Solid-solid reaction of CuFe2O4 with C in chemical looping system: A comprehensive study

Solid-solid reaction of CuFe2O4 with C in chemical looping system: A comprehensive study

Fuel 267 (2020) 117163 Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel Full Length Article Solid-sol...

2MB Sizes 0 Downloads 1 Views

Fuel 267 (2020) 117163

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Solid-solid reaction of CuFe2O4 with C in chemical looping system: A comprehensive study

T



Tianle Li, Qiao Wu, Wenju Wang , YuPeng Xiao, Chenlong Liu, Fufeng Yang School of Energy and Power Engineering, Nanjing University of Science and Technology, Nanjing 210094, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Solid-solid reaction Chemical looping CuFe2O4 Sol-gel DFT

Oxygen carriers are critical in chemical-looping system to transfer oxygen and heat. Spinel CuFe2O4 is a promising oxygen carrier due to its low decomposition temperature, good resistance to sintering and oxygen deficiency. In this study, a comprehensive investigation of solid-solid reaction of CuFe2O4 with C was conducted on a fixed-bed reactor. The XRD, Raman and SEM proved the primary structure of CuFe2O4 with uniform external morphology in size and shape. The weight loss of CuFe2O4 in TG showed that it could release O2 at about 700 °C. The XRD and Raman results of solid products for various time intervals showed that CuFe2O4 was mainly transformed into CuFeO2 and Fe3O4 with a rapid reaction rate at the initial stage, and then CuFeO2 was reduced to Cu and Fe3O4, Fe3O4 was slowly reduced to FeO eventually. The DFT results indicated that C was in favor of adsorbing on O-terminated surface, and CuFe2O4 transformed into CuFeO2 and Fe3O4 accompanied by the formation of COx.

1. Introduction Chemical looping system is a novel technology for the utilization of fuel, and it has gained much interest for its prominent merits. Chemical looping combustion (CLC) and chemical looping gasification (CLG) are the two most common chemical looping reactions, in which a solid fuel like coal or biomass is pyrolyzed to generate volatiles (gas) and char (solid), and then these products react with an oxygen carrier [1]. Fuel and air are never in direct contact in solid-solid reaction of CLC and CLG. Oxygen carriers circulate in the process to transfer heat and oxygen to the solid fuel [2–4]. Consequently, oxygen carriers are critical in chemical looping reaction to improve the redox and mechanical properties [5–7]. Different metal oxides have been proposed as possible candidates. Single OCs (oxygen carriers) were introduced at the beginning, such as Fe2O3-based [8,9], NiO-based [10,11] and CuO-based [12,13] ones. CuO has high oxygen mobility and high redox reactivity, but its melting point was low and it is prone to agglomerate [4,5,14]. NiO has high reactivity, but it has a thermodynamic restriction. Fe2O3 is cheap in cost, however, its oxygen transfer capacity and reactivity were relatively low [15]. Besides, the high thermo decomposition temperature is a common defect of these oxides [16]. Therefore, a single metal oxide is difficult to have a large-scale practical application. In addition, bimetallic OCs with a complementary synergistic effect have been



researched to avoid the disadvantages of single metal oxide [7,17–19]. Some perovskite-type oxides (such as LaFeO3 [20,21], CaMnO3 [22]) and spinel-type oxides (CuFe2O4 [23], NiFe2O4 [15], CaFe2O4 [24]) were used for chemical looping reaction. However, the oxygen capacity of the spinel-type oxides possibly is higher than that of the perovskites perovskite-type oxides, so different spinal oxides have been widely used in CLC, as well as in CLG. Compared with other spinel-type oxides, CuFe2O4 with cubic closepacked and tetragonal symmetry spinel structure is regarded as a promising OC [25]. CuFe2O4 has a better reactivity over its single reference oxides CuO and Fe2O3, it not only has the exothermic characteristics of CuO, but also has the lower cost and lower toxicity of Fe2O3 [14,26,27]. Futhermore, CuFe2O4 has low decomposition temperature, good resistance to sintering and oxygen deficiency [16,28,29]. It not only has the capacity to transfer the lattice oxygen directly, but also can decompose and emit O2 [14,27].Thus, CuFe2O4, as a competitive OC, is of great significance for chemical looping system. Gas-solid reactions of volatiles (gas) and OCs have been studied thoroughly in recent years [30–33]. However, few studies have been done in solid-solid reactions of char with OCs, and their mechanism needs to be further explored. In this work, CuFe2O4 was synthesized by using a novel sol-gel method, and fuel C was obtained from the sucrose pyrolysis. A comprehensive study of solid-solid reaction of CuFe2O4 with C was

Corresponding author. E-mail addresses: [email protected], [email protected] (W. Wang), [email protected] (F. Yang).

https://doi.org/10.1016/j.fuel.2020.117163 Received 10 November 2019; Received in revised form 30 December 2019; Accepted 20 January 2020 0016-2361/ © 2020 Elsevier Ltd. All rights reserved.

