Kinetics Investigation of Copper Ore Oxygen Carrier for Chemical Looping Combustion

Abstract

Chemical looping combustion (CLC) has been validated as one of the most promising technologies for fossil fuel combustion, which can produce high-purity CO₂ streams ready for capture and sequestration in power production. The selection of an appropriate oxygen carrier is one of the most important issues for the CLC process, and the reduction kinetics investigation of the oxygen carrier with fuel gas can provide the basis for CLC reactor design and simulation optimization. In this study, copper ore was chosen as an oxygen carrier, and the oxygen release property of copper ore under a nitrogen environment at various temperatures (1073–1193 K) was first investigated in a thermogravimetric analyzer (TGA). Subsequently, the reduction kinetics of copper ore with CO and H₂ were evaluated by the TGA at temperatures ranging from 773 K to 1073 K, using a continuous stream of 5, 10, 15, 20, 25, and 30 vol. % of CO or H₂ balanced by CO₂ or N₂. It was found that the reaction rates of these reactions accelerated with the increase in temperature and fuel gas concentration in inlet gas. Both the oxygen release process of copper ore and the reduction process of copper ore with reducing gases can be described by the unreacted shrinking core model (USCM). The reaction mechanism function for the oxygen-releasing and reduction process of copper ore oxygen carrier was varied. The activation energy of the oxygen-releasing process, reduction process with CO, and reduction process with H₂ were calculated as 99.35, 5.08, and 4.28 kJ/mol, respectively. The pre-exponential factor ranged from 1.96 × 10⁻¹ to 1.84 × 10³. The reaction order depended on the fuel gas, which was 1 and 0.86, respectively, for reaction with CO and H₂.

1. Introduction

The anthropogenic CO₂ emission has greatly increased over the past few decades due to the highly dependent of power and energy generation from fossil fuel utilization. The greenhouse effect caused (potentially) by massive CO₂ emissions is a growing global threat that has already brought great challenges to the living environment of human beings [1,2]. Under the context of carbon neutrality, extensive efforts have been made to reduce the anthropogenic CO₂ emission. The carbon capture, utilization, and sequestration (CCUS) technology is one of the most promising and efficient ways for CO₂ emission reduction [3,4]. Currently, most carbon capture technologies require multiple steps, including CO₂ adsorption and desorption, purification, and compression. These steps are highly energy-intensive and ultimately result in low energy efficiency of the whole CCUS process [5]. Chemical looping combustion (CLC) [6] has been proposed as a potential candidate for the capture of CO₂ with low economic costs. Based on the objective of generating a high-purity CO₂ stream, the CLC concept was first put forward by Lewis and Gilliland in 1954 [7]. Decades later, in the 1980s, Richter and Knoche [8] reviewed this concept as an effective process for power production. Nowadays, CLC has been widely recognized with the merits of inherent CO₂ separation, low energy penalty for CO₂ capture, and effective restraining of thermal NO_x formation [9,10,11].

In CLC, the traditional combustion process was completed by the cyclic reduction and oxidation of oxygen carriers in two separate reactors: the reduction in the fuel reactor (FR) and the oxidation in the air reactor (AR) [12]. Thus, to address the CLC process, an interconnected but atmosphere-isolated fluidized bed reactor is required. To be more specific, in the FR, the fuel reacted with the lattice oxygen provided by the solid oxygen carrier (MeO) instead of being oxidized directly by air, while in the AR, the reduced oxygen carrier (Me) is then generated by absorbing oxygen from the air and be ready for the next cycle. The oxygen carrier particles circulate between the AR and FR by the driving force of carrier gas to provide continuous lattice oxygen for fuel combustion. Theoretically, the outlet gas of FR contains CO₂ and H₂O only, and pure CO₂ can be obtained by a simple condensing process [13]. As the one-step traditional combustion process is split into two sub-steps in CLC, cascade utilization of energy and increased combustion efficiency can be ultimately achieved [14].

