Oxy-Fuel Combustion Characteristics of Pulverized Coal under O2/Recirculated Flue Gas Atmospheres

Oxy-fuel combustion is an effective technology for carbon capture and storage (CCS). Oxy-combustion for coal-fired power stations is a promising technology by which to diminish CO2 emissions. Unfortunately, little attention has been paid to the oxy-combustion characteristics affected by the combustion atmosphere. This paper is aimed at investigating the oxy-fuel combustion characteristics of Australian coal in a 0.3 MWth furnace. In particular, the influences of various oxygen flow rates and recirculated flue gas (RFG) on heating performance and pollutant emissions are examined in O2/RFG environments. The results show that with increases in the secondary RFG flow rate, the temperatures in the radiative and convective sections decrease and increase, respectively. At a lower oxygen flow rate, burning Australian coal emits lower residual oxygen and NO concentrations. In the flue gas, a high CO2 concentration of up to 94.8% can be achieved. Compared to air combustion, NO emissions are dramatically reduced up to 74% for Australian coal under oxy-combustion. Note that the high CO2 concentrations in the flue gas under oxy-coal combustions suggest great potential for reducing CO2 emissions through carbon capture and storage.


Introduction
Electricity is a crucial part of most industries and livelihoods. Although there has been rapid development of alternative energy, the major power supply is still mainly produced from combusting fossil fuels like coal and natural gas. Taking this into consideration for coal-fired power plants, it is necessary to develop novel technologies to mitigate their impact on the environment from pollutant emissions and to improve efficiency. In 1982, Abraham et al. [1] proposed the concept of oxy-fuel combustion to improve enhanced oil recovery by producing CO 2 -rich flue gas. Prior to combustion, an air separation unit is utilized to produce the oxygen required for oxy-fuel combustion. Using this method, fuel only combusts in an oxygen-diluted environment with recirculated flue gas (RFG), and CO 2 and H 2 O are mainly produced in the flue gas (FG). Therefore, a high concentration CO 2 stream occurs, because CO 2 can be easily separated. It is worth noting that oxy-fuel combustion can reduce the amount of flue gas treated, can produce high levels of CO 2 , and is feasible for retrofitting current power plants.
Oxy-coal combustion with recycled flue gas was first carried out by the International Flame Research Foundation (IFRF) at a pilot scale in the 1990s [2]. Subsequently, air separation technology has made considerable progress, promoting oxy-fuel combustion as it becomes a viable option for carbon is proportional to the fuel S, which is similar to combustion in air. It has been reported that the in-furnace SO 2 concentration significantly increases in an O 2 /RFG atmosphere. Nevertheless, in terms of the total sulfur mass output, SO 2 emissions are lower during oxy-fuel combustion than during air-combustion [2]. Recently, Duan et al. [13] found that when the combustion pressure was increased, the SO 2 emissions were significantly reduced, due to the enhanced self-desulfurization of ash under high pressure. This paper is aimed at investigating the oxy-combustion characteristics of Australian coal in a 0.3 MW th furnace. The influence of various oxygen flow rates and recirculated flue gas (RFG) on heating performance and pollutant emissions are examined in O 2 /RFG environments. Figure 1 schematically shows the major experimental apparatus, which is called the 0.3 MW th multi-fuel combustion test facility [30]. It contains a vertical furnace with a multi-fuel burner, a fuel and oxidant (air or oxygen/RFG) supply system, a post-combustion treatment unit (including a bag filter and heat exchanger (HEX)), and a measuring apparatus. More details are provided as follows.

