Sensitivity Analysis of Carbonate Looping Process Using Twin Fluidized Bed Model

CO2 as the major greenhouse gas plays an important role in environmental problems. Therefore, it is reasonable to find an effective technology for mitigating large CO2 emissions. Carbonating looping is an efficient post-combustion CO2 capture technology using limestone sorbent, which is more energy-saving than traditional technologies. In our research, a carbonate looping process model had been developed using ASPEN PLUS software. In detail, the sensitivity analysis of main parameters were adopted. The simulation results indicated that the CO2 capture rate of the whole process can achieve above 90% and CO2 concentration in the decarbonated flue gas was less than 4% under the carbonation temperature, the calcination temperature, the F0/FCO2 and the gas-solid separation were 630°C, 900°C, 0.04, 1.0, separately. This work gave an potential example for the retrofitting current coal-fired power plant combined with carbonate looping process.


INTRODUCTION
In recent years, scientists and researchers have put much more concern on the climate change issues, which surely confirm that global warming is mainly caused by greenhouse gas emission (GHG) (Safdarnejad et al., 2016;Cheng et al., 2019;Huang et al., 2019). GHG is mainly emitted from fossil fuel burning which is considered as the main source of global warming. Among all the GHG, CO2 is the principal gas which related to many large power plants (Symonds et al., 2017;Tritippayanon et al., 2019;Safarian et al., 2020). Therefore, it is necessary to develop effective technologies for reducing large CO2 emissions (Li et al., 2013;Jiang et al., 2019). CO2 capture and storage (CCS) is a promising technology to control GHG discharge, which has been widely used in these decades Liu et al., 2020). There are several methods for CO2 capture from power plants: (i) oxy-fuel combustion (Atsonios et al., 2015); (ii) precombustion (Solares et al., 2020); (iii) post-combustion (Alam et al., 2020). The post-combustion method is widely used method for CO2 capture in power stations at present. The flue gas passes through CO2 removal equipment to mitigate CO 2 emission (Liang et al., 2015;Martinez et al., 2018).
There have been many methods in the literature about post-combustion CO2 captures, such as absorption (Porter et al., 2015), adsorption (Shah and Imae, 2016), cryogenic (Yuan et al., 2014), chemical looping (Leion et al., 2008) and membrane (Zhao et al., 2018;Yang et al, 2019). However, these processes have some bottlenecks including enormous energy losses of 8-14% which increase fuel consumption and environmental concern regarding solvent regeneration (Lasheras et al., 2011;Oh et al., 2016;Liu et al., 2019a, b). In addition, some researchers have proposed that the chemical looping combustion consumes less energy to separate CO2 from flue gas, but this method still need full-scale experimental development (Abad et al., 2007;Leion et al., 2008). The carbonate looping (CL) is one of the promising methods for post-combustion CO2 capture, which uses calcium oxide (CaO) as absorption sorbents to reduce CO2 emission (Diego et al., 2017). The main equipment of this process is two interconnected fluidized beds. The principal of the carbonate looping process is shown in Fig. 1.
In Fig. 1, the flue gas which comes from an upstream power plant firstly enters the carbonator. In the carbonator, most of the CO2 react with CaO and convert to CaCO3. The exothermic carbonation reaction happens at 650°C and a pressure of 1 bar which can achieve a better conversion rate of CaO (Strohle et al., 2014). The CaCO3 then transfers to the second fluidized bed (calciner), where the CO2 is released from CaCO3 at about 950°C, forming high purity CO2 product and ready to compress for storage. The regenerated CaO from calcination reaction is then sent to the carbonator for another cycle. The calcination reaction is endothermic, therefore the extra heat is supplied by the combustion of fuel and pure oxygen. In the whole process, high temperature will cause deactivation of sorbent and the over-time cycle will also lead to solid losses. Therefore, make-up flow should be introduced into the system to keep the mass balance in the system (Jia et al., 2007;Zhang et al., 2016;Alonso et al., 2018). Compared with other post-combustion CO2 capture technologies, CL has extremely lower energy penalties and the sorbent is widely distributed in nature and easy to obtain (Romano, 2009;Pan et al., 2017). In addition, CL using indirectly heated method will be further improving CO2 capture efficiency (Hoeftberger et al., 2016;Reitz et al., 2016).
The feasibility of the CL process has been evaluated by many researchers through large-scale test rigs and simulations. In 2009, Symonds et al. (2009 testified that humid environment can increase carbonation conversion. In 2010, Alonso et al. (2010) designed and built a 30 MWth test facility in Spain, and conducted a series of experiments about CO2 capture performance. The results showed that CO2 capture efficiency can be achieved above 70% as the reactors had enough solid inventory. A 10kWth dual fluidized bed (DFB) was erected in the University of Stuttgart (Germany) by Charitos et al. (2010). Some parameters, such as carbonate temperature, calcium looping ratio, and carbonation reaction time, etc., were investigated through this experiment. The experiments demonstrated that the increased carbonate temperature and calcium looping ratio can significantly boost CO2 capture efficiency, meanwhile, the CO2 capture efficiency of the whole system can reach above 90%. On the other hand, Lasheras et al. (2011)  Present work focused on CO2 capture using carbonate looping based on earlier reported work by Shimizu et al. (1999). The flue gas came from 350 MWth net electrical power, which had a high temperature and flow rate. The whole process was introduced and set up into ASPEN PLUS software for numerical simulation of CO 2 capture performance. The carbonation and calcination temperature, make-up flow, gas-solid separation efficiency, and SO2 concentration in the flue gas were analyzed and optimal parameters were obtained. The work will provide important data for the application of carbonate looping.