Fuel 267 (2020) 117163

T. Li, et al.

conducted on a fixed-bed reactor. The solid products are identified by X-ray diffraction (XRD) and Raman. The gaseous reaction products are analyzed by an online gas analyzer. Thermodynamic analysis and density functional theory (DFT) were further conducted to explore the reaction mechanism of CuFe2O4 with coal.

2. Experimental 2.1. Materials preparation The copper ferrite (CuFe2O4) oxygen carrier was synthesized using a sol − gel technique. 0.01 mol Cu(NO3)2·3H2O and 0.02 mol Fe (NO3)3·9H2O were dissolved together in 100 mL deionized water to get a mixed solution. 0.3 M citric acid solution was subsequently added into the mixed solution to form a transparent solution. The molar ratio of Cu (NO3)2·3H2O to Fe(NO3)3·9H2O to citric acid was 1:2:3. Then, the mixed solution was stirred in a water bath at 80 °C until a transparent and viscous gel was obtained. The formed gel was dried in an oven at 100 °C for 12 h to expel residual water. After that, the as-prepared CuFe2O4 was further calcined at 850 °C in air for 3 h with a heating rate of 10 °C/ min. In addition, sucrose (AR) which contains less ash was put into a Muffle furnace at 600 °C for 3 h first. Then the honeycombed sucrose coke was ground into power and heated at 1000 °C in N2 to remove the volatile content utterly.

Fig. 1. Log (K) of Eqs. (1)–(7) varied with temperature.

3. Results and discussion 3.1. Thermodynamic analysis Although there are lots of limitations in thermodynamic equilibrium analysis, and many kinetic constraints are not considered, such as the effects of catalysts and temperature gradients [34,35]. The equilibrium calculation is also be of great significance for us to understand the detail in the solid-solid reaction process. The possible reactions related to the chemical looping reaction of CuFe2O4 with C have been listed as below. It contained the thermal decomposition reaction of copper-iron spinel [Eqs. (1) and (2)], the reaction of C with CuFe2O4 and its pyrolysis products CuFeO2 [Eqs. (3)–(7)]. Thus, the thermal equilibrium constants K was as a function of the final reaction temperature in equilibrium [36]. The values of the equilibrium constants K were calculated from the Gibbs free energy and the formula was Eq. (8) [37]. The Log values of K which indicated the direction of chemical reactions at different temperature were shown in Fig. 1.

2.2. Fixed-bed test The reduction of the CuFe2O4 was performed on a fixed bed, and a gas analyzer was used to obtain the gas product concentration of H2, CO, CO2 and CH4. In each run, 0.5 g CuFe2O4 and 0.167 g sucrose coke were mixed together evenly (with a CuFe2O4/C molar ratio of 1/4). This ratio was chosen based on the equivalent molar quantities of available carbon in the char and the available transferable oxygen in the lattice of the metal ferrite for gasification. Prior to each test, a programmable tube furnace was heated to 850 °C, the tubular reactor loaded with the samples was then put into it. Samples were held at 850 °C for 2, 5, 15 or 30 min. After that, the samples with desired reaction time were rapidly cooled in room temperature under N2 flow. The spent catalysts were finally collected for Raman and XRD analysis.

4CuFe2 O4 = 4CuFeO2 + 2Fe2 O3 + O2 (g)

(1)

3CuFe2 O4 = 3CuFeO2 + Fe3 O4 + O2 (g)

(2)

2.3. Characterization

4CuFe2 O4 + C= 4CuFeO2 + 2Fe2 O3 + CO2 (g)

(3)

The crystal structure and type of phases were identified by means of X-ray powder diffraction (XRD) at room temperature using a Rigaku D/ max-2500 instrument with CuKα radiation (λ = 0.154 nm). The Data were collected over the 2θ range of 10–80° using a scanning rate of 0.02°/step. The X-ray was generated by 40 kV and 40 mA power settings. Raman measurements of samples were recorded on an iHR550 with a 532 nmexcitation source. In addition, metal oxide is likely to be oxidized when exposed to laser irradiation, so the laser power of the Raman scattering spectra have been adjusted as low as possible to avoid oxidation. The external morphology of the CuFe2O4 was observed by a scanning electron micrograph(SEM) whose model is (COXEM)EM-30 Plus. Thermogravimetric analysis (TGA) was performed on a NETZSCH STA409 C/PC equipment. The gas analyzer (Gasboard-3100) is manufactured by Hubei CubieRuiyi instrument Co. Ltd. In our work, the gas concentrations CO, CO2 and CH4 were measured by NDIR non-spectral infrared method, and H2 concentrations was measured by TCD thermal conductivity technology.

