An experimental investigation of oxy-coal combustion in a 15 kW th pressurized fluidized bed combustor

Pressurized oxy-coal combustion is considered as a promising carbon capture technology due to its potential of high efficiency and low cost in CO 2 capture. However, experimental investigations of oxy-coal combustion under pressurized conditions are far less common than those under atmospheric pressure conditions and hence further research is needed to elucidate the effects of pressure on oxy-coal combustion in terms of combustion performance and emissions etc. In this study, a series of oxy-coal combustion experiments were carried out under the pressures from 0.1 MPa to 0.4 MPa on a 15 kWth fluidized bed combustion system. The effects of the combustion pressure and oxygen concentration in the oxidant on the temperature profile, unburnt carbon, combustion efficiency, fly ash composition and NO x emission were investigated. The experimental results have shown that the CO 2 concentration in the oxy-combustion flue gas under different combustion pressures have all exceeded 90%, which is beneficial to the carbon capture process. An increase in pressure is helpful to reduce the unburnt carbon in the fly ash and improve the combustion efficiency under all the tested oxy-combustion atmospheres. The NO x emission decreases with combustion pressure within the investigated range of 0.1 MPa to 0.4 MPa, while the reduction is more pronounced at lower pressures. Besides, the effect of combustion pressure on the chemical composition of the fly ash is found to be insignificant.


Introduction
Greenhouse gas (GHG) emissions have become an increasing concern as rising GHG emissions are closely related to the climate change observed over past decades, and therefore, the call for the application of CO2 capture and storage (CCS) technologies has become particularly strong in recent years. As one of the most developed CCS technologies, oxy-fuel combustion is considered as technically feasible and economically competitive for future commercial applications [1][2][3]. Research and development activities in oxy-combustion have been associated with both pulverized coal combustion and fluidized bed (FB) combustion [4][5]. Compared with pulverized coal oxy-fuel combustion, a critical advantage of fluidized bed combustion is the capability to reduce the recirculated flue gas flow for a given coal input, which can increase the net power generation efficiency significantly [5]. Besides, experimental results have shown that the advantages of fluidized beds under air combustion mode such as fuel flexibility, lower NOx emissions and better sulfur removal are all inherited by the oxy-fuel combustion mode [6][7][8].
Oxy-fuel combustion has shown advantages and potentials in both combustion and CO2 capture processes. However, the biggest obstacle to its application is the net efficiency loss associated with the cost of the air separation unit (ASU) and compression purification unit (CPU). It is estimated that the net efficiency can be reduced by more than 10% when a conventional air-fired coal power plant is converted to oxy-firing [9][10][11]. In order to improve the efficiency, the pressurized oxy-fuel combustion (POFC) technology has been proposed, and many theoretical and economic system analyses have been carried out on POFC technology [12][13][14][15][16]. Hong et al. [12] concluded that the elevated pressure raised the dew point and the available latent enthalpy of the flue gas, and hence was helpful to recover more thermal energy.
Furthermore, the higher pressure was beneficial to save energy for the downstream CO2 compression work. Gopan et al. [13] evaluated a staged pressurized oxy-combustion system, and the results showed that the penalty in plant efficiency caused by carbon capture could be reduced by 6 %. Similar positive effects of high pressure on the overall plant efficiency and CO2 capture have also been demonstrated by the simulation results of many other researchers [14][15][16]. The optimal pressure has been found to be system dependent as a result of the concurrence of the minimum compression power and the maximum thermal recovery.
Although simulation and theoretical calculation results have showed the advantages of pressurized oxy-fuel combustion technology in increasing the net power generation efficiency, there are few experimental studies on the POFC. Wang et al. [17] and Ying et al. [18] conducted oxy-combustion experiments on the pressurized thermogravimetric analyzer (PTGA) and investigated the effects of combustion pressure and oxygen concentration on the ignition and burnout of pulverized coals. Their results indicated that an increase in combustion pressure or oxygen concentration resulted in a decrease in the ignition temperature and an increase in the combustion rate. Lei et al. [19] studied the release behaviors of SO2, NO and NO2 of a Chinese bituminous coal by a PTGA. It was found that the conversion of fuel-N to NOx increased with increasing pressure. Lasek et al. [20][21] carried out pressurized oxy-fuel combustion experiments in a fluidized bed under the pressures from 0.1 MPa to 0.6 MPa. Their studies mainly focused on the emissions of NOx and N2O under the elevated pressure conditions, and a reduction in NOx emissions with an increase in pressure was observed. One possible reason for the contradictory NOx emission behavior may be the difference between the experimental facilities used by these researchers [19][20][21]. The whole combustion processes (e.g., pyrolysis, ignition and burnout) within the fluidized bed combustors should be quite different from those within the thermo-gravimetric analyzer.
The pressurized oxy-fuel fluidized bed combustion system used in this study was modified from the pressurized air-combustion system developed by the authors [22]. A series of fluidized bed combustion experiments has been conducted under different combustion pressures and oxy-fuel environments in order to elucidate the effects of pressure and O2 concentration on the oxy-coal combustion performance. Fig. 1 shows the schematic of the pressurized oxy-coal combustion system which was modified from the pressurized fluidized bed combustion system used with our previous research on pressurized air combustion [22]. The combustion system was designed as a CFB but it was always operated with the valve in the return leg closed, i.e., as a bubbling fluidized bed combustion system in this study. Only the essential and/or new information of the combustion system is included in this paper as other details can be found from our previous paper [22]. The water cooling probe located inside the bed zone allows the bed temperature to be controlled within a desired range.

