Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Combustion performance and high temperature hydrodynamics in a spouted and spout-fluid bed Ye, Bogang 1988

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

Item Metadata

Download

Media
831-UBC_1988_A7 Y42.pdf [ 6.31MB ]
Metadata
JSON: 831-1.0058774.json
JSON-LD: 831-1.0058774-ld.json
RDF/XML (Pretty): 831-1.0058774-rdf.xml
RDF/JSON: 831-1.0058774-rdf.json
Turtle: 831-1.0058774-turtle.txt
N-Triples: 831-1.0058774-rdf-ntriples.txt
Original Record: 831-1.0058774-source.json
Full Text
831-1.0058774-fulltext.txt
Citation
831-1.0058774.ris

Full Text

C O M B U S T I O N P E R F O R M A N C E A N D H I G H T E M P E R A T U R E H Y D R O D Y N A M I C S IN A S P O U T E D A N D S P O U T - F L U I D B E D By Mr. Bogang Ye B. A.Sc., Zhejiang University, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR T H E DEGREE OF M A S T E R OF APPLIED SCIENCE in T H E FACULTY OF GRADUATE STUDIES CHEMICAL ENGINEERING We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA October 1988 © Mr. Bogang Ye, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Chemical Engineering The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T' 1W5 Date: Oct. /7//1S8 Abstract Combustion of Minto coal, a sub-bituminous eastern coal which is highly agglomerating and has a high sulphur content, was carried out in a 0.15 m internal diameter half-column spout-fluid bed combustor in inert beds of sand, with limestone addition for sulphur capture. The average bed temperature ranged from 800 to 900°C, flue gas oxygen level was 2.5 to 11.0%, auxiliary to total air was 0 to 0.50, and Ca/S molar ratio was 2.5. High vale coal was employed in hydrodynamic runs. Aspects studied included combustion efficiency, sulphur capture efficiency, axial and radial temperature profiles, axial 02 and C02 concentration profiles, axial SO2 concen-tration profiles, minimum spouting velocity, spouting stability, and maximum spoutable bed height. The principal problem encountered with Minto coal in this equipment was agglom-eration during the heat-up period. A spout-fluid bed has proved to be great favourable for handling agglomerating coal relative to the standard spouted bed. When limestone was used as bed material, less sintering was encountered. However, limestone could not stand up to spouting for prolonged periods because of excessive attrition. Combustion efficiencies were found to be higher than 80% in the temperature range of 800 to 900°C without solid fines recycle. An increase of temperature between 800°C and 840°C was beneficial for combustion efficiency, while a further increase up to 885°C did not seem to have a significant effect on combustion efficiency. Increase of auxiliary/total air ratio was favourable to combustion efficiency at elevated temperatures. Sulphur capture efficiency passed through a maximum with increasing temperature between 800°C and 900°C The maximum value was obtained at around 830°C. n J\f0x emission increased linearly with increasing flue gas oxygen level. No abrupt temperature increase above the bed surface was observed in both spouted and spout-fluid beds investigated in the present study. Temperature may increase above the bed surface for low excess oxygen runs in view of the substantial amount of combustion found to occur in the freeboard. Temperatures were more uniform after the introduction of auxiliary air. Most oxygen was consumed below the bed surface. Axial profiles showed a signifi-cant SO2 jump in the spout over the bed height. Combustion and sulphation could be considered to occur in two main stages: (1) Combustion of carbon, at the same time as most of the sulphur is released. (2) Sulphation of the sorbent. The Mathur and Gishler (1955) and Wu et al. (1987) equations gave poor agreement with the minimum spouting velocity, Umil, over the entire range of temperature. For large particles Ums tended to increase with increasing temperature, while for small particles it decreased with increasing temperature. Gas viscosity should be taken into consider-ation for predicting Ums. A considerably greater effect of auxiliary to total air ratio, q/Qr, o n total minimum spouting velocity was found at elevated temperatures than at room temperature. At the maximum spoutable bed height, the value of Um/Umj was found to decrease with increasing temperature and to be smaller than unity at elevated temperatures. The McNab and Bridgwater (1977) expression correctly predicted the observed trends of Hm and worked reasonably well at high temperatures, although it was found to over-predict Hm at lower temperatures. Hm decreased with increasing temperature for all particle sizes, with a faster decrease for smaller particles. Fluidization in the annulus was never observed as the termination mechanism of spouting at high temperatures. iii Table of Contents Abstract ii List of Tables vii List of Figures viii Acknowledgement xii 1 Introduction 1 2 Literature Review . 7 2.1 Combustion in Spouted and Spout-fluid Bed 7 2.2 Hydrodynamics 10 2.2.1 Minimum Spouting Velocity 10 2.2.2 Maximum Spoutable Bed Height 13 2.2.3 Average Spout Diameter 15 3 Apparatus and Particle Properties 17 3.1 Experimental Equipment 17 3.1.1 Combustor . • . 17 3.1.2 Gas Flow and Coal Feeding System 19 3.1.3 Gas Sampling System . 21 3.1.4 Solid Sampling System 25 3.1.5 External Cooling System 25 i v 3.2 Experimental Instrumentation 28 3.3 Properties of Solids 31 3.3.1 Sand . . 31 3.3.2 Coal 32 3.3.3 Limestone 32 4 Experimental Technique and Problems Encountered 37 4.1 Initial Startup and Operation 37 4.2 Measurement Techniques 41 4.2.1 Temperature Measurement 41 4.2.2 Gas Concentration Measurement • • • • 41 4.2.3 Solid Sampling 43 4.3 Problems Encountered 45 4.4 Shutdown of the Unit 52 5 Temperature and Concentration Profiles 53 5.1 Temperature Profiles 53 5.2 0 2 and CO2 Concentration Profiles 56 5.3 SO2 Concentration Profiles 61 6 Combustion Performance 65 6.1 Combustion Efficiency 65 6.2 Sulphur Capture 70 6.3 NOx Emissions 73 7 Hydrodynamics 75 7.1 Minimum Spouting Velocity 75 7.1.1 Effect of Temperature and Particle Size 77 v 7.1.2 Comparision with Existing Correlations 77 7.1.3 Effect of Auxiliary Air 80 7.1.4 The Maximum Value of Um, 82 7.2 Maximum Spoutable Bed Height 84 7.2.1 Stability of Spouting 84 7.2.2 Effect of Temperature on Hm 86 7.2.3 Effect of Particle Diameter 91 7.2.4 Mechanism for Spout Termination 91 7.3 Average Spout diameter 92 8 Conclusions and Recommendations for Further Work 97 8.1 Conclusions . 97 8.2 Recommendations ; 99 Notation 101 Bibliography 105 Appendices 112 A Calbration Curves 112 B Derivation of T)SO2 115 C Experimental Data 117 v i List of Tables 3.1 Size distribution of silica sand 33 3.2 Size distribution of Minto coal 34 3.3 Analysis of Minto coal 35 3.4 Size distribution of Texada limestone 36 4.5 Chronology of combustion runs 46 7.6 Analysis of Highvale coal 76 C.7 Experimental data . 118 vu List of Figures 1.1 Schematic diagram of a spouted bed 2 1.2 Schematic diagram of a spout-fluid bed. a. with conical base distributor; b. with flat base distributor 5 3.3 Combustor 18 3.4 Overall schematic of equipment 20 3.5 Rotary Valve impeller 22 3.6 Flue gas sampling probe . 23 3.7 Continuous gas sampling system 24 3.8 Flue gas solid sampling probe 26 3.9 Cooling jacket . 27 3.10 Locations of thermocouples 29 3.11 Pressure tap 30 4.12 History of bed temperature and flue gas oxygen concentration (Run#880121, Ca/S = 2.25, H0 = 0.23 m) 38 4.13 Coal feed rate variation history during steady state operation 40 4.14 Thermocouple positions 42 4.15 Locations of gas sampling points in the combustor 44 4.16 Size distributions (Ca/S = 2.25, Bed material: limestone, H0 = 0.2 m) . 51 V l l l 5.17 Axial temperature profiles (Run#880216, q/Qr = 0.0, flue gas oxygen level: 11.2%, H0 = 0.23 m, U = 1.4 m/s, sand size: 1.4—2.0 mm, limestone si ze: 0.85 — 1.68 mm, coal (Minto) size: 0.85 — 1.68 mm, Ca/S = 2.5, m c = 0.905 kg/h) 54 5.18 Radial temperature profiles (Run#880216, q/Qr = 0.0, flue gas oxygen level: 11.2%, H0 = 0.23 m, U = 1.4 m/s, sand size: 1.4—2.0 mm, limestone s ize: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5, m c = 0.905 kg/h) 55 5.19 Axial temperature profiles (Run#880310, q/Qr — 0.25, flue gas oxygen level: 10.2%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 — 1.68 mm, coal (Minto) size: 0.85 — 1.68 mm, Ca/S = 2.5, mc = 1.214 kg/h) 57 5.20 Radial temperature profiles (Run#880310, q/Qr — 0.25, flue gas oxygen-level: 10.2%, H0 = 0.23 m, U - 1.85 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85—1.68 mm, coal (Minto) size : 0.85 — 1.68 mm, Ca/S = 2.5, m c = 1.214 kg/h) 58 5.21 Axial 02 and C02 concentration profiles (Run#880310, Tb = 840°C, q/QT = 0.25, flue gas oxygen level: 10.0%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4 — 2.0 mm, limestone siz e: 0.85 — 1.68 mm, coal (Minto) .size: 0.85 - 1.68 mm, Ca/S = 2.5, rhc = 1.214 kg/h) 59 5.22 Axial 02 and C02 concentration profiles (Run#880318, Tb - 840°C q/QT = 0.25, flue gas oxygen level: 3.4%, H0 = 0.23 m, U = 1.85 m/ s, sand size: 1.4 — 2.0 mm, limestone size : 0.85—1.68 mm, coal (Minto) size: 0.85—1.68 mm, Ca/S = 2.5, m c = 1.38 kg/h) 60 I X 5.23 Axial S02 concentration profiles (Run#880310, Tb = 840°C, q/QT = 0.25, flue gas oxygen level: 10.0%, H0 — 0.23 m, U = 1.85 m/s, sand size: 1.4 — 2.0 mm, limestone size: 0.85—1.68 mm, coal (Minto) size: 0.85 — 1.68 mm, Ca/S = 2.5, m c = 1.214 kg/h) 62 5.24 Axial S02 concentration profiles (Run#880318, Tb = 840°C, q/QT = 0.25, flue gas oxygen level: 3.4%, H0 = 0.23 m, U = 1.85 m/s, sand s ize: 1.4—2.0 mm, limestone size: 0.85 —1.68 mm, coal (Minto) size: 0.85 — 1.68 mm, Ca/S = 2.5, m c = 1.38 kg/h) 64 6.25 Packed bed reactor system 66 6.26 Effect of temperature on combustion efficiency (flue gas oxygen level: 3.25 - 3.45%, U=1.83 to 1.92 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85 — 1.68 mm, Ca/S = 2.5) . . 68 6.27 Combustion efficiencies obtained in the present work compare with those reported by Lim et al. (1988) 69 6.28 Effect of temperature on sulphur capture efficiency (flue gas oxygen level: 3.25 — 3.45%, U=1.78 to 1.95 m/s, sand size: 1.4 — 2.0 mm, limestone size: 0.85 - 1.68-mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5) 72 6.29 Effect of flue gas oxygen level and bed temperature on NOx emission; in flue gas. (Run#880403, U=1.85 m/s, sand size: 1.4 —2.0 mm, limestone s ize: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5) . 74 7.30 Effect of temperature and dp onT/ m s (Di = 15.9 mm) . . 78 7.31 Effect of temperature on UmS compared with correlations from the litera-tures (dp = 2.18 mm,.H0 = 0.233 m, D{ = 15.9 mm) 79 7.32 Effect of q/Qr on total minimum spouting velocity (dp = 1.54 mm, Ho = 0.23 mm) . 81 x 7.33 Effect of temperature and particle size on Um/Umf (Di — 15.9 mm, Dc = 156 mm, pp = 2660 kg/m3) 83 7.34 Effect of temperature and superficial gas velocity on spouting character-istics (H0 = 0.233 m, dp — 2.18 mm, Di = 15.9 mm, Dc = 156 mm, pp = 2660 kg/m3) 85 7.35 Effect of temperature and superficial gas velocity on spouting character-istics (H0 = 0.238 m, dp = 1.09 mm, Di = 15.9 mm, Dc — 156 mm, pp = 2660 kg/m3) 87 7.36 Effect of temperature and superficial gas velocity on spouting character-istics (H0 = 0.23 m, dp = 1.54 mm, Di — 15.9 mm, Dc = 156 mm, pp = 2660 kg/m3) 88 7.37 Effect of temperature and particle size on Hm 89 7.38 Effect of temperature on Hm compared with McNab and Bridgwater cor-relation (dp = 1.54 mm) 90 7.39 Effect of temperature on spout diameter (dp — 1.54 mm, H0 = 0.23 m, U/Ums = 1.10) a. Th = 23°C, Umt = 0.90 m/s; b. Tb = 420°C, Umt = 0.80 m/s; c. Tb = 500°C, U^, = 0.76 m/s; d. Tb = 600°C, Umt = 0.82 m/s.) 93 7.40 Effect of temperature on spout diameter (dp — 2.18 mm, H0 = 0.233 m, U/Ums = 1.10) a. Tb = 23°C, Umt = 1.14 m/s; b. Tb = 230°C, Ume = 1.15 m/s; c. Tb = 424°C, Umt = 1.15 m/s; d. Tb = 534°<7, Umt = 1.17 m/s.) 95 A.41 Calibration curve for the coal feeder (Mixture of Minto coal and Texada limestone, coal size: 0.85 — 1.68 mm, limestone size: 0.85 — 1.68, Ca/S = 2.5, kg limestone/kg coal — 0.571 113 A.42 Calibration curve for the S02 analyser 114 xi Acknowledgement I would like to express my appreciation to Dr. J.R. Grace and Dr. C.J . Lim for their excellent supervision and support. I am also indebted to many others who helped during the course of the work: To the Chinese government which provided financial support in the form of scholar-ship. To Mr. Pingnan Shi, Ms. Peng Hui, Mr. Xiaoming Cai, Dr. Jesse Zhu, Ms. Mei Mei, Mr. Guangxi Wu, and many other friends for their invaluable encouragement, support and help. To my parents for their encouragement and support. To my fiancee Miss Zhuang Qun for her encouragement, patience and understanding from thousands miles away. To Prof. Kefa Cen for his useful discussions, suggestions and encouragement. xn Chapter 1 Introduction Fluidization has been widely used as a gas-solid contacting technique in industry during recent decades. However, most applications have been for relatively fine particles (dp < 1mm), in order to achieve good quality fluidization (Geldart, 1986). Spouted beds provide an alternative technique for liandhng coarse particles (dp > 1mm) (Mathur and Epstein, 1974). Figure 1.1 illustrates schematically a typical cylindrical spouted bed column with a conical base. Under the condition of stable spouting, the bed can be characterized b3r two distinct zones, the dilute phase central core called the spout and the surrounding dense phase called the annulus. Gas is injected vertically from the bottom of the bed through the centrally located small opening called the orifice. Particles are entrained in the spout by the gas at high velocity, and then penetrate somewhat above the bed level in a region called the fountain where they fall back onto the annulus surface. In the annulus thej' travel downward uniformly as a moving packed bed. Thus a spouted bed is a composite of a dilute central gaseous spout carrying the solids upward and a dense downward-moving annulus with a countercurrent percolation of air. For combustion of coal spouted beds share most advantages with fluidized beds, in-cluding: (a) Sorbent material like limestone or dolomite can be added with the coal for sulphur capture, reducing the emission of SO2 • 1 Chapter 1. Introduction 2 FOUNTAIN BED SURFACE SPOUT ANNULUS SPOUT - A N N U L U S INTERFACE CONICAL B A S E FLUID INLET Figure 1.1: Schematic diagram of a spouted bed Chapter 1. Introduction 3 (b) The low bed operating temperature (typically 800 — 950°C) of combustion leads to low NOx emissions, lower cooling tube corrosion, and no ash slagging. (c) Because of the large amount of heat storage in the bed and long residence time of particles, resulting in substantial preheating of freshly fed material, it is possible to burn low heating value coals with high ash and moisture contents (Khoshnoodi and Weinberg, 1978; Arbib et al., 1981; Arbib and Levy, 1982; Lim et al., 1984; Lim et al., 1988) (d) Heat transfer coefficients are favourable between bed and cooling surface, leading to compact equipment (Lim et al., 1984). (e) Fine coal grinding is not needed. In addition, spouted beds have a number of advantages over conventional bubbling beds, in particular: (a) Spouted beds are capable of processing coarser particles than fluidized beds. (b) The pressure drop across spouted beds is typically only about two thirds that of fluidized beds. This can result in considerable savings in operating costs. (c) Fuel particles can be fed with the spouting air without the need for overbed feeding or extra air for feeding from below. (d) Higher attrition derived from the spout (Mathur and Epstein, 1974) provides benefits for the combustion of agglomerating solid fuels (Watkinson et al., 1983) which can lead to defluidization in fluidized beds. The attrition may also expose fresh surfaces, thereby improving sorbent utilization in capturing sulphur. Chapter 1. Introduction 4 (e) Turndown and load following are claimed (Shirley and Litt, 1987) to be superior in spouted bed units. The minimum spouting velocity increases with the bed depth and not only with the particle size and density (Mathur and Epstein, 1974) as the minimum fluidization velocity does (Geldart, 1986). Therefore, good spouting can be maintained over a wide range of flow rates by changing the amount of material in the bed. One disadvantage of a spouted bed is that despite its zone of high activity, much of the bed (annulus region) is relatively stagnant or at least moves slowly. In a large vessel where gas/solids