Effect of Lime Addition to CaSO ₄ Oxygen Carrier in Chemical Looping Combustion

Abstract

Chemical-looping combustion (CLC), which has the characteristic of greenhouse gas CO₂ inherent separation, is a novel combustion technology. In this study, CLC experiments of methane using CaSO₄ oxygen carrier with lime addition were carried out in a batched fluidized bed reactor, where the sample was exposed to alternate oxidizing and reducing conditions. The influences of temperature, calcium-to-sulphur ratio and lime particle size on the conversion of CaSO₄ and sulfur capture were investigated and a suitable operation condition was determined. Under the optimal operation condition, a multi-cycle test was performed to evaluate the cyclic redox behavior of the lime-promoted CaSO₄ sample. X-ray diffraction and a field emission scanning electron microscope were used to characterize the phase and surface morphology of the samples used. The results show that the addition of lime could improve the conversion rate of CaSO₄ and the capture efficiency of sulfur-containing gases. The operation conditions of calcium-to-sulfur ratio 0.8, lime particle size of 180-250 μm and operation temperature of 900 °C turned out to be the optimal conditions. Besides, the average desulphurization rate of lime was up to 78.77% during the cyclic test.

Keywords:
Chemical-looping combustion; lime; CaSO₄; sulfur capture

INTRODUCTION

It is well known that fossil fuel consumption is the major source of anthropogenic CO₂ emission. There are numerous techniques that can potentially be used for CO₂ separation and compression from combustion, but most of them are considered to be energy intensive (Stern and Hatton, 2014; Liu et al., 2005; Notz et al., 2007). Chemical-looping combustion (CLC) is a novel combustion technology where CO₂ is separated from the flue gases without such an energy-consuming gas separation process. (Ding et al., 2015; Abad et al., 2006; Johansson et al., 2006; Adánez et al., 2009).

CLC involves the use of oxygen carrier, which transfers oxygen from air to allow fuel oxidation, thus avoiding the direct contact between fuel and air. Typically, a dual fluidized bed system composed of an air reactor (AR) and a fuel reactor (FR) is widely applied in CLC process (Ding et al., 2012). In the FR, fuel (natural gas, refinery gas, synthesis gas from coal gasification, etc.) is oxidized via an oxygen carrier to produce CO₂ and steam. After a steam condensation process, high-purity CO₂ can be easily obtained without extra energy consumption. The reduced oxygen carrier is further transferred to the AR where it is re-oxidized by air. Since this is an exothermic reaction, the temperature of gases consisting of N₂ and residual O₂ increases substantially. After that, the oxygen carrier with high oxygen potential is recirculated to the FR for a new cycle. The reduction reaction is either endothermic or slightly exothermic depending on the kinds of the oxygen carrier and the fuel. The total amount of heat evolved from the two reactions is equivalent to that of normal combustion of the same fuel (Cao and Pan, 2006; Scott et al., 2006; de Diego et al., 2008).

CaSO₄, as a low cost potential oxygen carrier for CLC, has attracted increasing attention all over the world recently. ALSTOM Power Co. Ltd. (Andrus et al., 2009) studied the limestone-based chemical looping process, and a plant facility at a power scale of 3 MW_th using CaSO₄ as the oxygen carrier has been constructed and operated. Shen and Xiao at the Southeast University in China also investigated the feasibility of CaSO₄ as an oxygen carrier (Shen et al., 2008; Song et al., 2008a), reactivity test with gaseous and solid fuels (Zheng et al., 2010; Song et al., 2008b), and the reduction kinetics (Zheng et al., 2011; Xiao and Song, 2011). The results demonstrated that CaSO₄ was an alternative oxygen carrier with high oxygen capacity. However, it has been verified that the potential sulfur release results in a decline in the reactivity of CaSO₄ during the reduction/oxidation process. Thus, the sulfur release is a major challenge for the practical application of CaSO₄ oxygen carrier.

For CaSO₄/CaS oxygen carrier, the main reactions occuring in the FR and AR with methane as fuel are shown as follows:

$${\text{CaSO}}_{\text{4}}+{\text{CH}}_4\to \text{CaS}+{\text{CO}}_{\text{2}}+2{\text{H}}_{\text{2}}\text{O}\Delta H_{\text{298K}}^0=159.6\text{kJ/mol}$$ (1)

$$\text{CaS}+2{\text{O}}_{\text{2}}\to {\text{CaSO}}_{\text{4}}\Delta H_{\text{298K}}^0=-962.4\text{kJ/mol}$$ (2)

As illustrated in many previous investigations (Anthony and Granatstein, 2001; Mattisson and Lyngfelt, 1998; Tian et al., 2010; Ding et al., 2011), the emission of sulfur related to CaSO₄ and CaS varied with the operating conditions. In this paper, SO₂ and H₂S are mainly formed from the following side reactions during the reduction (R3, R4, R5) and oxidation (R5, R6).

$${\text{CaSO}}_{\text{4}}{\text{+CH}}_{\text{4}}\to {\text{CaO+CO}}_{\text{2}}{\text{+H}}_{\text{2}}{\text{O+H}}_{\text{2}}\text{S}\Delta H_{\text{298K}}^0=221.17\text{kJ/mol}$$ (3)

$${\text{4CaSO}}_{\text{4}}{\text{+CH}}_{\text{4}}\to 4{\text{CaO+CO}}_{\text{2}}{\text{+2H}}_{\text{2}}{\text{O+4SO}}_{\text{2}}\Delta H_{\text{298K}}^0=1220.80\text{kJ/mol}$$ (4)

$${\text{3CaSO}}_{\text{4}}\text{+CaS}\to 4{\text{CaO+4SO}}_{\text{2}}\Delta H_{\text{298K}}^0=1061.34\text{kJ/mol}$$ (5)

$$\text{CaS}+3/2{\text{O}}_{\text{2}}\to \text{CaO}+{\text{SO}}_{\text{2}}\Delta H_{\text{298K}}^0=-456.73\text{kJ/mol}$$ (6)

