UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Gas composition and temperature within a spouted bed gasifier Haji Sulaiman, Mohd Zaki 1984

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

Item Metadata

Download

Media
831-UBC_1984_A7 H34.pdf [ 7.28MB ]
Metadata
JSON: 831-1.0058902.json
JSON-LD: 831-1.0058902-ld.json
RDF/XML (Pretty): 831-1.0058902-rdf.xml
RDF/JSON: 831-1.0058902-rdf.json
Turtle: 831-1.0058902-turtle.txt
N-Triples: 831-1.0058902-rdf-ntriples.txt
Original Record: 831-1.0058902-source.json
Full Text
831-1.0058902-fulltext.txt
Citation
831-1.0058902.ris

Full Text

GAS COMPOSITION AND TEMPERATURE WITHIN A SPOUTED BED GASIPIER By MOHD ZAKI^HAJI SULAIMAN B.Sc. Imperial College London, 1980; ACGI A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Chemical Engineering) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLOMBIA July 1984 ®Mohd Zaki Haji Sulaiman, 1984 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I further agree that permission for extensive copying of t h i s t h e s i s for s c h o l a r l y purposes may be granted by the head of my department or by h i s or her representatives. I t i s understood that copying or pu b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Chemical Engineering The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date 2 4 t h July 1984  DE-6 (3/81) - i i -ABSTRACT Internal gas composition and temperature profiles have been measured in a spouted bed coal ga s i f i e r . These, data are essential to develop a mathematical model that can be used for scale-up of the system. As a f i r s t step'towards developing a kinetic model, the treatment of Yoshida and Kunii for the fluidized bed gasifier has been adapted to the spouted- bed using the one-dimensional and the streamtube gas flow models. Two Western Canadian bituminous and sub-bituminous coals having particle size -3.36 + 1.19 mm were gasified to produce low and medium c a l o r i f i c value gas. The effect of different operating variables including total bed height, average bed temperature and char recycling on the internal gas composition profiles were investigated. In a typical run eighteen gas samples were collected from different levels and radial positions in the bed. In addition the average axial temperature profiles along different sections were also recorded. Both radial and axial gas composition profiles exist in a spouted bed coal ga s i f i e r . The combustion reaction takes place in a very narrow zone close to the gas inlet in the annulus and in the lower section of the spout. The temperature variation in the spout was large, and the upper section of the spout was found to be the hottest region in the bed. A hydrogen/carbon monoxide ratio of about unity has been observed throughout the bed for the system producing low c a l o r i f i c value gas. In this case the contribution from pyrolysis and vo l a t i l e reforming - i i i -• reactions on the gas composition profiles appears to be significant, especially in the upper section of the bed. Gasifying char instead of coal altered the carbon monoxide and hydrogen distribution in the bed. Less hydrogen was produced. This pattern was however reversed when oxygen rather than air was used to gasify coal and was partly because of the high steam/oxygen ratio. Both the one-dimensional and the streamtube models gave relatively similar predictions of the average gas composition profiles and the overall gas composition based on kinetic expressions from the literature. However the predictions could not be matched with the experimental results. In modelling, the particle movement in the bed should be described. Pyrolysis should not be assumed instantaneous but i t s kinetics should be included. The streamtube model was superior to the one-dimensional model since i t was able to predict radial composition profiles throughout the annulus. These profiles however were much steeper than those found in the experiments. - i v -TABLE OF CONTENTS Page 1. INTRODUCTION 1 2. LITERATURE REVIEW AND BACKGROUND 3 2.1. The Spouted Bed as a Chemical Reactor 3 • 2.2 The Coal Gasification Process 7 2.3 Theory of Coal Gasification 12 2.3.1 Pyrolysis 12 2.3.2 Char-Gas Reaction 16 2.3.2.1 Volumetric Reaction 18 2.3.2.2 Surface Reaction. 22 2.3.2.3 Char-Oxygen Reaction 27 2.3.2.4 Char-Carbon Dioxide Reaction 33 2.3.2.5 Char-Steam Reaction 34 2.3.3 Gas-Gas Reaction 38 2.3.3.1 Water Gas Reaction 38 2.3.3.2 Gaseous Combustion 42 3. AIM AND SCOPE OF PRESENT WORK 43 4. EXPERIMENTAL APPARATUS 45 4.1 Gasification System 45 4.2 Gas Sampling 47 5. EXPERIMENTAL TECHNIQUE 50 5.1 Gasification Process 50 .5.2 Gas Sampling 50 5.3 Blockage of Sampling Probes 53 5.4 Gas Analysis 53 6. MODEL DEVELOPMENT AND SIMULATION 54 6.1 Development of Model 54 6.1.1 Dominant Reactions 54 6.1.2 Reaction Kinetics 55 - v -Page 6.1.3 Bed Hydrodynamics 55 6.1.4 One-dimensional Model 57 6.1.4.1 Reactions in the Spout 57 6.1.4.2 Reactions in the Annulus 64 6.1.5 Streamtube Model 66 6.1.5.1 Gas Flow Path 67 6.1.5.2 Reactions in the Spout . 68 6.1.5.3 Reactions in the Annulus 68 6.1.6 Outlet Component Gas Concentration 70 6.2 Calculation Procedure 70 6.3 Model Simulation and Discussion 71 6.3.1 Axial Gas Composition Profiles 71 6.3.2 Radial Gas Composition Profiles 77 6.3.3 Exit Gas Composition 81 6.3.3.1 Effect of Total Bed Height 84 6.3.3.2 Effect of the Air/Steam Ratio 87 6.3.3.3 Effect of the Average Bed Temperature. 89 6.3.3.4 Effect of Particle Size 89 7.. EXPERIMENTAL RESULTS AND DISCUSSION 93 7.1 General Consideration.. 93 7.1.1 Sampling Rate 93 7.1.2 Drying 95 7.1.3 Pyrolysis 98 7.2 Forestburg Coal-Air 99 7.2.1 Axial Gas Composition Profiles 99 7.2.2 Radial Gas Composition Profiles 110 7.2.3 Effect of Operating Condition 110 7.2.3.1 Char Recycle 110 7.2.3.2 Bed Height 114 7.2.3.3 Average Bed Temperature 116 7.3 Forestburg Coal-oxygen 116 7.3.1 Axial Gas Composition Profiles 117 - v i -Page 7.4 Balraer Coal-oxygen 122 7.4.1 Axial Gas Composition Profiles 122 7.5 Partially Devolatilised Forestburg Char-Air 125 7.5.1 Axial Gas Composition Profiles 126 7.6 Average Temperature Profiles 126 7.6.1 Forestburg Coal-Air 130 7.6.2 Forestburg Coal-Oxygen 132 7.6.3 Partially Devolatilised Forestburg Char-Air.... 134 8. COMPARISON BETWEEN MODEL AND EXPERIMENT.. 136 8.1 Estimation of Pyrolysis Product Distribution 136 8.2 Gaseous Combustion in the Fountain 137 8.3 Kinetic Parameters 138 8.4 Axial Gas Concentration Profiles 138 8.4.1 Profiles in the Annulus 138 8.4.2 Profiles in the Spout 141 8.5 Radial Gas Composition Profiles 142 8.6 Exit Gas Composition.... 142 9. CONCLUSION AND RECOMMENDATION 145 9.1 Conclusion 145 9.2 Recommendation 146 NOMENCLATURE 148 REFERENCES 153 APPENDIX - A 157 B 158 C 171 D 172 - v i i -LIST OF TABLES Table Page 1 Comparison of gasification processes 11 2 Classification of heating rates and temperature zone of pyrolysis 13 3 Approximate relative rates of gas-carbon reactions.... 17 4 Variations of K a values for gas-carbon reactions 19 5 Correlated rate equations for carbon-oxygen reaction.. 31 6 Correlated rate equations for carbon-carbon dioxide reaction 35 7 Correlated rate equations for carbon-steam reaction... 39 8 Kinetic parameters used in gasification modelling 56 9 Hydrodynamic parameters used in gasification modelling 58 10 Properties of coals tested 94 11 Effect of sampling rate on gas composition 96 12 Gas yield due to pyrolysis at 600°C for the coals tested 103 13 Properties of Forestburg coal and the pa r t i a l l y d e v o l i t i l i s e d Forestburg char used in the experiment.. 127 14 Operating conditions 1 158 15 Operating conditions II 159 16 Operating conditions III 160 17 Experimental results - Run 88 161 18 Experimental results - Run 89 162 19 Experimental results - Run 90 163 20 Experimental results - Run 91 164 21 Experimental results - Run 93 165 - v i i i -Page 22 Experimental results - Run 94 166 23 Experimental results - Run 96 167 24 Experimental results - Run 103 168 25 Experimental results - Run 106 169 26 Experimental results - Exit gas composition and average bed temperature 170 27 Bed geometry and properties of solids used in model simulation 172 - ix -LIST OF FIGURES Figure Page 1 Schematic diagram of a spouted bed 4 2 Concentration profiles of carbon and reacting gas as a function of carbon conversion for volumetric reaction 20 3 Concentration profiles of carbon and reacting gas as a function of carbon conversion for surface reaction 24 4 Schematic flow diagram for the gasification system.... 46 5 Gas sampling probe 48 6 Schematic diagram for the gas sampling line 49 7 Exact locations of sampling points in the bed 52 8 Axial gas composition profiles in the annulus predicted using the one-dimensional model blown with air and steam 72 9 Axial gas composition profiles in the annulus predicted using the one-dimensional model blown with oxygen and steam 74 10 Axial gas composition profiles in the spout predicted using the one-dimensional model 76 11 Carbon monoxide composition profile along streamlines in the annulus predicted using the streamtube model... 78 12 Comparison on the axial carbon monoxide and hydrogen profiles predicted along the centre of the annulus using both models 79 13 Radial gas composition profiles at the top of the bed predicted using the streamtube model 81 14 Radial carbon monoxide composition profiles in the annulus at various bed levels predicted using the streamtube model 82 15 Comparison between the radial hydrogen and carbon monoxide composition profiles predicted using both models... 83 Page 16 Effect of the total bed height on the exit gas composition predicted using the one-dimensional model 85 17 Comparison on the overall carbon monoxide and hydrogen composition predicted using both models 86 18 Effect of air/steam ratio on the exit gas composition predicted using the one-dimensional model 88 19 Effect of the average bed temperature on the exit gas composition predicted using the one-dimensional model 90 20 Effect of the particle size on the exit gas composition predicted using the one-dimensional model 91 21 Axial gas composition profiles along position B for the Forestburg coal-air system 100 22 Axial gas composition profiles along position A for the Forestburg coal-air system 105 23 Axial gas composition profiles along position C for the Forestburg coal-air system 107 24 Radial carbon dioxide and carbon monoxide composition profiles in the annulus for the Forestburg coal-air system I l l 25 Effect of char recycling on the axial gas composition profiles for the Forestburg coal-air system 112 26 Effect of the total bed height on the axial gas composition profiles for the Forestburg coal-air system 115 27 Axial gas composition profiles along position B for the Forestburg coal-oxygen system 118 28 Axial gas composition profiles along position A for the Forestburg coal-oxygen system 120 29 Axial gas composition profiles along position C for the Forestburg coal-oxygen system 121 - xi -Page 30 Axial gas composition profiles along positions A and B for the Balmer-oxygen system 123 31 Axial gas composition profiles along position A for the partially d e v o l i t i l i s e d Forestburg char-air system .... 128 32 Axial gas composition profiles along positions B and C for the partially d e v o l i t i l i s e d Forestburg char-air system 129 33 Axial temperature profiles in the bed for the Forestburg coal-air system 131 34 Axial temperature profiles in the bed for the Forestburg coal-oxygen system 133 35 Axial temperature profiles in the bed for the par t i a l l y devolitilised Forestburg char-air system.... 135 36 Comparison between the predicted and the experimental axial composition profiles for the Forestburg coal-air system 139 37 Comparison between the predicted and the experimental radial carbon dioxide and carbon monoxide composition profiles for the Forestburg coal-air system 143 38 Comparison between the predicted and the experimental overall gas composition for the Forestburg coal-air system 144 - x i i -ACKNOWLEDGEMENTS I would l i k e to thank, the following i n d i v i d u a l s and organizations f o r t h e i r contributions throughout t h i s study: - Dr. A.P. Watkinson and Dr. -C.J. Lim for t h e i r excellent supervision and guidance. - Mr. Gordon Cheng and Mr. Sam Low for t h e i r help i n preparing and running the experiments. - The Energy Research Laboratories, Department of Energy, Mines and Resources, Canada who funded the research contract under which t h i s work was done. - The Government of Malaysia and the U n i v e r s i t y of Malaya, Kuala Lumpur for the f i n a n c i a l a s s i s t a n t s h i p . I am also indebted to my wife, J u l i a n a for her continual support and encouragement throughout t h i s work. - 1 -1 . INTRODUCTION The d i r e c t use of s o l i d f u e l i s l i m i t e d and i t s use i s l e s s a t t r a c t i v e as compared to gaseous and l i q u i d f u e l s i n many a p p l i c a t i o n s . The reserves of s o l i d f u e l , e s p e c i a l l y c o a l , are abundant and extensive research has been di r e c t e d towards developing s u i t a b l e processes that can e f f i c i e n t l y u t i l i z e t h i s m a t e r i a l . An example of such a process which i s s t i l l being developed i s g a s i f i c a t i o n . Coal g a s i f i c a t i o n involves the production of gaseous fuels such as methane, hydrogen and carbon monoxide by allowing coal to react with steam and/or carbon dioxide at elevated temperatures. As in any other process i n v o l v i n g reactions of s o l i d s and gas, high conversion of coal i n g a s i f i c a t i o n depends mainly on the g a s - s o l i d contacting technique used. A number of r e l i a b l e g a s i f i e r s based on various g a s - s o l i d contacting techniques have been developed and some of these have been commercially proven. These include moving, entrained and f l u i d i z e d bed g a s i f i e r s . However there e x i s t some disadvantages for d i f f e r e n t types of g a s i f i e r s which r e s u l t i n s u p e r i o r i t y of one with respect to the other depending on a number of factors such as the type of coal to be processed, p a r t i c l e s i z e to be employed et c . Both moving and f l u i d i z e d bed g a s i f i e r s s u f f e r from a common disadvantage which i s the i n a b i l i t y to handle caking c o a l . Entrained bed g a s i f i e r s can accommodate caking . c o a l , however they can only be operated using fin e p a r t i c l e s , which brings with i t an economic penalty because of the cost of p u l v e r i z i n g the c o a l . Due to these problems, the spouted bed g a s i f i e r i s being - 2 -i nves t iga ted with the hope of prov id ing a s tab le and r e l i a b l e technique for process ing caking coals which are too coarse for entrained bed g a s i f i e r s . The o v e r a l l performance and the s e n s i t i v i t y to d i f f e r e n t operat ing va r i ab l e s have been studied for spouted bed g a s i f i e r s (13-16). Promising r e su l t s have been obtained when compared to other u n i t s . However no informat ion i s a va i l ab l e concerning the gas composit ion wi th in the g a s i f i e r . To provide a scale-up based on p r i n c i p l e s of reac tor des ign , some knowledge of the composit ion with in d i f f e r e n t zones of the spouted bed i s r equ i r ed . Since th i s in format ion must be generated exper imenta l l y , an in-bed study of the gas composit ion p r o f i l e i s r equ i r ed . Previous work on hydrodynamics c l e a r l y ind i ca ted that the performance of the spouted bed as a chemical reactor i s s t rong ly in f luenced by the bed diameter. As r e l a t i v e l y la rge bed diameters would be employed for commercial spouted bed g a s i f i e r s i . e . diameters much la rger than those being inves t iga ted cu r r en t l y , development of a mathematical model i s e s sen t i a l fo r scale-up purposes. A r e l i a b l e model can only be developed using informat ion obtained from an experimental study of the i n t e rna l composit ion p r o f i l e s i n a spouted bed coa l g a s i f i e r . - 3 -2. LITERATURE REVIEW AND BACKGROUND 2.1 The Spouted Bed as a Chemical Reactor The spouted bed has been developed as an alternative fl u i d - s o l i d contacting technique to the fluidized bed and has recently received much attention for industrial applications due to i t s capability of handling relatively coarse particles. A complete review on this subject is given by Mathur and Epstein (1) and w i l l not be presented here. In brief the spouted bed consists of a centrally located dilute phase concurrent-upward transport region (spout) surrounded by a dense-phase moving packed bed with countercurrent fl u i d percolation (annulus), as shown schematically in Figure 1. A number of research papers on the application of the spouted bed as a chemical reactor have been published including the development of appropriate mathematical models for the system. The models generally treat gas phase reactions which are catalysed by solids. In some cases experimental results were used to verify the models and to identify the v a l i d i t y of the various assumptions used. Mathur and Lim (2) introduced a one-dimensional model, where plug flow of gas was assumed both in the spout and the annular regions which were treated as two separate regions. The effect of gas back-mixing was neglected. The effect of the different operating variables on the overall conversion in a f i r s t order reaction was examined theoretically based on this approach. It was found that the overall extent of conversion of gas improved as the column diameter was increased and this - 4 -V ^ 0 -•>•'?. f> "•"•'•if.- -ire : 0 U N T A I N BED S U R F A C E S P O U T A N N U L U S • S P O U T - A N N U L U S I N T E R F A C E CONICAL B A S E FLUID I N L E T Figure 1 - Schematic diagram of a spouted bed. - 5 -improvement was mainly due to an increase in conversion in the annulus. In a l l cases the extent of conversion in the spout was not significant. The role of spout voidage in determining the overall conversion was relatively unimportant even for the fastest reaction. Measurements made by Lim (3) showed that the spout voidage ranged typically from unity to 0.65 at z = H. Poorer results were obtained by increasing the operating gas velocity above U m s. An improved overall conversion was obtained by increasing the bed height up to a certain value. Above this optimum height no significant changes were observed for slow reaction while for fast reaction, the conversion actually became poorer. Based on the observation made in a half-sectional column with nitrogen dioxide used as tracer, Lim and Mathur (4) proposed a more r e a l i s t i c streamtube model as a modification to the one-dimensional two region model. In this model plug flow of gas with dispersion along curved streamlines in the annulus was proposed, as supported by residence time distribution measurements. Assuming the va l i d i t y of this assumption, one would expect to observe a radial concentration profile in the annulus of a spouted bed chemical reactor. Vertical plug flow of gas in the spout, as assumed in the one-dimensional model, was also assumed in this case. In an attempt to verify the two models proposed by Mathur and Lim (2,4), Piccinini et al (5) studied the decomposition of ozone on an iron catalyst in a 0.15 m diameter column. Unlike Mathur and Lim (2) where a l l the hydrodynamic features were predicted from equations, these investigators used experimentally measured values of a l l the requisite - 6 -quantities except U A and e s as input to the models. In addition, the original one-dimensional model was modified by allowing diffusional mass transfer between the spout and the annulus in addition to transfer by bulk flow. The overall conversion predicted by both models was in good agreement with the experimental values, with an average deviation of about 5%. The inclusion of a non-zero mass transfer coefficient between the spout and the annulus in the one-dimensional model gave poorer results. Littman et al (6) examined the one-dimensional model proposed by Mathur and Lim (2) and theoretically simulated a f i r s t order reaction using the bed hydrodynamic equations different from those of the original study. The findings reported by Mathur and Lim were confirmed except for the effect of bed height on overall conversion where no optimum value was observed. Further work on assessing the performance of spouted bed as a chemical reactor was performed by Rovero et al (7), where the decomposition of ozone on an iron oxide catalyst was examined in 0.15 m and 0.22 m diameter beds. To aid in discrimination between the two proposed models both the radial and longitudinal profiles of ozone concentration were determined. As in a l l previous cases the overall conversion predicted by both models was in good agreement with experimental values. However the one-dimensional model failed to account for the radial concentration profiles observed while the streamtube model predicted a much steeper concentration gradient over most of the annulus. Contrary to expectations and findings reported by - 7 -a l l previous investigators, the effect of increasing bed diameter was to decrease the conversion. The results for the other variables investigated were in accord with the predictions made by Mathur and Lim (2) and by Littman et al (6). Even though the two models suggested by Mathur and Lim (2,4) were able to give good overall predictions for most cases investigated, they failed to describe closely the annulus concentration profiles observed. As suggested by Rovero et al (7), the failure may be due to the assumption of no radial dispersion which could be significant especially in large columns. In addition, depending on the size of column, the conical bottom region may form a large fraction of the bed where there is evidence of circulation (8). Rovero et al (9) have shown that the superficial annulus gas velocity in the conical region is in excess of that predicted by the Mamuro-Hattori flow distribution (1) which has been used in most work. It seems therefore that any r e a l i s t i c model for in-bed composition must account for radial concentration gradients in the gas phase. 2 . 2 T h e C o a l G a s i f i c a t i o n P r o c e s s Previous work on coal gasification producing either low, medium or high BTU fuel gas employed a variety of different reactors including moving, fluidized and entrained beds. Each of these reactor types has i t s own characteristics which wi l l influence the operating conditions to be selected and the type of coal to be used. Thus there exist - 8 -advantages/disadvantages of one type of reactor over the other depending on the required operating conditions ( 1 0 - 1 2 ) . The counter current direct contact heat exchange between gas and solids that takes place within a moving bed gasifier results in a relatively high overall conversion. However fine particles (< 3 mm) and caking coals cannot be handled by this type of reactor. The entrained bed reactor on the other hand can accept any type of coal and gives high throughput especially when operated at elevated pressure. Unfortunately the coal particle size must be smaller than about 7 5 um and carbon conversion may be low. The fluidized bed gasifier has a relatively uniform bed temperature due to i t s excellent mixing characteristics. However, a disadvantage of this reactor i s that d i f f i c u l t i e s are usually encountered when caking coals are used. This is due to the agglomeration that takes place, forming larger particles which cause the bed to collapse. The spouted bed coal gasifier was developed with the purpose of providing a smooth and a reliable system for gasifying caking coal. The problems of agglomeration could be overcome with the high gas velocity in the spout which tends to break up any agglomerates that might form within the bed, and the absence of a grid where agglomerates can build up. Systematic investigation of spouted bed coal gasification was initiated by group of investigators at the University of British Columbia. Foong et al (13) studied the effect of various operating parameters on the quality of gas produced in a 0 . 1 5 m diameter gasifier blasted with a i r , using inert bed of s i l i c a ( 1 . 