contact is important or in systems where particles may fuse, this slow-moving zone with little gas flow can be a serious liability. Various modifications to standard spouted beds have been made to overcome this problem (e.g. see Kono, 1981; Mathur and Epstein, 1974; Epstein and Grace, 1984). One of them is the spout-fluid bed (SFB), in which auxiliary air is supplied through a flat or conical perforated distributor as shown in Fig.1.2a and 1.2b, in addition to supplying spouting fluid through a centrally located opening. By providing fluidizing air in the annulus region, any tendency to form a stagnant zone is diminished. This results in improved overall gas/particle contact, while retaining the benefits of the spout zone. The SFB shows better solids mixing and annular solid-fluid contact than standard spouted beds (Chatterjee, 1970; Madonna et al., 1980) and less tendency for particles to agglomerate (Kono, 1981). In order to promote the application of spouted beds or SFBs to combustion or gasifi-cation of coal, a good understanding of their performance at high temperature is required. However, until now, almost no information exists about their hydrodynamic behaviour in the range of 700 — 900°C. It is obvious that one of the key advantages of spouted beds or SFB is the high attrition in the spout, which may result in higher sorbent utilization and reduced discharge of course material (Shirley and Litt, 1987). However, no experimental Chapter 1. Introduction F O U N T A I N • A N N U L U S S P O U T A U X I L I A R Y F L U I D I N L E T S P O U T I N G F L U I D I N L E T Figure 1.2: Schematic diagram of a spout-fluid bed. a. with conical base distributor with flat base distributor. Chapter 1. Introduction 6 data have been obtained on sorbent utilization. The main objective of the present research is to study the combustion performance of spouted and spout-fluid beds by burning Minto coal, which has a high sulphur content and a high tendency to agglomerate, with limestone addition for sulphur capture. The base conditions used in this work are: • Limestone size 0.85-1.68 mm • Coal size 0.85-1.68 mm • Sand size 1.40-2.00 mm Bed height Flue gas O2 level • Bed temperature Ca : S molar ratio 0.23 m ~ 3.5% ~ 840°C 2.5:1 Auxiliary to total air ratio 0.25 Superficial gas velocity 1.85 m/s The secondary objective is to study the hydrodynamics of the spouted bed and spout-fluid bed at elevated temperature up to 880°C. The hydrodynamic parameters studied are: minimum spouting velocity, spouting stability, and maximum spoutable bed height. Chapter 2 Literature Review 2.1 Combustion in Spouted and Spout-fluid Bed The spouted bed technique was developed by Mathur and Gishler (1955) for drying wheat. Since then, as an alternative to fluidized beds for handling coarse particles, spouted beds have been used for a large variety of processes such as drying of granular solids, tablet coating, solids blending, gasification of coal, and coal combustion. A complete review of spouted bed technology was presented in the monograph by Mathur and Epstein (1974). More recent reviews are given by Epstein and Grace (1984) and Bridgwater (1985). The application of spouting to combustion did not come until Khoshnoodi and Wein-berg (1978) used a spouted-bed combustor of 40 mm diameter to burn gaseous mixtures of methanol and air. They found that reactants of heat content equivalent to less than half the normal limit of flammability can be burned in such systems. This was because the spouted bed worked as a counter-current heat exchanger, leading to preheating of the gaseous mixture before it reached the foutain. It was observed that under certain conditions, presumably when viscosity increased sufficiently due to temperature rise, a second fluidized bed formed above the bed surface. This offers a means of dealing with fuels which contain a volatile fraction. Volatiles are apt to be driven off unless introduced near the base of the bed and the second bed provides a convenient method of bringing combustion to completion. 7 Chapter 2. Literature Review 8 A study of hydrodynamics of spouted beds under combustion conditions was con-ducted by Arbib et al. (1981). Two quartz columns of 53 mm and 38.8 mm diameters, particles in the 0.5 to 2 mm range, and methane/air reactants were employed. It was found that combustion reduced the minimum spouting velocity, and increased particle circulation rate, pressure drop and spout height in comparison with operation at identical conditions without combustion due to the higher viscosity and linear velocity of the hot gas. Also, the sudden rise in presssure drop which accompanies the spout collapse in the cold case is smoothed by combustion in pass through the pulsating and fluidized mode. Palsating combustion was more frequently observed with small than with large particles. The leanest mixture burnt at approximately 2/3 of the normal limit of flammability. Using a spouted bed combustor of 48 mm diameter, Arbib and Levy (1982) burned a variety of low calorific value liquid fuels, such as methanol-water mixtures and oil-water emulsion, using irregularly shaped tabular alumina particles as bed material. It was proved that a spouted bed combustor has the ability to burn low-quality fuels, which are incombustible by other methods without any auxiliary fuel. A study of a spouted bed combustion of propane/air through direct visual observation in a half cylindrical bed (90 mm ID x 500 mm height, flat base), using river sand and molding sand (0.85 - 1.0 mm), was reported by Khoe and Weve (1983). It was found that for a given propane to air ratio, nozzle size and bed height, the spouted bed exhibited the following regimes as the inlet mass flowrate was increased: (1) Internal spout. (2) Bubble eruption. (3) Regular pulsing spout. (4) Stable spouting. Spouted bed combustion of solid fuels was started at the University of British Co-lumbia in 1980. Low-grade solid fuels, including washery rejects with up to 89% ash and heating values as low as 3500 kJ/kg have been successfully burned in a 0.3 m diameter refractory-lined spouted-bed combustor (Lim et al., 1984). The overall bed-to-immersed tube heat transfer coefficient in a spouted bed was found to be close to that found in Chapter 2. Literature Review 9 fluidized bed combustors. Addition of auxiliary air, introduced to produce the spout-fluid bed combustion mode, resulted in axial temperature profiles which were more uniform than in a standard spouted-bed combustor. A study of coal combustion was carried out by Zhao et al. (1987a) in a 0.15-m diameter half-column spout-fluid bed combustor, in which a sub-bituminous Alberta coal was burned in inert beds of sand. It was found that high temperature promoted spouting stability leading to earlier appearance of pulsatory spouting, the jet-in-fluidized-bed regime, and slugging than for room temperature operation. While radial temperature profiles below the bed surface were quite uniform, axial profiles showed a significant temperature jump in the fountain region above the bed surface. Axial profiles of oxygen concentration fell abruptly at the bed surface due to rapid burning there. Addition of auxiliary air along the conical base of a spout-fluid bed led to more uniform axial temperature and oxygen concentration profiles. Coal burnout times in spouted and spout-fluid bed were measured by Zhao et al. (1987b) using small batches of Forestberg coal particles of 0.55 to 2.2 mm in diameter dropped into a semi-cylindrical column of diameter 0.15 m with inert 1.8 mm sand making up the vast majority of the bed particles. It was found that burnout times were in the range 20 to 60 s, increasing with increasing particle size, and significantly lower than for particles of comparable size in fluidized bed burnout tests. They also developed a simple model, which gave good agreement with the measured values, for predicting burnout times in spout-fluid beds. Lim et al. (1988) burned low-grade coal rejects in three different contacting modes: fluidized bed combustion, spouted bed combustion and spout-fluid bed combustion. The overall combustion behaviour was found to be broadly similar in the three cases for their 0.3 m diameter combustor. However, the spout-fluid bed tended to give some-what higher combustion efficiencies at low temperatures, greater temperature uniformity Chapter 2. Literature Review 10 and improved bed-to-surface heat transfer compared with the other modes of operation. Combustion efficiencies were over 90% providing that bed temperature was higher than about 870°C and solids captured in the primary cyclone were recycled to the bed. The particles tending to segregate, leaving char particles at or near the bed surface. They also pointed out that recycling of fines captured in cyclones was helpful in increasing the overall combustion efficiency. 2.2 Hydrodynamics 2.2.1 Minimum Spouting Velocity Spouted Beds The minimum fluid velocity at which a bed will remain in the spouted state is called the minimum spouting velocity ( t / m a ) . It is fount experimentally by decreasing the fluid velocity until the spout collapses to give a static bed in its random loose-packed condition. Uma depends on solid and fluid properties, bed geometry and bed depth. For a given material and fluid properties, it increases with increasing bed depth, increasing fluid inlet diameter, and with decreasing column diameter. However, for a large bed as reported by Peterson (1969 - 1972), Umil has a tendency to increase with increasing column diameter. More than a dozen correlations have been proposed for predicting Ums (Mathur and Epstein, 1974). The most widely used one seems to be the empirical equation of Mathur The above equation, developed with the help of dimensional analysis, was derived from results for a number of closely sized materials spouted in 76 to 305 mm diameter columns size and density of inert bed materials played a significant role, denser and coarser inert and Gishler (1955). Chapter 2. Literature Review 11 using air as well as water as the spouting fluids. Over the years, it has proved to be valid for a much wider range of conditions (Mathur and Epstein, 1974), which include not only a larger variety of solid materials, closely sized as well as with a wide spread, but also column diameters up to 610 mm. However, the absence of fluid viscosity in the equation has been questioned (Charlton et al, 1965), and the equation has not been very successful for large columns (Lim and Grace, 1987) or at high temperatures (Wu, 1988). Although no theoretical basis for Eq.(2.1) was claimed by its authors, Ghosh (1965) subsequently derived a similar equation based on a momentum exchange between the entering fluid and the entrained particles. U m ' \ o i J n JI n J" 2gH(pp-Pf) 3fcl.Dc Z V \ P f Wu et al. (1987) correlated their experimental data carried out in a 156 mm diameter half column using sand as bed material at temperatures up to 420°C with an empirical equation: ume = lo.e^] 1 0 5^] 0- 2 6^^]- 0- 0 9 5!^^] 0- 2 5^^ (2.3) t>c Pf It is claimed by its authors that H and (pp — pf)/pj should not be grouped as one single parameter and that Equation (2.1) overestimates the effect of bed height on Ums- Similar findings were reported by Manurung (1964). Grbavcic et al. (1976), based on their flow model, derived the following correlation for Ums: v ~a* H „ = l - [ l - - 5 r - ] 3 (2.4) I-a, 1 Hn_ where as is defined as the ratio of the area of the spout to that of the bed. Since at is much smaller than 1 in most cases, Equation (2.4) can be further simplified to Ums = UmJ[l - [1 - i 3 ] (2.5) Sim Chapter 2. Literature Review 12 Spout-fluid Beds A number of workers have investigated the effect of auxiliary air on the minimum total fluid flow required to give a spout-fluid bed. For spouting with aeration (UAH < ^m/)> Chatterjee (1977) found that, with the introduction of auxiliary air, the minimum total fluid flow (QT,msa) remains nearly constant and almost equal to the minimum spouting flow, i.e. QT,msa ^ Qm, (2.6) However, a correlation obtained by statistical analysis of experimental data by Dumis-trescu (1977) showed that UT,msa = 2 7 . 7 - ^ + Umt (2.7) where C/ m j was estimated from Equation (2.1). Thus the minimum total fluid superficial velocity with aeration {Ux^msa) is always greater than the minimum spouting velocity. As pointed out by Sutanto et al. (1985), based on their experimental results, the minimum total fluid fiowrate for spouting with aeration and spout-fluidization is always greater than the minimum spouting velocity. They obtained ( | P ) = * [ l - ( ^ ) ( § ^ ) ] (2-8) The empirical constant K varied between 1.27 and 1.61 for different system investigated by the authors. Based on the work.of Vukovic et al. (1972), Littman et al. (1974, 1976) showed that for any given bed height, H < Hm, the minimum spout-fluid fiowrate varies between Qms and Qmf (for Qmf > Qmt)- They obtained ^ = [ l - ^ - M 3 ] [ l - ^ ] ( 2 . 9 ) However, for a deeper bed such that Qms > Qmf, the above equation is no longer valid. Chapter 2. Literature Review 13 For spout-fluidization (UAH > Umf), Chatterjee (1974) employed the equation: UT,m.f = U m t + Umf(l - <f>) (2.10) The function <fr was determined empirically to be <f> = 0 . 2 0 d p - ° - 3 2 0 D , 0 - 2 3 5 ^ 0 1 6 0 (2.11) The value of Um, was obtained from the Mathur and Gishler correlation, Eq.(2.1). The Maximum Value of Ums The value of Ume at the maximum spoutable bed depth is termed Um, the maximum value of the minimum spouting velocity (Becker, 1961). For many materials, Um is expected to coincide with the minimum fluidization velocity since beyond Hm a spouted bed transforms into a fluidized bed. Experimental data show that Um often exceeds Umf. In the case of sand (dp = 0.42 — 0.83 mm), Um is approximately equal to Umj, while it is 33% higher than Umf for wheat ( dp = 3.2 x 6.4 mm) and 45% for semicoke (dp = 1 — 5 mm)(Mathur and Gishler, 1955; Dumitrescu and Ionescu, 1967). Values oiUm exceeding Umf by 10 — 33% have been reported by Becker (1961) for a variety of uniform size materials. Pallai and Nemeth (1969), based on their experiments, suggested that Um ~ 1.5Umf. On the other hand, the geometry of the spouting vessel also affects the ratio Um/Umf. For a fixed Di/Dc ratio, Um increases with increasing column diameter, while for a fixed value of Dc it increases with increasing orifice diameter. Thus, in general •—— = b = 1.0 to 1.5 (2.12) Umf 2.2.2 Maximum Spoutable Bed Height The maximum spoutable bed height, Hm, is the maximum height at which steady or stable spouting can be maintained. Mathur and Epstein (1974) suggested three distinct Chapter 2. Literature Review 14 mechanisms for spout termination beyond i f m . 1. Fluidization of Annular Solids 2. Choking of the Spout 3. Growth of Instability at the Spout-Annulus Interface At the maximum spoutable bed height, equation (2.1) becomes U™ = [ 7 r ] [ ^ ] 1 / 3 . 2gHm(pp- pf) As mentioned in the previous section, Um is closely related to Umf. Umf, on the other hand, can be estimated from the Ergun (1952) equation with substitution of the empirical approximation of Wen and Yu (1966), i.e. 1/^e^y = 14 and (1 — em/)/^>2e^y = 11, which yields Remf = d p U m f P f = 33.7(Vl + 35.9 x 10- 6 Ar - 1) (2.13) f1 where Ar = dp3(pp-pf)gpf/u2 (2.14) When equations (2.1a), (2.14) and (2.12) are combined to eliminate Um and £/ m / , the result is £> 2 r J D C l , , , r 568& 2 H m = [ ^ ] [ ^ ] 2 / 3 [ ^ T 3 ( V l + 3 5 - 9 x 1 0 ~ 6 A r ~ 1 ) 2 ( 2 - 1 5 ) McNab and Bridgwater (1977) found that Equation (2.15) gave the best fit to existing experimental data for i / m in gas-spouted beds with b = l . l l , although there was consid-erable scatter. Differentiating equation (2.15) with respect to Ar, after substituting for dp from equa-tion (2.14) and setting d i f m /d (Ar) equal to zero, yields the critical value, Ar=223,000, or ( f r U = 6Q-.61 n i n i L \ ] n ] 1 / 3 ( 2 - 1 6 ) Chapter 2. Literature Review 15 below which Hm increases with dp and above which Hm decreases as dp increases. For gas spouting (dp)^ is typically in the range 1.0- 1.5 mm. Wu et al. (1987) modified the equation of Malek and Lu (1965) to give the following empirical dimensionless correlation for Hm: ^ = 3 3 6 ^ ] ° - 7 5 [ f : ] 0 - 4 ^ ] 1 - 2 ^ 2 ( 2 - 1 7 ) For non-spherical particles, Morgan and Littman (1982) suggested the following ex-pressions: H^Di n o i 0 , 5.13 x l O - 3 2.54 x l O " 5 _ _ 0.218 + +. for A > 0.014 = 175(4-0.01) for 0.010 < A < 0.014 (2.18) where A = , (2.19) (pp-pf)gDi and Umf and Ut are the minimum fluidization and terminal settling velocity, respective^. Wu et al. (1987) compared the three existing Hm correlations given above with their experimental data at 420°C. The results showed that the McNab and Bridgwater (1977) equation gave better predictions than the other two. The modified Malek and Lu (1965) equation predicted the correct trend of Hm with increasing temperature, but it underpredicted Hm at elevated temperatures. Similar results were observed by Zhao et al.