The sulfur release will lead to atmospheric pollution, and decrease the oxygen transfer capacity of CaSO₄ oxygen carrier. Therefore，it is necessary to reduce the sulfur release in the CLC of CaSO₄ oxygen carrier. Limestone (CaCO₃) particles are widely used in circulating fluidized beds for sulfur capture (Saastamoinen and Shimizu, 2007; Bragança et al., 2003,2009; Sebag et al., 2001). The addition of limestone sorbent may be a reasonable method to capture sulfur and decrease the sulfur release during the reduction/oxidation process (Song et al., 2008b). If the fresh limestone particles are fed into the FR, CaCO₃ can be converted to CaO (R7) depending on the partial pressure of CO₂. Therefore, after CaO or CaCO₃ addition, the following reactions may take place in the reactors:

$${\text{CaCO}}_{\text{3}}\to \text{CaO}+{\text{CO}}_{\text{2}}\Delta H_{\text{298K}}^0=178.16\text{kJ/mol}$$ (7)

$${\text{CaO+H}}_2\text{S}\to {\text{CaS+H}}_{\text{2}}\text{O}\Delta H_{\text{298K}}^0=-61.40\text{kJ/mol}$$ (8)

$${\text{CaO+SO}}_{\text{2}}\to {\text{CaSO}}_{\text{3}}\Delta H_{\text{298K}}^0=-227.65\text{kJ/mol}$$ (9)

$${\text{CaO+3CO+SO}}_{\text{2}}\to {\text{CaS+3CO}}_{\text{2}}\Delta H_{\text{298K}}^0=-392.16\text{kJ/mol}$$ (10)

$${\text{CaCO}}_{\text{3}}{\text{+H}}_{\text{2}}\text{S}\to {\text{CaS+H}}_{\text{2}}{\text{O+CO}}_{\text{2}}\Delta H_{\text{298K}}^0=72.77\text{kJ/mol}$$ (11)

The oxygen carrier particles, desulphurization products (CaS and CaSO₃), and residual CaO are all transferred into AR and these bed materials can be oxidized to CaSO₄ in the oxidation condition. Besides, the SO₂ released in the AR may also be captured by CaO:

$${\text{CaO+SO}}_{\text{2}}+1/2{\text{O}}_2\to {\text{CaSO}}_{\text{4}}\Delta H_{\text{298K}}^0=-505.89\text{kJ/mol}$$ (12)

In other ways, the sulfur-containing gases released in the process of the fuel conversion, i.e., H₂S and SO₂, may have a positive effect on fuel gasification in CLC of solid fuel (Mattisson and Lyngfelt, 1999; Lyon and Cole, 2000). The influence of lime and limestone addition to ilmenite oxygen carrier was investigated in a laboratory-batched fluidized bed reactor (Teyssié et al., 2011) and in a continuous 10 kW_th CLC pilot reactor with solid fuel (Cuadrat et al., 2011), respectively. The addition of CaO to ilmenite improved the gas conversion, which can be explained by the catalytic effect of CaO on the conversion of CO to H₂ (by the water-gas shift reaction) and the high reactivity of ilmenite oxygen carrier toward H₂.

In this study, thermodynamics and kinetics of CaSO₄ oxygen carrier with lime addition in CLC of methane was discussed. To be more specific, the effects of temperature, calcium to sulphur ratio and lime particle size on conversion of CaSO₄ and sulfur capture were analyzed. Moreover, the used oxygen carriers were also characterized through X-ray diffraction (XRD) analysis and field emission scanning electron microscope (FESEM).

EXPERIMENTAL SECTION

Material

The oxygen carrier particle used in the present study was originated from the natural anhydrite ore (provided by the Wanbei Coal-Electricity Group CO., LTD), which was further crushed by pulverizer and sieved to a size range of 180-250 μm. The natural anhydrite ore is composed of CaSO₄ and a small proportion of other impurity, as presented in Table 1. The apparent density and bulk density of the particles are 2950 and 1510 kg/m³, respectively.

Table 1
Composition of natural anhydrite (wt.%).

Chemical composition	Value
CaSO4	95.02
CaO	1.25
SiO2	0.65
MgO	0.46
Al2O3	0.25
TiO2	0.05
Fe2O3	0.02
Crystallization water	2.30

As the main components of the limestone, CaCO₃ accounts for a mass fraction of 95.67 % and the other components are MgO (1.75 wt.%), SiO₂ (1.26 wt.%), Al₂O₃ (0.82 wt.%), and Fe₂O₃ (0.50 wt.%). In a previous sulfur capture process, the optimum calcinations temperature was found to be 950 °C (Doğu, 1981). Therefore, the limestone was fully calcined in a fixed bed with an inert gas flow of nitrogen at 950 °C until the CO₂ disappeared in the exhaust gas, and the lime obtained was used in the CLC experiments. The composition of the samples used was determined by XRD test using Cu-Kα radiation over a 2θ range of 15-85^o. The surface morphology of samples was measured by FESEM (Sirion 200, Holland).

Experimental Setup

Figure 1 shows the experimental apparatus used for testing the oxygen carriers. It consisted of a gas feed, a fluidized reactor, a tube furnace, a cooler and a gas analysis system. A porous distributor plate was located in the middle stainless-steel tube (I.D. =40 mm, length =900 mm). The reactor was heated in an electric furnace, and the furnace temperature was controlled by a K-type thermocouple between the reactor tube and the heater, while the temperature was also monitored by another K-type thermocouple inside the oxygen carrier bed materials. The reactor had two connected pressure taps in order to measure the differential pressure in the bed and monitor the fluidization state. The flow rates of fluidizing gas were measured by mass flow controllers. The product gas from the reactor flowed through a cooler filled with CaCl₂ desiccant where the steam was condensed, and then was sent to the gas analyzers. Three gas analyzers were used to measure the gas composition simultaneously. The concentrations of CO₂, CO and SO₂ as well as the outlet gas flow rate were measured by a GA-21 Plus flue gas analyzer, the H₂ concentration was measured by an online GASboard-3100 coal gas analyzer, and the concentrations of CH₄, O₂ and H₂S were measured by Geotech Biogas Check. Note that the gas concentration in the paper refers to the gas volume fraction in the outlet gases.