The selection of appropriate oxygen carriers is one of the key factors to successfully realize the CLC process. It is generally a kind of transition metal oxide, perovskite, natural ore, etc. The suitable oxygen carrier should have the following characteristics: enough oxygen-carrying capacity, stable cyclic redox reactivity, anti-sintering property at the temperature window of CLC, economically inexpensive and no secondary pollution [15]. Inert supports such as Al₂O₃, ZrO₂, TiO₂, CuAl₂O₄, and MgAl₂O₄ are generally used to support the active phase of the oxygen carrier to improve its redox reactivity as well as its mechanical strength. So far, the Cu-, Fe-, Ni-, Co-, and Mn-based oxygen carriers have received great attention, and extensive investigations have been conducted to evaluate the properties of the oxygen carriers [16,17,18,19]. Synthetic oxygen carriers generally show relatively high reactivity, whereas the expensive preparation process is one of the most troublesome problems to overcome. Copper ore, known for its low cost, rich reserves, and moderate reactivity, has been investigated as a suitable oxygen carrier for CLC [20,21,22,23,24].

Determination of the reaction mechanism and kinetic parameters of the oxygen carrier with fuel is of vital significance for the design and operation optimization of the CLC system [25,26,27,28,29]. Yu et al. [30] used a nucleation and nuclei growth model to describe the oxygen-releasing process of Cu-based oxygen carriers. Different inert support could make a difference in the distributed activation energy, which were 155.0 ± 2.2 kJ/mol, 152.9 ± 6.9 kJ/mol, and 144.9 ± 6.6 kJ/mol for CuO/TiO₂, CuO/ZrO₂, and CuO/SiO₂, respectively. However, the mechanism function of CuO reduction reaction was not affected by binders, and it was shown as $f(\alpha )=3(1-\alpha )[-\ln {(1-\alpha )}^{2/3}]$. García-Labiano et al. [31] investigated the reduction reaction kinetics of impregnated CuO/Al₂O₃ oxygen carrier with CO and H₂, using the unreacted shrinking core model (USCM) for a plate-like geometry. The activation energy for CO and H₂ were 14 and 33 kJ/mol, and reaction orders were 0.8 and 0.6 for CO and H₂, respectively. Reduction properties of binary mixtures of Cu-, Fe-, and Ni-based oxygen carriers under the reducing environment of H₂, CO, and CH₄ were studied by B. Moghtaderi et al. [32]. The activation energy, reaction order, and pre-exponential factor were extracted from TGA experimental measurements using the shrinking core model. As concluded from the experimental observations, the conversion time (time corresponding to full conversion) for any given mixture was the geometric mean value of the conversion time of the parent materials. The CSIC group [33] studied the reduction kinetics of Cu-, Ni-, and Fe-based with syngas (CO + H₂), using the grain model with spherical or plate-like geometry in the grain. The key kinetic parameters, such as the activation energy and the reaction order were calculated, with which a preliminary estimation of the solids inventory in the fuel reactor of a CLC system was conducted. Chuang et al. [34] pioneered the idea of using a batch-scale fluidized bed reactor to investigate the reaction kinetics of a co-precipitated mixture of CuO and Al₂O₃ with CO at the temperature range of 523–1173 K. A rapid-response mass spectrometer and an infra-red analyzer were applied to minimize sampling problems. The activation energies and pre-exponential factors for both reactions of CuO and Cu₂O with CO were measured, respectively. As a summary,

Table 1. A summary of the kinetics works over Cu-based oxygen carriers in chemical looping processes.

