Advanced Chemical Looping Materials for CO2 Utilization: A Review

Combining chemical looping with a traditional fuel conversion process yields a promising technology for low-CO2-emission energy production. Bridged by the cyclic transformation of a looping material (CO2 carrier or oxygen carrier), a chemical looping process is divided into two spatially or temporally separated half-cycles. Firstly, the oxygen carrier material is reduced by fuel, producing power or chemicals. Then, the material is regenerated by an oxidizer. In chemical looping combustion, a separation-ready CO2 stream is produced, which significantly improves the CO2 capture efficiency. In chemical looping reforming, CO2 can be used as an oxidizer, resulting in a novel approach for efficient CO2 utilization through reduction to CO. Recently, the novel process of catalyst-assisted chemical looping was proposed, aiming at maximized CO2 utilization via the achievement of deep reduction of the oxygen carrier in the first half-cycle. It makes use of a bifunctional looping material that combines both catalytic function for efficient fuel conversion and oxygen storage function for redox cycling. For all of these chemical looping technologies, the choice of looping materials is crucial for their industrial application. Therefore, current research is focused on the development of a suitable looping material, which is required to have high redox activity and stability, and good economic and environmental performance. In this review, a series of commonly used metal oxide-based materials are firstly compared as looping material from an industrial-application perspective. The recent advances in the enhancement of the activity and stability of looping materials are discussed. The focus then proceeds to new findings in the development of the bifunctional looping materials employed in the emerging catalyst-assisted chemical looping technology. Among these, the design of core-shell structured Ni-Fe bifunctional nanomaterials shows great potential for catalyst-assisted chemical looping.


Introduction
To bolster the rapid economic growth of modern society, the entire world is facing increasing demands on its energy sources. Although the consumption of non-fossil fuels (such as renewable energy and nuclear power) is projected to grow faster than fossil fuels (coal, petroleum and natural gas), the latter are expected to account for almost 80% of global energy use through 2040 [1,2]. Therefore, the continued dependence on fossil fuels for energy supply in the foreseeable future is inevitable. However, fossil fuel-based energy production is widely considered to be a leading cause for greenhouse gas emissions such as CO 2 , which is the largest contributor towards the global warming effect [3][4][5][6][7][8].
With the adoption of the Paris Agreement at the 21st Conference of the Parties of the United Nations Framework Convention on Climate Change (UNFCCC) in December 2015, scientists from all over the scientists from all over the world are now being challenged with a new goal: To limit the global temperature increase to 1.5 °C above pre-industrial levels [9].
Hence, there is an urgent need for strategies to reduce CO2 emissions in various related fields, especially industrial production and electricity generation, which contribute almost half of total emissions [10]. Carbon Capture and Storage (CCS) is a central strategy, as it not only reduces the associated greenhouse gas emissions in conformity with global targets, but also provides the opportunity to allow for a more sustainable use of fossil fuel energy [11]. The applications of the CCS scheme are reflected in two groups of technologies: One is the capture of CO2 from power plants, followed by compression, transportation and permanent storage [12]; another is the combination with strategies to improve energy efficiency for the separation of CO2, or the conversion of CO2 to value added chemicals [13][14][15]. The recent progress in CO2 capture and separation technologies with respect to the feasible processes, as well as new materials have been reported in some reviews [3,7,16].
In the past few decades, the large-scale capture of CO2 using amine absorbers became commercially available and was mainly applied in CO2 separation from flue gases [17]. This conventional amine scrubbing for CO2 capture however increases the energy requirements of a plant by 25-40% [18].
From an economic point of view, there is need to improve current technologies for lowering the overall cost of CO2 capture and separation processes, which can be implemented from two aspects: Upgrading the efficiency of capture materials to decrease material expense, or designing novel process concepts to lower equipment cost. Figure 1 shows a series of important developments of CO2 capture technologies [16]. Evidently, more time and funding investments are required to commercialize the innovative techniques that possess higher cost reduction benefits. Among the innovating technologies, chemical looping technology has the potential to achieve inherent separation of CO2 from a fuel feed, thus eliminating energy intensive gas-gas separation costs, expectedly providing highest cost reduction benefits once it is fully commercialized. Furthermore, this technique can also provide an approach to convert CO2 into value added chemicals and energy sources (syngas, CO) [19][20][21][22][23]. The concept chemical looping was first used by Ishida et al. [24] to describe a high-exergy efficiency combustion process that employs redox reactions of chemical intermediates. Later, this concept was extended to a broad range of applications, depending on the nature of a recyclable  The type A-2 process uses a regenerable solid sorbent to achieve the in-situ capture of CO2 through a cyclic carbonation-calcination reaction before, during or after the carbonaceous fuel combustion or reforming [27]. It provides a viable CO2 control technology, i.e., sorption-enhanced chemical looping (SE-CL), for process intensification [15] or high-purity chemicals (CO2-lean) production [28]. The SE-CL process embodies two steps: CO2 is in-situ absorbed through the carbonation of a metal oxide, forming metal carbonate in the carbonator, afterwards the latter is calcined at higher temperature to regenerate the metal oxide, thereby releasing CO2.
In group B, a similar metal oxide regeneration process is involved as in group A, but the reactions in the former are not intended to produce CO2. The type B-1, chemical looping reforming (CLR), aims to reform carbonaceous fuels to syngas (a mixture of H2 and CO) or chemicals (e.g., ethylene) [29]. Utilization of CO2 can be achieved if it is used as an oxidizing agent in the re-oxidation half-cycle, turning it into chemical looping dry reforming (CLDR). Compared to conventional dry reforming (DR), application of CLDR brings two main advantages: First, since the looping material is regenerated in each cycle, deposited carbon can be removed by reacting with CO2. Second, as CO2 is used as an oxidizing agent for material regeneration in the re-oxidation half-cycle, this process can The type A-2 process uses a regenerable solid sorbent to achieve the in-situ capture of CO 2 through a cyclic carbonation-calcination reaction before, during or after the carbonaceous fuel combustion or reforming [27]. It provides a viable CO 2 control technology, i.e., sorption-enhanced chemical looping (SE-CL), for process intensification [15] or high-purity chemicals (CO 2 -lean) production [28]. The SE-CL process embodies two steps: CO 2 is in-situ absorbed through the carbonation of a metal oxide, forming metal carbonate in the carbonator, afterwards the latter is calcined at higher temperature to regenerate the metal oxide, thereby releasing CO 2 .
In group B, a similar metal oxide regeneration process is involved as in group A, but the reactions in the former are not intended to produce CO 2 . The type B-1, chemical looping reforming (CLR), aims to reform carbonaceous fuels to syngas (a mixture of H 2 and CO) or chemicals (e.g., ethylene) [29]. Utilization of CO 2 can be achieved if it is used as an oxidizing agent in the re-oxidation half-cycle, turning it into chemical looping dry reforming (CLDR). Compared to conventional dry reforming (DR), application of CLDR brings two main advantages: First, since the looping material is regenerated in each cycle, deposited carbon can be removed by reacting with CO 2 . Second, as CO 2 is used as an oxidizing agent for material regeneration in the re-oxidation half-cycle, this process can utilize more Materials 2018, 11, 1187 4 of 32 CO 2 than conventional dry reforming; in case of CH 4 reforming, the CO 2 converted in CLDR is three times higher than in DR [23].
When using hydrocarbon fuels such as natural gas and biogas in CLDR, the kinetics of the reduction half-cycle is limited. The natural mixture of CH 4 with CO 2 (e.g., in biogas) significantly decreases the rate and degree of metal oxide reduction because CO 2 also acts as an oxidizing agent. Both CO 2 in the feed and as a product have this retarding effect on the metal oxide reduction, thus limiting the conversion of CO 2 to CO in the re-oxidation half-cycle. In order to overcome these drawbacks, an extra catalyst component can be introduced to convert CH 4 and CO 2 into CO and H 2 , which then both reduce the metal oxide during the reduction half-cycle. This not only increases the CH 4 and CO 2 conversions, but also diminishes the oxidizing atmosphere because both CO and H 2 are reducing agents. Based on this concept, catalyst-assisted chemical looping dry reforming (CCDR) was recently proposed as an intensified process with the goal of maximum utilization of two main greenhouse gases, CH 4 and CO 2 , in terms of CO yield. This process is generally implemented over a bifunctional reactor bed, composed of either a physical mixture of a reforming catalyst and a metal oxide looping material, or a bifunctional looping material [15,23,30]. The entire CCDR process yields the dry reforming of CH 4 into CO as an overall reaction, being a strongly endothermic process. Thermodynamic calculations indicate that an increase of temperature is necessary to obtain high CH 4 and CO 2 conversion, leading to larger energy requirements for heating the reactor, which highly increases the operating cost. Additionally, material deactivation due to active metal sintering will be aggravated under high-temperature conditions during dry reforming [31]. To counter these problems, CCDR was adjusted to yield a new process, called catalyst-assisted chemical looping auto-thermal dry reforming (CCAR) [20]. Here, a small amount of O 2 is added to CH 4 and CO 2 to partially convert CH 4 in an exothermic reaction (partial oxidation or combustion) [31][32][33][34][35], so that heat is generated in-situ, thereby compensating the strong endothermicity of the dry reforming reaction. Based on thermodynamic analysis, the whole CCAR process can run in thermoneutral conditions, i.e., ∆H = 0 kJ·mol −1 , with molar ratios of CH 4 :CO 2 :O 2 = 1:2.4:0.3 at 600 • C or with molar ratios of CH 4 :CO 2 :O 2 = 1:2:0.5 at 750 • C [20]. Owing to the inherent composition of biogas, typically containing CH 4 (45~75%), CO 2 (25~55%) and O 2 (0.01~5%) [36], the CCAR can efficiently utilize biogas as fuel feed in the reduction half-cycle.
