Experimental Study on Preheating and Combustion Characteristics of Semi-coke in 0.2MW Pilot Scale Test Rig

Paper from the AFRC 2015 conference titled Experimental Study on Preheating and Combustion Characteristics of Semi-coke in 0.2MW Pilot Scale Test Rig

Low-rank coal accounts for more than 50% of coal resources in China. Its direct combustion or gasification is not economic to energy efficiency. Utilizing low rank coal in stages is an important way to improve the comprehensive benefits. It means that oil and gas products are firstly extracted through an efficient process and the resultant semi-cokes are then combusted. However, because the semi-coke, almost no volatile, has difficulties in ignition, stabilization, and burn out, realizing high efficient combustion for semi-coke is a difficult problem and become a key research direction. The Institute of engineering thermophysics (IET), CAS has put forward a new process to combust semi-coke in a high efficient way and worked a lot of experiments study in a 30 kW test rig. In order to further investigate the combustion characteristics of semi-coke in high thermal capacity, a 0.2MW pilot scale test rig was built in 2014 in IET, CAS. The test rig consists of preheating system, combustion system and auxiliary system. The preheating system is a circulating fluidized bed with the riser 260 mm in diameter and 1500mm high, and the combustion system is a down-flow combustor (DFC) with 700 mm in diameter and 7000mm high. The semi-coke was firstly preheated to over 800°C by a CFB with the primary air, accounted for 10-30% of the stoichiometric air requirement. In the process of preheating, primary air took reactions with semi-coke and was converted to syngas, including CO, H2 and CH4. The preheating fuel, together with syngas, flows into the the DFC via a nozzle. Air supply to the DFC was staged to optimize performance. The secondary air was supplied from the nozzle, while the tertiary air was introduced at a position 1000 mm below the nozzle to ensure complete combustion. A reducing zone was created between the secondary and tertiary air ports in order to reduce NOx formation. The results showed that the semi-coke can be preheated to 850℃ steadily with primary air ratio of 0.15, and the maximum combustion temperature is about 1200℃ with the temperature difference about 300℃. As a result, the combustion efficiency is 98.8%, and NOx emission is 140 mg/m3 (6% O2). The experimental results can provide a good support for the development and application of semi-coke combustion technology.

AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America Experimental study on preheating and combustion characteristics of semi-coke in 0.2MW pilot scale test rig Jianguo Zhu, Kun He, Ziqu Ouyang, Qinggang Lu Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing 100190, China Abstract: Low-rank coal accounts for more than 50% of coal resources in China. Its direct combustion or gasification is not economic to energy efficiency. Utilizing low rank coal in stages is an important way to improve the comprehensive benefits. It means that oil and gas products are firstly extracted through an efficient process and the resultant semi-cokes are then combusted. However, because the semi-coke, almost no volatile, has difficulties in ignition, stabilization, and burn out, realizing high efficient combustion for semi-coke is a difficult problem and become a key research direction. The Institute of engineering thermophysics (IET), CAS has put forward a new process to combust semi-coke in a high efficient way and worked a lot of experiments study in a 30 kW test rig. In order to further investigate the combustion characteristics of semi-coke in high thermal capacity, a 0.2MW pilot scale test rig was built in 2014 in IET, CAS. The test rig consists of preheating system, combustion system and auxiliary system. The preheating system is a circulating fluidized bed with the riser 260 mm in diameter and 1500mm high, and the combustion system is a down-flow combustor (DFC) with 700 mm in diameter and 7000mm high. The semi-coke was firstly preheated to over 800°C by a CFB with the primary air, accounted for 10-30% of the stoichiometric air requirement. In the process of preheating, primary air took reactions with semi-coke and was converted to syngas, including CO, H2 and CH4. The preheating fuel, together with syngas, flows into the the DFC via a nozzle. Air supply to the DFC was staged to optimize performance. The secondary air was supplied from the nozzle, while the tertiary air was introduced at a position 1000 mm below the nozzle to ensure complete combustion. A