(1987), who obtained several experimental values of Hm at temperatures up to 640°C. 2.2.3 Average Spout Diameter The diameter of the spout is an important parameter for determining the flow distri-bution between spout and annulus (Epstein and Grace, 1984). A number of equation Chapter 2. Literature Review 16 are available for estimating the average spout diameter, Da (Mathur and Epstein, 1974; Littman, 1982; Wu et a l , 1987). Based on a force balance, Bridgwater and Mathur (1972) derived the following theo-retical equation after a number of approximations: /-r0.5 T-)0.75 JP. = 0.384[ 0 2 5 c ] (2.20) Pb where all variables were expressed in SI units. McNab (1972) correlated experimental data empirically by means of the dimensional correlation /O0.49 r)0.68 D. = 2.0[ ^ - ] (2.21) Pb over a wide range of experimantal data, with all variables expressed in SI units. Based on the theoretical model of Bridgwater and Mathur (1972), Wu et al. (1987) developed the following dimensionless correlation using their experimental data at tem-peratures up to 420°C: /^r0.433 r)0.583: 0.133 D - = 5 - 6 i r , c ] (2-22) {pbpfg)0-283 Wu et al. (1987) reported good agreement between their experimental and calculated values whereas Equation (2.21) overestimated the effect of temperature on the average spout diameter. Chapter 3 Apparatus and Particle Properties 3.1 Experimental Equipment 3.1.1 Combustor Experiments were carried out in a half-cylindrical spout-fluid bed combustion column with a half conical base. This was originally designed and set up by Zhao (1986). The combustor, constructed of stainless steel, is shown in Fig. 3.3. It is 1.06 m in height and 152 mm in I.D. with a wall thickness of 6.4 mm. The solid discharge lines are located at two different levels: 0.25 and 0.35 m above the inlet orifice. The half column is also furnished with seven temperature measuring ports along the curved back of the combustor, at intervals of 101.6 mm. The first port is located 178 mm above the central air inlet orifice. The half conical base, which has an included angle of 60° and is 0.13 m high, is described in detail by Zhao (1986). The diameter of the semi-circular orifice used in this work was 15-9 mm. The front stainless steel panel of the column has four rectangular openings on which either quartz glass windows or stainless steel plates can be mounted for direct -visual observation or full combustion runs, respectively. In the case of combustion runs, two pressure taps were mounted in the centre of the panels 10 mm and 345 mm above the central orifice to allow the pressure drop across the bed to be determined. For the combustion runs, the exterior of combustor was covered by ceramic fibre insulation which 17 Chapter 3. Apparatus and Particle Properties 18 8 Figure 3.3: Combustor 1. Spouting flow l i n e 2. Gas chamber and d i s t r i b u t o r 3. A u x i l i a r y flow l i n e 4. Solids discharge l i n e 5. Measuring port 6. Half-column 7".-Off-gas l i n e 8. Port f o r gas sampling probe 9. Front panel 10. Quartz glass window 11. Pressure taps ( A l l dimensions i n t h i s thesis are i n mm.) Chapter 3. Apparatus and Particle Properties 19 could be removed in sections to help control the temperature. The main reason for using a half-column spouted bed is that visualization of the inter-nal behavior and measurement of some hydrodynamic data become possible. However, the reliablity of data obtained from half column beds has been questioned and investi-gated by many workers. Mathur and Gishler (1955) showed that particle velocities at the wall and pressure drop profiles obtained in half columns were similar to those in full columns. Whiting and Geldart (1979) and Geldart et al. (1981) reported that hydrody-namic variables such as Umi and Hm measured in half and full columns were nearly the same. Lim (1975) indicated that the flat wall in a half column has a negligible effect on spout shapes. However, Rovero et al. (1985) showed that the frictional effect caused by the front wall cannot be overlooked when measuring particle velocities. The half column used by Zhao (1986) was modified for the present work by adding cooling air jackets at the curved back external surface of the column as described below. 3.1.2 Gas F low and Coa l Feeding System A schematic diagram of the overall equipment layout is given in Fig. 3.4. Coal premixed with limestone to a desired Ca/S ratio and stored in the coal hopper, was fed to the rotary valve coal feeder. It then dropped to the main spouting air pipe where it was pneumatically transported to the bottom of the combustor by the spouting air. For start-up this air could be preheated by three electric heaters each with power of 3.6 kW. Auxiliary air was introduced to the plenum chamber around the semi-conical base, and then went through the perforated distributor and the supporting screen to create a spout-fluid bed. Hot combustion gas, carrying some entrained coal fines, and limestone fines left the top of the combustor and passed through the cyclone to a scrubber. The cyclone, which is 75 mm in diameter (inlet pipe diameter is 25 mm) with a cylinder height of 125 mm and Chapter 3. Apparatus and Particle Properties 20 Figure 3.4: Overall schematic of equipment 1. Vacuum.pump 2. Fiowrate 3. S o l i d f i l t e r 4. Cyclone 5. ash receiving container 6. preheater 7. Coal hopper 8. Rotory valve 9. Gear reduction 10. Motor 11. Motor c o n t r o l l e r 12. Combustor 13. Scrubber 14. Manometer 15. Pressure taps Chapter 3. Apparatus and Particle Properties 21 cone height of 177 mm, is used for recovering particles which escape from the combustor. The flue gas scrubber is a packed countercurrent spray column 570 mm in diameter used for cleaning and cooling the off-gas. The rotary valve coal feeder, consisting of a brass casing and a rubber impeller, was connected to a variable speed motor by a speed reducer, the reduction ratio being 100:1. A small copper tube used for the pressure neutralization was connected from the top of the coal hopper to the coal feeder outlet tube. Details of the impeller are provided in Fig. 3.5. The coal hopper is 508 mm I.D. by 508 mm high with a cone height of 254 mm and included cone angle of 90°. This gives a storage volume of 0.07 m 3 , sufficient for feeding over a period of 40 to 50 hours. 3.1.3 Gas Sampling System The in-combustor gas sampling probe is described in detail by Zhao (1986). By rotating and moving the probe from the top of the combustor, the gas concentration at different positions inside the combustor could be measured. In order to be able to measure flue gas concentrations, a new gas sampling probe was designed and mounted in the outlet pipe leaving the combustor, as shown in Fig. 3.6. As illustrated schematically in Fig. 3.7, the continuous gas sampling system consists of two sampling probes, two cooling coils, a gas drying unit (60 mm I.D. by 200 mm high packed column filled with magnesium perchlorate [Mg(ClO4)2) powder), three rotame-ters, various analysers (described below) and a vacuum pump (Model 8730, E X T E C H ) . Gas samples were drawn through a stainless steel filter mounted at the tip of the probe to remove particulate material down to a size of 60 fim. This gas was passed through a cooling coil to be cooled nearly to room temperature, and then through the drying unit where most of the moisture was removed. The dry gas leaving the drier then went to the Chapter 3. Apparatus and Particle Properties 22 10 036 <P48 Figure 3.5: Rotary Valve impeller Samp le G a s F|ue G a s Figure 3.6: Flue gas sampling probe 1. Cooling c o i l 2. Cap 3. Sampling Tube'41-Porous s t a i n l e s s s t e e l f i l t e r 5. Flue gas pipe l i n e Figure 3.7: Continuous gas sampling system I. S0 a analyser 2. 0»analyser 3. C02/C0 analyser 4. CH4 analyser 5. Pump 6. N0X analyser 7. Rotameter 8. Gas drying unit 9. Cooling c o i l s 10. Sampling probes II. Combustor to Chapter 3. Apparatus and Particle Properties 25 following three lines: (1) to CH$ analyser and then CO2/CO analyser and finally through the continuous 0 2 monitor; (2) to the SO2 analyser, which included a small pump inside, with nitrogen dilution gas supplied from a cylinder; (3) to the NOx analyser followed by a vacuum pump. The method of isokinetic gas sampling, by which the sampling velocity is equal to the gas velocity at the sampling position, was used to measure gas concentrations inside the combustor (see Zhao (1986) for detailed estimation). In order to obtain a desired overall bed pressure so that the sampling flow rate would be enough for isokinetic sampling, an orifice was installed in the flue gas outlet pipe, at the inlet of the scrubber. The sampling rate was controlled by valves and monitored by rotameters as shown in Fig. 3.7. 3.1.4 Solid Sampling System Most of the ash leaving the" combustor was collected at the bottom of the cyclone by a conical ash container of bottom diameter 127 mm and height 178 mm shown in Fig. 3.4. The small amount of ash not captured by the cyclone was sampled by an isokinetic sampling probe illustrated in Fig. 3.8 and then collected by a ceramic filter of 45 mm I.D. and 25 mm length with a perforated bottom covered by filter paper, as shown in Fig. 3.4. The gas samples containing elutriated solid particles were drawn by a vacuum pump (Model DOA-V110-FS,WAINBEE), and the fiowrate was monitored by a rotameter. 3.1.5 External Cooling System A half-cylindrical cooling jacket was designed and mounted at the back of the combustor so that both bed temperature and flue gas oxygen level could be controlled. In the previous work by Zhao (1986), very high excess air levels (e.g. 100% or more) had to be used in order to keep the temperature to reasonable levels. As illustrated in Fig. 3.9, the jacket consists of four separate sections. The top three Chapter 3. Apparatus and Particle Properties A Flue gas Figure 3.8: Flue gas solid sampling probe 1. Flue gas pipe .2. s o l i d sampling probe Chapter 3. Apparatus and Particle Properties 27 Chapter 3. Apparatus and Particle Properties 28 sections have a height of 101.6 mm each, while the bottom one was 63.5 mm high. The width of the channel in each section was 6.35 mm. In order to create turbulent flow in the jacket which would increase the coolant-to-wall heat transfer coefficient, and also increase the surface area, the gaps in the jacket were filled with steel wool. Either water or air at room temperatures could be used as a cooling fluid. The overall fiowrate of the cooling fluid was monitored by a rotameter, and the fiowrate in each section was controlled by valves installed at the inlet tubes as shown in Fig. 3.9. Thermocouples were mounted in both the inlet and the outlet tubes. 3.2 Experimental Instrumentation Temperatures were measured by Chromel-Alumel K type thermocouples connected to a digital temperature display through a control board. Seven thermocouples located at different axial positions could be moved radially from the back of the combustor to five different radial positions at each level. These various positions are shown in Fig. 3.10. The coal feed rate was monitored by a digital display percentage speed controller connected to the variable speed monitor as shown in Fig. 3.4. The calibration curve is provided in Appendix A. All air flowrates were controlled and monitored by rotameters. A manometer connected to two pressure taps as shown in Fig. 3.4 was used to de-termine bed pressure drops. A 1.6 mm diameter steel bar was inserted in each presssure tap and could be moved horizontally to prevent the tap from being plugged by particles as shown in Fig. 3.11. ] Gas concentrations were determined and displayed continuously by the following gas analysers: (1) S02 analyser (Model 460P, E X T E C H ) . The NOVA electrochemical type S02 sensor was used to produce a small electrical output signal directly proportional to the Chapter 3. Apparatus and Particle Properties 29 - 0 No. Axial Position Z 7 788 6 686 5 584 4 483 3 38 1 2 • 2 80 1 1 78 No. Radial Position R E 3 D 16 C 36 B 56 A 76 Figure 3.10: locations of thermocouples Chapter 3. Apparatus and Particle Properties Figure 3.11: Pressure tap Chapter 3. Apparatus and Particle Properties 31 SO 2 level detected. The SO2 in the sample caused a reaction with an electrolyte within the cell to produce this output signal which was amplified and then displayed on the digital meter. The range of the display was 0-500 ppm. Because of the high sulphur content of the Minto Coal, the sampling gas was diluted by N2 before passing through the analyser. The calibration curve of the SO2 digital display is given in Appendix A. (2) 02 meter (Model 715, B E C K M A N ) . Oxygen was sensed utilizing TAI's patented Micro-fuel Cell which consumed 02 in the sampling gas and generated a proportional electric current. This was then amplified and read out on a built-in meter. The outputs could be selected over the range 0 — 5%, 0 — 10%, or 0 — 25% 02. (3) CO2/CO analyser (Model 732, W H I T T A K E R ENVIOR). Infrared spectral ab-sorption was used in the C02/CO analyser. The range of C02 display was 0 — 20% CO2, while two ranges, 0 — 0.2% and 0 — 0.5% CO could be selected for the CO display. (4) CH4 analyser (Model 730, W H I T T A K E R ENVIOR). Infrared spectral absorption was used in the CH4 analyser. The range was 0-1000 ppm CH4. (5) NOx analyser (Model 8840, E X T E C H ) . The NOx monitor was a gas-phase chemi-luminescence-detection device which performed a continuous dry analysis of NO, N02, and NOx, with variable ranges up to 0-500 ppm. 3.3 Properties of Solids 3.3.1 Sand Ottawa sand was used as bed material in this study. The sand, as received, had a wide size range and was screened to the desired size range before it was used. The mean particle diameter of the sand was determined from sieve analysis as 1 (3.23) S ( x t / d P i ) Chapter 3. Apparatus and Particle Properties 32 where x± is the weight fraction of particles with an average adjacent screen aperture size of dp.. The size distribution of the sand is presented in Table 3.1. The density of sand was measured by water displacement. The particles were first coated with a water sealant before the measurement because the sand particles may be permeable to water. Three measurements of the density were made and the average value, given in Table 3.1, was then used. The bulk density of loosely packed sand was determined by the following procedure. A graduated cylinder was partially filled with a known weight of sand and the top was then covered. The cylinder was inverted and returned quickly to the original upright position. The volume occupied by the sand after this loosening procedure was recorded and used to calculate the bulk density. Such tests were repeated several times to produce an average value. The loosely packed bed voidage was determined from the particle density and bulk density. 3.3.2 Coal A sub-bituminous Minto coal from New Brunswick was used as the fuel throughout the sulphur capture and combustion efficiency runs. The size distribution of the coal is given in Table 3.2. The proximate and ultimate analysis and calorific value of the coal, as determined by the Department of Energy, Mines and Resources Canada, are provided in Table 3.3. 