Compared to type B-1, the operation of type B-2 uses carbon neutral energy sources like solar and nuclear instead of carbonaceous fuels, providing a renewable chemical looping reforming (R-CLR) process. In R-CLR, considerable heat from solar or nuclear origin is imposed on the metal oxide to decompose it to metal and oxygen in the reducer. Subsequently, the metal is re-oxidized to metal oxide by steam or CO 2 , forming pure H 2 or CO [37].
In light of the above overview of chemical looping processes, the key to success for chemical looping technology are the looping materials. Their adequacy in reactivity at temperatures below the melting point ensures the energy conversion efficiency during chemical looping. On the other hand, the contamination resistance and physical strength of a looping material at high redox reaction temperatures determine the durability of a chemical looping process. Further, the low price and environmental compatibility of the looping materials play a crucial role in the sustainability of the chemical looping technology. Therefore, major efforts have been made to develop the ideal chemical looping materials in the past decade. During the early stages of development, research focused on the metal oxide materials with a single function as CO 2 accepter or oxygen storage, which are mainly used for conventional chemical looping combustion and reforming processes. The advances were reported in a few review papers [1,2,14,21,26,38,39]. More recently, a novel process, so-called catalyst-assisted chemical looping, was proposed with the aim of maximal CO 2 utilization. Since then, the development of chemical looping materials has been directed towards the bifunctional material that integrates catalyst and oxygen storage material in one unit, achieving both catalytic fuel conversion and chemical looping CO 2 utilization. In this review article, an update of the previous overview of chemical looping materials is given, together with an extension focusing on new insights in the development of bifunctional looping materials based on the latest studies.

Materials for Chemical Looping
Depending on the chemical looping process performed, the looping materials are mainly classified in two types: CO 2 carrier and oxygen carrier, which are required to have high reactivity and sufficient stability, and to be cost-effective and environmentally friendly [40].

CO 2 Carrier
Among all the solid CO 2 sorbents, CaO is a proper material to act as CO 2 carrier because of numerous advantages [41,42]: (1) low cost since it can be derived from limestone that is readily available; (2) non-toxic to the environment; (3) the spent materials can be further used in the cement industry after decarburization, supporting sustainable development in economy; (4) CaO may also serve as an alkaline catalyst in biomass conversion processes; and (5) the heat from the exothermic carbonation of CaO can be retrieved to compensate the high heat requirement during chemical looping. However, it is well known that the reactivity decay over cycles of CaO-based CO 2 sorbents, particularly natural material-based sorbents, is inevitable. The main causes for the deactivation are particle sintering and attrition. Hence, many efforts have been made to improve their sintering resistance ability and mechanical strength, including doping of the material to suppress sintering, thermal or chemical pretreatment to achieve superior structural properties, development of new synthetic sorbents with good CO 2 capture capacity and stability. Additionally, the development of feasible technologies for the reactivation of spent sorbents is of equal importance. The recent progress in CaO-based looping materials has been summarized by numerous review articles [43][44][45][46].

Oxygen Carrier
The nature of the oxygen carrier is a crucial matter, because a chemical looping process requires it to possess a certain redox activity and be able to withstand multiple cycles without large loss in the integrity of physical and chemical properties. Ideally, the oxygen carrier should function with reproducible kinetics at high temperature, and the materials that only show reactivity near or even above their melting point should be eliminated [1]. Oxygen transport capacity is another important property of an oxygen carrier. It is defined as the usable oxygen in the oxygen carrier during one redox cycle, which determines the fuel conversion and the circulation rate [2,47]. If chemical looping process is performed in a fluidized bed, the oxygen carriers additionally need to have sufficient mechanical strength to avoid attrition, which is quite common during their lifetime. In this sense, the attrition behavior of materials should be evaluated according to standard crushing tests under chemical looping operation conditions [39]. When carbonaceous and sulfur-containing fuels (fossil fuels and biomass) are used as feedstocks in chemical looping, issues like coke deposition, sulfur poisoning and high operation temperature inevitably result in deactivation of the oxygen carrier. Hence, properties like coke resistance, sulfur tolerance and thermal stability are of great importance for developing a proper oxygen carrier. In addition, the effect of ash formation on the performance of an oxygen carrier cannot be ignored, particularly in the utilization of biomass. Ash with low melting point alkali metals such as Na or K will promote agglomeration of the oxygen carrier [48,49], while high-acid ash (e.g., SiO 2 -rich ash) leads to the production of potassium silicates that facilitate particle sintering, suppressing the fuel conversion and CO 2 capture efficiency [50]. Therefore, the resistance of an oxygen carrier to agglomeration is also a critical factor for its practical application. Eventually, the reactivity of oxygen carrier includes not only the theoretical conversion of fuel, relying on the thermodynamic properties of the materials, but also transport and reaction kinetics, which depend on composition, structure as well as texture properties like particle size, porosity and specific surface area [1].
Among various transition metals, Ni, Fe, Cu, Mn and Co-based materials are commonly considered to serve as oxygen carriers, although they present some disadvantages: Low reactivity and Despite of the toxicity and high cost, Ni-based material still receives major attention as a reliable reforming catalyst due to its high catalytic activity with CH4 and other light hydrocarbons [51]. Almost complete CH4 conversion can be obtained over Ni-based oxygen carriers during a CLR process. However, particle sintering and the loss of metallic Ni, owing to partial re-oxidation into NiO, along with the repetition of cycles cause a significant loss of catalytic performance [52]. In addition, Ni-based materials are sensitive to deactivation by sulfur poisoning and coke deposition. Improved Ni-based materials are necessary to be designed to tackle these drawbacks for further application.
Fe-based materials are commonly known as a good option for oxygen carriers due to their low price, high mechanical strength and environmentally friendly nature [39]. They almost have no tendency for carbon or sulfur deposition [53], making them promising candidates in all chemical looping processes involving carbonaceous and sulfur-containing fuels. Further, iron oxide presents the highest oxygen storage capacity from CO2 (0.7 mol CO2/mol Fe) over a broad range of operating temperatures (600~1800 °C) [22]. Thus, chemical looping dry reforming over Fe-based oxygen carriers has been one of the target technologies for CO2 utilization. However, within the multiple oxidized states of Fe system, only the transformation from Fe2O3 to Fe3O4 is suitable for CLC process due to thermodynamic limitations, leading to low oxygen transport capacity of Fe-based oxygen carriers. Other drawbacks also impede the application of Fe-based oxygen carriers, including relatively low reactivity towards gaseous fuels and the agglomeration related to magnetite formation [54]. The development of new-generation Fe-based oxygen carriers should focus on solving these issues.
The flexible redox behavior between the oxidized (CuO) and reduced (Cu or Cu2O) forms of Cu ensures that Cu-based materials have both high reactivity and oxygen transfer capacity. The Despite of the toxicity and high cost, Ni-based material still receives major attention as a reliable reforming catalyst due to its high catalytic activity with CH 4 and other light hydrocarbons [51]. Almost complete CH 4 conversion can be obtained over Ni-based oxygen carriers during a CLR process. However, particle sintering and the loss of metallic Ni, owing to partial re-oxidation into NiO, along with the repetition of cycles cause a significant loss of catalytic performance [52]. In addition, Ni-based materials are sensitive to deactivation by sulfur poisoning and coke deposition. Improved Ni-based materials are necessary to be designed to tackle these drawbacks for further application.
Fe-based materials are commonly known as a good option for oxygen carriers due to their low price, high mechanical strength and environmentally friendly nature [39]. They almost have no tendency for carbon or sulfur deposition [53], making them promising candidates in all chemical looping processes involving carbonaceous and sulfur-containing fuels. Further, iron oxide presents the highest oxygen storage capacity from CO 2 (0.7 mol CO 2 /mol Fe) over a broad range of operating temperatures (600~1800 • C) [22]. Thus, chemical looping dry reforming over Fe-based oxygen carriers has been one of the target technologies for CO 2 utilization. However, within the multiple oxidized states of Fe system, only the transformation from Fe 2 O 3 to Fe 3 O 4 is suitable for CLC process due to thermodynamic limitations, leading to low oxygen transport capacity of Fe-based oxygen carriers. Other drawbacks also impede the application of Fe-based oxygen carriers, including relatively low reactivity towards gaseous fuels and the agglomeration related to magnetite formation [54]. The development of new-generation Fe-based oxygen carriers should focus on solving these issues.
The flexible redox behavior between the oxidized (CuO) and reduced (Cu or Cu 2 O) forms of Cu ensures that Cu-based materials have both high reactivity and oxygen transfer capacity. The advantages of Cu also include relatively low cost and toxicity, so Cu-based oxygen carriers have received greater attention from both industry and academia. Moreover, the negative influence of sulfur impurities on the performance of Cu-based oxygen carriers during a chemical looping process is not significant when the oxygen carrier-to-fuel ratio lies above 1.5 [53]. There are two main disadvantages that limit the application of Cu-based materials as oxygen carrier: One is attrition owing to their weak mechanical strength, another is agglomeration as a result of the low melting point of Cu. The former can be relieved through stabilizing Cu by inert promoters, such as Al 2 O 3 , ZrO 2 , TiO 2 and SiO 2 [55]. In order to solve the melting problem, keeping the operation temperature lower than 800 • C could be effective [39].