2 - 3.4 ram). The various - 9 -parameters investigated were bed temperature (1020 - 1220 K), coal feed rate (2.5 - 12 kg/h), steam feed rate (16 - 24 kg/h), particle size, bed height (0.25 - 0.6 m) and different types of coal (sub-bituminous Forestburg and highly caking Sukunka). Under the range investigated improved gas quality was obtained as the coal feed rate, bed temperature and the particle size were increased. Increasing the bed height up to a certain value improved the quality of gas produced but no further improvement was observed on increasing the height above this value. The coal feed rate was found to be an important factor in determining the smoothness and s t a b i l i t y of the overall operation, especially when gasifying caking coal. In assessing the effect of coal feed location, these investigators concluded that bottom feeding of coal minimizes the presence of tar in the product stream as compared to feeding the coal at the top of the bed. The fluid mechanical characteristics of spouted beds strongly depend on the column diameter. Hence a higher coal conversion and improved gas quality can be obtained when using a larger diameter gasi f i e r . Watkinson et al (14-15) investigated the performance of a 0.3 m diameter gasifier. A number of operating parameters were examined in addition to those investigated previously. The effect of char recycle and particle size distribution were studied at coal feed rates of about 50 kg/h. The conclusions of the previous investigators were confirmed in this work. Char recycle showed no significant improvement in the gas quality produced, but the efficiency over once-through operation improved. The gasifier efficiency was reported to be significantly - 10 -reduced by having a higher percentage of finer particles in the coal feed. This was mainly due to a large extent of particle elutriation from the bed. The advantage of the 0.3 m diameter gasifier over the smaller 0.15 m diameter was clearly demonstrated when a highly caking Sukunka coal was used. In this case a much higher percentage of carbon in the bed can be tolerated (= 37%) as compared to 10% for the smaller system. Other advantages such as higher gas c a l o r i f i c value, thermal efficiency and steam decomposition have been achieved from the bigger system. These gains were not solely related to the diameter increase i t s e l f . Watkinson et al (16) compared the performance of the spouted bed to the already established fluidized bed. In their experimental work, the effect of air/coal ratio and coal feed rate, keeping the other parameters constant, on the gas heating value, total gas yield etc. were investigated under both modes of operations using different types of coal. They concluded that no appreciable differences between spouted and fludized bed existed as far as yield and heating value were concerned. These workers however suggested that the advantage of one over the other could exist i f the systems were to be tested using other variables such as using oxygen rather than a i r , recycling of char and using the optimal particle size for each system. In addition, advantages or disadvantages may become obvious when comparisons were made using larger size reactors. The performance of the spouted bed gasifier investigated was considered promising when compared to the other systems as shown in Table 1. Table 1 Comparison of gasification processes* Reactor Type Process Fluidized Bamag-Winkler (17) Bed Westinghouse (18) Moving Bed Wellman-Galusha (17) Spouted Bed (13 & 15) Diameter (m) 5.5 0.5 3.05 0.15 0.305 Bed Depth (m) 3 0.40 1.0 Temperature (c) 800-1000 930-1020 900 800-930 805-860 Coal Throughput 0.136 0.32-0.60 0.123 0.188 0.23 (kg/s«m2) Gas Yield 4.45 4.00 3.05 3.22 (m /kg coal) Gas Calorific Value 3.91 3.7-4.5 5.97 3.6 4.36 (MJ/m3) Coal Type Sub. bit A. Sub. bit to High Volatile Sub bit Eastern b i t . bit A. (Forestburg) *Air and steam blown. A l l at atmospheric pressure except Westinghouse at 1545 KPa. - 12 -2.3 Theory of Coal G a s i f i c a t i o n 2.3.1 P y r o l y s i s Upon heating, coal undergoes physical and chemical changes to an extent which depends on a number of factors such as heating rate, temperature, coal type etc. Pyrolysis or devolatization is normally the f i r s t step through which such major changes take place. During this process there is a substantial weight loss' of coal due to the evolution of vo l a t i l e matter which cannot be described by a simple single step reaction but occurs in stages involving a series of complex reactions, producing both tar and gases (CO2, CO, CH^, H2 and other liquid hydrocarbons). Depending on the environmental conditions, these products can undergo further secondary reactions involving cracking, polymerisation and deposition onto the char. The proportions and distribution of gaseous components and liquid yield is generally determined by both heating rates and the temperature zone of pyrolysis which can be classified into different categories as given in Table 2. Typically, under flash pyrolysis condition with minimum vapour residence time a large proportion of tar and unsaturated hydrocarbon w i l l be produced (19). A number of investigators (20) have reported that volatile yields significantly greater than those indicated by proximate analysis can be.obtained from coal by flash pyrolysis. In.general, at any particular heating rate, both the total yield and gas/tar ratio increases with temperature. At a relatively low final temperature (= 500°C) increasing the heating rate w i l l increase the total v o l a t i l e yield while the gas/tar ratio - 13 -Table 2 Classification of heating rates and temperature zone of pyrolysis (19). A. Classification of Heating Rates Heating Rate Heat-Up Time to 1000°C °C/s f o r 1 0 0 y in Size 1. Slow heating « 1 20 min 2. Intermediate heating 5 ~100 10 s 4 min 3. Rapid heating 500 ~ 100,000 10 ms~2 s 4. Flash heating > 10 5 < 1 ms B. Classification of Temperature Zones for Carbonization of Coal 1. Low temperature carbonization 500°C 2. Intermediate temperature carbonization ^750°C 3. High temperature carbonizations 1000°C 4. Very high temperature carbonization > 1200°C - 14 -decreases. On the other hand, rapid heating at a much higher temperature 1000°C) wi l l increase both the total yield and gas/tar ratio. Several investigators (21-22) working on flash pyrolysis reported that the tar yield shows a maximum value at about 600°C. Pyrolysis is a complex process and a number of simplified kinetic models have been proposed to describe the reaction. According to Anthony et al (23) primary volatiles formed are of two categories: reactive and non-reactive. The non-reactive volatiles escape completely from the particle, while only a portion of the reactive volatiles escape. The remainder deposits inside the coal particle by polymerisation. Chang et al (20) working on flash pyrolysis reported that only a portion of the coal volatiles directly pyrolysed into gaseous product. The rest w i l l escape as tar, which could subsequently crack into smaller gas molecules until the system is quenched. Gas cracking was considered not to be important. In interpreting his data, Wen and Chen (24) assumed that coal w i l l i n i t i a l l y devolatilise to give tar. A portion of this tar w i l l escape, another portion w i l l crack into gaseous components while the remainder deposits back onto the char particles. According to this model no gas will be evolved directly from coal by pyrolysis. A simpler model of pyrolysis is normally more attractive for gasifier design purposes, as several reaction coefficients have to be determined for any one particular coal i f the above described models are to be used. These coefficients are normally d i f f i c u l t to generate. A review on coal pyrolysis by Anthony and Howard (25) indicated that most - 15 -investigators approximated the overall pyrolysis process as a f i r s t order decomposition with respect to the amount of volatiles V that s t i l l remain in the coal, i.e. where K' = pyrolysis rate constant, S~ Voo = V as t •»• °° The variation of K' with temperature is assumed to follow Arrhenius relationship expressed as follows: Values of K q and E' are to be experimentally determined for each particular coal. The above equation however does not predict the product distribution of the volatile released. Solomon (26) used the form of equation 2.1 above to predict the amount of each component released. In this case V and V^ , represent the amount of the particular component at time t and t -»• » respectively. The rate parameters K Q and E' w i l l have different values for different components and coal. According to Howard (27) the sequence of volatiles being released (after drying) i s : chemical water, carbon dioxide, carbon monoxide, higher hydrocarbons (tar, etc.), ethane, methane and f i n a l l y hydrogen. ;dv dt = K' (V_ - V) (2.1) = IC exp (-E'/RT) (2.2) - 16 -2.3.2 Char—Gas Reaction The heterogeneous reactions between char and gases are the main reactions that take place in a gasifier. These reactions are normally slow when compared to the homogeneous gas-gas reactions, and their rates of reactions are of considerable interest and importance for gasification studies. The following set of linearly independent gas-solid reactions are normally considered in describing gasification processes: Combustion tyC + 0 2 •* 2(ij> - 1) CO + (2 - ip) C0 2 (2.3) Gasification C + H20 ->- CO + H 2 (2.4) C + C0 2 •»• 2C0 (2.5) C + 2H2 + CH^ (2.6) The char-oxygen reaction is the fastest among a l l the reactions considered. This reaction, which is exothermic, acts as the main heat source for the intermediate endothermic char-steam and char-carbon dioxide reactions. The char-hydrogen reaction is known to be the slowest of the above four gas-solid reactions that takes place in a gas i f i e r . For comparison, the relative rates of a l l the four reactions considered are given in Table 3. The equilibirum consideration of gasification reactions (2.4) -(2.6) has been the commonly used procedure to obtain an approximate Table 3 Approximate relative rates of the gas-carbon reactions at 800° and 0.1 atmospheric pressure (41) Reaction Relative Rates c - °2 1 X 10 5 c - H20 3 c - CO 2 1 c - H 2 3 x i o - 3 - 18 -approachable conversion in a real system. The extent to which a given reaction w i l l proceed towards the right under given conditions i s reflected by the value of i t s thermodynamic equilibrium constant K. A large K value implies large equilibrium conversion; a small K implies . l i t t l e conversion to product. Variation of K values for graphite-steam, graphite-carbon dioxide and graphite-hydrogen reactions with temperature are given in Table 4. These K values are based on acti v i t i e s (K a) and in this case the activity of graphite is taken to be unity. It is clearly shown from these tables that high temperature favours the gasification of char with steam and carbon dioxide. In contrast the char-hydrogen reaction i s only favoured at low temperature and high pressure. The extent of this reaction is so small under a typical gasifying condition of 1 atm and 800°C and is normally neglected. This reaction has been ignored by other investigators (28-29) in their studies of coal gasification at atmospheric pressure. The char-gas reactions can be classified into two categories: volumetric and surface reactions as discussed below: 2.3.2.1 Volumetric Reaction Volumetric reaction is characterized by the chemical reaction as the rate controlling step in porous solids. Here the reacting gas diffuses into the interior of the particles causing the reaction zone to spread throughout the solid. As the reaction proceeds, an ash layer may build up at the outside surface of the particle whilst the reaction zone continues to shrink. Figure 2 shows typical carbon and gas concentration profiles as a function of carbon conversion. - 19 -Table 4 The variations of K a values for the graphite-steam, graphite-carbon dioxide and graphite-hydrogen systems with temperature (19) Temperature K p u n K p r n K o K C-H20 C-C0 2 C-H 2 298.16 1.005 x i o - 1 6 1.014 X IO" 2 1 7.902 x 10 8 600 5.058 X 10"5 1.870 X 10"6 1.000 x 10 2 800 4.406 X IO" 2 1.090 X IO" 2 1.4107 1,000 2.617 X 10° 1.900 X 10° 0.0983 1,500 6.081 X 10 2 1.622 X 10 3 0.00256 - 20 -Figure 2 - Concentration p r o f i l e s of carbon and reacting gas as a function of carbon conversion for volumetric r e a c t i o n . - 21 -Many investigators (30-32) concluded that the internal surface area due to the inside pores of the particle plays an important role in volumetric reactions. Most of this internal area is attributed to the micropores which are in the size range of 10-50A. This internal surface area varies as the reaction proceeds. It f i r s t increases to a maximum and then reduces to a value of zero at complete conversion as reported by Dutta and Wen (30). According to these workers each coal/char sample exhibits i t s own pore structure and the change of such structure with conversion and temperature gives rise to a characteristic rate-conversion curve for that individual coal or char. The following rate expression has been formulated by Dutta and Wen (30): If" av k v C g ( 1 " x ) < 2' 7 ) where x = fraction of carbon conversion of char due to char-gas reaction, -t = time, s ctv =>. relative pore surface area function, -C = gas concentration, mol/cm O k^ = volumetric rate constant for char-gas reaction (cm 3/mol) n*s~ 1 n = order of reaction cty describes the change in pore structure and is a function of - 22 -conversion and temperature for each type of coal in question. Equation 2.7 is applicable when resistance due to gas diffusion inside the particle is negligible. At high temperature, this resistance may become appreciable and the concept of effectiveness factor n is introduced as given by the following expression for n = 1. M vtanh M (2.8) where M - <M(1 - x) a ] o v 1/2 <j> = r /k C 7D To o v so' eo r = outside radius of particles, cm C = carbon concentration of char at zero conversion, mol/cm so ' D = effective d i f f u s i v i t y in solid at zero conversion, cm/s Thus the volumetric rate equation of a f i r s t order reaction i s expressed, including the interparticle diffusion as dx dt = net k C (1 - x) v v g (2.9) The reaction rate constant ky is assumed to follow Arrhenius relationship (30) given by k v = k Q exp(-E/RT) (2.10) 2.3.2.2 Surface Reaction This type of reaction normally takes place when the reacting gas can hardly penetrate into the interior of the solid particle, i.e. for - 23 -impervious solids and/or when the overall reaction is gas diffusion controlled. For this type of reaction, the reaction zone is confined to the surface of the unreacted core which shrinks as the reaction proceeds, while the product solids or ash layer builds up on the outside. Typical carbon and gas concentration profiles for such a reaction i s given in Figure 3. The resistance to the reaction is made up of the combination of resistances due to mass transfer of gas in the boundary layer surrounding the particle, gas diffusion in the porous ash layer and the surface reaction i t s e l f . Levenspiel (33) has formulated this type of reaction known as the shrinking core model expressed as follows: dx 3 P 2  R a t e = d l = ~ , 1 \ ( Y - l ) , 1 , ( 2 * u ) Mw ro p c ( k — k ~ Y — 2 > d i f f ash k Y s where = molecular weight of carbon, g/mol. r Q = outside particle radius including ash layer, cm P c = molar density of carbon in coal particle, mol/cm P = gas partial pressure, atm. k ^ ^ = diffusional reaction rate constant, g/cm2*atm*s k , = ash film diffusion rate constant, g/cm 2 ,atm»s ash ' & k = surface reaction rate constant, g/cm 2 ,atm«s s ' 0 RC 'C ' 0 Gas concentration profile Carbon concentration profile Figure 3 - Concentration profiles of carbon and reacting gas as a function of carbon conversion for surface reaction. - 25 -Y r c = core radius at time t, cm Depending on the system employed, the product or ash layer may segregate from the p a r t i c l e s as the reaction proceeds. This i s normally true for the spouted bed g a s i f i e r due to high a t t r i t i o n . In the absence of resist a n c e due to the ash l a y e r , the instantaneous r e a c t i o n rate from Equation 2.11 reduces to The mass transf e r through the boundary layer i s influenced by a number of f a c t o r s , such as the flow condition of the bulk gas ( i . e . laminar or turbulent) as well as the r e l a t i v e v e l o c i t y between the gas and the p a r t i c l e s . Turbulent flow w i l l bring the oxidant to the v i c i n i t y of the p a r t i c l e s surface while high r e l a t i v e motion between the gas and the s o l i d w i l l reduce e f f e c t i v e thickness of the boundary l a y e r and enhance transf e r rate through i t . The values of d i f f u s i o n a l r e a c t i o n rate constant, k ^ i f f . can be cal c u l a t e d to a f a i r degree of accuracy using the following p r e d i c t i v e equation (34-35): 3P Rate = M r p (-r-w o c k 1 (2.12) "diff 24(1 + 0.6 Re 1/2 Sc 1/3 ) D k AB (2.13) d i f f d R'T P - 26 -where Re = Reynolds number Sc = Schmidt number 2 DAT1 = diffusion coefficient of species A in B, cm/s A D d = particle diameter, cm P R' = universal gas cnstant (82.06 atm«cm /mol .°K) T = temperature, °K The ash film diffusion rate constant depends on both the gas d i f f u s i v i t y and the voidage of the ash layer. According to Wen and Chaung (36) the value of k a s h can be roughly estimated by the following correlation: where e is the voidage of the ash layer. The surface reaction rate coefficient i s assumed to follow the Arrhenius law (37) expressed as values of k s o and E for each coal are to be determined experimentally. In general, the shrinking core model does not f u l l y represent the actual gas-solid reactions especially when dealing with coal/char particles which are normally porous. Investigation by Smith and Taylor kash ~ k d i f f (2.14) k g = k s Q exp(-E/RT) (2.15) - 27 -(38) and F i e l d (39) on char combustion and Dutta and Wen (30) on char-carbon d iox ide react ions ind ica ted that the react ions ins ide the p a r t i c l e s do take place s imultaneously with the externa l r e a c t i o n . However the shr ink ing core model represented by Equation 2.11 i s simple and p r a c t i c a l l y use fu l for model l ing purposes. A number of i n ves t i ga to r s (36, 40) employed t h i s model in the i r work on coa l g a s i f i c a t i o n . 2.3.2.3 Char-Oxygen Reaction Char-oxygen reac t ion i s the f as tes t heterogenous gas-so l id r eac t ion that takes place in the g a s i f i e r (42) . The reac t ion i s exothermic and provides most of the heat necessary fo r the other endothermic g a s i f i c a t i o n reac t ion and can be represented as i|>C + 0 2 * 2(+ - 1) CO + (2 - i|>) C 0 2 (2.16) where i|> i s a system constant which depends on the r eac t ion cond i t ions and determines the primary product d i s t r i b u t i o n of CO and C0 2 i n the combustion products , takes a value of 2 when CO i s the d i r e c t product of combustion and a value of 1 when C 0 2 i s the only product . Several i n ves t i ga t i ons to p red ic t the combustion product d i s t r i b -u t i on and hence the va lue of I|I as a func t ion of temperature, p a r t i c l e s-i z e , carbon type and oxygen p a r t i a l pressure (43-45) gave no q u a n t i t -a t i v e conc lus ions . However there i s a general agreement that both CO and C 0 2 are the primary products of char-oxygen r e a c t i o n , with smal ler p a r t i c l e and/or higher temperatures favour ing the formation of CO. - 28 -Larger p a r t i c l e s and/or lower temperatures favour the CO2 formation. Arthur (46) investigated the combustion product d i s t r i b u t i o n s for two carbons of widely d i f f e r e n t r e a c t i v i t i e s (graphite and coal char) at atmospheric pressure and i n the temperature range of 400°C - 900°C and found that Z, the r a t i o of C0:C0 2, i s given by Z = 2500 exp (-6249/T) (2.17) In his experiment the oxidation of CO was suppressed using the i n h i b i t o r phosphoryl c h l o r i d e . Using the value of Z given by Equation 2.17 Wen and Dutta (19) suggested a c o r r e l a t i o n to predict the value of i|> expressed as follows: For d < 0.005 cm P — » - 2 \ \ \ (2.18) For 0.005 cm < d < 0.1 cm P — , Z(d - 0.005) * = z ~ T T [ ( 2 Z + 2 ) " Po.o95 ] > d P i n c m For dp > 0.1 cm t|» = 1.0. There i s a large volume of work published, dealing with the k i n e t i c studies of char-oxygen r e a c t i o n . Mulcahy and Smith (47) reported that the o v e r a l l combustion rate for p a r t i c l e s l a r g e r than 100 u i s d i f f u s i o n c o n t r o l l e d at temperatures above 1200°K. According to Smith and Tayler (38), coal/char p a r t i c l e s of about 90 ym in s i z e - 29 -w i l l burn under chemical reaction control rate up to about 750°K, whereas for smaller particles (- 20 \i) the chemical reaction control regime extends up to a higher temperature of about 1600°K. Field (39) in his experiment using particle sizes 28 y - 105 y concluded that char burns both internally as well as externally. This conclusion was based on the observation that both particle density and mean diameter reduce as a function of char burn-off. This observation is generally expected when working with small particles as the chemical reaction controls the overall rate of reaction, thus allowing the reaction to proceed inside the particles. Large particles are normally employed in spouted bed (= 2 mm) where diffusion is significant. Under this condition, the assumption of surface reaction for char combustion is j u s t i f i e d (42). The rate of surface reaction as given by Equation 2.12 can be used for this reaction. In this case Pg represents the oxygen partial pressure. The order of reaction n, for char-oxygen reaction has been reported to have values ranging between zero and unity (42, 48-51). However, there i s a general agreement based on a large volume of experimental data that the char combustion is f i r s t order with respect to oxygen partial pressure. Hence Equation 2.12 when applied to char combustion reduces to 3 P o Rate = w TT7T 2 . . / i \ (2.19 M r p (1/k,. c c + 1/k ) w o c d i f f s The value of k<jiff to be used in Equation 2.19 can be obtained with a fa i r degree of accuracy from Equation 2.13. The effect of - 30 -relative velocity and hence the Reynolds number on k ^ i f f i s not very significant for particles in the pulverized fuel size range (= 100 y) as concluded by Mulcahy and Smith (47). For such a condition Field et al (48) suggested a simpler expression to calculate the k ^ f f for char combustion given as follows: k = 24 i|P' , . cliff R'T d p U ' 2 U ; where R' = Universal gas constant (82.06 atm-cm /mol*°K) D' = diffusion coefficient of oxygen 4.26 ( T 1 8 0 0 ) 1 , 7 5 / P , cm2/s P = total pressure, atm ^ = mechanism factor given by Equation 2.18 dp = particle diameter, cm Equation 2.20 can be applied without much error to the fluidized bed char combustion where the Re is normally small. The rate of surface reaction for char combustion as a function of temperature has been determined involving char of different reactivity (37,38,48,52-53). Most of this work was conducted using particles in the pulverized coal size range. A number of these correlations suggested by different investigators are given in Table 5. According to Field (39), there is no variation in the rate of surface reaction with particle size. Field et al (48) reported approximate value of k s o and E/R for char-oxygen reaction which can be used together with Equation 2.15 to predict the rate of surface reaction for char combustion as a Table 5 Rate equations for carbon-oxygen reaction Rate = kP" °2 Ref k n Size (y) Temp. Range °K Particle Type Remarks 37 1.34 exp (-16405/T ) s (g/cm «s) 0 22, 49, 89 630 - 1812 Brown coal Experiments were performed i n fixed and entrained bed reactors under oxygen partial pressure up to 0.2 atm. Rate given is based on total surface area and is the rate of surface reaction. 38 -0.49 + (3.85 x 10_1* T ) (g/cm2«s) 0 20 - 100 1400 - 2000 Low rank non swelling coal The effect of mass transfer was eliminated by theoretical arguments. Rate equation is based on the external surface area of particle, and i s the rate of surface reaction. 