Pressurized oxy-coal fluidized bed combustion system
For an industrial oxy-fuel combustion system, the oxidant for combustion is a mixture of pure O2 and recycled flue gas (RFG). As the real flue gas recirculation is not available with the present pressurized fluidized bed combustion system, the oxidant, i.e. the inlet gas to the combustor, was replaced by a premixed flow of CO2 and O2 supplied by gas cylinders. The flow rates of CO2 and O2 were adjusted by mass flow controllers, to achieve a preselected concentration of O2 in the oxidant. The fly ash was collected with each test and then analyzed for chemical composition by use of an X-ray fluorescence analyzer (ARL9800XP+) after the tests. Most of the ash was taken from the pipe at the bottom of the cyclone separator, and only a small amount of the ash was found from the bag filter. Because the combustor was operated under pressure, two valves were used to take ash samples during operation. The detailed process is as follow: Before we took ash samples, both valves on the pipe (Fig. 1) were closed. Then, the first valve was opened so that ash entered into the pipe between the two valves. Next, the first valve was closed, and the ash samples were separated from the combustor. Finally, the second valve was opened, and the ash samples were taken out. The above process was repeated several times during a short time until all the ash was taken out, which ensures the next ash sample was produced under the new combustion condition.

Procedure and operating conditions
Before each experiment, the oxidant (inlet gas) was preheated to 600 C by the Particle size (m) electric preheater, and silica sand (2.3 kg) was introduced into the combustor and preheated to about 500 C. The pressure inside the combustor was adjusted to a desired value by the auto pressure controller. Then, proper amounts of coal and air were fed into the combustion zone continuously and the electric heaters around the combustor (except for the gas preheater) were turned off. Fig. 3 shows the profiles of bed temperatures and flue gas composition in a typical experiment (0.2 MPa).
Once the stable air combustion state was reached, the oxidant was changed from air to oxy-fuel oxidant (O2/CO2). The major operating parameters are summarized in It is worth mentioning that, in order to keep the same superficial gas velocity and excess air coefficient, the oxidant flow rate and coal feeding rate should be increased in proportion to the combustion pressure. As more coal is introduced into the combustor, more heat should be removed by the cooling system in order that the combustion temperature can maintain within a desired and safe range. In this study, due to the limited heat exchange capacity of the water cooling system, the coal