Similar to Fe, Mn has various oxidation states including MnO 2 , Mn 2 O 3 , Mn 3 O 4 and MnO, among which Mn 3 O 4 is the only oxide phase of Mn that exists at a temperature above 800 • C. Hence, only the redox transformation between Mn 3 O 4 and MnO is applicable for the CLC process, showing a relatively low oxygen transport capacity. Additionally, Mn oxides were found to show low reactivity towards CH 4 . Similar to Ni-based materials, deactivation of Mn-based materials caused by sulfur contaminants cannot be ignored [56,57]. Nevertheless, Mn-based materials are still accepted to serve as oxygen carriers because of their low toxicity and cheap price. Some related research indicates that Mn promoted with typical materials like SiO 2 , Al 2 O 3 , TiO 2 , ZrO 2 and MgAl 2 O 4 , can interact to irreversibly form highly unreactive phases [58,59], making them unsuitable as oxygen carriers. However, Mn-based materials using dopant-stabilized ZrO 2 (applicable dopants such as MgO, CaO, Y 2 O 3 , La 2 O 3 and CeO 2 [60]) as support proved to possess not only high reactivity but also high structural stability and agglomeration resistance during redox cycles, which is prospective as promising Mn-based oxygen carriers [61].
Owing to the high reactivity and oxygen transport capacity, Co-based materials have been increasingly considered as good oxygen carriers, even though they also suffer from cost and environmental concerns. As the highest oxidation state of Co, Co 3 O 4 is thermodynamically unstable above 900 • C and easily transformed into CoO. In this respect, the redox transformation between Co and CoO is commonly adopted in chemical looping processes [2]. However, CoO shows strong interaction with common promoter materials such as Al 2 O 3 , MgO and TiO 2 , forming unreactive phases like CoAl 2 O 4 , Mg 0.4 Co 0.6 O and CoTiO 3 , respectively. As a result, these oxygen carriers generally undergo fast deactivation during repeated cycles [62,63]. In contrast, the Co-based oxygen carrier with YSZ (yttria-stabilized zirconia) as support was found to exhibit good reactivity and high resistance to carbon formation in CLC process [62].
In summary, all of the above mentioned factors should be considered and balanced based on the involved feedstock and the desired chemical looping process when selecting an oxygen carrier. The total cost of the typical metal-based oxygen carriers follows the order: Co > Ni > Cu > Fe > Mn-based material (Figure 3), which includes the cost of active materials, promoter materials and the preparation section [2]. Overall, Fe-and Ni-based oxygen carriers are the most popular carriers for chemical looping, particularly for carbonaceous fuels-driven processes, because Fe oxides hardly suffer from carbon deposition and sulfur poisoning, while Ni shows excellent reactivity towards hydrocarbon fuels, comparable with the activity of noble metals (Pt, Pd and Rh).

Improvements to Oxygen Carriers
To meet the requirements of industrial chemical looping application, further improvements should be made to oxygen carriers to tackle their shortcomings. There are three main pathways: (1) the introduction of promoter and/or support; (2) the mixing of active metal oxides to integrate various advantages; and (3) the activation and stabilization of oxygen carriers by design of special structures. Here, we focus on reporting the advances related to improved Ni-and Fe-based oxygen carriers. The major improvement approaches, based on the promoters used, the material structure and the applied chemical looping processes, are summarized in Table 1.

Monometallic Materials
The promoter and support of monometallic materials can be classified as physically or chemically active based on their functions in oxygen carrier. The former usually serve as a physical barrier to improve thermal stability, while the latter contribute their redox properties to enhance reactivity [127].
Regarding to Ni-based oxygen carriers, series of heat-resistant oxides such as Al 2 O 3 [64][65][66], SiO 2 [58,59], TiO 2 [67,68], ZrO 2 and stabilized ZrO 2 [69][70][71], as well as spinels like NiAl 2 O 4 [72,73], MgAl 2 O 4 and CaAl 2 O 4 [74][75][76] are usually considered as adequate promoter materials. Among these, NiO/Al 2 O 3 as an oxygen carrier demonstrates excellent reactivity, high thermal stability and strong resistance to carbon formation during chemical looping operation. However, the formation of an inactive NiAl 2 O 4 spinel, owing to the strong interaction between NiO and Al 2 O 3 , leads to a partial loss of reactivity. As a result of NiAl 2 O 4 formation, an excess of Ni should be added during material preparation to retain free NiO in the oxygen carrier. Accordingly, the NiO/NiAl 2 O 4 oxygen carrier still shows high reactivity, although now a larger amount of Ni is required. Alternatively, the chemical passivation (i.e., the spinel formation behavior) can be used to enhance the inert function of Al 2 O 3 promoter. Addition of MgO or CaO during the preparation of the NiO/Al 2 O 3 oxygen carrier will form a more stable spinel like MgAl 2 O 4 or CaAl 2 O 4 . The presence of such spinel in the oxygen carrier effectively suppresses the formation of NiAl 2 O 4 , maintaining the high reactivity of NiO during redox cycles and eliminating the need for more Ni. Generally, the use of other promoter materials shows problems with reactivity and mechanical strength: The use of TiO 2 for Ni-based oxygen carrier was found to show slow reaction rate or even absence of reaction, because of the formation of the stable phase NiTiO 3 ; the NiO/SiO 2 oxygen carrier gives much lower reactivity, poor mechanical strength and deactivation with progress of cycling; the NiO/ZrO 2 or NiO/stabilized ZrO 2 oxygen carriers present good reactivity but their mechanical strength is weak.
The use of a reducible oxide, particularly CeO 2 , as support for a Ni-based oxygen carrier, presents an obvious potential for enhanced redox properties of the oxygen carrier through a synergistic effect between Ni and CeO 2 [77]. This synergistic effect may increase the utilization of Ni species due to the enhanced oxygen transfer from the CeO 2 support to the Ni particles via the Ni-CeO 2 interface, as shown in Figure 4: After the initial part of the reduction half-cycle, a Ni particle consists of two phases, i.e., metallic Ni (the outer layer of the particle) and NiO (the core of the particle). At the same time, CH 4 is cracked on the surface of metallic Ni to form Ni-bound carbonaceous species, which afterwards can be gasified by lattice oxygen derived from nearby CeO 2 . The lost lattice oxygen in the support is refilled by oxygen from the core of the Ni particle (where the NiO phase is present), resulting in the continued reduction of the Ni particle.

Monometallic Materials
The promoter and support of monometallic materials can be classified as physically or chemically active based on their functions in oxygen carrier. The former usually serve as a physical barrier to improve thermal stability, while the latter contribute their redox properties to enhance reactivity [127].
Regarding to Ni-based oxygen carriers, series of heat-resistant oxides such as Al2O3 [64][65][66], SiO2 [58,59], TiO2 [67,68], ZrO2 and stabilized ZrO2 [69][70][71], as well as spinels like NiAl2O4 [72,73], MgAl2O4 and CaAl2O4 [74][75][76] are usually considered as adequate promoter materials. Among these, NiO/Al2O3 as an oxygen carrier demonstrates excellent reactivity, high thermal stability and strong resistance to carbon formation during chemical looping operation. However, the formation of an inactive NiAl2O4 spinel, owing to the strong interaction between NiO and Al2O3, leads to a partial loss of reactivity. As a result of NiAl2O4 formation, an excess of Ni should be added during material preparation to retain free NiO in the oxygen carrier. Accordingly, the NiO/NiAl2O4 oxygen carrier still shows high reactivity, although now a larger amount of Ni is required. Alternatively, the chemical passivation (i.e., the spinel formation behavior) can be used to enhance the inert function of Al2O3 promoter. Addition of MgO or CaO during the preparation of the NiO/Al2O3 oxygen carrier will form a more stable spinel like MgAl2O4 or CaAl2O4. The presence of such spinel in the oxygen carrier effectively suppresses the formation of NiAl2O4, maintaining the high reactivity of NiO during redox cycles and eliminating the need for more Ni. Generally, the use of other promoter materials shows problems with reactivity and mechanical strength: The use of TiO2 for Ni-based oxygen carrier was found to show slow reaction rate or even absence of reaction, because of the formation of the stable phase NiTiO3; the NiO/SiO2 oxygen carrier gives much lower reactivity, poor mechanical strength and deactivation with progress of cycling; the NiO/ZrO2 or NiO/stabilized ZrO2 oxygen carriers present good reactivity but their mechanical strength is weak.
The use of a reducible oxide, particularly CeO2, as support for a Ni-based oxygen carrier, presents an obvious potential for enhanced redox properties of the oxygen carrier through a synergistic effect between Ni and CeO2 [77]. This synergistic effect may increase the utilization of Ni species due to the enhanced oxygen transfer from the CeO2 support to the Ni particles via the Ni-CeO2 interface, as shown in Figure 4: After the initial part of the reduction half-cycle, a Ni particle consists of two phases, i.e., metallic Ni (the outer layer of the particle) and NiO (the core of the particle). At the same time, CH4 is cracked on the surface of metallic Ni to form Ni-bound carbonaceous species, which afterwards can be gasified by lattice oxygen derived from nearby CeO2. The lost lattice oxygen in the support is refilled by oxygen from the core of the Ni particle (where the NiO phase is present), resulting in the continued reduction of the Ni particle.  Although Ni/CeO 2 oxygen carrier shows high reactivity towards CH 4 , high carrier utilization and high resistance to carbon deposition in CLC process, a significant loss in surface area owing to sintering of CeO 2 over repeated cycles indicates that the CeO 2 support must be stabilized to make it applicable for industry. It has been confirmed that doping CeO 2 with metals of different valency (such as La and Zr) can not only increase the thermal stability of CeO 2 by maintaining the high surface area and preserving oxygen vacancies, but also further enhance the reducibility by increasing the oxygen mobility [128,129].