3CuFe2 O4 + C= 3CuFeO2 + Fe3 O4 + CO2 (g)

(4)

2CuFe2 O4 + C= 2CuFeO2 + Fe2 O3 + CO(g)

(5)

3CuFeO2 + 2C = 3Cu + Fe3 O4 + 2CO (g)

(6)

3CuFeO2 + C= 3Cu + Fe3 O4 + CO2 (g)

(7)

K = −exp

Δr G∅ RT

(8)

From the picture we found that the thermodynamic Log values of Eqs. (1) and (2) were less than zero at 850 °C, but it maintained a slight upward trend and approached to zero. It meant that the thermal decomposition of copper-iron spinel was carried out reversely in thermodynamic equilibrium. However, it should be noted that the values of Log (K) become positive when the carbon was added into, which illustrated that the presence of carbon will facilitate the conversion of CuFe2O4. The results will provide a theoretical basis for the solid-solid reactions of C with CuFe2O4. 2

Fuel 267 (2020) 117163

T. Li, et al.

distinctive reflection plans (1 0 1), (1 1 2), (2 0 0), (1 0 3), (2 1 1), (2 0 2), (0 0 4), (2 2 0), (2 0 4), (3 1 2), (1 0 5), (3 0 3), (3 2 1), (2 2 4), (4 0 0), (4 1 3), (4 2 2) and (4 0 4). These narrow and strong peaks confirm the spinel cubic structure of CuFe2O4 and its good crystallinity, and similar results have been reported by Haija et al. [25]. However, the CuO impurity phase is observed in the CuFe2O4 samples, and it was also observed by others [38,39]. It was reported by Laokul et al. [39] that the CuO-phase could be found in the CuFe2O4 sample calcined at the highest temperature of 900 °C. The Raman spectroscopic analysis of fresh CuFe2O4 OC was carried out between 100 and 1000 cm−1 at room temperature, shown in Fig. 3. Four identified Raman peaks of CuFe2O4 are observed in the pattern, which are around 164, 461, 554 and 678 cm−1, and the similar results have been reported by Verma et al. [40]. The peaks at 461, 554 and 678 cm−1 are indexed to T2g, F2g, and A1g Raman active modes, respectively [41–43]. The peak at 678 cm−1 correspond to T-site mode that represents the local lattice effect in the tetrahedral sub-lattice. The peak at 461 cm−1 is related to the O-site mode, which is used to represent the effect of local lattice in the octahedral sublattice [44]. The external morphology of the CuFe2O4 OC calcined at 850 °C has been captured from the SEM (Fig. 4). It indicated the uniform size and shape of CuFe2O4 external morphology. Some uneven holes are observed owing to the release of gases during the calcination process. Citric acid complex agent contains a large amount of organic which promotes the formation of pores in the CuFe2O4 structure [45]. The overall improvement in material properties may be due to better interparticle connectivity, which was propound by Selvan et al. [46]. The TG and DTG analysis of spinel CuFe2O4 oxygen carrier was carried out to investigate its O2 emission behavior. The sample was heated from room temperature to 1000 °C at 5 °C/min under the pure N2 atmosphere (50 mL/min), and it was held for 30 min at final temperature. From Fig. 5a we draw that the initial decomposition of CuFe2O4 takes place at about 700 °C. The low decomposition temperature meant the low surface activation energy and higher decomposition reactivity of CuFe2O4 OC [47]. As the temperature increased, the decomposition rate also increased, and it reached a maximum of 0.171%/min at about 800 °C. The spinel was still not completely decomposed at 1000 °C, similar to the finding of Zhang et al. [16]. It was observed that the mass loss of copper ferrite reached about 6% at 1000 °C, related to the emission of O2 through direct decomposition of the prepared OC. However, the CuFe2O4 could not release O2 through direct decomposition of the CuFe2O4 in thermodynamics, this was ascribed to that the process of TG was a dynamic process, in which the carrier gas (N2) provided a favorable trend for the leaving of O2 which generated from the lattice oxygen. Then the XRD pattern of decomposition products of TG was shown in Fig. 5b. There are three phases including CuFeO2, Fe3O4 and Cu2O, their corresponding crystal planes have been given in the picture. The existing of CuFeO2 and Fe3O4 showed that the decomposition reaction of Eq. (2) was happened in the thermal decomposition process, similar

Fig. 2. XRD patterns of the prepared CuFe2O4.

Fig. 3. Raman spectrum of fresh CuFe2O4 at room temperature.

