Effects of pyrolyzed semi-char blend ratio on coal combustion and pollution emission in a 0.35 MW pulverized coal-fired furnace

Abstract

The effects of blend ratio on combustion and pollution emission characteristics for co-combustion of Shenmu pyrolyzed semi-char (SC), i.e., residuals of the coal pyrolysis chemical processing, and Shenhua bituminous coal (SB) were investigated in a 0.35 MW pilot-scale pulverized coal-fired furnace. The gas temperature and concentrations of gaseous species (O₂, CO, CO₂, NO_x and HCN) were measured in the primary combustion zone at different blend ratios. It is found that the standoff distance of ignition changes monotonically from 132 to 384 mm with the increase in pyrolyzed semi-char blend ratio. The effects on the combustion characteristics may be neglected when the blend ratio is less than 30%. Above the 30% blend ratio, the increase in blend ratio postpones ignition in the primary stage and lowers the burnout rate. With the blend ratio increasing, NO_x emission at the furnace exit is smallest for the 30% blend ratio and highest for the 100% SC. The NO_x concentration was 425 mg/m³ at 6% O₂ and char burnout was 76.23% for the 45% blend ratio. The above results indicate that the change of standoff distance and NO_x emission were not obvious when the blend ratio of semi-char is less than 45%, and carbon burnout changed a little at all blend ratios. The goal of this study is to achieve blending combustion with a large proportion of semi-char without great changes in combustion characteristics. So, an SC blend ratio of no more than 45% can be suitable for the burning of semi-char.

Keywords

pulverized-coal combustion / pyrolyzed semi-char / ignition characteristics / pollution emission / char burnout / blend ratio

Introductions

Grade utilization is one of the most important ways to use coal cleanly and efficiently. It facilities the reduction of coal consumption and prevention of deterioration of the environment. The comprehensive utilization of coal through pyrolysis can be regarded as a way of grade utilization, whose process is as follows [¹]: when peat, lignite, high-volatile and low-rank bituminous coals are heated rapidly in air-free condition at a temperature range of 500°C-700°C, a series of physical and chemical changes will happen in coal. Tar and light gases like CH₄, C₂H₆, CO will release during this process, and the residual solid is the so called pyrolyzed semi-char (SC). So far there are a lot of coal chemical industry companies in Shaanxi province in China to handle coal in this way and there are many advantages and application of the products for industry. For example, tar can replace part of crude oil and light gases like CO and CH₄ can be a substitute of nature gas. Since pyrolyzed semi-char has a high calorific value, it is usually used in the field of acetylene, ferroalloy, and silicon iron production. However, the limited consumption of semi-char for these downstream industries frequently causes an excess amount of semi-char, which seriously impedes the development of the whole coal chemical industry chain in China recently. It is urgent to find an adaptive solution to dispose the large amount of semi-char. A possible way to break the bottleneck is to burn semi-char to generate power in large capacity boilers. The application of semi-char combustion in power plants can dispose a large number of industrial wastes; besides, semi-char can become a coal substitute to save coal resources; moreover, if semi-char is used properly, it will help to reduce pollutant emission, slagging, and corrosion in power plant since semi-char is originated from high quality coal. However, considering its property of low volatile matter and high fixed carbon content, semi-char is closed to anthracite, which may cause unstable flame, high NO_x emission, and low combustion efficiency [²]. Therefore, it is necessary to introduce some auxiliary fuels such as easily ignited high-volatile bituminous coal by using the blending technology to improve the combustion characteristics of semi-char in utility boilers. If possible, the blending combustion technology can also fulfill new emission standard for air pollutants from power plants as dictated by Chinese regulations [³] and the contradiction between demanding of high-quality coal and coal reserve in China. Thus it is fundamental to have a deep understanding of blending combustion characteristics, such as ignition distance, burnout ratio, NO_x emission etc.

Ignition is a key procedure for the co-combustion technology since reactions in the proximity of burner zone become more complex when coals of different ranks are blended or co-fired [⁴], the flame flow pattern of blended coal is different from that of the component coal in the furnace, and is also related to the achievement of easy ignition, and good flame stability to ensure safe boiler operation. Stable ignition for blended combustion of semi-char and coal become a precondition for the practical utilization in large boiler furnaces.