Regarding Fe-based oxygen carriers, promoter materials such as Al 2 O 3 [78][79][80][81][82], MgO [83,130], TiO 2 [84][85][86], SiO 2 [87], MgAl 2 O 4 [88][89][90][91]103], CaO [92,93], ZrO 2 [94][95][96] and CeO 2 [22,[97][98][99][100] have received extensive scientific attention in the past decades. However, these promoters can easily lead to solid-solid transformation with iron oxides during a continuous cycling process [127]. Depending on the specific promoter, new phases like AlFe 2 O 4 , MgFe 2 O 4 , FeTiO 3 , and Fe 2 SiO 4 , Ca 2 Fe 2 O 5 and CeFeO 3 have a strong tendency to develop during the redox reactions. The so formed metal-ferrites can continue to serve as a physical barrier to prevent sintering of the iron oxide, but the overall oxygen carrier becomes less reactive in the presence of these phases, since the regeneration of the active iron oxides from the aforementioned metal-ferrites requires much stronger reducing and oxidizing conditions. As a result, the oxygen storage capacity decreases significantly [94,120]. Especially metal-ferrites such as Fe 2 SiO 4 and Ca 2 Fe 2 O 5 are unreactive and hard to decompose, and they usually cause a substantial decrease in the oxygen carrier reactivity and selectivity. Because the MgAl 2 O 4 spinel forms more easily, the addition of an appropriate amount of Mg (the atomic ratio of Mg:Al = 1:2) to Fe 2 O 3 /Al 2 O 3 could successfully alleviate the formation of FeAl 2 O 4 . Further, MgAl 2 O 4 itself is also favored to act as an oxygen carrier support owing to its high thermal and chemical stability, as well as to its specific heat capacity. While using Fe 2 O 3 /MgAl 2 O 4 as an oxygen carrier in a CLR process, an increase in activity and stability with the number of cycles was found. Marked by a strong inertness, the use of ZrO 2 as a promoter has received a great deal of attention. An extensive study on the Fe 2 O 3 /ZrO 2 oxygen carrier for chemical looping indicated that the physical structure of ZrO 2 was relatively stable during cycling and that no new phase formed [94]. Additionally, it was observed that the use of Ca-, Mg-and Ce-stabilized ZrO 2 as supports for Fe-based oxygen carriers yields a further enhancement of redox properties and thermal stability [54]. CeO 2 is an ideal candidate for promoting iron oxides in a chemical looping process, not only as a physical promoter, but also as a chemically active promoter due to its high activity towards methane or syngas oxidation by lattice oxygen, as well as its reasonable CO 2 and H 2 O re-oxidation capacity [22,127]. The redox couple Ce 4+ and Ce 3+ facilitates oxygen storage and release from its bulk fluorite lattice. When the Ce 4+ ions are replaced by lower valence cations (such as Fe 3+ ), oxygen vacancies are created, which are considered to be active sites. After consuming the available surface oxygen for oxidation, extra oxygen can be transferred from bulk to surface more rapidly in a CeO 2 -Fe 2 O 3 solid solution than in pure Fe 2 O 3 . The fast oxygen transfer from the bulk of Fe 2 O 3 to its surface is enabled by the creation of oxygen vacancies owing to the CeO 2 additive [1]. Therefore, the addition of CeO 2 in Fe-based oxygen carriers enhances the reducibility of Fe 2 O 3 by means of CeO 2 -Fe 2 O 3 interaction. Machida et al. [98] proposed the concept that CeO 2 and Fe 2 O 3 can respectively serve as an oxygen gateway and an oxygen reservoir in a combined system, opening up a new type of Ce 2 O 3 -grafted Fe 2 O 3 oxygen carrier that is different from the conventional Fe 2 O 3 /Ce 2 O 3 solid solution materials. To validate this idea, they grafted CeO 2 nanoparticles onto the surface of

Mixed Oxide Materials
It is well accepted that complex metal oxides may provide better properties than the individual metal oxides. Therefore, oxygen carriers based on mixed oxides have been investigated by combining different active metal oxides in a single particle or mixing different oxygen carriers, each composed of a single metal oxide. The main objective is to combine the complementary advantages of different oxygen carriers for the practical application in a chemical looping process.
The CoO-NiO supported YSZ oxygen carrier was found to have high reactivity towards CH4, complete absence of coke formation and good regenerative ability in natural gas-fueled CLC, even though the formation of a NiCoO2 solid solution decreased the redox reaction rate to some extent [101]. Hossain et al. [102] reported that the addition of Co into a Ni/Al2O3 material not only enhanced the reducibility of the oxygen carrier due to the suppression of NiAl2O4 formation, but also inhibited the agglomeration of particles by maintaining consistent Ni dispersion throughout the redox cycles.
The incorporation of Fe into Cu-based oxygen carriers can relieve their agglomeration problem, owing to the high melting point of Fe2O3 [131]. Wang et al. [103] evaluated the performance of series of Fe2O3-CuO/MgAl2O4 oxygen carriers for coke-oven-gas-based CLC in a laboratory pressurized circulating fluidized bed system. The best working oxygen carrier was found to be Fe2O3(45)-CuO(15)/MgAl2O4 (40), which showed high reactivity and circulation stability. It was also found that the introduction of a NiO-based oxygen carrier into the Fe2O3/CuO-based oxygen carrier bed facilitated the reduction of the operating temperature in CLC process, due to the high reactivity of Ni towards CH4. Further, a layer-by-layer mixing mode of a NiO-based oxygen carrier and a Fe2O3/CuO-based oxygen carrier in a fixed-bed reactor avoids the formation of NiFe2O4 spinel during the reduction half-cycle, ensuring a good redox rate, thereby providing a promising reactor configuration candidate for an enhanced-CH4-reforming chemical looping process [132].
The bimetallic Fe-Mn-O system shows higher CH4 combustion (close to 100%) in the absence of oxygen or air, higher CO2 selectivity (almost no toxic CO byproduct) and more stable multicycle CLC performance, compared to the individual Fe or Mn oxides. Additionally, owing to the low cost of both Fe and Mn, the bimetallic Fe-Mn oxide can be considered as a very competitive oxygen carrier for the chemical looping process [104]. Moreover, both Fe-Mn-O and Ni-Mn-O systems have oxygen uncoupling properties, as a result, FexMn(1−x)O and NiMn2O4 can be used as oxygen carriers for the CLOU process [105][106][107].

Mixed Oxide Materials
It is well accepted that complex metal oxides may provide better properties than the individual metal oxides. Therefore, oxygen carriers based on mixed oxides have been investigated by combining different active metal oxides in a single particle or mixing different oxygen carriers, each composed of a single metal oxide. The main objective is to combine the complementary advantages of different oxygen carriers for the practical application in a chemical looping process.
The CoO-NiO supported YSZ oxygen carrier was found to have high reactivity towards CH 4 , complete absence of coke formation and good regenerative ability in natural gas-fueled CLC, even though the formation of a NiCoO 2 solid solution decreased the redox reaction rate to some extent [101]. Hossain et al. [102] reported that the addition of Co into a Ni/Al 2 O 3 material not only enhanced the reducibility of the oxygen carrier due to the suppression of NiAl 2 O 4 formation, but also inhibited the agglomeration of particles by maintaining consistent Ni dispersion throughout the redox cycles.
The incorporation of Fe into Cu-based oxygen carriers can relieve their agglomeration problem, owing to the high melting point of Fe 2 O 3 [131]. Wang et al. [103] evaluated the performance of series of Fe 2 O 3 -CuO/MgAl 2 O 4 oxygen carriers for coke-oven-gas-based CLC in a laboratory pressurized circulating fluidized bed system. The best working oxygen carrier was found to be Fe 2 O 3(45) -CuO (15) /MgAl 2 O 4 (40) , which showed high reactivity and circulation stability. It was also found that the introduction of a NiO-based oxygen carrier into the Fe 2 O 3 /CuO-based oxygen carrier bed facilitated the reduction of the operating temperature in CLC process, due to the high reactivity of Ni towards CH 4 . Further, a layer-by-layer mixing mode of a NiO-based oxygen carrier and a Fe 2 O 3 /CuO-based oxygen carrier in a fixed-bed reactor avoids the formation of NiFe 2 O 4 spinel during the reduction half-cycle, ensuring a good redox rate, thereby providing a promising reactor configuration candidate for an enhanced-CH 4 -reforming chemical looping process [132].
The bimetallic Fe-Mn-O system shows higher CH 4 combustion (close to 100%) in the absence of oxygen or air, higher CO 2 selectivity (almost no toxic CO byproduct) and more stable multicycle CLC performance, compared to the individual Fe or Mn oxides. Additionally, owing to the low cost of both Fe and Mn, the bimetallic Fe-Mn oxide can be considered as a very competitive oxygen carrier for the chemical looping process [104]. Moreover, both Fe-Mn-O and Ni-Mn-O systems have oxygen uncoupling properties, as a result, Fe x Mn (1−x) O and NiMn 2 O 4 can be used as oxygen carriers for the CLOU process [105][106][107].