3.2. Characterization of fresh catalyst The crystal structural of spinel oxygen carrier is important for its oxygen transfer. Therefore, the phase of the sol-gel prepared CuFe2O4 and its detailed thermal decomposition products were identified using XRD analysis. First, the XRD pattern of CuFe2O4 sample is shown in Fig. 2. It indicated that the peaks of CuFe2O4 are in good agreement with the basis of well-defined XRD peaks, corresponding to the

Fig. 4. SEM images of CuFe2O4 under different magnifications (a) 5000 × and (b) 10000×. 3

Fuel 267 (2020) 117163

T. Li, et al.

Fig. 5. TG data of CuFe2O4 (a) and the XRD patterns of its products (b).

to the results of Liu et al. [4]. Therefore, the gaseous oxygen release of our CuFe2O4 was 4.44% theoretically according to the Eq. (2), and the result was inconsistent with the mass loss (6%) of copper ferrite in its TG curve. What cannot be ignored was the CuO impurity phase in the fresh OC, shown in Fig. 2. CuO has a high oxygen transfer capacity, and it could easily decompose into Cu2O and O2 at about 800 °C in N2, reported by the literatures [48,49]. The existing of Cu2O in the XRD pattern of Fig. 5b proved our conclusion about the releasing of O2 by CuO at 850 °C. Overall, both CuO and CuFe2O4 were responsible for the weight loss of OC. 3.3. Solid-solid reaction 3.3.1. Solid products analysis Fig. 6 depicted the XRD patterns of the solid products obtained from the reaction between CuFe2O4 and coke for various time intervals. The data indicated that the phase change of oxygen carrier occurred during the reaction with C. When the reaction time is 2 min, we find that the CuFe2O4 and CuO phase has been replaced by CuFeO2, Fe2O3, Fe3O4 and Cu. Lots of peaks of CuFeO2 and Fe3O4 were attributed to the occurrence of Eqs. (3), which was the primary reaction between CuFe2O4 and C. The presence of Cu and Fe2O3 was attributed to the reduction of

Fig. 6. XRD data of CuFe2O4 sampled at different reduction times with sucrose coke.

4

Fuel 267 (2020) 117163

T. Li, et al.

for the solid products obtained at 5 min and 15 min, except for the increasing diffraction intensity of Fe3O4. When the reaction time is 30 min, we find FeO phase, and it reveals the further reduction of Fe3O4. The Raman spectra of the solid products with different reaction time were shown in Fig. 7. At the beginning of 2 min, there are few peaks at 224, 289, 348, 410, 501, 608, 683 and 1313 cm−1. The peaks at 224, 410, 501, 608 and 1313 cm−1 are in good agreement with the typical Raman peaks of α-Fe2O3 [50], the peaks at 348 and 683 cm−1 correspond to the peaks of CuFeO2 [51], and the peak at 289 cm−1 is indexed to Fe3O4 [52]. The Raman results of products at 2 min are consistent with their XRD analysis. The Raman pattern of the product at 5 min is similar to that of 15 min and 30 min. Four dominant Raman bands at 195, 298, 524, and 656 cm−1 are observed, which are regarded as the typical peaks of Fe3O4. As reported by Guo et al. [52], the Raman bands at 298, 524 and 656 cm−1 correspond to the Eg, T2g(2) and A1g mode of Fe3O4, respectively. The difference is that the intensity for the Fe3O4 increased with increasing reaction times. The results revealed the controlled transformation of CuFe2O4 to Fe3O4 and Cu, which caused an increase in the content of Fe3O4. No FeO was found in the product at 30 min, possibly due to its low amount.

Fig. 7. Raman spectra of CuFe2O4 sampled at different reduction times with sucrose coke.

CuO and the occurrence of Eqs. (4) and (5), respectively. With the increase of reaction time, then the CuFeO2 and Fe2O3 were further reduced to Fe3O4 and Cu, corresponding to Eqs. (6) and (7). Wang et al. [49] obtained the same solid reaction products via the reaction between coal and CuFe2O4 at 900 °C for 15 min. There is no obvious difference

3.3.2. Gaseous products analysis The molar flow rates of gas components CO and CO2 with different reaction time were shown in Fig. 8. It is worth noting that the peak of CO2 reaches its maximum value (3 min) before CO (5 min). The