Oxygen Carrier	Testing Condition	Kinetics Parameters	Ref.
CuO/Al2O3	523–1173 K 1.1–9.77% CO	DRM n = 1.0, Ea = 28–52 kJ/mol	[34]
CuO/Al2O3	523–1173 K 1.1–9.77% H2	DRM n = 1.0, Ea = 58 kJ/mol	[35]
CuO/Al2O3	723–1073 K 5–70% CH4 5–70% H2 5–70% CO	SCM n = 0.4, Ea = 60 kJ/mol n = 0.6, Ea = 33 kJ/mol n = 0.8, Ea = 15 kJ/mol	[31]
CuO/Al2O3	873–1073 K 5–70% CH4 5–70% CO 5–70% H2	SCM n = 0.5, Ea = 106 kJ/mol n = 0.8, Ea = 11 kJ/mol n = 0.5, Ea = 20 kJ/mol	[36]
CuO	473–773 K 1.6–5% CO	DRM n = 0.7, Ea = 20–25 kJ/mol	[37]
CuO/Al2O3	773–1073 K 20–70% H2 20–70% CO	SCM n = 0.55, Ea = 30 kJ/mol n = 0.8, Ea = 16 kJ/mol	[32]
CuO/MgAl2O4	0–9% O2, 1148–1273 K 2.5–21% O2, 1123–1273 K	NNGM, n = 0.5, Ea = 270 kJ/mol L-H, n = 1.2, Ea = 32 kJ/mol	[38]
CuO/ZrO2	0–3.4% O2, 1048–1198 K	First order, n = 1.0, Ea = 58 kJ/mol	[39]
CuO/TiO2	0–3.4% O2, 1073–1173 K	First order, n = 1.0, Ea = 67 kJ/mol	[39]
CuO/SiO2	973–1173 K 0–0.875% O2	SCM Ea = 249 kJ/mol	[40]
CuO/Al2O3	573–1023 K 1.22–7.5% O2	DRM n = 1.0, Ea = 40–60 kJ/mol	[41]
CuO/Al2O3	773–1073 K 5–21% O2	SCM n = 1.0, Ea = 15 kJ/mol	[31]
CuO@TiO2-Al2O3	0–1.0% O2, 1083–1163 K 5.2–21% O2, 793–873 K 5–35% H2, 498–598 K 5–35% CO, 573–673 K 5–20% CH4, 948–1048 K	NNGM, n = 0.5, Ea = 217.2 kJ/mol CRM, n = 0.2, Ea = 87.5 kJ/mol SCM, n = 0.8, Ea = 44.5 kJ/mol SCM, n = 1.0, Ea = 40.1 kJ/mol NNGM, n = 0.6, Ea = 112.2 kJ/mol	[25]

Notes: DRM = diffusion reaction model; SCM = shrinking core model; NNGM = nucleation and nuclei growth model; L-H = Langmuir–Hinshelwood mechanistic model; CRM = chemical reaction model.

presents the relevant kinetics works of Cu-based oxygen carriers under the context of chemical looping in the literature.

To date, there have been relatively few kinetics studies of copper ore oxygen carriers in the literature. Since copper ore is expected to be a promising oxygen carrier candidate in the CLC process, this study aims to determine its reaction kinetics using TGA. The isothermal kinetic method was employed to determine the reaction mechanism of copper ore oxygen carrier, and the kinetic parameters, such as the activation energy and the pre-exponential factor of the oxygen release process, reduction with CO, and reduction with H₂, were first measured, respectively.

2. Experimental

2.1. Materials

The raw material of the Cu-based oxygen carrier used in this study was a kind of copper ore from Zhongtiaoshan of China, which contained a moderate level of Cu component after flotation. The preparation process of the copper ore oxygen carrier followed the following 3 steps: (1) the ore, after preliminary breaking, was dried at 378 K for 24 h in the drying cabinet and then pre-screened to the size range of 0.1–0.3 mm; (2) the particles within the size range of 0.1–0.3 mm were first calcined at 773 K for 5 h and then 1273 K for 10 h in an air-atmosphere muffle oven for the purpose of eliminating the sulfur content in the ore as well as enhancing the mechanical strength; (3) the copper ore particles after calcination were screened and sieved to 0.125–0.18 mm for the TGA experiment. According to the X-ray diffraction (XRD) and X-ray fluorescence (XRF) spectrometry analysis, the main active components of the calcined copper ore were CuO and CuO-Fe₂O₃ (CuFe₂O₄), as shown in

Table 2. Chemical content and ultimate analysis of the calcined copper ore.

Chemical Content (wt. %)					Ultimate Analysis (wt. %)
CuO	CuFe2O4	SiO2	CaSO4	Al2O3	Cu	Fe	Si	Ca	S	Al
21.04	70.05	5.53	2.29	1.08	48.88	45.00	7.07	1.31	1.47	1.63

.

