTAMETER SETTING 20 F i g . A.3 C a l i b r a t i o n curve for H„0 rotameter. 328 APPENDIX B SAMPLE CALCULATION FOR STEAM PARTIAL PRESSURE The steam p a r t i a l pressure in the gas i fy ing gas was c a l c u l a t e d as fo l lows: The pressure drop at N 2 i n l e t to reac tor i s 2.5 cm H.O X 98.07 — = 0.0024 atm 2 cm H 2 0 Thus, the t o t a l pressure of N 2 = 101.24 kPa at a temperature of 298K. Flowrate of N 2 at t h i s condi t ion i s 7.5 L m i n - * or 125 cm 3 s - * a. If temperature, T R , and pressure, Pf , i n the reac tor are 900 oC and  101 kPa, r e s p e c t i v e l y ; Volume N 2 at t h i s cond i t ion i s : T P 2 1 V = — — v N T P I 2 1 2 V At P.. „ = 0.3 » 2 ° 3 3 Thus V = ° - ^ V - — x 493 21 — - 211 376 — h U S V H 2 0 (V) 0.7 V N 2 0.7 X s 211. j / e s The s p e c i f i c volume of superheated steam at 0.1 MPa can be obtained from K e r r ' s steam t a b l e : 329 Table A . l S p e c i f i c volume of steam at var ious temperature and pressure SPECIFIC VOLUME, cm 3 /g \P[MPa] \ 0.95 0.10 0.11 T E M P , \ °C \ 800 5212 4952 4502 850 5455 5158 4711 880 5601 5212 4837 900 5698 5414 4921 930 5844 5552 5047 950 5941 5644 5131 At 9 0 0 ° C , 0.1 MPa, the s p e c i f i c volume of superheated steam i s v = 5 4 1 4 _9i5_ V H 2 0 g Thus the l i q u i d water needed i s V- n - 211.376 2 S i x 60Ls H O s min — j = 2.343 5414 cm mm We keep the t o t a l f lowrate , V^, of (N2+steam) into the reactor constant. i . e . V T = constant - V ^ ( y ) + = 493.21 2=2 + 211.376 ^ s s = 704.586 cm-330 b. At T R ° 9 3 0 ° C and atm, P„ ^ =0.3 and for a given P „ ^ of 0.3 atm: 2 2 3 3 V r\ / \ ~ 0.3 x 704.586 — = 211.376 — as i n (a) HjO (v) s s From Table A.1 V = 5552 ^ H 2 0 g Thus: o V' n . . = 211.376 S S L X ™JL x g m 2 _ 2 8 4 V ( 1 ) 3 m i n 5551 cm 3 cm 3 c. At T R = 9 3 0 ° C and Pt •= 1 atm and f o r a given ? n Q ~ Ops atm: 3 3 cm (~m V„ n . . = 0.15 x 704.586 = 105.688 — H 2 0 (v) s s 3 and V„ rt = 5552 — H 2 0 g Thus, V ' n = 105.688 — x ^ 4 x S = 1.142 - S -2 5552 cm-1 mxn The summary of V ' H Q ^ in jec ted and d i f f e r e n t P R Q and at d i f f e r e n t T R i s given in Table 3 .4 . 331 APPENDIX C TECHNIQUES OF SULPHUR REMOVAL; There i s an increas ing p u b l i c awareness of atmospheric p o l l u t i o n . The p r i n c i p a l cause of ac id r a i n i s genera l ly a t t r i b u t e d to the sulphur oxides , which are major p o l l u t a n t s wi th respect to h e a l t h , and n i trogen oxides emitted by c o a l - f i r e d power p l a n t s . Recent s tudies suggested that a 50% reduct ion i n emission of SO2 and NO may be necessary to avoid environmental damage by ac id depos i t ion [27]. The i n c r e a s i n g c o m p l e x i t y of p o l l u t i o n c o n t r o l equipment and technology poses a dilemma for the whole i n d u s t r i a l s ec tor . Numerous processes have been proposed to remove sulphur from the coa l before , during or a f t er combustion. P a r t i a l l y removing sulphur from the sulphur-bearing s o l i d fue l s can be done by pretreatment or by mechanical c l ean ing . Mechanical c leaning i s the i n i t i a l step to separate p a r t i c l e s which contain large amounts of inorganic sulphur compounds whose dens i ty i s greater than that of the r e s t . In chemical c l ean ing , the p a r t i c l e s are impregnated in a l k a l i s o l u t i o n fol lowed by a c a l c i n a t i o n step [68,69,74]. These processes only p a r t l y remove the su lphur , are expensive and reduce the q u a l i t y of the s o l i d fue l s [120]. The removal of sulphur during combustion by i n j e c t i n g pu lver i zed dry limestone and dolomite into the b o i l e r was f i r s t c a r r i e d out i n 1963 [130]. The pu lver ized p a r t i c l e s of limestone or dolomite are thermal ly decomposed at high temperature