Fig. 8. Gaseous products of the solid-solid reaction of CuFe2O4 with C. 5

Fuel 267 (2020) 117163

T. Li, et al.

3.4. Reaction mechanism CuFe2O4 with the antiferromagnetic arrangement is an important spinel oxide, and its unit cell is displayed in Fig. 9. In an ideal inverse CuFe2O4 spinel structure, the tetrahedral hollow sites are occupied by Fe atoms while the octahedral hollow sites are occupied by equal Cu and Fe atoms. The calculations were based on spin-polarized density functional theory method. The generalized gradient approximation (GGA) functional as parameterized by Perdew, Burke, and Ernzerhof (PBE) was used to explain the correlated 3d orbital of Cu and Fe atoms [53,54]. It has been reported that GGA can give accurately structural and electronic properties of spinels [55]. The Monkhorst-Pack grid k-points were used to provide cell parameters. A 4 × 4 × 4 set of k-points was used for bulk relaxation, and a 4 × 4 × 1 set of k-points was used for surface optimization as well as for C adsorption. A cutoff energy of 450 eV was used in all calculations. There are convergence criterions for unit cell optimization of CuFe2O4, which is the atomic forces of 2.0 × 10−3 Hartree/Å, maximum displacement of 5.0 × 10−3 Å and total energy variation of 1.0 × 10−5 Hartree, respectively. In order to find the ground state of spinel CuFe2O4, a full optimization of the cell parameters for bulk structure of CuFe2O4 was performed, shown in Fig. 10a, and the optimization results were listed in Table 1. The lattice parameter of optimized CuFe2O4 structure is a = b = 6.067 Å and c = 19.774 Å. Previous researches have demonstrated that (1 0 0) surface of CuFe2O4 has a better thermodynamic stability and lower catalytic activity [56,57]. Furthermore, Cu atoms, Fe atoms and the O atoms with different Cu/Fe coordination are exposed on this surface. Thus CuFe2O4 (1 0 0) surface was adopted in this study. Accordingly, two modeled slabs (named as Fe, Cu-terminated surfaces and O-terminated surface) were built, as shown in Fig. 10(b and c). Since the asymmetrical forces were performing on atoms near the surface are asymmetric, the surface relaxation should be considered to improve the accuracy of calculations. Table 2 showed the variations of surface displacement and energy for these two slabs before and after surface relaxation. It was found that the relaxed layers had a tendency to move to the bottom, which meant that the slab layers become more compact and the relaxed surfaces are more stable under the effect of surface relaxation. These results fully showed that surface relaxation could not be neglected for an accurate result. In addition, the adsorption property reflects the nature of CuFe2O4 catalyst, and it can serve as a basis of the comprehensive mechanism for the chemical looping reaction process of CuFe2O4 with C. The free molecules (carbon) were added into the slab model, and the adsorption energies (Eads) of carbon on different slab were calculated by Eq. (9).

Fig. 9. Crystal structure of CuFe2O4.

Fig. 10. The optimized structure of CuFe2O4 (1 0 0) orientation (a). The settings and possible adsorption sites of two cleaved surfaces: O-terminated surface (b) and Fe, Cu-terminated surface (c). Table 1 The optimized cell parameters of bulk CuFe2O4.

CuFe2O4

a = b (Å)

c (Å)

RFe−O (Å)

RCu−O (Å)

6.067

19.774

2.038

2.012

Eads = Emole + slab − Emole − Eslab Table 2 The variation of slab displacement and energy before and after relaxation.

Fe, Cu-terminated surface O-terminated surface

△Z (Å)

△E (eV)

−0.235 −0.154

−1.04 −1.59

(9)

where Eslab, Emole and Emole+slab denote the slab [O-terminated surface or Fe, Cu-terminated surface], free molecule (carbon), and total energy of slab with molecule, respectively. Fig. 11a showed that C molecule was adsorbed on the top of the surface O atom, and Fig. 11b showed that C molecule was adsorbed on the surface Fe atom and Cu atom. As shown in Table 3, the adsorption energies of C on Fe, Cu-terminated surface and O-terminated surface was −2.69 eV and −4.03 eV, respectively. The adsorption energies of two configurations were negative, it suggested that the adsorption of C molecule on CuFe2O4 (1 0 0) surface is exothermic. The results indicated that the C was in favor of adsorbing on O-terminated surface. It was found that the interaction between adsorbed carbon and surface lattice oxygen was easily to form COX at a high temperature. Due to that the oxygen vacancies arose, which resulted in an oxygen diffusion from sub-layers to the surface, and it also explained the facilitating effect of C on the spinel CuFe2O4 decomposition in thermodynamic analysis.

generation of CO2 was due to the reaction of the CuFe2O4 with C, and the CO was due to the reaction of product CuFeO2 with C. It is found that the solid-solid reaction is basically over in 30 min, shown in Fig. 8d. Furthermore, the integral results on curves at 30 min showed that the mass of the C contained in CO and CO2 was 10.7 mg in total. The carbon conversion is only 10.7%, and the low conversion was possibly related to the reduction of solid contact surfaces, and only the carbon in surface reacted.

6

Fuel 267 (2020) 117163

T. Li, et al.

Fig. 11. Optimized structures of C adsorption on the O-terminated surface (a) and Fe, Cu-terminated surface (b).

Acknowledgments

Table 3 Adsorption energies of C on the surface of CuFe2O4.

