Oxygen-Induced Elemental Mercury Oxidation in Chemical Looping Combustion of Coal.

Mercury emission is an important issue during chemical looping combustion (CLC) of coal. The aim of this work is to explore the effects of different flue gas components (e.g., HCl, NO, SO2, and CO2) on mercury transformation in the flue gas cooling process. A two-stage simulation method is used to reveal the reaction mechanism of these gases affecting elemental mercury (Hg0) oxidation. Furthermore, using this method, Hg0 oxidation by eight oxygen carriers (Co3O4, CaSO4, CeO2, Fe2O3, Al2O3, Mn2O3, SiO2, and CuO) commonly used in CLC are investigated and their Hg0 oxidation efficiencies were compared with the existing experimental results. The results show that HCl, NO, and CO2 promote Hg0 oxidation during flue gas cooling, while SO2 inhibits Hg0 oxidation. The stronger the oxygen release capacity of oxygen carriers, the higher the oxidation efficiency of Hg0 becomes. The order of Hg0 removal efficiency from high to low is Co3O4, CuO, Mn2O3, CaSO4, Fe2O3, CeO2, Al2O3, and SiO2, and this sequence is in good agreement with the existing experimental results. Different flue gas components directly or indirectly affect the O2 content, thus affecting the content of gaseous oxidized mercury (Hg2+). Different oxygen carriers have different oxygen release capacities and different Hg0 oxidation efficiencies. Therefore, O2 is the core species affecting the mercury transformation in CLC.

Introduction

One of the biggest challenges facing the world today is global warming. Coal combustion produces a large amount of carbon dioxide, which is the main reason for the greenhouse effect in the atmosphere. To mitigate climate change, CO2 emission control and carbon capture utilization and storage (CCUS) are potentially viable approaches and promising options to alleviate increasing carbon dioxide (CO2) emissions in human society.[1−3] China’s energy structure of rich coal, poor oil, and little gas determines that coal will occupy an important energy position for a long time. However, coal combustion releases a large amount of trace element mercury. Generally speaking, mercury in coal-fired flue gas mainly exists in three forms: elemental mercury (Hg0), oxidized mercury (Hg2+), and particle-bound mercury (HgP).[4,5] Among them, Hg2+ is easily soluble in water and can be removed using a wet flue gas desulfurization device,[6] HgP generally exists on the surface of solid particles such as fly ash and can be removed using an electrostatic precipitator or a bag filter,[7] and Hg0 is volatile and insoluble in water, so it is difficult to be effectively removed using the existing flue gas purification equipment.[8,9] Considering the high toxicity, mobility, and bioaccumulation of Hg0, it is necessary to effectively treat Hg0 produced in the combustion process.[10,11]

Among the carbon capture technologies, chemical looping combustion (CLC) has recently emerged as a promising option to facilitate CO2 inherent separation at a low cost. It is a new combustion technology containing a circulating combustion system composed of an air reactor (AR) and a fuel reactor (FR).[3] Different from the traditional combustion mode, the oxygen required in CLC is not directly provided by air, but by the oxygen carrier (OC). OC circulates between FR and AR to realize its reduction and regeneration.[12] Therefore, the core of CLC is to select appropriate OC, which can not only provide lattice oxygen[13,14] but also catalyze the oxidation of Hg0 released by coal combustion and effectively reduce the emission of Hg0.[15]