The 0.3 MW th Multi-Fuel Combustion Test Facility
Appl. Sci. 2020, 10, x FOR PEER REVIEW 3 of 17 furnace SO2 concentration significantly increases in an O2/RFG atmosphere. Nevertheless, in terms of the total sulfur mass output, SO2 emissions are lower during oxy-fuel combustion than during aircombustion [2]. Recently, Duan et al. [13] found that when the combustion pressure was increased, the SO2 emissions were significantly reduced, due to the enhanced self-desulfurization of ash under high pressure. This paper is aimed at investigating the oxy-combustion characteristics of Australian coal in a 0.3 MWth furnace. The influence of various oxygen flow rates and recirculated flue gas (RFG) on heating performance and pollutant emissions are examined in O2/RFG environments. Figure 1 schematically shows the major experimental apparatus, which is called the 0.3 MWth multi-fuel combustion test facility [30]. It contains a vertical furnace with a multi-fuel burner, a fuel and oxidant (air or oxygen/RFG) supply system, a post-combustion treatment unit (including a bag filter and heat exchanger (HEX)), and a measuring apparatus. More details are provided as follows.  Figure 2a depicts the combustion furnace, which is composed of three sections: the radiative, convective, and flue gas sections. The adjustable swirl burner is set at the top of the radiative section, and is available to burn both liquid and solid fuel, as shown in Figure 2b. The swirl strength varies with changes in the angle of the swirl generator. The radiative section is a down-fire combustion chamber (with a maximal heat release rate of 0.3 MWth), which has a 56 cm inner diameter and is 305 cm in height. It is composed of five sections of steel-made shell, with three layers of firebricks inside. Each section has a window available to observe flame characteristics and stability. In addition, each section is equipped with an R-type thermocouple, (Tw1-Tw5), as illustrated in Figure 2a. The convective section is composed of six sections of steel-made shell, and is equipped with five K-type thermocouples (Tg1-Tg5). The upstream parts of the convective section near the radiative section are cooled with a water-cooling system because of high-temperature flue gas, and then the downstream parts are cooled by an air-cooling system. The third section is the flue gas section, which includes a gas sampling port, a pulse-jet cleaning bag house, a heat exchanger (shell-and-tube type), and a chimney. The operating pressure is maintained at +100 Pa in order to prevent air leakage [ Figure 2a depicts the combustion furnace, which is composed of three sections: the radiative, convective, and flue gas sections. The adjustable swirl burner is set at the top of the radiative section, and is available to burn both liquid and solid fuel, as shown in Figure 2b. The swirl strength varies with changes in the angle of the swirl generator. The radiative section is a down-fire combustion chamber (with a maximal heat release rate of 0.3 MW th ), which has a 56 cm inner diameter and is 305 cm in height. It is composed of five sections of steel-made shell, with three layers of firebricks inside. Each section has a window available to observe flame characteristics and stability. In addition, each section is equipped with an R-type thermocouple, (T w1 -T w5 ), as illustrated in Figure 2a. The convective section is composed of six sections of steel-made shell, and is equipped with five K-type thermocouples (T g1 -T g5 ). The upstream parts of the convective section near the radiative section are cooled with a water-cooling system because of high-temperature flue gas, and then the downstream parts are cooled by an air-cooling system. The third section is the flue gas section, which includes a gas sampling port, a pulse-jet cleaning bag house, a heat exchanger (shell-and-tube type), and a chimney. The operating pressure is maintained at +100 Pa in order to prevent air leakage [31].

The 0.3 MWth Multi-Fuel Combustion Test Facility
system. The air supply system includes primary air, used to fluidize solid fuel, and secondary air, used as a source of oxidants. Both of their supply rates are controlled by a valve opening intended to meet the oxidant demand. In addition, both primary air and secondary air can be replaced with RFG, which herein is called either 1st RFG or 2nd RFG, respectively, in the oxy-fuel combustion case. Oxygen (purity 99.9%), fed by a commercial high pressure oxygen cylinder, is introduced and mixed with the 2nd RFG to form an O2/RFG mixture, as shown in Figures 1 and 2b. Two kinds of flue gas are defined, wet flue gas and dry flue gas, depending on whether or not the flue gas flows through the shell and tube heat exchanger. Primary air and secondary air in airfiring mode can be replaced with wet RFG or dry RFG in oxy-fuel combustion mode to meet the experimental needs, as shown in Figures 1 and 2b

Fuel and Oxidant/Recirculated Flue Gas (RFG) Supply System
The fuel supply system is comprised of three parts: gaseous fuel (liquefied petroleum gas, or LPG), liquid fuel (fuel oil or diesel), and solid fuel (pulverized coal or biomass), as illustrated in Figure 2b. The LPG, fed from a high-pressure cylinder, serves as a pilot flame for the combustion furnace. The liquid fuel (diesel) is stored in a daily tank, transported by a pump, and is used for preheating the furnace. The solid fuel, fed by a coal feeder, is transported by primary air to the burner. The feed rates of the gaseous and liquid fuels are controlled by adjusting the valve opening, while the supply rate of coal depends on the speed of the rotary feeder. An air supply system and an oxygen supply system are combined into an oxidant/RFG supply system. The air supply system includes primary air, used to fluidize solid fuel, and secondary air, used as a source of oxidants. Both of their supply rates are controlled by a valve opening intended to meet the oxidant demand. In addition, both primary air and secondary air can be replaced with RFG, which herein is called either 1st RFG or 2nd RFG, respectively, in the oxy-fuel combustion case. Oxygen (purity 99.9%), fed by a commercial high pressure oxygen cylinder, is introduced and mixed with the 2nd RFG to form an O 2 /RFG mixture, as shown in Figures 1 and 2b.
Two kinds of flue gas are defined, wet flue gas and dry flue gas, depending on whether or not the flue gas flows through the shell and tube heat exchanger. Primary air and secondary air in air-firing mode can be replaced with wet RFG or dry RFG in oxy-fuel combustion mode to meet the experimental needs, as shown in Figures 1 and 2b. In this study, only dry RFG is used. The dry RFG can be used to transport coal particles from coal feeders to furnaces and other uses, while the wet RFG can be used to control the combustion temperature, due to its much higher water vapor content than dry RFG. Although wet RFG is thermodynamically advantageous over dry RFG, the dry RFG will be needed if the water vapor or SO 2 content in the RFG is limited [21]. In addition, to focus on the effect of CO 2 , the dry RFG is applied in the combustion experiments in this study. As a fuel carrier gas, the volume flow rate of the 1st RFG is maintained at 70 Nm 3 /h to ensure a sufficient gas velocity to transport pulverized coal from the coal feeder into the furnace. When used as a feed oxidizer gas stream, oxygen is mixed with the secondary RFG to form a CO 2 -rich mixture of 2nd RFG and O 2 , and the pulverized coal is then combusted with oxygen/RFG in the combustion chamber.