Figure 1
Schematic diagram of the experimental setup.

Experimental Procedure

The experiments were performed in a laboratory-batched fluidized bed reactor under atmospheric pressure. Pure natural anhydrite or the mixture of natural anhydrite and lime were used as the oxygen carrier. In each run, 100 g of quartz sand with size range of 1500-2000 μm, served for preheating and distributing the reactant gas uniformly through the bed, was added above the porous plate. Then a new batch of 50 g of natural anhydrite as well as a certain amount of lime (0 g, 10 g, 20 g, 30 g, 40 g and 50 g) was placed on the quartz sand bed with the inside thermocouple located in the middle of the layer. During the preheating period, the reactor was purged with nitrogen gas flow. After the temperature reached the desired value (850, 900, 925 °C) and kept stable, the reduction gas of 25 vol.% CH₄ balanced by N₂, which was selected to avoid exceeding the terminal velocity of the particles due to the volume expansion of the reactant gas, was continuously introduced and the reduction time was 20 min. To avoid CH₄ and O₂ contact, nitrogen was introduced for a period of 5 min after the reducing period. After purging with N₂, the bed materials were exposed to the gas environment of 10 vol.% O_2, balanced by N₂ for the re-oxidation of the oxygen carrier. Then the fluidizing gas was switched to N₂ for the next cycling test. It should be noted that the oxidizing medium with lower O₂ concentration was used instead of air, which was to avoid the release of excess amounts of heat in the re-oxidation process.

When the test was finished, the system was again purged with nitrogen until the product gas was cleared away and the furnace was shut down. The sample was cooled in the environment of nitrogen to ambient temperature and was collected for further analysis. It should be noted that, as an inert gas, the N₂ was not considered in the results of this study. The experimental conditions are presented in Table 2.

Table 2
Experimental condition.

Oxygen carrier	natural anhydrite with lime
Pressure (atm)	1
Reaction temperature (°C)	850, 900, 925, 950
Particle size (μm)	natural anhydrite: 180-250
	lime: 125-180, 180-250, 250-350
Particle mass (g)	natural anhydrite: 50
	lime: 10, 20, 30, 40, 50
Reduction gas	CH4/N2 = 25/75 %
Reduction gas flow (ml/min)	800
Reduction time (min)	20
Oxidation gas	O2/N2 = 10/90 %
Oxidation gas flow (ml/min)	1200
Oxidation time (min)	46
Diluent N2 flow (ml/min)	400 (reduction) and 0 (oxidation)
Sweeping gas flow (ml/min)	1000

Data Evaluation

The solid conversion of the oxygen carrier (X) is defined as:

$$X=\frac{m-m_{\text{red}}}{m_{\text{ox}}-m_{\text{red}}}$$ (13)

The conversion of the CaSçO₄ oxygen carrier as a function of time during the reduction period can be calculated based on the mass balance of oxygen from the CaSO₄ to the product gases, and the relationship is shown as below:

$$X_{\text{r}\text{,i}}=X_{\text{r}\text{,i-1}}-{\displaystyle {\int }_{t_{i-1}}^{t_i}\frac{M_o{\dot{n}}_{\text{out}}}{m_{\text{ox}}{R}_o}}(4y_{{\text{CO}}_{\text{2}}\text{,out}}+3y_{\text{CO}\text{,out}}-y_{{\text{H}}_{\text{2}}\text{,out}}+2y_{{\text{SO}}_{\text{2}}\text{,out}})dt$$ (14)

$$\frac{d{X}_{\text{i}}}{dt}=\frac{M_o{\dot{n}}_{\text{out}}}{{\text{m}}_{\text{ox}}{R}_o}(4y_{{\text{CO}}_{\text{2}}\text{,out}}+3y_{\text{CO}\text{,out}}-y_{{\text{H}}_{\text{2}}\text{,out}}+2y_{{\text{SO}}_{\text{2}}\text{,out}})$$ (15)

Similarly, the solid conversion during the oxidation period can be calculated as below:

$$X_{\text{o}\text{,i}}=X_{\text{o}\text{,i-1}}+{\displaystyle {\int }_{t_{i-1}}^{t_i}\frac{2M_o}{m_{\text{ox}}{R}_o}}\left[({\dot{n}}_{\text{in}}{y}_{{\text{O}}_{\text{2}}\text{,in}}-{\dot{n}}_{\text{out}}(4y_{{\text{O}}_{\text{2}}\text{,out}}+\frac{1}{2}{y}_{\text{CO}\text{,out}}+y_{{\text{CO}}_{\text{2}}\text{,out}}+y_{{\text{SO}}_{\text{2}}\text{,out}})\right]dt$$ (16)

$$\frac{d{X}_o}{dt}=\frac{2M_o}{m_{\text{ox}}{R}_o}\left[({\dot{n}}_{\text{in}}{y}_{{\text{O}}_{\text{2}}\text{,in}}-{\dot{n}}_{\text{out}}(4y_{{\text{O}}_{\text{2}}\text{,out}}+\frac{1}{2}{y}_{\text{CO}\text{,out}}+y_{{\text{CO}}_{\text{2}}\text{,out}}+y_{{\text{SO}}_{\text{2}}\text{,out}})\right]$$ (17)

Note that some of the oxygen in the air reactor was consumed by the deposited carbon, and some of the oxygen was released in the form of SO₂.