2.2. Conversion Determination

This study only considered the oxygen-releasing mechanism and the reduction kinetics of copper ore with CO and H₂. The conversion X of the oxygen-releasing process and the reduction process of the oxygen carrier can be calculated as,

$$X_i=\frac{m_{\mathrm{ox}}-m_i}{m_{\mathrm{ox}}-m_{\mathrm{re},i}}$$

(1)

where m_ox and m_re is the weight of the sample at the fully oxidized and reduced state, respectively; m is the weight of the oxygen carrier at time t; i represents the oxygen-releasing process or the reduction process with fuel gas. To be more specific, for the oxygen-releasing process, m_re,i equals the sample weight when the CuO content in the ore was completely decomposed into Cu₂O, while for the reduction process with fuel gas, m_re,i equals the sample weight when CuO and Fe₂O₃ in the oxygen carrier were fully reduced to Cu and Fe₃O₄, respectively.

2.3. Pre-Experiment in TGA

To minimize the potential influence of internal and external diffusion on the experimental result, preliminary experiments were first carried out to obtain a suitable mass sample and gas flow rate. The sample weight was varied in the range of 7–15 mg, and the gas flow rate in the range of 40–90 mL/min, while the reaction temperature was always maintained at 1123 K. Figure 1a shows the oxygen-releasing conversion ratio of copper ore under different sample weight in N₂ atmosphere (with the N₂ flow rate maintained at 80 mL/min). As can be seen, when the sample mass ≤ 9 mg, the global reaction rate inside the particles was not controlled by gas diffusion. Figure 1b shows the conversion ratio under different N₂ flow rates (with the sample mass maintained at 9 mg). The conversion rate with a gas flow rate ≥ 80 mL/min was almost the same. Thus, it can be concluded that the reaction rate was not controlled by internal and external diffusion when the sample mass was ≤9 mg and the gas flow rate was ≥80 mL/min.

2.4. Formal Experiment in TGA

For the determination of the oxygen-releasing kinetics, isothermal decomposition experiments of copper ore at various temperatures (1073 K, 1113 K, 1153 K, and 1193 K) were conducted, while for the reduction reaction of copper ore with CO and H₂, 4 relatively low temperatures (773 K, 873 K, 973 K, and 1073 K) and 6 different reducing gas concentrations (5 vol. %, 10 vol. %, 15 vol. %, 20 vol. %, 25 vol. %, and 30 vol. %) were considered. For each test, the sample weight was always chosen at nearly 9 mg, and the gas flow rate was always maintained at 80 mL/min. To be noted, the different reducing gas composition was realized by mixing with different volumes of inert carrier gas: when CO was used as the fuel gas, CO₂ was chosen as the inert gas so as to avoid the influence of carbon decomposition on the experimental result; when H₂ was used as the reducing gas, N₂ was introduced as the inert carrier gas (due to the limitation of the TGA setup, it was not allowed to introduce steam as a carrier gas). Moreover, for the reduction experiments of copper ore with CO or H₂, an N₂ sweep step is a must before and after the introduction of fuel gas into the TGA; when the temperature was higher than 1073 K, the copper ore would release oxygen in N₂ atmosphere to some extent. Considering that, the highest temperature for the reduction experiments of copper ore with fuel gas was 1073 K.

2.5. Isothermal Kinetic Analysis Method

The function used to describe the gas-solid reaction rate can be expressed in the differential form and the integral form, as illustrated in Equations (1) and (2), respectively [42].

$$\frac{dX}{dt}=A\exp (-\frac{E}{RT})f(X)C^n$$

(2)

$$G(X)={\displaystyle \int \frac{1}{f(X)}}dX=A\exp (-\frac{E}{RT})C^{n}t$$

(3)

where X is the conversion ratio of the oxygen carrier; t is the time during the reaction process; A is the pre-exponential factor; E is the distributed activation energy; R is the gas constant (R = 8.314 J/(mol K)); C is the fuel gas concentration; n is the reaction order; f(X) and G(X) represent the reaction mechanism function in the differential and integral form, respectively.