reducing zone was created between the secondary and tertiary air ports in order to reduce NOx formation. The results showed that the semi-coke can be preheated to 850℃ steadily with primary air ratio of 0.17, and the maximum combustion temperature is about 1200℃ with the temperature difference about 360℃. As a result, the combustion efficiency is 98.8%, and NOx emission is 140 mg/m3 (6% O2). The experimental results can provide a good support for the development and application of semi-coke combustion technology. Key words: Semi-coke, Preheating, Combustion, 0.2MW 1. Introduction 1   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America Low-rank coal accounts for more than 50% of coal resources in China. Its direct combustion or gasification is not economic to energy efficiency. Utilizing low rank coal in stages is an important way to improve the comprehensive benefits. It means that oil and gas products are firstly extracted through an efficient process and the resultant semi-cokes are then combusted. However, because the semi-coke, almost no volatile, has difficulties in ignition, stabilization, and burn out, realizing high efficient combustion for semi-coke is a difficult problem and become a key research direction. At present, most investigations on semi-coke were based on thermogravimetry (TG) system or laboratory-scale reactors. Shaw [1, 2] investigated the combustion of semi-coke from New Zealand coals using thermogravimetric analysis(TGA), and they pointed out that the differences in reactivity between semi-coke from different coal ranks are significant. Xiangmei Meng[3] and Blasi [4] investigated the semi-coke combustion characteristics and found that semi-coke combustion rate increased with increasing either O2 concentrations or combustion temperatures. In addition to the TG analyzer some lab-scale reactors were used to investigate semi-coke combustion. Kelebopile [5] investigated pore development and combustion behavior of semi-coke in a drop tube furnace, the studies give a result that the semi-coke reactions occur close to the inside of the particle. It can be seen that most of these aforementioned investigations into semi-coke combustion were conducted with TG system and laboratory-scale reactors. There are few investigations whether the semi-coke can be combusted well in industrial device. The Institute of engineering thermophysics (IET), Chinese academy of sciences (CAS) has put forward a new process to combust semi-coke in a high efficient way and made a lot of experiments in a 30 kW test rig. The investigation results showed that this technique is available to low-volatile semi-coke. The combustion efficiency can reach up to 98%, and the NOx emission is about 140 mg/Nm3[6-10]. In order to further investigate the combustion characteristics of semi-coke in a higher thermal capacity, a 0.2MW pilot scale test rig was built in 2014 in IET, CAS. The aim of this paper is to analyze the preheating characteristics, combustion characteristics, and NOx emission characteristics in the 0.2MW test rig. Also, there is a comparision between the result of 0.2MW test rig and that of 30 kW test rig, to provide a foundation for the development and application of preheated 2   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America combustion technology. 2. Experimental section 2.1 Experimental set-up The test rig diagram, shown in Fig. 1, is composed of a circulating fluidized bed (CFB) to preheat semi-coke, a down-fired combustor (DFC) to finish high efficient combustion, and an auxiliary system including air system, water circulation system, and automatic controlling system. The primary air was supplied into the CFB for preheating semi-coke, the secondary air was supplied into the nozzle to make a fast reaction with preheated fuel, and the tertiary air was supplied into the DFC at different positions to have a complete combustion and a lower NOx emission. The CFB mainly consists of riser, cyclone and return valves. The primary air, accounting for 10-30% of the stoichiometric air requirement, was supplied to the riser as both fluidizing and reacting agent. In the process of preheating, primary air takes reactions with semi-coke in the CFB. As a result, the semi-coke can be preheated to over 800°C and the primary air was converted into syngas, including CO, H2 and CH4. Due to small primary air ratio, the CFB is strong reducing atmosphere, and no oxygen exited in the outlet of the CFB. In addition, different to ordinary CFB, this CFB has no air-distributor, which is convenient to use in some thermal power plants with pulverized fuel being transported directly