Continuing the comparison of Ni-and Fe-based oxygen carriers in Section 2.2: Ni shows best activity in CH 4 conversion, but low resistance to carbon formation, whilst Fe has almost no trend to carbon deposition, with high oxygen storage capacity from CO 2 , but low reactivity towards CH 4 . Due to its low price and non-toxicity, Fe is advantageous from an economic and environmental point of view compared to Ni. Aiming to combine the complementary advantages of these two constituent metals, many efforts have been made to develop bimetallic Fe-Ni oxygen carriers via two different main strategies, physical [108,133] and chemical [23,[108][109][110][111][112] mixing. Johansson et al. [133] reported that the addition of only 3 wt % NiO in the bed of a Fe-based oxygen carrier was sufficient to reach a very high CH 4 conversion in CLC process. This is probably due to a synergy effect, i.e., Ni catalyzes the reforming of CH 4 to CO and H 2 , which then both react with iron oxide at a considerably higher rate than CH 4 . In the case of chemically mixed oxygen carriers, in which the NiFe 2 O 4 spinel usually forms, the CH 4 conversion and reaction rate in CLC process follow the ranking: NiO > NiFe 2 O 4 > Fe 2 O 3 [109], further indicating that alloying a less reactive metal-Fe-with a highly reactive one-Ni-can significantly improve the activity of the resulting bimetallic oxygen carrier. Compared to pure NiO and Fe 2 O 3 , NiFe 2 O 4 is more suitable to serve as oxygen carrier for the CLR process, because the limited oxygen transfer capacity of NiFe 2 O 4 avoids the complete oxidation of CH 4 , which results in an increased selectivity to syngas [112]. Alternatively, the addition of small amount of Fe into a Ni-based oxygen carrier or catalyst can obviously enhance the carbon resistance. Theofanidis et al. [51] prepared a series of bimetallic Fe-Ni/MgAl 2 O 4 catalysts with different Fe/Ni molar ratios (from 0 to 1.5) to investigate the role of Fe on enhanced carbon resistance during CH 4 dry reforming. They found that the process of dry reforming over bimetallic Fe-Ni catalyst follows the Mars-van Krevelen mechanism, where CO 2 oxidizes Fe to iron oxide and CH 4 is decomposed on Ni sites to H 2 and surface carbon. The latter, simultaneously, can be oxidized by lattice oxygen from iron oxide, yielding CO (Figure 6a). During dry reforming, a Fe-Ni alloy is formed, which still constitutes an active phase and can be regenerated to metallic Ni and Fe 3 O 4 by re-oxidizing by steam or CO 2 . Compared to monometallic Ni/MgAl 2 O 4 , the carbon deposition on the bimetallic Fe-Ni/MgAl 2 O 4 is significantly suppressed (Figure 6b).

Perovskite Structure
Oxygen carriers with perovskite structure are receiving increasing attention owing to their enhanced redox properties, high oxygen mobility and thermal stability. Perovskite-structured oxide materials are typically formulated as ABO 3 , where A is a large alkaline earth or rare earth element and B is a smaller transition metal cation [134]. Where the A-site cation is responsible for the thermal resistance, the B-site cation will be the contributor to the reactivity, e.g., the catalytic activity in chemical looping reforming of CH 4 . As a result, the selection of B-site cations is crucial in the design of promising perovskite-structured oxygen carriers. In addition, the interaction between the A and B cations can affect the reactivity and selectivity during fuels conversion as well as the stability of the structure [1].
As well-established transition metal for oxygen carriers, Fe 3+ has certainly been considered as a B-site cation. The performance of AFeO 3 (A = La, Nd and Eu) was first investigated for selective oxidation of CH 4 by Dai et al. [135]. It was found that, among the three AFeO 3 candidates, LaFeO 3 showed the best performance with respect to syngas production and stability during the redox cycling between CH 4 and air at 900 • C. Because of the kinetic limitations, the reduction of LaFeO 3 by CH 4 is basically performed from Fe 3+ to Fe 2+ , while deeper reduction is very difficult to achieve [113]. Further, partial substitution of the A-site cation by a low oxidation-state metal such as Sr leads to the formation of oxygen vacancies and a fraction of B-site cations in a higher valence state. A highly variable oxygen non-stoichiometry in the perovskite structure is hence introduced that can readily provide lattice oxygen, while subsequently the removed oxygen species can be replenished by gaseous oxygen. Substitution of La 3+ by Sr 2+ could cause electronic imbalance, but it can be compensated by the transformation of a fraction of Fe 3+ to Fe 4+ . Therefore, La 1−x Sr x FeO 3 (x = 0 to~1) perovskite oxides have been widely investigated as oxygen carriers in CLR processes [114,115,136]. Li et al. [136] proposed that there are two kinds of oxygen species on such perovskite oxides: One is the weakly-bound oxygen species which is active in complete oxidation of CH 4 to CO 2 and H 2 O, another is strongly-bound oxygen which is responsible for partial oxidation of CH 4 to CO and H 2 . The activity of such perovskite materials was found to show a decrease when x increased above 0.5, probably owing to the suppression of CH 4 decomposition by the Sr substitution [115]. The best performance was presented by La 0.7 Sr 0.3 FeO 3 , and during the reduction half-cycle of chemical looping steam reforming a maximum H 2 yield was obtained over a simple mechanical mixture of La 0.7 Sr 0.3 FeO 3 with an additional 5 wt % NiO [114]. On the other hand, the partial substitution of active B-site cations by aliovalent transition metal cations can lead to the enhancement of catalytic activity due to synergistic valence changes and introduction of non-stoichiometry-related microstructural defects into the lattice. Lori et al. [116] modified the "base case" material, La 0.7 Sr 0.3 FeO 3 , by transition metals such as Ni, Co, Cu and Cr. All the resulting perovskite-structured materials, with the general formula La 0.7 Sr 0.3 M y Fe 1−y O 3 (M is the additional transition metal), show the capability to act as suitable oxygen carriers in CLR process. The modification with partial substitution of Fe by Cr shows the best improvement compared to others. Additionally, La 0.7 Sr 0.3 Cr 0.1 Fe 0.9 O 3 was found to have the optimum composition, showing highest CH 4 conversion and syngas production, zero carbon deposition as well as good redox properties.
The application of CH 4 -based CLR over the perovskite-structured oxygen carriers, that provide lattice oxygen for CH 4 conversion, represents an alternative in terms of high selectivity to syngas and elimination of the air separation process. However, the oxygen storage capacity of perovskite-structured metal oxides is relatively low compared to traditional metal oxides, and the oxygen transport capacity of the perovskites is less due to its slow reduction rate. Both disadvantages limit the application of the perovskite-structured oxide materials as industrial oxygen carriers.

Core-Shell Structure
Fast redox kinetics and hence high oxygen transport rates of oxygen carriers are considered to be obstructed by the solid-state diffusion limitation due to the formation of a dense oxide overlayer during oxidation [137]. The transition from a kinetically controlled regime to a diffusion limited regime occurs when the overlayer reaches a critical thickness. Hence, an oxygen carrier with smaller particle size (i.e., nano-sized oxygen carrier) is expected to suffer less from such diffusion limitations by shifting the transition point towards close to complete oxygen carrier conversion, because, from a geometrical point of view, the fraction of the total particle volume constituted by a layer with constant thickness strongly increases with decreasing particle diameter [14]. Moreover, owing to the relatively slower oxidation kinetics with CO 2 compared to air or steam, the use of nano-sized metal particles is expected to maximize the reactivity when CO 2 is used as oxidizing agent [120]. In addition, it is believed that carbon deposition is not energetically favored over small metal particles, especially for Ni. As a consequence, the use of nano-sized Ni as oxygen carrier is very beneficial to chemical looping dry reforming of CH 4 [138]. However, the fast deactivation of nanoparticles, due to severe sintering during high-temperature processes like chemical looping, poses a big issue for practical application. To counter this, oxygen carriers with a core-shell structure are designed, allowing stabilization of nano-size active metals by a thin and porous encapsulating shell (i.e., metal core @oxide shell nanomaterial). This approach provides a promising strategy, not only to address the stability requirement, but also to improve the performance in terms of reactivity, redox properties and coke resistance, depending on the employed shell materials. Such core-shell structured nanomaterials have shown excellent performance in methane reforming [117,118,[138][139][140] and chemical looping processes [14,40,[119][120][121][122][123][124]141].