Model Description
In the present research, a carbonate looping process model was established using ASPEN PLUS software and the process flow diagram was demonstrated as shown in Fig. 2. A coal-fired power plant that has 350 MWth net electrical power and 36% net efficiency was used as an upstream plant in this study. The sulfur dioxide gas was removed when the flue gas passed through the carbonate looping process. The main parameters of flue gas were shown in Table 1.
The flue gas initially entered the carbonator and the two chemical reactions occurred in carbonator: The CO2 absorption reaction (R1) happened in the first fluidized bed (CARBONATE) and the SO2 absorption reaction rate (R2) was set as 100%. The solid stream and the  decarbonated flue gas were separated by the cyclone (SEP1), in which gas-solid efficiency was 100%. After separation, the decarbonated flue gas was cooled down to around 100°C (COOLER-1) and discharged directly into the atmosphere. On the other hand, the solids were partly removed from the system (PURGE) due to the deactivation of sorbent and the rest solids were transferred into the calcinator (CALCINER).
In the calcinator, the reversed reaction happened at around 900°C and CO2 released from CaCO3. Because the reversed reaction was endothermic, the whole system needed external heat, which we adopted coal combustion to supply the heat. The coal was first pulverized into powder (DECOMP) and pushed into the calcinator to combust with pure oxygen. Meanwhile, the make-up flow (MAKE-UP) consisted of 100% CaCO3 continually entered the calcinator for the supplement of emitted sorbent, which made sure the mass balance of the whole process.
After the reaction in the calcinator, the stream passed through the separator (SEP) first to discharge the CaSO4, the rest stream then flowed to the second cyclone (SEP2), where the CO2 of the whole system was separated from the solid and part of the CO2 (CO2-2) was recycled to the calcinator, the another was compressed for storage. Meanwhile, the CaO sorbent was sent into the carbonator for another cycle.

Simulation Assumptions
The simulation of the carbonate looping process was based on ASPEN PLUS and the main equipment parameters were shown in Table 2. To simplify the calculation, the following assumptions were adopted: (1) The reaction time did not affect the whole process; (2) There was no pressure loss in the whole process; (3) The combustion of coal was complete; (4) The by-product (ash and CaSO4) was completely extracted from the process.

Process Evaluation
To evaluate the performance of the whole carbonate looping process, two main indexes were introduced into this work: CO2 capture in the carbonator, and CO2 capture in the whole process. The calculation of these two indexes was summarized as follows: The CO2 capture rate in the carbonator, Xcarb, was defined as the relationship between CO2 concentration in the decarbonate flue gas, CO2, flue out, and the CO2 concentration in the flue gas, CO2, flue gas.
The CO2 adsorption efficiency in the whole process, XCO2, was explained as the relationship between captured CO2, and produced CO2. The produced CO2 mainly contained two parts: the make-up flow, CO2, make up, and the combustion of the coal, CO2, coal.