3.3.3 Limestone Texada limestone (97% CaCOs, 1% MgCOz) was used as sorbent for SO2 removal in this study. The particle size distribution appear in Table 3.4. Chapter 3. Apparatus and Particle Properties 33 Table 3.1 Size D i s t r i b u t i o n of S i l i c a Sand Sieve Size (mm) Wt. % 1.68 - 2.00 42.33 1.40 - 1.68 32.59 1.18 - 1.40 22.09 1.00 - 1.18 2.50 0.85 - 1.00 0.31 0.71 - 0.85 0.19 Mean d = 1.56 mm Density = 2661 kg/m Chapter 3. Apparatus and Particle Properties 34 Table 3.2 Si2e D i s t r i b u t i o n of Minto Coal Sieve Size (nun) wt. % 1 .18 - 1.68 25.89 1.00 - 1.18 30.82 0.85 - 1.00 19.95 0.71 - 0.85 16.84 0.51 - 0.71 5.49 0.35 - 0.50 0.69 < 0.35 0.32 Mean d = 0.99 mm Density = 1267 kg/m Chapter 3. Apparatus and Particle Properties Table 3.3 Analysis of Minto Coal Proximate Analysis (Wt. %) Moisture 1.53 Ash 18.78 V o l a t i l e 32.91 Fixed Carbon 46.78 Ultimate Analysis (Wt. %) Carbon 63.78 Hydrogen 4.13 Nitrogen 0.67 Sulphur 7.09 Oxygen (by d i f f . ) 4.02 Ash 18.78 Moisture • 1.53 C a r l o r i f i c Value: 27,310 kJ/kg Chapter 3. Apparatus and Particle Properties 36 Table 3.4 Size D i s t r i b u t i o n of Limestone Sieve Size (mm) Wt. % 1.40 - 1.68 9.07 1.18 - 1.40 20.62 1.00 - 1.18 24.97 0.85 - 1.00 27.43 0.71 - 0.85 .15.25 0.50 - 0.71 2.67 < 0.50 0.0 Mean d = 1.02 mm Density = 2601 kg/m Chapter 4 Experimental Technique and Problems Encountered 4.1 Initial Startup and Operation Before start-up of the equipment, the Minto coal was mixed with limestone to a desired Ca/S ratio. Approximately 30 kg of the mixture was poured into the coal hopper be-fore the operation started. Other preparatory action included cleaning of the combustor, cleaning of the pressure taps, cleaning of the cyclone and isokinetic solid sampling line, changing the solid filter and the material in the gas drying unit, checking the thermo-couples,or and th testing leakage in the combustor and the electric heater section, and calibrating the feeder system for the particular solid/feed mixture. Start-up of the combustor was achieved by first turning on the spouting air at a level sufficient to prevent particles from settling down in the inlet pipe. A calculated amount of inert bed material, limestone or sand, was then poured into the combustor from the top to provide the desired bed height. The air flowrate was then gradually increased and the bed pressure drop monitored until stable spouting was reached. The preheater was then switched on and the bed heated up slowly b}7 the hot air. Fig. 4.12 shows a typical history of bed temperature and flue gas oxygen concentra-tion. When the bed temperature was brought up to 450°C, a process which required about 3 h, coal feeding was begun at a small feed rate. The bed temperature continued to rise once the coal started burning, with the rate of temperature rise controlled by the coal feed rate. A careful process was then followed where the coal feed rate was gradually 37 Chapter 4. Experimental Technique and Problems Encountered 39 increased and the air fiowrate gradual!}' decreased as the bed temperature rose slowly. It took about two more hours to bring the bed temperature up to 840°C. At this point the spouting air was switched to room temperature air, and the more careful process of controlling both bed temperature and flue gas oxj'gen level began. The overall period, from the beginning of the run to achieving a steady state condition of desired bed temperature and flue gas oxygen level, required a total time of 7-9 hours when no serious problems were encountered. Once the correct temperature was reached, the air flow rates were adjusted to match predetermined values. The parameters which could be used to control the bed temperature were: • coal feed rate • air flow rates in the different sections of the cooling jacket • amount of external insulation It was found that the effects of air flowrates in the cooling jacket on bed temperature were relatively minor compared with the influences of the other two parameters. Since the spouting air flow rate was fixed, coal feed rate was the only parameter which could be varied to control the excess oxygen level. Due to the small bed-to-wall heat transfer coefficient when air was used as the cooling medium, the system had to be operated at the maximum cooling air flow rate through each of the jacket cooling surfaces to bring the flue gas oxygen level down to 3.0% to 4.0% in the temperature range of 800°C to 900°C. To maintain the steady state condition, minor changes of coal feed rate were then used to control the excess oxygen level, while the amount of insulation was altered to minimize the consequent temperature variation . Fig. 4.13 illustrates the coal feed rate variation history during typical "steady state" operation. The measuring period was initiated and terminated according to the starting and ending of both cyclone Chapter 4. Experimental Technique and Problems Encountered 40 c u E c < i i CM < e E E (H/6>j) eioy paaj p<o Figure 4.13: Coal feed rate variation history during steady state operation Chapter 4. Experimental Technique and Problems Encountered 41 and isokinetic solid sampling. Thus, the average coal feed rate is determined by m C l + mC2At2 + ••• + rn^ Atn At! + At2 + • • • + Atn E ^ a ^ A f . ) (4.24) Where m c is average coal feed rate, and rhc. (i = 1,2,... ,n) are coal feed rates corre-sponding to the time intervals At ; (i = 1, 2 , . . . ,n). The bed height was monitored and controlled by the bed pressure drop and overflow solid discharge valve. 4.2 Measurement Techniques 4.2.1 Temperature Measurement Temperatures were measured by seven thermocouples mounted at the back of the com-bustor with a spacing of 102 mm, as shown in Figs. 4.14 and 3.10. When the desired steady state condition was reached, all seven thermocouples were moved to position C, where the first set of readings was taken. Then the thermocouples were moved to position E, D, B, and A respectively, to obtain the four readings. Finally, the thermocouples were moved to position C again to check the reproducibility. During these measurements, the temperature variation was usually within ±1°C. The overall time required to complete the temperature profile measurement was about 15 minutes. 4.2.2 Gas Concentration Measurement Gas analysers had to be turned on to be warmed up at least two hours before the measurement was started. The flue gas was sampled by the flue gas sampling probe shown in Fig. 3.6. The probe was fixed in the outlet pipe of the combustor as shown in Fig. 3.7. Chapter 4. Experimental Technique and Problems Encountered Figure 4.14: Thermocouple positions Chapter 4. Experimental Technique and Problems Encountered 43 Axial gas concentration profiles were measured by a mobile gas probe which was described in detail by Zhao (1986). Fig. 4.15 shows the locations in the combustor where the samples were obtained. By rotating the gas probe, gas concentrations could be measured both in the annulus and in the spout. When measurements were made in the annulus, the gas probe was rotated to such an angle that the gas concentrations at the middle point from the wall to spout-annulus interface were measured. Isokinetic sampling, by which the sampling velocity is equal to the gas velocity at the sampling position, was employed to ensure the accuracy of the measurements. The sampling was started from the lowest position, and the probe was then gradually raised to obtain a set of data. A repeat measurement was then carried out for at least one position to ensure reproducibility. Since the range of the SO2 analyser is only 0 — 500 ppm, the sampled gas was diluted by N2 before entering the analyser when the SO2 concentration exceeded 500 ppm. The total time required to finish determination of a gas concentration profile was about one hour. 4.2.3 Solid Sampling Most solids leaving the combustor were captured by the cyclone. The small amount of solids not captured by the cyclone was sampled by an isokinetic sampling probe shown in Fig. 3.8. Once the desired steady state condition was reached, time was set to zero and samplings started simultaneously for both the cyclone and downstream isokinetic probe. The cyclone sampling was started by closing the ball valve between the cyclone and the ash receiver. Then an empty ash container was installed and the ball valve was quickly reopened. As soon as the sample was collected, a cap was put on the container to cover the sample and prevent contact with air while the sample cooled. When cooled to room temperature, the samples were transferred to glass containers and sealed for Figure 4.15: Locations of gas sampling points in the combustor Chapter 4. Experimental Technique and Problems Encountered 45 analysis or storage. The isokinetic solid sampling started by turning on the vacuum pump and adjusting the sampling fiowrate to the desired value. When the sampling terminated, the ash collected by the filter was transferred to a glass bottle for storage. It was found in the experiments that the amount of ash captured by the filter was typically only about 0.5% of that collected by the cyclone. 4.3 Problems Encountered During the course of the present work, a total of 34 combustion runs were attempted, as shown in Table 4.5 in form of a chronology. However, most of the tests had to be aborted due to a number of problems encountered which required special corrective action. The difficulties included: (1) Leakage of the combustor. In two runs solids and gas started leaking from the bottom of the combustor when the temperature reached about 850°C. This was due to improper installation of the front plate and the thermal expansion at high temperature. (2) Overcooling. Because of the limited power of the electric preheater, the bed temper-ature was not able to rise to high enough to initiate burning of coal, when water was used as the external cooling median. Three runs failed due to this problem before air cooling was used instead of water. Cooling water has to be fed with the bed at low temperature to prevent the sudden generation of steam in the cooling jacket, which could damage the equipment. It was found in one run that bed tem-perature was still only 245°C after 4 hour of heating, while normally two and a half hours, were required to bring the temperature up to 400°C Another run showed Table 4.5: Chrono' Dav/Mo/Yr Coal type/Bed m a t e r i a 1/Ca :S /a:Q Durat ion 26/05/86 Forestberg/Sand/O.0/0.0 3.5 h 12/06/86 Forestberg/Sand/0.0/0.0 7.5 h 19/09/86 Forestberg/Sand/0.0/0.0 4 h 10/12/86 Forestberg/Sand/0.0/0.0 3 h 12/02/87 Forestberg/Sand/0.0/0.0 3.5 h 26/02/87 Forestberg/Sand/0.0/0.0 4 h 03/04/87 Minto/Sand/2.25/0.0 7.5 h 30/04/87 • Minto/Sand/2.25/0.0 5.5 h 21/05/87 Minto/Sand/2.25/0.0 2 h 26/05/87 Minto/Sand/2.25/0.0 14.5 h 09/06/87 09/07/87 Minto/L imestone/2 Minto/Limestone/2 25/0.0 25/0.0 • 6 h 12 h Dgy of Combustion Runs p> Comments F u r t h e r reduced bed h e i g h t ( 2 0 c m ) . Obtained S O 2 d a t a , Oz p r o f i l e , and temperature p r o f i l e . A f t e r k e e p i n g s t a b l e at 830°C f o r one and h a l f h o u r s , bed temperature rose g r a d u a l l y and f i n a l l y reached another s t a b l e p o i n t at 908°C, w i t h o u t c h a n g i n g c o a l f e e d r a t e and s p o u t i n g a i r f i o w r a t e . D i f f i c u l t t o c o n t r o l bed temperature and f l u e gas 0 2 l e v e l s . No s i n t e r i n g at 910°C. Feeder problem. 3 G l a s s windows were used to see how s p o u t - f l u i d bed works at e l e v a t e d t e m p e r a t u r e . Bed temperature was Pp brought up to 580°C. No c o n t r o l problems. a. a To t e s t heat l o s s from the s t e e l f r o n t w a l l v s . bed temperature. Bed temperature was brought up t o 980°C. n Coal s t a r t e d l e a k i n g from the bottom of the f r o n t w a l l a when Tb reached 850°C. Bed temperature n o n u n i f o r m i t y . 5" D i f f i c u l t t o a c h i e v e s t a b l e c o n d i t i o n s longer then 10 min . H <5 Unable to b r i n g bed temperature up to 250°C by u s i n g water coo 1ing. a a A i r c o o l i n g . Bed temperature n o n u n i f o r m i t y . Feeder problem. p> a A i r c o o l i n g . Feeder problem. Q " T3 Water c o o l i n g . Reduced c o o l i n g s u r f a c e . Not enough O heate r power. 21 Much improved c o n t r o l . Some S0 2 d a t a . Bubble formed ° at Tb=830"C. Feeder problem. A f t e r f e e d i n g c o a l , bed . p r e s s u r e drop i n c r e a s e d from 10.8 cmH 0 to 16.4 cmH z0. ^ No s i n t e r i n g was found a t Tb=860°C. n c Reduced bed h e i g h t ( 2 5 c m ) . Some S 0 2 d a t a . S i n t e r i n g a at Tb=880°C. Heater problem. Obtained c y c l o n e s a m p l i n g . Low S 0 2 c a p t u r e . Ran out of c o a l . S l u g g i n g . 01 Dav/Mo/Yr Coal type/Bed materia1/Ca:S /cr.Q Durat ion 24/07/87 Minto/Limestone/2.25/0.0 2 h 28/07/87 Minto/Limestone/2.25/0.0 4 h 04/08/87 Minto/Limestone/2.25/0.0 9.5 h 03/01/88 Minto/Sand/2.25/0.0 10 h 18/01/88 Minto/Sand/2.25/0.0 4.5 h 21/01/88 Minto/Sand/2.25/0.0 11.5 h 05/02/88 Minto/Sand/2.5/0.0 8.5 h 16/02/88 Minto/Sand/2.5/0.0 9.5 h 22/02/88 - Minto/Sand/2.5/0.0 9.5 h 25/02/88 Minto/Sand/2.5/0.0 5 h 26/02/88 Minto/Sand/2.5/0.0 1 h 28/02/88 Minto/Sand/2.5/0.2 6.5 h 02/03/88 Minto/Sand/2.5/0.25 6 h 10/03/88 Minto/Sand/2.5/0.25 14.5 h 13/03/88 Minto/Sand/2.5/0.25 7 h 18/03/88. Minto/Sand/2.5/0.25 13 h Comments Heater problem. Coal f e e d e r b l o c k e d . S m a l l e r c o a l s i z e . O b t a i n e d ash sample. High S 0 2 c a p t u r e . Feeder u n s t a b l e . Obtained ash sample. S 0 2 d a t a . Feeder problem. Bed temperature dropped towards the end. Not enough coo 1ing. S i n t e r i n g . Some S 0 2 d a t a . Cyclone b l o c k e d . Bed temperature was not u n i f o r m and f l u c t u a t e d . SO2 d a t a . Towards the end, 0 2 l e v e l keep d e c r e a s i n g even when c o a l feed r a t e was d e c r e a s e d . S i n t e r i n g at Tb=860°C. S p o u t i n g c o l l a p s e d because of leakage i n the h e a t e r . Minimum s p o u t i n g v e l o c i t y was about t w i c e normal. High p r e s s u r e drop a c r o s s s c r u b b e r . Leakage i n the h e a t e r . S i n t e r i n g o c c u r r e d a f t e r a d j u s t i n g f l u e gas v a l v e to i n c r e a s e bed p r e s s u r e . Leakage i n the h e a t e r caused the c o l l a p s e of s p o u t i n g . 8 I S. 3 8 C n> b o 8 a o c a. Bed temperature n o n u n i f o r m i t y . S p o u t i n g c o l l a p s e d . Some S0 2 d a t a . O b t a i n e d 0 2 , C 0 2 , S 0 2 , and bed temperature p r o f i l e s . Ash sample. D i s c h a r g i n g bed m a t e r i a l caused sudden i n c r e a s e and n o n u n i f o r m i t y of bed temperature. 0o meter f a i l e d t o work. Obtained SO d a t a , ash sample, bed te m p e r a t u r e p r o f i l e , SO2 p r o f i l e , 0 2 p r o f i l e , and C0 2 p r o f i l e . Dav/Mo/Yr Coal type/Bed materia1/Ca:S /Q:Q 21/03/88 Minto/Sand/2.5/0.25 22/03/88 Minto/Sand/2.5/0.25 28/03/88 Minto/Sand/2.5/0.25 03/04/88 Minto/Sand/2.5/0.50 Durat ion Comments 7.5 h Temperature at c e r t a i n p o i n t was much lower than the s u r r o u n d i n g temperatures i n bed. 9.5 h S0 2 d a t a . Ash samples. Sand escaped from the back thermocupple p o r t . 12.5 h SO2 d a t a . Ash sample. D i s c h a r g i n g bed m a t e r i a l caused a sudden i n c r e a s e and n o n u n i f o r m i t y of bed temperature. 14 h Obtained S 0 2 d a t a , NO* d a t a . S 0 2 a n a l y s e r f a i l e d . (Run* i s named a f t e r the date of the r u n . Example: Run#880318 i s c o r r e s p o n d i n g to the date 18/03/88.) 