The research on the co-combustion of semi-char and coal is rare and the experiments are mostly conducted in small lab-scale devices [⁵,⁶]. The experiment in TGA found that the ignition and burnout temperature of the blends decrease with the increase in high volatiles coal. The results from DTG experiment show that the burnout ratio of the blends increase non-linearly and NO reduces as the blending ratio of high volatiles coal increases. The emission of fine particulate matter is also reduced in the co-combustion of coal and semi-char [⁷]. There are also some researches on the co-combustion of coal and biochar [⁸,⁹] or petroleum char [¹⁰], whese results indicate that the combustion characteristics of the hard ignition fuel is improved during co-combustion. But biochar and petroleum are quite different from semi-char due to the lower energy density of biochar and higher ash content, and the lower heat value of petroleum char, which provide a little help in co-combustion of semi-char and coal. Since semi-char is obtained from coal, the research of blended combustion of different ranks of coals may provide a reference to study the co-combustion characteristics of semi-char and coal. Although a lot of work has been done about binary coals combustion [¹¹–¹³], they are mostly performed in TGA or DTF. The drawback of small scale experimental facility lies in the fact that the experiment condition is very different from that in a real boiler furnace. Besides, the physical and chemical structures are also different between semi-char and coal. Therefore, the experience obtained from the experiments in TAG or DTF cannot provide reference for guiding the co-combustion of semi-char and coal in reality.

Therefore, it is urgent to do some researches which are close in nature to the boiler furnace in power plants to provide a higher level of confidence on combustion behavior prior to the large scale use of semi-char in real. Pilot-scale test is more suitable under this situation, but no relevant reference has been found so far. The objective of this paper is to investigate the ignition and pollution emission characteristics of blended fuel of semi-char and high volatile bituminous coal in a pilot-scale pulverized coal furnace. The resulting data will enhance the understanding of co-combustion characteristics of semi-char, coal and provide reference for developing a blended combustion technology for coal and semi-char in power plants.

Experiment

Experimental system

The experiment is conducted in a 0.35 MW pulverized coal-fired system (PCFS), as shown in Fig. 1, which consists of a furnace body, a fuel feeding system, an air supply system, a propane stable combustion system, an air preheating system, and a measurement and sampling system, whose main design parameters are listed in Table 1. The PCFS is refitted from the system used in Ref. [¹⁴]. Figure 2 depicts the specific structure of the fuel jet burner, propane burners, and second air pipes on the top of the furnace. The fuel feeding system, second air and propane stable combustion system are all arranged symmetrically.

In the PCFS system, induced draft fan is started first to form a negative pressure in the furnace, and then the propane stable combustion system is started to heat the furnace. The thermal power of the propane output is 100 kW. In order to ensure the safe operation of the boiler, the pressure in the furnace is always maintained at –100 Pa. When the temperature in the middle of the furnace reaches 400°C, the forced draft fan is started to supply primary air and second air to ignite the blended fuel. The blended fuel is stored in two bins and transported into the furnace by the primary air. The fuel feeding rate is controlled by adjusting the rotation speed of the feeder. Once the blended fuel is ignited and the fuel feeding rate reaches the designed value, the thermal power of the propane output decreases to 50 kW and the total thermal power in the furnace is stabilized at 350 kW. The measurement begins when the temperature change in the furnace is less than 10 K and it is about 4 hours to reach to this state.

In flame temperature measurement

The measurement holes can be found in Fig. 1. There are 6 horizontal holes on the side of the furnace to measure the radial temperature, and the vertical distance between the burner outlet and center line of the hole is denoted by Z below. The vertical distance between every two measurement holes is 160 mm from Z=180 to 980 mm. The measuring points are arranged at r = 0, 20, 40, 60, 80, 100, 120, 150, 200, and 300 mm radically at each cross section. The thermocouples for measuring radial temperatures are corundum armored type-S thermocouples with an external diameter of 16 mm and an accuracy of ±0.25%T (T is the temperature measured) over a measuring range of 600°C–1600°C.

There is also a hole between two coal burner jets at the top in the center of the furnace, which is used to measure the axial temperature. An inconel armored type-K thermocouple with a diameter of 8 mm and an accuracy of ±0.75%T (T is the temperature measured) over a measuring range of 0°C–1300°C was employed to measure the temperature along the furnace axial center line. This temperature is referred to as the “axial temperature” in the present paper. The distance from the burner outlet to the measuring point varies from 20 to 900 mm, and is also denoted by Z.