The reactor temperature of CLC is about 800–1000 °C. It is generally considered that when the temperature is higher than 800 °C, the mercury species mainly exist as Hg0. When the temperature is reduced to 700 °C, part of Hg0 will be oxidized to Hg2+. Mendiara et al. used Fe2O3 as OC to study the mercury release in CLC.[16,17] It was found that the fraction of the mercury in coal vaporized in the FR mainly depended on the temperature of the FR and the coal used. At the same time, the species of mercury was measured. It was found that Hg0 was the main species in the FR. Pérez-Vega et al. used CuO as the OC, and it was found that owing to the presence of copper oxide as the oxygen carrier, which has oxygen uncoupling ability, the volatilized Hg was further oxidized to Hg2+.[18] Ma et al. found that in the presence of hematite in the FR, the concentration of Hg0 decreased and the concentration of Hg2+ increased, indicating that hematite promoted the conversion of Hg0 to Hg2+.[19] Only a small amount of mercury was adsorbed by the OC and transported to the AR together with the carbon residue, which was released in the form of Hg0(g) and Hg2+(g) or left in OC and coal ash in the form of HgP. Ji et al. used ferrous ore as OC to study the release characteristics of mercury in bituminous coal in CLC. It was found that most of the mercury in coal was released in FR, and the rest was released in AR. In particular, Hg0 is the main species of released mercury.[20] However, with the increase of FR temperature, the amount of Hg0 may decrease and the amount of Hg2+ may increase. This can be attributed to different flue gas components.[21,22]

At present, the mechanism of Hg0 oxidation by oxygen carriers in the furnace has been reported,[23] but the effects of flue gas components produced in the gasification process on mercury transformation in the cooling stage are not clear. Certainly, oxygen carriers may have a significant influence on Hg0 oxidation through the coal gasification process. In the CLC process, coal-fired flue gas contains many gas components such as Cl2, HCl, SO2, and CO. According to early studies, the presence of Cl2/HCl in flue gas is the main reason for the migration and transformation of mercury,[24] and Hg0 oxidation reactions occur during flue gas cooling.[25] However, the actual flue gas is too complex and the mechanism by which these flue gas components affect Hg0 oxidation is not yet clear.

In this paper, we propose a two-stage simulation method to understand the effects of oxygen carriers on the gasification process, and the effects of gas components produced in the gasification process on the transformation of mercury in the flue gas cooling stage. The reaction paths are potentially revealed. The Hg0 oxidation efficiencies of OCs in FR are simulated and compared with the experimental results. This paper provides a theoretical basis for simulating the mercury transformation mechanism in CLC, and also provides a technical method for the selection of oxygen carriers with Hg0 oxidation ability.

Ranking Method

Thermodynamic Simulation

To evaluate Hg0 oxidation efficiencies of different OCs, the speciation of mercury in the system should be calculated. Therefore, a thermodynamic calculation was performed by the Equilib module in FactSage 5.2. Chemical equilibrium calculation is performed by means of a general Gibbs energy minimization algorithm. The Equilib module calculates the conditions for multiphase, multicomponent equilibria, with a wide variety of tabular and graphical output modes, under a large range of possible constraints through Gibbs energy minimization based on the ChemApp algorithm.[26]

When the chemical equilibrium is investigated, the standard Gibbs free energy change (ΔG) of the reaction is often used. A possible way to determine whether some specific chemical processes can occur spontaneously is by calculating the ΔG energy of the reaction. The definition of ΔG can be given by eq .where ΔH is the enthalpy, T is the absolute temperature, and ΔS is the entropy.

According to eq , there are three possibilities for ΔG in any process, namely:

ΔG < 0: the reaction is likely to proceed spontaneously without involving external energy.

ΔG = 0: the chemical reaction is at a thermodynamic equilibrium.

ΔG > 0: the possibility of the reaction to proceed spontaneously without external energy is very small.

The priority of thermodynamic simulation is to predict Hg0 oxidation efficiency. However, the methodology has not been established. Our previous research proposed that Cl2 generated in the “Hg0 + HCl + MeO” system can be used to rank the ability of different oxygen carriers for Hg0 oxidation.[23] However, Hg0 oxidation efficiency cannot be predicted for different systems containing different oxygen carriers. On top of our previous research, our aim is to establish a methodology for the calculation of Hg0 oxidation efficiency.

The typical experimental results obtained by An et al. were chosen for the development of the simulation methodology. According to the previous results obtained in a fluidized bed using CuFe2O4 as the oxygen carrier to oxidize Hg0, it is found that when 1 L/min (STP) gas flow (N2/H2O = 50/50%) was used as the gasification atmosphere, the Hg0 proportion with respect to total mercury measured at the tail gas outlet after 40 min was 60.92%.[27] Thermodynamic simulation is attempted to match this result.