Results and discussion
One important target for oxy-fuel combustion is to obtain high CO2 concentration in the flue gas and hence improve the CO2 capture efficiency. In this study, the oxy-coal previous studies [23][24]. With an increase in combustion pressure, the temperature of the bed zone under the same combustion atmosphere increases, e.g., T1 with oxy-21 atmosphere increases from 790 C to 810 C when the pressure increases from 0.1 MPa to 0.2 MPa. Previous experiments [25] had showed that the higher combustion pressure was beneficial to accelerate the combustion rate of coal particles, and the higher oxygen partial pressure increased not only the reaction rates but also the particle temperature.
The temperature increase with pressure changing from 0.1 MPa to 0.2 MPa is mainly owing to the effect of the elevated pressure on the coal combustion, whereas the temperature increase from 0.2 MPa to 0.4 MPa is a combined result of the higher pressure and lower gas velocity. The lower gas velocity leads to a longer residence time of the coal particle, which is beneficial to the burnout of the coal and release more heat.  As many researchers [2,4] have pointed out, the higher specific heat of CO2 than that of N2 led to a sharp decrease in combustion temperature when the combustion environment was changed from air to oxy-21 atmosphere. In many previous atmospheric oxy-fuel combustion studies [24,26], it was impossible to stabilize the temperature once the combustion environment was changed to oxy-21 atmosphere (21 vol% O2/79 vol% CO2) from air. However, some other studies [20][21]27] showed that this switch only decreased the combustion temperature by less than 100 C, and it was possible to achieve a steady combustion state. The main reason that leads to the difference in the combustion temperature achieved with oxy-21 with different studies lies with the temperature of the oxidant or the inlet gas. It is obvious that the temperature of the oxidant affects the combustion temperature significantly. In this study, the oxidant was heated to about 600 C by the electric preheater, and all the experimental results were based on this experimental setting. Although the high oxidant temperature is helpful to achieve the steady combustion state and maintain the combustion temperature within a suitable range, it is worth mentioning that the energy input of the electric preheater under oxy-21 atmosphere is higher than that under air (even though the oxidant was heated from 30 to 600 C under both conditions), and the negative effect of higher specific heat of CO2 than nitrogen on the combustion temperature is offset by the preheater to some extent. It should also be pointed out that it was not necessary to heat the oxidant (air or the mixture of oxygen and CO2) to 600 C, and we plan to decrease the oxidant temperature from 600 C to 200 C which is more likely to be the case for actual industrial processes in our future experiments.

Unburnt carbon
This section focuses on the discussion of the unburnt carbon in fly ash as the mass

CO emissions
Normalized mass emissions of CO under different pressures and atmospheres are shown in Fig. 7 which reveals that the CO emissions under the oxy-combustion atmospheres are higher than those under air. This is mainly due to the higher concentration of CO2 with oxy-coal combustion. The reactions involving char during the combustion process can be written as: When the oxidant is changed from air to the oxy-atmosphere, the high concentration of CO2 is beneficial to its reduction reaction with carbon (R4), and hence the CO emission increases significantly. As the previous studies [26][27][28]  other studies [27][28], the higher CO emission in this study is mainly caused by the lower height of the combustor. In addition, previous studies [4,[29][30] showed that the O2 staging and recycled flue gas were beneficial to reduce CO emissions. In practice, about 70-75 % flue gas will be circulated back to the combustor and CO emissions will be reduced proportionally [30].