48 8710 exp (-17976/Tg) (g/cm »s) 1 varied 750 - 1650 varied Effect of mass transfer was eliminated by using small particles and/or high gas velocity or correction made to the observed rate for this effect using theoretical relationship. Rate equation i s based on the external surface area, and is the rate of surface reaction. Table 5 continued... Ref k n Size (lim) Temp. Range °K Particle Type Remarks 71 20.4 exp (-9596/T,) (g/cm •s*atm) 1 6, 22, 49, 78 1400 - 2200 semi anthracite Experiment was performed in a laminar flow reactor using N 2 - 0 2 mixture with variable 0 2 concentration of 0.1 - 0.2 atm. Diffusion effect was taken into account from theoretical relationship. Rate equation is based on the total surface area, and is the rate of surface reaction. 51 55 exp (-20200/Tg) (g/cm »s«atm) 1 6, 22, 49, 78 1400 - 2200 semi anthracite Data obtained from referece 71 was used to correlate rate equation based on external surface area, and is the rate of surface reaction. - 33 -function of temperature. These are 2 k = 8710 g/cm *atm«s so & E/R = 17,967°K 2.3.2.4 Char-Carbon Dioxide Reaction Based on a number of studies in this area, the rate of char-carbon dioxide reaction can be correlated using the Langmuir-type adsorption equation expressed as follows (54-56). Both ?i and P3 are the partial pressures of carbon dioxide and ?2 i s that of carbon monoxide. According to this equation carbon monoxide would have an inhibiting effect on the char-carbon dioxide reaction. This was found to be the case as reported by Wen and Lee (19) and Batcheleder et al (54). The constant kj depends on the rate of adsorption of carbon dioxide on carbon, k2 is the equilibrium constant for the adsorption-desorption of carbon monoxide on the carbon surface and k3 depends primarily on the rate of reaction of the adsorbed carbon dioxide molecules on carbon as well as on the value of k^. These rate constants are however d i f f i c u l t to generate from experimental data as they are interrelated and become less attractive for gasifier design. Dutta and Wen (30) investigated the rate of this reaction for a - 34 -number of coal samples at different sizes (-20 +35, -60 +100, -35 +60, -100 mesh) in the temperature range of 840 - 1100°C. They concluded that the reaction is best described by volumetric reaction according to Equation 2.7. On the other hand, Amundson et al (40, 57-58) and Wen and Chaung (36) assumed the applicability of surface reaction for the char-carbon dioxide reaction in their work. The order of reaction for the char-carbon dioxide reaction has been found to have values ranging from zero to unity with respect to the partial pressure of carbon dioxide depending on the operating temperature and pressure. Wen and Dutta (42) suggested that a f i r s t order reaction can be assumed at low pressure up to 1 atm. As the pressure is increased typically above 15 atm, zeroth order reaction i s approached. The reported values for the activation energy l i e in the range of 55 - 65 kcal/mol. A typical value of 59 kcal/mol has been reported by Dutta and Wen (30). Differences in the values published are mainly due to the different reactivity of coal used under different experimental conditions. A number of rate equations for the char-carbon dioxide reaction correlated by different investigators are given in Table 6. 2.3.2.5 Char-Steam Reaction The mechanism for the char-steam reactions has been found to be similar to the char-carbon dioxide reaction (54,59) and the rate equation can be correlated by the Langrauir adsorption isotherm, represented by Equation 2.21. In this case Pi and P3 are the Table 6 Rate equations for carbon-carbon dioxide reaction R a t e = k P C 0 2 Ref Temp. Range °K Particle Type Remarks 71 1.35 exp (-16300/T ) (g/cm »s«atm) 63.5 exp (-19500/T ) y S (g/cm •S'atm) 1123 - 1223 1223 - 1573 g r a p h i t e g r a p h i t e Experiment was conducted in pure C0 2 using cylindrical particles. A f i r s t order reaction with respect to steam partial pressure was assumed. The rate equations given are based on the external surface area, and is the rate of surface reaction. 72 1.2 x IO"5 exp (-17600/T ) (g/cm2«s) 1013 - 1133 There were no indication to the type of carbon used in the experiment or the area to which the rate equation was based on. The rate equation given i s the rate of the overall reaction. - 36 -partial pressures of steam and P2 is the partial pressure of hydrogen. Thus the order of the char-steam reaction varies with the steam partial pressure in a similar way to that in which the char-carbon dioxide reaction varies with carbon dioxide partial pressure as supported by Walker et al (41). The retarding effect of hydrogen in the char-steam reaction has been reported, while there is s t i l l uncertainty about the effect of carbon monoxide with some data showing inhibition while the others show no effect at a l l on this reaction (54,59-61). In a manner similar to carbon-carbon dioxide reaction, Reide (60) and Klei (61) concluded that under a typical gasification temperature, the overall rate of char-steam reaction is chemical reaction controlled and i s appropriately described by the volumetric rate equation given by Equation 2.7. However these investigators were unable to correlate their experimental data based on this model but instead assumed the v a l i d i t y of surface reaction. Wen and Chaung (36) also assumed the surface reaction model to describe the char-steam reaction in their modelling of the entrained bed coal g a s i f i e r . Johnson (62) concluded that char-steam reaction should be described by the volumetric reaction and proposed the following equation to describe the rate equation: | | = f L k x ( l - x ) 2 / 3 exp(-ax 2) (2.22) fraction of carbon conversion of char f e(8467/T Q) o where x = f L = f = relative reactivity factor of coal/char T = maximum temperature to which char has been exposed prior to gasification, °K P P CO Ho exp(9.0201 - 31705/T) (1 1) P K? k - H?0 1  K I p p [1 + exp(-22.16 + 44787/T) (| + 16.35 + 43.5 ^ - ) ] 2 H 20 H 20 H 20 ? n n t PIT , P.. _ = partial pressures of CO, H, and H90 OU t i 2 t i 2 U £ R E = 1 Q(7.49 - 7079/T) a = 5 2 - 5 % + ° - 5 2 1 \ P H , 0 1 + 5 4 , 3 P % ' * 0.707 P H 2 0 + 0.5 P H ^ 2 P H 2 0 Different values of the activation energy for the char-steam reaction have been reported in the literature which l i e approximately in the range 45-65 kcal/mole. These values are close to those of the char-carbon dioxide reaction (55-65 kcal/mol). The rates of the char-carbon dioxide and char-steam reaction are typically in the same order of magnitude. This view has been supported by a number of investigators (22,30,41,54). Walker and Kini (22) suggested that the rate of the char-steam reaction is about three times that of the char-carbon dioxide reaction while Batchelder (54) reported that these two reactions have equal rates under a typical gasification temperature. - 38 -A few of the char-steam rate equations are given in Table 7. 2.3.3. Gas-Gas Reactions 2.3.3.1. Water Gas S h i f t Reaction The water gas reaction is the reaction between steam and carbon monoxide and can be represented as follows CO + H20 C0 2 + H 2 (2.23) This homogeneous gas-gas reaction is slightly exothermic having heat of reaction of about -7.8 kcal/mol (63). According to Johnstone et al (64), the water gas reaction which is catalysed by the carbon surface and the impurities present, is fast and rapidly approaches equilibrium. There are s t i l l however contradicting views reported on the mechanism of the water gas reaction. According to Lowry (11) this reaction does not take place in the gas phase but purely on the carbon surface, while Haslam (65) reported that to some extent the reaction does take place in the gas phase. Batchelder et al (54) found that the water gas reaction attained equilibrium much more rapidly at the carbon surface than in the gas phase indicating that there i s a parallel gas phase reaction. Gwosdz (66) reported that the interaction of carbon monoxide and steam takes place on the carbon surface and the proportions of gaseous components produced correspond to the equilibrium established at the surface. Different approaches in treating the water gas reaction have been Table 7 Rate equations for carbon-sfceam reaction Rate = kP n or Rate = k, C n H20 1 H20 Ref k Unit of Rate n Temp. Range °K Particle Type Remarks 60 k : = 11.94 exp (-8238/Tg) mol/cm2 «s 1 773 -1173 graphite A graphite rod was used and placed in a gaseous mixture of N 2 - H20 with steam partial pressure varied from 7-12 mm Hg. Rate equation given is the overall rate of reaction based on the external surface area. Effect of mass transfer was found to be small. 64 k 1 =2 x 10 8 exp (-32005/T ) s mol/s «g of carbon 0.58 1033 - 1089 Activated carbon The sample particle was placed in a gaseous mixture of argon and steam with mole fraction of steam varied from 0.107 - 0.534. The rate equation is the overall rate of reaction. However mass ransfer resistance was found to be small. Table 7 continued. Ref k Unit of R n Temp. Range °K Particle Type Remarks 73 k = 3190 exp (--25020/T ) s g/cm «s'atm 1 1273 - 1433 graphite Rate of reaction was g/cm2«s«atm determined in an environment k = 19.2 exp (--17680/T ) s 1 1123 - 1273 of pure steam at 1 atm. First order reaction with steam partial pressure was assumed. The rate equations given are the rate of surface reaction based on the external surface area. 74 k (-= 3.66 x 103 -26700/T ) exp g/cm2»s 0 1256 - 1589 petroleum coke The rate equations are for the actual rate of reaction s g/cm •s under the experimental k = 1.95 x 109 exp 0 1256 - 1589 electrode condition based on the -41600/T ) s graphite external surface area. However due to small particle sizes used (63 - 74 u) the rate can be assumed to be the rate of surface reaction. - 41 -used by many workers on gasifier modelling. The majority of them assumed that the reaction goes to equilibrium (36, 67), while the others (39) take into account the reaction kinetics. Purdy et al (68) found that the inclusion of the water gas reaction kinetics improved their prediction of the performance of the gasifier under study compared to the equilibrium assumption of the water gas reaction. The equilibrium constant for the water gas reaction has been correlated by Parent and Katz (63) and follows the Van't Hoff Isochore relationship. K W G = 0.0265 exp (3955.7/T) (2.24) The catalytic effect of impurities present in coal on the water gas reaction makes i t d i f f i c u l t to establish a general rate equation for this reaction. Singh and Saraf (69) studied this reaction over an iron-base catalyst and proposed the following equation: Rate = F (2.77 x 105) x ( x _ „ - x* ) exp(-27760/1.987T) x w CO CO p(0.5-P/250) e x p ( _ 8 . 9 1 + 5553/T) ( g - m o l / s of ash) (2.25) where F w = relative catalytic of ash to that of iron-base catalyst. P P 1 C 02 H2 x* = — r—-—-1 co p L K p u . J ws H^ O P = total pressure, atm P__ , PtI , P77 ^ = partial pressure of CO2, H2 and H 2 0 CO2 t i 2 H2O respectively, atm - 42 -K w g = equilibrium constant for water gas reaction (eqn. 2.24) 2.3.3.2 Gaseous Combustion A number of the gaseous components that are generated in a gasifier can undergo further combustion with oxygen to produce more heat. These include hydrogen, carbon monoxide and most of the components released due to pyrolysis. The rates of gaseous combustion are much faster than those of the gas-solid reactions (36,42). According to Wen and Chaung (36) gaseous combustion can be considered instantaneous and therefore can be assumed to be completed as long as there is enough oxygen. - 43 -3 . AIM AND SCOPE OF PRESENT WORK A l l the previous studies on spouted bed coal gasification concentrated on assessing the performance of the system based on the overall conversion achieved from relatively small systems (13-15). In order to improve the understanding of the spouted bed coal gasifier in general, i t i s necessary to obtain some information regarding the internal behaviour of the system which is expected to be related to the hydrodynamics of spouted beds. Previous studies on spouted beds (1) indicated that the bed hydrodynamics strongly depend on the column diameter. As i t is very l i k e l y that large bed diameters would be employed i f spouted bed gasifiers are to be chosen for commercial operations, the development of a reliable mathematical model is essential for design and scale-up purposes. It is therefore the aim of the present work to perform a study of the in-bed gas composition and temperature. In addition the theoretical model developed by Mathur and Lim (2,4) i s applied to gasification and tested against the experimental data obtained. Based on the observation reported by Rovero et al (7) regarding the existence of axial and radial profiles of gaseous components in their studies on the decomposition of ozone in a spouted bed and the RTD measurement of tracer gas made by Lim and Mathur (4), the following experimental measurements w i l l be performed. - axial and radial concentration profiles of gaseous components and gas temperature profile in the annulus. - 44 -- axial concentration profiles of gaseous components and gas temperature profile in the spout. The effect of the following operating variables on the resulting profiles w i l l be studied: - operating with a char bed as opposed to an inert bed - gasification with air/steam mixtures - gasification with oxygen/steam mixtures - variation in types of coal fed to the gasifier - bed height - char rather than coal used as feed Gaseous samples collected w i l l be analysed for the following components: oxygen, hydrogen, methane, nitrogen, carbon dioxide and carbon monoxide. - 45 -4. EXPERIMENTAL APPARATUS 4.1 Gasification System The existing 0.3 m diameter unit built under a contract awarded by Energy Research Laboratories, CANMET was used throughout the work.. Detailed equipment description and designs are given elsewhere (75) and only a brief description w i l l be presented here. The schematic diagram of the system is given in Figure 4. Coal which i s stored in a 45-gallon drum is fed to and metered through a vibra screw feeder and pneumatically transported using air or oxygen to either the top or the bottom- of the bed. For this study only bottom feeding was used. In this case coal feeding into the bottom of the reactor i s made through a 31.8 mm inlet o r i f i c e . The reactor i s 0.5 m OD x 1.8 m high cylindrical column with a 60° conical base. To tolerate high gasification temperature the column i s refractory-lined to an inside diameter of 0.305 m. Solid entrainment is reduced by provision -of a 0.77 m ID x 0.62 m high disengaging section above the cylindrical section. The spouting medium which is also the gasification medium is injected into the reactor through the same coal inlet o r i f i c e . Hot gaseous products with some entrained coal fines leave the top of the gasifier and pass through a series of two cyclones provided with char receivers at the bottom, three double pipe heat exchangers and f i n a l l y to a knock-out drum and a dust f i l t e r before being burnt in the incinerator. In a number of selected runs, char collected from the f i r s t cyclone was recycled into the reactor via a rotary valve. TRANSPORT OXYGEN/AIR STEAM AIR PROPANE *l TO * INCINERATOR GAS SAMPLE -txd—5 BED SAMPLE 1 COAL DRUM 7 CHAR RECEIVERS 2 VIBRA SCREW FEEDER 8 ROTARY VALVE 3 PREHEATER 9 HEAT EXCHANGERS 4 RUPTURE DISK 10 KNOCK- OUT DRUM 5 REACTOR AND FILTER 6 CYCLONES II 12 ORIFICE METER EXHAUST FAN Figure 4 - Schematic flow diagram for the g a s i f i c a t i o n system. - 47 -For start-up purposes a propane fired gas burner is provided in order to preheat the reactor to the desired experimental temperature. 4.2 Gas Sampling A gas sampling probe was designed as shown in Figure 5. The probe consists of a 356 mm long x 6.35 mm OD SS 316 tube. Mounted at one end of the tube is a 12.7 mm OD porous stainless steel f i l t e r with an average pore opening of 90 \i. In order to be able to measure the gas temperature at the suction point a 1.59 mm (1/16") Chromel-Alumel thermocouple is inserted inside the tube. This arrangement is achieved by having a branch tee as shown in the diagram. A valve is incorporated so that the probe can be isolated from the sampling line whenever necessary. The schematic diagram for the sampling line is shown in Figure 6. The gas sample drawn from the gasifier and which is free from any solid material due to the stainless steel f i l t e r mounted at the tip of the probe is passed through a knock-out drum where most of the condensed steam is removed. The dry gas leaving the knock-out drum is collected in a gas sampling tube fitted with two valves at both ends. A vacuum pump is used to suck the gas sample at a flowrate measured using a rotameter installed in the sampling line . A nitrogen line taken from a high pressure nitrogen cylinder is provided to clear off any possible blockage caused by the accumulation of coal dust on the stainless steel f i l t e r . - 48 -CO Figure 5 - Gas sampling probe. (—\ V 7 > TO VENT 1 PROBE z GASIFIER WALL 3 N 2 CYLINDER 4 ISOLATING VALVE 5 KNOCK-OUT DRUM 6 GAS SAMPLING TUBE 7 ROTAMETER 8 VACUUM PUMP Figure 6 - Schematic flow diagram for the gas sampling l i n e . - 50 -5. EXPERIMENTAL TECHNIQUE 5.1 G a s i f i c a t i o n Process Detailed experimental techniques for the o v e r a l l g a s i f i c a t i o n process have been described by Watkinson et a l (76) and only a b r i e f d e s c r i p t i o n w i l l be presented here. A known amount of crushed coal or i n e r t ( depending on the bed material selected) was i n i t i a l l y charged into the reactor and was then heated up using combustion products from the external nozzle mixed propane burner. In t h i s case the gas from the burner formed the major portion of the spouting f l u i d . As soon as a temperature of about 450°C was attained i n the bed, continuous coal feeding into the reactor was i n i t i a t e d , while the propane input was slowly reduced to zero. Due to the coal combustion in s i d e the reactor, the bed temperature continued to r i s e to the desired value and the steam flow was turned on. The a i r to coal r a t i o was then adjusted and the system was allowed to reach a steady state which normally took about one hour a f t e r feed adjustments before any gas sampling could be performed. The g a s i f i c a t i o n experiments were c a r r i e d out by Mr. Gordon Cheng and Mr. Sam Low as part of the requirements of the research contract. 5.2 Gas Sampling Six d i f f e r e n t a x i a l positions and three d i f f e r e n t r a d i a l p o s i t i o n s corresponding to each l e v e l have been chosen for t h i s study. This required six sampling probes, one to be placed at each l e v e l . - 51 -These various positions are shown schematically in Figure 7. An isokinetic gas sampling technique was used throughout this work. Hence at a particular location, the gases velocities inside the sampling probe and that inside the gasifier should be equal. Each of these velocities and therefore the actual volumetric gas flowrate to be drawn were theoretically determined. The Mamuro-Hattori equation (1) was used to determine the axial gas flow distribution in the annulus, while the axial flow distribution in the spout was obtained from a mass balance as described by Mathur and Lim (2). For this purpose the McNab equation (1) was used to calculate the average spout diameter. The gas sampled from a chosen position was f i r s t allowed to pass through the sampling line for about 8 min before i t was collected and isolated from the line by closing the two valves at the ends of the gas sampling tube. Typically i t took about 2-3 min to purge out the whole sampling line completely. Meanwhile, the actual gas temperature was recorded at every 2 min interval. The sample collected was subsequently stored in a pressure-lock gas syringe for analysis at a later time. While working on a particular probe, a l l the other five remaining were pulled out to positions such that their tips would remain inside the gasifier wall. This procedure was practised to minimise the disturbance of the gas flow path which would have been caused by the presence of these probes. There was no set sequence for any particular sampling position to be chosen. This was done at random. In order to construct the average gas temperature profiles in the gas i f i e r , instantaneous gas temperatures at a l l the 18 locations - 52 -19.8cm No. z (cm) r (cm) No. z (cm) r (cm) No. z (cm) r (cm) 6A 65 11.4 4A 45 11.4 2A 25 11.4 6B 65 6.3 4B 45 6.3 2B 25 6.3 6C 65 0 4C 45 0 2C 25 0 5A 55 11.4 3A 35 11.4 1A 8.5 5.0 5B 55 6.3 3B 35 6.3 IB 9.1 4.0 5C 55 0 3C 35 0 1C 11.4 0 Figure 7 - Exact locations of sampling points in the bed. - 53 -involved were taken almost instantaneously. Three different sets of readings, taken at the beginning, in the middle and at the end of the sampling process were recorded and averaged. 5.3 Blockage of Sampling Probes Blockage of the gas sampling probes was experienced; however this was not frequent. The main reason for the blockage was the accumulation of coal fines on the f i l t e r elements mounted at the tips, and these were removed by flushing with nitrogen. In order to e f f i c i e n t l y flush away these particles, the whole sampling line was isolated beforehand by closing the isolating valve as indicated in Figure 6. 5.4 Gas Analysis The gas analysis was performed using a Hewlett-Packard 5710A gas chromatograph. The unit was equipped with a thermal conductivity detector and an automatic integration system. Two different columns were used to separate the gaseous components. These were a 3.2 mm x 2.13 m molecular sieve A column to separate hydrogen, oxygen, nitrogen, methane and carbon monoxide and a 3.2 mm x 3.96 m Porapak Q column to separate carbon dioxide. Throughout the analysis both the oven and the detector were maintained at temperatures 80°C and 200°C respectively. To calibrate the G.C. five gas samples having known quantities of different species were injected into the system. A calibration table as shown typically in Appendix C w i l l then be generated, which w i l l be used as the internal standard. Before performing any analysis a standard gas sample was f i r s t analysed to ensure that the calibration is correct. - 54 -6. MODEL DEVELOPMENT AND SIMULATION Steady state models of char gasification in a spouted bed blown with air/oxygen and steam have been developed based on the general one-dimensional and the strearatube spouted bed models proposed by Lim and Mathur (2,4) and the gasification model of Yoshida and Kunii (76) which they applied to the fluidized bed. In both cases isothermal gasifiers have been assumed. 6.1 Development of Model 6.1.1 Dominant Reactions The following reactions have been considered to describe the gasification process gasification: C + H20 + CO + H 2 (6.1) C + C0 2 * 2C0 (6.2) combustion: C + 0 2 •»• C0 2 (6.3) CO + 1/2 0 2 > C0 2 (6.4) H 2 + 1/2 0 2 + H20 (6.5) water gas shift: CO + H 2 0 ^ ± C 0 2 + H 2 (6.6) Following the correlation proposed by Wen and Lee (19) to determine the product distribution of char combustion, carbon dioxide has been assumed to be the only product (Equation 6.3). The water gas shift reaction was - 55 -assumed to be in equilibrium throughout the bed. At a typical gasification temperature and atmospheric pressure, the rate of reaction between hydrogen and char was found to be insignificant (28-29). This reaction has been ignored in the present model. 6.1.2 Reaction Kinetics A simple reaction model based on the particle geometric surface area has been used to describe' reactions 6.1 - 6.3. In this case the resistances due to gas transport in the boundary layer and in the ash layer have been ignored and the rate is assumed to depend on temperature and the partial pressure of reacting species given as follows: Rate = k „ exp(-E/RT)Pn (6.7) overall v g The different kinetic parameters chosen for this simulation are given in Table 8. The rates of gas phase combustion given by equation 6.4 - 6.5 are normally fast compared to char combustion as concluded by Wen and Chaung (36). These rates of reaction are usually high and can be considered to be instantaneous. Hence these reactions were assumed complete as long as there was oxygen existing. 6.1.3 Bed Hydrodynamics The presently available correlations on the hydrodynamics of spouted beds have been used for this purpose, with the assumption of Table 8 Kinetic parameters used in gasification modelling A. Reaction kinetic *Rate = k „ exp(-E.,/RT) X? overall^ r i i Reaction k E./R X. n Unit of Rate Ref. overall^ i i Char-02 8710 Char-C02 4 x 108 **Char-H20 1.2 x 109 18000 P n (atm) gl s cm' 48 2 9 7 9 0 C„n (mol/ni ) CO 2 2 9 7 9 0 CJJ n (mol/m3) mol/m" mol/m' 3 0 B. Equilibrium constant for water-gas shift reaction (63) K = 0.0265 exp (3955/T) ws r *Based on external suface area **Assuming rate constant of C/H20 reaction is three times the rate constant for the C/C02 reaction. - 57 -their validity at high temperatures. These different correlations are given in Table 9. 