In furnace flue gas components measurement

There are other 6 horizontal holes arranged on the side of the furnace to measure the radial flue gas composition in the furnace, and the center line of the radial temperature measure holes and radial flue gas measure holes are vertical to each other. Radius flue gas measuring holes range from Z=20 to 820 mm along the axial distance, and the measuring points are arranged at r = 0, 20, 40, 60, 80, 100, 120, 150, and 200 mm at each cross section. The vertical distance between each two flue gas measuring holes is also 160 mm. The flue gas components are analyzed by a sampling system demonstrated in Fig. 3, which mainly consists of a water-cooled sucking probe, a fiber filter, a drying bottle, and a flue gas analyzer.

The fuel gas sampling process is as follows: A centrally-located sampling pipe surrounded by a stainless steel tube with high pressure water is inserted into the furnace along the radius direction to suck the hot flue gas. The flue gas is quickly cooled by water, passes through the fiber filter and the drying bottle, and is then analyzed online by a GASMET DX-4000 portable FTIR gas analyzer. The measured gas species include O₂, CO, CO₂, and NO_x (NO, NO₂), with an accuracy of ±2% for the measurement range. Each sensor in the gas analyzer is calibrated prior to the measurement.

Solid residuals collection

Solid residual (fly ash) is collected in the center line at Z = 820 mm by the same sampling system used for flue gas measurement, as shown in Fig. 3. The difference is that the FTIR gas analyzer is replaced by a vacuum pump while the ash filter is replaced by an ash collector. The ash samples are extracted and sealed for off-line analysis. The filter diameter in the ash collector is 0.3 µm.

The burnout rate is calculated using Eq. (1) to analyze the burnout ratio of the blended fuel [¹⁵].

(1){\mathrm{BR}}_{\mathrm{C}}=\left[\mathrm{1}-{\displaystyle \frac{A_{\mathrm{0}}}{\mathrm{1}-A_{\mathrm{0}}}}{\displaystyle \frac{\mathrm{1}-A_i}{A_i}}\right]\times \mathrm{100}\mathrm{\%,}

where BR_C is the burnout ratio, A₀ is the ash weight fraction in the input blended fuel, and A_i is the ash weight fraction in the collected solid residual.

Fuel properties and experimental operation parameters

The tested fuel SC in the blended fuel is obtained from Shaanxi Coal and Chemical Industry Group Co. Ltd. Shenhua bituminous coal (SB) is obtained from Inner Mongolia Autonomous Region, China. The proximate and ultimate analytical data for SC and SB are presented in Table 2. The two fuels are milled separately both to a fineness of R₉₀ = 9% and are blended in advance before injecting into the furnace. The weight blend ratio of SC in the blended fuel are 0%, 30%, 45%, 60%, and 100%.

The main operating parameters for each case of experiments are given in Table 3. It is believed that the small fluctuations caused by experiment condition difference and fuel disparity can be neglected for this scale of experimentation. To guarantee the ignition behavior in a steady state, the fuel feeder is calibrated with the fuels in advance to make sure that the fuel feeding rate is in the designed condition, and the temperature change is maintained to be less than 10°C in the furnace.

Results and discussion

Effect of blend ratio on ignition and stable combustion for blended fuel

Effect of blend ratio on ignition temperature and standoff distance of blended fuel

The axial temperature in the furnace is a good indicator of ignition characteristics of PC jets [¹⁶,¹⁷]. Figure 4 exhibits the axial temperature change at different blend ratios. As can be seen in Fig. 4, for all blends, the axial temperature first increases from Z=0 to 60 mm due to the fact that the two parallel pulverized jets does not intersect at the central axis of the furnace, and then decreases from Z=60 to 90 mm. The reason for this is that the two parallel pulverized jets intersects at the center line and absorbs heat from the gas around in this range. The temperature difference became obvious when Z is larger than 90 mm with the increase in the blend ratio. In the low temperature region (Z is less than 500 mm), the pure bituminous coal always has the highest axial temperature which decreases with the increase in the blend ratio. The reason for this is that during the ignition process, volatile first burns in the initial ignition stage and then induces char burning [¹⁸]. For pure bituminous coal, it has the highest volatile content compared to other blends. The volatile combustion releases a lot of heat rapidly to heat flue gas to a temperature high enough to ignite char in a short axial distance. Then, volatile and char burns simultaneously and the oxygen concentration is still high at this moment, thus the axial temperature increases rapidly in this range. With the increase in the blend ratio, the volatile concentration is decreased. The reaction between the volatile and oxygen releases less heat. The char is ignited in a longer distance, the reaction zone is prolonged, and the high temperature zone shifts down, thus the axial temperature in the low temperature region is gradually decreased with the increase in the blend ratio. The axial temperature for 100% blend is obviously lower than that of the other blends. This is because the semi-char has a too low content of volatile to incur homogeneous ignition, and the ignition of the char is slower than that of the volatile matter.