The initial conditions for simulation are taken from the paper published by An et al. High mercury-containing lignite (YN) was used in the simulation, and the elements contained in YN mainly included C, H, O, N, S, Cl, and Hg.[27] The gasification atmosphere was N2 and H2O. The oxygen carrier was CuFe2O4, and the experimental temperature was 850 °C. In the experiment, 10 g YN and 1 L/min (STP) gas flow (N2/H2O = 50/50%) were used. Thus the initial amount of these substances in the simulation is shown in Table .

Table 1

Initial Quantity of Each Substance in the Thermodynamic System

species	C	H	O	N	S	Cl	N2	H2O	Hg
n/mol	0.634	0.350	0.01	0.0021	0.00137	6.58 × 10–6	0.89	0.89	1.26 × 10–8

By observing the experimental system published previously, two stages may be involved, as shown in Figure . The first stage is the coal gasification process which occurs in the furnace. The second stage is the cooling stage which occurs outside the furnace and before analysis. In this stage, the flue gas after coal gasification, which comes out of the furnace passes through the ice bath and then enters the mercury analyzer. Both processes are simulated using the Equilib module in FactSage. The Reaction module in the thermodynamic calculation software FactSage is used to calculate Gibbs free energy change (ΔG).

Figure 1

Two-stage simulation method.

The gasification process of coal and CuFe2O4 under a N2 and H2O atmosphere is first simulated, and then the gasification products of this process are taken as the reactants in the second stage to continue to simulate mercury transformation during cooling.

Experimental Apparatus

Experiments were carried out in a drop furnace. In the experiments, eight oxygen carriers, including CaSO4, Co3O4, Mn2O3, Fe2O3, CuO, CeO2, SiO2, and Al2O3 were adopted. These eight substances were acquired from Sinopharm Chemical Reagent Co., Ltd. Zhundong coal was used in the experiment, and it had a fine grain of 80–100 mesh.

As can be observed from Figure , the experimental setup was composed of five main parts: the gas feed system, coal gasification section, gas cooling section, analysis instrument for Hg0 (VM3000, mercury analyzer), and the exhaust gas treatment device (with activated carbon). Approximately 5 g of Zhundong coal and oxygen carriers with different contents were packed into the drop furnace in each experiment. The total amount of mercury was about 0.7 μg. The temperature of the reactor was controlled at 850 °C. In the reduction reactor, the CO2 flow rate was 5 L/min and the H2O flow rate was 1 L/min. The four impact bottles in the gas cooling section were respectively filled with 1 mol/L KCl solution and silica gel particles. KCl solution was mainly used to absorb Hg2+ in gas. Silica gel particles were used to remove moisture in gas and to prevent moisture from entering VM3000 and damaging the instrument.

Figure 2

CLC experimental system.

Results and Discussion

Validation of the Simulation Method

In the first step of the simulation, the only uncertainty is the initial content of CuFe2O4. To determine the content of CuFe2O4, different contents of oxygen carriers are optimized in the calculation. In the experiment, the gasification products first passed through the ice bath and then entered the mercury analyzer. Therefore, the gas temperature entering the mercury analyzer was about 0–10 °C. In the CuFe2O4 system with different quantities, the gasification process of coal and CuFe2O4 under a N2 and H2O atmosphere is first simulated, and the gasification product of this process is then taken as the reactant in the second stage to simulate mercury transformation in the cooling process. Finally, the proportion of Hg0 at 0–10 °C can be obtained by the thermodynamic method. The proportion of Hg0 in the CuFe2O4 system with different contents can be obtained by averaging the proportion of Hg0 at 0 and 10 °C.

As shown in Figure , when the amount of OC is 2–3 mol, the proportion of Hg0 decreases significantly. Therefore, an optimal value is selected in this interval to make the proportion of Hg0 consistent with the experimental results of An et al. (60.92%).[27] The calculation results of the first stage show that the gas components after gasification contain mainly CO2, SO2, H2, CO, HCl, NO, O2, etc.