6.1.4 One—Dimensional Model This model assumed plug flow of gas both in the spout and in the annulus. In developing the equations describing the gasification process in spouted bed, the procedure proposed by Yoshida and Kunii (76) , has been adopted. Using this method the change in gas concentration due to gas-solid and gas-gas reaction was accounted for separately. U a + dUa C a i + dC a i Aa Ea ANNULUS UT T f z + Az As £s SPOUT 1 T ' S I i Species 1 H 20 2 CO 3 4 c o 2 5 0 2 6 N 2 6.1.4.1 Reactions i n the Spout A. Changes due to gas-solid reactions A balance over spout height AZ for component i gives Table 9 Hydrodynamic parameters for model Parameter Equation Ref. Maximum spoutable bed height H = 0.105 d (d /d ) 0 , 7 5 (d /d,)° m c c p c i •4 X 2/p p 1' 2 1 Annulus gas velocity (superficial) U = 0.88 U . (1 - (1 - z/H ) 3) a mr m 5 Minimum spouting velocity 1/3/ 2g H U -lI = (d /d ) (d./d ) / E_ ms p c i c \J Pf) 1 Mean spout diameter n n no P 0 - 4 9 A °-68/ °-41 D = 0.118 G d / p, s c b 1 Annulus voidage e = 0.43 a Spout voidage e = 1 - 0 . 2 z/H s 6 Spout gas velocity (superficial) «s = I U s A c - U a ( A c - A s > ] / A s Minimum fludization velocity Umf = °' 5 ( U b f + *tf> 77 Velocity beginning of fluidization (Re), . = [(18.1) 2 + 0.0192 G a ] 0 , 5 br - 18.1 77 Velocity total fluidization (Re) f c f = [ ( 2 4 ) 2 + 0.0546 Ga] °* 5- 24 77 - 59 -Uo K C Q, = [Aa U + d(U A J ] [C . + d C .] s s s i L s S S S SI S1 J + U ir D C . dz + I R. A (1 - e ) dz (6.8) r s s i . i s s J where Rj, j = 1,2,3 are the rate of reactions represented by equations 6.1 - 6.3 respectively expressed as kmol/(m of particle volume s). Ignoring any change in volumetric flux due to changes in molar flux and pressure: r TTD dz s s s Combining Equation 6.8 and 6.9 and assuming a constant spout diameter, gives the following simplified equations for individual components: d C s l - R l ( 1 " £s> dz U s dC . (R + 2R,) (1 - e ) s2 _ 1 2 s dz — U s d C s 3 R l ( 1 ~ £ s ) dz -U s dC s 4 (R 3 - R 2) (1 - s s) (6.10) dz U s - 60 -d C s 5 - R 3 ( 1 - £ s )  d z " U s dC , dz Equation 6.10 can be rewritten to give the following expression for the change in component concentration A C S - L over bed height Az resulting from gas-solid reaction -R (1 - e ) AC = — _ Az (6.11) U s .(R. + 2R„) (1 - £ ) AC = — Az (6.12) 3 2 U s R, ( 1 - e ) AC - — — A z (6.13) S J ^ s (R- - R ) (1 - e ) AC , = — - =— — Az (6.14) S ^ U s AC = — ^ — Az (6.15) 8 3 U s AC g 6 = 0 (6.16) - 61 -Thus the component concentration CV^ resulting from gas-solid reactions are given by: C . 1 = C , + AC . s i s i s i C ' = C 0 + AC „ s2 s2 s2 Cs3' = Cs3 + A C s 3 <6-17> C . ' = C , + AC • s4 s4 s4 C ' = C , + AC c s5 s5 s5 Cs6' = Cs6 B. Changes Due to gaseous combustion On the assumptions that the gaseous combustion proceeds instantaneously, and as long as there is oxygen present the change in component concentrations AC' S£ due to gaseous combustions is given by A C s l ' = Cs3' A C s 2 ' - - Cs2' AC s 3' = -C g 3- (6.18) AC . 1 = C .' s4 s2 - 62 -AC ' = -{C J + C ') s5 s2 s3 AC , ' - = 0 s6 Combining Equations 6.17 and 6.18 gives the component concentration C s i " after taking into consideration both the gas-solid reaction and the gaseous combustion which can be represented as follows: C s l " = C s l + A C s l + Cs3 + A C s 3 . Cs2" * 0 Gs3" = 0 (6.19) Cs4" = Cs4 + A Cs4 + Cs2 + A Cs2 C = C , + AC , - C , - AC', - C , - AC , s5 s5 s5 s3 s3 s2 s2 s6 s6 According to Yoshida and Kunii (76) who modelled fluidized bed coal gasification, the conversion predicted by taking into consideration gaseous combustion was very close to conversion where such reactions are neglected. The importance of considering such reactions in spouted bed char gasification modelling w i l l be discussed later. - 63 -C. Equilibrium of Water Gas Shift Reaction If ACS is the individual change in molar concentration of H20, CO, H2 and C02 due to water gas shift reaction, Yoshida and Kunii (76) state that the following relationship must be satisf i e d . ( C s 4 " + ACs) ( C s 3 " + ACs) K. ws (c " - AC ) (c - AC ) s2 s s i s A r <Kws Cs2" Csl"> - « W Cs 3"> s = C ." + C " + K C " + K C " ( 6 ' 2 0 ) s4 s3 ws s2 ws s i D. Equations Describing Reaction i n the Spoafe In the situation where CS2" and Cs3" equal zero, Equation 6.20 is not applicable and the fi n a l component concentration Cs-j/ ' ' at height z + Az equals Cs^" represented by Equation 6.19. On the other hand i f the gaseous combustion is neglected or when there i s no oxygen present, the f i n a l concentration C s i ' " at height z + AZ can be obtained by combining Equations 6.17 and 6.20: C . 1 " = C . + AC . - AC si s i s i s C 0 ' " = C „ + A C 0 - A C s2 s2 s2 s C o " ' = C - + AC,+ AC (6.21) s3 s3 s3 s C . ' ' • = C . + AC . + AC s4 s4 s4 s - 64 -C ' ' ' - C - + AC s 5 s 5 s 5 s6 s6 6.1.4.2 Reactions i n the Annulus A. Changes Due to Gas-Solid Reactions A balance over the annulus height Az for component i gives U a A a C a i + V D s C s i d z = t A a U a + d ( A a U a ) ] [ C a i + d C a i ^ + ER.A (1 - e ) dz (6.22) J a a where U r is given by U = -L- d (u A ) (6.23) r TTD dz a a s Combining 6.22 and 6.23 and rewriting the resulting equations as the change in concentration AC ai over the annulus height z + AZ gives -R.(l-e ) (C . - C .) dU (C . - C ,) dA .„ r 1 a 7 si a l y a , si a l ' a, 0. . AC , = [ + - j — + T —] Az (6.24) al U U dz A dz ( V 2R 2)(l-e a) ( C g 2 - C a 2) dU a ( C s 2 - C a 2) dA A Ca2 = C U + U d7 + A • ~ d l ] A Z ( 6 * 2 5 ) - 65 -. c _ . R l ^ - e a > . ( C s 3 - C a 3 > d U a + ^ C s 3 " C a 3 > d A a , . (f. _ A C a 3 - [ U + U d i + A d 7 ] A z ( 6 ' 2 6 ) a a a (R - R 2 ) ( l - e a ) (C 4 - C ) dU (C -'C ) dA A Ca4 - I " I + U " d t + A i f ] A Z < 6 ' 2 7 > - V 1 - 6 ^ + ( Cs5 - Ca5> d Ua + <Cs5 ' Ca5> d A 3 ] . A Ca5 = + U d7 + A dz"1 A Z ( 6 ' 2 8 ) AC a 6 = 0. (6.29) B. Changes Due to Gaseous combustion and Water Gas Shift Equilibrium  and Equations Describing Reactions in the Annulus Following the same procedure previously adopted, the expressions for ACa due to gaseous combustion, water gas shift reaction and the f i n a l molar concentration at annulus height z + Az were obtained which were in the same form as Equations 6.19 - 6.21, but at conditions referring to those in the annulus. - 6 6 -6.1.5 Streamtube Model Q(J-I) N+l N jjj-l 2 I In this model as proposed by Lim and Mathur (4) ver t i c a l plug flow of gas in the spout and plug flow along curved streamlines in the annulus have been assumed. No mixing or dispersion was allowed in the - 67 -direction of flow and normal to the streamlines. 6.1.5.1 Gas Flow Path To determine the gas flow path in the annulus, grid points were constructed by dividing the bed into M equal vertical intervals and the annulus width at the top into N equal horizontal intervals with each interval representing one flow path. Letting Q(J-l) be the volumetric flow between streamlines J - l and J , taking overall gas mass balance at any level I gives Q(J-l) = U (I)ir[R(J-l, I ) 2 - R(J,I) 2] (6.30) a where U a(I) is the superficial annulus gas velocity at level I and R(J,I) specifies the ordinate of the streamline J at level I (the column axis i s taken as the reference). The f i r s t streamline (J=l) was assumed to coincide with the column wall. Letting the interval between streamlines J and J+l be the flow path J, the superficial gas velocity U a z(I) along flow path J was calculated using the relationship. U (I) - U (I)/Cos 9 (6.31) az a a In this model 6 a was taken to be the angle formed between the line joining the centre of flow path J at levels I and 1-1 and the horizontal. Denoting PL(I) as the distance between these two centre points, the total path length for flow path J from the spout-annulus interface to the top of the bed at I=M was given by E PL(I). I - 68 -6.1.5.2 Reactions in the Spout Since the same assumptions were made with regard to the gas flow in the spout as in the one-dimensional model, the reactions in the spout were described by the same equations, i.e. equations 6.17 - 6.21. 6.1.5.3 Reactions i n the Annulus A. Changes Due to Gas-Solid Reactions A component mass balance over length AL along the flow path J gives U A C , = [A U + d(A U )][C . + dC .] az az azi 1 az az az az azi a z i J + E R. A (1 - e ) dL (6.32) l az az J - 69 -On the assumption of no change in volumetric flux due to the change in molar flux and pressure, Equation 6.32 reduces to the following expressions for the change in the individual component concentration A C a z i over the path lengh AL due to the gas-solid reactions: -R. (1 - e ) AC , L — — ^ (6.33) azl U az (R + 2R?)(1 - e ) AC . i ^ — AL (6.34) azz u az V 1 - ^ A Caz3 = U *> ( 6 * 3 5 ) az (R - R )(1 - e ) AC . =^ — AL (6.36) az4 U az _ R 3 ( 1 " e a z } AC _ — — AL (6.37) az5 U az AC , = 0 (6.38) az6 In this case e a z = ea« B. Changes Due to Gaseous Combustion and Water Gas Shift Equilibrium Following the same procedure adopted in a l l the previous cases, the expressions for AC a z due to the water gas shift equilibrium and the f i n a l component concentrations at the path length L + AL were - 70 -derived and were in the same form given by Equations 6.18 - 6.20 but at conditions referring to those in the flow path J . 6.1.6 Outlet Component Gas Concentration The f i n a l component gas concentration leaving the gasifier was calculated by combining the streams leaving each region. Thus the exit gas concentration C e i for component i can be expressed as One-dimensional model: U u A C „. + U „ A C ... n aH a aHi sH s sHi ... C . = rr—r (6.39) ei U A s c Streamtube model: N-l _ C e i • t , ^ UaH AazHj CazHij1 + UsH A s C s H i . ( 6' 4 0> U s A c Any reaction in the fountain has been ignored. 6.2 Calculation Procedure The equations describing the change in gas concentration due to the gas solid reactions given by Equations 6.10 - 6.15 for the spout, Equations 6.24 - 6.28 for the annulus in the one dimensional model and Equations 6.33 - 6.37 for the flow path J in the streamtube model are non-linear d i f f e r e n t i a l equations and were solved using the 4th order Runge-Kutta technique. - 7 1 -The component concentrations i n the one dimensional model were obtained by a l t e r n a t e l y solving the equations for the spout and the annular region over any incremental height Az. Both of these regions have i d e n t i c a l i n i t i a l boundary conditions which were equal to the i n l e t gas component concentration. For the streamtube model, the gas concentration p r o f i l e s i n the spout were i n i t i a l l y obtained using the i n l e t gas concentration as the boundary conditions. The gas concentrations along the gas flow path J were then solved using the gas concentration in the spout taken at the mid point between streamlines J and the J + 1 as the i n l e t c o n d i t i o n . 6 . 3 Model Simulation and Discussion A number of simulated runs were performed using both the one-dimensional and the streamtube models to predict the a x i a l and the r a d i a l gas composition p r o f i l e s as well as the o v e r a l l conversion achieved from a spouted bed coal g a s i f i e r . The column geometry used i n the experimental system has been chosen for t h i s purpose, using air/oxygen and steam as the only g a s i f i c a t i o n medium. In a d d i t i o n , the importance of some operating v a r i a b l e s such as the t o t a l bed height and the bed temperature i n determining the g a s i f i e r performance was also i n v e s t i g a t e d . 6 . 3 . 1 Axial Gas Composition Profiles Figure 8 shows the a x i a l gas composition p r o f i l e s i n the annulus predicted using the one-dimensional model. For both carbon monoxide and hydrogen, the gas composition increases with height, reaching a maximum - 7 2 -Figure 8 - Axial gas composition profiles in the annulus predicted using the one-dimensional model. - 73 -value in the conical region and reduces to a minimum at the cone-cylinder intersection. The concentration of combustible species increases steadily thereafter with the carbon monoxide showing a sharper incr ease than the hydrogen. The C—CO2 reaction was partly responsible for this sharp increase in carbon monoxide as a function of height. The low H2/CO ratio predicted throughout the bed was due to the small amount of steam used in the feed, i.e. air/steam (w/w) = 8. These proportions of H 2 and CO would actually become closer to each other i f the steam/ oxygen ratio in the feed stream was higher. This i s clearly shown in Figure 9 which gives the axial gas composition profiles in a spouted bed gasifier blown with oxygen/steam. In this case an oxygen/steam (w/w) ratio of unity was used as the feed. The appearance of maximum and minimum values for these gases was due to the change in the annulus area along the bed height in the conical region. The combined effect of the variation in annulus area Ag and the annulus superficial velocity U a with bed height causes equations 6.25 and 6.26 which described the change in carbon monoxide and hydrogen concentration to change sign as the calculation proceeds from the bed inlet to the cone-cylinder intersection. On the other, hand the carbon dioxide composition shows a maximum value at the cone-cylinder interface and steadily decreasing values thereafter. The sharp increase in carbon dioxide close to the bed inlet was due to the product of combustion as shown by a sudden diminishing of oxygen. Within this bed height the term R3 in Equation 6.27 becomes dominant to give a positive increase in carbon dioxide concentration. The oxygen concentration f a l l s to zero in that f i r s t few centimeters. Once the oxygen concentration diminishes R3 becomes zero - 74 -Figure 9 - A x i a l gas composition p r o f i l e s i n the annulus predicted using the one-dimensional model blown with oxygen rather than a i r . - 75 -and further increase in heights cause the carbon dioxide concentration to decrease. The corresponding spout gas composition profiles are shown in Figure 10. In contrast to those in the annulus there is hardly any reaction taking place in this region except for some combustion. The oxygen content thus remains high. The main reason for the absence of any significant reactions is the high gas velocity in the spout, which results in a very small change in the components' concentration along the spout height as represented by Equation 6.10. The spout voidage has no significant effect in determining the gas concentration profiles in the annulus. Using an average constant value of 0.65, a value measured by Lim (3) at z = H, produced an almost indistinguishable result from the one shown in Figure 10. Rovero et al (7) made the same conclusion regarding the effect of voidage on conversion in the spout. In order to investigate the importance of carbon monoxide and hydrogen oxidation in the modelling of spouted bed char gasification, a run was simulated in which these two reactions were ignored. In comparing the results, no significant differences could be detected. There was hardly any reaction between steam and carbon dioxide with char in the spout and the char oxidation in the annulus was almost instantaneous in u t i l i z i n g the oxygen as soon as i t enters the reactor. Hence carbon monoxide and hydrogen never coexist with oxygen anywhere in the gasifier and very l i t t l e error would be introduced by ignoring the combustion of carbon monoxide and hydrogen in models where radial mixing is assumed not to be present. - 76 -F i g u r e 1 0 - A x i a l g a s c o m p o s i t i o n p r o f i l e s i n t h e s p o u t p r e d i c t e d u s i n g t h e o n e - d i m e n s i o n a l m o d e l . - 77 -The t o t a l bed height was found to have no s i g n i f i c a n t e f f e c t on the a x i a l gas composition p r o f i l e s . As shown in Table 9 a l l of the c o r r e l a t i o n s used in p r e d i c t i n g the bed hydrodynamics except for the spout voidage and the minimum spouting v e l o c i t y were independent of the t o t a l bed height. As has already been shown the e f f e c t of spout voidage on conversion was n e g l i g i b l e , while a constant overal gas mass flowrate rather than a constant U s/U m s , , , . s mt> r a t i o has been used in the p r e d i c t i o n . However, under a given operating condition U m s was used as a check, to ensure that the r a t i o U s/U m s i s greater than 1. Figure 11 shows the a x i a l p r o f i l e s for carbon monoxide along the d i f f e r e n t flow path predicted using the streamtube model. For a l l the flow paths, the gas composition r i s e s s t e a d i l y along the path length from the spout-annulus i n t e r f a c e to the top of the bed, with the one c l o s e s t to the wall increasing most sharply and then l e v e l l i n g o f f as i t reaches the bed top. For comparison the average p r o f i l e predicted by the one-dimensional model was also p l o t t e d . The only d i f f e r e n c e that can be observed i s the absence of any maximum/minimum value in the c o n i c a l region for the streamtube model as compared to the predictions of the one-dimensional model. This d i f f e r e n c e i s a t t r i b u t e d to the independence of the equations describing the change in the gas concentration along the path length i n the annulus area A a (Eqn. 6.33 - 6.37). Plo t t e d i n Figure 12 are the a x i a l carbon monoxide and hydrogen composition p r o f i l e s along the centre of the annulus, obtained using the one-dimensional and the streamtube models. The p r e d i c t i o n s from both models are r e l a t i v e l y close except for the lower section of the bed. - 78 -F i g u r e 11 - C a r b o n m o n o x i d e p r o f i l e a l o n g s t r e a m l i n e s i n t h e a n n u l u s p r e d i c t e d u s i n g t h e s t r e a m t u b e m o d e l . - 79 -F i g u r e 1 2 - C o m p a r i s o n o n t h e a x i a l h y d r o g e n a n d c a r b o n m o n o x i d e p r o f i l e s a l o n g t h e c e n t r e o f a n n u l u s ; , o n e -d i m e n s i o n a l m o d e l ; , s t r e a m t u b e m o d e l . - 80 -6 . 3 . 2 R a d i a l G a s C o m p o s i t i o n P r o f i l e s Figure 13 shows the radial gas composition profiles at the top of the bed predicted using the streamtube model. As expected more carbon monoxide and hydrogen are found close to the column wall due to the longer path length and gas residence time in the corresponding flow paths. Meanwhile the carbon dioxide composition rises steadily from i t s original value in the spout to a maximum and decreases steadily towards the column wall. This maximum value, located close to the spout-annulus interface is due to the char combustion as the gas from the spout entered the corresponding flow path. At the same time the oxygen composition is reduced to zero. Figure 14 shows the radial carbon monoxide composition profiles at different levels in the bed which can be deduced directly by following a vertical straight line drawn at any bed level z in Figure 11. As shown there i s a large variation in gas composition along the middle section of annulus. In contrast to the streamtube model, the one dimensional model f a i l s to give any radial variation in the gas composition profiles as shown in Figure 15 for carbon monoxide and hydrogen. This is not surprising as uniform radial gas concentration has been one of the assumptions made in the i n i t i a l model development. 6 . 3 . 3 E x i t G a s C o m p o s i t i o n In predicting the exit gas composition the spout gas and the annulus gas have simply been added, assuming that no reaction occurs in the freeboard. Obviously in a reactor operating at 1073 K, reaction - 81 -Figure 13 - Radial gas composition profiles at the top of the bed predicted using the streamtube model. - 82 -Figure 14 - Radial carbon monoxide composition the annulus at various bed levels predicted using the streamtube model. - 83 -Dimensionless radius (r/R c) Figure 15 - Comparison of the radial hydrogen and carbon monoxide profiles in the annulus at the top of the bed; , streamtube model; , one-dimensional model. - 84 -would occur. These gas-gas reactions in the fountain w i l l be considered later in Chapter 8, where comparison w i l l be made between the predicted and the experimental results. 6.3.3.1 Effect of Total Bed Height Figure 16 gives the variation in the component gas composition as a function of the total bed height predicted using the one-dimensional model. The proportions of the carbon monoxide and hydrogen in the exit stream increase as the total bed height increases, while that of oxygen steadily decreases. As discussed earlier and from Figures 8 and 10, a l l of the carbon monoxide and hydrogen in the product stream is due to the reaction in the annulus. A l l of the oxygen in the annulus is consumed almost instantaneously, while in the spout there is insufficient reaction to consume the oxygen. High total bed height results in high carbon dioxide concentration in the spout at the top of the bed and low concentration in the annulus measured at the same level as shown in Figures 8 and 10. The combination of these two streams results in the observed carbon dioxide concentration profiles in which carbon dioxide increases to a maximum and then decreases thereafter as the total bed height is increased. Hence improved gas quality from a spouted bed gasifier would be obtained as the total bed height increased. The gasifier would give the best performance when operated at i t s maximum spoutable height. Theoretical investigation by Littman et al (6) on catalytic reaction in the spouted bed gave the same conclusion with regard to the effect of the total bed height on the overall conversion. - 85 -Figure 16 - Effect of the total bed height on the exit gas composition predicted using the one-dimensional model. Assume no reaction of spout and annulus gas. Overall volumetric gas flow is kept constant. - 86 -0 4 8 12 16 2 0 24 Gas composition , vol % (dry) Figure 17 - Comparison of the exit carbon monoxide (^) and hydrogen (#) composition predicted using the one-dimensional model (abscissa) and the streamtube model (ordinate). Assume no reaction between spout and annulus gas. Overall volumetric gas flow i s kept constant. - 87 -Under the same operating conditions the variation in the exit gas composition with the total bed height was also predicted based on the streamtube model. No significant differences between these two models could be detected as shown in Figure 17, which gives the comparison for carbon monoxide and hydrogen. 6 . 3 . 3 . 2 Effect of the Air/Steam Ratio Figure 18 shows the variation in the product gas distribution as a function of the inlet air/steam ratio predicted using the one-dimensional model. In general much higher percentage of carbon monoxide and hydrogen and hence improved gas quantity could be obtained at low air/steam ratios. Increasing the air/steam ratio causes both the hydrogen and carbon- monoxide to decrease with the hydrogen composition f a l l i n g off more sharply than that of the carbon monoxide. Operating at low air/steam would give a high proportion of hydrogen in the product gas. On the other hand fuel gas richer in carbon monoxide would be produced by maintaining high air/steam ratios. L i t t l e variation in carbon dioxide composition was observed within the range investigated. The oxygen fraction increases slightly as the air/steam ratio increases, since more unreacted oxygen in the spout w i l l leave the gas i f i e r . As in the previous.case, predictions using the streamtube model gave a trend undistinguishable from the simpler model. In practice i t is very d i f f i c u l t to operate at a low air/steam ratio in order to maintain a reasonably high gasification temperature, as very l i t t l e oxygen i s available for combustion. Extra heat can, of course, be supplied by recirculating hot solids from an external - 88 -Figure 18 - Effect of air/steam ratio on the exit gas composition predicted using the one-dimensional model. Assume no reaction between spout and annulus gas. - 89 -combustor, but here we are considering only systems blown with gases containing oxygen. For these systems when oxygen is used instead of ai r , a high proportion of steam is required to control the temperature, particularly i f operation in the dry-ash mode is desired. 