8 -a fa •a s B-a £>' c CD a a. *n 5 3 n o c a a <-> n OO Chapter 4. Experimental Technique and Problems Encountered 49 that temperature dropped from 380°C to 310°C after turning the cooling water on, while later on the temperature stayed at 330°C without any tendency to rise. (3) Heater problem. Some runs were stopped because one or two stages of the electrical preheater refused to work. (4) Feed problems. The coal feeder proved to be very troublesome in the early runs: (i) Particles larger than 1.68-2.00mm tended to block the first feeder before a bigger feeder was fabricated. (ii) The early feeder could not run more than 4 hours. The impeller stopped intermittently because of excessive friction. (iii) Motor speed instability at low feed rate led to difficulty in reading and resulted in inaccuracy. This was solved by using a bigger ratio speed reduction. (iv) Coal refused to drop from the hopper occasionally until a pressure neutral-ization tube was mounted connecting the top of the coal hopper to the coal feeder outlet tube. (5) Segregation of coal and limestone in the coal hopper. When bigger coal particles were mixed with smaller limestone particles, segregation occurred in the coal hopper. This led to inaccuracy of feeding and difficulty in maintaining desired conditions. (6) Burnout of outlet cooling tube. Since the original cooling system was designed for water, a copper tube was used to carry the coolant (water) out instead of a steel tube. As air was used as the coolant, in one case, the air outlet temperature reached 650°C, and this caused burnout of the tube. Chapter 4. Experimental Technique and Problems Encountered 50 (7) Sintering in the reactor. In early runs, Minto coal caused serious sintering problems. Agglomeration was found at the back wall, even when the bed temperature indi-cated by the thermocouples was only 850°(7. This sintering was probably caused by poor circulation of solids,and the limited bed-to-wall heat transfer coefficient with air as coolant. Thus, the large heat release due to the oxidation of coal at the elevated temperature cannot be taken away fast enough locally, causing an increase of the local temperature, resulting in local agglomeration. This problem was suc-cessfully solved by introducing auxiliary air to create a spout-fluid bed. Spout-fluid beds show better solid mixing, better annular solid-gas contact (Yang, 1983) and more uniform temperature (Lim et al, 1984; Lim et al, 1988). (8) Excessive attrition of limestone. In the early runs, limestone was used as the initial bed material. After a run of 8 hour duration, large amplitude fluctuations of the bed pressure drop 'indicated that slugging was occurring, and steady state conditions could not be maintained successfully. The size distributions of coal, fresh limestone, bed material (limestone) after a 10 hour run (#870804) are plotted in Fig. 4.16. The results show that after the 10 hour run the mean diameter of the limestone had fallen from 1.18 mm to 0.32 mm. In other words, the particles changes from group D to group B according to Geldart (1973), and stable spouting would not be expected with the given orifice size (Chandnani and Epstein, 1986; Grace and Lim, 1987). However, one benefit was found when limestone was used as bed material: there was less tendency to agglomerate. No sintering was found, even when the temperature reached 940°C. This may be partially due to the more favourable heat transfer for smaller particles (Zhang et al, 1984; Lim et al, 1988). (9) Plugging of coal feed fine caused by preburning of coal before entering the bed. This was solved by feeding the coal slowly at the beginning of runs. Chapter 4. Experimental Technique and Problems Encountered 52 (10) Plugging in scrubber. This increased the overall bed pressure and consequently caused leakage in the heater. As a result, less air passed through the combustor and the spouting collapsed. (11) Not enough cooling. When air was used as the cooling medium, it was unable to bring the excess oxygen level down to 3.5-4.0% while the bed temperature was kept in the range of 800-840°C. This was solved by adding steel wool in the cooling jacket to increase the coolant side heat transfer coefficient and removing some insulation from the reactor. (12) Malfunctioning of gas analysers. The 02 sensor, S02 analyser, and C02 analyser were found to be troublesome in several runs. (13) Problems controlling bed height. As bed material was discharged, the following phenomena occurred: (i) The bed temperature suddenly increased. (ii) The temperature in the fountain regime and freeboard dropped quickly. (iii) The difference between the bed temperature and fountain temperature could reach 100°C, while during stable operation there was normally not much dif-ference (see Chapter 5). Discharging small amounts of material each time improved the problem. 4.4 Shutdown of the U n i t When a run was completed, the coal feeding rate was gradually reduced and preheated air was switched on to avoid a too rapid temperature drop. Chapter 5 Temperature and Concentration Profiles 5.1 Temperature Profiles Figures 5.17 and 5.18 show the typical temperature profiles in pure spouted beds (q/Qr — 0.0) for 11.2% oxygen in the flue gas. Generally, temperature decreases as the height in the reactor increases, except for those points in and above the spout where temperature slightly increases above the bed and then decreases. This result differs from earlier work (Zhao et al., 1987) where the temperature was increased markedly in the fountain region indicating that combustion was intensified above the bed surface. This early finding may be due to the considerable amount of fine coal particles contained in the fuel. These tend to burn in the freeboard rather than raining back onto the annulus like the bigger particles. It would appear that combustion was not intensified in the fountain region for large excess air in the present experiments because of a smaller fines content. Temperature may increase above the bed surface for low excess oxygen runs in view of the substantial amount of combustion found to occur in the freeboard (see Fig. 5.22). The temperature gradients in the radial direction, both in the bed and in the fountain region, were minor. Relatively uniform temperatures in the active zone (in and near the spout) were found, due to the good solid mixing in this region. A slight decrease of temperature occurred near the curved back wall of the column, presumably due to the heat losses and poor solid mixing in the outer part of the annulus. The temperature decreased faster near the back wall as the height increased in the freeboard. 53 Chapter 5. Temperature and Concentration Profiles 54 900 900 Distance above Inlet Orifice (mm) Figure 5.17: Axial temperature profiles (Run#880216, q/Qr = 0.0, flue gas oxygen level: 11.2%, HQ - 0.23 m, U = 1.4 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5, m c = 0.905 kg/h) Chapter 5. Temperature and Concentration Profiles 55 900 800-O o 0) 3 o Q. E 700-600-500-400-300 V 178 mm a b o v e o r i f i c e 0 280 mm a b o v e o r i f i c e m 381 mm a b o v e o r i f i c e 0 483 mm a b o v e o r i f i c e 584 mm a b o v e o r i f i c e a 686 mm a b o v e o r i f i c e • 788 mm a b o v e o r i f i c e i 10 I 20 i 30 i i 40 50 60 70 80 Distance f rom centre (mm) Figure 5.18: Radial temperature profiles (Run#880216, q/QT = 0.0, flue gas oxygen level: 11.2%, H0 = 0.23 m, U = 1.4 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto). size: 0.85 - 1.68 mm, Ca/S = 2.5, m c = 0.905 kg/h) Chapter 5. Temperature and Concentration Profiles 56 Figures 5.19 and 5.20 show typical temperature profiles in spout-fluid beds (q/Qr — 0.25). Instead of decreasing continuously, the temperatures increased slightly until the maximum temperature was reached 150 mm above the bed surface, and then decreased as the height in the reactor increased. This suggests that combustion is delayed, presumably due to the higher superficial velocity (1.85 m/s) compared to the value (1.39 m/s) in Fig.5.17. There was no marked increase of temperature in the fountain region, consistent with the result in Fig. 5.17. However, the slight increase of temperature above the bed surface over a relatively wide range (up to 150 mm above the bed surface) despite heat loss to the wall indicates that appreciable combustion occurs in the fountain region. As pointed out by Grace and Mathur (1978) and Arbib et al (1981), the fountain may play a very important role in spouted-bed reactors. Temperatures were more uniform radially, with little decrease of temperature near the curved back wall of the column, both in the bed and in the fountain region. The reduced radial temperature gradients are presumably due to better solid mixing when auxiliary air is introduced. 5.2 02 and C02 Concentration Profiles Combustion is strongly affected by temperature and by local oxygen concentrations. The 02 arid C02 concentration profiles help to analyse combustion behaviour in the reactor. Figures 5.21 and 5.22 give typical 02 and C02 concentration profiles for different flue gas oxygen levels in a SFB. The effects of other operating conditions and solid properties on Pi and CO2 profiles were not studied in this work. For a higher flue gas 02 level as shown in Figure 5.21, the oxygen concentrations in and above the spout tend to be higher than that in and above the annulus. The oxygen concentration decreases sharply as the height increases from 125 mm to 200 mm in the spout and then maintains an almost constant level above the bed surface. This indicates Chapter 5. Temperature and Concentration Profiles 57 900-i 800-700-CL E 600-500-400-300-V 3 mm f r o m c e n t r e O 16 mm f r o m c e n t r e O 36 mm f r o m c e n t r e ^ 56 mm f r o m c e n t r e A 76 mm f r o m c e n t r e 100 200 300 400 500 600 700 Distance above Inlet Orifice (mm) 800 900 Figure 5.19: Axial temperature profiles (Run#880310, q/Qr = 0.25, flue gas oxygen level: 10.2%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85— 1.68 mm, Ca/S = 2.5, mc = 1.214 kg/h) Chapter 5. Temperature and Concentration Profiles 58 900-r 800-700-O 600-o 500-CL E ^ 400-300-O • o V 178 mm a b o v e o r i f i c e 0 280 mm a b o v e o r i f i c e A 381 mm a b o v e o r i f i c e 0 483 mm a b o v e o r i f i c e m 584 mm a b o v e o r i f i c e a 686 mm a b o v e o r i f i c e • 788 mm a b o v e o r i f i c e 200-100-1 10 20 30 40 50 60 70 Distance f rom centre (mm) 80 Figure 5.20: Radial temperature profiles (Run#880310, q/QT = 0.25, flue gas oxygen level: 10.2%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5, m c = 1.214 kg/h) Chapter 5. Temperature and Concentration Profiles 59 25-C o c Q) O c o o 20-15-10-5-0-V Oz i n and above spout A O2 i n and above annulus ^ C0 2 i n and above spout O COgin and above annulus —i 1 1 1 1 1 100 200 300 400 500 600 Distance above Inlet Orifice (mm) Figure 5.21: Axial 02 and C02 concentration profiles (Run#880310, Th '= 840°C, q/QT = 0.25, flue gas oxygen level: 10.0%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5, m c = 1.214 kg/h) Chapter 5. Temperature and Concentration Profiles 60 25-20 H c o c O c o o 15-10-0-r o • Oz i n a n d a b o v e s p o u t V O2 i n a n d a b o v e a n n u l u s _ C 0 2 i n a n d a b o v e s p o u t A C O ^ i n a n d a b o v e a n n u l u s V •A 8 I* 100 200 300 400 500 600 700 Distance above Inlet Orifice (mm) 800 Figure 5.22: Axial 02 and C02 concentration profiles (Run#880318, Tb = 840°C qjQT = 0.25, flue gas oxygen level: 3.4%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4-2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5, m c = 1.38 kg/h) Chapter 5. Temperature and Concentration Profiles 61 that most of the coal particles entrained by the spouting air start to burn in the spout below the bed surface (at about 125 mm above the inlet orifice). The slight decrease of the oxygen concentration in the freeboard is no doubt caused by the radial mixing of low-oxygen gas from the annulus with high-oxygen gas which has emerged from the spout. The oxygen concentrations decrease gradually above the annulus until a minimum is reached, followed by a slight increase in the freeboard caused by radial mixing of the gases. At a low flue gas oxygen level as shown in Fig. 5.22, the oxygen concentrations in and above the spout show great similarity with those in and above the annulus. The concentrations decrease substantially below the bed surface, followed by a gradual decrease above the bed surface. This indicates that most of the coal starts to burn in the spout and is consumed below the bed surface. However, the profiles indicate that an appreciable amount of the combustion occurs in the freeboard for this lower excess case. 5.3 S02 Concentrat ion Profiles Figure 5.23 shows axial S02 concentration profiles in a spout-fluid bed at a high excess oxygen level. A sharp increase of SO2 concentration in the spout was found, followed by a sudden decrease. The maximum value was found at 200 mm above the inlet orifice, i.e. 30 mm below the bed surface. A gradual decrease occurred in the fountain region and in the freeboard above the spout. The abrupt S02 concentration change in the spout indicates that sulphur is released very quickly after coal particles are entrained in the bed, and most sulphur is captured in a very short distance after it is released from the coal in the spout. From the corresponding oxygen concentration profile shown in Fig 5.21, most carbon is consumed within the first 200 mm above the orifice, the region in which the S02 concentration increases to a maximum value. It might be concluded from the Chapter 5. Temperature and Concentration Profiles 62 18000- \7 i n a n d a b o v e s p o u t 4 i n a n d a b o v e a n n u l u s 16000-14000-£ 12000-C o 10000-1 <D E o > to 8000-6000-4000-2000-0-100 200 300 400 500 Distance above Inlet Orifice (mm) 600 Figure 5.23: Axial S02 concentration profiles (Run#880310, Tb = 840°C7, q/QT = 0.25, flue gas oxygen level: 10.0%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4 — 2.0 mm, limestone size: 0.85-1.68 mm, coal (Minto) size: 0.85-1.68 mm, Ca/S = 2.5, mc = 1 214 kg/h) Chapter 5. Temperature and Concentration Profiles 63 above evidence that combustion and sulphation happen in two stages: (1) Combustion of carbon, at the same time as most of the sulphur is released. (2) Sulphation of the sorbent However, no abrupt SO2 concentration change was found in the annulus. Instead, there was a gradual increase in the annulus and the fountain region above the annulus, followed by a slight decrease in the freeboard, presumably due to radial gas mixing. Figure 5.24 shows the axial S02 concentration profiles in a SFB at a low flue gas oxygen level. A sharp increase of SO2 concentration was found at the beginning, followed by a sharp decrease in the upper part of spout and the fountain region and a gradual decrease in the freeboard. The maximum S02 value was found at 200 mm above the inlet orifice, in accord with Fig 5.23. However, there was a gradual increase in the annulus, followed by a more gradual increase in the fountain region and the freeboard above the annulus, due to radial mixing of the gas. Chapter 5. Temperature and Concentration Profiles 64 4000 3500-_ 3000-E CL D_ O 2500-CD E o > <S 00 2000-1500-i n a n d a b o v e s p o u t i n a n d a b o v e a n n u l u s 1000-500-100 200 300 400 500 600 700 Distance above Inlet Orifice (mm) 800 Figure 5.24: Axial S02 concentration profiles (Run#880318, Tb = 840°C, q/QT = 0.25, flue gas oxygen level: 3.4%, H0 = 0.23 m, U = 1.85 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85-1.68 mm, coal (Minto) size: 0.85-1.68 mm, Ca/S = 25 m = 1 38 kg/h) Chapter 6 Combustion Performance 6.1 Combustion Efficiency The combustion efficiency is defined as the amount of carbon burnt per unit time in the reactor divided by the amount of carbon introduced to the reactor per unit time under steady state operation, i.e. Vc = ^ x 100% (6.25) where rnbc is the carbon burnout rate in the reactor, m c is average coal feed rate during the' solid sampling period determined from equation (4.24), and Cu is the carbon content in the coal from ultimate analysis. Since m b c is difficult to obtain, rjc is usually determined from the following equation: Vc = [1 - - C - L ± £ L \ x 100% (6.26) mcCu where m c c is the unburnt carbon captured in the cyclone per unit time, and rhcf is the unburnt carbon in the flue gas sampled hy an isokinetic sampling probe as shown in Fig. 3.8. In the experiments, m cy was found to be negligible compared with rhcc. The carbon content in the ash captured in cyclone was determined by burning the ash sample in a Marphi Stove at 850°C for 8 hours. The total weight loss is considered to be the sum of the carbon content and CO'2 content in the ash sample, The latter is due to uncalcined CaCOz- The CO2 content in the ash sample was determined by heating the ash sample in a packed bed reactor described as shown in Fig.6.25 at 750°C for 2 hours in an inert 65 Chapter 6. Combustion Performance 66 X N, n Figure 6.25: Packed bed reactor system 1. Thermocouple 2. Ash container 3. Reactor 4. Ceramic pack 5. Cooling unit 6. Packed preheater 7. Ash sample 8. Heating element 9. Insulation 10. Rotameter Chapter 6. Combustion Performance 67 atmosphere of N2 to prevent the burning of carbon. Thus, m a is obtained by . m i m2 mcc = ma( ) (6.27) mat\ mas2 where mj is the total weight loss in the ash sample of weight maai burnt in the Marphi Stove, m2 is the weight loss in the ash sample of weight mat2 heated in the reactor shown in Fig.6.25, with the introduction of N2, and rha is the weight of ash captured in the cyclone per unit time under steady state operation. The results of the Minto coal combustion tests are shown in fig. 6.26. In general, combustion efficiencies obtained in this combustor were higher than 80%. The relatively lower combustion efficiencies compared with whose reported by Lim et al. (1988) (shown in Fig. 6.27) in spouted and spout-fluid beds may be partially due to the lack of solid recycle in the equipment, since the elutriation of carbon represents the main source of loss in combustion efficiency. A slight improvement of combustion efficiency with recycle of the solids caught in the cyclones was reported by Lim et al. (1988). In fluidized bed combustors, it has been reported that cyclone-fines recycle not only increases limestone utilization