into boiler. The primary air carries the pulverized semi-coke into the bottom of the riser, 150mm in diameter and 1500mm high. The gas leaving the CFB is named "high-temperature syngas", and the solids resulting from the CFB are named "preheated semi-coke". The gas and solids were guided by a tube, 57 mm in diameter and 1100mm long, into the nozzle of the DFC. In this test rig, the CFB was made of high temperature resistant alloy steel, and was insulated with 100mm thick fiber cotton. The down-fired combustor mainly includes combustor, furnace wall and cooled heating surface. The furnace wall is 360mm thick and made of four layers of heat insulating material. The combustor is 700 mm in diameter and 6000mm long. The nozzle, with the structure shown in Fig.2, was installed at the top center of the furnace. There are 4 spray pipes in the same circle, 250 mm in diameter, and the diameter of each spray pipe is 38 mm. The tertiary air was divided into several layers to control combustion process and reduce NOx emission. The cool air was supplied to the pipe, installed in the furnace, to absorb heating from preheated semi-coke combustion, and the hot air was emitted into the air. 3   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America Four Ni-Cr/Ni-Si thermocouples in the CFB and Ten Pt/Pt-Rh thermocouples in the DFC are used to measure the temperature. The gas generated in the DFC is measured by a Gasmet FTIR DX-4000 analyzer with an accuracy of ±2% in volume concentration. The O2 concentration is measured by the KM9106 with an accuracy of ± 0.1% in volume concentration. Hot air Preheated semi-coke and syngas Secondary air Cool air Cool tower Air fan Cool water Tertiary air Primary air Coal feeder Water pump Water tank Chimney Air compressor Air heater Air heater Flue gas cooler Bag filter Induct fan Tertiary air Secondary air Air fan Fig.1 Schematic of the 0.2MW pilot scale test rig Preheated fuel port Secondary air port 4   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America Fig.2 Nozzle structure 2.2 Pulverized semi-coke characteristics The pulverize semi-coke used in the experiments was from Shenmu, China. The proximate and ultimate analysis results are shown in Table 1. It can be seen the volatile contant is 11.33%, lower than that of bituminous coal or other fuels. Table 1 Proximate and ultimate analysis of semi-coke Parameter Value Proximate analysis (wt%) Moisturea 12.9 Volatile matter Fixed carbon Ash a 11.33 a 60.34 a 15.43 Ultimate analysis (wt%) Carbona 63.81 Hydrogena 1.18 Oxygen a Nitrogen Sulfur 5.72 a 0.56 a 0.39 Lower heating value (MJ/kg) 22.23 Figure 3 shows the size distribution of pulverized semi-coke. The size distribution of pulverize semi-coke was obtained using a Malvern Mastersizer 2000 laser analyzer. The size range of pulverized semi-coke is 0-0.5mm, and the d50 is 103 um. Cumulative Volume (%) 100 80 60 40 20 0 0.01 0.1 1 size (mm) Fig.3 The size distribution of pulverized semi-coke 2.3 Experimental conditions The experimental conditions are listed in Table 2. The superficial gas velocity is 1.0 m/s in the CFB riser, which is lower than that of ordinary CFB for coal combustion with the superificial gas velocity of 3-5 m/s. The reason is that the fuel has been pulverized and the size is small. About the ratio of circulating-solids-to-feed, it is 20 in design conditions and not easy to directly measure in hot operation. The superficial gas velocity is the same as 1.0 m/s in the DFC, and the residence time of 5   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America pulverized semi-coke is about 6 s. Table 2 Experimental conditions Parameter Unit Value Feed rate kg/h 30.0 3 Primary air flow Nm /h 31.0 λCFB - 0.17 3 Secondary air flow Nm /h 46.0 λRZ - 0.42 3 Tertiary air flow Nm /h Tertiary air position mm λ 144.0 1500,3000 - 1.2 The air equivalence ratio in the CFB is defined as λCFB. The air equivalence ration in the reducing zone of the DFC is defined as λRZ and the excess air ratio in this system is defined as λ. They are expressed as follows. A1 Astoi (1) A1 + A2 Astoi (2) A1 + A2 + A3 Astoi (3) λCFB = λRZ = λ = where Astoi is the air flow for stoichiometric incineration (Nm3/h), A1is the primary air flow for preheating supplied to the CFB (Nm3/h), A2 is the secondary air flow supplied to the DFC (Nm3/h) by the nozzle, and A3 is the tertiary air flow supplied to the DFC at different positions (Nm3/h). In the experiment, λCFB was set to 0.17, i.e., 17% of the stoichiometric air was supplied to the CFB for preheating process. λRZ was set to 0.42, i.e. a strong reducing zone was formed in the area between secondary airport and tertiary airport. 