Effect of Carbonation Temperature
Carbonation temperature had an effect on the CO2 capture rate which had been presented in Fig. 3. The basic simulation conditions were shown in Tables 1 and 2.
In Fig. 3, with increasing carbonation temperature from 600 to 650°C, the CO2 capture rate in carbonator increased steadily first and then dropped sharply to 82%. For CO2 concentrations in the decarbonated flue gas, when carbonation temperature increased, it decreased firstly and then increased rapidly. This trend showed that high carbonation temperature can boost the conversion rate of the absorption reaction (R1), which was a benefit for CO2 capture. When the carbonation temperature was 630°C, the CO2 capture rate in the carbonator reached the highest value which was 90% and CO2 concentration in the decarbonated flue gas was 2.5% which was the lowest point. However, the higher carbonation temperature was not good for the CO2 capture. When the temperature was above 630°C, the over-high temperature had a bad effect on both the CO2 capture rate and CO2 concentrations in the flue gas. That was because CaO absorption was an exothermic reaction and higher temperature was detrimental to the adsorption reaction.
The effect of carbonation temperature on CO2 capture of the whole process and CaCO3 circulating flow rate were shown in Fig. 4 and obtained parameters were represented in Tables 1 and 2. It can be seen from Fig. 4 that the CO2 capture rate of the whole process and CaCO3 circulating flow rate had the same trend when the carbonation temperature was increased from 600 to 650°C. The increasing carbonation temperature had two contrary effects. On the one hand, when the temperature increased from 600 to 630°C, more CO 2 was captured and more limestones were generated and circulated in the whole process, therefore, both CO2 capture rate and CaCO3 circulating flow grew. On the other hand, over-high temperature hindered the absorption reaction rate. That was why both two lines dropped sharply when the temperature increased from 630 to 650°C. From the obtained results, it was concluded that the carbonation temperature significantly affected the performance of the process. Meanwhile, when the carbonation temperature was 630°C, the CO2 capture rate and CaCO 3 circulating flow rate had the highest value. So, we chose 630°C as the optimal carbonation temperature for further considerations.

Effect of Calcination Temperature
When the calcination temperature changed from 850 to 1250°C, the carbonation temperature and coal mass flow were set as 630°C and 120 t h -1 , respectively. The variation of CO 2 capture rate in the carbonator and the whole process, CO2 concentrations in the decarbonated flue gas, and CaCO3 circulating flow along with the calcination temperature were calculated and analyzed. Fig. 5 showed that the CO 2 capture rate in the carbonator decreased firstly and then increased with the calcination temperature rose, on the contrary, CO2 concentrations in the decarbonated flue gas rose firstly and then dropped down. The reason for the trend was that the rising temperature caused the deactivation of sorbent which called sintering effect. Sintering can weaken CO2 capture capacity and reduce cycle times of the sorbent. When the calcination temperature was 1175°C, the CO2 capture rate in the carbonator was only 80% and CO2 flue gas reached the highest point of 4.5%. That was unacceptable for the system. However, as the calcination temperature varied from 1150 to 1250°C, more CaSO4 decomposed into CaO and make-up introduced into the whole process. Thus, the more CO2 was capture in the carbonator, which increased the CO2 capture rate in the carbonator and decreased the CO2 concentration in the flue gas.
The effect of calcination temperature on the CO2 capture rate in the whole process and CaCO3 circulating flow was shown in Fig. 6. As the calcination temperature increased, the CO2 capture rate in the whole process and CaCO3 circulating flow showed the same pattern (See Fig. 5). Sintering can change and deactivate the micro-granular of the sorbent, which caused a rapid decline of the CO2 capture rate. On the other hand, when the temperature was higher than 1175°C, the CaSO4 decomposition and make-up flow lines moved upward by indicating a sharp increment of the CO2 capture rate.
The relationship between the energy consumption and the calcination temperature was shown in Fig. 7. The energy consumption increased steadily with the calcination temperature increased. This behavior was observed because a higher calcination temperature needed more coal combustion to provide extra heat, and it increased the system load and caused much more energy consumption.
From the above-obtained results (Figs. 5, 6 and 7), it can be concluded that when the calcination was 900°C, the CO2 capture rate in carbonator and the whole process, CaCO3 circulating flow and energy consumption were found to be 89%, 91%, and 74.9 MW, respectively, which were the best in the process. Therefore, 900°C was chosen as the optimal calcination temperature.