becoming ca lc ines which react with SO2 332 forming calc ium sulphate s o l i d that i s caught i n the dust c o l l e c t o r s a long w i t h the ash. The drawbacks of t h i s technique are that the c a l c i n e s are not f u l l y react ive due to high temperature c a l c i n a t i o n and s h o r t r e s i d e n c e t imes. Consequently, large amounts of sorbent are needed. These drawbacks often make t h i s technique not economica l ly a t t r a c t i v e . In r e c e n t y e a r s , t h e r e has been growing i n t e r e s t i n chemical processes for the removal of sulphur compounds from the f lue gas stream. Many of these involve treatment of the f lue gas stream from the plant wi th a l k a l i n e l i q u i d in large scrubbers p r i o r to d i scharge . Although the e f f i c i e n c y of sulphur removal i s h igh , the equipment i s c o s t l y , and i t may be d i f f i c u l t to add to many e x i s t i n g power p l a n t s . Another wet s c r u b b i n g t r e a t m e n t i n v o l v e s the r e a c t i o n between SO2 and f i n e l y p u l v e r i z e d CaC03 i n the l i q u i d phase. Limestone suspended i n water flows countercurrent ly to the e f f luent gas stream. SO2 d i s so lved In the l i q u i d phase r e a c t s w i t h CaCC"3 forming CaSC>4. To achieve a high percentage sulphur removal, a large amount of p u l v e r i z e d limestone has to be used which consequently leads to the expensive d i sposa l of a voluminous s o l i d waste p r o d u c t . D e s p i t e s u b s t a n t i a l i n t e r e s t i n s u b s e q u e n t development work on s c r u b b e r s , t h e i r o p e r a t i o n s are considered by the major users to be u n r e l i a b l e and expensive . The h i s t o r y of e f f o r t s to remove sulphur a f t er combustion from the stack gases i s therefore not encouraging [137]. A more v e r s a t i l e route to u t i l i z a t i o n of high sulphur s o l i d fuels i s through g a s i f i c a t i o n . Most of the sulphur in the s o l i d i s p r i m a r i l y converted to the form of hydrogen su lphide . The hydrogen sulphide gas 333 should be removed from the product gas b e f o r e b e i n g used i n many a p p l i c a t i o n s such as m e t a l l u r g i c a l reducing gas, f or e l e c t r i c power p lant and synthes is gas. When the product gas i s to be used in the power p lant or to generate steam, i t might be economical to remove hydrogen sulphide p r i o r to combustion ra ther than having to remove SO2 from large volume of combusted gas [183]. G e n e r a l l y , the removal of hydrogen sulphide from the product gases can be c a r r i e d out e i t h e r at low temperature (below 1 2 0 ° C ) or at high t e m p e r a t u r e (near the t empera ture at which the gas i s produced). Conventional low temperature techniques, us ing solvent-gas so lu t ions to take up the ac id gases, are capable of removing almost a l l H2S and removing about 99% of the CO2 [131]. The removal of H2S involves wet scrubbing techniques using mono-, d i - or t r i - e thanolamine [149,99] or di - i sopropanolamine so lut ions [200]. The processes have been employed by the n a t u r a l gas and petroleum i n d u s t r i e s f o r the c leaning "sour gas" and i s appropriate for the removal of H2S from a gas mixture . The disadvantage of the low temperature removal techniques i s that the d e s u l p h u r i z a t i o n process works only when the temperature of the product gas i s l o w e r e d . T h u s , l a r g e r heat exchanger s u r f a c e i s r e q u i r e d . Depending on the subsequent use of the c lean gas, a gas reheat ing step may be needed. In the case of the c lean gas to be used at h i g h p r e s s u r e i n an advanced power cyc l e [197] or subjected to f u r t h e r process ing at high temperature or burnt d i r e c t l y , the r e s u l t i n g l o s s of s e n s i b l e heat as w e l l as the f u e l v a l u e of condensable components in coo l ing the gas to low temperature makes wet scrubbing thermal ly less e f f i c i e n t [149] compared to h igh temperature H2S removal. 