but also improves the overall combustion efficiency significantly ( Lessig et al., 1983; Newby et al., 1983; Wen and Chen, 1980). The amount of auxiliary air seems to have a significant effect on combustion .efficienc}'. The combustion eficiency increased from 86.4% to 90.8% as q /Qr increased from 25% to 50% while bed temperature, superficial velocity and flue gas oxygen level were kept unchanged. A considerably higher combustion efficiency was observed at 840°C than at 800°C, while no significant difference was found as the temperature increased from 840°C to 885°C. Chapter 6. Combustion Performance 68 100-1 90-o c w 80-c o E o O 70-60-J i i i i i 1 1 r 800 810 820 830 840 850 860 870 880 890 Temperature (°C) Figure 6.26: Effect of temperature on combustion efficiency (flue gas oxygen level: 3.25 - 3.45%, U=1.83 to 1.92 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5) Chapter 6. Combustion Performance 69 r^  o m m m • • * o o o il li II r- r- r-cy O " o* c r a 4 t r e m TJ *d • • CD t—i CD CD • rQ & CD o o o a •H u •H •H o TJ Q a il CU TH rH rH u N m m P •H l M 1 1 13 4-1 U 4-1 4-1 «H o rH O o O CO rH ex, O-, ex CO 4-1 CO CO 4J a CD O <3 CD m CO e CD •H J-l | J PM o o o O) o CO o o o o o o O) CD o U J c c U J O o CO o CD O O c a CD < t • c c U J % ' AGNHioiddB N o n s n a w o o Figure 6.27: Combustion efficiencies obtained in the present work compare with those reported by Lim et al. (1988) Chapter 6. Combustion Performance 70 6.2 Sulphur Capture Sulphur in coal occurs in two major forms: inorganic and organic sulphur. The organic sulphur is the sulphur bound to the hydrocarbon matrix and usually constitutes about one half to one third of the total sulphur content of coal (Attar, 1978). The inorganic sulphur is mostly in the form of pyrite and marcasite but can also be presented as sulfates (i.e. FeS04, Fe2(S04)3 and CaS04). A l l sulphur forms, except CaS04, are converted to SO2 during the coal combustion process. The SO2 captured by limestone can be represented by the following overall reaction: CaC03 + S02 + ^02 CaS04 + C02 (6.28) However, the direct reaction of SO2 with CaC03 is usually ver}r slow because of the small surface area of raw limestone. Fortunately, CaC02 is unstable at the conditions found in spouted and spout-fluid bed combustors and reaction (6.28) takes place in two steps: rapid decomposition of the CaCOz to CaO, termed calcination, i.e. CaC03 ^ CaO + C02 (6.29) followed by sulphation of CaO according to CaO + S02 + ^02 — CaSOA (6.30) The calcination of limestone before reacting with SO2 is necessary because the sulphation rate greatly depends on the porosity and surface area of the particles. An increase in temperature leads to an increase in the extent of sintering of CaO (Szekely, 1976) resulting in the loss of surface area, while a decrease in temperature results in a lower calcination rate. For a CO2 partial pressure of 15 kPa, the calcination of limestone only takes place at temperatures higher than 780°C (Hills, 1967). Thus, the sulphur capture efficiency is expected to pass through a maximum with increasing temperature. Chapter 6. Combustion Performance 71 In practical cases, a higher than stoichiometric amount of limestone is necessar}' to achieve a good sulphur capture. In man}' fluidized beds,a Ca/S molar ratio of 3 to 4 is usually needed for 90% sulphur capture (Walker et al., 1979). The weight of limestone required per unit weight of coal is computed for a given Ca/S ratio using the relation: where. Ca/S is the Calcium to Sulphur molar ratio, Su the weight percent of sulphur in the coal, WcaCO* the weight percent of CaCOz in the limestone, and M the molecular weight (Johnson et al., 1978). The sulphur capture efficiency is calculated from the following relation (see Appendix B): SQ2(ppm) x 10- 6 x 273(? r Vso, = 11 . 2Su v , — J x 100% (6.32) U 0 2 x i o o X 2 7 3 +T0)mc Fig.6.28 shows the effect of temperature on sulphur capture efficiency. While it is impossible to keep all other conditions totally unchanged, it is seen that variations -in the flue gas oxygen level, and superficial gas velocities are small. As discussed above, the sulphur capture efficiency increases to a maximum value and then decreases with increasing temperature in the range 800°C to 900°C7. The maximum value was obtained around 830°C The sulphur capture presented in Figure 6.28 are somewhat lower than reported by Grace and Lim (1987b) for Minto coal combustion in a circulating fluidized bed combustor. Higher sulphur capture may be expected with the recycle of cyclone fine particles. Because of the difficulties of the experiments, other parameters such as superficial gas velocities, limestone particle size, auxiliary-to-total air ratios and Ca/S ratio were not varied in this work. Chapter 6. Combustion Performance 72 70-i 10-i i i i i i i i i I I — 800 810 820 830 840 850 860 870 880 890 900 Temperature (°C) Figure 6.28: Effect of temperature on sulphur capture efficiency (flue gas oxygen level: 3.25 - 3.45%, U=1.78 to 1.95 m/s, sand size: 1.4 - 2.0 mm, limestone size: 0.85 - 1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5) Chapter 6. Combustion Performance 73 6.3 N0X Emissions The NOx emissions obtained in the spout-fluid bed with q/Qr — 0.5 are plotted in Fig. 6.28. It can be seen that NOx emission increases linearby with increasing flue gas oxygen level. Chapter 6. Combustion Performance 74 Figure 6.29: Effect of flue gas oxygen level and bed temperature on N0X emission in flue gas. (Run#880403: U=1.85 m/s, sand size: 1.4-2.0 mm, limestone size: 0.85-1.68 mm, coal (Minto) size: 0.85 - 1.68 mm, Ca/S = 2.5) Chapter 7 Hydrodynamics In order to achieve direct observation and measurement, the lower two steel windows in the front plate were replaced by two quartz glass windows of thickness 6.4 mm throughout the hydrodynamic runs. Four narrow size ranges (2.0 - 2.36 mm, 1.68 - 2.0 mm, 1.40 -1.68 mm, and 1.0 - 1.18 mm) of silica sand were employed as the bed material. Highvale coal was used instead of Minto coal to avoid the sintering difficulties described in Chapter 5. The size of Highvale coal used was 0.6 to 1.68 mm, and the proximate and ultimate analyses of this coal appear in Table 7.6. Another independent variable examined was the ratio of the flow of auxiliary air to the total air flow, this ratio ranging from 0.0 to 0.62. The fluid density and viscosity were also varied by changing the bed temperature from 20°C to 880°C The dependent Irydrodynamic parameters investigated were minimum spouting velocity, maximum spoutable bed height, and average spout diameter. 7.1 Minimum Spouting Velocity The minimum spouting velocity was measured by observing the bed through the trans-parent front panel. The gas fiowrate was first increased to a value above the minimum spouting condition, held until a stable bed temperature was reached and then decreased slowly until spouting collapsed. During this process, the temperature change was al-ways kept within ±1°C. The gas flowate at which the fountain just collapsed was taken as the minimum spouting fiowrate. The superficial gas velocity at minimum spouting was then determined by dividing the volumetric minimum spouting fiowrate (at the bed Chapter 7. Hydrodmamics 76 T a b l e 7 . 6 : A n a l y s i s o f H i g h v a l e C o a l P r o x i m a t e A n a l y s i s (Wt. %) V o l a t i l e M a t t e r 2 8 . 6 F i x e d Carbon 3 5 . 0 Ash 16 .6 M o i s t u r e 19 .8 U l t i m a t e A n a l y s i s ( d r y b a s i s ) (Wt. %) Carbon 58 .7 Hydrogen 3 .9 N i t r o g e n 0 .9 S u l p h u r 0 .2 Oxygen (by d i f f . ) 15 .8 Ash 20 .6 H i g h e r H e a t i n g V a l u e : 22 .8 M J / k g Chapter 7. Hydrodynamics 77 temperature) by the cross-sectional area of the bed. 7.1.1 Effect of Temperature and Particle Size Fig.7.30 illustrates the effect of bed temperature and particle size on Um, for a fixed orifice diameter and bed height. For the biggest particles ( <fp=2.18 mm ), Ums remains almost the same for Tb < 400°C, and it then increases significantly for the subsequent increase of temperature (i.e. for Tb > 400°C). For the medium size particles (f£p=1.54 mm), however, Um, first decreases slightly until a temperature of about 500°C is reached and then increases gradually with increasing bed temperature. For the smallest particles (<fp=1.09 mm), however, Ums is unchanged for Tb < 100°C, decreases gradually between 100°C and 600°C, and remains constant again for Tb > 600°C It may because that for the big particles, the gas density is important and since pf decreases as temperature increases leading to the increase of £ / m j with temperature. For the small particles, however, the viscosity may be important and since /z increases with temperature resulting in the decrease of Ums. For the medium particles, Ums may be effected by the combined factors. At any given temperature, the values of Ums are higher for larger particles, but the effect of dp on f/m„ appears to be different at the different bed temperatures. At higher temperatures the effect of particles size seems stronger than at the lower temperatures. 7.1.2 Comparis ion wi th Exis t ing Correlations Experimental values of Umi were compared with the values calculated by the Mathur and Gishler (1955) correlation, Equation (2.1), and the Wu et al. (1987) correlation, Equa-tion (2.3), as shown in Fig. 7.31. Both equations underpredicted Uma. Equation (2.3) appeared to work better than Equation (2.1). Furthermore, the trends of Ume with in-creasing temperature predicted by both equations are inconsistent with the experimental data. For the particle size of 2.18 mm, the experimental results show that Umil remained Chapter 7. Hydrodynamics 78 1.8-1.6-1.4-s 1.2-1.0-0.8-0.6-O d j ,=2 .18 mm, H Q = 0 . 2 3 3 m V d p = 1 . 5 4 mm, H o = 0 . 2 3 m A dj> = 1 .09 mm, H o = 0 . 2 3 8 m O O ftp o. o Q> 0.4-200 400 600 Temperature (°C) —i 800 1000 Figure 7.30: Effect of temperature and dp on Um, (D; = 15.9 mm) Chapter 7. Hydrodynamics 79 1.8 1.6 1.4 H 1 vt 3 1.2 H 1.0 H O E x p e r i m e n t a l d a t a O O 0.8H 0.6H 0.4 H Wu e t a l . : ( 1 9 8 7 ) M a t h u r ;anxi G i s h l e r ( 1 9 5 5 ) 200 400 600 800 1000 Temperature ( C) Figure 7.31: Effect of temperature on Uma compared with correlations from the literatures (dp = 2.18 mm, H0 = 0.233 m, D{ = 15.9 mm) Chapter 7. Hydrodynamics 80 almost unchanged for Tb < 400°C, while a considerable increase of t7m j I was predicted by both equations. For Tb > 600°C7, the effect of temperature appears to be stronger than predicted. For fixed values of D{, Dc, H and pp, since pp^> pf, from Equation (2.1) the effect of Pj and dp on r / m 5 can be expressed as Un. = fci-^= (7-33) where k\ = [^][^] 1 / ' 3\J2gHp p is constant for the given conditions. Similarly, Equation (2.3) gives j 1.05 where k2 = 10 .6[^] 1 ' 0 5 [^] 0 ' 2 6 6 [^]~ 0 ' 0 9 5 p p 0 - 2 5 6 \ /2gH is constant for the given conditions. For a given dp, Equations (7.33) and (7.34) suggest that Ume always increases with increasing temperature because of the decrease in pf. They further indicate that the effects of temperature on Ums are generally the same as those plotted in Fig. 7.31, being flatter for smaller particles and steeper for bigger particles. However, the experimental data in Fig.7.30 show that Um! decreases with increasing temperature for smaller particles and increases with increasing temperature for bigger particles. The effects of dp and temperature appear to be much more complex than those predicted. The improper predictions may be due to inadequate knowledge of how to include dp in Equations (2.1) and (2.3) and the absence of fluid viscosity. A more complex equation is necessary to predict t7m 4 correctly at high temperatures. 7.1.3 Effect of Auxiliary Air After auxiliary air is introduced through the conical perforated distributor to the standard spouted bed, it is called a spout-fluid bed. Figure 7.32 shows the effect of q /Qr on the total minimum spouting velocity, t7j-m,,. Fluidization on top of the annulus was observed Chapter 7. Hydrodynamics 81 Figure 7.32: Effect of q/Qr on total minimum spouting velocity (dp = 1.54 mm, H0 = 0.23 mm) Chapter 7. Hydrodynamics 82 for all points with q/Qr > 0.35 and Tj, > 655°C. For those points at room temperature, stable spouting was observed for all values of q/Qr up to 0.55. The minimum temperature and minimum q/Qr ratio required to create spout fluidization were not studied in this work. It can be seen from Fig.7.32 that at high temperatures (TJ, = 655, 710, 820°C) UT,™* increases with increasing q/Qr ratio and the increase is faster at a higher q/Qr value, although there is some overlap at high temperatures. On the other hand, at room temperature Ur,m$ only increases slightly with increasing q/Qr ratio. 7.1.4 The M a x i m u m Value o f ^ , The minimum spouting velocity measured at the maximum spoutable bed height is termed Um. It has been claimed (Mathur and Gishler, 1955; Becker, 1961; Dumitresu and Ionescu, 1967; Pallai and Nemeth, 1969; Mathur and Epstein, 1974; Epstein and Grace, 1984) that Um is closely related to the minimum fluidization velocit3r, Umf, for the given material, and that the ratio of Um/Umf varies between 1.0 and 1.5. Fig.7.33 shows the effect of temperature on Um/Umf- The value of Umf was calculated from the equation: = d£j^EL = JC? + C2Ar - C, (7.35) where Ar is given by Equation (2.14). Values of C\ and C 2 used here are 27.2 and 0.0408 respectively, as proposed by Grace (1982). It can be seen from Fig.7.33 that generally Um/Umf decreases with increasing temperature. For the smallest particles (<ip=1.09 mm), Um/Umf decreases faster than for dp > 1.54 mm. For the three bigger particle sizes (<fp=1.54, 1.84, and 2.18 mm), Um/Umf tends to be smaller than 1.0 when Tb > 200°C, whereas for the smallest particles (cfp=1.09 mm), the ratio was found to be less than unity when Tb > 50°C The values of Um/Umf which are smaller than 1.0 at the elevated temperatures indicate that the mechanism for spout termination beyond Hm is unlikely Chapter 7. Hydrodynamics 83 • d p = 2 . 1 8 mm V dp =1 . 84 mm O d p = 1 .54 mm A dp = 1 - 0 9 mm O 1.05-I \ A O 0.90-1 A 0.85H \ —i 1 1 1 1 1 100 200 300 400 500 600 700 Temperature (°C) Figure 7.33: Effect of temperature and particle size on Um/Umj (D{ mm, P p — 2660 kg/m3) = 15.9 mm, Dc = 156 Chapter 7. Hydrodynamics 84 to be fluidization of annular solids. This is consistent with the observed results described in the next section. 7.2 Maximum Spoutable Bed Height Hm was determined by increasing the bed temperature slowly while keeping the bed height unchanged until stable spouting could not be obtained for any gas fiowrate. The bed height was then taken as Hm at the corresponding loosely-packed bed temperature. After each measurement was finished, solids were discharged until the next height was reached and the same measurement procedure-was then repeated. 7.2.1 Stability of Spouting Fig.7.34 shows the spouting characteristics for different superficial velocities and bed temperatures at dp — 2.18 mm. For bed temperatures well below TBD (line BD in the figure), increasing the bed temperature had no observable effect on the bed behaviour. However, as the bed temperature approached Tjg the spout-anulus interface near the bed surface was found to be slightly unstable. At TB the intensity of this instability increased leading to pulsing and choking in the upper spout. Fluidization in the annulus was never observed. Similar findings on the transition at Hm were reported by Wu et al. ~(1987). The gas superficial velocity, U, also affects the stability of spouting. At temperatures below TEF (line E F in the figure), increasing U did not affect the characteristics of spouting. However, for TEF •< Tb < Tp, stable spouting only existed over a limited range of U. This range was found to be approximately U M S < U < 1.2[/m3 for the given particle properties and bed geometry. Beyond l.2Ums (line D E in the figure), pulsatory spouting was observed. The range shrinked as TB increased between TQ and TR- For bed temperatures higher than T B , the regime transferred to pulsatory spouting and no Chapter 7. Hydrodynamics 85 Figure 7.34: Effect of temperature and superficial gas velocity on spouting characteristics (H0 = 0.233 m, dp = 2.18 mm, D{ = 15.9 mm, Dc = 156 mm, P p = 2660fc 5/m 3) Chapter 7. Hydrodynamics 86 stable spouting could be obtained for any U. TB is taken as the temperature at which the existing bed height becomes the maximum spoutable bed height. The corresponding graphs for dp = 1.09 mm and dp = 1.54 mm are shown in Fig. 7.35 and Fig. 7.36, respectively. 