According to the axial temperature distribution in the low temperature region, it is evident that the bituminous coal in the blended fuel exerts more influence on the initial reactive characteristics during ignition. More specifically, it is the volatile in the bituminous coal that promotes ignition. In the high temperature region (Z is more than 500 mm), the axial temperature disparity between different blends gradually lessens as the axial distance increases and the curves for all blends nearly overlap at last, showing that the combustion intensity at this stage are similar for all blends. The increment in axial temperature is more obvious with the increase in the blend ratio in the high temperature region. The reason for this is that the fuel of low blended ratio reaches the steady state sooner due to the contribution of the volatile combustion to increase the particle temperature [¹⁹], and the char combustion is the dominant reaction in this region, but more oxygen is consumed by the volatile in the low temperature zone while the high temperature zone is severe depleted of oxygen, leading to a weak combustion of the char in low blend ratio, and the degree is alleviated with the increase in the blend ratio. Especially, when the blend ratio is over 45%, the fuel of the higher blend ratio has more oxygen to react with in the high temperature zone. The char ignition is the dominant process at the very beginning due to its very low volatile content for 100% blend, and the reaction is weak at the low temperature region, thus the oxygen concentration is still very high at the high temperature region. Besides, the unburned semi-char has absorbed adequate heat in the low temperature zone and the flue gas temperature is over 1000°C in the high temperature region, which is very favorable for semi-char combustion, thus the axial temperature increases fast and is gradually closed to other blends in the high temperature zone. These results suggest that the bituminous coal in the blended fuel has an obvious influence on the ignition stage, semi-char can get the same combustion intensity as other blends at the burnout stage, and the combustion intensity of semi-char is not weaker than other blends once it is ignited stably.

This paper defines standoff distance as the position where the axial temperature increases fastest, which is calculated using the inflection point from the axial temperature versus the axial distance curve (d²T/dZ²=0) [²⁰]. The axial temperature increases sharply as the axial distance increases from 90 to 550 mm for all blends, thus the standoff distance is in this range. There may be more than one zero point on the second-derived curves for some cases, and the unreasonable points are excluded through observation of the real flame. The method to calculate the standoff distance for 0% is displayed in Fig. 5. The standoff distance for other blends are calculated using the same method. The calculated results for different blend ratios are listed in Table 4. Table 4 also lists the ignition temperature (IT), which refers to the flue gas temperature at the ignition point.

It can be seen from Table 4 that the standoff distance increases monotonically with the increase in the blend ratio. Since the other experiment parameters are the same for all blends except changing the blend ratio, the radiation heat and convection heat absorbed at the same axial distance for all blends are similar, thus the different standoff distance with the increase in blend ratio is mainly caused by the difference in fuel property. Since the ignition of high-volatile bituminous starts with volatile [²¹], the ignition model may convert to the combined ignition model with the increase in the blend ration and eventually convert to heterogeneous model for pure semi-char. The induced effect of volatile is essential to the easy ignition of the blended fuel.