334 E f f i c i e n c y losses w i l l a lso occur in the use of hot reducing gas for d i r e c t reduct ion of i ron ore . U s i n g ca lc ium-based sorbents such as limestone and dolomite to capture hydrogen sulphide at high temperature may p r o v i d e a u s e f u l cleanup technique for f l u i d i z e d bed g a s i f i c a t i o n processes [110]. A s i m i l a r method f o r S0 2 removal i n f l u i d i z e d bed combustion processes has a lready been shown to be a promising technique [189,77-79]. In both processes , the h igh temperature c a p a b i l i t y of sorbent permits i n - s i t u d e s u l p h u r i z a t i o n of the producer gas with h igh sorbent u t i l i z a t i o n . 335 APPENDIX D SAMPLE CALCULATIONS OF De AND K A. CALCULATION OF De: Gra in model proposed the r e l a t i o n s h i p between r e a c t i o n time and convers ion for the l a t e r stage of as fo l lows: P t ' - 6 5 § T r e 2 I i - 3 ( i - x S B ) 2 / 3 + 2 ( 1 - X S B ) ] The p l o t of t vs [ 1 - 3 ( 1 - X S B ) 2 / 3 + 2 ( 1 - X S B ) ] w i l l y i e l d a slope of P" v CaO o S l ° p e 6DTC7 r S or: P D e = 6C g ( s lope) r g CaO 2 n i t n i "K mo 1 _ _ _ _ mo 1 e P CaO = 3.32 g/cm-3 x - j ^ — — = 0.059 cm r g = Measure by using S . E . M . photos. H S C B = Avg. Cone. HoS during the run - 1050 ppm - 1050 — 6 mole gas Assumming: v = ^ B l p i i m atm cm 3 1203K - „ , , „ „ cm 3 = 1 mole x 82.06 — — x — — = 98,718.2 — -gmol K 1 atm mole C„ - 1050x10-6 ?!2l!L2i2S 1 mole gas 8 mole gas 98,718.2 cm J 336 l rsci \n~& mole = 1.064 x 10 —rr— cm-" Run 40: slope = 46,739 min from S . E . M . photos: d g = 2.5 um -• r g = 1.25 x 10" cm 0.059 De = mol cm 6 x 46,739 min x 60 s x 1.064 x 10~ 8 mole x ( 1 . 2 5 x l 0 - 4 ) 2 cm 2 min cm-1 De = 5.167 x 10~ 9 cm 2 = 5.167 x 1 0 - 1 3 m 2 Results obtained by s i m i l a r means for other runs are shown i n Table 6.4. b. CALCULATIONS OF k s _• In the e a r l y stage of the r e a c t i o n , the g r a i n model p r e d i c t s the t ime-convers ion as fo l lows: CaO A p l o t of t vs [ 1 - ( 1 - X ) 1 / 3 ] w i l l y i e l d a slope of . CaO s l o p e " " T V r B or: k = CaO C g Slope 337 For Run 40: 0.0592 m o 1 * x (1 .25xl0" 4 ) cm , cm k = ; 1 . 0 6 4 x l 0 - 8 mole x 7360.92 min x 60 s ~ c i ? min = 1 . 9 7 x l 0 - 4 cm s S i m i l a r c a l c u l a t i o n s were performed for a l l other runs. The r e s u l t s are tabulated i n Table 6.4. DATA FOR TESTING THE G.M AND THE C.M A. For runs with calcine of d i f f e r e n t c a l c i n a t i o n conditions. Using the G.M i n the early stage of reaction. * FITTING EXPERIMENTAL DATA INTO THE GRAIN MODEL. * * USING DATA OF RUNS WITH SORBENT OF DIFFERENT * CALCINATION CONDITIONS. * * M = 1 - (1-XSB)**l/3 TIME, M FOR M FOR M FOR MIN. RUN 69 RUN 54 RUN 40 t = 300min. t = 120min. t = 40m.ii 30.0 8.841E-4 8.841E-4 4.769E-4 40.0 3.915E-3 3.915E-3 1.492E-3 50.0 5.834E-3 5.834E-3 2.406E-3 60.0 6.543E-3 6.543E-3 3.130E-3 80.0 1.102E-2 1.102E-2 5.868E-3 100.0 1.568E-2 1.448E-2 8.607E-3 120.0 1.965E-2 1.809E-2 1.153E-2 140.0 2.341E-2 2.139E-2 1.454E-2 160.0 2.688E-2 2.428E-2 1.733E-2 180.0 3.077E-2 2.762E-2 2.059E-2 200.0 3.422E-2 3.066E-2 2.359E-2 220.0 3.738E-2 3.347E-2 2.625E-2 240.0 4.100E-2 3.630E-2 2.903E-2 260.0 4.390E-2 3.918E-2 3.166E-2 280.0 4.719E-2 4.208E-2 3.422E-2 300.0 5.014E-2 4.426E-2 3.667E-2 320.0 5.326E-2 4.609E-2 3.846E-2 340.0 5.385E-2 4.793E-2 4.027E-2 360.0 5.571E-2 4.940E-2 4.208E-2 339 B. For runs with sorbent of d i f f e r e n t c a l c i n a t i o n conditions. Using the G.M i n the la t e r stage of conversion. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * FITTING EXPERIMENTAL DATA INTO THE GRAIN MODEL, * * USING DATA OF RUNS WITH SORBENT OF DIFFERENT * * * * CALCINATION CONDITION. * * * * M = 1 - 3(1-XSB)**2/3 + 2(1-XSB) * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * TIME, M FOR M FOR M FOR MIN. RUN 40 RUN 54 RUN 69 t = 40min. t = 120min. t = 300min. 30.0 6.82E-7 2.34E-6 2.34E-6 40.0 6.67E-6 4.59E-5 4.59E-5 50.0 1.73E-5 1.02E-4 1.02E-4 60.0 2.93E-5 1.28E-4 1.28E-4 80.0 1.03E-4 3.62E-4 3.62E-4 100.0 2.21E-4 6.63E-4 7.30E-4 120.0 3.96E-4 9.70E-4 1.14E-3 140.0 6.28E-4 1.35E-3 1.62E-3 160.0 8.90E-4 1.74E-3 2.13E-3 180.0 1.25E-3 2.25E-3 2.78E-3 200.0 1.64E-3 2.67E-3 3.43E-3 220.0 2.03E-3 3.29E-3 4.09E-3 240.0 2.48E-3 3.86E-3 4.90E-3 260.0 2.94E-3 4.49E-3 5.61E-3 280.0 3.43E-3 5.16E-3 6.47E-3 300.0 3.93E-3 5.70E-3 7.29E-3 320.0 4.32E-3 6.18E-3 7.94E-3 340.0 4.73E-3 6.67E-3 8.39E-3 360.0 5.16E-3 7.08E-3 8.97E-3 340 C . For runs of d i f f e r e n t sorbent sizes. Using the C M i n the early stage of reaction. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * FITTING EXPERIMEMTAL DATA INTO THE CONTINUOUS * * * * FIRST-ORDER MODEL, USING DATA OF RUNS WITH * * * * DIFFERENT CALCINE PARTICLE SIZES. * * * * M = -Ln(l-XSB) = Kt * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * TIME, -Ln(l-XSB) -Ln(l-XSB) -Ln(l-XSB) MIN. RUN 76 RUN 78 RUN 77 dia.=0.68mm dia.=1.0mm dia.=1.3mm 30. 0.002656 0.002656 0.002656 40. 0.011744 0.011744 0.011744 50. 0.017518 0.017518 0.017518 60. 0.019716 0.019716 0.019716 80. 0.033220 0.028500 0.028526 100. 0.047384 0.035850 0.037915 120. 0.060300 0.039482 0.047592 140. 0.073075 0.042800 0.056270 160. 0.084420 0.045670 0.063799 180. 0.096380 0.049064 0.071230 200. 0.107393 0.051317 0.078072 220. 0.117825 0.053195 0.082840 240. 0.128323 0.055510 0.087343 260. 0.138683 0.057402 0.091182 280. 0.148766 0.059560 0.095265 300. 0.157920 0.060782 0.098050 320. 0.165420 0.061770 0.097490 340. 0.171165 0.062675 0.100480 360. 0.177320 0.063395 0.101924 . For runs of d i f f e r e n t sorbent sizes- Using the G.M in the early stage of reaction. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * FITTING EXPERIMENTAL DATA INTO THE GRAIN MODELS USING DATA OF RUNS WITH DIFFERENT CALCINE PARTICLE SIZES. M = 1 - (1-x) ** 1/3 **************************************************** TIME, M M M MIN. RUN 76 RUN 77 RUN 78 dia.= 0.68mm dia.= 1.0mm dia.= 1.3mm 30. 0.885E-3 0.885E-3 0.885E-3 40. 3.907E-3 3.907E-3 3.907E-3 50. 5.822E-3 5.822E-3 5.822E-3 60. 6.551E-3 6.551E-3 6.551E-3 80. 11.013E-3 9.464E-3 9.456E-3 100. 15.670E-3 12.559E-3 11.880E-3 120. 19.903E-3 15.739E-3 13.074E-3 140. 24.064E-3 18.582E-3 14.166E-3 160. 27.478E-3 21.042E-3 15.108E-3 180. 31.616E-3 23.496E-3 16.222E-3 200. 35.165E-3 25.688E-3 16.960E-3 220. 38.514E-3 27.236E-3 17.576E-3 240. 41.872E-3 28.695E-3 18.333E-3 260. 45.176E-3 29.937E-3 18.952E-3 280. 48.389E-3 31.256E-3 19.658E-3 300. 51.279E-3 32.155E-3 20.057E-3 320. 53.647E-3 32.619E-3 20.381E-3 340. 55.458E-3 32.939E-3 20.675E-3 360. 57.393E-3 33.404E-3 21.090E-3 For runs with sorbent of d i f f e r e n t sorbent sizes, using the G.M i n the l a t e r stage of conversion * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * FITTING EXPERIMENTAL DATA INTO THE GRAIN * * * * MODEL, USING DATA OF RUNS WITH SORBENT * * * * OF DIFFERENT CALCINATION CONDITIONS * * * * M = 1 - 3(l-X)**2/3 + 2(1-X) VS. TIME * * * ************************************************* TIME, M(RUN 76) M(RUN 77) M(RUN 78) MIN. dia.= 0.68mm dia.= 1.0mm dia.= 1.3mm 30. 2 .35E-6 2 .35E-6 2 .35E-6 40. 4 .57E-6 4 .57E-6 4 .57E-6 50. 0 .000100 0 .000100 0 .000100 60. 0 .000128 0 .000128 0 .000128 80. 0 .000361 0 .000267 0 .000266 100. 0 .000729 0 .000469 0 .000420 120. 0 .001173 0 .000735 0 .000508 140. 0 .001710 0 .001023 0 .000596 160. 