The sulfur release rate is defined as the ratio of moles of sulfur exiting the reactor on a dry basis to moles of sulfur-containing compound entering the reactor:

$${\chi }_s={\displaystyle {\int }_{t_{i-1}}^{t_i}\frac{{\dot{n}}_{\text{out}}(y_{{\text{SO}}_{\text{2}}\text{,out}}+y_{{\text{H}}_{\text{2}}\text{S}\text{,out}})}{n_{CaS{O}_4}}}dt\times 100\%$$ (18)

The desulphurization rate can be defined through the sulfur release rate, and can be calculated as follows:

$${\eta }_{\text{s}}={\displaystyle {\int }_0^{t_i}\frac{n_{\text{out}}^{\text{o}}(y_{{\text{SO}}_{\text{2}}\text{,out}}^o+y_{{\text{H}}_{\text{2}}\text{S}\text{,out}}^o)-{\dot{n}}_{out}(y_{{\text{SO}}_{\text{2}}\text{,out}}+y_{{\text{H}}_{\text{2}}\text{S}\text{,out}})}{n_{\text{out}}^{\text{o}}(y_{{\text{SO}}_{\text{2}}\text{,out}}^o+y_{{\text{H}}_{\text{2}}\text{S}\text{,out}}^o)}}dt\times 100\%$$ (19)

An index S _I can be expressed as the relative intensity of the major peak of CaO divided by the relative intensity of the major peak of CaSO₄:

$$S_I=\frac{I_{\text{CaO}}}{I_{{\text{CaSO}}_{\text{4}}}}$$ (20)

The equilibrium constant of a chemical reaction (K_p) is the value of the reaction quotient when the reaction has reached equilibrium. For a general reaction aA(g) + bB(s) = pD(g) + qE(s), where A, B and D, E denote the reactants and products, respectively, and a, b, p and q refer to the stoichiometric number), it can be calculated as follows:

$$K_p=\frac{C_D^{\text{p}}}{{\text{C}}_{\text{A}}^{\text{a}}}=\frac{{\text{P}}_{\text{D}}^{\text{p}}}{{\text{P}}_{\text{A}}^{\text{a}}}$$ (21)

The standard Gibbs free energy of a reaction is related to the temperature, the ideal gas constant and the equilibrium of the specified reaction, and it can be expressed as:

$$\Delta \text{G}=-RT\ln K_{\text{p}}$$ (22)

RESULTS AND DISCUSSION

Desulphurization mechanism

Even though a great variety of reactants and products existed in these complex reactions, as illustrated in R1-R12, there were three kinds of solid matter, which were CaSO₄, CaS and CaO, respectively. Therefore, as one of the products in the side reactions, lime was added to CaSO₄ oxygen carrier in CLC of methane to achieve the desulphurization goal.

Thermodynamic analysis of the relevant reactions which illustrate the products distribution was discussed to investigate the desulphurization mechanism in this section. Figure 2 shows the equilibrium constants (K _p) of the CaSO₄ reduction reactions with lime addition as a function of temperature range from 600 °C to 1200 °C. The reduction of CaSO₄ to CaS (R1) was the main reduction reaction due to the larger equilibrium constant, and the reaction equilibrium would be shifted toward the right side with the increase of temperature. Meanwhile, the equilibrium constants of the SO₂ or H₂S generation reactions (R3, R4, R5) also increased with temperature, which was consistent with the previous investigations (Song et al., 2008a,b; Zheng et al. 2010,2011). As a result, the temperature between 850 °C and 950 °C was chosen as the experimental temperature in the present work.

Figure 2
Equilibrium constants of relevant reactions in the reducing atmosphere

The lime particles changed the fluidization state in the fluidized bed so that the contact between the sulfur-containing gases and the CaO particles was improved, which means that the reaction rate would be enhanced. The standard Gibbs free energy of the reactions R8 and R9 was close to zero in the temperature range from 850 °C to 950 °C, indicating that the two reactions were reversible reactions and neither the forward nor the reversed reaction dominated. Therefore, the reactions R8 and R9 were more likely to be affected by the addition of the reactants or products, which was our starting point on sulfur-containing gases removal. The forward reactions corresponded to SO₂ or H₂S release (R3, R4) were dominant due to the CH₄-rich environment. However, after the lime addition, the reactions with smaller equilibrium constants (R8, R9) would go toward to the direction in which the sulfur-containing gases were consumed.

During the process of CaS oxidization, the standard Gibbs free energy of the relevant reactions was calculated over a wide range of temperatures, as shown in Figure 3. The standard Gibbs free energy of the reaction R2 was less than those of reactions R6 and R12 in the temperature range of 800-1000 °C, which demonstrated that the reaction R2 took precedence over reactions R6 and R12, and the main product of CaS oxidization was CaSO₄ rather than CaO. Similarly, the reaction direction of R9 was more probably influenced by the addition of lime because the Standard Gibbs free energy approximated to zero in the temperature range from 850 °C to 950 °C. As the residual CaO circulated to the air reactor, SO₂ released in the oxidization period could be removed through the forward reaction of R9. Based on the thermodynamic analysis, the idea of suppressing the sulfur-containing gases release by lime addition was feasible in the reduction and oxidization conditions, and the influence of the lime addition to CaSO₄ oxygen carrier was experimentally investigated in the next sections.

Figure 3
Standard Gibbs free energy of relevant reactions in the oxidizing atmosphere

Effect of temperature

To evaluate the influence of lime addition on the solid conversion and sulfur capture efficiency of CaSO₄ oxygen carrier in CLC, experiments using natural anhydrite as the reference group were conducted at different temperatures of 850, 900, 925 and 950 °C. The added weight of lime was set to 10 g and the particle size was 180-250 μm. The concentrations of SO₂ and H₂S in the experiment at 950 °C exceeded the measuring range of the gas analyzers; thus, these results were not exhibited here.

Figure 4 shows the variation of the product gases concentrations on a dry basis as a function of time at 900 °C. During the reduction period, CO₂ and H₂O were produced by the reaction between CaSO₄ and parts of CH₄. It could be seen that the CO₂ concentration in the experiment of CaSO₄ with 10 g of lime addition was almost twice that without lime, which indicated that lime could improve the reactivity of CaSO₄ to CaS (R1). At the same time, a small part of CH₄ was partially oxidized to CO and H₂O, while another small part of CH₄ was decomposed into H₂ and carbon, and the carbon deposited on the particle surface. The concentrations of the by-product gases (i.e., H₂S, SO₂, H₂ and CO) moderately decreased with the addition of lime compared to that without lime addition, which demonstrated that the by-product gases release could be evidently suppressed by lime addition.