Effect of blend ratio on stable combustion of blend fuels

Effect of blend ratio on radial temperature distribution

The fire is stable because, first, the operating parameters such as stable flowrate and temperature of primary and secondary air, and negative pressure at the furnace outlet are stable; next, the flue gas temperature change is less than 10°C; and finally, the visual observation of the flame is stable. Radial temperature can reflect the characteristics of stable combustion. Radial flame temperature distribution along the axial distance for different blend ratios are illustrated in Fig. 6. For 0% blend ratio, the radial temperature at the center is lower than that in the peripheral region at Z=180 mm, showing the fact that the ignition first occurs near the jet center. This is because the high-temperature flue gas produced from propane combustion first contacts with the peripheral part of the jets, the fuel is ignited from the edge of the jets, and then gradually spread into the center of the jets, resulting in the burning throughout the jets. The temperature in the center at Z =340 mm is higher than that at the two sides, which means that the center of the jets is ignited thoroughly and the flame is stable at this section. Then the radial temperature is drastically increased from Z=340 mm to Z=500 mm. This may result in a convert of combustion from volatile to char. The small increment in radial temperature from Z=180 mm to Z=340 mm can also illustrate that volatile releases less heat than char during combustion. The increment in radial temperature after Z=500 mm is small, showing that char combustion is in a stable state. The radial temperature is stable when Z reaches 820 mm. The reason for this is that oxygen is deficit at this section, which inhibit the combustion behavior.

For the 30% blend, the variations of radial temperature at the cross sections of Z=180 mm and Z=340 mm are nearly 0%, showing that the 30% blend also incurs volatile ignition at first. The increment in radial temperature from Z=500 mm to Z=660 mm is more obvious compared to 0%. This is because the combustion zone moves down for the 30% blend due to its longer standoff distance, thus the stable region for the 30% blend is also more delayed than that for the 0% blend. After Z=660 mm, char combustion is inhibited because the oxygen is deficit, thus, the radial temperature changes a little. For the 45% blend, the variation of radial temperature is similar to that for the 30% blend at all cross sections. The difference is that the increment in radial temperature from Z=660 mm to Z=820 mm is more obvious than that for the 30% blend. This shows that the position of char stable combustion for the 45% blend is further back than that for the 30% blend. For the 60% blend, the variation of radial temperature in the region of r=±60 mm are similar to that of the 0% blend at Z=180 mm and Z=340 mm, while the radial temperature is almost the same at other radial position at the two cross sections. This indicates that although the fuel has already been ignited at Z=340 mm, the ignition intensity is weak. The reason for this may be that the volatile content in the 60% blend is too low to catch fire, instead the ignition of volatile and char start simultaneously, leading to a weak combustion in the first two cross sections. The variation of radial temperature for the 60% blend is similar to that of the 45% blend after Z=500 mm, showing that the combustion mode in this zone is the heterogeneous combustion of char. For the 100% blend, the radial temperature in center at Z=180 mm and Z=340 mm are both lower than that at its two sides, showing that the ignition has not yet been stable. Since the volatile content in pure SC is less than 10%, the 100% blend is in a heterogeneous ignition mode and char combustion is the dominated reaction in the whole process. As the char combustion rate is slower than volatile, the oxygen consumption is also slower than other blends, thus the increment still exists at Z=980 mm, which means that the flame length is prolonged and the primary combustion zone is moved down with the increase in blend ratio, and the stable combustion zone moved down too.

Effect of blend ratio on radial flue gas distribution

The radial gas distribution can also reflect the ignition process. Figures 7, 8, and 9 show the variation of O₂, CO, and CO₂ in radial gas. At Z=20 mm, O₂ concentration is the highest in the center and then gradually decreases from r =0 to r=120 mm, showing that the ignition initially occurs at the peripheral part of the jets, which is consistent with the conclusion achieved from the radial temperature. CO and CO₂ concentrations at this section are very low for all blends, showing that the reaction in this section is weak. At Z=180 mm, O₂ concentration decreases the fastest from 0–40 mm for all blends, and reaches a valley at r=40 mm, but O₂ increases with the increase in blend ratio, while CO and CO₂ concentration also decreased with the increase in blend ratio, which indicate that the ignition region is mainly in the periphery of the jet center and has not been fully ignited in this section. The reaction intensity is low for the higher blend ratio, which is caused by the lower reactivity of SC in blended fuel. At Z=340 mm, O₂ concentration is nearly 0 in the region of r=±40 mm, but CO concentration is over 10000 ppm and CO₂ concentration is over 10% for all blends except that of the 100% blend, showing that the fuel has been ignited thoroughly and is in a stable combustion state. The concentrations of CO and CO₂ are inversely proportional to the blend ratio, showing that the combustion intensity is weakened with the increase in blend ratio. O₂ concentration is still very high for the 100% blend, which is similar to that at Z=180 mm, indicating that the reaction is still weak in this section and ignition has not been stable. At Z= 500 mm, O₂ and CO₂ concentrations for the 0% and 30% blends are similar to that of the former cross section, showing that the combustion for the two blends is very stable. O₂ concentration are nearly zero at r=60 mm for the 45% and 60% blends, showing that the reaction zone continues to expand for the two blends in this section. O₂ concentration for the 100% blend ratio is still higher than other blends and there is no region with O₂ concentration of nearly zero, showing that the combustion intensity of the 100% blend is not as strong as that for other blends. CO for all blends except the 100% blend is the highest in this section. CO concentration for the 100% blend is the highest at r=60 mm, which is 17300, showing that pure semi-char has been ignited stably, and the reaction zone deviates farther from the center than other blends. At Z =660 and 820 mm, O₂ concentration for all blends changes a little, showing that all the blends are in a stable state and the reaction is at a reducing atmosphere.