6.3.3.3 Effect of the Average Bed Temperature Figure 19 shows the variation of the component gas composition as a function of the average bed temperature again assuming simple mixing of spout and annulus gas without reaction. As expected on increasing the average bed temperature, the proportion of carbon monoxide and hydrogen in the product stream increases. This was due to the strong dependence of both the C-CO2 and the C-H2O reactions on the reaction temperature. The decrease in the oxygen composition at high reaction temperature was due to the increase in the extent of combustion in the spout to give a high carbon dioxide composition in this region. At the same time higher temperature results in a much lower carbon dioxide composition in the annulus. The combination of these two streams at the top of the bed lead to the observed carbon dioxide profile, showing a minimum value at around 820°C and a slight increase as temperatures increase. However at a much higher temperature when a l l of the oxygen has been consumed, a decrease in the carbon dioxide composition is expected. As in a l l earlier cases, both the streamtube and the one-dimensional models gave the same prediction. 6.3.3.4 Effect of Particle Size Figure 20 shows the variation in the overall gas composition as a function of particle diameter. - 90 ->» T3 O > o CL £ o o CO o CD 28 24 20 75 I 6 1 dp = 2.1 mm H = i.2m A i r/H 2 0 (w/w) = 8 750 8 0 0 8 50 9 0 0 Bed temperature °C Figure 19 - Effect of the average bed temperature on the exit gas composition predicted using the one-dimensional model, Assume no reaction between spout and annulus gas. - 91 -28 2 4 -_ 2 0 O C o o Q. E o o if) o I 2 8 0 1 1 T= 1073 K _ H = 1.2 m A i r /H 2 0 (w/w) = 8 Balance N 2 ; ^ H 2 i i Particle diameter, dp( mm) Figure 20 - E f f e c t of the p a r t i c l e size on the e x i t gas composition predicted using the one-dimensional model. Assume no no reaction between spout and annulus gas. Ov e r a l l volumetric gas flow i s kept constant. - 92 -As shown in Table 9 particle diameter is an important variable . which wi l l affect most of the hydrodynamic properties of a spouted bed. In general, under the same gas spouting flowrate a large particle diameter w i l l give a high gas velocity in the annulus and hence a much lower velocity in the spout than a smaller particle. Thus an improved reaction in the annulus is normally obtained from small size particles while large particles result in an increase in the extent of conversion in the spout. Combining these two streams gives the effect of the particle size on the overall gas composition as shown in Figure 20. The carbon monoxide increases steadily to show a maximum value at the particle diameter 2.5 mm and decreases slightly thereafter. No such pattern i s observed for a l l the other gas components. Prediction using the streamtube model shows similar variation in the overall gas composition as a function of particle size. - 93 -7. EXPERIMENTAL RESULTS AND DISCUSSION Gas composition profile measurements were attempted in over eighteen runs of the gasifier. Western Canadian coals were used: sub-bituminous Forestburg coal and bituminous Balmer coal. Both the proximate and ultimate analysis on these two coals are given in Table 10. About ten of the attempted runs were successful in that complete data were collected for the given modes of operation and operating conditions. The coal particle size used throughout the present study is -3.36 + 1.19 mm. The present work is part of an ongoing research programme on coal gasification in spouted beds which has been in progress over the past five years. A l l the profile measurements were made during runs performed by G. Cheng and S. Low who are looking into various aspects of gasification. Thus to avoid any confusion on the data compiled the run numbers of Cheng and Low have been adopted in this work. In presenting the results a l l gas compositions are expressed on a moisture-free basis. The axial gas concentration profiles are presented for three different radial positions in the bed referred to here as positions A, B and C. The exact locations of these three different positions in the bed are given in Figure 7 — p o s i t i o n A is nearest the wall, and position C i s on the centre-line of the reactor. 7.1 General Considerations 7.1.1 Sampling Rate Throughout the present work the intention was to sample the gas - 94 -Table 10 Properties of coals tested Ultimate Analysis (% wt MAF) Forestburg Balmer C 74.87 88.71 H 4.44 4.86 S 0.64 0.21 N 1.59 1.14 0 18.46 5.07 Proximate Analysis % Moisture 23.71 1.03 % Volatiles 29.44 20.93 % Fixed Carbon 33.25 66.67 % Ash 13.60 11.37 Free Swelling Index 1 1.5 - 95 -isokinetically. Due to the variations in gas composition and temperature from point to point in the bed and from one run to another, an assumed gaseous environment and bed temperature have been chosen in order to estimate the gas sampling'rate (Appendix A). Therefore some variation between the estimated sampling velocity and the actual gas velocity in the gasifer at the point of suction i s expected to exist. However from the gas analyses given in Table 11 which were collected from point 2A (Figure 7) in Run 89 at three different flowrates, the effect of sampling rate on the gas composition seems to be very small even at four times the estimated isokinetic rate. In the worst case the present estimated rate can only be a few percent in excess of the actual isokinetic value. Thus i t can be concluded that the procedure adopted in estimating the isokinetic sampling rates from different positions in the bed can be used without introducing any significant error in the results obtained. 7 .1 .2 Drying No steam was added into the system when gasifying Forestburg coal since this coal as fed contained 24% moisture. Thus a l l of the steam available for the reaction must have been released from the fresh coal particles as they dried. According to the measurements made by Lim (3) using a particle size and column diameter relatively close to the one employed in the present study, the particle velocity in the spout ranges from 3 m/s at the bottom to 0.8 m/s at the top of a 0.9m bed. Based on these values and for a bed of 1 m high, the average residence time of - 96 -Table 11 Effect of sampling rate on gas composition u /*u s iso Gas Component 1 3 4 Gas Composition, Vol °A : (dry) H 30.307 31.131 31.379 CO 2 8.872 8.103 8.242 0 2 0.109 0.09 0.167 N 2 28.44 29.086 28.854 CHi* 3.428 3.347 3.110 CO 28.844 28.235 28.249 *Estimated as in Appendix A. - 97 -a freshly fed coal particle in the spout w i l l be in the order of one second. To estimate the amount of water released within this time due to drying, the equation proposed by Mcintosh (78) can be used: t = time, s h = heat transfer coefficient, W/m '°K dp — particle diameter, m k — thermal conductivity of coal, W/°K.m L = latent heat of vaporization of water, J/kg MC = coal moisture content at time t (fraction, dry basis) MCQ = MC at t = 0 T ^ = adiabatic wet bulb temperature, °K T = gas temperature, °K § P w c = density of raw coal, kg/m The heat transfer coefficient in the spout can be estimated from the following equation (1) L P w c [ M C o / ( 1 + M C o ) ] dp  h ( T g - TWb) 2 f l n m w i MC , , Bi ,MC N x [ T (1 - B l ) ( l - — ) + — ( ! - ( — ) o o )] (7.1) where Bi = Biot number, Nu = 2 Pr 1/3 Re 0.55 (7.2) [1 - (1 - e) T73 - 98 -where e in the spout can be taken as 0.85(3), and the Nusselt and Reynolds numbers are based on particle diameter. For the present case using coal of average particle size of 2.1 mm and an average temperature of 800°C in the spout, the average heat transfer coefficient in this region has been estimated to be 736 W/m2«°K. Using an estimated value of 0.069 W/m«°K for the thermal conductivity of coal (79), calculations suggest that about 30% of the water remains inside the coal particles as i t leaves the spout. Thus the assumption of complete drying within the spout may not be j u s t i f i e d for relatively large coal particles or when a shallow bed is used. 7.1.3 Pyrolysis The rate of pyrolysis which i s known to be much slower than drying depends on a number of factors including the particle temperature. On the assumption of no radial temperature gradient within the particle, the temperature Tp at any time t attained by a particle placed in a gaseous environment of temperature T^ can be deduced from the following unsteady state equation: T - T -hA t ( T / M . . I P I ^ I (7.3) b po p where Tp Q is the i n i t i a l particle temperature. Under the same operating conditions discussed in section 7.1.2 the same coal particle would leave the spout with an average particle temperature of 450°C taking Tp Q as 25°C. Applying equations 2.1 and 2.2 to describe the rate of pyrolysis and taking the values of KQ' and E' as 0.019s - 1 and - 99 -19.72 kJ/mol respectively for Forestburg coal (79) i t would appear that only about 0.1% of the volatiles in the fresh feed coal are released in the spout. As the particles w i l l only spend a very brief period in the fountain (« 0.1 s) most of the pyrolysis w i l l take place in the annulus where the particles residence time w i l l be much longer. According to Al-Jarallah (79) there is a substantial radial temperature gradient within a 2.3 mm coal particle during the heating up period. In this case devolatilisation in the spout w i l l be much less than what has been estimated. 7.2 Forestburg Coal-Air 7.2.1 Axial Gas Composition Profiles A. Gas Composition Along Position B Figure 21 shows the axial gas composition profiles taken along position B which i s roughly 9 cm from the reactor wall. As shown the oxygen concentration drops almost instantaneously to essentially zero as soon as i t enters the gasifier due to the high rate of char combustion. The decrease in the oxygen concentration is accompanied by a sharp increase to a maximum value in the- carbon dioxide concentration indicating that carbon dioxide forms the major product of combustion. Whether there is any carbon monoxide produced in addition to carbon dioxide as a result of char combustion could not be confirmed in this study. The drop in carbon dioxide composition in the conical region which is in line with the model prediction as shown in Figure 8 can be explained by the C-CO2 reactions. In addition, the effect of gas - 100 -28 • H 2 • CO A C0 2 24 O CH 4 • 0 2 20r-H Balance N 2 Figure 21 - Axial gas composition profiles along position B for Forestburg coal-air system (Run 9 3 ) . - 101 -circulation in this region (8) may have also contributed to the drop in concentration. However, contrary to the theoretical prediction which shows a continuous decrease in the carbon dioxide composition throughout the bed, the experimental profile rises again from a value of 11.4% at a height of 50 cm. Beyond this level the composition remains relatively constant. The existence of such a profile towards the top of the bed i s thought to be due to the competition between the amount released by pyrolysis and that consumed by the C-CO2 reaction. As discussed earl i e r , the fresh feed coal that enters the annulus from the top is s t i l l in the i n i t i a l stages of pyrolysis. Thus there are both pyrolysis and gasification reactions in the upper section of the annulus. Recent work on pyrolysis (79) showed that a high yield of carbon dioxide (15% w/w MAF coal) could be obtained by pyrolysing Forestburg coal at 600°C. Both the carbon monoxide and hydrogen show similar profiles with the gas composition increasing along the bed height. The existence of points of inflection in the composition curves at the bed height of 25 cm which i s close to the cone/cylinder intersection (at z = 20 cm) may have been the effect of gas recirculation in the conical region. Relatively similar profiles for these two components have also been predicted for this region solely from a theoretical approach as is shown in Figure 8. In this case maximum and minimum peaks were shown which were purely due to the column geometry for this lower,section of the bed as has already been discussed in the previous chapter. Presumably the observed profiles occur not only because of gas recirculation but also are to some extent due to the conical bottom employed. The gasification - 102 -reactions must have been responsible for the increase in both carbon monoxide and hydrogen composition towards the top of the annulus. However, based on a simulated model using the operating conditions employed in this run as the input, gasification alone would have not resulted in the relatively equal proportion of hydrogen and carbon monoxide measured throughout the bed. Possibly in addition to gasification there are other sources of hydrogen and carbon monoxide which may be significant. Pyrolysis could be one of these sources. The H2/CO ratio released during pyrolysis depends very much on temperature with more carbon monoxide being released at low temperature. As much as 5% (w/w MAF Coal) of carbon monoxide could be obtained at 600°C while the corresponding figure for hydrogen is 1% (79). Due to the large amount of volatiles present in the Forestburg coal, reforming reactions of these volatiles with steam probably occurs. Taking carbon, hydrogen and oxygen as the only components present in the v o l a t i l e s , the following basic reaction can be assumed: CmHn 0 + ( m - H 2 ° * m C 0 + ( f + m - 1) H 2 (7.4) The values of m and n for Forestburg coal can be estimated from the data reported by Al-Jarallah (79). A simple calculation performed to estimate the amount of carbon monoxide and hydrogen released separately by reforming reactions pyrolysis and gasification showed that a l l of these reactions are about equally important as sources of carbon monoxide and hydrogen in the bed. Particle segregation may also play an - 103 -Table 12 Gas yield due to pyrolysis at 600°C for Forestburg coal and Balmer coal (79) Gas Yield, % (w/w) MAF coal Coal H 2 CO C0 2 CHL Forestburg 1.2 5.0 15.0 3.7 Balmer 1.0 0 0 5.3 - 104 -important role especially in the upper section of bed, where improved gas-solid reactions are achieved from smaller size particles. The coal as fed contains particles from 1.19 mm to 3.36 mm. Furthermore the particles in the bed wi l l keep reducing in size as the reactions proceed until they are entrained away. In such a system where particles of different sizes exist, the smaller sizes tend to congregate near the top of the bed (80). Low methane composition throughout the bed has been observed increasing from zero at the inlet to only about 3.5% at z = 65 cm. This low value is expected as a l l of the methane in the bed comes from pyrolysis rather than the C-H2 reaction which is insignificant under the present condition. B. Gas Composition Along Position A Figure 22 shows the gas composition profiles along position A which i s close to the column wall. There i s no significant difference with regard to both the oxygen and methane composition profiles to those previously observed along position B. Both the hydrogen and carbon monoxide show similar concentration profiles, increasing smoothly from zero at the inlet to about 23% at z = 65 cm. As discussed earlier the high concentrations at the upper section of the bed for these two components may be due to pyrolysis and reforming reaction in addition to the normal gas-solid reactions. However, in contrast to the profiles along position B, no point of inf l e c t i o n along the concentration curves i s observed in the conical - 105 -24 -H Balance N2 • H 2 • CO A CO2 O C H 4 A 0 2 0.8 1.0 z (m ) Figure 22 - A x i a l gas composition p r o f i l e s along p o s i t i o n A for the Forestburg c o a l - a i r system (Run 93). - 106 - \ region. A similar profile for carbon monoxide concentration has been predicted from the theoretical approach using the streamtube model as shown in Figure 12. Thus i t is possible that the actual gas flow pattern along this position is similar to that assumed in the stream tube model, i.e. plug flow of gas from inlet o r i f i c e to the top of bed. For the case of position B which is close to the annulus centre the plug flow pattern in the conical region could not exist due to gas recirculation. Presumably this effect is less experienced along this lower section of position A due to i t s close location to the wall, thus allowing the plug flow pattern of gas to be maintained. In general the carbon dioxide composition profile shows a pattern similar to that observed along position B. As argued in the previous case, the high carbon dioxide concentration at the top of the bed may be due to the additional amount released from pyrolysis. The sudden jump in the concentration between the bed level z = 25 cm and z = 35 cm is partly due to the effect of char recycling which w i l l be discussed in the next section. C. Gas Composition along position C Figure 23 shows the gas composition profiles taken along the spout. Both the carbon monoxide and hydrogen show similar profiles with very low concentrations at the lower section of the bed and then increase steadily to about 13% at the bed height z = 65 cm. The high concentrations observed in the upper section of the column contradict - 107 -H 2 8 — • H 2 • CO A CO2 24 O C.H4 A 0 2 z (m ) Figure 23 - Axial gas composition profiles along position C for the Forestburg coal-air system (Run 93). - 108 -the model predictions as shown in Figure 10. According to the model, both of these components should not be present anywhere in the spout. The high compositions measured may have been contributed by the vo l a t i l e s . However from what has been discussed above the fresh feed coal in the spout is only in the very i n i t i a l stage of pyrolysis and i t is very unlikely that they w i l l release any susbstantial amount of v o l a t i l e s . Another source of volatiles in the upper section of the spout is the particles which emerge from the annulus which are s t i l l undergoing pyrolysis. These particles which cross-flow into the spout in the upper section of the bed normally spend very l i t t l e time in the inner upper part of the annulus*. Due to segregation these particles are usually large in size, in which case pyrolysis can last for several seconds. As in the annulus, the methane composition remains relatively low throughout the bed. A value of about 1.8% has been measured at the bed level z = 65 cm. It is believed that the presence of methane in this region as well as in the annulus is solely due to pyrolysis. Judging from i t s profile in Figure 23 and Figures 21 and 22 for the annulus region, i t is evident that there is s t i l l coal pyrolysis taking place even in the upper section of the bed. The oxygen concentration curves shows both a minimum and maximum peak within the conical region before f i n a l l y reducing to zero. The total disappearance of oxygen may have not been entirely due to char combustion as shown by the model prediction in Figure 10. Possibly i t is the combination of char, volatiles and gaseous combustion that - 109 -consumes a l l of the available oxygen in the region. Interestingly the carbon dioxide profile traces a curve which turns out to be the mirror image of the oxygen profile. Its high concentration in the upper section of the bed is mainly due to the product of combustion and may be to some extent contributions from pyrolysis and reforming reactions. The fluctuations in both the carbon dioxide and oxygen composition profiles in the conical region are due to combustion and gas recirculation. Throughout this work i t has been assumed that the spout in the system is stable in the vertical position and i t s centre coincides with the centre of the column. A l l the samples taken along the column axis are assumed to represent those in the spout. Piccinini et al (5) however observed that the spout in their system tended to wander, resulting in poor reproducibility of results for the region. Unfortunately, no such observation can be made here. Comparing the measured composition profiles obtained along this region for a l l the runs, relatively similar shapes have been maintained by a l l the gas components. From this observation i t can be concluded that those assumptions made earlier are valid. The gas composition profiles in the conical region as clearly shown in Figures 21-23 have been repeatedly observed in a l l the runs. Thus experimental errors should be ruled out with regard to the measured experimental points at the different bed levels in the conical region. - n o -7.2.2 Radial Gas Composition Profiles Plotted in Figure 24 are the radial gas composition profiles in the annulus for carbon dioxide and carbon monoxide taken at different bed levels for Run 93. As shown there is a variation in the radial component gas composition within the annulus i t s e l f , which becomes more significant as one goes deeper into the bed. In terms of the models discussed, the existence of such a radial variation can only be explained by assuming the v a l i d i t y of steamtube model to describe the gas flow pattern in the annulus. 7.2.3 Effect of Operating Condition 7.2.3.1 Char Recycle In order to study the effect of the recycle of char from the external cyclone on the gas composition profiles experimental runs using both modes of operation, i . e . with and without char recycling, were performed. Char re-enters the bed at a single point in the annulus and i s conveyed with a small flow of a i r . Plotted in Figure 25 are the measured profiles along position A obtained from Run 90 (char recycling) and Run 89 (no char recycling). In this case a direct comparison can be made to distinguish any differences that might exist. It i s clearly shown from these plots that the gas composition curve traces by each component except carbon dioxide show similar shapes. As far as these components are concerned, no differences can be detected to discriminate these two modes of operation. The differences in the actual composition values are due to the slight variation in the operating variables - I l l -Dimensionless radius ( r/R c ) Figure 24 - Radial gas composition profiles for Forestburg coal-air system (Run 93): a) carbon dioxide, b) carbon monoxide. - 112 -z (m) Figure 25 - Effect of char recycling on composition profiles for the Forestburg coal-air system: a) Run 90 (recycling), b) Run 89 (no recycling) . - 113 -employed, as i t was not possible to maintain identical conditions in both cases. The only distinguishable trend which can differentiate the two systems is shown by the carbon dioxide composition profile. The sudden increase in the carbon dioxide concentration between the bed level z = 25 cm and z = 35 cm is partly due to the contribution from the char combustion with the incoming air which is used to transport the recycling char. For this system the recycling inlet to the reactor is located close to the bed position z = 35 cm. This sudden increase in carbon dioxide between the two bed levels has also been observed in a l l runs operated with char recycling. Only a gradual increase in the carbon dioxide composition between these two bed levels is observed in runs where no char recycling are employed as is shown in Figure 25b for Run 89. There are, however, no differences shown on the profiles along position B and C. This could be due to the location in the bed from which the samples were taken. In this case they are further away from the column wall and the char recycling i n l e t . The decline in the component gas compositions beyond the bed height H in Run 89 is considered to be due to gas mixing in the fountain. The same factor i s also responsible for the decline observed in Run 90. However, judging from the position at which the drop in gas compositions took place, higher bed height should have actually been employed in Run 90. The indicated bed height in Figure 25 as well as in a l l other figures was actually taken at the end of the operation. In a number of occasions, a slight reduction in the bed height during the course of the experiment was experienced. - 114 -7.2.3.2 Bed Height Plotted in Figure 26 are the axial gas composition profiles along position B taken for Runs 91 and 93, which were operated under different total bed heights as indicated. The same general trends in the gas composition profiles are s t i l l being maintained in both cases except for carbon dioxide where a minor difference exists. In this case the drop in the composition in the upper section of the conical region as shown in Run 93 i s no longer observed when operated at a lower bed height as in Run 91. Instead the carbon dioxide composition increases continuously throughout the bed. As discussed earlier, the effect of pyrolysis can be significant depending on the location in the bed. Since some of the particles entering the annulus are s t i l l undergoing pyrolysis, the top of the bed can be affected most by this process. Thus the increase in the carbon dioxide composition throughout the conical region in run 91, which is in this case located close to the bed top, i s probably due to the contribution from pyrolysis, rather than experiemental error since the same trends are also observed in most other runs operated under relatively deep and shallow beds respectively. As shown in Figure 26b the increase in the gas composition in the fountain is mainly due to gas mixing beyond the bed as evident by the corresponding drop in the profiles along position A. In addition there is also some contribution from pyrolysis. The extent of gas-solid reaction in the fountain is not known and i t s contribution to the change in gas concentration in the region cannot be assessed. - 115 -• H 2 • CO A C0 2 O CH 4 A 0 2 O > C o O CL E o o to o z (m) Figure 26 - Effect of the total bed height on composition profiles along position B for Forestburg coal-air system: a) Run 93, b) Run 91. - 1 1 6 -The gas composition profiles along position A are similar to those observed in Run 90 as illustrated in Figure 24a. A comparison of Figure 24a and Figure 21 shows no major difference in the concentration curves within the bed between these two cases. The same conclusion can also be made with regard to the profiles along position C. As in the previous case, this comparative study is based on the results obtained from two systems operated under similar rather than identical operating conditions. Thus, as expected there exists some variation in the actual gas composition at any given position. 