The first-generation metal core @oxide shell nanomaterial of particular interest is a metal@SiO 2 core-shell material, because of its easy preparation and low cost, providing great potential to be developed as commercial material, as well as the easy adjustment of its structural properties, such as porosity, specific surface area and shell thickness through modification of the synthesis method of the SiO 2 shell [117,119]. Ni@SiO 2 core-shell materials have received broad interest for CH 4 reforming, as they turned out to be good redox catalysts or oxygen carriers, owing to their high activity and stability as well as their potential to resist coke deposition. Fe 2 O 3 @SiO 2 core-shell materials (Figure 7a) however, show poor utilization of the oxygen carrier due to the formation of Fe 2 SiO 4 during the reduction half-cycle. The active Fe cannot easily be restored from this Fe 2 SiO 4 phase in the CO 2 re-oxidation half-cycle, resulting in deactivation of material along with the proceeding cycles. To counter this deactivation, an additional air oxidation step is required to fully liberate active Fe from the Fe-silicate (Figure 7b), which leads to an increasing cost with process complexity [14,120]. during oxidation [137]. The transition from a kinetically controlled regime to a diffusion limited regime occurs when the overlayer reaches a critical thickness. Hence, an oxygen carrier with smaller particle size (i.e., nano-sized oxygen carrier) is expected to suffer less from such diffusion limitations by shifting the transition point towards close to complete oxygen carrier conversion, because, from a geometrical point of view, the fraction of the total particle volume constituted by a layer with constant thickness strongly increases with decreasing particle diameter [14]. Moreover, owing to the relatively slower oxidation kinetics with CO2 compared to air or steam, the use of nano-sized metal particles is expected to maximize the reactivity when CO2 is used as oxidizing agent [120]. In addition, it is believed that carbon deposition is not energetically favored over small metal particles, especially for Ni. As a consequence, the use of nano-sized Ni as oxygen carrier is very beneficial to chemical looping dry reforming of CH4 [138]. However, the fast deactivation of nanoparticles, due to severe sintering during high-temperature processes like chemical looping, poses a big issue for practical application.
To counter this, oxygen carriers with a core-shell structure are designed, allowing stabilization of nano-size active metals by a thin and porous encapsulating shell (i.e., metalcore@oxideshell nanomaterial). This approach provides a promising strategy, not only to address the stability requirement, but also to improve the performance in terms of reactivity, redox properties and coke resistance, depending on the employed shell materials. Such core-shell structured nanomaterials have shown excellent performance in methane reforming [117,118,[138][139][140] and chemical looping processes [14,40,[119][120][121][122][123][124]141]. The first-generation metalcore@oxideshell nanomaterial of particular interest is a metal@SiO2 coreshell material, because of its easy preparation and low cost, providing great potential to be developed as commercial material, as well as the easy adjustment of its structural properties, such as porosity, specific surface area and shell thickness through modification of the synthesis method of the SiO2 shell [117,119]. Ni@SiO2 core-shell materials have received broad interest for CH4 reforming, as they turned out to be good redox catalysts or oxygen carriers, owing to their high activity and stability as well as their potential to resist coke deposition. Fe2O3@SiO2 core-shell materials (Figure 7a) however, show poor utilization of the oxygen carrier due to the formation of Fe2SiO4 during the reduction halfcycle. The active Fe cannot easily be restored from this Fe2SiO4 phase in the CO2 re-oxidation halfcycle, resulting in deactivation of material along with the proceeding cycles. To counter this deactivation, an additional air oxidation step is required to fully liberate active Fe from the Fe-silicate (Figure 7b), which leads to an increasing cost with process complexity [14,120]. Although the above three-stage chemical looping dry reforming process allows sufficient utilization of the oxygen carrier via regaining Fe2O3 from the Fe2SiO4 phase in the air reactor, the low CO2 conversion in the CO2 reactor still limits the process efficiency in terms of CO2 utilization. Therefore, there is need to further explore reproducible core-shell structures. Li's group designed a Figure 7. TEM image of a core-shell structured Fe 2 O 3 @SiO 2 oxygen carrier (a) and schematic representation of a three-stage chemical looping dry reforming process for CO 2 utilization (b). Reproduced with permission from Bhavsar S. et al. [14], Catalysis Today; published by Elsevier, 2014.
Although the above three-stage chemical looping dry reforming process allows sufficient utilization of the oxygen carrier via regaining Fe 2 O 3 from the Fe 2 SiO 4 phase in the air reactor, the low CO 2 conversion in the CO 2 reactor still limits the process efficiency in terms of CO 2 utilization. Therefore, there is need to further explore reproducible core-shell structures. Li's group designed a core-shell Fe 2 O 3 @La x Sr 1−x FeO 3−δ material (Fe 2 O 3 @LSF), composed of a nanoscale Fe 2 O 3 core (~50 nm) and a perovskite-structured La x Sr 1−x FeO 3−δ shell (thickness of~10 nm), which was successfully synthesized by using a sequential sol-gel method [121]. The performance of such material was investigated in CH 4 partial oxidation: In this well-defined structure (Figure 8), syngas production from CH 4 is enhanced by the highly selective LSF shell; the activity is improved with nano-size Fe 2 O 3 which can provide more readily accessible lattice oxygen; the coating of Fe 2 O 3 with a stable perovskite-structured shell does not hinder the recyclability of the material, but rather prevents Fe 2 O 3 from sintering and carbon deposition, while improving the oxygen ion and electron fluxes. Thus, the proposed Fe 2 O 3 @LSF core-shell material was suggested to be a promising oxygen carrier for CLR processes. Later on, this group investigated the effect of core and shell compositions of MeO x @LSF (Me = Mn, Co and Fe) materials on the performance for CLR [122]. They found that the presence of Sr in the shell is favorable to stabilize Fe in the core-shell structure (i.e., prevents irreversible Fe leaching to the surface) by forming strontium hexaferrite (SrFe 12 O 19 ) or a Sr deficient perovskite. However, these phases are more difficult to reduce, leading to a decrease of the overall reduction rate of the Fe 2 O 3 @LSF material. As a result, its redox activity is weakened. core-shell Fe2O3@LaxSr1−xFeO3−δ material (Fe2O3@LSF), composed of a nanoscale Fe2O3 core (~50 nm) and a perovskite-structured LaxSr1−xFeO3−δ shell (thickness of ~10 nm), which was successfully synthesized by using a sequential sol-gel method [121]. The performance of such material was investigated in CH4 partial oxidation: In this well-defined structure (Figure 8), syngas production from CH4 is enhanced by the highly selective LSF shell; the activity is improved with nano-size Fe2O3 which can provide more readily accessible lattice oxygen; the coating of Fe2O3 with a stable perovskite-structured shell does not hinder the recyclability of the material, but rather prevents Fe2O3 from sintering and carbon deposition, while improving the oxygen ion and electron fluxes. Thus, the proposed Fe2O3@LSF core-shell material was suggested to be a promising oxygen carrier for CLR processes. Later on, this group investigated the effect of core and shell compositions of MeOx@LSF (Me = Mn, Co and Fe) materials on the performance for CLR [122]. They found that the presence of Sr in the shell is favorable to stabilize Fe in the core-shell structure (i.e., prevents irreversible Fe leaching to the surface) by forming strontium hexaferrite (SrFe12O19) or a Sr deficient perovskite. However, these phases are more difficult to reduce, leading to a decrease of the overall reduction rate of the Fe2O3@LSF material. As a result, its redox activity is weakened. More suitable shell materials with chemical stability should be considered in order to maintain the long-term activity of core-shell structured Fe-based oxygen carriers. ZrO2 is of great interest as shell material due to its strong thermal and chemical stability. Li et al. [142] proposed a Ni@ZrO2 nanocomposite, constructed from nano-Ni particles encapsulated in a nano-ZrO2 framework, showing a strong ability to resistant sintering and coke deposition under reforming conditions. ZrO2 exists in different phases, among which the t-ZrO2 phase is dominant. It easily forms in nanoparticles and ultrathin films [143], where it contributes to a high specific surface area and oxygen conductivity. In a similar way, a core-shell material Fe2O3@ZrO2, with a thin, porous ZrO2 shell surrounding a nanosized Fe2O3 core, can exhibit high reactivity and stability in chemical looping processes. The Fe2O3@ZrO2 material can be synthesized via nanocoating of Fe2O3 with ZrO2 through hydrolysis with a ZrO2 precursor [144,145]. In this process, two challenges need to be addressed: (1) the high tendency of nanoparticles to aggregate during the coating process, resulting in multiple cores within one shell [144]; (2) generally, the as-prepared Fe2O3 particles are non-densified, providing an unstable support for the shell, which easily collapses upon high-temperature reaction. To counter these drawbacks, a novel core-shell structured nanomaterial was proposed, Fe2O3/ZrO2@ZrO2 [40]. The core consists of More suitable shell materials with chemical stability should be considered in order to maintain the long-term activity of core-shell structured Fe-based oxygen carriers. ZrO 2 is of great interest as shell material due to its strong thermal and chemical stability. Li et al. [142] proposed a Ni@ZrO 2 nanocomposite, constructed from nano-Ni particles encapsulated in a nano-ZrO 2 framework, showing a strong ability to resistant sintering and coke deposition under reforming conditions. ZrO 2 exists in different phases, among which the t-ZrO 2 phase is dominant. It easily forms in nanoparticles and ultrathin films [143], where it contributes to a high specific surface area and oxygen conductivity. In a similar way, a core-shell material Fe 2 O 3 @ZrO 2 , with a thin, porous ZrO 2 shell surrounding a nano-sized Fe 2 O 3 core, can exhibit high reactivity and stability in chemical looping processes. The Fe 2 O 3 @ZrO 2 material can be synthesized via nanocoating of Fe 2 O 3 with ZrO 2 through hydrolysis with a ZrO 2 precursor [144,145]. In this process, two challenges need to be addressed: (1) the high tendency of nanoparticles to aggregate during the coating process, resulting in multiple cores within one shell [144]; (2) generally, the as-prepared Fe 2 O 3 particles are non-densified, providing an unstable support for the shell, which easily collapses upon high-temperature reaction. To counter these drawbacks, a novel core-shell structured nanomaterial was proposed, Fe 2 O 3 /ZrO 2 @ZrO 2 [40]. The core consists of Fe 2 O 3 nanoparticles decorating a stable ZrO 2 and it is coated afterwards with a porous thin ZrO 2 layer (Figure 9) . Fe2O3 nanoparticles decorating a stable ZrO2 and it is coated afterwards with a porous thin ZrO2 layer (Figure 9).