Figure 3

Effect of the OC amount on the proportion of Hg0.

The second process is the cooling of the gasification product. Sliger et al. investigated the homogeneous oxidation of chlorine and mercury by the kinetic model. They believe that the Hg0 oxidation occurs in the cooling process of flue gas.[25] According to the thermodynamic results, Hg2+ mainly exists in three forms HgCl2, HgS, and HgO. Since the amount of total mercury is small, it is only 1.26 × 10–8 mol, and in addition, the main mercury species in the second stage are Hg0 and HgCl2. Therefore, the amount of HgCl2 is mainly used as the evaluation standard of the Hg0 oxidation efficiency. In the following part, we specifically discuss the impact of different flue gas components on Hg0 oxidation in this system.

Influence of Gas Compositions

To explore the influence of different flue gas components on Hg0 oxidation in the flue gas cooling process, sensitivity analysis of HCl, NO, H2, CO, SO2, and CO2 was carried out. Seven working conditions including “All,” “Remove HCl,” “Remove NO,” “Remove SO2,” “Remove CO2,” “Remove H2,” and “Remove CO” were designed. “All” represents the case with all the gasification species included, whereas “Remove gas” represents the case with all the gasification species except for the indicated gas. It is found that in the cooling process from 800 °C to 0 °C, H2 and CO have little impact on Hg0 oxidation, while HCl, NO, SO2, and CO2 have different degrees of impact on Hg0 oxidation at different temperatures as shown in Figure . Therefore, next, we mainly discuss how these four gases affect Hg0 oxidation.

Figure 4

Variation of HgCl2 with temperature under seven working conditions: (a) “All” is compared with “Remove HCl”, “Remove NO”, “Remove SO2”, and “Remove CO2” and (b) “All” is compared with “Remove H2” and “Remove CO”.

As can be seen from Figure a, when HCl is removed, the amount of HgCl2 in the whole process is reduced. Therefore, HCl promotes Hg0 oxidation in the whole temperature range. For NO, the effect is not obvious in the range of 100–800 °C. In the range of 0–100 °C, when NO is removed, HgCl2 is reduced, and thus NO plays a promoting role. After SO2 is removed, HgCl2 increases significantly when the temperature is lower than 400 °C. Therefore, SO2 has an obvious inhibitory effect on Hg0 oxidation in this temperature range. When CO2 is removed and the flue gas is cooled down from 400 to 300 °C, HgCl2 decreases. Therefore, CO2 gas is conducive to the oxidation of Hg0 in the temperature range of 300–400 °C.

Effect of HCl

Since HCl is conducive to the oxidation of Hg0 in the whole temperature range, to see more clearly how HCl affects the amount of HgCl2, 300 °C is taken as the discussion temperature. Comparing “All” and “Remove HCl” systems, the main substances which are changed include HCl, HgCl2, Cl2, Cl, and HOCl as shown in Table . When HCl is present, the amounts of HCl, Cl2, Cl, and HOCl increase, resulting in an increased Hg0 oxidation efficiency.

Table 2

Amount of Different Species at 300 °C in “All” and “Remove HCl” Systems

	n/mol			n/mol
species	All	Remove HCl	species	All	Remove HCl
H2O	1.06	1.06	O2	2.63 × 10–11	2.63 × 10–11
N2	0.89	0.89	NO	1.28 × 10–13	1.28 × 10–13
CO2	6.34 × 10–1	6.34 × 10–1	HgO	2.08 × 10–14	2.33 × 10–14
SO2	7.91 × 10–6	7.91 × 10–6	Cl2	8.47 × 10–15	2.93 × 10–19
HCl	6.58 × 10–6	3.87 × 10–8	H2	7.72 × 10–15	7.72 × 10–15
O2S(OH)2	2.36 × 10–6	2.36 × 10–6	Cl	9.08 × 10–16	5.35 × 10–18
SO3	3.13 × 10–7	3.13 × 10–7	OH	8.46 × 10–16	8.46 × 10–16
Hg	1.12 × 10–8	1.26 × 10–8	HOCl	7.76 × 10–16	4.57 × 10–18
HgCl2	1.37 × 10–9	5.34 × 10–14	CO	1.12 × 10–16	1.12 × 10–16

For HCl, the increase in the amounts of HCl, Cl2, Cl, and HOCl is the key to the increased Hg0 oxidation efficiency. The Gibbs free energy change (ΔG) for the reactions related to the conversion of Hg0 to HgCl2 in the “All” system are calculated.