7.2.3.3 Average Bed Temperature The effect of average bed temperature can be illustrated by comparing Figure 25a and Figure 22 to give the composition profiles along position A for Run 91 and 93 which were operated at average bed temperature of 800°C and 909°C respectively. As concluded earlier these profiles as well as those along positions B and C show no major differences. Within this temperature range the C-H2O and C-CO2 reactions are s t i l l low and the change in the carbon monoxide, hydrogen and carbon dioxide concentrations due to these reactions are s t i l l insignificant to be noticed. Relatively similar gas heating values have been produced by gasifying Forestburg coal between temperatures 7 90 and 930°C. (13) 7.3 Forestburg Coal - Oxygen In this study the effect of using oxygen rather than air on the gas profiles in the gasifier was investigated. There was however a small - 117 -flow of air added to the system through the feed line to transport the fresh feed coal into the gasifier. To control the gasification temperature a large amount of steam has to be added to the system to give H20/02 ratio in the order of 3.45 (w/w). Only one set of readings on the gas composition profiles was successfully obtained under this operating condition. Hence the effect of other operating variables could not be examined. 7.3.1 Axial Gas Composition Profiles A. Gas Composition along position B Plotted in Figure 27 are the gas composition profiles along position B. Comparison with Figure 21 where air was used instead shows that there are major differences in the gas composition pattern between these two systems. Both the carbon dioxide and hydrogen profiles show very l i t t l e change throughout the bed. The concentrations of CO, C0 2 and H2 increase sharply within a few centimeters from the inlet o r i f i c e in the lower section of the conical region and remain relatively constant thereafter. The slight increase followed by a continuous decrease in the hydrogen composition in the fountain is due to gas mixing in the region as evident by the shape of the corresponding hydrogen composition profiles along positions A and C (Figures 28 and 29 respectively). The same factor is also responsible for the observed carbon monoxide profiles in the fountain. A large ratio of H2/CO i s consistently produced throughout the bed, which is due to a high - 1 1 8 -H > 0 0.2 0.4 0.6. 0.8 1.0 z (m) Figure 27 - Axial gas composition profiles for Forestburg coal-oxygen system along position B (Run 96). - 119 -H 20/0 2 ratio in the feed. As in the previous system the sharp reduction in the oxygen concentration is accompanied by a rapid increase to a maximum value in the carbon dioxide concentration followed by a continuous decrease throughout the bed. The C/CO2 reaction is partly responsible for this decrease. Due to mixing the gas concentration rises again in the fountain as shown. No methane was found in the bed. It may have been burned away as soon as i t was released. A sli g h t l y higher proportion of hydrogen as compared to carbon monoxide has been predicted using the char gasification model. This i s not in a very good agreement with the experimental data. An estimate of the amount of carbon monoxide and hydrogen released individually from the volatiles reforming, pyrolysis and gasification indicated that the effects of pyrolysis may not be very significant when compared to gasification and reforming reactions. From this argument, i t is possible that reforming reaction may be an important potential source of carbon monoxide and hydrogen in addition to gasification reactions. Bo Gas Composition along positions A and C Figure 28 shows the gas composition profiles along postion A. In general these profiles are similar to those along position B. The trends shown in the bed can be explained in the same manner as discussed above while those in the fountain are possibly due to gas mixing. Figure 29 shows the gas composition profiles along position C at the reactor centre line . As in the case of the Forestburg coal-air - 120 -< H 0 0.2 0.4 0.6 0.8 1.0 z (m) Figure 28 - Axial gas composition profiles for Forestburg coal-oxygen system along position A (Run 96). - 121 -0 0.2 0.4 0.6 0.8 1.0 z ( m ) Figure 29 - Axial gas composition profiles for Forestburg coal-oxygen system along position C (Run 96). - 122 -system the profiles along this position are not in accordance with theoretical predictions for the spout where only oxygen and carbon dioxide are supposed to be present. Comparing this plot and those in Figures 27 and 28 shows that their shapes are relatively similar. Based on this comparison i t is l i k e l y that these profiles do not represent those in the spout, in which case the spOut might have wandered off the column axis under this operating condition. However, no confirmation can be made at this stage as there are no other data available under this operating condition. 7.4 Balmer Coal — Oxygen Four different runs were attempted to gasify Balmer Coal using oxygen and steam. However most of the runs lasted for about two hours except Run 103 where a total running time of three hours was achieved. Longer running times were prevented by a pressure build-up in the gasi f i e r , in which case the system had to be shut down immediately. Hence only one set of readings taken from Run 103 w i l l be discussed in this section. Even for this run, complete results could not be presented as there was no time to collect any samples along position C (spout). Due to some modifications carried out on the gasifier samples from the conical region could not be collected. 7.4.1' Axial Gas Composition Profiles Plotted in Figure 30 are the gas composition profiles observed along positions A and B respectively. Comparing these two plots with Figure 30 - Axial gas composition profiles for Balmer coal-oxygen system in Run 103: a) Along position A, b) Along position B. - 124 -those in Figures 27 and 28, i.e. the composition profiles obtained from the Forestburg coal-oxygen system clearly shows that the composition curves in the bed traced by a l l the components are relatively similar. The difference in the actual composition values are due to the differences in the coals used and to different operating conditions used, mainly the average bed temperature. For this run the average bed temperature was 913°C compared to 830°C in Run 96. As in a l l the previous cases, the sharp decrease in the oxygen composition is expected which is due to combustion, while a l l the methane present in the bed which shows a maximum composition of about 2% comes from pyrolysis. The steady rise in the carbon monoxide and hydrogen compositions throughout the bed along both positions A and B are due to gasification reactions and to some extent the volatiles reforming. As is evident by the methane composition profile, there is s t i l l pyrolysis taking place in the lower section of the bed. An estimate on the amount of hydrogen released due to pyrolysis showed that i t s contribution to the observed hydrogen profile was not significant, and certainly was insignificant i n the case of carbon monoxide. This coal as fed contains about 21% volatiles which i s about half of that in Forestburg coal. Recent attempt to pyrolyse this coal showed that no carbon monoxide and carbon dioxide were released mainly due to i t s low oxygen content. In addition about 5.3% (MAF Coal) of methane were obtained at 600°C as compared to 1% for hydrogen (79). From this observation, i t can be argued that the role of pyrolysis in determining the gas composition profiles in the bed - 125 -may not be very important for this system. The continuous decrease in the carbon dioxide composition along both positions was partly due to OC0 2 reaction. As mentioned above there was no contribution from pyrolysis. As in the case of the Forestburg coal-oxygen system the high H2/CO ratio observed throughout the bed is due to the high H2O/O2 ratio in the feed. In this case a ratio of 2.87 (w/w) has been used. 7.5 P a r t i a l l y Devolatilised Forestburg Char-Air A run was performed (Run 106) using partially devolatilised Forestburg char as the feed in order to examine the differences between coal and char as feed material. For this purpose the same operating conditions as in Run 93 were chosen so that direct comparison of the measured composition profiles between these two cases can be made. However there were s t i l l some variations in the actual conditions employed. Air/fixed carbon and steam/fixed carbon ratios of 4.6 and 0.723 respectively have been used in this run. The corresponding figures for Run 93 are 5.4 and 0.713. Thus an average bed temperature of 900°C has been achieved in this run compared to 800°C for Run 93. The char sample was prepared in different batches by heating Forestburg coal over a period of 2 - 3 hours at a temperature of 400 -500°C. A spouted bed combustor equipped with an external gas burner was used for this purpose. The combustion products of propane and air from an external burner was used as the heating medium. To avoid any - 126 -combustion of coal, the propane/air ratio entering the burner was carefully controlled so that no oxygen was present in the product stream leaving the burner. The properties of this sample are given in Table 13. As in the case of the Balmer-oxygen system, gas samples from the conical region could not be collected. 7.5.1 Axial Gas Composition Profiles Plotted in Figures 31 and 32 are the gas composition profiles along positions A, B and C. The penetration of oxygen into the bed i s much deeper in this case as shown by the axial oxygen composition profiles. This observation strongly suggests that the rate of char combustion in this system is lower. Thus the drop in both hydrogen and carbon monoxide compositions along the bed is due to gaseous combustion. According to Johnson (62) exposing char samples to a high temperature prior to gasification reduces i t s reactivity. However this finding could not be confirmed in this study. A higher proportion of carbon monoxide as compared to H 2 was measured throughout the bed. . This is in contrast to the profiles observed in Run 93 as shown in Figures 21 - 23. This is an important difference between these two systems as the effects of operating conditions should be ruled out. However more experimental data are required to confirm these profiles. 7.6 Average Temperature Profiles During the course of the experiments three different sets of - 127 -Table 13 Properties of Forestburg coal and the partially devolatilised Forestburg char Ultimate Analysis (% wt MAF) Forestburg Partially Devolatilised Forestburg Char C H S N 0 74.87 4.44 0.64 1.59 18.46 84.71 2.55 0.51 1.80 10.43 Proximate Analysis % Moisture % Volatiles % Fixed Carbon % Ash 23.71 29.44 33.25 13.60 1.64 15.64 65.60 17.12 - 128 -H T3 O > c o '35 o Q. £ o a tn o CD 32 28 24 2 0 I 6 I 2 8 Balance N2 • H 2 • CO A C 0 2 O C H 4 A 0 2 0.4 0.6 z (m) 0.8 1.0 Figure 31 - Axial gas composition profiles along position A for partially devolatilised Forestburg char-air system (Run 106). 24 20 - 129 -"o > c o o CL E o o <fi o O 8 0 16 I 2| 8 0 0 • H 2 • CO C 0 2 O C H 4 • 0 2 Balance N 2 Balance N 2 0.2 0.4 0.6 z (m ) OB LO Figure 32 - A x i a l gas composition p r o f i l e s f o r p a r t i a l l y d e v o l a t i l i s e d Forestburg c h a r - a i r (Run 106): a) Along p o s i t i o n B, b) Along p o s i t i o n C. - 130 -instantaneous temperature readings at a l l of the points involved were recorded. These readings, which were taken at the beginning, in the middle and at the end of the experiment were then averaged to give the average axial temperature profiles along positions A, B, and C. 7.6.1 Forestburg Coal-Air Figure 33 shows the average temperature profiles along positions A, B, and C taken for Run 93. There exists a large variation in the gas temperature along the spout referred to here as position C. The rapid rise from 700°C at the bed level z = 25.cm to about 800°C at the bed level z = 35 cm is due to combustion as evident by the rapid drop in the oxygen concentration within the same bed levels shown in Figure 23. It is the combination of combustion and low incoming gas and solid temperatures that results in a very large temperature difference in the spout. The temperature profile along position B which is close to the annulus centre is similar to that observed in a moving bed gasi f i e r . In this case the combustion reaction i s responsible for the high temperature at the bottom while the drop in temperature towards the top is due to gasification reactions. The temperature profile along position A increases steadily to maximum value in the lower half of the column and reduces thereafter. This low temperature at the bottom is the result of the endothermic ' C-H2O and C-CO2 reactions. The slight increase in values observed may have been due to the inflow of heat from the neighbouring region which is partly shown by the simultaneous decrease in temperatures along position B. - 131 -0 0.2 0.4 0.6 0.8 1.0 z (m) Figure 33 - Axial temperature profiles for Forestburg coal-air system (Run 93) along positions: O » A; • , B; ^ , C . - 132 -In addition to the axial variation in gas temperatures there are also significant radial temperature gradients, not only between the spout and annulus, but also within the annulus i t s e l f , especially in the lower section of the bed. 7.6.2 Forestburg Coal-Oxygen The average temperature profiles along positions A, B and C taken for Run 94 are plotted in Figure 34. As in the previous case the rapid increase in temperture along the spout i s due to the heat released from combustion. According to Figure 29, there is s t i l l a substantial amount of oxygen along the centre of the column in the fountain. Hence combustion must also be responsible for the sharp increase in the gas temperature observed along this region. Both the temperature profiles along positions A and B show similar trends increasing throughout the bed as well as in the fountain. The steady increase in temperature in the fountain is the result of gas mixing within the region to give a higher average gas temperature than that in the annulus. A very shallow bed was employed in this run. Thus i t is possible that the continuous increase in temperature along the bed in the annulus is due to the higher temperature of the incoming solids from the fountain rather than combustion. Judging from the oxygen concentration along positions A and B as shown typically in Figures 27 and 28, i t is unlikely that any combustion takes place in the upper section of the bed. - 133 -" " "0 0.2 0.4 0.6 0.8 1.0 z (m) Figure 34 - Axial temperature profiles for Forestburg coal-oxygen (Run 94) along positions: O > A; • , B; A , C - 134 -7.6.3 P a r t i a l l y D e v o l a t i l i s e d Forestburg Char-Air Figure 35 shows the average axial temperature profiles along positions A, B and C. Judging from the amount of oxygen in the bed as indicated in Figures 31 and 32, i t is, evident that the increase in temperature along the bed and in the fountain i s due to combustion. - 135 -Figure 35 - Axial temperature profiles for partially devolatilised Forestburg char-air (Run 106) along positions: O > A; • , B; A , C . - 136 -8. COMPARISON BETWEEN MODEL AND EXPERIMENT A model simulation of Run 93 has been performed and compared with the experimental data, based on the following assumptions in addition to those mentioned earlier in Chapter 6. 1. Effect of char recycling is neglected. 2. The gasification reactions and pyrolysis are treated separately. The gaseous products due to these two processes are combined in the fountain. The gas phase in this region i s assumed to be perfectly mixed. 3. Gas phase combustion i s assumed to continue in the fountains, as long as there is oxygen present. The gas components leaving the gasifier are in equilibrium with respect to the water gas shift reaction. 8.1 Estimation of Pyrolysis Product Distribution The composition of the product of pyrolysis (weight fraction) can be estimated from the following equations proposed by Rejan (81) CELu = 0.201 - 0.469 X + 0.241 X (8.1) H vm vm Ho - 0.157 - 0.868 X m + 1.388 X 2 (8.2) ^ vm vm COo = 0.135 - 0.900 X + 1.906 X 2 c vm vm (8.3) - 137 -CO = 0.428 - 2.653 X + 4.845 X 2 (8.4) vm vm HoO = 0.409 - 2.389 X + 4.554 X 2 (8.5) vm vm Tar = -0.325 + 7.279 X - 12.880 X vm vm (8.6) 8.2 Gaseous Combustion i n the Fountain Only the combustion of carbon monoxide and hydrogen given by Equations 6.4 and 6 . 5.respectively are considered to be s i g n i f i c a n t . These reactions are assumed complete as long as there i s oxygen e x i s t i n g . Methane oxidation can be neglected as i t s concentration i n the fountain, which s o l e l y comes from p y r o l y s i s i s generally low. The proportions of oxygen shared between these two reactions are estimated from the r a t i o of t h e i r rates of r e a c t i o n . According to Sundaresan and Amundson (58) the combustion rates of carbon monoxide (R^) and hydrogen (Rg) can be represented as follows: R A (8.7) R, B (8.8) Thus the r a t i o R ^ R B O R R A B i s § i v e n b y R AB (8.9) - 138 -From their.analysis which is applicable for reactions occurring in an environment of high oxygen concentration, the value of kg is very much bigger than k^. Under the present operating conditions, the oxygen concentration in the fountain is always low and i t is very unlikely that the proposed rate constants are s t i l l valid. Due to the lack of other data, an assumed value of 1 has been assigned to the ratio k^:kg. Thus Equation (8.9) reduces to CC0 H 2 8.3 Kinetic Parameters The kinetic parameters for both types of coal under study are not available. Thus the same parameters as given in Table 8 have been chosen for this purpose. In this case the comparison w i l l only focus on the trends predicted by the theoretical model and that obtained from experimental measurements rather than the actual values. 8.4 Axial Gas Composition Profiles 8.4.1 Profiles i n the Annulus Plotted in Figure 36 are the profiles for carbon dioxide, carbon monoxide and hydrogen predicted using the streamtube model and those measured along positions A and B. The predicted trends for both carbon monoxide and hydrogen are in line with the measured profiles where the gas composition increase continuously along the bed. The effect of pyrolysis could not be included in the model since there are no - 139 -CO • A H 2 CO2 4 — 6 -8 — (b) • H 2 • CO A C 0 2 i • C 0 2 A A CO 3 - Comparison between experimental and predicted axial profiles for Run 93 (Forestburg coal-air) using the streamtube model: a) Along position A, b) Along position B. - 140 -mathematical expressions to describe the particle movements in the bed. In addition the possibility of volatiles reforming in the system was not considered. As discussed earlier the inclusion of these two factors may be essential in order to accurately.predict the hydrogen and carbon monoxide composition profiles in the bed. The predicted carbon dioxide profiles are not in agreement with that experimentally measured. Pyrolysis may be partly responsible for this difference. Unlike the streamtube model, the one-dimensional model can only predict the average gas composition profiles, which is assumed to be the profile along the centre of the annulus. Exact comparisons could not be made here as no experimental data along the annulus centre were taken. However as already been discussed in Chapter 6, the profiles along the centre of the annulus predicted by both models (Figure 12) are relatively close giving a higher proportion of carbon monoxide than hydrogen throughout the bed. The predicted and measured oxygen concentrations are in good agreement, and both drop sharply to zero very close to the bed i n l e t . This i s due to the very high combustion rate compared to the other reactions. The C-H2 reaction has been ignored in this model as this reaction has been assumed negligible under the present operating conditions. In addition to the contribution from pyrolysis and to some extent the volatiles reforming reactions, there are several other factors which may have resulted in the difference between the predicted and measured profi l e s . The rate equations which describe reactions 6.1 - 6.3 were assumed to follow a surface reaction kinetic model. Under the typical - 141 -gasification temperature and the relatively coarse particle size employed in the present study the gasification reactions have been found to take place simultaneously on the inside as well as on the outside surface of the particle (30). Thus these reactions may be more appropriately represented by the volumetric reaction model, in which the rate strongly depend on the fractional carbon conversion in the particle since the inside pore structure changes as the reactions proceed.. Unfortunately this reaction model could not be applied here because the particle movements could not be described. There are also some errors introduced in predicting the profiles within the conical region by ignoring the effect of gas recirculation in the region. In a l l the simulations performed, an isothermal gasifier has been assumed. Judging from Figure 33 such an assumption is j u s t i f i e d especially when dealing with the annulus since the measured temperature is relatively constant to within about 50°C. 8.4.2 P r o f i l e s i n the Spout The difference between the predicted and the measured profiles in the spout can be obtained by comparing Figures 10 and 23. There i s s t i l l a substantial amount of oxygen predicted in the spout which has not been observed experimentally. In addition large proportions of carbon monoxide and hydrogen have been measured in the upper section of the spout which are not in accordance with the predicted trends. Thus i t appears that the contributions from pyrolysis and volatiles reforming play a major role in determining the axial gas composition profiles in the spout. - 1 4 2 -8.5 Radial Gas Composition Profiles Figure 37 shows the comparison between the predicted and measured radial composition profiles of carbon dioxide and carbon monoxide at z = 65 cm. As shown experimentally, there is a radial variation in the gas composition within the annulus. Thus in this respect the streamtube model will be superior to the one-dimensional model since i t is able to predict radial composition profiles over the entire annulus, eventhough the predicted gradients were much too steep when compared to the actual measured values. It is l i k e l y that an improved prediction could be obtained by allowing some form of gas dispersion normal to the streamlines, which was not included in the present model. 