Effect of Make-up Flow
Make-up flow (F0) was related to CO2 concentration in the flue gas (FCO2), therefore F0/FCO2 had been chosen to be a variable in further study. The carbonator and calcination temperature, coal consumption was set at 630°C, 900°C, and 120 t h -1 , respectively. The effect of F0/FCO2 on CO2 capture in the carbonator and CO2 concentration in the decarbonated flue gas were calculated and analyzed.
It can be seen from Fig. 8 that, with the increase of makeup flow, CO2 capture in the carbonator had a significant increase and CO2 concentrations in the decarbonated flue gas decreased, which indicated that more make-up flow was good for the system. On the other side, a certain level of make-up flow increase was not adequate for CO2 capture conditions. When the F0/FCO2 was above 0.04, the trend of CO2 capture rate in the carbonator and CO2 concentrations in the decarbonated flue gas was stabilized gradually due to the oversaturation of CaO in the system. The trend showed that the appropriate amount of CaCO3 supplement was necessary, and it can update the deactivated sorbent. Fig. 9 reflects the effect of make-up flow on the CO2 capture rate in the whole process and CaCO3 circulating flow rate, which indicated the same increase pattern when the make-up flow increased. When F0/FCO2 was lower than 0.04, the CO2 capture rate of the whole process increased rapidly. This was observed because of the initial CaCO3 produced by carbonation was fewer than the initial circulation flow rate and when the CaCO3 stream and the make-up flow mixed, the circulating flow rate surged.
In combination with Figs. 8 and 9, when the optimal F0/FCO2 was 0.04, the CO2 capture rate in the carbonator and whole process, CO2 concentration in the decarbonated flue gas and CaCO3 circulating flow rate reached the best value, which were 86.4%, 91.6%, 3.1%, and 642.9 t h -1 , respectively.

Effect of Gas-solid Separation Efficiency
When the gas-solid separation efficiency varied from 0.95 to 1.0, the carbonation temperature, calcination temperature, and F0/FCO2 were set at 630°C, 900°C, and 0.04, respectively. The change of CO2 capture in the carbonator and CO2 concentrations in the decarbonated flue gas were calculated and analyzed. From Figs. 10 and 11, it can be seen that the CO2 concentrations in the decarbonated flue gas and CaCO3 circulating flow rate were linear with gas-solid separation efficiency, which indicated that gas-solid separation efficiency had a great impact on the system stability. The CO2 concentrations in the decarbonated flue gas linearly decreased with the gas-solid separation efficiency increased, however, CO2 capture rate in the carbonator and the whole process and  CaCO3 circulating flow rate showed the opposite trend. When the gas-solid separation efficiency was in the range of 0.95 to 1, the CO2 capture rate in the carbonator increased from 87% to 89%, the CO2 concentrations in the decarbonated flue gas decreased from 3.1% to 2.59%, and the CaCO3 circulating flow also increased accordingly. Therefore, the gas-solid separation efficiency directly affected the system performance. It is observed that the higher gas-solid separation efficiency can achieve more favorable CO2 adsorption rate. Herein, the design of the cyclone should try to ensure a high gas-solid separation efficiency. When the gas-solid separation efficiency was too low, the loss of effective CaO sorbents increased, and the concentration of the sorbent gradually decreased after circulation, which caused CO2 cannot be captured. In the present work, we chose 1.0 as gas-solid separation efficiency.