0 .002267 0 .001310 0 .000678 180. 0 .002936 0 .001630 0 .000781 200. 0 .003622 0 .001946 0 .000853 220. 0 .004336 0 .002185 0 .000916 240. 0 .005113 0 .002422 0 .000996 260. 0 .005938 0 .002635 0 .001064 280. 0 .006800 0 .002869 0 .001144 300. 0 .007620 0 .003035 0 .001191 320. 0 .008325 0 .003123 0 .001229 340. 0 .008886 0 .003183 0 .001265 360. 0 .009504 0 .003273 0 .001316 343 F. For runs with calcine of d i f f e r e n t c a l c i n a t i o n conditions. Using the CM i n the earcly stage of reaction. * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * FITTING EXPERIMEMTAL DATA INTO THE CONTINUOUS * * * * FIRST-ORDER MODEL, USING DATA OF RUNS WITH * * * * CALCINE OF DIFFERENT CALCINATION CONDITIONS * * * * M = -Ln(l-XSB) = Kt * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * TIME, -Ln(l-XSB) -Ln (1-XSB) -Ln (1-XSB) MIN. RUN 69 RUN 54 RUN 40 t = 300min. t = 120min. t = 40min. 30. 2.654E-3 2.654E-3 1.430E-3 40. 1.177E-2 1.177E-2 4.480E-3 50. 1.755E-2 1.755E-2 7.226E-3 60. 1.969E-2 1.969E-2 9.404E-3 80. 3.325E-2 3.325E-2 1.765E-2 100. 4.741E-2 4.374E-2 2.593E-2 120. 5.954E-2 5.477E-2 3.480E-2 140. 7.107E-2 6.486E-2 4.395E-2 160. 8.175E-2 7.375E-2 5.245E-2 180. 9.376E-2 8.403E-2 6.241E-2 200. 1.045E-1 9.343E-2 7.160E-2 220. 1.143E-1 1.021E-1 7.980E-2 240. 1.256E-1 1.109E-1 8.839E-2 260. 1.347E-1 1.199E-1 9.651E-2 280. 1.450E-1 1.290E-1 1.045E-1 300. 1.543E-1 1.358E-1 1.121E-1 320. 1.613E-1 1.416E-1 1.177E-1 340. 1.661E-1 1.473E-1 1.233E-1 360. 1.720E-2 1.520E-1 1.290E-1 APPENDIX E PLASMA SPECTROSCOPY RESULTS OF REPRESENTATIVE SAMPLES OF SORBENTS EMPLOYED IN THIS STUDY SAMPLE I .D. QN-1 Calcined dolomite, t=120 min. DL1 Calcined dolomite, t=90 min. QN-3 Limestone t=120 min. QN-4 Raw dolomite DETECTION LIMIT MAJOR COMPONENTS Percent (Z) S i l i c a S i 0 2 1.93 2.27 1.98 1.22 0.50 Alumina A1 2 0 3 0.12 0.17 0.50 0.17 0.30 Iron F e 2 0 3 0.21 0.21 0.22 0.11 0.05 Calcium CaO 57.2 57.0 94.8 31.4 0.01 Magnesium MgO 38.7 39.9 1.45 20.5 0.01 Sodium Na 20 0.01 0.01 0.03 0.01 0.01 Potassium K 2 0 0.03 0.03 0.04 0.02 0.01 Sulphur S0 3 - - - - 0.01 Loss on Ignition L . O . I . 0.74 L 0.25 45.6 0.01 TRACE COMPONENTS Parts Eer M i l l i o n (P.P.M.) Barium Ba L L L L 10. Manganese Mn 200. 210. 110. 150. 30. Strontium Sr 71. 77. 1460. 75. 10. Titanium T i L L L L L L = Less than detection l imit 345 APPENDIX F : TABLE F . l : D E T A I L S O F G A S I F I C A T I O N E X P E R I M E N T S A T V A R I O U S CONDITIONS. p « n = 30 .30 kPa ( E x c e p t Run 95 at P „ _ = 15 .15 k P a ) , T = H 2 ° H 2 ° T e m p e r a t u r e i n ° C , d = P a r t i c l e d i a m e t e r i n mm and t = pc Time i n m i n . Carb. Conv. Carb. Conv. Carb. Conv. Initial Carb. Overall Carb. Conv. Sp Run Conditions after Pyrol., in the over- In Gas i f . , for Gas If., Carb. inGaslf . , m2/g C mole all Process, mole mole Conv., X Q mole X T = 930 848,85 dpc = 0.68 0.55597 t = 60 0.55597 0 9.86403 0.05331 0 8.166 T = 930 81 dpc = 0.68 0.55597 t = 105 1.20902 0.65305 9.86403 0.1160 0.0662 24.400 T = 930 79 dpc = 0.68 0.55597 t = 150 2.10487 1.5489 9.86403 0.2020 0.1570 40.1616 T = 930 80 d ^ = 0.68 0.55597 t = 220 3.60765 3.05168 9.86403 0.3462 0.31012 45.2020 T = 930 83 dpc = 0.68 0.55597 t = 300 4.99747 4.4415 9.86403 0.47960 0.45027 31.2777 7 = 930 35 dpc = 0.68 0.55597 t =360 5.82512 5.26915 9.86403 0.55903 0.53418 28.5725 T = 930 86 dpc = 0.68 0.23453 t = 720 7.70303 7.4685 10.18547 0.73925 0.7333 6.7676 346 APPENDIX F: (Continued) Carb. Conv. Carb. Conv. Carb. Conv. Initial Carb. Overall Carb. Conv. Sp Run Conditions after Pyrol., In the over- In Gas If., for Gas If., Carb. InGasif., m2/gC mole all Process, mole mole Conv., X Q mole X 22 T = 850 dpc = 2 t = 360 0.21612 1.63679 1.42067 10.20388 0.157 0.1392 32.609 24 T = 880 dpc = 2 t =360 0.15698 3.14226 2.9853 