7.2.2 Effect of Temperature on Hm In order to determine the effect of temperature on Hm, Equation (2.15) was differentiated with respect to Ar while other variables were kept constant. The results showed that for all values of Ar, dHm/dAr > 0. This indicates that i 7 m increases with Ar. For gas spouting, Ar decreases with increasing temperature and therefore Hm decreases with increasing temperature. This trend was supported qualitatively by the experimental results as shown in Fig.7.37. It is also shown in the figure that Hm decreases faster for the smallest particles (dp=1.09 mm), with increasing temperature. The direct comparison of the experimental data with the McNab and Bridgwater (1977) equation, Equation (2.15) is given in Figure 7.38. This shows that Equation (2.15) overpredicts Hm at room temperature and that the deviation between the predicted and experimental values tends to be smaller at higher temperatures. The deviations are no doubt partlj- because the McNab and Bridgwater equation is based on the Mathur and Gishler (1955) £/m i ! equation which has already been found to give large errors (Fig. 7.31) and also because we have found different value of b then the 1.11 assumed (Fig. 7.33). It seems that Equation (2.15) provides a better prediction for Hm at elevated temperatures. This result is consistent with that reported by Zhao et al. (1987), who compared their experimental data at temperatures up to 640°C with predicted values calculated from the correlations of McNab and Bridgwater (1977) , Malek and Lu (1965). and Morgan and Littman (1982). The latter two correlations were in poor agreement with the experimental data. Similar trends have been observed by Wu et al. (1987). Chapter 7. Hydrodynamics 87 900-r 800 700 H 600H o 2 sooH ZJ Q. 400 -\ E 300H 200 H 100H P u l s a t o r y S p o u t i n g Stabler-S p o u t i n g 0.6 0.7 0.8 0.9 Superf ic ial Velocity U (m/s) Figure 7.35: Effect of temperature and superficial gas velocity on spouting characteristics (H0 = 0.238 m, dp = 1.09 mm, D{ = 15.9 mm, Dc = 156 mm, pp = 2660 kg/m3) Chapter 7. Hydrodynamics 88 900 800H 700 600 H o O 500 13 - t — a Q. 400-| E 300 H 200H tooH o-J P a c k e d b e d P u l s a t o r y S p o u t i n g S t a b l e S p o u t i n g 0.6 0.8 1 1.2 Superf ic ia l Velocity U (m/s) Figure 7.36: Effect of temperature and superficial gas velocity on spouting characteristics (HQ = 0.23 m, dp = 1.54 mm, D{ = 15.9 mm, Dc = 156 mm, pp = 2660 kg/m3) Chapter 7. Hydrodynamics 89 0.15 H I I 1 1 1 1 0 200 400 600 800 1000 Temperature (°C) Figure 7.37: Effect of temperature and particle size on H, Chapter 7. Hydrodynamics 90 1000 Temperature (°C) Fi gure 7.38: Effect of temperature on Hm compared with McNab and Bridgwater corre-lation (dp = 1.54 mm) Chapter 7. Hydrodynamics 91 7.2.3 Effect of Particle Diameter According to Equation (2.15), there exists a critical value of dp below which Hm increases with dp and above which Hm decreases as dp increases. This critical value is given b}' {dp)crit = 60.6[ , ^ , ] 1 / 3 (7.36) 9{Pp-pf)Pf For air spouting of sand particles at atmospheric pressure, the critical values of dp were calculated to be 1.42, 2.05, 2.66, 3.21, 3.75 and 3.99 mm at 23, 223, 423, 623, 823 and 923°C, respectively. The qualitative effect of increasing dp predicted above was consis-tent with experimental findings shown in Fig. 7.37. At temperatures between 200°C and 300°C7, Hm values were very close for t£p=1.84 mm and 2.18 mm, indicating that {dp)crit w a s reached around 1.84 — 2.18 mm in this temperature range. At temperatures higher than 400°C, Hm increases with increasing dp, because (dp)^ > 2.18 mm. At temperatures lower than 100°C, there were too few experimental data to examine the predicted trends. 7.2.4 Mechanism for Spout Termination Most published works assume that spout termination is due to fluidization of annular solids and most correlations for predicting Hm are based on this mechanism (Littman and Morgan, 1977). However, as mentioned in section 7.2.1, fluidization on top of the annulus did not occur at the elevated temperatures. This result is consistent with Fig. 7.33, where Um/Umj was found to be smaller than 1.0 at high temperatures. This suggests that UAHm is smaller than I7my, since UsHm ^ UAHm • Possible alternative mechanisms for spout termination at high temperatures are choking of the spout or the growth of instabilities at the spout-annulus interface as described in section 7.2.1. Chapter 7. Hydrodynamics 92 7.3 Average Spout diameter Figs. 7.39 and 7.40 show the photographs of spout shape and diameter. It can be seen that spout shape and diameter do not change much with temperature over the range covered. This is consistent with the results reported by Wu et al. (1987). Chapter 7. Hydrodynamics 93 Figure 7.39: Effect of temperature on spout diameter (dp = 1.54 mm, H0 = 0.23 m, U/Umt = 1.10) a. Tb = 23°C7, Ume = 0.90 m/s; b. Tb = 420°C, Umt = 0.80 m/s; c. Tb = 500°C, C/ m J = 0.76 m/s; d. r b = 600°C7, Um. = 0.82 m/s.) Chapter 7. Hydrodynamics 94 Continue Fig.7.39 Figure 7.40: Effect of temperature on spout diameter (dp = 2.18 mm, H0 = 0.233 m, U/Umt = 1-10) a. Tb = 23°C, Ums = 1.14 m/s; b. Tb = 230°C, Umt = 1.15 m/s; c. Tfc = 424°C, C/TO, = 1.15 m/s; d. Tfc = 534°C, Um, = 1.17 m/s.) Continue Fig.7.40 Chapter 8 Conclusions and Recommendations for Further Work 8.1 Conclusions In spite of a number of difficulties, it was demonstrated that it is possible to burn Minto coal, a highly agglomerating coal with a high sulphur content, in small scale spouted and spout-fluid beds with limestone addition for sulphur capture. Auxiliary air, (i.e. a spout-fluid bed) has proved to be very beneficial for handling the agglomerative coal relative to the standard spouted bed. The combustion efficiency of Minto coal in the spout-fluid bed was generally higher than 80% in the temperature range of 800 to 900°C without recycle of the cyclong fines. An increase of temperature between 880°C and 840°C was beneficial for combustion efficienc3r, while a further increase up to 885°C did not seem to have a significant effect on combustion efficiency. An increase of auxiliary/total air ratio was favourable to combustion efficiency at elevated temperature. Sulphur capture obtained in this equipment was relatively low, generally in the range 40 to 60% for a Ca : S molar ratio of 2.5. The optimum temperature for sulphur capture was found to be around 830°C Limestone could not stand up to spouting as a bed material for prolonged periods because of excessive attrition. However, limestone as bed material was found to be beneficial for avoiding agglomerating problems. The NOx emissions in the spout-fluid bed were in the range of 380 to 400 ppm for the operating conditions studied. NOx emission increased linearly with increasing flue gas oxygen level. 97 Chapter 8. Conclusions and Recommendations for Further Work 98 At high flue gas oxygen level, no abrupt temperature increase above the bed sur-face was observed in the spouted and spout-fluid beds investigated in the recent stud}-. Temperature may increase above the bed surface for low excess oxygen runs in view of the substantial amount of combustion found to occur in the freeboard (see Fig. 5.22). More temperature uniformity could be achieved for the spout-fluid bed. Most oxygen was consumed below the bed surface. However, for the low excess air case, an appreciable amount of oxygen was reacted in the freeboard. Sulphur appeared to be substantially released within the bed height. Combustion and sulphation could be considered to occur in two main stages: (1) Combustion of carbon, at the same time as most of the sulphur is released. (2) Sulphation of the sorbent. Experimental values of the minimum spouting velocity, Umi, were compared to the Mathur and Gushler (1955) and Wu et al. (1987) equations. These gave poor agreement over the entire range of temperature. Different trends of Ums for different particle sizes were found. For 'bigger particles Ume tended to increase with increasing temperature, while for small particles it decreased as temperature increased at high temperatures. The total minimum spouting velocity was found to increase considerably with increasing q/Qr ratio at elevated temperatures, while it only increased slightly at room temperature. At the maximum spoutable bed height, the value of Um/Umf was found to decrease with increasing temperature and to be smaller than 1.0 at elevated temperatures. Fluidization in the annulus was never observed for as the termination mechanism of spouting at high temperatures. The McNab and Bridgwater (1977) expression correctly predicted the observed trends of Hm and worked reasonably well at high temperatures, although it was found to overpredict at lower temperatures. Hm decreased with increasing temperature for all particle sizes, with a faster decrease for smaller particles. Chapter 8. Conclusions and Recommendations for Further Work 99 8.2 Recommendations More variables affecting the combustion efficiency and sulphur capture, such as sand par-ticle size, limestone particle size, coal particle size, bed height, superficial gas velocity, Ca/S ratio, excess air, auxiliary/primary air ratio, and orifice diameter, should be exam-ined. A mathematical model should be established and tested against the experimental data. A separate feeding of limestone and coal is necessary to avoid the segregation in the coal hopper, because it was found in this work that coal and limestone tended to segregate in the hopper for a mixture of limestone/coal volume ratio smaller than 1.0 and limestone particle size smaller than coal particle size, leading to the inaccuracy of coal feed rate. Hence, premixing of limestone and coal may not be a good idea for particles of considerable size difference. It ma3' be a better idea to test combustion efficiency seperately from testing sulphur capture since the problem becomes more complicated and very difficult to handle with the addition of limestone. Recycle of cyclone fines is necessary to examine the combustion efficiency. Special attention must be paid to the effect of gas viscosit}', gas density and particle diameter on hydrodynamic variables. A detailed study on the effects of different inde-pendent variables on minimum spouting velocity at high temperatures is necessarj'. Ums is a function of the following parameters: •Um,, = f(H, dp, Di} Dc, (pp-pf)g, pf, p) (8.37) It may be expressed in the following form: Rem, = K^T(fn§-T[y/l + K2Ar - K3]d (8.38) where Remi = c7madp/9y/\i is the Reynolds number under the minimum spouting condi-tion. The reason is that the form provided by equation (7-35) predicts the similar trends Chapter 8. Conclusions and Recommendations for Further Work 100 of Ume as a function of temperature and dp as observed, i.e. equation (7.35) predicts that Umf increases with increasing temperature for bigger particles and decreases with increasing temperature for smaller particles (Grace, 1982). The group (J^) or (Jp)may be replaced by (_}:), depending on the experimental results. Notation Roman Letters a Constant in Equation (8.38) a. As/Ac AA Cross-sectional area of the annulus, m2 Ac Cross-sectional area of the column, m2 As Cross-sectional area of the spout, m2 b Constant in Equation (8.38) c Constant in Equation (8.38) cu Carbon content of coal from ultimate analysis, % Constant in Equation (7.35) d Constant in Equation (8.38) dp Average particle diameter, m Critical value of dp for Hm, m Mean diameter of adjacent screen apertures, m Dc Column diameter, m Di Orifice diameter, m D, Average spout diameter, m g Acceleration due to gravity, m/s2 G PfU, kg/m2s h H/Hm 101 Chapter 9. Notation H,H0 Loosely packed bed height, m Hm Maximum spoutable bed height, m k Constant in Equation (2.2) *. i o . « [ A i " , [ f t r " « i " , u " / v ' l " v ^ f f Ki,K2,K3 Constant in Equation (8.38) mi Total weight loss in the ash sample of weight M a a l ) kg m 2 Weight loss in the ash sample of weight m a 5 2 , kg m a Weight of ash captured in cyclone per unit time, kg/h m a s i Weight of ash sample, kg m M 2 Weight of ash sample, kg rhbc Carbon burnout rate in the reactor, kg/h rhc Average coal feed rate, kg/h mcc Unburnt carbon captured in the cyclone per unit time, kg/h rhcf Unburnt carbon in the flue gas, kg/h m Coal feed rate corresponding to the time intervals (i=l,2,...,n) AU (i=l,2;...,n), kg/h Mcacc-z Molecular weight of CaC03, g/mole Ms Molecular weight of S, g/mole n Number of particles accelerated per unit time q Auxiliary air flow rate, m3/s qmia Auxiliary air flowrate under the condition of minimum spouting with aeration, m3/s Qmj Minimum fluidization flowrate, m3/s Qms Minimum spouting flowrate, m3/s Chapter 9. Notation Qmsa Spout-fluid fiowrate under the condition of minimum spouting with aeration, m3/s Qx Total air fiowrate, m3/s Q.T,mta Total fluid fiowrate under the condition of minimum spouting with aeration, m3/s Qr,Tb Total air fiowrate at Tb, m3/s Su Sulphur content of coal from ultimate analysis, % Tb Average bed temperature, °C TB,TD,TEF Temperature in Fig. 7.33, °C T0 Reference temperature, °C U Gas superficial velocity, m/s Um Minimum spouting velocity at the maximum spoutable bed height, m/s Minimum fluidization velocity, m/s Minimum spouting velocity, m/s Terminal velocity of a particle, m/s Annular gas velocity at H , m/s Annular gas velocity at Hm, m/s Spout gas velocity at Hm, m/s Total minimum spouting velocity, m/s Total fluid superficial velocity with aeration, m/s Total fluid superficial velocity for spout-fluidization, m/s Wcaco3 Weight percent oiCaCOz in limestone, % X{ Weight fraction of particles with an average Umf TT ^ ms Ut UAH UAHM UsHm Ur,ms UT,mta Ux,msf Chapter 9. Notation (i=l,2,...,n) adjacent screen aperture size of dPi Greek Letters pb Bulk solids density, kg/m3 pf Fluid density, kg/m3 pp True particle density, kg/m3 Emf Voidage at minimum fluidization <f> Shape factor of the particles p. Fluid viscousity, kg/m-s r\c Combustion efficiency. % 77so 2 Sulphur capture efficiency, % At{ Time intervvals, min (i=l,2,...,n) Dimensionless Group A PfUmfUt/[(pp- pf)gDi] Ar dp3(pp- pf)pfg/fi2 Remf UmfdpPf/p, Rem, Umtdppf/p Bibl iography [1] Arbib, H. A. and Levy, A. , Combustion of Low Heating Value Fuels and Wastes in the Spouted Beds, Can. J . Chem. Eng., Vol. 60, 528-531, 1982. [2] Arbib, H. A. , Sawyer, R. F. and Weinberg, J., The Combustion Characteristics of Spouted Beds, 18th International Symposium on Combustion, The Combustion Inst., 233-241, 1981. [3] Attar, A. , Fuel, Vol. 57, 201, 1978. [4] Becker, Ha. A. , An Investigation of Laws Governing the Spouting of Coarse Particles, Chem. Eng. Sci., Vol. 13, 245, 1961. [5] Bridgwater, J . , Chapter 6 in Fluidization, 2nd ed., Edited by Davidson, J . F., Clift, R. and Harrison, D., London, 201-224, 1985. [6] Bridgwater, J . and Mathur, K. B. , "Prediction of Spout Diameter in a Spouted Bed - A Theoretical Model", Powder Technol., Vol. 6, 183, 1972. [7] Chandnani, P. P. and Epstein, N . , Spoutability and Spout Destabilization of Fine Particles with a Gas, in Fluidization, ed. K. Ostergaard and A. Sorensen, Engng. Foundation, New York, 233-240, 1986. [8] Charlton, B, G., Morris, J . B. , and Wolloams, G. H. , An Experimental Study of Spouting Beds of Spheres, Rep. Mo. AERE-R4852, U .K. At. Energy Authority, Harwell, 1965. 105 Bibliography 106 [9] Chatterjee, A. , Spout-fluid Bed Technique, Ind. Eng. Chem. Progress Des Develp., No. 2, Vol. 9, 340-341, 1970. [10] Dumitrescu, C. and Ionescu, D., The Spouted Bed, an aspect of the fluidized bed. Rev. Chim. (Bucharest) Vol. 18, 552, 1967. [11] Dumistrescu, C , The Hydrodynarnical Aspects of a Spouted Bed Modified by the Introduction of an Additional Flow, Revista de Chimie 28(8), Roumania, 746-754, 1977. [12] Epstein, N . and Grace, J . R., Spouting of Particulate Solids, in Hankbook of Powder Science and Technology (Edited by M . E. Fayed and L. Otten), Chap. 11, 507-536, Van Nostrand Reinhold, New York, 1984. [13] Ergun, S., Fluid Flow through Packed Columns, Chem. Eng. Progr., Vol. 48(2), 89, 1952. [14] Geldart, D., Hems worth, A . , Sundavadra, R. and Whiting, K. J. , A comparison of Spouting and Jetting in Round and Half-round Fluidized Beds, Can. J . Chem. Eng., Vol. 59, 6.38, 1981. [15] Geldart, D., Type of Gas Fluidization, Powder Technology, Vol. 7, 285-292, 1973. [16] Geldart, D., Gas Fluidization Technology, A Wiley - Interscience publication, 1986. [17] Ghosh, B. , A Study on the Spouted Bed, Part I: A Theoretical Analysis, Indian Chem. Eng., Vol. 56, 533, 1978. [18] Grace, J . R., Fluidized Bed Hydrodynamics, Chapt. 