As can be seen from Equation (3), when the reaction temperature and the fuel gas concentration were both fixed, the value of $A\exp (-\frac{E}{RT})C^n$ was a constant. Thus, the reaction mechanism function G(X) that could make $G(X)=const\times t$ would be selected. In other words, there could be a mechanism function that made G(X), and t conform to the linear relationship, and the one that exhibited the best linear correction would be the most possible reaction mechanism function. For each experiment with the constant temperature and constant fuel gas concentration, a slope of a linear value could be attained, noted as k_i, and i represented the number of the experiment.

$$ki=A\exp (-\frac{E}{RT})C^n$$

(4)

When taking the logarithm of Equation (4) on both sides, it can be rewritten as

$$\ln ki=\ln A\exp (-\frac{E}{RT})C_{}^n=n\ln C_{}+\ln A\exp (-\frac{E}{RT})$$

(5)

As can be seen in Equation (5), when the reaction temperature T was fixed, lnk_i and lnC showed a linear relationship, in which the slope of the line was the reaction order and the Y-intercept of the line was $\ln A\exp (-\frac{E}{RT})$. For the TGA experiments of copper ore with CO or H₂, there were 6 experiments (6 different fuel gas concentrations) under each reaction temperature. Thus, a total of 4 groups (4 different reaction temperatures) of slope n and Y-intercept $\ln A\exp (-\frac{E}{RT})$ under each reaction temperature T could be obtained by linear fitting. And for the Y-intercept, it can be rewritten like this:

$$b=\ln A\exp (-\frac{E}{RT})=-\frac{E}{RT}+\ln A$$

(6)

By linear fitting, a linear relation between b and $\frac{1}{T}$ can be acquired, of which $-\frac{E}{R}$ was the slope and lnA was the Y-intercept of the line. Now that all the parameters in Equation (2) can be determined, it should be noted that, for the oxygen-releasing process of copper ore, the determination of the kinetic mechanism was almost the same as above, except that there was no fuel gas concentration $C^n$ for the oxygen-releasing process. Therefore, the determination of the n value was eliminated, while the calculation of activation E and pre-exponential factor A were as above.

Several commonly used kinetic mechanism functions of the solid-gas reaction are listed in

Table 3. Common kinetic mechanism functions of the gas–solid reaction.

Reaction Model	Symbol	f(X)	G(X)
nucleation and nuclei growth (n = 1)	A1(X)	$1-X$	$-\ln (1-X)$
nucleation and nuclei growth (n = 2)	A2(X)	$2(1-X){(-\ln (1-X))}^{1/2}$	${(-\ln (1-X))}^{1/2}$
nucleation and nuclei growth (n = 3)	A3(X)	$3(1-X){(-\ln (1-X))}^{2/3}$	${(-\ln (1-X))}^{1/3}$
2D-diffusion model	D2(X)	${(-\ln (1-X))}^{-1}$	$(1-X)\ln (1-X)+X$
3D-diffusion model (Jander)	D3(X)	$(3/2){(1-X)}^{2/3}(1-{(1-X)}^{1/3})$	${(1-{(1-X)}^{1/3})}^2$
3D-diffusion model (Grinstling)	D4(X)	$(3/2)({(1-X)}^{-1/3}-1)$	$(1-2X/3)-{(1-X)}^{2/3}$
zero order contraction model	R1(X)	$1$	$X$
2D-contraction model	R2(X)	$2{(1-X)}^{1/2}$	$1-{(1-X)}^{1/2}$
3D-contraction model	R3(X)	$3{(1-X)}^{2/3}$	$1-{(1-X)}^{1/3}$

. For the determination of the reaction kinetic mechanism of a certain reaction process, it was necessary to apply all the possible G(X) into the Equation $G(X)=const\times t$: the G(X) that exhibited the best linear correction with t could be selected as the actual kinetic mechanism function.