10.26302 0.3C156 0.2909 47.96 20 dpc = 2 t = 360 0.14962 3.91855 3.7689 10.2704 0.37606 0.3670 45.0887 27 T = 930 dpc = 2 t = 360 0.22209 4.47643 4.25434 10.19791 0.430 0.4172 36.991 33 T = 930 dpc = 1-0 t = 360 0.38402 5.13132 4.7473 10.0360 0.49245 0.4730 30.4337 90 T = 880 dpc = 0.14 t = 360 0.27925 4.51731 4.23806 10.14075 0.43352 0.41792 40.86 91 T = 900 dpc = 0-14 t = 360 0.48683 5.77953 5.29270 9.93317 0.55466 0.53283 26.64 92 T = 930 dpc = 0.14 t = 360 0.64766 7.38774 6.77008 9.80234 0.7900 0.69066 13.59 87 T = 930 dpc = 0.14 t = 360 0.30924 5.84924 5.54000 9.62701 0.56135 0.54793 10.52 95 T = 930 dpc = 0.14 t = 360 0.29899 3.37971 30.7772 9.63726 0.32406 0.30409 31.13 T I M E M I N 10. 2 0 . 3 0 . 4 0 , 5 0 . 6 0 . 8 0 . 100 120 . 1 4 0 . 1 6 0 . 1 8 0 , 2 0 0 . 2 2 0 . 2 4 0 . 260 . 2 8 0 . 0 300 .0 3 2 0 . 0 3 4 0 . 0 3 6 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 . 0 0 U T P U T 3 0 I N T V . C A R B O N GMOLE 0 . 0 0 . 0 0 . 0 5 2 9 3 0 . 2 3 2 8 8 . 0 0 6 4 0 . 0 1 9 9 7 1 3628 14803 18753 , 1 8 0 9 0 1 4869 . 2 0 7 6 8 . 2 7 6 4 1 0 . 2 6 8 1 9 0 . 2 1 0 7 9 0 . 2 1 4 0 5 0 . 2 0 4 4 2 0 . 18617 0 . 1 7 8 6 2 0 . 1 9 0 7 6 0 . 1 7 2 3 7 0 . C . 0 , 0 . 0 , 0 . 0 . 0 , 0 , 0 U T P U T 3 5 I N T V . C A R B O N GMOLE 0 . 0 0 . 0 0 3 9 8 0 . 0 6 8 5 2 0 . 3 8 6 3 5 0 . 0 5 1 4 2 0 . 0 6 2 2 6 , 3 3 7 9 3 , 3 7 5 3 3 . 4 4 1 7 9 . 4 2 8 5 0 3 8 8 0 4 , 3 9 0 7 2 .41 135 0 . 3 9 6 5 4 0 . 3 3 5 3 5 0 . 3 5 6 1 3 0 . 3 1 4 1 1 0 . 3 2 3 6 4 0 . 3 0 2 5 6 0 . 3 0 3 2 0 0 . 2 9 4 8 9 0 U T P U T 3 3 I N T V . C A R B O N 0 U T P U T 2 0 I N T V . C A R B O N O U T P U T 2 2 I N T V . C A R B O N 0 U T P U T 2 4 I N T V . C A R B O N OUTPUT 18 I N T V . C A R B O N GMOLE GMOLE GMOLE GMOLE GMOLE 0 . 0 0 . 0 0 . 0 0 . 0 0 5 8 2 0 . 0 0 . 0 0 3 0 6 0 . 0 0 9 2 0 0 . 0 0 6 4 3 0 . 0 0 8 8 9 0 . 0 0 . 0 3 0 2 8 0 . 0 6 1 9 0 0 . 11837 0 . 0 4 9 8 9 0 . 0 1 9 7 5 0 . 3 2 4 0 9 0 . 0 8 6 8 7 0 . 0 7 9 7 4 0 . 0 6 8 8 4 0 . 0 4 0 2 5 0 . 0 6 5 8 5 0 . 0 1 9 8 0 0 . 0 0 5 9 2 0 . 0 1 6 3 5 0 . 0 0 4 7 6 0 . 0 0 7 6 0 0 , 0 1 0 5 4 0 . 0 0 5 6 6 0 . 0 0 7 1 9 0 . 0 0 3 8 2 0 . 3 4 4 1 0 0 , 2 1 6 7 2 0 . 0 6 8 8 3 0 , 1 4 123 0 . 0 0 7 4 8 0 . 3 6 6 2 6 0 . 2 8 4 4 8 0 . 0 8 0 4 1 0 , 1 6 0 9 7 0 . 0 0 8 7 8 0 . 3 7 6 8 5 0 . 2 6 9 9 6 0 , 0 8 0 2 2 0 , , 1 5 1 4 1 0 . 0 0 8 7 6 0 . 35051 0 . 2 5 0 9 2 0 . 0 7 6 9 4 0 , , 1 9 8 1 9 0 . 0 0 9 9 5 0 . 3 4 7 9 8 0 . 3 0 2 3 5 0 , . 0 8 3 9 3 0 . 1 9 2 6 2 0 , 0 0 9 2 7 0 . 34 180 0 . 2 9 2 6 8 0 , 0 8 6 9 0 0 , , 2 0 7 0 2 0 , 0 1 2 4 3 0 . 3 3 5 1 8 0 . 2 6 6 7 4 0 , , 0 9 6 8 8 0 . 1 8 4 8 6 0 . 0 1 3 6 2 0 . 3 4 2 5 4 0 . 2 6 6 1 5 0 , . 1 1 0 1 1 0 . 2 6 7 4 2 0 , 0 1 2 4 1 0 . 3 0 9 2 9 0 . 2 7 2 9 7 0 , , 1 1 3 3 9 0 . 2 5 1 2 5 0 , , 0 1 5 6 1 0 . 2 9 2 2 2 0 . 2 6 8 1 2 0 , , 1 1 8 3 3 0 . 251 17 0 . , 0 1 6 9 9 0 . 3001 1 0 . 2 4 7 4 2 0 , , 1 2 0 2 4 0 . 2 4 5 6 3 0 . 0 1 4 5 4 0 . 2 7 0 7 7 0 . 2 5 2 9 6 0 , , 1 0 9 1 3 0 . 2 4 2 9 6 0 . 0 1 4 4 2 0 . 2 5 9 5 9 0 . 2 2 9 5 4 0 . 0 9 0 8 6 0 . 2 3 0 4 1 0 . 0 1 3 9 8 0 . 2 5 9 9 3 0 . 2 5 2 4 0 0 . 0 8 7 9 9 0 . 2 2 2 9 7 0 . 0 1 2 3 9 0 . 2 4 8 1 7 0 . 21561 0 . 0 9 6 5 1 0 . 2 1 8 2 7 0 . 0 1 3 0 0 Appendix F (Cont inued) : Amount of carbon g a s i f i e d as a f u n t i o n o f t ime. 0 U T P U T 9 1 O U T P U T 9 2 0 U T P U T 9 0 0 U T P U T 8 8 0 U T P U T 8 7 > T IME MIN I N T V . C A R B O N GMOLE I N T V . C A R B O N I N T V . C A R B O N I N T V . C A R B O N INTV CARBON GMOLE GMOLE GMOLE GMOLE ipend to . 0 0 . 0 0 . 0 0 . 0 0 . 0 0 . 0 X 20 . 0 0 . 0 0 . 0 0 . 0 0 . 0 0 . 0 30 . 0 0 . 0 3 5 5 0 0 . 0 5 0 3 5 0 . 0 6 4 6 4 0 . 0 0 . 0 O o 40 . 0 0 . 2 8 6 7 5 0 . 2 9 3 6 8 0 . 0 9 3 3 6 0 . 0 3 3 7 4 0 . 0 4 4 7 8 rt H-3 C n> p. 50 . 0 0 . 0 7 8 4 6 0 . 17244 0 . 0 4 4 7 6 0 . 1 6 3 0 5 0 . 2 0 8 1 9 60 . 0 0 . 0 8 6 1 2 0 . 101 19 0 . 