Effect of SO2 Concentration in the Flue Gas
When the SO2 concentrations in the flue gas varied from 0.1% to 1.0%, the carbonation temperature, calcination temperature, F0/FCO2, and gas-solid separation efficiency was set at 630°C, 900°C, 0.04 and 1.0, separately. The change of CO2 capture rate in the carbonator and CO2 concentrations in the decarbonated flue gas were calculated and analyzed.
It can be seen from the Fig. 12 that the SO2 concentrations in the flue gas had a great effect on the CO2 capture rate in  the carbonator. When the SO2 concentrations in the flue gas increased from 0.1% to 1%, the CO2 capture rate in the carbonator decreased from 88.3% to 63.7%, which dropped by 21.6%. This trend also showed in the carbonator and CaO not only reacted with CO2, but with SO2, and meanwhile, the density of CaSO4 was larger than CaCO3, which was not good for CO2 diffusion process and greatly reduced the CO2 capture rate in the carbonator. Therefore, the CO2 concentrations in the decarbonated flue gas increased. Fig. 13 showed CO2 capture rate in the whole process and CaCO3 circulating flow rate decreased with SO2 concentrations in the flue gas increased. That was because the large increase of SO2 in the flue gas, the more CaO was converted into CaSO4. Meanwhile, CaSO4 was not completely decomposed at 900°C in the calcinator. Therefore, more CaSO 4 was discharged from the system, which significantly reduced the CO2 capture in the whole process and CaCO3 circulating flow.
Combining Figs. 12 and 13, when the SO2 was introduced into the system, the CaO sorbent reacted with SO2 firstly and then with CO2, which formed the stable CaSO4 solid and reduced circulating flow rate in the whole system. Therefore, the SO2 concentration in the flue gas had a huge influence on CO2 capture performance and to ensure the CO2 capture rate, the SO2 was mostly removed from the flue gas before came into the process and SO2 concentration in the flue gas was set as 0.1 in the work.
Under the condition that the carbonation temperature was 630°C, the calcination temperature was 900°C, the F0/FCO2 Fig. 11. CO2 capture rate in the whole process and CaCO3 circulating flow along with the changes of gas-solid separation efficiency. was 0.04, the gas-solid separation was 1.0 and the SO2 concentration in the flue gas was 0.1, the simulation was carried out, see Table 3. The results could satisfy for the CO2 capture rate of over 90% and CO2 concentration in the decarbonated flue gas was less than 4.0%. By contrast, Lasheras et al. (2011) obtained CO2 capture rate of 80% in a 1052 MWel coal-fired power plant. The carbonate looping used in this study can achieve above 90% CO2 capture rate. Thus, the results of this work was reasonable.

CONCLUSIONS
In order to remove and capture CO2 from flue gas, the carbonate looping process was more suitable for operating conditions with large flow flue gas. In the work, the proposed carbonate looping process was established with ASPEN PLUS to obtain optimal parameters of the carbonate looping process. The process adopts a reversible reaction CaO + CO2 ↔ CaCo3 into account and formed a circulation that would obtain higher CO2 product gas and continually recycle the sorbent. Also, due to particle attrition and sintering of CaO, the make-up flow should be introduced into the system. Meanwhile, the sensitivity analysis of several parameters was conducted and the optimal parameters were selected and listed in Table 3. The main results of this work were as follows: Fig. 13. CO2 capture rate in the whole process and CaCO3 circulating flow rate along with the changes of SO2 concentrations in the flue gas. Table 3. Summary of carbonate looping process.

Results parameters
Numerical value Carbonation temperature (℃) 630 Calcination temperature (℃) 900 F0/FCO2 0.04 Gas-solid separation efficiency 1.0 SO2 concentration in the flue gas 0.1 1. The proper temperature increase of carbonate and calcination would be beneficial to the reaction, which can increase the CO2 capture rate. However, the over-high temperature can cause deactivation of sorbents and decrease of the solid circulating flow; 2. The increase of make-up flow increased the solid circulating flow and the activated CaO sorbent which significantly promoted CO2 capture rate. when the F0/FCO2 was 0.04, the CO2 capture rate in the whole system was 91.6% and reached the highest value; 3. The gas-solid efficiency and SO2 concentrations in the flue gas both had huge influence on the CO2 capture rate. When the gas-solid efficiency decreased from 1.0 to 0.95, the CO2 capture rate dropped by 2%. Meanwhile, as the SO2 concentrations in the flue gas increased from 0.1% to 1.0%, the CO2 capture rate decreased by 21.6%; 4. When the carbonation temperature was 630°C, the calcination temperature was 900°C, the F0/FCO2 was 0.04, the gas-solid separation was 1.0 and the SO2 concentration in the flue gas was 0.1, the target of the CO2 capture of > 90% and CO2 concentrations in the decarbonated flue gas < 4.0% could be achieved.
The simulation results were applicable to the current situation of most stations which provide a good example for the utilization of CO2 capture in power plants. However, there were also some important parameters, such as pressure drop, solid fraction, superficial velocity of carbonator, and so on, which we did not discuss in the work. These parameters also play an indispensable role in system performance, which will be a challenge for us in the future.

ACKNOWLEDGMENTS
This work was supported by the Department of Science & Technology of Shandong Province (No. ZR2018LB025).