8.6 Exit Gas Composition The predicted exit gas compositions are not in good agreement with the experimentally determined data as shown by the plot in Figure 38. In this case the one dimensional model has been used. The same distributions would have also been obtained from the streamtube model. The similarity between these two models in this aspect has been discussed previously in Chapter 6. The observed variations between the predicted and the experimental values are directly related to the model's i n a b i l i t y to closely predict the gas composition profiles within the bed. - 143 -T3 O > C o to o CL £ o o co o O Dimensionless radius (r/Rc) Figure 37 - Comparison between experimental and predicted radial carbon monoxide and carbon dioxide profiles for Run 93 (Forestburg coal-air): , one-dimensional modal; , streamtube model. - 1 4 4 -Figure 38 - Comparison between the experimental and predicted exit gas composition using the one-dimensional model for Run 93 (Forestburg coal-air). - 1 4 5 -9 CONCLUSION AND RECOMMENDATION 9.1 Conclusion The following conclusions can be drawn from this study: - There i s a radial gas composition profile within the annulus of a spouted bed coal gasifier. - The combustion reaction takes place within a very narrow region close to the gas inlet in the annulus and in the lower section of the bed in the spout. - Pyrolysis in a spouted bed coal gasifier is not instantaneous. This process and the volatiles reforming reactions should be accounted for in order to construct the gas composition profiles both in the annulus and in the spout for the Forestburg coal-air system. There i s a substantial amount of pyrolysis, especially in the upper section of the bed. - No significant effects of the total bed height, char recycling and the average operating temperature (within range investigated) are observed on the measured gas composition profiles in the bed for the Forestburg coal-air system. - The contribution from pyrolysis on the gas composition pattern in the bed for Forestburg coal-oxygen and Balmer coal-oxygen appears to be insignificant. However the effects of the volatiles reforming reactions as the source of carbon monoxide and hydrogen could be important. - 146 -- There is a large axial temperature variation in the spout to give the hottest spot in the upper section. For the Forestburg coal-air system operated under a relatively deep bed, the temperature profile along the centre of the annulus resembles those in a moving bed gasifier. - The H2/CO ratio along the bed of a spouted bed gasifier i s lower when partially devolatilised char instead of coal is used as the feed. - Both the one-dimensional model and the streamtube model give a very similar prediction in terms of the overall exit gas composition as well as the gas composition profiles along the centre of the annulus, but they could not be matched with the experimental" results. - The streamtube model is superior to the one-dimensional model as i t is able to predict radial gas composition profiles over the entire annulus which have been observed experimentally. - It i s necessary to include the particle movement within the bed in order to accurately predict the performance of a spouted bed coal gasifier. 9.2 Recommendation The following should be included for future work: - Further study the effect of char recycle, bed height and temperature on the gas composition profiles for the Forestburg coal-oxygen and Balmer coal-oxygen system. - 147 -Further confirm Chose p r o f i l e s measured along the spout. In th i s case i t i s necessary to construct the actual spout regio within the bed by taking r a d i a l pressure drop p r o f i l e s at d i f f e r e n t bed l e v e l s . Investigate whether the water gas s h i f t r eaction i s i n equi l i b r i u m throughout the bed. For this purpose the steam composition p r o f i l e s should also be determined. Perform more experiments on char g a s i f i c a t i o n . Improve the g a s i f i c a t i o n model. In t h i s case the p a r t i c l e movement in the bed should be included as well as some form o r a d i a l d i s p e r s i o n i n the annulus. - 148 -NOMENCLATURE 3 9 cross-sectional area of annulus, m 2 cross-sectional area of column, m surface area of a single particle, raz 2 cross-sectional area of spout, m concentration of component i in annulus gas, mol/m C at z = H, mol/m 3.X concentration of component i along streamline in steamtube model, mol/m3 C*azi at z = H, mol/m3 concentration of component i leaving reactor, mol/m gas concentration, mol/cm unless otherwise specified carbon concentration of char; C , same at zero so' conversion, mol/cm unless otherwise specified concentration of component i in spout gas, mol/m C . at z = H, mol/m3 s i specific heat capacity, J/kg:K 2 diffusion coefficient; D', same for oxygen, cm /s effective d i f f u s i v i t y in solid; D , same at zero J ' eo' 2 conversion, cm/s column diameter, m gas inlet diameter, m particle diameter, m unless otherwise specified spout diameter, m - 149 -E' a c t i v a t i o n energy fo r p y r o l y s i s , kj/mol E a c t i v a t i o n energy fo r char-gas r e a c t i o n , kJ/mol a parameter used in Equation 2.34 f-^ , f r e l a t i v e r e a c t i v i t y fac tor fo r char-H20 r eac t ion used for Johnson k i n e t i c s (Eqn. 2.22) G gas mass flow rate per uni t of column c r o s s - s e c t i o n , kg/m 2 .s Ga Ga l i l eo number g a c ce l e r a t i on of g r a v i t y (9.81 m/s2) H bed depth, unexpanded, m H m maximum spoutable bed depth, m h heat t r ans fe r c o e f f i c i e n t , W/m . °K K equ i l i b r i um constant fo r char-gas r eac t i on K w g equ i l i b r i um constant for water gas s h i f t r eac t ion k' , k^ p y ro l y s i s rate constant , s - 1 kl> k 2 , ^3 constants of Langmuir-Hinshelwood type rate expressions f o r char-C0 2 and char-H 20 r eac t i on k T ra te constant for char-H^O reac t i on in Johnson k i n e t i c (Eqn. 2.22) k ash f i lm d i f f u s i o n rate constant , g/cm .atm.s ash k c thermal conduc t i v i t y of c o a l , W/m.°K m k , . c c d i f f u s i o n a l r eac t ion rate constant for char-gas r e a c t i o n , d l f f g/cm 2 .atm.s k o v e r a l l rate constant for char-gas r eac t i on as def ined by o v e r a l l „ , -, , 2 Equation 6 .7 , g/cm .atm.s k , k sur face r eac t i on rate constant for char-gas r eac t i on s so , 2 _ g/cm 'atm*s - 150 -k , k volumetric rate constant for char-gas reaction V V° (cmVmoDn.s-1 L latent heat of vapourisation of water, J/kg MC coal moisture content, fraction, dry basis; MCQ, same at zero time Mw molecular weight of carbon, g/mol nip mass of a single particle, kg/m Nu Nusselt number n order of reaction with respect to gaseous reactant concentration or partial pressure P total pressure, atm Pl,P2, P3 Reactants partial pressure in Langumuir type rate equation, atm P gas partial pressure, atm © PL path length along streamline, m Pr Prandalt number Q Q inlet gas flowrate, kg/h Q(J-l) volumetric gas flow along streamline J, m /s R universal gas constant (8.314 J/mol.°K) R' universal gas constant (82.06 atm.cm3/mol.°K) R column radius, m c Re Reynolds number (Re) ^ Reynolds number at beginning of fluidization (Re)t£ Reynol'ds number at complete fluidization R. rate of j t h reaction given by equation 6.1-6.3, J kmol/m3 of particle volume.s - 151 -r c core radius at time t in the core shrinking model; r , same at t = 0, m unless otherwise specified r radial distance from the column axis, m unless otherwise specified Sc Schmidt number T temperature, °K T Q maximum temperature to which char has been exposed prior to gasification, °K T w b adiabatic wet bulb temperature, °K t time, s U superficial gas velocity in annulus, m/s 3. U ajj U"a at z=H, m/ s U _ superficial gas velocity along streamline, m/s U b f superficial gas velocity at beginning of fluidization m/s U m£ minimum fluidization velocity, m/s U m g minimum spouting velocity, m/s U r radial gas cross flow velocity at spout-annulus interface, m/s U g gas spouting velocity, m/s U g superficial gas velocity in spout, m/s U t£ superficial velocity at complete fluidization, m/s u g gas sampling velocity; u^ s o> same at isokinetic condition, m/s V volatiles loss from particle up to time t, fraction of original coal weight - 152 -X proximate vo l a t i l e matter content of coal, gm/gm of coal ^ (daf) x fraction of carbon conversion of char due to char-gas reaction Y r c / r Q z vertical distance from gas in l e t , m unless otherwise specified Greek Symbols a a parameter in Johnson kinetic, equation 2.22 a relative pore surface area function in volumetric reaction model e voidage of the ash layer annulus voidage e voidage along streamline e g spout voidage 3 solid bulk density, kg/m except for calculating Dg (Table 9) where i t should be expressed in g/cm 3 P c molar density of carbon in coal particle, mol/cm 3 p£ gas density, kg/m 3 p particle density, kg/m except for calculating H m (Table 9) where i t should be expressed in g/cm 3 P w c raw coal density, kg/m n effectiveness factor product distribution of char combustion defined by Equation 2.17 X shape factor of particle. X = 1 for spherical particle - 153 -REFERENCES 1. Mathur, K.B. and Epstein, N. "Spouted Beds" Academic Press, New York (1974). 2. Mathur, K.B. and Lim, C.J., Chem. Eng. Sci. (1974), 29, 789. 3. Lim, C.J. PhD Thesis Univ. British Columbia (1975). 4. Lim, C.J. and Mathur, K.B., AlChEJ (1976), 22, 674. 5. Pi c c i n i n i , N., Grace, J.R. and Mathur, K.B., Chem. Eng. Sci. (1979), 34, 1257. 6. Littman, H., Narayanan, P.V., Tomlins, A.H. and Friedman, M.L., A.I.Ch. E Symp. Ser. No. 205 (1981), 7J_, 174. 7. Rovero, G., Pi c c i n i n i , N., Grace J.R., Epstein N. and Brereton C.M.H., Chem. Eng. Sci. (1983), 38, 557. 8. Van Velzen D., Flamm, J.J. and Langenkamp H., Can. J. Chem. Eng. (1974), 52, 143. 9. Rovero, G., Brereton, C., Epstein, N., Grace, J.R., Casalegno, L. and P i c c i n i n i , N., Can. J. Chem. Eng. (1983), 3_> 289-296. 10. Verma, A., Chem. Tech. (1978), 8_, 372. 11. Lowry, H.H. (Ed). "Chemistry of Coal U t i l i s a t i o n . Suppl. Volume" John Wiley and Sons, New York (1963). 12. Probstein, R.F. and Hicks, R.E. "Synthetic Fuel" McGraw-Hill (1982). 13. Foong, S.K., Lim, C.J. and Watkinson, A.P., Can. J. Chem. Eng. (1980) , 58, 84. 14. Watkinson, A.P., Foong, S.K., Cheng, G. and Prakash, C.B., Advances in Coal U t i l i s a t i o n Technology IV. (1981). 15. Foong, S.K., Cheng, G. and Watkinson, A.P., Can. J. Chem. Eng. (1981) , 59, 625. 16. Watkinson, A.P., Cheng, G. and Prakash, C.B., Paper presented at the 32nd Canadian Chemical Engineering Conference, Vancouver, October (1982) . 17. Hendrickson, T.A. "Synthetic Fuels Data Handbook" Cameron Inc., Denver (1975). 18. Yasinky, J.B. "Coal Technology 78 - Vol I" Pre. of Int. Coal Ut i l i s a t i o n Conf. Houston, Texas Oct. (1978) p. 22. - 154 -19. Wen, C.Y. and Stanley Lee, L. (Ed) "Coal conversion Technology" Addison-Wesley Publishing Company (1979). 20. Chang, P.W., Durai-Swamy, K., Knell, E.W., Coal Technology, 6_, 20-27. 21. Edwards, J.H. and Smith, I.W., Fuel (1980), 59_, 674. 22. s Tyler, R.J., Fuel Fuel (1980), 59, 218. 23. Anthony, D.B., Howard, J.B., Holtel, J.C. and Meissner, H.P., Proc. of the 15th Symposium on Combustion (Int). Combustion Institute, Pittsburg PA (1974). 24. Wen, C.Y. and Chen, L.H., Paper presented at ACS National Meeting, Washington, D.C. Sept. (1979). 25. Anthony, D.B. and Howard J.B., J.A.I. Che. E. (1976), 2_2, 625. 26. Solomon, P.R., Report R76-952588-2, United Technologies Research Center East Hartford, Conn. (1977). 27. Howard, H.C. "Pysolytic Reactions in Coal" in Chemistry of Coal Utilis a t i o n p. 340-394. Suppl. Volume. H.H. Lowry (Ed) Wiley, N.Y. (1963). 28. Haggerty, J.F. and Pulsifer, A.H., Fuel, (1972) 51_, 304. 29. Jensen, G.A., Ind. Eng. Chem. Process. Des. Dev. (1975), 308. 30. Dutta S. and Wen, C.Y., Ind. Eng.Chem. Process. Des. Dev. (1977), 16, 20. 31. Gulbransen, E.A. and Andrew, K.F., Ind. Eng. Chem. (1952), 44, 1039. 32. Walker, P.L. Jr., and Ki n i , K.A., Fuel (1965), 44, 453. 33. Levenspiel, 0. "Chemical Reaction Engineering" Wiley, N.Y. (1972. 34. Essenhigh, R.H. , Froberg, R. and Howard, J.B., Ind. Eng. Chem. (1965), 57_, 33. 35. Laurendeau, N.M., Prog. Energy Combust. Sci. (1978), 4_, 221. 36. Wen, C.Y. and Chaung, T.Z., Ind. Eng. Chem. Process Des. Dev. (1979), JjJ., 684. 37. Heeyoung, Y., Wei, J. and Denn, M.M., A.I. Che. E. (1978), 24, 885. 38. Smith, I.W. and Taylor, R.J., Combustion Sci. Technology (1974, 9_ , 87. - 155 -39. Field, M.A., Combust. Flame (1969), _L3 , 237. 40. Caram, H.S. and Amundson, N.R. , Ind. Eng. Fundam. (1977), 16_, 171. 41. Walker P.L. Jr., Rusinko, F. Jr., and Austin, L.G.-, Advances in Catalyst Volume 11. Academic Pres. N.Y. (1959), p. 133. 42. Wen, C.Y. and Dutta, S. "Coal Conversion Technology, 4_, 40-51. 43. Philips, R., Vastola, F.J. and Walker, P.L., Carbon (1969), _7, 495. 44. Day, R.J. PhD Thesis. The Pennsylvania State Univ. (1949). 45. Lewis, W.K., G i l l i l a n d , E.R. and Paxton, R.R., Ind. Eng. Chem. (1954), 46, 1327. 46. Arthur, J.R., Trans. Faraday Soc. (1957), 4_7,, 164. 47. Mulcahy, M.F.R. and Smith, I.W., Rev. Pure Appl. Chem. (1969), 1_9, 81. 48. Field, M.A.,. G i l l , D.W., Morgan, B.B. and Hawksley, P.G.W. "Combustion of Pulverised Coal" BCURA, Leatherhead (1967). 49. Dutta, S. and Wen, C.Y., Ind. Eng. Chem. Process. Des. Dev. (1977), 1_6, 31. 50. Gray, D., Cogoli, J.G. and Essenhigh, R.H., Paper 6 in Coal gasification, Advances in Chemistry series, 131, American Chemical Society, Washington, D.C. (1976), 72-91. 51. Smith, I.W. and Tayler, R.J., Fuel (1972),_51_> 312-322. 52. Tu, CM., Davis, H. and Hottel, H.C., Ind. Eng. Chem. (1934), 26, 749. 53. Nagle, J. and Strickland-Constable, R.F., Proc. 5th. Conf. on Carbon (1962), 1_, 154. 54. Batchelder, H.R., Busche, R.M. and Armstrong, W.P., Ind. Eng. Chem. (1953), 45, 1856. 55. Long, F.J. and Syker, K.W., Proc. Roy. Soc. London Seri. A. (1948) , 193_, 357. 56. Reif, A.E., J. Phys. Chem. (1952), 56, 785. 57. Sundaresan, S. and Amundsen, N.R., Ind. Eng. Chem. Fundam. (1980), 19_, 344. 58. Sundaresan, S. and Amundsen, N.R., Ind. Eng. Chem. Fundam. (1980), 19_, 351. 59. Long, F.J. and Sykes, K.W., Proc. Roy. Soc. London Ser. A. (1948), 193, 377. - 156 -60. Riede, B.E. and Hannesian, D., Ind. Eng. Chem. Process. Des. Dev. (1975), 14, 70. 61. Kl e i , H.E., Sahagian, J. and Sundstrom, D.W., Ind. Eng. Chem. Process. Des. Dev. (1975), 14_, 470. 62. Johnson, J.L. "Kinetics of coal gasification." John Wiley (1979). 63. Parent, J.D. and Katz, S., IGT Research Bulletin 2 (Jan. 1948). 64. Johnstone, J.F., Chen, C.Y. and Scott, D.S., Ind. Eng. Chem. (1952), 44_, 1564. 65. Haslam, R.T., Ind. Eng. Chem. (1923), 1_5, 679. 66. Gwosdz, J., J. Soc. Chem. Ind. (London), (1919), 38, A125. 67. Parker, A.S. and Hottel, H.C., Ind. Eng. Chem. (1936), 2_8, 1334. 68. Purdy, M.J., Felder, R.M. and Fe r r e l l , J.K., Ind. Eng. Chem. Process Des. Dev. (1981), 20, 675. 69. Singh, CP.P. and Saraf, D.N., Ind. Eng. Chem. Process Des. Dev. (1977), 16_, 313. 70. Smith, I.W., Combustion and Flame (1971), 1_7, 421. 71. Mayers, M.A., J. Am. Chem. Soc. (1934), 56, 70. 72. Pursley, S.A., Matula, R.A. and Witzell, O.W., J. Phys. Chem. (1966), 70, 3768. 73. Mayers, M.A., J. Am. Chem. Soc. (1934), 56^ 1879. 74. Dotson, J.A., Koehler, W.A. and Holden, J.H., Ind. Eng. Chem. (1957), 49, 149. 75. Watkinson, A.P., Foong, J. and Cheng, G., Final Year Report (1978-1979) DSS Contract 01SU 23440-7-9073/11. 76. Yoshida, K. and Kunii, D., Journal of Chemical Eng. of Japan (1974) , 7_, 34-39. 77. Vaid, R.P. and Sen Gupta, P., Can. J. Chem. Eng. (1978), 56, 292. 78. Mcintosh, M.J., Fuel (1976), 5_5, 47-52. 79. Al-Jarallah, A., PhD Thesis Univ. British Columbia (1983). 80. Uemaki, 0., Yamada, R. and Kugo, M., Can. J. Chem. Eng. (1983), 61, 303-307. 81. Rejan, R., PhD Thesis West Virginia University (1978). - 157 -APPENDIX A Estimation of Isokinetic Sampling Flowrate The gaseous environment varies with respect to i t s temperature and composition from different points in the bed and from one run to the other. In order to estimate the sampling rate a gas temperature of 900°C and a gas mixture of 10% steam/90% air (v/v) has been assumed. These values were chosen based on the typical results obtained in some of the previous runs. Below is a sample calculation for point 6A. Height above gas i n l e t , z (figure 7) = 0.65 m Coal particle diameter = 0.0021 m •Column diameter = 0.305 m Gas inlet diameter = 0.0318 m Coal particle density =1.3 Mg/m Maximum spoutable bed height, ILj (Table 10) = 2.03 m Minimum fluidization velocity, IL^f (Table 10) = 0.89 m/s Superficial gas velocity, U a (Table 10) = 0.55 m/s ' I n t e r s t i t i a l gas velocity — = 1.29 m/s a Sampling probe cross-sectional area (1/4" dia. tube) = 3.17 x 10 - 5 m2 Isokinetic sampling volume at 900°C = 4.1 x 10~5 m3/.s Assuming that the gas leaving the water separator is saturated, the gas — 5 3 sampling volume at 21°C to be recorded by the rotameter is 1 x 10 m/s or 1.27 f t 3 / h r . APPENDIX B Operating Conditions and Experimental Results Table 14 Operating conditions I Run No. Coal Type H (cm) Wet Coal Feed kg/h Char Recycle kg/h Cyclone Catch kg/h 88 F 72 40.20 0 N/A 89 F 38 44.40 0 N/A 90 F 22 43.80 12.96 4.97 91 F 30 50.40 3.57 7.25 93 F 97 55.80 3.57 8.51 94 F 23 47.58 0 6.40 96 F 35 48.00 0 7.97 103 B 1.28 37.20 0 16.29 106 FC 41 33.42 0 N/A F = Forestburg B = Balmer FC = Partially devolitilised Forestburg Char N/A = not available Table 15 Operating conditions II Run No. Steam Flowrate Air Flowrate Pure 0 2 Flowrate Pure N 2 Flowrate kg/h kg/h kg/h kg/h 88 9.53 92.64 0 0 89 10.53 113.23 0 0 90 10.38 103,39 0 0 91 11.95 100.61 0 0 93 13.23 100.17 0 0 94 119.28 0 31.47 0 96 83.38 0 25.43 9.41 103 72.55 0 26.41 0 106 15.84 100.17 0 0 Table 16 Operating conditions III Run No. H 20/air H 20/0 2 H20/dry Coal Air/dry Coal 0 2/dry Coal (w/w) (w/w) (w/w) (w/w) (w/w) 88 0.103 - 0.31 3.02 89 0.093 - 0.31 3.34 90 0.100 - 0.31 3.09 91 0.119 - 0.31 2.62 93 0.132 - 0.31 2.35 94 - 3.79 3.29 - 0.87 96 - 3.27 2.28 - 0.70 103 - 2.87 1.98 - 0.72 106 0.158 - 0.48 3.05 Table 17 Experimental r e s u l t s : Run 88 Gas Composit ion, v o l . % (dry) Point No. Temp. Ave. Temp. :  °C °C H 2 C 0 2 0 2 N 2 CH4 CO 6A - - - - - - — -6B - - 11.232 15.436 0.180 63.444 1.673 8.045 6C - - 8.510 10.844 1.312 67.421 0.874 11.039 5A - - 31.439 18.190 0.344 26.014 6.080 17.934 5B - - 15.631 15.003 0.187 57.022 2.002 10.162 5C - - 10.567 14.215 0 65.054 1.227 8.937 4A - - 21.354 14.703 0.134 44.430 2.872 16.505 4B - - 15.696 14.037 0.199 55.739 2.037 12.292 4C - - 8.897 13.984 0.189 68.097 0.930 7.903 3A - - 22.292 8.318 0.253 49.396 2.248 17.493 3B - - 13.556 12.281 0.149 60.691 1.253 12.073 3C - - 5.179 13.369 0.237 75.353 0.418 5.445 2A - - 15.108 3.356 1.997 64.795 0.997 13.747 2B - - 8.434 7.974 0.279 74.099 0.415 8.799 2C - - 0.850 13.063 4.791 80.159 0 1.137 1A 1 R — — 8.709 3.837 0.164 75.183 0.197 11.910 1 Jj 1C — — 1.309 14.927 1.301 80.050 0 2.417 Table 18 Experimental results: Run 89 Gas Composition, vol. % (dry) Point No. Temp. Ave. Temp. °C °C H 2 C0 2 0 2 N 2 CH4 CO 6A — - 18.050 8.002 0.161 50.847 1.537 21.403 6B - - 13.253 10.709 0.322 58.703 1.346 15.667 6C - - 11.016 12.862 0.050 63.412 1.264 11.394 5A - - 21.672 7.845 0.138 45.862 1.997 22.487 5B - - 13.878 10.212 0.110 57.723 1.400 16.676 5C - - 8.421 13.222 0.127 67.451 0.879 9.900 4A - - 30.307 8.872 0.109 28.440 3.428 28.844 4B - - 14.653 9.585 0 58.336 1.374 16.052 4C - - 3.897 14.330 0.115 75.656 0.570 5.432 3A - - 32.047 7.652 0.070 27.242 4.432 28.551 3B - - 17.613 8.262 0.230 55.846 1.863 16.187 3C - - 2.941 12.798 • 3.712 77.130 0.279 3.141 2A - - 27.463 3.141 0.090 45.740 2.727 20.831 2B - - 12.264 5.960 0.772 68.251 1.039 11.714 2C - - 0.599 6.730 10.823 81.105 0 0.744 1A - - 11.967 1.547 0.207 67.527 0.424 18.328 IB - - - - - - - -1C - - 0.206 13.175 6.217 80.402 0 0 Table 19 Experimental results: Run 90 Gas Composition, vol. % (dry) Point No. Temp.* Ave. Temp. 3C °C H 2 C0 2 0 2 N 2 CEk CO 6A 899 - 16.491 9.528 0.088 50.479 1.317 21.648 6B 937 - 12.361 11.059 0.115 59.138 1.084 16.243 6C 973 - 8.531 12.374 0.162 67.831 0.748 10.354 5A 893 - 22.048 8.406 0.057 42.114 1.727 25.648 5B 933 - 10.883 12.687 0.967 60.394 0.999 14.070 5C 959 - 10.404 12.520 0.205 56.525 0.910 9.975 4A 761 - 22.420 8.974 0.142 51.859 3.511 13.094 4B 805 - 11.826 14.304 0.127 62.392 1.632 9.720 4C 840 - 5.728 12.817 1.458 73.426 0.713 5.859 3A 801 - 27.929 10.886 0.130 33.089 3.497 24.469 3B 841 . - 14.519 11.579 0.080 57.162 1.502 15.218 3C 884 - 3.334 10.484 4.869 76.088 0.295 4.931 2A 830 - 24.391 3.350 0.100 48.557 1.888 21.715 2B 859 - 6.242 10.017 0.109 74.135 0.310 9.179 2C 820 - 1.162 9.540 8.523 78.878 0.054 1.842 1A 837 - 8.225 3.837 0.148 72.010 0.186 15.594 IB 855 - 5.140 8.299 2.449 73.948 0.147 10.016 1C 766 - 1.089 11.662 7.239 77.719 0 2.291 *Recorded while sucking sample at any partcular point. Table 20 Experimental results: Run 91 Gas Composition, vol. % (dry) Point No. Temp. Ave. Temp. °C °C H 2 CO 2 o 2 N 2 CHi+ CO 6A 785 — 25.041 11.711 3.455 37.795 3.540 18.458 6B 792 — 22.086 13.044 1.595 42.295 3.077 17.904 6C 805 - 12.000 14.960 4.283 56.065 1.601 11.089 5A 764 - 27.718 19.666 0.312 30.418 6.501 15.322 5B 780 - • 20.604 17.744 0.131 43.436 4.429 13.655 5C 795 - 11.170 13.987 1.461 60.434 2.332 10.615 4A - - 23.278 14.013 0.541 25.610 3.169 19.414 4B - - 17.599 14.341 0.150 49.297 2.231 16.382 4C - - 9.384 12.370 2.086 66.566 1.021 8.565 3A 764 - 25.722 15.087 0.492 36.478 3.946 18.275 3B 781 - 15.066 14.728 0.147 56.330 1.879 11.850 3C 790 - 3.558 7.000 10.170 76.204 0.334 2.733 2A 780 - 20.147 6.645 0.367 52.587 1.666 18.589 2B 813 - 4.357 13.090 0.146 77.872 0.266 4.274 2C 702 - 0.785 8.699 9.735 79.834 0 0.947 1A 735 - 2.610 2.771 14.727 76.450 0.111 2.930 IB 740 - 4.307 12.416 0.126 77.166 0.129 5.850 1C 680 - 0.126 8.969 9.500 80.798 0 0.607 Table 21 Experimental results: Run 93 Point No. Temp.* °C Ave. Temp.** °C Gas Composition, vol. ' I (dry) H 2 . C0 2 °2 N 2 CH^ CO 6A 798 775 24.407 13.643 0.246 35.624 3.222 28.858 6B 803 782 19.623 13.722 0.205 41.684 2.571 19.118 6C 807 790 11.380 13.545 0.300 59.892 0.732 13.052 5A 726 776 . 22.833 14.006 0 36.348 4.819 22.000 5B 741 782 18.093 14.300 0.009 51.005 3.633 12.960 5C 753 792 12.037 13.180 0.300 61.759 1.674 11.05 4A 794 775 21.373 12.380 0 42.327 2.663 21.257 4B 807 784 15.206 12.970 0 54.787 1.649 15.388 4C 825 801 10.426 12.979 0.300 59.847 0.956 10.082 3A 746 773 22.465 14.584 0.326 41.555 3.389 17.681 3B 764 790 12.767 13.984 0.290 61.488 1.498 9.973 3C 755 797 3.964 12.098 1.811 78.270 0.442 3.415 2A 779 767 18.978 6.368 0.315 37.100 1.373 18.340 2B 809 803 6.676 11.516 0 75.016 0 6.792 2C 687 706 0.186 6.049 14.019 79.746 0 0 1A 751 763 8.051 8.756 0 74.334 0.232 8.627 IB 775 785 4.541 13.343 0 76.357 0.134 5.631 1C 659 680 0.210 8.840 10.784 79.248 0 0.919 *Recorded while sucking sample at any particular point. **Will give temperature profile along radial positions A, B, C. Averaged over the run. Table 22 Experimental results: Run 94 Gas Composition, vol. % (dry) Point No. Temp. Ave. Temp.** °C °C H 2 C0 2 0 2 N 2 CH^ CO 6A — 867 — — — _ _ 6B - 887 - - - - -6C - 881 - - - - -5A - 852 - - - - -5B - 880 - - - - -5C - 894 - - - - -4A - 830 - - - - -4B - 862 - - - - -4C - 870 - - - -3A - 816 - - - - -3B - 855 - - - - -3C - 842 - - - - -2A - 804 - - - - -2B - 844 - - - - -2C - 762 - - - -1A - 785 - - - - -IB - 823 - - - - -1C — 720 — — — — — — **Will give temperature profile along radial positions A, B, C. Averaged over the run. Table 23 Experimental results: Run 96 Point No. Temp.