Effect of blend ratio on NO_x emission and burnout

Effect of blend ratio on radial NO_x and HCN distribution

Figures 10 and 11 show the radial NO_x and HCN distribution. HCN is decreased with the increase in blend ratio in all sections, and the concentration for the 100% blend is nearly zero for all sections. There are mainly two ways in the formation of HCN [²²], to produce HCN by the rapid decomposition of fuel-N at high temperatures [²³], and to produce HCN by the reaction of C(N) caused by the NO/char reaction and H₂, whose specific mechanism can be expressed as [²⁴]

(1)\mathrm{NO}+\mathrm{2C}\to \mathrm{C(N)}+\mathrm{CO\; or\; C(O)}

(2)\mathrm{C}(\mathrm{N})+\mathrm{1}/\mathrm{2}{\mathrm{H}}_{\mathrm{2}}\mathrm{or\; H}\to \mathrm{C}(\mathrm{H},\mathrm{N})\to \mathrm{HCN}

(3)\mathrm{C}(\mathrm{N})+\mathrm{C}(\mathrm{H})\to \mathrm{C}(\mathrm{H},\mathrm{N})\to \mathrm{HCN}

Before the HCN concentration reaches the peak value corresponding to Z=500 mm, the HCN mainly comes from the volatile-N and is as the intermediate to produce NO_x. The decrease at Z=660 mm is due to the fact that it reacts with NO because both HCN and NO are decreased. A little increase at Z=820 mm comes from the reaction between NO and CH_i. These results indicate that volatile-N first converts into HCN as the intermediate product and then HCN is oxidized to NO_x, and after that part of the NO_x is reduced to N₂ or other products while char-N is directly converted to NO_x through heterogeneous reaction on char surface for 100%.

Nearly no NO_x is generated at Z=20 mm for all blends. Then, it gradually increases from the periphery of the axial center at Z=180 mm, and the higher NO_x concentration in the center than that on its two sides is caused by the corrected calculation at the 6% O₂ under a high oxygen concentration condition. All blends, except the 100% one, has a peak value at Z=340 mm in this section, and NO_x is mainly distributed near the center, which shows that the fuel combustion is most intense in this cross section. When Z exceeded 340 mm, NO_x is decreased compared to that of the former section, with the exception of the 100% blend, which still increases, showing that these blends, but the 100% blend, are in a reduced condition at this distance. A possible reason is that the ignition distance is longer at the blend ratio of 60% and 100%, the oxygen content after Z= 500 mm is relatively higher, the reduction zone is delayed, and the temperature is also higher, which are in favor of NO generation. The 100% blend comes into reduction zone after Z=660 mm, showing that the primary zone is obviously prolonged for pure semi-char combustion. At Z=820 mm, flue gas components for the blends less than 100% are stable, which can be taken as the outlet of primary combustion zone at Z=820 mm. In this section, the NO_x concentration is 442, 407, 425, 481, and 510 mg/m³ at 6% O₂ for the blend ratios of 0%, 30%, 45%, 60%, and 100%, respectively. These results indicate that a suitable blend ratio of semi-char in bituminous coal can reduce emission of NO_x, which is no more than 45% as is found in this paper.