3.Results and Discussion 3.1 Preheating characteristics in the circulating fluidized bed Fig.4 shows the time series temperature profile with 17% of the stoichiometric air supplied to the CFB. The temperature profile indicated that the preheating process was stable and the temperature could reach up to 8500C. The process of preheating was energy self-sustaining, and no additional energy was needed to maintain steady operation. 6   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America Tempereture ( ℃ ) 900 850 800 axial distance 50mm 750 axial distance 750mm axial distance 1450mm 700 0 15 30 45 60 75 90 105 120 Time (min) Fig.4 Temperature variations with time in the CFB As a comparision, the λCFB was 0.37 with the same preheating temperature for semi-coke in the 30 kW test rig, higher than that of the 0.2MW test rig. The main difference between the two test rigs is heat capacity, resulting to the variations of primary air equivalence ratio with the same preheating temperature. Increasing the heat capacity to a larger scale, the λCFB can be further decreased, important to design a new preheating device or operate it well. 3.2 Combustion characteristics in the DFC The syngas and preheated fuel flowed into the DFC where the temperature exceeded the self-ignition point, and the flame was stable. To reduce nitrogen oxide emissions, air-staging technology was also used in the DFC. The time series of temperature data in the DFC are shown in Fig.5, and the temperature profile along the axis of the DFC is shown in Fig. 6. It is obvious that the operation is very stable and the temperature curves are very flat. 1200 Temperature ( ℃ ) 1000 800 600 400 Top 50 mm Middle 750 mm Bottom 1450 mm 200 0 0 15 30 45 60 75 90 105 Time (min) Fig.5 Temperature variations with the time in the DFC 7   120 AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America 1200 Temperature ( ℃ ) 1000 800 600 400 200 0 0 1000 2000 3000 4000 5000 6000 DFC axis distance (mm) Fig.6 Temperature profile along the axis of the DFC There are two temperature peaks in the DFC: one is at the position of 1000mm below the nozzle with 11800C, and the other is at the positon of 3000mm below the nozzle with 10800C. The temperature profile is relevant to the tertiary air inlet position. In this experiment, tertiary air was supplied to the DFC at the position of 1500mm and 3000mm below the nozzle seperately. After the 2000mm below the nozzle, the temperature starts to increase agin due to unreacted combustible species, including fixed carbon and syngas, takes exothermic reactions when it meets oxygen. From the Fig.6, the maximum temperature difference is 3650C in the DFC, lower than that of pulverized boiler especiaclly for low-volatile fuel. In the experiments, the position of tertiary air inlet was changed, and the temperature profile along the axis of the DFC are showned in Fig. 7. The earlier the tertiary air enters the DFC, the earlier the temperature reachs up to the peak. If the tertiary air all was supplied at the upper part of the DFC, the combustion temperature may be higher than 15000C, not better for controlling combustion process and fuel-N transformation route. 8   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America 1200 Tempereture ( ℃ ) 1000 800 600 400 900and1200 mm 1500and1800 mm 200 2100and2400 mm 2700and3000 mm 0 0 1000 2000 3000 4000 5000 6000 DFC axial distance (mm) Fig.7 Temperature profiles at different cases The combustion efficiency is defined as follows [24]: η= H LHV − H CO / C × ARC H LHV (4) where HLHV is the low heating value of semi-coke, HCO/C is the low heating value of unburned carbon and canbon monoxide, ARC is the ash contant of the semi-coke. In the experiment, fly ash was sampled and the combustible content was 2.8%. the calculated combustion efficiency was 98.8%. 3.3 NOx emission Because the maximum flame temperature was below 15000C, thermal NOx could be ignored in this experiment [11].Prompt NOx was a very small fraction of the total NOx emissions. Therefore, the nitrogen stayed preheated fuel, as well as HCN and NH3 in the syngas, which may contribute to NOx formation during combustion. In this experimental condition, the NOx emission is 140 mg/Nm3 (@6%O2). The effect of preheating temperature on NOx emission is shown in Fig. 8. The preheating temperature increases from 8500C to 9500C, the NOx emission decreases from 225 mg/Nm3 to 207mg/Nm3 (@6%O2). This experiment results is totally similar to those of 30 kW test rig [8]. The reason is that when preheating temperature increases, more fuel-N was released in the CFB. The released-N in the CFB, with a strong reducing atmosphere, can be transformed into N2, resulting to a lower NOx emission. 