NOx emissions
Both NO and NO2, collectively termed as NOx, have been measured in each test.
NO has always been the dominant part of NOx (> 95%) for all the tests of this study.
The results of NOx emissions under different combustion pressures and atmospheres are plotted in Fig. 9. Fig. 9 shows that the NOx emission decreases when the combustion atmosphere is switched from air to oxy-21 and increases with an increase in O2 concentration in the oxidant. This agrees with most of the previous studies [21,[27][28] where NOx emissions were found to decrease in the order of air > oxy-30 > oxy-21 under the same pressure. Reactions (5)-(7) show the main fuel-NOx formation mechanism during the combustion process. The fuel nitrogen is released with the volatiles and converted to the NOx precursors, i.e., cyanide (HCN) and ammonia (NH3) etc. Then, depending on the reaction conditions, NH3 could be oxidized to NO (R6) or The NOx emissions are mainly affected by O2 concentration, temperature and CO concentration under the same combustion pressure. Generally, the higher O2 concentration and higher temperature often lead to higher NOx emissions while the higher CO concentration is helpful to reduce NOx emissions. The reduction of NOx emissions resulted from the change of combustion environment from air to oxy-21 is the combined effects of lower combustion temperature and higher CO concentration [32]. As shown in Fig. 4 and Fig. 7, changing the environment from air to oxy-21 leads to a sharp decrease in the bed zone temperature at 0.1 MPa (from 850 to 790 C), while the CO emission increases significantly from 460 to 2185 mg/MJ, which is beneficial to the reduction of NO emission (R8). An increase in NOx emission is observed when the combustion environment is switched from oxy-21 to oxy-25 or oxy-30 as the higher partial pressure of O2 and the higher combustion temperature result in more NH3 or other fuel-N intermediates being oxidized to NOx [33]. Besides, the lower CO emissions also caused the higher NOx emission as less NO was reduced by CO (R8).
The lower gas velocity under oxy-25 or oxy-30 than that under oxy-21 atmosphere should be beneficial to the reduction of NOx emissions (R8). However, the higher NOx emissions under oxy-25 or oxy-30 atmosphere indicates that the effect of lower gas velocity on NOx emissions is inferior to the effect of higher O2 concentration and higher combustion temperature in this study.
As shown in Fig. 4

Elemental analysis and chemical composition of fly ash
The results of the elemental analysis of the fly ash samples are shown in Table 3.  Table 3. From Table 3 Where Ashcoal and Ashfly ash are the mass fractions of ash in the raw coal and the collected fly ash, respectively. Carboncoal, Hydrogencoal and Nitrogencoal are the carbon, hydrogen and nitrogen contents of coal, respectively, whereas Carbonfly ash, Hydrogenfly ash and Nitrogenfly ash are the carbon, hydrogen and nitrogen contents of the fly ash, respectively.
The chemical compositions of fly ash (excluding the unburnt carbon/combustibles) under different combustion pressures and atmospheres are presented in Fig. 10 and Fig.   11. As many researchers had indicated that oxy-fuel combustion didn't have a significant effect on the chemical composition of the fly ash [4,37], the compositions of the fly ash under the same pressure are almost identical. It can be seen from Fig. 10 that the fly ash mainly consists of SiO2, Al2O3 and Fe2O3 while the other species only make up small fractions. Agreeing with most of the previous studies [28,[38][39], the mass fraction of the solid-state SO3 in the fly ash decreases in the order air > oxy-21 > oxy-25 > oxy-30. The reduction is mainly the result of the higher combustion temperature, which increases the release of coal-S to gaseous SO2 and SO3, and hence results in lower S retention in the fly ash. Besides, it is observed that the mass fraction of Fe2O3 increases as the oxidant is changed from air to oxy-21 atmosphere, and then it decreases with an increase in O2 concentration in the oxidant. The higher mass fraction of Fe2O3 under oxy-21 atmosphere than that under air contradicts with the experimental results from Sheng et al. [4,40], which showed that more iron melted into the glass silicates and less was oxidized. It is believed that the differences in reactors (drop tube furnace and fluidized bed) result in the opposite trend, however, further research is needed to confirm this. The results indicate that an increase in pressure from 0.1 to 0.4 MPa has no significant impact on the ash composition, except for the slight reductions in SO3 (solid-state) and CaO, which is the result of higher CO2 concentration and higher combustion temperature. Specifically, the high CO2 partial pressure inhibits the sulfate formation of alkaline earth metals, and the higher combustion temperature increases the release of coal-S to gaseous SO2 and SO3. Till now, there are only a few reports about pressurized oxy-coal combustion, and none of them presents the result of ash composition. Previous experiments [17][18] conducted in pressurized thermogravimetric analyzer showed that the reaction pressure affected the transformations of minerals in the fly ash due to the variations of the ignition mechanism under different pressures. However, the critical pressure for the ignition changed from heterogeneous mode to homogeneous mode was over 1 MPa which is higher than that the highest pressure investigated in this study. Figure 11. Chemical composition of the fly ash (oxy-30 atmosphere) (1) After the steady oxy-combustion condition is achieved, the CO2 concentrations in the dry flue gases have exceeded 90% under all the tested pressures, which is beneficial to the CO2 capture and utilization processes.