Figure 4(b) and (d) show the variation of gas concentrations during the oxidation processes. Oxygen was not observed from the outlet during the early stage, which implied that the reaction rate at the initial time was so fast that a rapid conversion was obtained. Then the reaction rate gradually slowed down, and the O₂ concentration increased to 10 vol.% at the end of the oxidation period. That meant the conversion of CaS to CaSO₄ was completed at last, which could be proved by the XRD analysis. Note that little amount of CO₂ and SO₂ were observed at the beginning of the oxidation. The release of CO₂ can be explained as follows: some carbon was generated and then deposited on the surface of the particle in the reduction period, while in the oxidation period, this carbon would be oxidized by oxygen to produce CO₂. According to the literature, the SO₂ release could mainly be attributed to the side reactions R5 and R6 (Song et al., 2008b).

Figure 4
Product gas concentrations and dry gas flow during the first cycle: (a) and (b) CaSO₄ without lime, (c) and (d) CaSO₄ with lime. T = 900 °C, lime mass = 10 g, lime particle size = 180-250 μm. The dashed lines indicate the switch to an inert atmosphere.

The effects of temperature and the lime addition on the conversion rate of CaSO₄ during the reduction/oxidation period are shown in Figure 5. It could be observed that the degree of reduction and oxidation increased with temperature, which was in agreement with some other investigations (Song et al., 2008a; Zheng et al., 2011). This result also indicated that the conversion rate of CaSO₄ increased with the temperature in the reduction period. Besides, the conversion rate of CaSO₄ in the experiment with lime addition was larger than that without lime addition. It should be noted that the conversion rate of CaSO₄ showed the same value in the oxidation period, because the reaction rate was controlled by the oxygen aeration rate (Su et al., 2016). The improvement of the conversion rate may be attributed to the catalysis of lime on the water-gas shift reaction (Teyssié et al., 2011). During the oxidation period, the oxygen carrier conversion could not be restored to the same level as before. As will be discussed later, this phenomenon resulted from the sulfur release during the reduction/oxidation process. Note that the conversion rate of CaSO₄ with lime addition at 900 °C was even slightly larger than that without lime at 925 °C. This could be attributed to the sulfidation reaction of lime, which was favorable to the cyclic redox behavior of CaSO₄ oxygen carrier. Overall, the lime in the reactor could have a beneficial effect on the conversion of CaSO₄.

Figure 5
Oxygen carrier conversion (X) as a function of time in the temperature range of 850 -925 °C.

Figure 6 shows the sulfur release rate of the CaSO₄ oxygen carrier as a function of time during the first cycle. A positive correlation can be observed between the sulfur release rate and the temperature. As illustrated above, this may also be explained by the side reactions, including the competing reduction (R3, R4), solid-solid reaction (R5) and partial oxidation (R6). Satisfactorily, the sulfur release rate of CaSO₄ with lime addition was significantly lower than that without lime addition at the same temperature. And the desulphurization rates after the lime addition were up to 35.13%, 62.90% and 50.50% at 850, 900 and 925 °C, respectively. Based on the above analysis, the subsequent experimental temperature for CaSO₄ oxygen carrier with lime addition was set as 900 °C.

Figure 6
The sulfur release rate (X _s) of oxygen carrier as a function of time.

Effect of the calcium to sulfur ratio

The influence of the calcium to sulfur ratio (0.2, 0.4, 0.6, 0.8, 1.0, correspond to added weight of lime: 10g, 20g, 30g, 40g, 50g) on the performance of CaSO₄ and sulfur release was studied under the same particle size (180-250 μm) at 900 °C. The results are shown in Figure 7 and Figure 8, respectively. It can be seen from Figure 7 that the conversion rate of CaSO₄ increased with the calcium to sulfur ratio increasing in the reduction period. Interestingly, the conversion rate of CaSO₄ with 40 g of lime addition was very close to that with 50 g of lime addition. As shown in Figure 8, the lime addition resulted in the decrease of the sulfur release rate. Generally speaking, the more lime was added, the more the sulfur release decreased. The curves of the sulfur release rate had little differences between the calcium to sulfur ratios of 0.8 and 1.0 during the reduction period, while the sulfur release rate of CaSO₄ with 40 g lime addition was slightly lower than that with 50 g lime addition during the oxidation period. It was likely that the excess lime would cover the surface of CaS particles and reduce its contact with O₂, thus resulting in more sulfur release. The desulphurization rates of the various lime addition weights from 10 g to 50 g reached 62.90%, 67.98%, 76.28%, 81.67% and 80.97%, respectively. Considering the above results, 0.8 for the calcium to sulfur ratio was recommended in practical applications.

Figure 7
Oxygen carrier conversion (X) as a function of time at 900 °C with various masses of lime.

Figure 8
The sulfur release rate (X _s) of oxygen carrier as a function of time at 900 °C with various masses of lime.

Effect of lime particle size

The lime particle size is a significant factor that influences the desulfurization rate in a traditional combustion system (Saastamoinen and Shimizu, 2007). In this section, the influence of lime particle size (range from 125 μm to 350 μm) on the desulfurization rate and the performance of CaSO₄ was investigated. The temperature was set to be 900 °C and the added weight of lime was 40 g.

The effect of lime particle size on the performance of the lime-promoted CaSO₄ oxygen carrier as a function of time at 900 °C is shown in Figure 9. The particle size of lime had little effect on the conversion of CaSO₄ oxygen carrier, in addition to a little difference in the final stage of oxidation, where the conversion of the oxygen carrier with larger lime particle size was slightly lower than that with the smaller one. In fact, it resulted from the sulfur release, which was illustrated in Figure 10. As shown in Figure 10, the decrease of the lime particle size resulted in the decrease of the sulfur release rate. Besides, it should be noted that the sulfur release rate of CaSO₄ with the lime particle size of 125-180 μm was approximately equal to that with 180-250 μm. Corresponding to the various particle size ranges of the lime (125-180 μm, 180-250μm, 250-350μm), the desulphurization rates were up to 79.38%, 80.97% and 81.56%, respectively. As mentioned in the literature (Saastamoinen and Shimizu, 2007; Yrjas et al., 1996), lime with smaller size had a higher reactivity. However, the small particles were more inclined to attrition and fragmentation, and the small particles could be carried out by the gas stream. Therefore, the 180-250 μm particle size of lime was recommended.