0 7 6 4 9 0 . 0 4 0 7 1 0 . 0 5 6 2 7 N 8 0 . 0 0 . 3 0 6 7 9 0 . 4 4 4 7 0 0 . 2 3 8 9 3 0 . 2 3 5 7 9 0 . 2 6 6 7 7 time. r 100 . 0 0 . 3 1 8 8 4 0'. 5 0 6 0 3 0 . 2 5 2 5 9 0 . 2 6 4 7 9 . 0 . 4 3 1 9 2 time. o c 3 120 . 0 0 . 3 9 6 4 7 0 . 4 4 9 7 8 0 . 3 3 2 6 1 0 . 3 6 2 2 8 0 . 4 7 5 5 6 O 140 . 0 0 . 3 8 5 1 6 0 . 5 2 2 6 5 0 . 2 8 6 4 0 0 . 3 6 9 2 4 0 . 4 4 8 7 2 E carbc 160 . 0 0 . 3 6 3 9 6 0 . 4 9 9 0 3 0 . 2 8 7 9 9 0 . 3 0 9 8 4 0 . 4 5 3 3 8 E carbc 180 , 0 0 . 3 9 8 4 6 0 . 5 6 9 5 7 0 . 3 2 1 6 2 0 . 3 6 2 8 9 0 . 4 6 5 4 5 Dn gas: 2 0 0 , 0 0 . 3 8 6 0 4 0 . 5 0 5 0 2 0 . 2 8 1 7 1 0 . 3 1 2 1 8 0 . 4 4 9 0 0 Dn gas: 2 2 0 . 0 0 . 3 8 3 10 0 . 4 6 8 8 8 0 . 2 9 8 7 3 0 . 3 1 6 0 0 0 . 3 6 6 2 3 i-h H' 2 4 0 . 0 0 . 3 6 0 4 4 0 . 4 0 5 6 7 0 . 3 1 0 3 9 0 . 2 9 2 2 5 0 . 3 4 8 8 3 ro 2 6 0 . 0 0 . 3 6 3 7 2 0 . 4 4 8 7 4 0 . 2 9 4 3 1 0 . 2 8 2 4 7 0 . 3 6 6 2 8 w 2 8 0 . 0 0 . 3 6 8 9 9 0 . 4 4 8 7 3 0 . 2 9 9 3 1 0 . 2 9 8 6 1 0 . 3 1 8 3 3 300 . 0 0 . 3 3 1 4 3 0 . 4 1087 0 . 2 5 7 6 8 0 . 2 6 2 0 6 0 . 3 1 5 5 2 c 3 rt 320 . 0 0 . 3 2 8 9 5 0 . 3 8 5 6 8 0 . 2 8 8 3 6 0 . 2 6 1 0 1 0 . 3 1 6 7 7 ion 3 4 0 . 0 0 . 3 0 2 5 1 0 . 3 5 4 6 7 0 . 2 4 9 7 6 0 . 2 4 4 9 4 0 . 2 6 9 4 1 o Hi 360 . 0 0 . 2 9 7 8 6 0 . 3 5 0 0 7 0 . 2 3 7 6 7 0 . 2 1 0 5 1 0 . 2 4 7 8 3 OO T I ME M I N .RES 20;' INTERVAL. H 2 S . G M O L E l l l l l l i ? : N T E R V A L / : I H 2 S ' -GMOLE :::|E.S:2;4|: iN'tERVAL GMOLE H2S RES.2.7:. INTERVAL GMOLE 'H2S ,;;RES.'3P:;: ^INTERVAL GMOLE #ES3.3. H 2 S : INTERVAL'H2S GMOLE RES3;.5:; > INTERVAL GMOLE : H 2 S RES 8.7 • INTERVAL GMOLE H 2 10 . 0 0 . 0 c.pl; ;qpl ;;;;;;;;;; 0: 0 .0 20 . 0 0 . 0 0. O:0i'^:"':'< • :o.o.--; • ' ;• ; :£ C. 0 . . 0 30 . 0 0 . 2 3 8 5 4 E - 0 2 0 . 9 6 8 2 9 E - 03 0.87459E •03 0. 29508E •02 0 . 16722E -02 0. 10877E-02 0 . 35026E - 0 2 0 . 0 4 0 . 0 0 . 4 25 3 4 E : 0 . 2 , ; . • 6 ; ;2 9.2:;W E:- 0.2 i ?§l3:ij:9:3|;i ;:6'3.3|OE i l l t||:|||:io;| l:P|? 0 .79517E*02 : | $ | § | § $ g § .T::Q2;: yp..57328E • 0 2 . 5 0 . 0 - C . ' . O S 2 1 E - 0 2 '. ; ; : oi i i i l l i i Wi •M. 0. WB l l l l l l l i l l i l l ;6:; 232gis' f i l l .0..; 1 584 1 E •02 60 . 0 0 . 4 9 7 2 6 E - 0 3 0 . 65378E- 03 0 . 8 0 3 0 8 E • 0 3 0. 47390E -03 0.74090E -03 0. 52140E-03 0 .52548E -03 0. 15250E - 0 2 80 . 0 0 . 4 S 2 1 5 E - 0 2 : p : 3 5 0 7 8 E i : p | | | 6 p | | ; •02 M •02 ;;S||3f|Pl; •02 | | | | | | j 4 9 l l p j l ; : : O X 6 5 : 5 . 4 J E ; .-.02 ; 0 , 1 0 4 1 4 E •0,1. , 1 0 0 . 0 . . 0 . 46328E-02 ' o ! . 27858E-•02 . 0 39746E •02 0. -02 0 . 422046 -'02: 0.68824E -O2 0..7 .4 6 '4 9E. $?:-iv' 0 .1101 17E - 0 1 1 2 0 . 0 0 . 47628E-02 0. , 30426E-•02 0.39134E -02 0. 57107E -02 0.43079E •02 0 .72204E-02 0 . 74 102E •02 0 . 1 0 2 1 1 E • 0 1 1 4010: : w ;;;|3||;i;|:; MM III 0. i | p j : | | | 0 ; 2 | : 0 .48377E: •02 0 .76265E; •02 ::p::-;3947;0E •02 ' . 5 0 . 0 M : 29185=:! IP • d.;l566l:4t 111 0.471B3E: 11 illllllillil Illlllll •02 :;6.;;9.925.4E -02 180 . 0 0 . 4 6 5 0 8 E - 0 2 0 .30627E' •02 0. 39666E -02 0. 52250E •02 0.42545E- •02 0 .65327E-02 0 .71557E ' •02 0 .98894E •02 2 0 . 0 . 0 ; 0. 39 754S-02;. ;0. 27224E IP ;O ; . ;351 .34E. -P'2: 0 ;47;|';15|. •02 0.44375E •02 0 59489E-02 O . 7 2 9 0 3 E ' 02 ; p y 9 5802E-'0.2: 22C . 0 • o:>:3 7:a V 31 :TiO?;V !$ ; i : 8769E :'6:4230-11 If! Wm Iblllflllpll i:0'il : '.6:;:92.1?4E- •02 2 4 0 . 0 0 . 3 6 8 2 5 E - 0 2 0 .27432E • 02 0. 34333E •02 0. 41 176E •02 0. 3667 1E-•02 0 .53004E-02 0 .63082E- 02 0 .92466E- 02 2 6 0 . 0 0 . 3 7 9 9 3 E . ; p 2 0 ..2go.i..9.e; J : 0 | | 0.3.5S3JE; P:?V .0. :429.;1:S|, ?.P.2: : 0 .32337E ' ;02;;: i;p.;;.5ig4B.E::j'P2;:f ::P.v;8 2.1.8 2Ef 02•:: 0 ;.;911.46E - 02. 280 ."0 0 . 3 4 1 0 S E - 0 2 ;26 i;:92l' | 0 2 | '0:;;309;23E 111! w '355'f9t: t i p \ •02' fb i tp i7 j l i ; Bl l Wk' 6.. o b s d S E - 02 300 . 0 0 . 