8.1 in Handbook of Multiphase Systems, ed. G. Hestroni, Hemispher, Washington, 1982. Bibliography 107 [19] Grace, J . R. and Lim, C J. , Permanent Jet Formation in Beds of Particulate Solids, Can. J . Chem. Eng., Vol. 65, 160-162, 1987a. [20] Grace, J . R. and Lim, C. J. , Circulating Fluidized Bed Combustion of Coal, Wood-wastes and Pitch, Final report, prepared for Energy, Mine and Resourcds Canada under contract 24St,23440-6-9007, 1987b. [21] Grace, J . R. and Mathur, K . B . , Height and Structure of the Fountain Region above Spouted Beds, Can. J . Chem. Eng., Vol. 56, 533-537, 1978. [22] Grbavcic, Z. B. , Vukovic, D. V . Zdanski, F. K. and Littman, H. , Fluid Flow Pattern, Minimum Spouting Velocity and Pressure Drop in Spouted Beds, Can. J . Chem. Eng., Vol. 54, 332, 1976. [23] Hills, A. W. D., Trans. Inst. Min. Metall. Sect. C, Vol. 76, 241, 1967. [24] Johnson, I., Shearer, J. , Snyder, R. and Vogel, G. J. , 13th Intersocietjr Energy Conversion Engineering Conf., New York, P. 523, 1978. [25] Khoe, G . . K . and Weve, D., Visual Observations of Spouted Bed Gas Combustion Modes and their Flow Regimes, Can. J . Chem. Eng., Vol. 61, 460-467, 1983. [26] Khoshnoodi, M . and Weinberg, F. J . , Combustion in Spouted Beds, Combustion and Flame, Vol. 33, 11-21, 1978. [27] Kono, H. , Granulation of Small Granules from Fine Powder in Spouted Fluidized . Bed Granulators, Int. Symp. on Powder Technology, Kyoto, Japan, 1981. [28] Lessig, W. S., Callahan, S. F., Rickman, W. S., Manaker, A . M . and Harness, J . L., 7th Int. Conf. on Fluidized Bed Combustion, Springfield, p. 761, 1983. Bibliography 108 [29] Lim, C. J . , Gas Residence Time Distribution and Related Flow Patterns in Spouted Beds, Ph.D. Thesis, University of B.C. , Vancouver, 1975. [30] Lim, C. J. , Barua, S. K. , Epstein, N . , Grace, J . R. and Watkinson, A. P., Spouted Bed and Spout-fluid Bed Combustion of Solids Fuel, in Fluidized Combustion, Pro-ceedings of International Conference on Combustion, London, England, Institute of Energy, London, 72-79, 1984. [31] Lim, C. J. , Watkinson, A. P., Khoe, G. K. , Low, S., Epstein, N . and Grace, J . R., Spouted, Fluidized and Spout-fluid Bed Combustion of Bituminous Coals, Fuel, to be published, 1988. [32] Littman, H. Morgan, M . H., Vukovic, D. V . , Zdanski, F. K. and Grbavcic, Z. B., A Theory for Predicting the Maximum Spoutable Height in a Spouted Bed, Can. J. Chem. Eng., Vol. 55, 499,1977. [33] Littman, H. , Vukovic, D. V. , Zdanski, F. K . and Grbavcic, Z. B. , Pressure Drop and Fiowrate Characteristics of a Liquid Phase Spout-fluid Bed at the Minimum Spout-fluid Fiowrate, Can. J . Chem. Eng., Vol. 52, 174-179, 1974. [34] Littman, H. , Vukovic, D. V , Zdanski, F. K. and Grbavcic, Z. B. , Basic Relations for the Liquid Phase Spout-fluid Fiowrate, Fluidization Technology (Edited by Keairns. D. L.) 1, Heisphere, 373-386, 1976. [35] Littman, H. , "The Measurement and Prediction of the Maximum Spoutable Height, Spout Diameter, Minimum Spouting Velocity and Pressure Drop at Minimum Spout-ing in Spouted Beds", Lecture Notes for C.S.Ch.E. Continuing Education Course on Spouted Beds, Vancouver, 1982. Bibliography 109 [36] Madonna, L. A. , Boornazian, L. , Bencel, B. K. and Geveke, D., Some Characteristics of a Spout-fluid Bed, Int. Conf. on Alternative Energy Sources, Edited by T. N . Veziroglous, No. 6, Vol. 3, 257-282, 1980. [37] Malek, M . A. and Lu, B. C. Y . , Pressure Drop and Spoutable Bed Height in Spouted Beds, Ind. Eng. Chem. Process Des. Develop., Vol. 4, 123, 1965. [38] Manurung, F., Studies in the Spouted Bed Techique with Particular Reference to Low Temperature Coal Carbonization, Ph.D Thesis, Univ. of New South Wales, Kensington, Australia, 1964. [39] Mathur, K . B . and Epstein, N . , Spouted Beds, Academic Press, New York, 1974. [40] Mathur, K. B. and Gishler, P. E., A Technique for Contacting Gases with Solid Particles, A.I.Ch.E. J . , Vol. 1, 157, 1955. [41] McNab, G. S., "Prediction of Spout Diameter ", Brit. Chem. Eng. and Proc. Tech., Vol. 17, 532, 1972. [42] McNab, G. S. and Bridgwater, J. , Spouted Beds - Estimation of Spouted Pressure Drop and the Particle Size for Deepest Bed, Proc. European Congress on Particle Technology, Nuremberg, 1977. [43] Morgan, M . H. and Littman, H. , Predicting the Maximum Height in Spouted Beds of Irregularly Shaped Particles, Ind. Eng. Chem. Fundam., Vol. 21, 23, 1982. [44] Newby, R. A. , Vaux, W. G., Ulerich, N . H. , Ranadive, A. Y . and Keairns, D. L. , 7th Int. Conf. on Fluidized Bed Combustion, Springfield, p. 1076, 1983. [45] Pallai, I. and Nemeth, J . , Analysis of Flow Forms in a Spouted Bed Apparatus by the So-called Phase Diagram, Int. Congr. Chem. Eng.(CHISA), 3rd, Prague, Paper Bibliography 110 No. C2.4, Sept. 1969. [46] Peterson, W. S., Shem. Div., Nat. Res. Council of Can., Ottawa, Personal Commu-nications, 1969-1972. [47] Rovero, G., Piccinini, N . and Lupo, A. , Solids Velocities in Full and Half-sectional Spouted Beds, Entropie, Vol. 124, 13, 1985. [48] Shirley, F. W. and Litt, R.D., Advanced Spouted-fluidized Bed Combustion Concept, Proc. 9th Internat. Conf. on Fluidized Bed Combustion, Vol. 2, 1066-1073, 1987. [49] Sutanto, W., Epstein, N . and Grace, J . R., Hydrodynamics of Spout-fluid Beds, Powder TechnoL, Vol. 44, 205-212, 1985. [50] Szekeby, Evans, J . W. and Sohn, H. Y. , Gas-solid Reactions, Academic Press, 1976. [51] Vukovic, D. V. , Zdanski, F. K. and Grbavcic, Z. B., Effect of Annular and Nozzle Flow of Fluid on the Behaviour of a Spouted Fluidized Bed, Int. Congr. Chem. Eng.(CHISA), 4th, Prague, Czechoslovakia, Paper No. C3.8, Sept. 1972. [52] Walker, D. J., Mcllroy, R. A. and Lange, H. B. , Combustion, Vol. 50, 26, 1979. [53] Watkinson, A. P., Cheng, G. and Prakash, C.B., Comparison of Coal Gasification in Fluidized and Spouted Beds, Can. J . Chem. Eng., Vol. 61, 468 - 474, 1983. [54] Wen, C. Y . and Chen, L. H. , 6th Int. Conf. on Fluidized Bed Combustion, Spring-field, p. 1115, 1980. [55] Wen, C. Y . and Yu, Y . H. , A Generalized Method for Predicting the Minimum Fluidization Velocity, A.I.Ch.E. J . , Vol. 12, 610, 1966. Bibliography 111 [56] Whiting, K . J . and Geldart, D., A Comparison of Cylindrical and Semi-cylindrical Spouted Beds of Coarse Particles, Chem. Eng. Sci., Vol. 35, 1499-1501, 1979. [57] Wu, S., Hydrodynamics of Gas Spouting at High Temperature, Master Thesis, Uni-versity of U.C. , 1986. [58] Wu, S. W. M . , Lim, C. J . and Epstein, N . , Hydrodynamics of Spouted Beds at Elevated Temperatures, Chem. Engng. Commun., 1987. [59] Yang, W. C. and Keairns, D. L., Studies on the Solid Circulation Rate and Gas Bypassing in a Spouted Fluid-bed with a Draft Tube, Can. J . Chem. Eng., Vol. 61, 349, 1983. [60] Zhang, H. , Cen, K . and Huang, G., Internat. Chem. Engng., Vol. 24, 158, 1984. [61] Zhao, J . , Coal Combustion in Spouted and Spout-fluid Beds, M.A.Sc. Thesis, Uni-versity of B.C. , 1986. [62] Zhao, J. , Lim, C. J . and Grace, J . R., Flow Regimes and Combustion Behaviour in Coal-burning Spouted and Spout-fluid Bed, Chem. Eng. Sci., Vol. 42, No. 12, 2865-2875, 1987a. [63] Zhao, J. , Lim, C. J . and Grace, J . R., Coal Burnout Times in Spouted and Spout-fluid Beds, Chem. Eng: Res. Des., Vol. 65, 426-429, 1987b. Appendix A Calbration Curves 112 Appendix A. Calbration Curves 113 o m o * * • , *~ O O (u/6>j) spy 6uip89j JDOQ Figure A.41: Calibration curve for the coal feeder (Mixture of Minto coal and Tex-ada limestone, coal size: 0.85 - 1.68 mm, limestone size: 0.85 - 1.68, Ca/S = 2.5, kg limestone/kg coal = 0.571 q I in Appendix A. Calbration Curves 114 Figure A.42: Calibration curve for the SO2 analyser Appendix B Derivation of rjso2 Sulphur capture efficiency is defined as Sulphur leaving with flue gas ^s°2 Sulphur entering with fuel Volume of SO2 released per kg coal with no capture is 25" — Nm3/kq coal pso2 x 100 . where Su is the sulphur content of the coal from the ultimate analysis and pso2 — 2.93 fcg/iVm3 is the density of SO2 under standard conditions. At the bed temperature. Tb, the above expression becomes 25" 273 -f Tb 3 Pso2 x 100 273 The total air fiowrate at Tb is m /kg coal . 273 + T b 3 / l Qr,rh = QT x m/h where To is the reference temperature. The measured flue gas S02 volumetric fiowrate at Tb is S02(PPm) x 10- 6 x ^ ^ xQT m3/h where S02{ppm) is the measured volume fraction in the flue gas. At a coal feed rate of m c kg/h, the sulphur capture efficiency can be written as c n = i _ S02(PPm) x 10- 6 ( l j^)Qr  7?5°2 — i is* WZTZ+T,^ \pso XlOO/V 273 ' c 115 Appendix B. Derivation of r)so2 So that Vso2 = [1 S02{Ppm) x 10" 6 x 273Q r ( 2S» P50 2X100 )(273 + T 0)m c ] x 100% Appendix C Experimental Data 117 Appendix C. Experimental Data Table C.7.A Ef f e c t of temperature on minimum spouting v e l o c i t y Condition: D^=15.9 mm, Dc=156 mm, ^=2660 kg/m dp=2.18 mm (H=0.233 m) 23 1 . 14 482 1 .25 122 1 . 1 1 479 1 .20 180 1 . 12 462 1 .21 224 1 . 15 450 1 . 17 237 1 . 15 445 1 . 24 280 1 . 18 440 1 . 18 300 1 . 18 427 1 .21 340 1 . 18 418 1 .23 396 1 . 17 370 1 . 19 420 1 . 15 347 1 .21 460 1 . 14 300 1 . 20 502 1 . 17 293 1 .20 534 1 . 17 230 1 . 17 577 1 .20 206 1 .15 593 1 .22 190 1 . 11 599 1 .22 600 1 .23 630 1 .26 666 1 .31 669 1 . 29 673 1 . 32 706 1 . 33 708 1 . 38 734 1 . 38 728 1 . .41 740 1 . .43 790 1 . .48 820 1 . .48 8 30 1 . 44 844 1 . 52 845 1 . 46 865 1 . 44 845 1 . 41 834 1 . .45 820 1 . . 39 " 802 1 . 41 780 1 . 38 760 1 . 42 725 1 . 35 703 1 . 39 694 1 . 36 676 1 . 38 657 1 . . 35 650 1 . . 32 617 1 . . 36 610 1 . .24 590 1 . 33 550 1 . .23 526 1 . .25 518 1 . 23 504 1 . . 28 500 1 . . 18 Appendix C. Experimented Data dp =15.4 mm (H=0.23 m) Tb (°6) U m s (m/5) 23 0.90 23 0.85 250 0.86 390 0.77 420 0.80 500 0. 76 600 0.82 610 0.83 700 0.84 770 0.90 655 0.85 660 0. 84 675 0. 85 715 0.88 708 0.84 750 0.87 780 0.90 775 0.90 820 0.91 860 0.85 dp = 1.09 mm (R=0.238 m) 22 , 0 .72 52 J 0 .71 80 j 0 . 72 110 I 0 .71 132 ! 0 .71 139 I 0 .70 151 ! 0 . 70 168 i 0 .69 210 ' 0 .69 235 0 .65 240 0 . 70 295 0 .67 335 0 62 341 0 64 364 0 62 378 0 60 416 0 60 520 0 58 567 0 59 584 0. 58 620 0. 55 630 0. 57 635 0. 54 650 0. 56 700 0. 53 735 0. 54 754 0. 56 770 . 0. 53 780 ; 0. 56 805 0. 54 812 0. 57 835 0. 54 851 ! 0. 56 857 0. 53 Appendix C. Experimental Data Table C.7.B Ef f e c t of q/QT on t o t a l minimum spouting v e l o c i t y (df=1.54 mm, H=0.23 m, D;=15.9 mm, Dc=156 mm, $,=2660 kg/n Tb=25°C q/QT U (rv/s) 0.0 0. 147 0.291 0.551 0.82 0.84 0.85 0.90 T b=655°C q/Qr 0.0 0.373 0.562 0.562 U > 0.85 1 .04 1 .42 1 .38 Tb=710°C q/Q T 0.0 0. 379 0 . 563 U(™/s) 0.84 1 .08 1.31 Tb=820°C q/Q T 0.0 0.447 0.612 0.91 1 .02 1 .34 Table C.7.C Ef f e c t of temperature and p a r t i c l e s i z e on W l U f (D£=15.9 mm, Dc = 156 mm, Pp =2660 kg/m ) d^=1.54 mm Tb U m / U ^ dp =2.18 mm Tb tu/u* 50 110 110 175 205 235 295 365 365 530 1 .07 1 .03 1 .04 1 .03 0.99 1 .00 0.98 0.98 0.99 0.94 170 230 305 565 690 1 .00 0.99 0.94 0.95 0.93 df> = 1.09 mm Tb WU^. 25 1.04 70 0.95 110 0.94 150 0.89 230 0.86 dp=1.84 mm Tb Um/Um-f 110 1.06 190 , .02 270 o.97 350 o g 4 555 0.97 685 o.93 Appendix C. Experimental Data 121 Table C.7.D Ef f e c t of temperature and p a r t i c l e s i z e on H (D^=15.9 mm, Dc=156 mm, =2660 kg/m ) dp=1.54 mm T-bCC) H (m) 50 110 175 205 235 295 365 530 710 605 0.510 0.446 0.420 0. 372 0. 348 0.315 0.275 0.255 0. 185 0.230 dp=2.18 mm Tb (°C) H (m) 170 230 305 565 690 880 720 0.480 0.420 0. 369 0. 320 0. 260 0.210 0.235 dp=1.09 mm T b ( ° C ) H (m) 25 70 1 10 150 230 165 0.436 0. 377 0.315 0. 255 0. 180 0.238 dp=l.84 mm T b(°C) H (m) 1 10 190 270 350 555 685 880 0.515 0.448 0. 395 0.348 0. 305 0.250 0 . 195 Appendix C. Experimental Data 122 Table C.7.E Temperature p r o f i l e s (Run #880216, q/QT=0.0, fl u e gas oxygen level:11.2%, K^O.23 m, U=1.4 m/s, sand s i z e : 1.4-2.0 mm, limestone size:0.85-1.68 mm, coal (Minto) s i z e : 0.85-1.68 mm, Ca/S=2.5, mc =0.905 kg/h) Distance above i n l e t o r i f i c e (mm) 278 380 482 584 686 788 Tb(°C) (from centre) (3mm) (16mm) (36mm) (56mm) (76mm) 1 7 6 760 781 794 785 763 767 774 777 769 735 751 754 754 750 717 714 715 709 701 643 672 678 668 659 533 606 629 601 590 453 541 570 529 509 320 Distance above Tb(°C) i n l e t o r i f i c e (from centre) (ram) (3mm) (16mm) (36mm) (56mm) (76mm) 178 832 8 3 5 838 8 3 7 826 280 838 f 4 2 840 8 3 5 811 381 836 °42 840 8 3 8 8 2 4 483 794 7 9 6 795 7 Q 1 7 5 8 584 730 7 4 2 744 ? 3 2 6 6 7 686 643 ° 7 * 875 6 5 4 528 788 589 6 1 2 6 1 4 580 407 Appendix C. Experimental Data Table C.7.F A x i a l Oi and C0 2concentration p r o f i l e (q/QT=0.25, H=0.23 m, sand s i z e : 1.4-2.0 mm, limestone s i z e : 0.85-1.68 mm, coal (Minto) s i z e : 0.85-1.68 mm, T b=840°C, U=1.85 m/s) Run #880310: f l u e gas oxygen l e v e l : 10.0%, m =1.214 kg/h Distance above o r i f i c e (mm) 127 203 ; 305 305 406 406 508 508 0 2 (%) (spout) Distance above o r i f i c e (mm) 18. .7 11 .0 203 12. .0 203 11 .8 203 11 . .3 305 11 .  1 406 10, .0 508 10 .5 0 2 (%) (annulus) 9.5 9.0 8.5 6.5 6.5 7.0 Distance above o r i f i c e (mm) C0 2(%) (spout) 127 127 203 203 305 305 406 508 0.2 0.8 Distance above o r i f i c e (mm) 203 203 305 406 508 C0 Z(%) (annulus) 9.2 9.6 10.5 10.8 10.2 Appendix C. Experimental Data 124 Table C.7.G A x i a l 0 2 and C0 2 concentration p r o f i l e (q/Q-r =0.25, H=0.23 m, sand s i z e : 1.4-2.0 mm, limestone s i z e : 0.85-1.68 mm, coal (Minto) s i z e : 0.85-1.68 mm, Tb=840°C, U=1.85 m/s) Run #880318: fl u e gas oxygen l e v e l : 3.4%, mc =1.38 kg/h Distance 0 2(%) C0 2(%) Distance 0 2(%) C0 2 above o r i f i c e (spout) (spout) above o r i f i c e (annulus) (annulus) (mm) (mm) 101.6 i HI 1 -3 1 IO-1 .6 j " 7 8 , 1.4 1 2 7 f - I 3 1 127 \\l 4.4 152.4 5 6.7 152.4 11-9 6.0 152.4 7.0 203 9 0 8.4 203 °\ 8.6 254 7 9 9.0 254 7.2 g.7 3 5 5 6 5.1 5 305 6.7 10.6 3 5 5 6 5.4 n 6 355.6 11.3 508 3 7 14.0 4-3 13.1 3-7 14.0 406 508 609.6 J • 1 14.7 711.2 3 - 3 14.4 Appendix C. Experimented Data 125 Table C.7.H A x i a l SOj. concentration p r o f i l e s (Tb=840 C, q/Qy =0.25, H=0.23 m, U=1.85 m/s, sand s i z e : 1.4-2.0 mm, limestone s i z e : 0.85-1.68 mm, coal (Minto) s i z e : 0.85-1.68 mm, Ca/S=2.5) Run #880310: flue gas oxygen l e v e l : 10.0%, mc=1.214 kg/h Distance SO2 Distance SO2 above o r i f i c e concentration above o r i f i c e concentration (mm) (ppm) (mm) (ppm) 127 2694 203 2909 127 2874 203 3113 127 3143 203 3410 203 12519 305 4116 203 16843 406 4116 305 3253 508 3616 305 3459 305 3706 305 3912 406 2389 406 2594 508 2043 508 208fl Run #880318: f l u e gas oxygen l e v e l : 3.4%, mc=1.38 kg/h Distance S0 2 Distance S0 2 above o r i f i c e concentration above o r i f i c e concentration (mm) (ppm) (mm) (ppm) 101.6 1555 101.6 1092 127 1 102 127 2128 152.4 2183 152.4 1498 203 3584 203 1931 254 2791 254 1991 254 2871 305 1895 355.6 2719 355.6 1832 355.6 2925 355.6 2902 508 2637 406 2075 508 2198 609.6 2348 711.2 , 2568 Appendix C. Experimental Data Table C.7.I Ef f e c t of temperature on combustion e f f i c i e n c y (03/8=2.5, sand s i z e : 1.4-2.0 mm, limestone s i z e : 0.85-1.68 mm, coal (Minto) s i z e : 0.85-1.68 mm) q/QT=0.25 Run U f l u e gas m Tb Combustion (m/s) 0 2 (%) (kg/h) C c ) (%) #880318 1.83 3.45 1.64 825 83.1 #880318 1.84 3.0 1.58 835 85.8 #880318 1.85 3.3 1.51 840 86.5 #880322 1.78 3.45 1.89 800 78.7 #880328 1.92 3.45 1.86 885 86.4 q/Q T = =0.50 Run U f l u e gas t mc Tb Combustion (m/s) o 2 <%) (kg/h) CC) CO #880403 1.92 3.25 1.40 885 90.8 Table C.7.J Ef f e c t of temperature on sulphur capture e f f i c i e n c y (Run #880403, sand s i z e : 1.4-2.0 mm, limestone s i z e : 0.85-1.68 mm, coal (Minto) s i z e : 0.85-1.68 mm, Ca/S= 2.5, q/QT=0.50) mc f l u e gas U Tb sulphur (kg/h) 0 2 l e v e l (m/s) (°C) capture 1.62 3.4 1.78 800 41.5 1.55 3.45 1.83 827 54.5 1,55 3.45 1.83 827 56.0 1.55 3.40 1.85 846 53.0 1.52 3.15 1.90 875 50.5 1.45 3.45 .1.92 885 46.0 1.40 3.25 1.92 885 42.0 1.38 3.05 1.92 885 43.0 1.40 3.20 1.92 885 40.0 1.43 3.50 1.94 894 39.5 1.43 3.50 1.94 894 43.5 1.43 3.40 1.95 900 38.0 Appendix C. Experimental Data 127 Table C.7.K Eff e c t of f l u e gas oxygen l e v e l and bed temperature on NOx emission (Run #880403, q/Q^O.50, Ca/S=2.5, sand s i z e : 1.4-2.o mm, limestone s i z e : 0.85-1.68 mm, coal (Minto) s i z e : 0.85-1.68 mm, H=0.23 m) Tb=840 C (U=1.85 m/s) fl u e gas NOK emission 0a(%) (g/kg coal) 3.3 4.7 4.75 5.1 6.0 5.5 7.5 5.9 10.2 6.7 Tb= 870 C (U=1.89 m/s) f l u e gas NOx emission 0 2(%) (g/kg coal) 2.9 4.8 Tb=885 C (U=1.92 m/s) f l u e gas NOx emission 0 2 (%)'••. (g/kg coal) 3.25 5 . 2 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0058774/manifest

Comment

Related Items