3. Results and Discussion

3.1. Kinetic Determination for the Oxygen Releasing Process

Isothermal experiments of the oxygen-releasing process of copper ore under different temperatures were first conducted in TGA, and the curve of the conversion ratio vs. time was illustrated in Figure 2a. As seen, the oxygen-releasing rate was highly dependent on the decomposition temperature, which generally increased with the temperature increasing. Moreover, for the 1193 K case, the oxygen-releasing process was accomplished in 10 min, while for the 1073 K case, the copper ore could not release its active oxygen completely within the setup time range (30 min) of the experiment.

Based on the TGA experimental data of the oxygen-releasing process of copper ore under different temperatures, together with the isothermal kinetic analysis method previously demonstrated, the kinetic mechanism of the oxygen-releasing process of copper ore was determined. After multiple linear fitting attempts, it was found that $G(X)=1-{(1-X)}^{1/3}$ exhibited the best linear relationship with t, which corresponded to the three-dimensional (3D)-contraction model. Therefore, $G(X)=1-{(1-X)}^{1/3}$ was chosen as the kinetic mechanism function for the oxygen-releasing process. By analyzing the data obtained from experiments under four different temperatures, a fitted straight line was acquired and shown in Figure 2b, of which the slope $k=-\frac{E}{R}=-11950$, and the Y-intercept $b=\ln A=7.52$. Finally, the activation energy and the pre-exponential factor can be determined as $E=99.35\mathrm{kJ}/\mathrm{mol}$ and $A=1.84\times {10}^3$, respectively.

3.2. Kinetic Determination for the Reduction of CO

For the determination of the kinetic mechanism of copper ore with CO, experiments of different temperatures (773 K, 873 K, 973 K, and 1073 K) and various CO concentrations (5 vol. %, 10 vol. %, 15 vol. %, 20 vol. %, 25 vol. %, and 30 vol. %) have been investigated in the TGA. The curve of the conversion ratio vs. time for each case is shown in Figure 3. To be noted, when calculating the conversion ratio, the complete reduction of copper ore into Cu and Fe₃O₄ was determined to be the benchmark reaction. As can be seen, both the reaction temperature and CO concentration would affect the reduction conversion rate of copper ore, which, with the increase in temperature and CO concentration, the conversion rate would also increase to some extent.

The experimental results obtained from TGA were analyzed by linear fitting using the same isothermal kinetic analysis method mentioned above. The kinetic mechanism function $G(X)=X$ was found to show the best linear relation with t, which corresponded to the zero-order contraction model. Within the reaction process, the solid grain size of the copper ore oxygen carrier remained unchanged while the reaction interface migrated to the center of the core. The fuel gas went through the product layer and reacted with the unreacted solid grain at the interface of the core. Reactions did not take place within the core, but the radius of the core shrank gradually until reactions were completed. With the selected kinetic mechanism function $G(X)=X$, experimental data under four different temperatures and six different CO concentrations was used to linearly fit four lines, as shown in Figure 4a. The average slope value of the lines was the reaction order of the reduction of copper ore with CO, which $n=1$. The four lines in Figure 4a would cut four intercepts in the Y-axis. Using the four Y-intercepts, together with the corresponding $\frac{1}{T}$, another line could be linearly fitted, as shown in Figure 4b. The slope $k=-\frac{E}{R}=-609$, and the Y-intercept $b=\ln A=-1.62$. Thus, the activation energy could be determined as $E=5.08\mathrm{kJ}/\mathrm{mol}$ and the pre-exponential factor $A=1.96\times {10}^{-1}$.

3.3. Kinetic Determination for the Reduction with H₂

As refer to the experimental settings of the reduction with CO, TGA experiments of 4 different temperatures and six different H₂ concentrations were also conducted for the reduction of copper ore with H₂, a total of 24 cases, and the conversion ratio of the copper ore oxygen carrier under different experimental conditions were shown in Figure 5. As usual, when calculating the conversion ratio, the complete reduction of copper ore into Cu and Fe₃O₄ was determined to be the benchmark reaction. As seen from the figures, the conversion ratio of copper ore with H₂ under different temperatures and different H₂ concentrations generally showed a linear relationship with t, except for the unstable stage at the end of the reaction. The possible G(X) was selected to examine its linear relation with t, and it was found that $G(X)=X$ exhibited the best linear relation with t. Thus, $G(X)=X$ was determined as the kinetic mechanism function for the reduction of copper ore with H₂. To be noted, for the experimental condition of the same reaction temperature and the same fuel gas concentration, a shorter time was needed for the full conversion of copper ore with H₂ than with CO, which means that H₂ shows a higher reactivity with copper ore than CO.