The reaction pathways of HCl related reactions to generate Cl2, Cl, and HOCl are shown in R1–R3 in Table . The global reaction for the direct oxidation of Hg0 by HCl is R4. Due to the high energy barrier of the direct reaction between Hg0 and HCl, it is calculated as ΔG > 0 at 300 °C. This is consistent with previous results of Wilcox et al.[28] Widmer et al. proposed eight elementary reactions for mercury oxidation including R5–R12.[29] It is calculated that R5, R9, R10, and R11 are the main reactions for Hg0 oxidation in the “All” system. The reaction pathway may be described as follows. Hg0 is oxidized to HgCl (R5), and then HgCl reacts with Cl2, Cl, and HOCl to form HgCl2 (R9–R11).

Table 3

Gibbs Free Energy Change of Each Reaction at 300 °C

number	reaction	ΔG (kJ mol–1)	number	reaction	ΔG (kJ mol–1)
R1	2HCl + 0.5O2 = Cl2 + H2O	–19.87	R2	HCl + OH = Cl + H2O	–57.46
R3	HCl + O = HOCl	–166.39	R4	Hg0 + 2HCl = HgCl2 + H2	46.62
R5	Hg0 + Cl = HgCl	–57.74	R6	Hg0 + Cl2 = HgCl + Cl	122.47
R7	Hg0 + HOCl = HgCl + OH	111.42	R8	Hg0 + HCl = HgCl + H	319.25
R9	HgCl + Cl2 = HgCl2 + Cl	–91.18	R10	HgCl + Cl = HgCl2	–271.39
R11	HgCl + HOCl = HgCl2 + OH	–102.24	R12	HgCl + HCl = HgCl2 + H	105.59
R13	SO2 + 0.5O2 + H2O = H2SO4	–159.38	R14	SO2 + Cl2 + 2H2O = 2HCl + H2SO4	–295.92
R15	SO2 + 0.5O2 = SO3	–44.91	R16	SO3 + H2 = SO2 + H2O	–170.50
R17	N2 + O2 = 2NO	166.32	R18	CO2 = 0.5O2 + CO	233.16

Effect of NO

Because NO is conducive to Hg0 oxidation in the range of 0–100 °C, it is studied separately. Figure shows that the amounts of HgCl2 in the two systems are similar in the temperature range of 70–100°C. When the temperature is lower than 70 °C, the amount of HgCl2 in the “Remove NO” system decreases rapidly. Therefore, NO is conducive to Hg0 oxidation within 0–70 °C. To more clearly compare the changes in the amount of substances of each species in two systems, 30 °C is taken as the investigation temperature.

Figure 5

Amount of HgCl2 changes with temperature for “All” and “Remove NO” systems.

Table shows that NO most directly affects the amount of NH3 and NH4Cl (s). Compared with the “Remove NO” system, NH3 and NH4Cl (s) in the “All” system decrease. Similarly, gases including H2, CH4, CO, and HgCl2 decrease, while HCl and CuCl(s) increase.