This work is financially supported by the National Natural Science Foundation of China, China (No. 21676148), and the Fundamental Research Funds for the Central Universities, China (No. 30918012202).

eV Fe, Cu-terminated surface O-terminated surface

−2.69 −4.03

References 4. Conclusion

[1] Saito T, Lin S. Coal char reaction with oxygen carrier in chemical looping combustion. Energy Fuels 2019. [2] Yongzhuo L, Qingjie G. Investigation into syngas generation from solid fuel using CaSO4-based chemical looping gasification process. Chin J Chem Eng 2013;21:127–34. [3] Yang J, Ma L, Yang J, Guo Z, Liu H, Zhang W. Chemical looping gasification of phosphogypsum as an oxygen carrier: the Ca and S migration mechanism using the DFT method. Sci Total Environ 2019;689:854–64. [4] Liu F, Liu J, Yang Y, Wang Z, Zheng C. Reaction mechanism of spinel CuFe2O4 with CO during chemical-looping combustion: an experimental and theoretical study. Proc Combust Inst 2019;37:4399–408. [5] Abad A, Adánez J, García-Labiano F, Luis F, Gayán P, Celaya J. Mapping of the range of operational conditions for Cu-, Fe-, and Ni-based oxygen carriers in chemical-looping combustion. Chem Eng Sci 2007;62:533–49. [6] Siriwardane R, Tian H, Simonyi T, Poston J. Synergetic effects of mixed copper–iron oxides oxygen carriers in chemical looping combustion. Fuel 2013;108:319–33. [7] Siriwardane RV, Ksepko E, Tian H, Poston J, Simonyi T, Sciazko M. Interaction of iron–copper mixed metal oxide oxygen carriers with simulated synthesis gas derived from steam gasification of coal. Appl Energy 2013;107:111–23. [8] Berguerand N, Lyngfelt A. Operation in a 10 kWth chemical-looping combustor for solid fuel—testing with a Mexican petroleum coke. Energy Procedia 2009;1:407–14. [9] Leion H, Mattisson T, Lyngfelt A. Solid fuels in chemical-looping combustion. Int J Greenhouse Gas Control 2008;2:180–93. [10] Bolhàr-Nordenkampf J, Pröll T, Kolbitsch P, Hofbauer H. Performance of a NiObased oxygen carrier for chemical looping combustion and reforming in a 120 kW unit. Energy Procedia 2009;1:19–25. [11] Shen L, Wu J, Xiao J. Experiments on chemical looping combustion of coal with a NiO based oxygen carrier. Combust Flame 2009;156:721–8. [12] Leion H, Mattisson T, Lyngfelt A. Using chemical-looping with oxygen uncoupling (CLOU) for combustion of six different solid fuels. Energy Procedia 2009;1:447–53. [13] Adánez-Rubio I, Gayán P, García-Labiano F, Luis F, Adánez J, Abad A. Development of CuO-based oxygen-carrier materials suitable for chemical-looping with oxygen uncoupling (CLOU) process. Energy Procedia 2011;4:417–24. [14] Wang B, Yan R, Zhao H, Zheng Y, Liu Z, Zheng C. Investigation of chemical looping combustion of coal with CuFe2O4 oxygen carrier. Energy Fuels 2011;25:3344–54. [15] Wang B, Xiao G, Song X, Zhao H, Zheng C. Chemical looping combustion of highsulfur coal with NiFe 2 O 4-combined oxygen carrier. J Therm Anal Calorim 2014;118:1593–602. [16] Zhang P, Yu B, Zhang L. Mechanism of oxygen releasing of copper ferrite in the formation of the corresponding oxygen-deficient compound. Sci China, Ser B Chem 2009;52:101. [17] Ksepko E, Łabojko G. Effective direct chemical looping coal combustion with bimetallic Fe–Cu oxygen carriers studied using TG-MS techniques. J Therm Anal Calorim 2014;117:151–62. [18] Fan Y, Siriwardane R. Novel new oxygen carriers for chemical looping combustion of solid fuels. Energy Fuels 2014;28:2248–57. [19] Zhang Y, Wei T, Xu K, Ren Z, Xiao L, Song J, et al. Catalytic decomposition action of hollow CuFe 2 O 4 nanospheres on RDX and FOX-7. RSC Adv 2015;5:75630–5. [20] Tang M, Xu L, Fan M. Progress in oxygen carrier development of methane-based chemical-looping reforming: a review. Appl Energy 2015;151:143–56. [21] Mihai O, Chen D, Holmen A. Catalytic consequence of oxygen of lanthanum ferrite perovskite in chemical looping reforming of methane. Ind Eng Chem Res 2010;50:2613–21. [22] Galinsky N, Sendi M, Bowers L, Li F. CaMn1− xBxO3− δ (B= Al, V, Fe, Co, and Ni) perovskite based oxygen carriers for chemical looping with oxygen uncoupling (CLOU). Appl Energy 2016;174:80–7. [23] Liu C, Wang W. Chemical looping gasification of pyrolyzed biomass and coal char with copper ferrite as an oxygen carrier. J Renewable Sustainable Energy