* °C Ave. Temp. °C Gas Composition, vol. 5 I (dry) H 2 c o 2 o 2 N 2 CH^ CO 6A 785 — 37.201 28.872 3.007 11.333 0 19.586 6B 793 - 39.694 30.646 1.316 5.933 0 22.411 6C 805 - 29.290 36.655 1.368 15.069 0 17.618 5A - - 41.832 26.732 0.815 10.609 0 20.012 • 5B - - 44.725 30.295 0.523 5.050 0 19.407 4A 743 _ 47.838 27.371 0.766 1.200 0 22.826 4B 751 - 46.611 28.550 0.492 1.571 0 22.776 4C 770 - 32.602 22.772 1.666 18.981 0 23.979 3A 820 - 49.588 20.924 0.410 1.140 0 27.939 3B 835 - 45.222 27.340 1.408 4.940 0 21.089 3C 918 - 24.989 42.754 3.017 13.964 0 14.275 2A 810 - 50.418 22.604 0.522 1.485 0 24.971 2B 853 - 43.010 29.516 0.495 7.247 0 19.732 2C 867 - 45.325 24.876 1.530 12.890 0 15.379 1A 754 - 46.430 32.745 0.180 0.584 0 20.061 IB 765 - 43.562 33.948 0.783 2.225 0 19.483 1C 815 — 21.966 46.103 3.316 15.296 0 12.319 *Recorded while sucking sample at any particular point. Table 24 Experimental results: Run 103 Point No. Temp. °C Ave. Temp. °C H, Gas Composition, vol. % (dry) CO, 0, CHL CO 6A 6B 6C 5A 5B 5C 4A 4B 4C 3A 3B 3C 2A 2B 2C 1A IB 1C 43.831 40.670 42.636 39.142 40.615 36.270 39.032 34.528 37.925 29.822 28.471 26.377 28.214 29.246 30.206 29.645 33.590 34.758 36.012 43.822 0.573 1.212 0.300 0.504 1.130 3.61 0.080 2.968 0.133 0.096 0 0 0 0 0 0 0 0 0 0 1.405 1.532 2.444 1.636 2.049 1.538 1.633 1.839 1.107 0.957 25.720 30.209 26.406 29.476 26.000 28.917 25.635 29.907 24.823 25.303 00 Table 25 Experimental results: Run 106 Point No. Temp. °C Ave. Temp. °C Gas Composition, vol. ) 1 (dry) H 2 C0 2 °2 N 2 CH^ CO 6A 981 904 8.020 13.471 0.355 66.656 0.265 11.232 6B 885 931 11.692 12.890 0.951 59.667 0.460 14.340 6C 908 921 7.094 9.521 6.900 69.195 0 7.290 5A 943 896 12.315 13.666 0.070 59.612 0.440 13.890 5B 867 917 13.878 9.852 1.501 54.350 0.616 19.803 5C 897 922 10.593 9.048 5.391 61.868 0.350 12.651 4A 946 886 14.538 9.806 1.341 56.420 0.417 17.478 4B 857 898 14.497 8.808 2.012 50.575 0.817 23.291 4C 902 918 9.010 6.130 5.201 68.094 0.354 11.211 3A 922 880 24.692 6.438 0.028 36.612 0.906 31.325 3B 874 881 11.916 6.435 5.642 60.671 0.500 14.838 3C 904 900 7.253 10.986 2.739 71.229 0.226 7.568 2A 913 873 27.652 2.067 1.196 36.284 0.764 32.037 2B 855 879 8.264 9.807 3.720 69.989 0.231 7.989 2C 913 850 4.802 12.506 2.264 75.604 0 4.821 1A - - - - - - - -IB - - - - - - - -1C — — — — — — — — *Recorded while sucking sample at any particular point. **Will give temperature profile along radial positios A, B, C. Averaged over the run. Table 26 Experimental results: exit gas composition and average bed temperature Gas Composition, vol. % (dry) Run No. Average Bed Temp. °C H 2 C0 2 0 2 N 2 CH^ CO 88 776 13.601 14.971 0 58.136 2.020 11.271 89 877 12.780 12.660 0 58.020 1.290 15.250 90 907 12.800 12.440 0 58.210 1.120 15.430 91 786 14.431 14.611 0 56.586 1.810 12.561 93 800 14.190 16.350 0 57.460 2.140 9.860 94 960 35.266 41.376 0 0.640 2.600 20.118 96 788 36.850 34.200 0 8.09 0 20.860 103 918 37.759 29.918 0 0 3.492 28.831 106 901 11.011 14.231 0 61.686 0.300 12.771 Gas Chromatograph Calibration Table APPENDIX C IL RT LEVEL ft NT flMT/flREfl 1 3 . 697 1 2 .048Q0E+01 8 . 2 3 9 3 2 E - 0 4 o 1.63300E+01 3 . 1 2 1 3 7 E - 0 4 3 1 .22900E+01 8 . 0 1 6 5 3 E - W 4 4 8 . 1 9 0 0 O E + 0 0 7 . 7 2 5 3 8 E - 0 4 5 4 . 1 0 0 0 6 E + 0 0 7 . 6 8 4 4 5 E - 0 4 5 .421 1 2 . 0 1 1 0 0 E + 0 1 1 . 0 2 1 0 1 E - 0 2 ? 1.60960E+01 9 . 9 3 1 3 2 E - 0 3 3 1 .20700E+01 9 . 9 0 7 9 4 E - W 3 4 8 .040Q0E+00 9 . 6 7 6 3 7 E - 0 3 5 4 . 0 2 0 0 0 E + 0 0 8 . 3 9 4 9 9 E - 0 3 "7 7. 852 1 1 . 0 0 0 0 0 E - 0 7 2 . 2 4 1 4 0 E - G 9 4 . 2 0 0 0 0 E + 0 0 7 . 1 9 3 5 6 E - 0 3 3 8 . 4 0 0 0 0 E + 0 0 6 . 9 9 0 3 2 E - 0 3 4 1 .26000E+01 6 . 9 9 2 4 9 E - 0 3 5 1 . 6 3 0 0 0 E + 01 6 . 9 5 2 1 3 E - 0 3 4 9 . 198 1 3 . 2 6 9 0 0 E + 0 1 8 . 5 2 4 4 4 E - 0 3 4 . 1 9 6 0 0 E + 0 1 8 . 9 6 3 0 9 E - 0 3 3 5. 12100E+01 9 . 0 6 2 3 6 E - 0 3 4 6 . 0 4 8 0 0 E + 0 1 9 . 1 9 6 9 3 E - 0 3 5 6 . 9 7 3 0 0 E + 0 1 9 . 2 2 0 4 5 E - 0 3 5 1 2 . 3 1 4 1 4 . 0 3 0 0 0 E + 0 0 5 . 0 3 8 1 9 E - 0 3 3.26000E+W0 4 . 6 1 0 5 1 E - 0 3 3 2 . 4 5 0 0 0 E + 0 0 5 . 1 1 6 4 1 E - U 3 4 1 .*63000E+W0 5 . 4 2 0 0 3 E - Q 3 5 8 . 16000E-01 5 . 7 1 7 5 4 E - 0 3 6 1 6 . 2 5 3 1 2 .26400E+01 9 . 6 7 0 5 8 E - 0 3 1 .81100E+01 9 . 5 0 W 0 5 E - 0 3 3 1 .35800E+01 9 . 6 2 1 3 2 E - 0 3 4 9 . 0 6 0 0 0 E + 0 0 9 . 9 6 0 4 3 E - 0 3 5 4 .530O0E+00 9'. 6 7 7 9 9 E - 0 3 -1 72-APPENDIX D Computer Simulation Table 27 Bed geometry and properties of solids used in model simulation Parameters Values Column ID 0.305 m Cone angle (included) 60° Height of conical section 19.8 cm Orifice diameter 3.18 cm Particle density 1.3 g/cm Solid bulk density 0.523 g/cm3 Annulus voidage 0.43 Particle diameter (spherical) 2.1 mm -173-Computer Programme for the One-dimensonal Model C T h i s programme s i m u l a t e s the a x i a l gas c o m p o s i t i o n C p r o f i l e s and t h e e x i t gas c o m p o s i t i o n . C Water gas s h i f t e q u i l i b r i u m i s assumed.Bed 1s C assumed to be 1sothermal.The f o u r t h o r d e r Runge-C K u t t a t e c h n i q u e from a l i b r a r y s u b r o u t i n e has C been used t o s o l v e the d i f f e r e n t i a l e q u a t i o n s . C Gas d i s t r i b u t i o n due t o p y r o l y s i s has been C t h e o r e t i c a l l y p r e d i c t e d . T h i s model 1s based on C t he gas p l u g f l o w i n ? a n n u l u s and s p o u t . C C H - T o t a l bed h e i g h t C TEMP -Bed t e m p e r a t u r e C P - O p e r a t i n g p r e s s u r e C RWSE - F r e q u e n c y f a c t o r f o r water gas s h i f t r e a c t i o n C EWSE - A c t i v a t i o n energy f o r water gas s h i f t r e a c t i o n C RSC - F r e q u e n c y f a c t o r f o r C/H20 r e a c t i o n C ESC - A c t i v a t i o n energy f o r C/H20 r e a c t i o n C RCOC - F r e q u e n c y f a c t o r f o r C/C02 r e a c t i o n C ECDC - A c t i v a t i o n energy f o r C/C02 r e a c t i o n C ROC - F r e q u e n c y f a c t o r f o r C/02 r e a c t i o n C EOC - A c t i v a t i o n e nergy f o r C/02 r e a c t i o n C DC -Column diameter(m) C OP - P a r t i c l e dlameter(m) C DI - O r i f i c e i n l e t diameter(m) C RAWC -Coal d e n s i t y ( g / m 3 ) C RAWB -Bulk d e n s i t y o f coa1(g/m3) C TGF - O v e r a l l gas" f l o w r a t e ( k g / h ) C N -Number o f DE's to be s o l v e d C X - I n i t i a l bed h e i g h t C Y(1) - I n i t i a l H20 cone, i n spout(mol/m3) C Y(2) - I n i t i a l H20 cone, i n annulus(mol/m3) C Y(3) - I n i t i a l CO cone, i n spout(mol/m3) C Y(4) - I n i t i a l CO cone, i n annu1us(mol/m3) C Y(5) - I n i t i a l H2 cone, i n spout(mol/m3) C Y(6) - I n i t i a l H2 cone, i n annu1us(mo 1/m3) C Y(7) - I n i t i a l 02 cone, i n spout(mpo1/m3) C Y(8) - I n i t i a l 02 cone, i n annul us(mol/m3) C Y.(9) - I n i t i a l C02 cone. In spout(mo 1/m3) C Y(10) - I n i t i a l C02 cone, i n annulus(mol/m3) C CN2 - I n i t i a l N2 Cone. 1n spout & annu1us(mol/m3) C CUMF -A f a c t o r i n Mammuro-Hattor 1 e q u a t i o n C K -Number o f s t e p s i n the i n t e g r a t i o n C EA -Bed p o r o s i t y tn a n n u l u s C VIS -Gas v i s c o s i t y ( N s / m 2 ) IMPLICIT REAL*8( A-H.O-Z) DIMENSION Y ( 1 0 ) , F ( 1 0 ) , T ( 1 0 ) , S ( 1 0 ) , G ( 1 0 ) DIMENSION HA(27).HS(27).CMS(27).CMA(27).OA(27),0S(27). *CDA(27).CDS(27),HEI(27),H20A(27) COMMON/BLKA/UMFH,RAWGH,DP,CUMF,TEMP COMMON/BLKB/H,CN2,P,RWSE,EWSE,RSC,ESC.PIE,DC. #RC0C,ECDC,ROC,EOC,AS,HM,RAWC.GMF.K COMMON/BLKC/I EXTERNAL FUNC DATA N.X/10..00300/ DATA Y/1.9O70O.1.907D0,0.DO,0.DO,0.DO.0.DO,1.998D0.1.99800, MO.00000..00000/ K = 27 H=0.97DO TEMP=1073.00 CN2=7.515D0 -1 74-P = 1 .DO RWSE=.026500 EWSE=3955.DO RSC=1.2D9 ESC=29790.D0 RCDC=4.0D8 ECDC=29790.D0 R0C=8S0.D0 EOC=18000.DO DC=.305D0 RAWC=1.306 DP=.0021000 DI=.031800 TGF=118.000 RAWB=.52306 CUMF=0.9D0 PIE=4.DO*DATAN(1.DO) ACH=PIE*DC*DC/4.DO C Maximum Spoutable bed height HM HM=.105D0*0C*((DC/DP)**.75D0)*((DC/DI)**.400)/ U((RAWC/1.D6)**1.2D0) C C Mean spout diameter GMF=(TGF)/360O.DO DS=.11800*((GMF/ACH)** .4900)*(DC**.6800)/ #((RAWB/1.D6)«*.41D0) AS=PIE*DS*0S/4.D0 AAH=ACH-AS C C Calcu l a t e H1,HMIN,set E required f or DRKC H1=(H-X)/64.DO HMIN=0.01D0*H1 E=1.D-5 C C a l l DRKC to solve DE's DELZ=(H-X)/ FLOAT(K) Z=X+0EL2 WRITE(6,101) 101 F0RMAT(/'PERCENTAGE DRY GAS COMPOSITION IN SPOUT'/) DO 40 1=1,K CALL DRKC(N.X.Z.Y,F.H1.HMIN,E.FUNC.G.S.T) C C Concentration adjustment due to water s h i f t reaction EWS=RWSE*DEXP(EWSE/TEMP) DELCS = (EWS*Y( 1)*Y(3)-Y(5)*Y(9))/(Y(5)+Y(9)+EWS*(Y(1)+Y(3) ) ) DELCA = (EWS*Y(2)*Y(4)-Y(6)*Y(10))/(Y(6) + Y(10) + EWS*(Y ( 2) + Y(4))) Y(1)=Y(1)-0ELCS Y(2)=Y(2)-DELCA Y(3)=Y(3)-0ELCS Y(4)=Y(4)-0ELCA Y(5)=Y(5)+DELCS Y(6)=Y(6)+0ELCA Y(9)=Y(9)+0ELCS Y(10)=Y(10)+DELCA IF(Y(7).LT.O.)Y(7)=0.D0 IF(Y(8).LT.O.)Y(8)=0.D0 Z=X+DELZ H20A(I)=Y(2 ) C Calcu l a t e percentage dry gas composition at each level TM0A=CN2+Y(4)+Y(6)+Y(8)+Y(10) -175-TMDS=CN2+Y(3 )+Y(5 ) +Y(7)+Y(9) CMS( I )=Y(3 )/TMDS*100.DO CMA( I )=Y(4)/TMDA* 100.DO HS ( I )=Y (5 )/TMDS*100 .DO HA( I )=Y(6)/TMOA * 100.DO 0S ( I )=Y (7 ) /TMDS*100 .D0 OA( I )=Y (8 )/TMDA*10O.D0 CDS ( I )=Y (9 )/TMDS*100 .DO CDA( I )=Y (10 )/TMDA*100.DO HE I ( I )=X W R I T E ( 6 , 2 1 ) H E I ( I ) , C M S ( I ) , H S ( I ) , C D S ( I ) . 0 S ( I ) 21 F 0 R M A T ( 2 X . 5 F 1 0 . 3 ) 40 CONTINUE C UAH=CUMF*UMFH*(1 .D0-( (1 .DO-H/HM)**3) ) USH=(GMF-(UAH*AAH*RAWGH))/(AS*RAWGH) C C a l c u l a t i o n o f p e r c e n t a g e d r y gas c o m p o s i t i o n ( e x i t ) CH20E=( (UAH*AAH*Y (2 ) )+ (USH*AS*Y (1 ) )+ .0738800 ) CCOE=( (UAH*AAH*Y (4 ) )+ (USH*AS*Y (3 ) )+ .04084D0) CH2E=( (UAH*AAH*Y(6 ) ) +(USH*AS*Y ( 5 ) )+ .1266D0) C02E=( (UAH*AAH*Y (8 ) )+ (USH*AS*Y (7 ) ) ) CC02E=( (UAH*AAH*Y (10 ) )+ (USH*AS*Y (9 ) )+ .01377D0) CH4E=.011050 CN2E=((UAH*AAH*CN2)+(USH*AS*CN2)) C Changes i n c o n c e n t r a t i o n due t o gas c o m b u s t i o n and C wa te r s h i f t e q u i l i b r i u m a d j u s t m e n t DE0=C02E/(CC0E+CH2E) CH20E1=CH20E+2.DO*0E0*CH2E CC0E1=CC0E-2.DO*0E0*CC0E CH2E1=CH2E-2.DO*0E0*CH2E CC02E1=CC02E+2.DO*DE0*CC0E C02E=O.DO I F ( C C 0 E 1 . L E . O . ) G 0 TO 222 I F ( C H 2 E 1 . L E . O . ) GO TO 222 DELCF=(EWS*CH20E1*CC0E1-CH2E1*CC02E1)/ #(CH2E1+CC02E1+EWS*(CH20E1+CCOE1)) CH20F=CH20E1-DELCF CC0F=CC0E1-DELCF CH2F=CH2E1+0ELCF-CC02F=CC02E1+DELCF TCED=CC0F+CH2F+C02E+CC02F+CN2E+CH4E CMD=CC0F/TCED*100.DO HD=CH2F/TCED-100.00 0D=C02E/TCED*1O0.D0 CCD=CC02F/TCED*100.D0 CH4F=CH4E/TCED*100.D0 WR ITE (S .56 ) 56 FORMAT( / ' EXIT GAS COMPOSIT IONS v/v D R Y ) ' / ) WRITE(6.50)CMD 50 F0RMAT(2X, 'CARBON MONOXIDE = ' . 2 X . F 1 0 . 3 / ) WRITE (6 .55)HD 55 FORMAT (2X , 'HYDROGEN= ' . 2X . F10 .3/ ) WRITE (6 .SO)0D 60 F O R M A T ( 2 X , ' O X Y G E N = ' . 2 X . F 1 0 . 3 / ) WR ITE (6 ,65 )CCD 65 F0RMAT(2X, 'CORBON D I O X I D E = ' . 2 X . F 1 0 . 3 / ) WR ITE (6 ,66 )CH4F 66 F O R M A T ( 2 X , ' M E T H A N E = ' . 2 X . F 1 0 . 3 / ) GO TO 74 222 WR ITE (6 ,224 ) -1 76-224 F 0 R M A T ( / 2 X , ' A L L CO OR H2 CONSUMED BY COMBUSTION IN FOUNTAIN'/ ) 74 WRITE(6 ,75) 75 FORMAT(/12X, 'PERCENTAGE DRY GAS COMPOSITION IN ANNULUS'/) WRITE (6 ,80) 80 F O R M A T ( 2 X , ' H E I G ( M ) ' , S X , ' C MONOX ' ,7X , 'HYDROGEN ' , # 9 X , ' C D I O X ' , 1 0 X , ' O X Y G E N ' / ) WR ITE (6 ,85 ) (HE I ( I ) , C M A ( I ) , H A ( I ) . C D A ( I ) , O A ( I ) , 1 = 1,K) 85 FORMAT(5F12.3 ) WR ITE (6 ,111 )H ,DP ,TEMP,RSC ,RCDC 111 FORMAT(/2X, ' H = ' . F 6 . 3 , 2 X , ' D P = ' , F 6 . 5 . 2 X , ' T E M P = ' . F 1 0 . 3 , # 2 X , ' R S C = ' . E 1 0 . 3 . 2 X . ' R C C D = ' . E 1 0 . 3 ) STOP END C C S u b r o u t i n e FUNC SUBROUTINE F U N C ( X . Y . F ) IMPLICIT REAL*8 (A-H.O-Z ) DIMENSION Y ( 1 0 ) , F ( 1 0 ) COMMON/BLKA/UMFH,RAWGH,DP,CUMF,TEMP C0MM0N/BLK8/H,CN2,P ,RWSE ,EWSE ,RSC.ESC,P IE ,DC, #RCDC,ECDC,ROC.EOC.AS.HM,RAWC,GMF,K COMMON/BLKC/I C C C a l c u l a t e gas d e n s i t y m s p o u t . a n n u l us and UMF VIS=4.S1D-5 RAWG=P*1.D5*29.D-3/(8 .314D0*TEMP) GAN=(DP* *3)* ( (RAWC/1.D3)-RAWG)*RAWG*9.81DO/(VIS**2) R E B F = ( ( ( 1 8 . 1 D O * * 2 ) + . 0 1 9 2 D O * G A N ) * * . 5 ) - 1 8 . 1 0 0 R E T F = ( ( ( 2 4 . D 0 * * 2 ) + . 0 5 4 S D O * G A N ) * * . 5 ) - 2 4 . D O UBF=REBF*VIS/(DP*RAWG) UTF=RETF*VIS/(DP*RAWG) UMF=.5D0*(UBF+UTF) IF ( I . L T . K ) GO TO 25 UMFH=UMF RAWGH=RAWG C S u p e r f i c i a l v e l o c i t y i n a n n u l u s 25 UA=CUMF*UMF* (1 .D0-( (1 .DO-X/HM)**3 ) ) C C P o r o s i t y i n s p o u t ES=1 .D0- .2D0*X/H C C A v e r a g e s u p e r f i c i a l v e l . i n spou t and d e r i v a t i v e o f C a n n u l u s a r e a d i v i d e by a n n u l u s a r e a ( D I V A ) I F ( X . G E . O . 1 9 8 0 0 ) GO TO 10 CRAD = X * (DTAN(P I E /S .DO ) )+0 .0400D0 AC=PIE*CRAD*CRAD DA1=0TAN(P IE/S .D0) DAZ=2 .D0*P I E * ( ( X *DA1*DA1 )+0 .0400D0*0A1 ) GO TO 20 10 AC=P I E *0C*0C/4 .D0 DAZ=0. 20 US=(GMF-(UA*(AC-AS)*RAWG))/(AS*RAWG) DIVA=OAZ/(AC-AS) C C A n n u l u s v e l . g r a d i e n t DUA DUA=3.ODO*CUMF*UMF*( (1.DO-X/HM)**2)/HM I F ( Y ( 7 ) . L T . 0 . )Y (7 )=0.D0 I F ( Y ( 8 ) . L T . O . )Y (8 )=0.D0 C -1 77-C T o t a l gas c o n c e n t r a t i o n i n spout TCS=Y(1)+Y(3 )+Y(5 )+Y(7 )+Y(9 )+CN2 C T o t a l gas c o n c e n t r a t i o n i n a n n u l u s TCA=Y(2)+Y(4)+Y(6)+Y(8)+Y(10)+CN2 C Rate c o n s t a n t f o r C/H20 r e a c t i o n R1=RSC*(DEXP(-ESC/TEMP) )*6.DO/DP C C Rate c o n s t a n t f o r C/C02 r e a c t i o n R2=RCDC*(DEXP(-ECDC/TEMP) )*6.DO/DP C C Rate c o n s t a n t f o r C/02 r e a c t i o n R3=R0C* (DEXP (-E0C/TEMP) ) *6 .D5/ (DP*12 .DO) C C C DE f o r s team 1n s p o u t F( 1 ) = -R1*Y (1 ) * (1 .DO-ES ) /US C C DE f o r s team 1n a n n u l u s F ( 2 ) = - ( R 1 * Y ( 2 ) ) * . 5 7 D O / U A + ( Y ( 1 ) - Y ( 2 ) ) * #(DUA/UA+DIVA) C C DE f o r C Monox 1n s p o u t F ( 3 ) = ( R 1 * Y ( 1 ) + 2 . D O * R 2 * Y ( 9 ) ) * ( 1 . D O - E S ) / U S C C DE f o r C Monox 1n a n n u l u s F ( 4 ) = ( R 1 * Y ( 2 ) + 2 . D 0 * R 2 * Y ( 1 0 ) ) * . 5 7 D O / U A + #{Y (3 )-Y(4 ) ) * (DUA/UA+DIVA) C C DE f o r Hyd rogen m spou t F (5 )=R1*Y (1 ) * (1 .DO-ES ) /US C C DE f o r Hyd rogen i n a n n u l u s F ( S )= (R1*Y (2 ) ) * . 57D0/UA+(Y (5 ) -Y ( S ) ) * (DUA/UA+OIVA ) C C DE f o r Oxygen i n s p o u t F ( 7 )=-R3*Y (7 ) /TCS* (1 .D0-ES ) /US C C OE f o r Oxygen In a n n u l u s F (8 ) = -R3*Y(8 ) /TCA* .S7DO/UA+(Y (7 )-Y (8 ) ) * (DUA/UA+DIVA) C C DE f o r C D i o x i n s p o u t F ( 9 ) = ( R 3 * Y ( 7 ) / T C S - R 2 * Y ( 9 ) ) * ( 1 . D 0 - E S ) / U S C C OE f o r C D i o x i n a n n u l u s F ( 1 0 ) = ( R 3 * Y ( 8 ) / T C A - R 2 * Y ( 1 0 ) ) * . 5 7 D O / U A + # (Y (9 )-Y (10) )- (DUA/UA+DIVA) RETURN END -1 78-Computer Programme for the Streamtube Model. C T h i s programme u s e s the s t r e a m t u b e model to C s i m u l a t e g a s i f i c a t i o n p r o c e s s . Water gas s h i f t C e q u i l i b r i u m i s a s sumed .Bed i s assumed to be C i s o t h e r m a l . T h e f o u r t h o r d e r Runge-Kut ta t e c h n i q u e C f r om a l i b r a r y s u b r o u t i n e has been u s e d to s o l v e C t h e d i f f e r e n t i a l e q u a t i o n . C IMPLICIT REAL*8 (A-H ,0-Z ) COMMON/BLKA/TEMP,UMF,RAWG,CUMF,HM.AS,HC, #RB ,DC ,P I E ,GMF ,H COMMON/BLKB/UST,EA COMMON/BLKC/R1,R2,R3,CN2 DIMENSION R ( 8 1 , 4 1 ) , 0 ( 4 0 ) , H S ( 4 0 ) , H H ( 4 0 ) , Y ( 5 ) DIMENSION F ( 5 ) , Y Y ( 5 ) , F F ( 5 ) , T H ( 8 1 ) , U A 1 ( 8 1 ) . R C ( 8 1 , 4 1 ) DIMENSION H 2 0 S ( 4 0 ) , C O S ( 4 0 ) , H 2 S ( 4 0 ) , C 0 2 S ( 4 0 ) , 0 2 S ( 4 0 ) DIMENSION T ( 5 ) , S ( 5 ) , G ( 5 ) EXTERNAL FUNC1,FUNC2 DATA Y / 1 . 9 0 7 3 D O . O . D O . O . D O . O . D O , 1 . 9 9 7 6 D 0 / C RWSE - F r e q u e n c y f a c t o r f o r water gas s h i f t r e a c t i o n C EWSE - A c t i v a t i o n e n e r g y f o r water gas s h i f t r e a c t i o n C RSC - F r e q u e n c y f a c t o r f o r C/H20 r e a c t i o n C ESC - A c t i v a t i o n e n e r g y f o r C/H20 r e a c t i o n C RCDC - F r e q u e n c y f a c t o r f o r C/C02 r e a c t i o n C ECDC - A c t i v a t i o n e n e r g y f o r C/C02 r e a c t i o n C ROC - F r e q u e n c y f a c t o r f o r C/02 r e a c t i o n C EOC - A c t i v a t i o n e n e r g y f o r C/02 r e a c t i o n C DC -Column d i a m e t e r ( m ) c DP - P a r t i c l e d i a m e t e r ( m ) c DI - O r i f i c e i n l e t d i a m e t e r ( m ) c RAWC -Coa l d e n s l t y ( g / m 3 ) c RAWB -Bu lk d e n s i t y o f c o a l ( g / m 3 ) c TGF - O v e r a l l g a s s f l o w r a t e ( k g / h ) c NN -Number o f D E ' s t o be s o l v e d c N -Number o f h o r i z o n t a l g r i d p o i n t s c M -Number o f v e r t i c a l g r i d p o i n t s c X - I n i t i a l bed heght (m) c Y (1 ) - I n i t i a l H20 c o n e , i n spout (mo l/m3) c Y (2 ) - I n i t i a l CO c o n e , i n spout (mo l/m3) c Y (3 ) - I n i t i a l H2 c o n e , i n spout (mo l/m3) c Y ( 4 ) - I n i t i a l C02 c o n e . 1n spout (mo l/m3) c Y (5 ) - I n i t i a l 02 c o n e . 1n spout (mo l/m3) c CUMF -A f a c t o r 1n Mammuro-Hattor1 e q n . c K -Number o f s t e p s i n t h e i n t e g r a t i o n c EA -Bed p o r o s i t y i n a n n u l u s c VIS -Gas v i s c o s 1 t y ( N s / m 2 ) c RB - R a d i a l d i s t a n c e f r om c e n t r e o f co lumn t o c w a l l a t the b a s e o f g a s i f l e r ( m ) c CN2 _ i m t i a l N2 c o n c . m spout (mo l/m3) c HC - H e i g h t o f c y l i n d r l c a l s e c t i o n ( m ) c DZS - I n c r e m e n t a l s p o u t h e i g h t to be s p e c i f i e d c r e q u i r e d t o c a l c u l a t e h e i g h t s o f s t r e a m l i n e s c i n s p o u t c ER - T o l e r a b l e e r r o r t o be s p e c i f i e d to c a l c u l a t e c v o l u m e t r i c f l o w a l o n g s t r e a m ! 1 n e s ( m 3 / s ) . T h e s e c f l o e r a t e s a r e u s e d t o c o n s t r u c t s t r e a m l i n e s c a l o n g the s p o u t . c c c c H TEMP P - T o t a l bed h e i g h t -Bed t e m p e r a t u r e - O p e r a t i n g p r e s s u r e -179-C ZO -Spout h e i g h t a t i n l e t = 0 . H=.97DO CN2=7.515D0 P=1.DO RWSE=.02S5DO EWSE=3955.DO RSC=3.D8 ESC=29790.DO RCCD=1.D8 ECCD=29790.D0 ROC=860.00 EOC=18000.DO DC=.305D0 RAWC=1.3D6 DP=.002 1D0 D I=.0318D0 TGF =118.OODO RAWB=.523DS CUMF=.9D0 EA=.43D0 TEMP=1073.D0 N=21 M=41 NN=5 RB=.0400D0 HC=.198D0 DZS=.010D0 ER=1.D-5 ZO=O.DO C P IE=4.DO*DATAN(1 .DO) ACH=PIE*DC*DC/4.00 C C Mean s p o u t d i a m e t e r and max s p o u t a b l e h e i g h t GMF=(TGF)/360O.D0 DS = .11800* ( (GMF/ACH)* * . 4 9 0 0 ) * ( D C * * . 6 8 D 0 ) / # ( (RAWB/1 .D6 )** .41D0) AS=P IE*DS*DS/4.D0 H M = . 1 0 5 D 0 * 0 C * ( ( D C / D P ) * * . 7 5 0 0 ) * ( ( D C / D I ) * * . 4 D 0 ) / # ( ( RAWC/1 .D6 ) * *1 .200 ) C C Minimum f l u l d l z i n g v e l o c i t y and a n n u l u s v e l o c i t y C a t t op o f bed V IS=4.D-5 RAWG=P*1.D5*29.D-3/ (8 .314D0*TEMP) GAN=(DP**3)*( (RAWC/1.D3)-RAWG)*RAWG*9.81DO/ # (V IS**2 ) R E B F = ( ( ( 1 8 . 1 D 0 * * 2 ) + . 0 1 9 2 D 0 * G A N ) * * . 5 ) - 1 8 . 1 D 0 RETF = ( ( ( 2 4 . D 0 * * 2 ) + . 0 5 4 6 D 0 * G A N ) * * . 5 ) - 2 4 . D O . ' UBF=REBF*VIS/(DP*RAWG) UTF=RETF*VIS/(DP*RAWG) UMF=.5DO*(UTF+UBF) UAH=CUMF*UMF*( 1 . DO-( 1 . DO-H/HM.) **3 ) C C D i v i d e t op o f bed i n t o N-1 s e c t i o n s and comput C g r i d p o i n t s a l o n g top o f bed and co lumn wa l l C NM1=N-1 MM1=M-1 R(M, 1 )=DC/2.DO -180-RI=(DC-DS)/ (2 .DO*NM1) DO 10 d=1,NM1 R (M. J+1)=R(M, J )-R I 10 CONTINUE DZA=(H-Z0)/MM1 Z = ZO DO 20 1=1,MM1 I F ( Z . L E . H C ) R ( I , 1 )=Z* (DTAN(P I E/6 .DO) )+RB I F ( Z . G T . H C ) R ( I ,1 )=DC/2.D0 Z=Z+DZA 20 CONTINUE C C C a l c u l a t e v o l u m e t r i c f l o w f rom each s t r e a m t u b e DO 30 J=1,NM1 0 ( J )=P I E * (R (M . J ) * *2-R (M , J+1 ) * *2 ) *UAH*EA 30 CONTINUE C C Compute h e i g h t o f s t r e a m t u b e a l o n g spout Z1=0.DO 0AO=O.DO DO 40 J=1,NM1 uJ=0 55 Z=Z1+DZS 60 UA=CUMF*UMF* (1 .D0-(1 .DO-Z/HM)**3 ) I F ( Z . L E . H C ) CRAD=Z*(DTAN(P IE/6.D0) )+RB I F ( Z . G T . H C ) CRAD-0C/2.DO AC=P IE«CRAD*CRAD QA=UA*(AC-AS)*EA DOA=OA-QAO E=DABS(Q(d)-DQA) I F ( E . L E . E R ) GO TO 45 I F ( J J . E O . 1 ) GO TO 65 I F ( D O A . G T . O ( d ) ) GO TO 50 C C I n c r e m e n t a l s e a r c h Z1=Z DQ1=0QA GO TO 55 C C L i n e a r I n t e r p o l a t i o n 50 Jd=1 Z2 = Z 0Q2=D0A F 1=001-0( J ) F2=DQ2-Q(d) 75 Z 3 = ( Z 1 * F 2 - Z 2 * F 1 ) / ( F 2 - F 1 ) Z=Z3 GO TO 60 65 F3=DQA-0(J ) I F ( F 1 * F 3 . L E . O . ) GO TO 70 Z1=Z3 F 1 =F3 GO TO 75 70 Z2=Z3 F2 = F3 GO TO 75 45 HS(d )=Z QAO=OA Z1=Z -181-C Compute mid p o i n t f o r e a c h s t r e a m t u b e a l o n g s p o u t I F ( J . E 0 . 1 ) HH( J )=HS( J ) /2 .DO I F ( J . G E . 2 ) HH ( J )= (HS ( J )+HS (d-1 ) ) /2 .D0 40 CONTINUE C C Ra te c o n s t a n t f o r C/H20 R1=RSC*DEXP(-ESC/TEMP)*6.DO/DP C Ra t e c o n s t a n t f o r C/C02 R2=RCCD*DEXP(-ECCD/TEMP)*6.DO/DP C Ra te c o n s t a n t f o r C/02 R3=R0C*DEXP (-E0C/TEMP)*6 .D5/ (12 .D0*DP ) C C Compute gas c o n c e n t r a t i o n f r o f i l e s a t a l l i n l e t C t o s t r e a m t u b e s C C a l c u l a t e H1,HMIN-. s e t ERC r e q u i r e d by DRKC ' WR ITE (6 ,77 ) 77 FORMAT( ' STREAMTUBE INLET CONDITIONS(DRY) '/ ) WR ITE (6 .78 ) 78 F0RMAT (2X , ' INLAT H T ' , 4 X , ' C M O N O X ' , 4 X , ' H Y D ' , 4 X , ' C D I O X ' , 4 X , ' O X Y ' / ) ERC=1.D-5 X = ZO DO 100 d=1,NM1 H1=(HH(U )-X )/64 .D0 HMIN=.0100*H1 ZZ=HH(J) CALL DRKC (NN ,X ,ZZ .Y . F .H1 ,HM IN . ERC . FUNC1 , G . S . T ) X = Z2 C C o n c e n t r a t i o n a d j u s t m e n t due t o s h i f t r e a c t i o n WSE=RWSE*DEXP(EWSE/TEMP) DELC= (WSE*Y (1 ) *Y (2 ) -Y (3 ) *Y (4 ) ) / (Y (3 ) + Y(4)+WSE*(Y(1) + Y ( 2 ) ) ) Y (1 )=Y (1 )-DELC Y (2 )=Y (2 )-DELC Y (3 )=Y(3 )+DELC Y (4 )=Y(4 )+DELC I F ( Y ( 5 ) . L E . 0 . ) Y ( 5 ) = 0 . H20S ( J )=Y (1 ) C0S ( J )=Y (2 ) H2S (d )=Y (3 ) C02S (d )=Y (4 ) 02S (U )=Y (5 ) C C C a l c u l a t e p e r c e n t d r y ga s c o m p o s i t i o n i n s p o u t TMDG=CN2+Y(1)+Y(2)+Y(3)+Y(4)+Y(5) CMD=Y(2)/TMDG*100.D0 HD=Y(3 ) /TM0G»10O.DO CDD=Y(4 )/TMDG*100.00 0D=Y(5)/TMDG*10O.D0 WR ITE (6 ,90 ) HH( J ) ,CMD,HD,CDD,00 90 F0RMAT(5F12 .3 ) 100 CONTINUE C C Compute R (M, J ) XX=O.DO DO 110 J=1,NM1 YY (1 )=H20S ( J ) YY (2 )=COS( J ) YY (3 )=H2S ( J ) YY (4 )=C02S (d ) YY (5 )=02S (d ) K=1 -182-DO 120 1=2,M TH ( I )=FL0AT ( I -1 ) *DZA+ZO I F ( T H ( I ) . L T . H S ( J ) ) GO TO 120 I F ( U . E Q . N M 1 . A N D . I . E Q . M ) K=1 I F ( I . E O . M ) GO TO 125 UA1(1) =CUMF*UMF*(1 .D0~(1.DO-TH( I )/HM)**3 ) W=(R ( I . J ) * *2 ) -0 ( J ) / (UA1 ( I )*P IE*EA) R( I , J+1)=DSORT(W) C C C a l c u l a t e c e n t r e p o i n t a t e a c h l e v e l i n s t r e a m t u b e 125 RC ( I . J ) = ( R ( I , J+ 1 )+R ( I . >J ) ) /2 . DO I F ( K . E Q . 1) P L = D S O R T ( ( T H ( I ) - H H ( J ) ) * » 2 + ( R C ( I , d ) -OS/2 . DO)**2) I F (K . E 0 . 2 ) PI.=DSQRT( (TH( I )-TH( I-1 ) )-*2 # + ( R C ( I , J ) - R C ( I - 1 . J ) ) * * 2 ) I F ( K . E O . 1 ) UAA=CUMF*UMF*(1.DO-(1.DO-HH( J )/HM)**3) I F ( K . E 0 . 2 ) UAA=UA1( I -1) I F ( K . E Q . 1 ) A=TH( I )-HH( J ) I F ( K . E 0 . 2 ) A=TH( I )-TH(1-1) UST=UAA*A/PL C C C a l c u l a t e gas c o n c e n t r a t i o n p r o f i l e s i n s t r e a m t u b e ZF=XX+PL H1=PL/64 .D0 HMIN=.01D0*H1 CALL D R K C ( N N , X X , Z F , Y Y , F F , H 1 , H M I N , E R C , F U N C 2 , G , S , T ) C C C o n c e n t r a t i o n a d j u s t m e n t due t o water s h i f t WSE=RWSE*DEXP(EWSE/TEMP) DELC=(WSE*YY (1 ) *YY (2 ) -YY (3 ) *YY (4 ) ) / ( YY (3 )+YY (4 ) #+WSE*(YY(1 )+YY(2) ) ) YY (1 )=YY (1 )-DELC YY (2 )=YY (2 )-DELC YY(3 )=YY(3 )+DELC YY(4 )=YY(4 )+DELC I F ( Y Y ( 5 ) . L T . O . ) YY (5 )=0 . C C C a l c u l a t e p e r c e n t d r y ga s c o m p o s i t i o n TMDGA=CN2+YY(1)+YY(2)+YY(3)+YY(4)+YY(5) CM0A=YY(2)/TMDGA*1OO.DO HDA=YY(3)/TM0GA*1OO.DO H CDDA=YY(4)/TMDGA*100.DO 0DA=YY(5)/TMDGA*1OO.DO K = 2 W R I T E ( S . 1 4 0 ) T H ( I ) . J , R C ( I , J ) , I 140 F O R M A T ( 2 X , ' H E I G = ' , F 1 0 . 4 , 4 X , ' d = / , I 2 , 4 X , ' D I S FR C E N T R E = ' , F 1 0 . 5 , #4X,12) WRITE(6, 145)CMDA.HDA,CODA,ODA 145 FORMAT(4F10 .3/ ) 120 CONTINUE 110 CONTINUE STOP END C C S u b r o u t i n e FUNC1 SUBROUTINE F U N C 1 ( X . Y , F ) IMPL ICIT REAL*8 (A-H ,0-Z ) COMMON/BLKA/TEMP,UMF,RAWG,CUMF,HM,AS,HC,RB , DC, #P IE ,GMF,H COMMON/BLKC/R1,R2,R3,CN2 DIMENSION Y ( 5 ) . F ( 5 ) -183-c C S u p e r f i c i a l v e l o c i t y i n a n n u l u s UA=CUMF*UMF*(1.D0-(1.DO-X/HM)**3) C C P o r o s i t y i n s p o u t E S = 1 . D 0 - . 2 D 0 » X / H C C V e l o c i t y i n s p o u t I F ( X . L E . H C ) CRAD=X*(DTAN(PIE/6.DO))+R8 I F ( X . G T . H C ) CRAD=DC/2.D0 AC=PIE*CRAD*CRAD US=(GMF-UA*(AC-AS)*RAWG)/(AS*RAWG) TCS=Y( 1 ) + Y (2 ) + Y (3 )+Y(4 ) + Y (5 ) + CN2 F (1 )=-R1*Y (1 ) * (1 .D0-ES ) /US F ( 2 ) = ( R 1 * Y ( 1 ) + 2 . D 0 * R 2 * Y ( 2 ) ) * ( 1 . D O - E S ) / U S F ( 3 )=R1*Y (1 ) * ( 1 .D0-ES ) /US F ( 4 ) = ( R 3 * Y ( 5 ) / T C S - R 2 * Y ( 4 ) ) * ( 1 . D O " E S ) / U S F (5 )=-R3*Y (5 ) /TCS* (1 .DO-ES ) /US RETURN END C C S u b r o u t i n e FUNC2 SUBROUTINE FUNC2 (XX .YY , FF ) IMPLICIT REAL*8 (A-H ,0-Z ) COMMON/BLKB/UST,EA COMMON/BLKC/R1,R2.R3.CN2 DIMENSION Y Y ( 5 ) . F F ( 5 ) TCA=YY(1)+YY(2)+YY(3)+YY(4)+YY(5)+CN2 FF( 1 ) = -R1*YY( 1 )*•( 1 .DO-EA)/UST FF (2 ) = (R1*YY( 1 )+2 .DO*R2*YY (4 ) ) * (1 .DO-EA )/UST FF (3 )=R1*YY (1 ) * (1 .DO-EA ) /UST F F (4 )= (R3*YY (5 ) /TCA-R2*YY (4 ) ) * ( 1 .DO-EA ) /UST FF (5 )=-R3*YY (5 ) /TCA* (1 .DO-EA ) /UST RETURN END 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0058902/manifest

Comment

Related Items