Table 4

Amount of Different Species at 30 °C in “All” and “Remove NO” Systems

	n/mol			n/mol
species	All	Remove NO	species	All	Remove NO
N2	8.91 × 10–1	8.91 × 10–1	NH3	4.32 × 10–10	2.25 × 10–8
CO2	6.34 × 10–1	6.34 × 10–1	CH4	3.15 × 10–11	1.19 × 10–6
H2O	6.68 × 10–2	6.68 × 10–2	CO	1.43 × 10–12	1.99 × 10–11
HCl	1.29 × 10–6	2.48 × 10–8	HgCl2	3.77 × 10–15	9.97 × 10–12
Hg	1.26 × 10–8	1.26 × 10–8	NH4Cl(s)	5.25 × 10–6	6.56 × 10–6
H2	1.20 × 10–8	1.68 × 10–7	CuCl(s)	3.78 × 10–8	0

As given in Table , NO in “All” system directly affects the amounts of NH3 and NH4Cl (s) through R19 and R20. In detail, NO consumes a large amount of NH3 to produce N2 and H2O (R19). However, the amounts of N2 and H2O are much greater than that of NO, the increase of N2 and H2O cannot be clearly observed in the calculation results. Therefore, after NH3 is consumed, NH4Cl (s) produced by the reaction of NH3 with HCl decreases (R20). At the same time, HgCl2 increases as more HCl is involved in the oxidation of Hg0 rather than reacting with NH3. The reduction of CO and H2 may be due to the reaction with O2 to produce CO2 and H2O (R21 and R22). When CO and H2 are consumed, the generated CH4 decreases (R23).

Table 5

Gibbs Free Energy Change of Each Reaction at 30 °C

number	reaction	ΔG (kJ mol–1)	number	reaction	ΔG (kJ mol–1)
R19	4NO + 4NH3 + 2O2 = 4N2 + 6H2O	–1628.44	R20	NH3 + HCl = NH4Cl(s)	–177.03
R21	CO + 0.5O2 = CO2	–256.84	R22	H2 + 0.5O2 = H2O	–228.38
R23	CO + 3H2 = CH4 + H2O	–149.18

Effect of SO2

Since the amount of HgCl2 in the “Remove SO2” system tends to be stable at temperatures below 300 °C, and it is taken as the investigation temperature. In comparison with the “Remove SO2” system, in the “All” system, the quantities of SO2, SO3, O2S(OH)2, H2SO4 (H2O)6(liq), HCl, and H2 increase, while HgCl2, O2, NO, HgO, Cl2, and Cl decrease as seen in Table .

Table 6

Amounts of Some Species at 300 °C in “All” and “Remove SO2” Systems

	n/mol			n/mol
species	All	Remove SO2	species	All	Remove SO2
H2O	1.06	1.06	HgCl2	1.37 × 10–9	1.26 × 10–8
N2	0.89	0.89	O2	2.63 × 10–11	6.76 × 10–4
CO2	6.34 × 10–1	6.34 × 10–1	NO	1.28 × 10–13	6.47 × 10–10
SO2	7.91 × 10–6	1.52 × 10–9	HgO	2.08 × 10–14	1.94 × 10–13
HCl	6.58 × 10–6	6.55 × 10–6	Cl2	8.47 × 10–15	4.22 × 10–11
O2S(OH)2	2.36 × 10–6	2.30 × 10–6	H2	7.72 × 10–15	1.54 × 10–18
SO3	3.13 × 10–7	3.03 × 10–7	Cl	9.08 × 10–16	6.42 × 10–14
Hg	1.12 × 10–8	2.07 × 10–11	H2SO4(H2O)6(liq)	1.36 × 10–3	8.39 × 10–6

In the system where SO2 exists, the amount of SO2 increases, and then O2 is consumed to generate H2SO4 (R13), resulting in the increase of H2SO4 (H2O)6(liq). Meanwhile, the higher concentration of SO2 in the flue gas will also react with Cl2 and H2O to produce HCl and H2SO4 (R14). Therefore, Cl/Cl2 decreases and HCl increases. However, according to the above analysis, Hg0 cannot directly react with HCl to produce HgCl2. Therefore, the presence of SO2 consumes Cl and Cl2, which is the main reason for the reduction of the Hg0 oxidation efficiency. The increase of SO3 is due to the oxidation of SO2 by O2 (R15). In “Remove SO2,” the amount of SO3 is more than that of SO2, so SO3 reacts with H2 to produce SO2 and H2O (R16). In the system where SO2 exists, the amount of SO2 is more than that of SO3, so SO2 consumes O2 to produce SO3 (R15). Under such a circumstance, H2 is not consumed, and therefore, the amount of H2 increases.