In this study, a comprehensive investigation of solid-solid reaction of spinel CuFe2O4 with C was performed on a fixed-bed reactor under atmospheric pressure. XRD and Raman were employed to characterize the evolution of oxygen carrier phases during the reaction of spinel and carbon. There are three main stages for the reaction of C with CuFe2O4. In the first stage, CuFe2O4 is reduced into CuFeO2 and Fe3O4 with a high reaction rate. In the second stage, the CuFe2O4 was mainly reduced to Cu and Fe3O4. In the last stage, the product Fe3O4 is slowly transformed to FeO even to Fe. The uniform size and shape of the CuFe2O4 and uneven holes were observed from SEM, the better grainto-grain connectivity improved the properties of CuFe2O4. The thermodynamic calculation and TG were carried out to gain a comprehensive insight into the reaction process. TG data showed that the initial decomposition of CuFe2O4 took place at about 700 °C, and the decomposition rate reached a maximum of 0.171%/min at about 800 °C. The results of thermodynamic calculation indicated that the thermal decomposition of CuFe2O4 is impossible at 850 °C in theory, but the reaction of carbon with spinel will greatly facilitate the conversion of CuFe2O4. The gaseous product were analyzed by an online gas analyzer. The low conversion (10.7%) was possibly related to the insufficient mixing of CuFe2O4 with carbon and the reduction of solid contact surfaces. The DFT results of carbon oxidation over the CuFe2O4 (1 0 0) surface indicated that the C was in favor of adsorbing on Oterminated surface, so lattice oxygen was easily released to form COx. Overall, this work could provide useful information for the practical application of CLC and CLG system.

CRediT authorship contribution statement Tianle Li: Conceptualization, Methodology, Writing - original draft. Qiao Wu: Writing - review & editing, Visualization. Wenju Wang: Supervision, Visualization, Formal analysis, Project administration. YuPeng Xiao: Investigation, Data curation. Chenlong Liu: Software, Validation. Fufeng Yang: Funding acquisition.

Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

7

Fuel 267 (2020) 117163

T. Li, et al.

[41] Balaji M, Chithra Lekha P, Pathinettam Padiyan D. Core–shell structure in copper ferrite–polyaniline nanocomposite: Confirmation by laser Raman spectra. Vib Spectrosc 2012;62:92–7. [42] Wang Z, Lazor P, Saxena SK, O’Neill HSC. High pressure Raman spectroscopy of ferrite MgFe2O4. Mater Res Bull 2002;37:1589–602. [43] Anandan S, Selvamani T, Prasad GG, Asiri AM, Wu JJ. Magnetic and catalytic properties of inverse spinel CuFe2O4 nanoparticles. J Magn Magn Mater 2017;432:437–43. [44] Liu Y, Zhang Y, Feng J, Li C, Shi J, Xiong R. Dependence of magnetic properties on crystallite size of CoFe2O4 nanoparticles synthesised by auto-combustion method. J Exp Nanosci 2009;4:159–68. [45] Costa AF, Pimentel PM, Aquino FM, Melo DMA, Melo MAF, Santos IMG. Gelatin synthesis of CuFe2O4 and CuFeCrO4 ceramic pigments. Mater Lett 2013;112:58–61. [46] Selvan RK, Augustin CO, Berchmans LJ, Saraswathi R. Combustion synthesis of CuFe2O4. Mater Res Bull 2003;38:41–54. [47] Yu B, Zhang P, Zhang L, Chen J, Xu J. Studies on the preparation of active oxygendeficient copper ferrite and its application for hydrogen production through thermal chemical water splitting. Sci China, Ser B Chem 2008;51:878–86. [48] Hosseini D, Imtiaz Q, Abdala P, Yoon S, Kierzkowska A, Weidenkaff A, et al. CuO promoted Mn 2 O 3-based materials for solid fuel combustion with inherent CO 2 capture. J Mater Chem A 2015;3:10545–50. [49] Wang B, Ma Q, Wang W, Zhang C, Mei D, Zhao H, et al. Effect of reaction temperature on the chemical looping combustion of coal with CuFe2O4 combined oxygen carrier. Energy Fuels 2017;31:5233–45. [50] De Faria D, Venâncio Silva S, De Oliveira M. Raman microspectroscopy of some iron oxides and oxyhydroxides. J Raman Spectrosc 1997;28:873–8. [51] Pavunny SP, Kumar A, Katiyar R. Raman spectroscopy and field emission characterization of delafossite CuFeO 2. J Appl Phys 2010;107:013522. [52] Guo C, Hu Y, Qian H, Ning J, Xu S. Magnetite (Fe3O4) tetrakaidecahedral microcrystals: synthesis, characterization, and micro-Raman study. Mater Charact 2011;62:148–51. [53] Dudarev S, Botton G, Savrasov S, Humphreys C, Sutton A. Electron-energy-loss spectra and the structural stability of nickel oxide: An LSDA+ U study. Phys Rev B 1998;57:1505. [54] Liechtenstein A, Anisimov V, Zaanen J. Density-functional theory and strong interactions: orbital ordering in Mott-Hubbard insulators. Phys Rev B 1995;52:R5467. [55] Yang Y, Liu J, Zhang B, Liu F. Mechanistic studies of mercury adsorption and oxidation by oxygen over spinel-type MnFe2O4. J Hazard Mater 2017;321:154–61. [56] Zasada F, Grybos J, Indyka P, Piskorz W, Kaczmarczyk J, Sojka Z. Surface structure and morphology of M [CoM′] O4 (M= Mg, Zn, Fe, Co and M′= Ni, Al, Mn, Co) spinel nanocrystals DFT+ U and TEM screening investigations. J Phys Chem C 2014;118:19085–97. [57] Jiang Z, Zhang W, Shangguan W, Wu X, Teraoka Y. Adsorption of NO molecule on spinel-type CuFe2O4 surface: a first-principles study. J Phys Chem C 2011;115:13035–40.