The determination for the kinetic parameters of copper ore with H₂ was similar to that of copper ore with CO, and the analyzing results are shown in Figure 6. As concluded from Figure 6a, the reaction order was obtained as $n=0.86$. And from Figure 6b, the slope $k=-\frac{E}{R}=-515$, and the Y-intercept $b=\ln A=-1.37$. Therefore, the activation energy E and pre-exponential A factor were determined as $E=4.28\mathrm{kJ}/\mathrm{mol}$ and $A=2.54\times {10}^{-1}$, respectively.

The oxygen-releasing and reduction kinetics parameters of the copper ore oxygen carrier attained in this study are summarized in

Table 4. Summary of the oxygen releasing and reduction kinetics parameters of the copper ore oxygen carrier in CLC.

	Kinetics Model Equation	n (Dimensionless)	E (kJ/mol)	A (atm−n·s−1)
N2	$G(X)=1-{(1-X)}^{1/3}$	-	99.35	1.84 × 103
CO	$G(X)=X$	1	5.08	1.96 × 10−1
H2	$G(X)=X$	0.86	4.82	2.54 × 10−1

. We note here that the activation energies determined for the reduction of copper ore with CO and H₂ were very low (smaller than 10 kJ/mol) in this work. This might suggest that the kinetics tests were somehow affected by gas diffusion. In fact, considering the relatively high reduction temperatures (773 K to 1073 K) adopted in this work, gas diffusion can hardly be eliminated. Nevertheless, the reason for conducting the kinetics test within the high-temperature range was to keep the testing condition as close as possible to the real CLC process. With these regards, it might be more appropriate to conduct the reduction kinetics tests of copper ore with CO/H₂ under fluidized bed conditions in future studies, minimizing the gas diffusion effect as much as possible.

4. Conclusions

In this work, the oxygen-releasing and reduction (with CO or H₂) kinetics of copper ore oxygen carriers were investigated in a TGA. For the oxygen-releasing process, the kinetics tests were conducted within the temperature range of 1073–1173 K in pure N₂. With respect to the reduction kinetics investigation, the tests were carried out at temperatures of 773–1073 K, using either 5–30% H₂/N₂ or 5–30% CO/N₂ as the reducing atmosphere. After several model fitting attempts, the oxygen-releasing process of the copper ore was found to be well described by the 3D-contraction model, i.e., $G(X)=1-{(1-X)}^{1/3}$. Eventually, the activation energy and the pre-exponential factor for the oxygen-releasing process were determined as $E=99.35\mathrm{kJ}/\mathrm{mol}$ and $A=1.84\times {10}^3$, respectively. The reduction of the copper ore by either CO or H₂ under the tested conditions in this work can both be predicted by the zero-order contraction model, i.e., $G(X)=X$. The activation energy for the reduction of copper ore with CO and H₂ were calculated as 5.08 kJ/mol and 4.28 kJ/mol, respectively, and the corresponding pre-exponential factor were 1.96 × 10⁻¹ and 2.54 × 10⁻¹, respectively. In addition, the reaction order for CO and H₂ were 1 and 0.86, respectively. The kinetics parameters of the copper ore attained from TGA are expected to be useful for the CLC reactor design and reaction process modeling. Nevertheless, the low activation energy (<10 kJ/mol) attained for the reduction of copper ore with CO or H₂ suggested a potential gas diffusion effect in the kinetics tests. This can be mainly attributed to the insufficient gas-solid contact in the TGA apparatus and the relatively high reduction temperature adopted. Therefore, conducting the reduction kinetics tests of copper ore with CO/H₂ under fluidized bed conditions and lower temperatures would be recommended in future studies.