Effect of CO2

For CO2, it is conducive to Hg0 oxidation in the range of 300–400 °C. Thus 300 °C is taken as the investigation temperature. Table shows that the CO2, O2, and CO gases in the “All” system increase, as well as HgCl2, HgO, O2S(OH)2, SO3, NO, Cl2, Cl, and HOCl. Since the amount of CO2 is 0.634mol, accounting for a large part of the whole system, the overall C and O will be greatly reduced after CO2 is removed. On the contrary, in the presence of CO2, O2 and CO also increase, and the increase of O2 leads to the increase of HgO. At the same time, the increased O2 reacts with HCl to produce Cl2 (R1), which eventually leads to the increase of HgCl2. It can be seen that the addition of CO2 leads to the increase of O2, which is the key to the improvement of Hg0 oxidation efficiency. The increase of O2 lead to the increase of Cl2, Cl, HOCl, SO3, and NO (R1–R3, R15, and R17).

Table 7

Amount of Some Species at 300 °C in “All” and “Remove CO2” Systems

	n/mol			n/mol
species	All	Remove CO2	species	All	Remove CO2
H2O	1.06	1.06	O2	2.63 × 10–11	2.18 × 10–13
N2	0.89	0.89	NO	1.28 × 10–13	1.16 × 10–14
CO2	6.34 × 10–1	9.70 × 10–8	HgO	2.08 × 10–14	2.41 × 10–15
SO2	7.91 × 10–6	7.91 × 10–6	Cl2	8.47 × 10–15	8.89 × 10–16
HCl	6.58 × 10–6	6.58 × 10–6	H2	7.72 × 10–15	7.36 × 10–14
O2S(OH)2	2.36 × 10–6	3.28 × 10–7	Cl	9.08 × 10–16	2.56 × 10–16
SO3	3.13 × 10–7	3.28 × 10–8	OH	8.46 × 10–16	2.38 × 10–16
Hg	1.12 × 10–8	1.24 × 10–8	HOCl	7.76 × 10–16	8.14 × 10–17
HgCl2	1.37 × 10–9	2.11 × 10–10	CO	1.12 × 10–16	1.63 × 10–22

Effects of Different Oxygen Carriers

This part aims to analyze the Hg0 removal performance of common oxygen carriers in CLC using the simulation method discussed previously. Considering that the O element in the OC can promote the oxidation of Hg0, the oxygen content of other oxygen carriers will be controlled to provide the same amount of oxygen atoms. By comparing the Hg0 oxidation efficiency of eight oxygen carrier systems, the order of Hg0 removal performance of oxygen carriers can be obtained. The order of Hg0 removal efficiency from high to low is Co3O4, CuO, Mn2O3, CaSO4, Fe2O3, CeO2, Al2O3, and SiO2.

In the process of CLC, due to the addition of different oxygen carriers, the reactions will be different. Therefore, the gasification products in this process will also be different. It is well known that the conversion of Hg0 to Hg2+ depends largely on these gases. According to the first stage simulation, the gases produced in the gasification process mainly include CO2, SO2, H2, CO, HCl, NO, and O2. Among these gas components, the content of O2 is closely related to the Hg0 removal efficiency. In Co3O4 and CuO systems, with relatively high oxygen content, their Hg0 removal efficiency is high. At the same time, due to the different content of O2, the amount of Cl and Cl2 produced by the homogeneous reaction is also different. It can be found from Figure that the oxidation efficiency of Hg0 is directly related to the amount of O2, Cl, and Cl2.

Figure 6

Amount of gasification components of each system.