9   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America 250 NOx concentration( mg/m3) 240 230 220 210 200 800 850 900 950 1000 Temperature ( ℃ ) Fig.8 NOx emission with different preheating temperatures The NOx emission variation with secondary air ratio is shown in Fig.9. When secondary air ratio increases from 0.25 to 0.65, the NOx emission also increases from 225 mg/Nm3 to 294 mg/Nm3. Increasing the secondary air ratio, it means that a lot of air was supplied to the nozzel. The reaction velocity with preheated fuel and oxygen could be faster near the nozzle, a relative week reducing atmosphere exits in this area, favorable to promote NOx formation. NOx concentration( mg/m3) 300 250 200 0.2 0.3 0.4 0.5 0.6 Secondary air ratio (%) 0.7 Fig.9 NOx emission with secondanry air ratio 4. Conclusions In his paper, a process of pulverized semi-coke preheating, combustion, and NOx emission was described and analyzed. The main conclusions are summarized as follows. 1) In the CFB without air distributor, the preheating process is stable and the preheating temperature can reach up to 850 0 C, higher than the ignition point. 2) When the heat capacity of the test rig increases from 30 kW to 0.2MW, the primary air equivalence ratio decreases 10   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America from 0.37 to 0.17. With the heat capacity increases, the primary air equivalence ratio could be further decreased, very important to design a new preheating device. 3) There are two temperature peaks in the down-fired combustor with two layers of tertiary air, seperately at the position of 1500mm and 3000mm below the nozzle. The temperature profile is relevant to tertiary air inlet positions. In this experiment, the maximum temperature is about 12000C, and the temperature difference is about 3600C. The combustion efficiency is 98.6%. 4) The NOx emission decreases with the increase of preheating temperature, and with the decrease of secondary air ratio. The NOx emission is 140 mg/Nm3 (@6%O2). Acknowledgments This study is supported by the Stragetic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA07030100) References: [1] B. Basil Beamish, Karen J. Shaw, K. A. Rodgers, et al. Thermogravimetric determination of the carbon dioxide reactivity of char from some New Zealand coals and its association with the inorganic geochemistry of the parent coal[J]. Fuel Processing Technology, 1998, 53(3): p. 243-253. [2] Karen J. Shaw, B. Basil Beamish, and K. A. Rodgers. Thermogravimetric analytical procedures for determining reactivities of chars from New Zealand coals[J]. Thermochimica Acta, 1997, 302(1-2): p. 181-187. [3] Xiangmei Meng, Wiebren De Jong, Fatemeh Sadat Badri, et al. Combustion study of partially gasified willow and DDGS chars using TG analysis and COMSOL modeling[J]. Biomass and Bioenergy, 2012, 39(0): p. 356-369. [4] Carmen Branca and Colomba Di Blasi. Thermogravimetric analysis of the combustion of dry distiller's grains with solubles (DDGS) and pyrolysis char under kinetic control[J]. Fuel Processing Technology, 2015, 129(0): p. 67-74. [5] Leungo Kelebopile, Rui Sun, Hui Wang, et al. Pore development and combustion behavior of gasified semi-char in a drop tube furnace[J]. Fuel Processing Technology, 2013, 111(0): p. 42-54. [6] Qinggang Lu, Jianguo Zhu, Tianyu Niu, et al. Pulverized coal combustion and NOx emissions in high temperature air from circulating fluidized bed[J]. Fuel Processing Technology, 2008, 89(11): p. 1186-1192. [7] Ziqu Ouyang, Jianguo Zhu, and Qinggang Lu. Experimental study on preheating and combustion characteristics of pulverized anthracite coal[J]. Fuel, 2013, 113(0): p. 122-127. [8] Ziqu Ouyang, Jianguo Zhu, Qinggang Lu, et al. The effect of limestone on SO2 and NOX emissions of pulverized coal combustion preheated by 11   AFRC 2015 Industrial Combustion Symposium, Sep.9-11,2015, Salt Lake City, America circulating fluidized bed[J]. Fuel, 2014, 120(0): p. 116-121. [9] Jianguo Zhu, Qinggang Lu, Tianyu Niu, et al. NO emission on pulverized coal combustion in high temperature air from circulating fluidized bed - An experimental study[J]. Fuel Processing Technology, 2009, 90(5): p. 664-670. [10] Jian-Guo Zhu, Yao Yao, Qing-Gang Lu, et al. Experimental investigation of gasification and incineration characteristics of dried sewage sludge in a circulating fluidized bed[J]. Fuel, 2015, 150(0): p. 441-447. [11] W. D. Fan, Z.C. Lin, Y.Y. Li et al. Effect of Temperature on NO Release during the combustion of Coals with Different Ranks. EnergFuel. 2010,24:1573-1583. 12