Figure 9
The effect of lime particle size on the conversion of CaSO4 as a function of time at 900 °C.

Figure 10
The effect of lime particle size on the sulfur release rate as a function of time at 900 °C.

Multicycle test

In Figure 11(a), the outlet concentrations of CO₂, CO, CH₄ and O₂ are shown as a function of time. Most of the inlet CH₄ reacted with CaSO₄ to produce CO₂ and H₂O. The CO₂ concentration increased gradually in the first three reductions, and then decreased in the last two cycles. Meanwhile, the concentrations of CO and H₂ increased in the later stage of the fourth or fifth reduction process, which implied that the oxygen transfer capacity of CaSO₄ oxygen carrier decreased. Overall, it could be concluded that the reactivity of CaSO₄ increased for the initial cycles, but slightly decreased after three cycles. The above experiments illustrated that CaSO₄ oxygen carrier should not be reduced deeply in the reduction period. If the CaSO₄ oxygen carrier is reduced deeply, more CO and H₂ would be produced, instead of CO₂ and H₂O.

Figure 11(b) shows the variation of SO₂ and H₂S concentrations as a function of time during the cyclic test. It was clear that the H₂S and SO₂ concentrations were relatively stable during these five reduction reactions, but the SO₂ concentration increased during these five oxidation reactions, which may be due to the better adaptability of lime in the reduction process than in the oxidation process.

Figure 11
Products gas concentrations as a function of time during the five cycles: (a) outlet CO₂, CH₄, H₂, CO and O₂ concentrations, (b) outlet SO₂ and H₂S concentrations.

The conversion of oxygen carrier during the five cycles test is shown in Figure 12. It could be seen that the conversion of the oxygen carrier decreased gradually after reduction, which indicated that more oxygen in the oxygen carrier was utilized. The conversion could not restore the same level as before after oxidation, which could be ascribed to the release of sulfur. The majority of sulfur-containing gases were captured by the lime compared to the sulfur release of CaSO₄ oxygen carrier without lime addition (Song et al., 008b). However, the average desulphurization rate of lime was only 78.77% and the residual sulfur-containing gases would escape from the reactor, due to the limited contact between lime and the high-speed gas flow in the fluidized bed reactor. If the oxygen carrier particle made up of CaSO₄-core and pore-riched lime-shell was used, a better desulphurization rate could be achieved.

Figure 12
Variation of the oxygen carrier conversion as a function of time.

Characterization analysis

Phase characterization

The XRD patterns of the CaSO₄ oxygen carrier with and without lime addition are shown in Figure 13. It was evident that the main crystalline phase in the oxygen carrier was CaSO₄, as Figure 13(a) shows. Combined with the gas concentration analysis above, it could be concluded that most of CaSO₄ at 850 °C was not converted in the reduction period due to the low reactivity of CaSO₄. The index S _I, which meant the ratio of the relative intensity of the major peak for CaO phase to that for CaSO₄ phase, was introduced to express the relative content of CaO phase in the used oxygen carrier semi-quantitatively. The major peak of CaSO₄ phase was located at 2θ=25.52^o, while the major peak of CaO phase was located at 2θ=37.35^o. According to Figure 13, the index S _I and the relative standard deviation of S _I were calculated and presented in Table 3. The relative standard deviation of S _I could be used to evaluate the content of sulfur release or sulfur removal. The larger the relative standard deviation was, the more the sulfur was released from CaSO₄ oxygen carrier. It could be found from Table 3 that the sulfur release increased with temperature. Besides, the sulfur release decreased with the lime weight increase at the same temperature, which was consistent with the gas concentration analysis. The full-width half-maximum (FWHM) of the majorpeak could be used to represent the grain size of CaSO₄ in the used oxygen carrier. The FWHM of the sample (e) was 0.1574, while the value increased to 0.1771 after the 5^th cycle, which suggested that the grain size of CaSO₄ decreased.

Figure 13
XRD patterns of a CaSO₄ oxygen carrier: (a) at 850 °C, (b) at 900 °C, (c) 925 °C, (d) at 900 °C with 10 g lime, (e) at 900 °C with 40 g lime and (f) at 900 °C after the 5^th cycle oxidation.

Table 3
The index S _I and the relative standard deviation of S _I.

Sample	Index S I	Theoretical index of S I a	Relative standard deviation (%)
(a)	0.0138	0.0132	4.55
(b)	0.0271	0.0132	105.30
(c)	0.0820	0.0132	521.21
(d)	0.2014	0.1113	80.98
(e)	0.6573	0.5453	20.54
(f)	1.8160	1.6970	11.40

^a The theoretical index of S _I comes from the refinement of the XRD spectrum according to the different mixing ratio of natural anhydrite and lime, and the content of CaO in natural anhydrite is also taken into account.

Surface morphology

Figure 14 shows the FESEM images of the surface morphology of the oxygen carrier particles after the oxidation. After the first cycle, the surface topography and the grain structures at 850 °C in Figure 14(a) were basically similar to that at 900 °C in Figure 14(b). However, as shown in Figure 14(c) , a part of the grains on the external surface of the used oxygen carrier began sintering at 950 °C. The grain sintering on the external surface resulted from CaO formation (i.e., sulfur release) in the used oxygen carrier (Song et al., 2008b). So CaO formation from CaSO₄ was unfavorable. When lime was added to CaSO₄, the surface of the samples became rougher and the size of grains dramatically declined, as shown in Figure 14(d-f) . The change of surface morphology was beneficial to the diffusion of reactant gases into the core of oxygen carrier particles, thus facilitating the gas-solid reaction. Therefore, the addition of lime improved the performance of CaSO₄ oxygen carrier.