In our existing experiments, the order of Hg0 removal efficiency from high to low is Co3O4 > Mn2O3 > Fe2O3 > CuO > CaSO4 > CeO2 > SiO2 > Al2O3. The proportions of Hg0 and Hg2+ released with respect to total mercury were 5.01, 6.34, 6.46, 9.66, 33.7, 68.99, 75.96, and 91.24%, respectively. As shown in Figure , the concentrations of Hg0 and Hg2+ were integrated to find the total amount of gaseous mercury released during the experiment, the less elemental mercury was released, the more oxidized mercury was produced. It can be found that there is a difference between the experiment and the simulation, because the simulation only considers the gas interaction in the cooling process, but does not consider the conversion of mercury by OC itself. Both homogeneous and heterogeneous reaction pathways are important for Hg0 removal. OC can promote the conversion of Hg0 to Hg2+/HgP. Pérez-Vega et al. used CuO as OC and found that HgP accounted for 59.4% of the total Hg.[18] It was reported that mercury on OC (Co3O4@TiO2@Fe2O3) was mainly desorbed in the form of HgO, which was beneficial for mercury removal.[15] It was found that 24.14% of the mercury in the coal migrated to the OC (CuFe2O4) with the forms of Hg0, HgO, and HgCl2 in CLG.[27] Ma et al. analyzed the mercury on the oxidized OC (CuO@TiO2-Al2O3) by X-ray photoelectron spectra (XPS). XPS spectra over the spectral regions of Hg 4f were evaluated, and HgCl2, HgO, HgS, and Hg0 were detected.[13] Therefore, the heterogeneous reaction between mercury and OC needs a further in-depth study.

Figure 7

Total amount of gaseous mercury.

O2-Induced Hg0 Oxidation

From the discussion in previous sections, it can be found that different gas components affect the content of O2, thus indirectly affecting the oxidation efficiency of Hg0. As shown in Figure , O2 is in the core position, around which are different flue gas components, and the outermost ones are the stable species. In addition to O2, Cl, Cl2, HOCl, and HgCl are also more active gas components. For example, the presence of HCl consumes O2 and generates Cl2, which results in an increase in the amount of HgCl2; NO indirectly affects HCl as NH3 and consumes O2; the presence of SO2 also consumes O2 first, and then Cl2 and Cl, so it can inhibit the Hg0 oxidation; the amount of CO2 is large, and it can decompose into O2 at a certain temperature, so as to promote Hg0 oxidation.

Figure 8

O2-induced Hg0 oxidation in chemical looping combustion.

It can be seen from the above discussion that different oxygen carriers affect the Hg0 oxidation efficiency, which is closely related to O2. The stronger the oxygen release capacity of oxygen carriers, the higher the oxidation efficiency of Hg0. Moreover, the oxygen carrier with stronger oxygen release is also conducive to combustion, which is more suitable for CLC.

Conclusions

A two-stage simulation method is used in this paper. The first stage is the gasification process of coal and the oxygen carrier, and then the gasification products of this process are used as the reactants in the second stage. The thermodynamic method is used to simulate the transformation of mercury in the cooling process. The simulation results are in good agreement with the experimental results, indicating that this method is reliable.

In the flue gas cooling process, HCl, NO, and CO2 contribute to the transformation of mercury, and SO2 inhibits Hg0 oxidation. HCl reacts with O2 to form Cl/HOCl, which oxidizes HgCl to HgCl2, thus promoting Hg0 oxidation; the presence of NO indirectly affects the content of HCl, resulting in more HCl participating in Hg0 oxidation; a large amount of CO2 decomposes into O2 and reacts with HCl to promote Hg0 oxidation; SO2 consumes O2 and Cl2, and inhibits Hg0 oxidation.

By comparing the Hg0 oxidation efficiency of eight oxygen carrier systems, the order of Hg0 removal efficiency from high to low is Co3O4, CuO, Mn2O3, CaSO4, Fe2O3, CeO2, Al2O3, and SiO2, and the O2 content in gasification products in these systems is also in this order.

Different flue gas components directly or indirectly affect the O2 content, through which the Hg2+ content is affected. Different oxygen carriers have different oxygen release capacities and different Hg0 removal effects. Therefore, O2 is the core affecting Hg0 transformation. The selection of oxygen carriers with strong oxygen release is more conducive to the removal of Hg0 in chemical looping combustion.