2018;10:063101. [24] Miller DD, Siriwardane R. CaFe2O4 oxygen carrier characterization during the partial oxidation of coal in the chemical looping gasification application. Appl Energy 2018;224:708–16. [25] Haija MA, Abu-Hani AF, Hamdan N, Stephen S, Ayesh AI. Characterization of H2S gas sensor based on CuFe2O4 nanoparticles. J Alloy Compd 2017;690:461–8. [26] Wang B, Zheng Y, Liu Z, Zhao H, Zheng C, Yan R. Investigation of chemical looping combustion of coal with Fe2O3-based combined oxygen carrier. J Eng Thermophys 2010;31:1427–30. [27] Wang B, Wang W, Ma Q, Lu J, Zhao H, Zheng C. In-depth investigation of chemical looping combustion of a Chinese bituminous coal with CuFe2O4 combined oxygen carrier. Energy Fuels 2016;30:2285–94. [28] Wang B, Cao Y, Li J, Wang W, Zhao H, Zheng C. Migration and redistribution of sulfur species during chemical looping combustion of coal with CuFe2O4 combined oxygen carrier. Energy Fuels 2016;30:8499–510. [29] Zhang P, Zhang L, Nie R, Wang J. Preparation of CuFe2O4 and its preliminary application study in two-step thermochemical water-splitting cycle. Int J Nucl Hydrogen Prod Appl 2008;1:219–28. [30] Li F, Kim HR, Sridhar D, Wang F, Zeng L, Chen J, et al. Syngas chemical looping gasification process: oxygen carrier particle selection and performance. Energy Fuels 2009;23:4182–9. [31] Mattisson T, García-Labiano F, Kronberger B, Lyngfelt A, Adánez J, Hofbauer H. Chemical-looping combustion using syngas as fuel. Int J Greenhouse Gas Control 2007;1:158–69. [32] Linderholm C, Mattisson T, Lyngfelt A. Long-term integrity testing of spray-dried particles in a 10-kW chemical-looping combustor using natural gas as fuel. Fuel 2009;88:2083–96. [33] Kathe MV, Empfield A, Na J, Blair E, Fan L-S. Hydrogen production from natural gas using an iron-based chemical looping technology: Thermodynamic simulations and process system analysis. Appl Energy 2016;165:183–201. [34] Wang B, Yan R, Lee DH, Liang DT, Zheng Y, Zhao H, et al. Thermodynamic investigation of carbon deposition and sulfur evolution in chemical looping combustion with syngas. Energy Fuels 2008;22:1012–20. [35] Wang W, Wang Y. Thermodynamic analysis of hydrogen production via partial oxidation of ethanol. Int J Hydrogen Energy 2008;33:5035–44. [36] Melgar A, Pérez JF, Laget H, Horillo A. Thermochemical equilibrium modelling of a gasifying process. Energy Convers Manage 2007;48:59–67. [37] Atkins PW, De Paula J, Keeler J. Atkins' Physical Chemistry. Oxford University Press; 2018. [38] Yang H, Yan J, Lu Z, Cheng X, Tang Y. Photocatalytic activity evaluation of tetragonal CuFe2O4 nanoparticles for the H2 evolution under visible light irradiation. J Alloy Compd 2009;476:715–9. [39] Laokul P, Amornkitbamrung V, Seraphin S, Maensiri S. Characterization and magnetic properties of nanocrystalline CuFe2O4, NiFe2O4, ZnFe2O4 powders prepared by the Aloe vera extract solution. Curr Appl Phys 2011;11:101–8. [40] Verma K, Kumar A, Varshney D. Effect of Zn and Mg doping on structural, dielectric and magnetic properties of tetragonal CuFe2O4. Curr Appl Phys 2013;13:467–73.

8