CONCLUSIONS

In this paper, CLC experiments of methane using CaSO₄ oxygen carrier with lime addition were carried out in a laboratory-batched fluidized bed reactor. The idea of removing sulfur-containing gases through lime addition was thermodynamically feasible in alternate oxidation and reduction processes. The effects of temperature, the calcium to sulfur ratio and lime particle size on CaSO₄ conversion and sulfur capture were investigated and a suitable operation condition was determined. Under the optimal condition, the multi-cycle test was conducted to evaluate the cyclic redox behavior.

Figure 14
FESEM micrographs of a CaSO₄ oxygen carrier: (a) at 850 °C, (b) at 900 °C, (c) 925 °C, (d) at 900 °C with 10 g lime, (e) at 900 °C with 40 g lime and (f) at 900 °C after the 5^th cycle of oxidation.

The addition of lime could improve the conversion rate of CaSO₄, which might be explained by the catalysis of lime on the water-gas shift reaction and the surface morphology change of the oxygen carrier after the lime addition. Both the reaction rate and the conversion of CaSO₄ were improved at high temperature. Besides, the sulfur-containing gases were captured by the lime additive. The analyses of the sulfur release and the XRD verified that the more lime was added, the higher the desulphurization rate achieved. The optimal operation condition was described as follows: 0.8 for the calcium to sulphur ratio, 180-250 μm for the particle size range of added lime and 900 °C for the operation temperature. The majority of sulfur-containing gases were captured during the cyclic test and the average desulphurization rate of lime was up to 78.77%.

NOMENCLATURES

• C  - the molar concentration of the reactant or product (mol/L)
• C _CaO  - relative intensity of the major peak of CaO in the XRD spectra of used oxygen carrier
• C _CaSO4   - relative intensity of the major peak of CaSO₄ in the XRD spectra of used oxygen carrier
• K _p  - equilibrium constants of reactions
• m   - mass of oxygen carrier (kg)
• m _ox   - mass of the fully oxidized oxygen carrier (kg)
• m _red   - mass of the fully reduced oxygen carrier (kg)
• M _o   - atomic weight of oxygen (kg/mol)
• n _out   - molar flow of the gas exiting the reactor after condensation (mol/s)
• C ⁰ _out   - molar flow of the gas exiting the reactor without lime after condensation (mol/s)
• n _CaSO4   - molar of sulfur-containing compound entering the reactor (mol)
• P  - the pressure of the reactant gases or product gases (Pa)
• R  - constant of the ideal gases, 8.314 J mol^-1K^-1
• R _o   - oxygen ratio of the oxygen carrier, defined as R_o = (m_ox - m_red) / m_ox
• T  - temperature(°C)
• t   - time (s)
• dX/dt   - the reaction rate of oxygen carrier (1/s)
• X _o   - oxygen carrier conversion during the oxidation
• X _r   - oxygen carrier conversion during the reduction
• y _{i, out}   - outlet mole fraction of the gas species i exiting the reactor after condensation
• y ⁰ _i,out   - outlet mole fraction of the gas species i exiting the reactor without lime after condensation
• Greek lettersΔH⁰_298K  - enthalpy change of the reaction under room temperature and atmospheric pressure (kJ/mol)
• ΔG  - standard Gibbs free energy of reactions (kJ/mol)
• χ _s   - sulfur release rate of oxygen carrier with lime (%)
• χ ⁰ _s   - sulfur release rate of oxygen carrier without lime (%)
• η _s   - desulphurization rate of lime (%)

ACKNOWLEDGEMENTS

This work was supported by the National Natural Science Foundation of China (No.51406110), the Foundation for Outstanding Young Scientist in Shandong Province (No.BS2014NJ013) and the Youth Science Fund Project of Shandong Academy of Sciences (No. 2014QN014). The authors were also grateful to the analytical and testing centers of Shandong Province and Huazhong University of Science and Technology for XRD, FESEM-EDS and BET measurements.