(a) (b) (c) Applying Fe2O3 nanoparticles supported on stable crystalline ZrO2 as core material is advantageous; the Fe2O3/ZrO2 core avoids nanoparticle aggregation during the coating process, while at the same time providing a dense, stable support for the shell. The core material is covered with a ZrO2 shell by means of a general nanocoating process [146,147], using P-123 (a nonionic amphiphilic surfactant) as an active template layer on the core surface, upon which the shell is formed. Eventually, the template layer is calcined to decompose, leaving a mesoporous ZrO2 shell, permeable to reduction and oxidation gases to reach the iron oxide [148]. Compared to non-coated material, the core-shell Fe2O3/ZrO2@ZrO2 nanomaterial, designed to possess a stable Fe2O3/ZrO2 core and a mesoporous t-ZrO2 shell, exhibited excellent redox activity (higher CO2 conversion to CO at 650 °C) and stability (stable CO production, thermal-stable structure and high capacity to resist particle sintering) during 100 cycles of chemical looping conversion of CO2 ( Figure 10).  Applying Fe 2 O 3 nanoparticles supported on stable crystalline ZrO 2 as core material is advantageous; the Fe 2 O 3 /ZrO 2 core avoids nanoparticle aggregation during the coating process, while at the same time providing a dense, stable support for the shell. The core material is covered with a ZrO 2 shell by means of a general nanocoating process [146,147], using P-123 (a nonionic amphiphilic surfactant) as an active template layer on the core surface, upon which the shell is formed. Eventually, the template layer is calcined to decompose, leaving a mesoporous ZrO 2 shell, permeable to reduction and oxidation gases to reach the iron oxide [148]. Compared to non-coated material, the core-shell Fe 2 O 3 /ZrO 2 @ZrO 2 nanomaterial, designed to possess a stable Fe 2 O 3 /ZrO 2 core and a mesoporous t-ZrO 2 shell, exhibited excellent redox activity (higher CO 2 conversion to CO at 650 • C) and stability (stable CO production, thermal-stable structure and high capacity to resist particle sintering) during 100 cycles of chemical looping conversion of CO 2 ( Figure 10). Applying Fe2O3 nanoparticles supported on stable crystalline ZrO2 as core material is advantageous; the Fe2O3/ZrO2 core avoids nanoparticle aggregation during the coating process, while at the same time providing a dense, stable support for the shell. The core material is covered with a ZrO2 shell by means of a general nanocoating process [146,147], using P-123 (a nonionic amphiphilic surfactant) as an active template layer on the core surface, upon which the shell is formed. Eventually, the template layer is calcined to decompose, leaving a mesoporous ZrO2 shell, permeable to reduction and oxidation gases to reach the iron oxide [148]. Compared to non-coated material, the core-shell Fe2O3/ZrO2@ZrO2 nanomaterial, designed to possess a stable Fe2O3/ZrO2 core and a mesoporous t-ZrO2 shell, exhibited excellent redox activity (higher CO2 conversion to CO at 650 °C) and stability (stable CO production, thermal-stable structure and high capacity to resist particle sintering) during 100 cycles of chemical looping conversion of CO2 ( Figure 10).

Bifunctional Materials for Catalyst-Assisted Chemical Looping
When using Fe-based oxygen carrier as looping material, the reduction rate of iron oxide by hydrocarbon fuel is very slow, especially in the presence of CO 2 (originating from feed and product). As a result, the conversion of fuel and CO 2 is limited. Through addition of a catalyst component (commonly Ni), a bifunctional material, possessing an extra catalytic function for decomposition of hydrocarbon fuel into syngas (the latter ensures the formation of metallic Fe for adequate CO 2 cyclic conversion), brings chemical looping towards a new stage-catalyst-assisted chemical looping dry (or auto-thermal dry) reforming (CCDR or CCAR), two processes which are designed for efficient conversion of hydrocarbon fuels accompanied by maximum CO 2 utilization. Figure 11 sketches the CH 4 -based catalyst-assisted chemical looping processes over a Ni-Fe bifunctional material. In the reduction half-cycle, the Ni catalyst converts CH 4 and CO 2 in syngas (a mixture of H 2 and CO), which reduces the iron oxide oxygen carrier. Reduction of iron oxide by H 2 and CO is significantly faster and deeper than by CH 4 [23,149], which emphasizes the importance of the Ni catalyzed CH 4 reforming process. Upon re-oxidation, the oxygen carrier is regenerated by feeding extra CO 2 , thereby producing CO. In such processes, the Ni-Fe bifunctional material is required to provide not only an optimum activity towards conversion of CH 4 and CO 2 into CO and H 2 , but also provide redox properties for interaction with both reducing gases (CO and H 2 ) and oxidizing gases (CO 2 and H 2 O). Therefore, the material must show functions of both reforming catalyst and oxygen carrier, i.e., exhibit a bifunctional character.

Bifunctional Materials for Catalyst-Assisted Chemical Looping
When using Fe-based oxygen carrier as looping material, the reduction rate of iron oxide by hydrocarbon fuel is very slow, especially in the presence of CO2 (originating from feed and product). As a result, the conversion of fuel and CO2 is limited. Through addition of a catalyst component (commonly Ni), a bifunctional material, possessing an extra catalytic function for decomposition of hydrocarbon fuel into syngas (the latter ensures the formation of metallic Fe for adequate CO2 cyclic conversion), brings chemical looping towards a new stage-catalyst-assisted chemical looping dry (or auto-thermal dry) reforming (CCDR or CCAR), two processes which are designed for efficient conversion of hydrocarbon fuels accompanied by maximum CO2 utilization. Figure 11 sketches the CH4-based catalyst-assisted chemical looping processes over a Ni-Fe bifunctional material. In the reduction half-cycle, the Ni catalyst converts CH4 and CO2 in syngas (a mixture of H2 and CO), which reduces the iron oxide oxygen carrier. Reduction of iron oxide by H2 and CO is significantly faster and deeper than by CH4 [23,149], which emphasizes the importance of the Ni catalyzed CH4 reforming process. Upon re-oxidation, the oxygen carrier is regenerated by feeding extra CO2, thereby producing CO. In such processes, the Ni-Fe bifunctional material is required to provide not only an optimum activity towards conversion of CH4 and CO2 into CO and H2, but also provide redox properties for interaction with both reducing gases (CO and H2) and oxidizing gases (CO2 and H2O). Therefore, the material must show functions of both reforming catalyst and oxygen carrier, i.e., exhibit a bifunctional character. In previous work [23], a bifunctional 5 wt % Ni/CeO2-Fe2O3 material was prepared by impregnation of the required amount of Ni on a 50 wt % CeO2-Fe2O3 oxygen carrier. The working principle of the bifunctional material in CCDR is shown in Figure 12a. During the reduction halfcycle, the mixture feed of CH4 + CO2 is converted on Ni sites to CO and H2, which in turn reduce the oxygen carrier component (from Fe3O4 to metallic Fe), yielding CO2 and H2O. Subsequently, the reduced oxygen carrier is re-oxidized (from Fe to Fe3O4) by extra CO2, producing CO in the reoxidation half-cycle. During H2-TPR of the Ni/CeO2-Fe2O3 bifunctional material, Fe2O3 is reduced to Fe3O4, and then FeO, finally transforming into metallic Fe. At around 350 °C metallic Ni appears due to the reduction of NiO. Interaction between Ni and Fe leads to Ni-Fe alloy formation from 580 °C onwards. This alloy remains stable up to 600 °C during CO2-TPO but is completely decomposed into Ni and Fe3O4 at 700 °C (Figure 12b). The addition of Ni to CeO2-Fe2O3 shows great promise towards generation of CO from CO2 in CCDR. As seen in Figure 12c, the position of the CeO2 (111) peak shows In previous work [23], a bifunctional 5 wt % Ni/CeO 2 -Fe 2 O 3 material was prepared by impregnation of the required amount of Ni on a 50 wt % CeO 2 -Fe 2 O 3 oxygen carrier. The working principle of the bifunctional material in CCDR is shown in Figure 12a. During the reduction half-cycle, the mixture feed of CH 4 + CO 2 is converted on Ni sites to CO and H 2 , which in turn reduce the oxygen carrier component (from Fe 3 O 4 to metallic Fe), yielding CO 2 and H 2 O. Subsequently, the reduced oxygen carrier is re-oxidized (from Fe to Fe 3 O 4 ) by extra CO 2 , producing CO in the re-oxidation half-cycle. During H 2 -TPR of the Ni/CeO 2 -Fe 2 O 3 bifunctional material, Fe 2 O 3 is reduced to Fe 3 O 4 , and then FeO, finally transforming into metallic Fe. At around 350 • C metallic Ni appears due to the reduction of NiO. Interaction between Ni and Fe leads to Ni-Fe alloy formation from 580 • C onwards. This alloy remains stable up to 600 • C during CO 2 -TPO but is completely decomposed into Ni and Fe 3 O 4 at 700 • C (Figure 12b). The addition of Ni to CeO 2 -Fe 2 O 3 shows great promise towards generation of CO from CO 2 in CCDR. As seen in Figure 12c, the position of the CeO 2 (111) peak shows a downward shift indicating partial reduction. This shift is quite significant in Ni/CeO 2 -Fe 2 O 3 , while it remains limited in CeO 2 -Fe 2 O 3 and doesn't occur at these temperatures in pure CeO 2 , meaning the partial reduction is enhanced by the addition of Ni, leading to deeper reduction of the oxygen carrier. As a result, a higher CO yield can be obtained during the reduction half-cycle when feeding CO 2 . The CO yield over the Ni/CeO 2 -Fe 2 O 3 bifunctional material is proven to be 10 times higher than over the CeO 2 -Fe 2 O 3 oxygen carrier (Figure 12d).