Flue Gas Treatment Unit
The post-combustion treatment (flue gas treatment) unit is utilized to remove suspended particles and to cool down the flue gas, for the purpose of reducing pollutant emissions and testing different operating conditions with various RFG formulations. The post-combustion treatment unit is composed of three types of equipment: a pulse-jet cleaning bag house, a heat exchanger, and a flue gas recirculation system. The pulse-jet cleaning bag house has a capacity of 400 Nm 3 /h and a dust load of 2000 mg/m 3 , with a pressure pulse of 150 mmAq and a maximum temperature resistance of 400 • C.

Measurement Devices
The control room of the 0.3 MW th furnace is equipped with both a monitoring and control system, which are used to monitor the real-time status of the entire test furnace, e.g., primary and secondary air supply rate, coal feed rate, recirculated flue gas flow rate, stack pressure, oxygen flow rate, and temperature distributions (in the radiative section and convective section).
The sample ports for measuring temperatures using R-and K-type thermocouples are demonstrated in Figure 2a. The measured ranges are from 0 • C to 1450 • C for R-type thermocouples and from 0 • C to 1250 • C for K-type thermocouples, with accuracies of ±1.5 • C and ±2.2 • C, respectively. The measured temperature fluctuated, but about 95% of the recorded temperature fell within ±10 • C of the mean temperature. By assuming those errors are independent of each other, the overall errors are ±[(10) 2 + (1.5) 2 ] 0.5 = 10.11 • C and ±[(10) 2 + (2.2) 2 ] 0.5 = 10.24 • C, respectively, by the root-sum-square method [32].

Experimental Procedures
The Australian coal tested in this research is sub-bituminous coal. The particle size of pulverized coal is an important factor influencing flame stability and combustion characteristics. A pulverizing mill (Raymond mill) was used here to pulverize the Australian coal. Then, the coal particles were sieved and selected to ensure a suitable size of 75 µm (200 mesh), based on the industry standard [33]. Finally, the pulverized fuels were introduced into the combustion chamber by primary air/RFG and combusted with secondary air (or a mixture of O 2 and RFG). Table 1 provides the proximate and ultimate analysis of Australian coal. All of the experiments presented in this study adhere to the following three procedures [31,34]. The first step is furnace preheating. The goal of furnace preheating is to minimize the effect of furnace temperature fluctuations on combustion characteristics and to provide a steady, high-temperature atmosphere for test fuel combustion. The second procedure is combustion tuning. The aim of combustion tuning is to reduce either the air or flue gas flow rate as low as possible, in order to obtain the minimum excess oxygen for complete combustion. The third procedure is long-term, steady combustion. With appropriate air/oxygen or RFG flow rates, test fuel combustions require 4 h to investigate long-term combustion characteristics.
The coal feed rate depends on the experimental conditions, and a detailed coal feed rate for each experiment is provided in Table 2, which also shows the specific parameters of the coal in the oxy-combustion experiments. The parameters include the coal feed rate (heat release rate), the primary RFG flow rate, the secondary RFG flow rate, and the oxygen feed rate. In this study, the heat release rate was fixed at 165 kW th , and the primary RFG was maintained at a constant value (70 Nm 3 /h), in order to investigate the effects of varied oxygen and secondary RFG feed rates on oxy-combustion characteristics. The purpose of furnace preheating is to provide a steady, high-temperature environment for the subsequent experiment, and to mitigate the influence of furnace temperature variations that occur during the experiment on the combustion characteristics. The preheating process can be divided into three phases. First, diesel with a calorific value of 36.14 MJ/L is combusted with air. Initially, the diesel feed rate is maintained at 12 L/h with a secondary air supply rate of 130 Nm 3 /h for 3 h. Then, the diesel flow rate increases to 16 L/h with a secondary air supply rate of 210 Nm 3 /h for the next 3 h. The process ends with the diesel flow rate rising incrementally to 20 L/h at a secondary air flow rate of 260 Nm 3 /h for long-term (about 16 h) preheating, to ensure that the walls of the furnace reach a quasi-steady condition. That is, so the temperature distributions do not vary dramatically with time, as shown in Figure 3.