3 1 9 2 4 E - 0 2 0 .24312E -02 0.28128E •02 0 , 33556E -02 0 . 30830E- 02 0 .44072E-02 0 . 59053E- 02 0 .88268E- 02 3 2 0 . 0 0 . 3 0 6 1 . 5 E - 0 2 0 , ; 2 3 6 8 . 4 E 0 ' . ; 2.6.6 9CE SiOV3j 0 ;3.253.3E •02 :0; .;3 3 53 4E - 02 .'0y37:S0:B|:r:02:: ;p:>52743E':- .02.'; :: .0.83 143E r 02 3 40 ..0 ', 0 . 3 0 5 9 S E - 0 2 • f i | | l l o J l 111-1' i l l '02 . p i : : ; ; p | | | | 4 f E l 111 i i s l l p i E ; - 02 . 3 6 0 . C i 0 . 3 0 1 1 3 E - 0 2 0 . 20223E •02 0.25484E -02 0. 31576E •02 0 .27113E- 02 0. 35980E-02 0 .50719E- 02 0. 77057E- 02 R E S 8 8 R E S 8 9 R E S 9 0 - RES 91 R E S 9 2 R E S 9 4 R E S S 5 .. TIME I N T E R V A L H2S I N I E R V A L H2S I N T E R V A L H2S : I N T E R V A L H2S • I N T E R V A L H2S. I N T E R V A L . H 2 S I N T E R V A L h 2 S WIN GMOLE GMOLE GMOLE GMOLE " ' GMOLE GMOLE GMOLE i'p-'.o; M. illilllii 11 Kil l : III Illlllli w. lllllll 11 0 l:-ff|ff 2 0 . 0 :dJb : 'i' ;^ : :- : :-- : ' ; ;" , : ; :- l l 0. 11. .611 III 0 . oil'''-!-' 0 fO-•'"•:' 30 .0 0 . 0 0 . 0 0. 13326E • 0 2 0. 1 7 2 8 5 E • 0 2 0 1 7 9 3 7 E - 0 2 0. 0 0 0 CO .0 0 . 3 4 1 5 5 S - C 2 •ii 3Q5 ; 28E ; - .02:. I l l .17538E 111: Ill 5C.67CE. •S 111 S75.3SE I l l i !!• 5 S 1 5 2 S fill; 11 4 6 3 5 4 E •0.2 50 . 0 0 . 1 5 3 8 7 E . - 0 2 . Q. 1 3 2 0 2 E - 02; :. 2 3 1 5 8 E l i P l :|6|| 3-:579.E fill fl? P l l : s i 1 7 0 4 9 E w 1 7 3 8 0 c - 0 2 5 0 . 0 0 . 1 5 3 3 5 E - 0 2 0 . 1 2 7 8 5 E - 02 0. 15793E - 0 2 0. 2 1 3 6 3 E - 0 2 0. 1 2 3 1 4 E - 0 2 0. 1 5 3 3 5 E - 0 2 0 1 5 6 6 8 E - 0 2 60 ;:p; 0 . 8 7 01 .5^ -0 2 : :;q; 67 2 8 J 3 | ; I I IP .B|i7.54;E 111! 0, :7,54.3;:9E III 0. l l l l l l Hit Il-• 10673E IIIJI; 1:1 .6S67.4E : 0 2 '.CO . 0 0.S)A67i'0Z- 0 6454 i e • 0 2 l l l l l l t f l - 0 2 III 7 4 4 8 5 E 1® l|o r8^?785E l:0?l l i 1.080.1.E 01 111 C 9 6 7 4 E • 0 2 1 2 0 . 0 0 . 8 3 8 5 2 E - 0 2 0 6 6 1 2 0 E 02 0 6 3 6 2 1 E • 0 2 0. 78 1 7 1E - 0 2 0 8 8 3 3 7 E • 0 2 0. 107 7 1E • 0 1 0 7 0 5 4 0 E - 0 2 1 4 0 . C i .0.3.4 67 5E- . .02 : p B 5 2 4 6 E .III 0 lillsl 111 0 imp: -02 0 9.4 9-1.5 S. •pa 0. :;105;5:2.| nil l l ; 7 0 9 8 4 E - 0 2 1 5 0 . 0 .o l8.10.55E;- ,P2. ; 11 §11111 02 0 l l l l l l -02 Iii 74iB.3:ljB 111 :t:b..; 945.1SE -02 II Wmii Illi 11 . 7.6 20 4 E. 11 1 8 0 . 0 Q . 8 3 7 1 3 E - 0 2 0 6 3 8 9 5 E 02 0 6 1 5 4 6 E - 0 2 0. 7 5 6 9 5 E - 0 2 0 9 2 3 2 0 E - 0 2 0. 1 0 3 6 9 E - 0 1 0 6 9 8 6 8 E 0 2 200 . 0 0 . 3 2 4 2 9 E - 0 2 . p 5 8 3 3 2 E •02 : : 0 .59 1.53 E l i t 0 , 7.8.40 2E mi w. 9048.8E l l l l l 0. 1024 7E; 111 •II:: 6 7.70BE 0 2 2 2 0 . 0 C : 8 2 i 2 9 E - 0 2 . 0 l l l i l t l -03:: 0. l l l l l l Illi 0. l l l l l l i l l i 0 l l l l l l l i l l 0, llplfl 111! •III | | ; | | | | : 0 2 240 . 0 0 . 78 1 1 7 E - 0 2 0 .6 1290E • 0 2 0 5 9 6 3 6 E • 0 2 0. 7 4 9 3 2 E - 0 2 0. 8 8 4 4 4 E • 0 2 0. 9 8 7 0 8 E • 0 2 0 6 7 7 0 8 E 02 2 5 0 0 o.7 9 7 4 4 E - 0 2 .0 .S0959E • 0 2 III 55.2 28E 'III 11-7 319CE 111 :: 0. 88.762E • 0 2 11 96.8 7 5E 'III fi- S 7 1 5 5 E I2-2.80 o; 0 . 7 5 2 5 C E - . 0 2 •o . 5 0 9 5 9 E 111 111 l l l l l l fill p i . 73190E rill Ill 0 7 7 4 6 c . - i l l ill 943 .JCE fill l l- 6 4 8 2 0 E 0 2 30C 0 o . 7 4 2 9 2 E - 0 2 0 , 5 7 4 7 0 E - 0 2 0 5 2 5 7 7 E - 0 2 0. 6 3 7 2 5 E - 0 2 0. 8 4 2 3 8 E • 0 2 0. 8 6 2 0 7 E • 0 2 0. 6 1 8 9 3 E - 0 2 320 0 0 . 6 8 1 1 - 2 E - C 2 0 •ll- Ill- 4 8 7 0 5 E | 0 2 ! i l l 60S93E . •I2j i l l 7 7Q4 8E w$ i l - GS 152c I l l i Ml 5 7 1 2 3 S - C 2 .'3 40 .0 O . 6 3 2 4 2 E - 0 2 0 . 5 3 9 1 0 E Ill l-lli III 6Q999E 111: | f | 7 ;45 l : 9El •I-2II li! 8 .3899E fill i l l 5 l | | | : E - 0 2 360 .0 0 . 6 5 0 7 9 E - 0 2 0 . 5 1B4'4E - 0 2 0 . 4 4 5 2 2 E • 0 2 0. 5 4 8 6 5 E • 0 2 0. 7 2 8 1 2 E 02 0. 8 1 2 2 5 E - 0 2 0. 4 7 4 2 6 E - 0 2 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0058890/manifest

Comment

Related Items