REFERENCES

• Abad, A., Mattisson, T., Lyngfelt, A., and Rydén, M., Chemical-looping combustion in a 300 W continuously operating reactor system using a manganese-based oxygen carrier, Fuel, 85, 1174-1185 (2006).
• Adánez, J., Dueso C., de Diego, L. F., García-Labiano, F., Gayán P. and Abad, A., Methane combustion in a 500 Wth chemical-looping combustion system using an impregnated Ni-based oxygen carrier, Energy & Fuels, 23, 130-142 (2009).
• Andrus, H. E., Chiu, J. H., Thibeault, P. R. and Brautsch, A., Clean Coal and Fuel Systems, 34th International Technical Conference. Clearwater, Florida (2009).
• Anthony, E. J. and Granatstein, D. L., Sulfation phenomena in fluidized bed combustion systems, Progress in Energy and Combustion Science, 27, 215-236 (2001).
• Bragança, S. R., Jablonski, A. and Castellan, J. L., Desulfurization kinetics of coal combustion gases, Brazilian Journal of Chemical Engineering, 20, 161-169 (2003).
• Bragança, S. R. and Castellan, J. L., FBC desulfurization process using coal with low sulfur content, high oxidizing conditions and metamorphic limestones, Brazilian Journal of Chemical Engineering , 26, 375-383 (2009).
• Cao, Y. and Pan, W. P., Investigation of chemical looping combustion by solid fuels. 1. Process analysis, Energy & Fuels, 20, 1836-1844 (2006).
• Cuadrat, A., Linderholm, C., Abad, A., Lyngfelt, A. and Adánez, J., Influence of limestone addition in a 10 kWth chemical-looping combustion unit operated with petcoke, Energy & Fuels, 25, 4818-4828 (2011).
• de Diego, L. F., Ortiz, M., Adánez, J., García-Labiano, F., Abad, A. and Gayán, P., Synthesis gas generation by chemical-looping reforming in a batch fluidized bed reactor using Ni-based oxygen carriers，Chemical Engineering Journal, 144, 289-298 (2008).
• Ding, N., Zhang C. W., Luo C., Zheng Y. and Liu Z. G., Effect of hematite addition to CaSO₄ oxygen carrier in chemical looping combustion of coal char, RSC Advances, 5, 56362-56376 (2015).
• Ding, N., Wang, W. R., Zheng, Y., Luo, C., Fu, P. F. and Zheng, C. G., Development and testing of an interconnected fluidized-bed system for chemical looping combustion, Chemical Engineering & Technology, 35, 532-538 (2012).
• Ding, N., Zheng, Y., Luo, C., Wu, Q. L., Fu, P. F. and Zheng, C. G., Development and performance of binder-supported CaSO₄ oxygen carriers for chemical looping combustion, Chemical Engineering Journal , 171, 1018-1026 (2011).
• Doğu, T., The importance of pore structure and diffusion in the kinetics of gas-solid non-catalytic reactions: Reaction of calcined limestone with SO₂, Chemical Engineering Journal , 21, 213-222 (1981).
• Hossain, M. M. and de Lasa, H. I., Chemical-looping combustion (CLC) for inherent CO₂ separations-a review, Chemical Engineering Science, 63, 4433-4451 (2008).
• Johansson, M., Mattisson, T. and Lyngfelt, A., Creating a synergy effect by using mixed oxides of iron- and nickel oxides in the combustion of methane in a chemical-looping combustion reactor, Energy & Fuels, 20, 2399-2407 (2006).
• Liu, H., Zailani, R. and Gibbs, B. M., Comparisons of pulverized coal combustion in air and in mixtures of O₂ /CO₂, Fuel , 84, 833-840 (2005).
• Lyon, R. K. and Cole, J. A., Unmixed combustion: An alternative to fire, Combustion and Flame, 121, 249-261 (2000).
• Mattisson, T. and Lyngfelt, A., The reaction between limestone and SO₂ under periodically changing oxidizing and reducing conditions-effect of temperature and limestone type, Thermochimica Acta, 325, 59-67 (1999).
• Mattisson, T. and Lyngfelt, A., Reaction between sulfur dioxide and limestone under periodically changing oxidizing and reducing conditions-effect of cycle time, Energy & Fuels, 12, 905-912(1998).
• Notz, R., Asprion, N., Clausen, I. and Hasse, H. Selection and pilot plant tests of new absorbents for post-combustion carbon dioxide capture. Chemical Engineering Research & Design, 85, 510-515 (2007).
• Saastamoinen, J. J. and Shimizu, T., Attrition-enhanced sulfur capture by limestone particles in fluidized beds, Industrial & Engineering Chemistry Research, 46, 1079-1090 (2007).
• Scott, S. A., Dennis, J. S., Hayhurst, A. N. and Brown, T., In situ gasification of a solid fuel andCO₂ separation using chemical looping, AIChE Journal, 52, 3325-3328 (2006).
• Sebag, M. G., Bragança S. R., Normann. M. and Jablonski, A., The behavior of heavy metals in the process of desulfurization of Brazilian coal combustion gases by the addition of limestone, Brazilian Journal of Chemical Engineering , 18, 139-147 (2001).
• Shen, L. H., Zheng, M., Xiao, J. and Xiao, R., A mechanistic investigation of a calcium-based oxygen carrier for chemical looping combustion, Combustion & Flame, 154, 489-506 (2008).
• Song, Q. L., Xiao, R., Deng, Z. Y., Shen, L. H., Xiao, J. and Zhang, M. Y., Effect of temperature on reduction of CaSO₄ oxygen carrier in chemical-looping combustion of simulated coal gas in a fluidized bed reactor, Industrial & Engineering Chemistry Research , 47, 8148-8159 (2008a).
• Song, Q. L., Xiao, R., Deng, Z. Y., Zheng, W. G., Shen, L. H. and Xiao, J., Multicycle study on chemical-looping combustion of simulated coal gas with a CaSO₄ oxygen carrier in a fluidized bed reactor, Energy & Fuels, 22, 3661-3672 (2008b).
• Stern, M. C. and Hatton, T. A., Bench-scale demonstration of CO₂ capture with electrochemically-mediated amine regeneration, RSC Advances , 4, 5906-5914 (2014).
• Su, M., Ma, J., Tian, X., Zhao, H., Reduction Kinetics of Hematite as Oxygen Carrier in Chemical Looping Combustion, Fuel Processing Technology, DOI: 10.1016/j.fuproc.2016.05.015 (2016).
  » https://doi.org/10.1016/j.fuproc.2016.05.015
• Tian, H. J., Guo, Q. J., Yue, X. H. and Liu, Y. Z., Investigation into sulfur release in reductive decomposition of calcium sulfate oxygen carrier by hydrogen and carbon monoxide, Fuel Processing Technology, 91, 1640-1649 (2010).
• Teyssié, G., Leion, H., Schwebel, G. L., Lyngfelt, A. and Mattisson, T., Influence of lime addition to ilmenite in chemical-looping combustion with solid fuels, Energy & Fuels, 25, 3843-3853 (2011).
• Xiao, R. and Song, Q. L., Characterization and kinetics of reduction of CaSO₄ with carbon monoxide for chemical-looping combustion, Combustion & Flame , 158, 2524-2539 (2011).
• Yrjas, K. P., Zevenhoven, C. A. P. and Hupa, M. M., Hydrogen sulfide capture by limestone and dolomite at elevated pressure. 1. Sorbent performance, Industrial & Engineering Chemistry Research , 35, 176-183 (1996).
• Zheng, M., Shen, L. H. and Xiao, J., Reduction of CaSO₄ oxygen carrier with coal in chemical-looping combustion: Effects of temperature and gasification intermediate, International Journal of Greenhouse Gas Control, 4, 716-728 (2010).
• Zheng, M., Shen, L. H., Feng, X.Q. and Xiao, J., Kinetic model for parallel reactions of CaSO₄ with CO in chemical-looping combustion, Industrial & Engineering Chemistry Research , 50, 5414-5427 (2011).

Publication Dates

• Publication in this collection
  Jan 2018

History

• Received
  05 May 2016
• Reviewed
  19 Sept 2016
• Accepted
  01 Nov 2016