The Oxy-Fuel Combustion Process
Oxy-fuel combustion is different from air combustion. The oxidant is supplied from pure oxygen instead of air. As a result, the total feed rate of the O2/recirculated flue gas (RFG) mixture decreases significantly, and so does the flue gas heat loss. In this regard, the flame temperature increases significantly. Thus, it is necessary to utilize RFG to lower the temperature. Figure 4 illustrates the configuration of the oxygen and RFG supply system. Primary recirculated flue gas (1st RFG) and secondary recirculated flue gas (2nd RFG) flow through the primary and secondary air supply systems, respectively. That is, air is replaced with RFG. Thus, oxygen mixes with secondary RFG to form an O2/RFG mixture. In oxy-fuel combustion with RFG, the flow rates of primary RFG (1st RFG) and oxygen are fixed, and the flow rate of secondary RFG (2nd RFG) is gradually adjusted. The residual oxygen in the flue gas involved in the combustion process is recirculated back to the furnace. Therefore, the oxygen/RFG blend ratio should take the residual oxygen [O2] into consideration. In this study, the effect of the oxygen/RFG blend ratio ( O2 Ψ ) on the combustion characteristics of oxy-coal firing is examined. The oxygen/RFG blend ratio (i.e., the oxygen concentration (vol %) used in the RFG for each of the oxyfuel tests) is defined as [30] (vol.%)

The Oxy-Fuel Combustion Process
Oxy-fuel combustion is different from air combustion. The oxidant is supplied from pure oxygen instead of air. As a result, the total feed rate of the O 2 /recirculated flue gas (RFG) mixture decreases significantly, and so does the flue gas heat loss. In this regard, the flame temperature increases significantly. Thus, it is necessary to utilize RFG to lower the temperature. Figure 4 illustrates the configuration of the oxygen and RFG supply system. Primary recirculated flue gas (1st RFG) and secondary recirculated flue gas (2nd RFG) flow through the primary and secondary air supply systems, respectively. That is, air is replaced with RFG. Thus, oxygen mixes with secondary RFG to form an O 2 /RFG mixture.

The Oxy-Fuel Combustion Process
Oxy-fuel combustion is different from air combustion. The oxidant is supplied from pure oxygen instead of air. As a result, the total feed rate of the O2/recirculated flue gas (RFG) mixture decreases significantly, and so does the flue gas heat loss. In this regard, the flame temperature increases significantly. Thus, it is necessary to utilize RFG to lower the temperature. Figure 4 illustrates the configuration of the oxygen and RFG supply system. Primary recirculated flue gas (1st RFG) and secondary recirculated flue gas (2nd RFG) flow through the primary and secondary air supply systems, respectively. That is, air is replaced with RFG. Thus, oxygen mixes with secondary RFG to form an O2/RFG mixture. In oxy-fuel combustion with RFG, the flow rates of primary RFG (1st RFG) and oxygen are fixed, and the flow rate of secondary RFG (2nd RFG) is gradually adjusted. The residual oxygen in the flue gas involved in the combustion process is recirculated back to the furnace. Therefore, the oxygen/RFG blend ratio should take the residual oxygen [O2] into consideration. In this study, the effect of the oxygen/RFG blend ratio ( O2 Ψ ) on the combustion characteristics of oxy-coal firing is examined. The oxygen/RFG blend ratio (i.e., the oxygen concentration (vol %) used in the RFG for each of the oxyfuel tests) is defined as [30] (vol.%)  In oxy-fuel combustion with RFG, the flow rates of primary RFG (1st RFG) and oxygen are fixed, and the flow rate of secondary RFG (2nd RFG) is gradually adjusted. The residual oxygen in the flue gas involved in the combustion process is recirculated back to the furnace. Therefore, the oxygen/RFG blend ratio should take the residual oxygen [O 2 ] into consideration. In this study, the effect of the oxygen/RFG blend ratio (Ψ O2 ) on the combustion characteristics of oxy-coal firing is examined. The oxygen/RFG blend ratio (i.e., the oxygen concentration (vol %) used in the RFG for each of the oxy-fuel tests) is defined as [30]  Q 2nd RFG are the volume flow rates of the primary and secondary RFG, respectively. Therefore, the oxygen concentration (vol %) utilized in the O 2 /RFG mixture for each of the oxy-fuel combustion experiments can be calculated.
It is significant to investigate the influence of flue gas recirculation on the characteristics of oxy-fuel combustion, in order to maintain the similar combustion temperature and heat transfer characteristics in the boiler as in the conventional coal-fired power plant. In this study, the feeding rate of the pulverized coal was controlled at 21.54 kg/h (with heat releasing rate of 165 kW th ). The volume flow rate of the primary RFG (1st RFG) was maintained at 70 Nm 3 /h to ensure that gas velocity was sufficiently large to carry pulverized coal from the coal feeder to the furnace. Meanwhile, the secondary RFG (2nd RFG) is gradually adjusted from low flow rate to high flow rate. Notably, in order to maintain the oxygen/RFG blend ratio Ψ O2 (calculated using Equation (1)) in the range of around 18%-30%, as in [22], and to maintain the turndown ratio of the flow meter, the flow rate of the 2nd RFG must not be less than 50 Nm 3 /h. Therefore, for different recirculated flue gas ratios (oxygen/RFG blend ratios), it is easier to observe the heat transfer characteristics in the radiative zone in a short time. In addition, if the 2nd RFG is lower than 50 Nm 3 /h, there will not be enough high-concentration carbon dioxide (recirculated gas) to absorb the high heat generated during the oxy-coal combustion, resulting in burner damage for long-term operation. Therefore, the 2nd RFG was controlled in the range of 50-220 Nm 3  respectively) were used to assess the influence of the oxygen/RFG blend ratio on oxy-coal combustion characteristics. For each case, the temperatures (in the radiative and convective sections) and the compositions (in the flue gas) were recorded after at least 30 min of stable operation.