Effect of blend ratio on burnout

Figure 12 shows central carbon burnout of different blends at Z=820 mm. It can be observed that carbon burnout is the highest for the 0% blend. The burnout rate for the blends of 30% and 45% is lower than that of the 0% blend. The reason for this is that there exists a competition between volatile and semi-char to react with oxygen. It is obviously noticed that volatile reacts more easily with oxygen for the two blends, and char combustion is suppressed, thus the overall burnout rate is lower. When the blend ratio is 60%, the reaction between oxygen and semi-char is dominant, and the combustion is more intense than that of the 30% and 45% blends once it is ignited stably, which leads to a higher burnout rate than that of the blends of 30% and 45%. For the 100% blend, the reaction zone is much moved down due to its longer standoff distance, thus the burnout rate is the lowest of all the blends at this distance. There is a small decrease in burnout ratio when the blend ratio is less than 45% and NO_x emission is close to the 0% blend. The combined NO_x emission concentration and semi-char blend ratio in utilization may be not larger than 45%. Besides, the carbon burnout rate is far below the rate at the furnace outlet in the boiler. The reason for this is that the excess air ratio is 0.8 for different blend ratios, the primary combustion zone is in an oxygen deficient environment, and the combustion time is also short in this zone, which lead to a low carbon burnout rate at the exit of primary combustion zone.

Conclusions

A pilot-scale experiment was conducted to study the effect of semi-char blend ratio on combustion, NO_x emission, and burnout characteristics during co-combustion with bituminous coal. The conclusions are as follows:

The ignition behavior of co-combustion is weakened with the increase in blend ratio of SC, the ignition standoff distance increases from 142 mm to 483 mm with the increase in semi-char blend ratio from 0% to 100%, and the ignition temperature increases from 932°C to 1100°C.

The distance between the burner outlet and stable combustion zone increases with the increase in blend ratio. However, the delay effect is not obvious when the blend ratio is less than 45% and can be neglected when the blend ratio is less than 30%.

The NO_x emission at the primary outlet center first decreases as SC blend ratio increases from 0% to 30%, and then gradually increases after the ratio is over 60%. The NO_x concentration for the 45% blend is 425 mg/m³ at 6% O₂. There are two reasons for the lowest NO_x emission of the 30% blend. One is that the combustion intensity for the 30% blend is weaker than that for the 0% blend, thus the production of NO_x for the 30% blend is less than that for the 0% blend. The other is that there are more carbon content in the 30% blend than in the 0% blend, which is favorable for reducing NO_x. The highest NO_x emission is mainly caused by its longer standoff distance.

The burnout rate first decreases as the blend ratio increases from 0% to 45%, and then increases at a blend ratio of 60%. The 0% blend has the highest burnout rate while the 100% blend has the lowest burnout rate of 72.32%. The decline of the burnout rate is mainly caused by the competition of oxygen between volatile and char. Combining the NO_x emission and burnout rate, the suitable blend ratio for semi-char in co-combustion is no more than 45%.

References

[1] Gong Z, Liu Z, Zhou T, Lu Q, Sun Y. Combustion and NO emission of Shenmu char in a 2 MW circulating fluidized bed. Energy & Fuels, 2015, 29(2): 1219–1226

[2] Yao Y, Lü Q, Zhu J, Ouyang Z, Zhou Z. Experimental study on preheating process of fine char in a circulating fluidized bed. Proceedings of CSEE, 35(17): 4417–4422 (in Chinese)

[3] de Foy B, Lu Z, Streets D G. Satellite NO₂ retrievals suggest China has exceeded its NO_x reduction goals from the twelfth Five-Year Plan. Scientific Reports, 2016, 6(1): 35912

[4] Ulloa C, Borrego A G, Helle S, Gordon A L, García X. Char characterization and DTF assays as tools to predict burnout of coal blends in power plants. Fuel, 2005, 84(2–3): 247–257

[5] Wang P, Wang C, Tao Z, Yuan M, Du Y, Zhang J, Che D. Co-combustion characteristics of semi-coke and coal under air condition. In: Proceedings of ASME 2018 Power Conference, Lake Buena Vista, FL, USA, 2018

[6] Zhang J, Jia X, Wang C, Zhao N, Wang P, Che D. Experimental investigation on combustion and NO formation characteristics of semi-coke and bituminous coal blends. Fuel, 2019, 247(7): 87–96

[7] Huang Q, Li S, Li G, Zhao Y, Yao Q. Reduction of fine particulate matter by blending lignite with semi-char in a down-fired pulverized coal combustor. Fuel, 2016, 181(10): 1162–1169