Temperature Distributions in Oxy-Combustion of Australian Coal
For Australian coal under oxy-fuel combustion, Figure 5 shows the temperature variations in the radiative and convective sections with an oxygen supply of 41.5 Nm 3 /h, while Figure 6 displays those with an oxygen supply of 38.5 Nm 3 /h. The results show similar trends in the temperature variations under these two operating conditions. That is, when the coal feed rate (21.54 kg/h) and primary RFG (1st RFG) flow rate (70 Nm 3 /h) are fixed at a constant oxygen supply (41.5 Nm 3 /h or 38.5 Nm 3 /h), with an increase in the flow rate of secondary RFG (2nd RFG), the temperature in the radiative section decreases, while the flue gas temperature in the convective section increases, This is because a larger flow rate of 2nd RFG (a higher flue-gas recirculation rate) leads to a lower flame temperature, and therefore a reduction in the radiative heat transfer [14,15]. On the other hand, carbon dioxide and water vapor have high thermal capacities compared to nitrogen, which will lead to an enhancement in the heat transfer in the convective section of the furnace. Thus, higher flue gas recirculation results in a higher gas temperature in the convective section. In addition, a significant temperature drop between T g1 and T g2 will be observed because of the cooling system between the radiative and convective sections.

Emissions in Oxy-Combustion of Australian Coal
For Australian coal with an oxygen supply rate of O2 Q  = 41.5 Nm 3 /h, Figure 7a shows variations in the O2 concentrations and CO2 emissions with the 2nd RFG flow rate, while Figure 7b shows variations in the CO, SO2, and NO emissions with the 2nd RFG flow rate. As shown in Figure  7a, for all operating conditions (with various 2nd RFG flow rates), the residual oxygen concentration in the flue gas ranged from 8% to 10%, while the CO2 concentration in the flue gas ranged between 87% and 90%. On the other hand, as shown in Figure 7b, CO emissions remained nearly constant, with increases in the 2nd RFG flow rate. With increases in the 2nd RFG flow rate, the NO emission decreased, due to the decrease in the peak flame temperature, which inhibits thermal NO formation through the absorption of combustion heat by the relatively cooler flue gas, as well as the reduction

Emissions in Oxy-Combustion of Australian Coal
For Australian coal with an oxygen supply rate of O2 Q  = 41.5 Nm 3 /h, Figure 7a shows variations in the O2 concentrations and CO2 emissions with the 2nd RFG flow rate, while Figure 7b shows variations in the CO, SO2, and NO emissions with the 2nd RFG flow rate. As shown in Figure  7a, for all operating conditions (with various 2nd RFG flow rates), the residual oxygen concentration in the flue gas ranged from 8% to 10%, while the CO2 concentration in the flue gas ranged between 87% and 90%. On the other hand, as shown in Figure 7b, CO emissions remained nearly constant, with increases in the 2nd RFG flow rate. With increases in the 2nd RFG flow rate, the NO emission decreased, due to the decrease in the peak flame temperature, which inhibits thermal NO formation through the absorption of combustion heat by the relatively cooler flue gas, as well as the reduction

Emissions in Oxy-Combustion of Australian Coal
For Australian coal with an oxygen supply rate of . Q O2 = 41.5 Nm 3 /h, Figure 7a shows variations in the O 2 concentrations and CO 2 emissions with the 2nd RFG flow rate, while Figure 7b shows variations in the CO, SO 2 , and NO emissions with the 2nd RFG flow rate. As shown in Figure 7a, for all operating conditions (with various 2nd RFG flow rates), the residual oxygen concentration in the flue gas ranged from 8% to 10%, while the CO 2 concentration in the flue gas ranged between 87% and 90%. On the other hand, as shown in Figure 7b, CO emissions remained nearly constant, with increases in the 2nd RFG flow rate. With increases in the 2nd RFG flow rate, the NO emission decreased, due to the decrease in the peak flame temperature, which inhibits thermal NO formation through the absorption of combustion heat by the relatively cooler flue gas, as well as the reduction in the oxygen concentration in the combustion zone, which reduces the availability of oxygen for the formation of fuel NO. Meanwhile, the SO 2 emissions were higher at a higher 2nd RFG flow rate. In these experiments, the flue gas was recirculated without SO 2 removal. Accordingly, there were significant increases in the SO 2 concentrations in the flue gas, due to the accumulated effects of flue gas recirculation and the reduced volume of the flue gas [35].
Appl. Sci. 2020, 10, x FOR PEER REVIEW 10 of 17 in the oxygen concentration in the combustion zone, which reduces the availability of oxygen for the formation of fuel NO. Meanwhile, the SO2 emissions were higher at a higher 2nd RFG flow rate. In these experiments, the flue gas was recirculated without SO2 removal. Accordingly, there were significant increases in the SO2 concentrations in the flue gas, due to the accumulated effects of flue gas recirculation and the reduced volume of the flue gas [35].  Figure 8a, the residual oxygen concentration in the flue gas ranged between 3% and 5%. Meanwhile, the CO2 concentration in flue gas remains at about 94% (average value), and reached 94.8% when the 2nd RFG flow rate was 170 Nm 3 /h. It is noteworthy that the high CO2 concentration in the flue gas suggests great potential for reducing CO2 emissions through carbon capture and storage, thus mitigating global climate change. Figures 7a and 8a also demonstrate that a higher oxygen supply results in a higher residual O2 and less CO2 in the flue gas. The main reason for these results is that the flue gas contains O2, CO2, and other gaseous molecules, and CO2 and O2 are the two main components. Therefore, Figures 7a  Figure 8a, the residual oxygen concentration in the flue gas ranged between 3% and 5%. Meanwhile, the CO 2 concentration in flue gas remains at about 94% (average value), and reached 94.8% when the 2nd RFG flow rate was 170 Nm 3 /h. It is noteworthy that the high CO 2 concentration in the flue gas suggests great potential for reducing CO 2 emissions through carbon capture and storage, thus mitigating global climate change.
from 3% to 5% [37]. It is worth noting that the CO2 concentration in the flue gas remained at a high average value, about 94%, while the residual O2 concentration in the flue gas was controlled in the range of 3%-5%. This result is in good agreement with the findings of Khare et al. [37]. In the present study, it was also found that in order to achieve the same performance in an existing furnace, the oxygen concentration (about 23%) in oxy-fuel combustion is higher than the oxygen concentration in air, as reported by Wang et al. [38].    7a and 8a also demonstrate that a higher oxygen supply results in a higher residual O 2 and less CO 2 in the flue gas. The main reason for these results is that the flue gas contains O 2 , CO 2 , and other gaseous molecules, and CO 2 and O 2 are the two main components. Therefore, Figures 7a and 8a indicate that a lower residual O 2 concentration corresponds to a higher CO 2 concentration [36]. The oxygen concentration in the flue gas at . Q O2 = 38.5 Nm 3 /h is lower than that at . Q O2 = 41.5 Nm 3 /h; therefore, the CO 2 concentration at . Q O2 = 38.5 Nm 3 /h is larger than that at . Q O2 = 41.5 Nm 3 /h. In air-firing mode, 20% excess air is typically utilized. Oxy-coal combustion requires excess O 2 to reach a residual O 2 concentration in the flue gas similar to that of air combustion, which ranges from 3% to 5% [37]. It is worth noting that the CO 2 concentration in the flue gas remained at a high average value, about 94%, while the residual O 2 concentration in the flue gas was controlled in the range of 3%-5%. This result is in good agreement with the findings of Khare et al. [37]. In the present study, it was also found that in order to achieve the same performance in an existing furnace, the oxygen concentration (about 23%) in oxy-fuel combustion is higher than the oxygen concentration in air, as reported by Wang et al. [38].  (Figure 7b), perhaps because SO 2 converts into SO 3 with greater oxygen supplementation [39]. As shown in Figure 8b, during oxy-combustion with recirculated flue gas, the concentrations of SO 2 (in the range of 850-1180 ppm) are about three to four times larger than those (302 ppm) in air-combustion conditions [40], which agrees well with Tan et al.'s study [35]. In addition, increasing the secondary RFG flow rate corresponds to the enhancement of swirling in the swirl burner, leading to better mixing and more complete combustion; therefore, the concentration of CO in the flue gas is reduced. It was also found that at a fixed secondary RFG flow rate, a lower oxygen supply rate at . Q O2 = 38.5 Nm 3 /h (Figure 8b) results in a higher CO emission than with a higher oxygen supply rate, at . Q O2 = 41.5 Nm 3 /h (Figure 7b), due to the combined effects of increased oxygen concentration in the inlet oxidizer (RFG/oxygen mixture) and strengthened swirling.
The emission of NO decreased with increases in the 2nd RFG flow rate, due to a lower flame temperature. Operating conditions at an oxygen supply rate of 38.5 Nm 3 /h result in lower NO emissions compared to those with an oxygen supply rate of 41.5 Nm 3 /h. That is, the NO concentration in the flue gas increased with increases in the O 2 concentration in the inlet O 2 /RFG mixture. This phenomenon is in agreement with the findings in the literature [41][42][43]. Okazaki and Ando [41] reported that a decrease in the concentration of NO under oxy-combustion conditions was mainly due to the reburning of the NO in the recirculated flue gas. Hu et al. [42,43] found that an increase in the flow rate of recirculated flue gas led to a decrease in NO emissions. In other words, if the flow rate of the recirculated flue gas decreased, the concentration of oxygen in the burner zone increased, which contributed incrementally to NO emissions. Furthermore, the NO concentration under oxy-combustion conditions was 189.5 ppm (at 6% O 2 ), with a secondary RFG flow rate of 170 Nm 3 /h, while the NO concentration under air combustion conditions was 739 ppm (at 6% O 2 ) [40]. This indicates a 74% NO reduction under oxy-firing conditions compared to the air-firing mode. This significant decrease in NO emissions under oxy-combustion conditions coincides with the findings of Nozaki et al. [44] and our previous work [31].
To summarize, the results for Australian coal under oxy-combustion conditions with varied RFG flow rates and different amounts of oxygen supply indicate the same temperature distribution trends in both the radiative and convective sections of the furnace. That is, the furnace wall temperatures in the radiative section decreased while the gas temperatures in the convective section increased with increases in 2nd RFG. In addition, compared to those at . Q O2 = 41.5 Nm 3 /h, with a lower oxygen supply ( . Q O2 = 38.5 Nm 3 /h), the residual oxygen and NO emissions were lower, while SO 2 , CO 2 , and CO emissions were higher.