[8] Liu G, Liu Q, Zhu B, Ren S, Meng F, Lin F. Co-combustion characteristics and reaction kinetics of pine char and anthracite blends. Journal of Iron & Steel Research, 2016, 28(5): 11–17 (in Chinese)

[9] Wang P, Wang G, Zhang J, Lee J Y, Li Y, Wang C. Co-combustion characteristics and kinetic study of anthracite coal and palm kernel shell char. Applied Thermal Engineering, 2018, 143(10): 736–745

[10] Clements B R, Zhuang Q, Pomalis R, Wong J, Campbell D. Ignition characteristics of co-fired mixtures of petroleum coke and bituminous coal in a pilot-scale furnace. Fuel, 2012, 97(6): 315–320

[11] Wang H, Zhang J, Wang G, Xu R, Zhang P, Liu S, Song T. Characteristics and kinetic analysis of co-combustion of brown coal and anthracite. Journal of Thermal Analysis and Calorimetry, 2016, 126(2): 447–454

[12] Faúndez J, Arias B, Rubiera F, Arenillas A, García X, Gordon A L, Pis J J. Ignition characteristics of coal blends in an entrained flow furnace. Fuel, 2007, 86(14): 2076–2080

[13] Sahu S G, Mukherjee A, Kumar M, Adak A K, Sarkar P, Biswas S, Tiwari H P, Das A, Banerjee P K. Evaluation of combustion behaviour of coal blends for use in pulverized coal injection (PCI). Applied Thermal Engineering, 2014, 73(1): 1014–1021

[14] Zeng G, Sun S, Yang X, Zhao Y, Zhao Z, Ye Z, Gao J. Effect of the primary-air velocity on ignition characteristics of bias pulverized-coal jets. Energy & Fuels, 2017, 31(3): 3182–3195

[15] Li Z, Liu Y, Chen Z, Zhu Q, Jia J, Li J, Wang Z, Qin Y. Effect of the air temperature on combustion characteristics and NO_x emissions from a 0.5 MW pulverized coal-fired furnace with deep air staging. Energy & Fuels, 2012, 26(4): 2068–2074

[16] Prationo W, Zhang J, Cui J, Wang Y, Zhang L. Influence of inherent moisture on the ignition and combustion of wet Victorian brown coal in air-firing and oxy-fuel modes: part 1: the volatile ignition and flame propagation. Fuel Processing Technology, 2015, 138(10): 670–679

[17] Liu C, Li Z, Zhao Y, Chen Z. Influence of coal-feed rates on bituminous coal ignition in a full-scale tiny-oil ignition burner. Fuel, 2010, 89(7): 1690–1694

[18] Faúndez J, Arenillas A, Rubiera F, García X, Gordon A L, Pis J J. Ignition behaviour of different rank coals in an entrained flow reactor. Fuel, 2005, 84(17): 2172–2177

[19] Molina A, Murphy J J, Winter F, Haynes B S, Blevins L G, Shaddix C R. Pathways for conversion of char nitrogen to nitric oxide during pulverized coal combustion. Combustion and Flame, 2009, 156(3): 574–587

[20] Jovanovic R, Milewska A, Swiatkowski B, Goanta A, Spliethoff H. Sensitivity analysis of different devolatilisation models on predicting ignition point position during pulverized coal combustion in O₂/N₂ and O₂/CO₂ atmospheres. Fuel, 2012, 101(11): 23–37

[21] Riaza J, Khatami R, Levendis Y A, Alvarez L, Gil M V, Pevida C, Rubiera F, Pis J J. Single particle ignition and combustion of anthracite, semi-anthracite and bituminous coals in air and simulated oxy-fuel conditions. Combustion and Flame, 2014, 161(4): 1096–1108

[22] Park D C, Day S J, Nelson P F. Nitrogen release during reaction of coal char with O₂, CO₂, and H₂O. Proceedings of the Combustion Institute, 2005, 30(2): 2169–2175

[23] Glarborg P, Jensen A D, Johnsson J E. Fuel nitrogen conversion in solid fuel fired systems. Progress in Energy and Combustion Science, 2003, 29(2): 89–113

[24] Lin J Y, Zhang S, Zhang L, Min Z, Tay H, Li C Z. HCN and NH₃ formation during coal/char gasification in the presence of NO. Environmental Science & Technology, 2010, 44(10): 3719–3723

RIGHTS & PERMISSIONS

Higher Education Press