Modified Combustion Efficiency (MCE) in Oxy-Combustion of Australian Coal
Combustion efficiency (CE) is defined as the ratio of carbon released as CO 2 divided by the carbon released in the form of all carbon-containing products of combustion (such as CO 2 , CO, non-methane hydrocarbons, and PM 2.5 ) [45,46]. It can be used to determine the completeness of the carbon oxidation that occurs during combustion. Since CE is difficult to measure, the modified combustion efficiency (MCE), which is defined as the ratio of carbon emitted as CO 2 divided by carbon emitted as CO 2 plus CO [47], is typically used as an estimate of CE [48,49]. The modified combustion efficiency (MCE; η) was calculated using the following equation [31,[50][51][52]: where [CO 2 ] is volume concentration of CO 2 on a dry basis, and [CO] is volume concentration of CO on a dry basis. Pure flaming combustion usually exhibits an MCE approaching 99% [49]. The MCEs have also been reported to be greater than 99.9%, such as when burning gaseous fuels (Wu et al. [50]), solid fuels (Arvanitoyannis et al. [51]; Bilsback et al. [52]), and liquid fuels (Huang et al. [31]). In this investigation, CO and CO 2 were measured. However, non-methane hydrocarbons and particulate matter content were not measured. Therefore, Equation (2) was used to calculate the modified combustion efficiency (MCE), as reported in our previous study [31] and other literature [50][51][52]. Based on Equation (2), modified combustion efficiencies above 99.9% were obtained in the present study, which agrees well with Liu and Shao's study [36]. This indicated that excellent combustion efficiency was achieved.
In view of the above discussion, we realized that unburned carbon (UC) in fly ash (FA) is also an indicator of combustion efficiency [53]. The high proportion of UC in FA reflects a large amount of energy loss corresponding to low combustion efficiency. In addition, UC is an obstacle to the beneficial use of FA, especially in the cement and concrete industries. In this study, ash analyses were not carried out. However, before starting this work, a loss on ignition (LOI) analysis of air-fuel combustion was performed in various positions, including the ash pot, convective section, and bag house under the following operating conditions: a coal feed rate of 21.54 kg/h and an air supply rate of 210 Nm 3 /h. It is worth noting that the LOI level of each position was lower than 6% (by mass). In other words, fly ash can meet the CNS 3036 [54] and ASTM C618 [55] requirements. Liu and Shao [36] reported that oxy-coal combustion is expected to achieve higher combustion efficiency (close to 100%) than air-coal combustion (greater than 99%). The modified combustion efficiency (MCE) of oxy-coal combustion was higher than 99.9%, which supports the argument of Liu and Shao [36].

Conclusions
The oxy-combustion characteristics of Australian coal in a 0.3 MW th furnace were investigated. The influence of various oxygen flow rates and recirculated flue gas (RFG) on heating performance and pollutant emissions are examined in O 2 /RFG environments. The general conclusions drawn from results of this study are as follows.
With increases in the secondary RFG flow rate, the temperatures (T w1 -T w5 ) in the radiative section decreased, while the temperatures (T g1 -T g5 ) in the convective section increased. When firing Australian coal in the oxy-combustion mode at an oxygen supply of 38.5 Nm 3 /h, with various RFG flow rates, the CO 2 concentration could be stably maintained at about 94%. High CO 2 concentrations in the flue gas of up to 94.8% can be achieved. It is noteworthy that the high CO 2 concentration in the flue gas suggests great potential for reducing CO 2 emissions through carbon capture and storage, thus mitigating global climate change. Furthermore, a higher O 2 supply rate led to higher O 2 and NO concentrations in the flue gas, while it lowered CO, SO 2 , and CO 2 concentrations in the flue gas when firing Australian coal. With increases in the secondary RFG flow rate, the NO emissions decreased while SO 2 emissions increased. Compared to air combustion, a significant reduction in NO emissions of as high as 74% could be achieved under oxy-combustion conditions.
For oxy-coal combustion, an appropriate amount of RFG is required, which not only reduces the flame temperature, but also enhances the mixing of fuel and oxidant in the furnace, resulting in more complete combustion. High CO 2 concentration in the flue gas of up to 94.8% can be achieved when the residual oxygen concentration in the flue gas is controlled in the range of 3-5 vol %. The energy requirement for CO 2 compression is reduced in all CO 2 sequestration options as the purity of the CO 2 increases. It is suggested that 95% should be a reasonable CO 2 purity [56]. Thus, in this study, the high CO 2 concentration in the flue gas around 95% suggests that it is possible and beneficial to achieve easy carbon capture and storage. While this is true, however, in view of the research results, one realizes that using oxy-fuel combustion requires more capex (oxygen plant, flue gas recirculation) and produces a lower power output (oxygen plant and gas compressor power demand) than a conventional coal plant. In addition, CCS requires storage sites for CO 2 , which are not available in many locations. As such, there are still no commercial oxy-fuel plants in commercial operation, and one of the largest demo plants was the 30 MW th plant at Callide in Australia. The oxy-fuel combustion process is still in the development stage, and a lot of research is still needed to fully clarify the consequences of its implementation in power plants. Furthermore, it would be better to handle economic aspects of operational cost with a comparison with conventional combustion.
In the present study, coal was directly burned under oxy-fuel combustion conditions to achieve high CO 2 concentrations, which is beneficial to carbon capture and storage (CCS). It is recognized that coal gasification is a process that reduces the CO 2 emission and emerges as a clean coal technology. Accordingly, coal can be gasified in a fuel-rich (hypoxic) combustion mode, and then the resulting synthesized gases can be properly burned under oxy-fuel combustion [57]. The coal gasification process is controlled by several operating parameters, including temperature, pressure, reaction time, coal rank, porosity, and choice of catalyst. The reaction temperature is one of the parameters that has the most influence on the gasification reaction. For example, the carbon conversion and gasification rate increased with increases in the reaction temperature. In addition, the mineral matter contained in coal has negative impacts on the environment and equipment, because it produces by-product slag and soot, causes equipment corrosion, and reduces the overall combustion rate. It should be noted that an understanding of the operating parameters influencing coal gasification and the oxy-fuel combustion characteristics of the resulting synthesized gas is important and worth further study.