Materials 2018, 11, x FOR PEER REVIEW 20 of 33 a downward shift indicating partial reduction. This shift is quite significant in Ni/CeO2-Fe2O3, while it remains limited in CeO2-Fe2O3 and doesn't occur at these temperatures in pure CeO2, meaning the partial reduction is enhanced by the addition of Ni, leading to deeper reduction of the oxygen carrier. As a result, a higher CO yield can be obtained during the reduction half-cycle when feeding CO2. The CO yield over the Ni/CeO2-Fe2O3 bifunctional material is proven to be 10 times higher than over the CeO2-Fe2O3 oxygen carrier (Figure 12d). An alternative towards the bifunctional material for CCDR processes is provided by spinel ferrites (AFe2O4), possessing both a catalytic element (A) and an oxygen carrier element (Fe). This ferrite-based material is readily prepared via co-precipitation of all elements, unlike the additional modification of a Fe-based oxygen carrier by a catalyst component. Various Me-ferrites (Me = Ni, Co, Cu, Mn, Zn) have been investigated in chemical looping processes [112,150,151], among which the NiFe2O4 and CoFe2O4 show certain catalytic activity and redox stability during chemical looping dry reforming of CH4. Given the increase of alcohol feedstocks from chemical conversion of biomass, these ferrite-based bifunctional materials were further tested in chemical looping reforming of alcohol fuels, such as methanol and ethanol, for the production of value added fuels such as syngas. An alternative towards the bifunctional material for CCDR processes is provided by spinel ferrites (AFe 2 O 4 ), possessing both a catalytic element (A) and an oxygen carrier element (Fe). This ferrite-based material is readily prepared via co-precipitation of all elements, unlike the additional modification of a Fe-based oxygen carrier by a catalyst component. Various Me-ferrites (Me = Ni, Co, Cu, Mn, Zn) have been investigated in chemical looping processes [112,150,151], among which the NiFe 2 O 4 and CoFe 2 O 4 show certain catalytic activity and redox stability during chemical looping dry reforming of CH 4 . Given the increase of alcohol feedstocks from chemical conversion of biomass, these ferrite-based bifunctional materials were further tested in chemical looping reforming of alcohol fuels, such as methanol and ethanol, for the production of value added fuels such as syngas. Dharanipragada et al. [125] investigated the performance of NiFe 2 O 4 -or CoFe 2 O 4 -CeZrO 2 bifunctional materials in a CCDR process with two different alcohol fuels, methanol and an ethanol-water mixture (1:1 molar ratio). In this case (Figure 13), methanol or ethanol acts as feed in the reduction half-cycle, which is first converted into CO and H 2 (as well as some CH 4 when using the ethanol-water mixture), that simultaneously reduce the material, yielding CO 2 and H 2 O. In the re-oxidation half-cycle, CO 2 is fed to regenerate the material and produce CO. Dharanipragada et al. [125] investigated the performance of NiFe2O4-or CoFe2O4-CeZrO2 bifunctional materials in a CCDR process with two different alcohol fuels, methanol and an ethanol-water mixture (1:1 molar ratio). In this case (Figure 13), methanol or ethanol acts as feed in the reduction half-cycle, which is first converted into CO and H2 (as well as some CH4 when using the ethanol-water mixture), that simultaneously reduce the material, yielding CO2 and H2O. In the re-oxidation half-cycle, CO2 is fed to regenerate the material and produce CO. During prolonged cycling, the material suffers from deactivation due to phase segregation and particle sintering. It was found that CoFe2O4-CeZrO2 with low ferrite content (20 wt %) had good redox activity and stability, but the material with 80 wt % CoFe2O4 underwent deactivation owing to phase segregation into Co and Fe3O4. On the other hand, all NiFe2O4-CeZrO2 materials suffered from phase segregation as well as sintering, and the separated Ni and Fe3O4 phases could not be restored into the original spinel phase (NiFe2O4). In the methanol-based CCDR process, carbon formation was observed on both NiFe2O4-and CoFe2O4-CeZrO2, but significantly more intense on the former (Figure  13b), indicating a lower carbon-resistance ability of the Ni-ferrite. The deposited carbon in turn enhanced the CO yield during the re-oxidation half-cycle, because of the reaction between CO2 and carbon. This phenomenon was verified experimentally, showing a more elevated CO yield than the theoretical value after material re-oxidation (as seen in Figure 13a). Furthermore, it was suggested that air or O2 would be needed for complete gasification of the deposited carbon.
Of course, the deactivation of such bifunctional materials due to carbon deposition during CCDR cannot be ignored. In order to tackle this drawback, the upgraded technology of CCAR can provide a prospective solution. The feasibility of the CCAR process over a Ni-Fe/MgAl2O4 bifunctional material was studied through both thermodynamic analysis and prolonged cycling experiments [20]. In the reduction half-cycle, a mixture of CH4, CO2 and O2 in a 1:1:0.5 molar ratio was first converted over the Ni catalyst into a mixture of CO and H2, which both reduce iron oxide to metallic Fe ( Figure  14a). The high endothermicity of dry reforming as well as the coke formation could be mitigated in this half-cycle under auto-thermal conditions. The steady-state reforming of the CH4:CO2:O2 =1:1:0.5 mixture over Ni-Fe/MgAl2O4 at various temperatures showed the auto-thermal dry reforming reduction regime was established in the temperature range from 550 °C to 750 °C, where the O2 is completely consumed and the conversion of CH4 and CO2 respectively reaches 90% and 80%, yielding During prolonged cycling, the material suffers from deactivation due to phase segregation and particle sintering. It was found that CoFe 2 O 4 -CeZrO 2 with low ferrite content (20 wt %) had good redox activity and stability, but the material with 80 wt % CoFe 2 O 4 underwent deactivation owing to phase segregation into Co and Fe 3 O 4 . On the other hand, all NiFe 2 O 4 -CeZrO 2 materials suffered from phase segregation as well as sintering, and the separated Ni and Fe 3 O 4 phases could not be restored into the original spinel phase (NiFe 2 O 4 ). In the methanol-based CCDR process, carbon formation was observed on both NiFe 2 O 4 -and CoFe 2 O 4 -CeZrO 2 , but significantly more intense on the former (Figure 13b), indicating a lower carbon-resistance ability of the Ni-ferrite. The deposited carbon in turn enhanced the CO yield during the re-oxidation half-cycle, because of the reaction between CO 2 and carbon. This phenomenon was verified experimentally, showing a more elevated CO yield than the theoretical value after material re-oxidation (as seen in Figure 13a). Furthermore, it was suggested that air or O 2 would be needed for complete gasification of the deposited carbon.
Of course, the deactivation of such bifunctional materials due to carbon deposition during CCDR cannot be ignored. In order to tackle this drawback, the upgraded technology of CCAR can provide a prospective solution. The feasibility of the CCAR process over a Ni-Fe/MgAl 2 O 4 bifunctional material was studied through both thermodynamic analysis and prolonged cycling experiments [20]. In the reduction half-cycle, a mixture of CH 4 , CO 2 and O 2 in a 1:1:0.5 molar ratio was first converted over the Ni catalyst into a mixture of CO and H 2 , which both reduce iron oxide to metallic Fe (Figure 14a). The high endothermicity of dry reforming as well as the coke formation could be mitigated in this half-cycle under auto-thermal conditions. The steady-state reforming of the CH 4 :CO 2 :O 2 =1:1:0.5 mixture over Ni-Fe/MgAl 2 O 4 at various temperatures showed the auto-thermal dry reforming reduction regime was established in the temperature range from 550 • C to 750 • C, where the O 2 is completely consumed and the conversion of CH 4 and CO 2 respectively reaches 90% and 80%, yielding syngas with a H 2 :CO ratio close to unity (Figure 14b). During the re-oxidation half-cycle, CO is produced alongside the regeneration of iron oxide by CO 2 (Figure 14a). The overall CO yield depends on the degree of Fe 3 O 4 reduction reached in the reduction half-cycle, which strongly depends on the ratio (R c ) between the present reducing (CO + H 2 ) and oxidizing (CO 2 + H 2 O) gases. Based on thermodynamic analysis, high conversion of Fe 3 O 4 to metallic Fe can be reached if R c > 2 and T > 600 • C in the present study (Figure 14c). syngas with a H2:CO ratio close to unity (Figure 14b). During the re-oxidation half-cycle, CO is produced alongside the regeneration of iron oxide by CO2 (Figure 14a). The overall CO yield depends on the degree of Fe3O4 reduction reached in the reduction half-cycle, which strongly depends on the ratio (Rc) between the present reducing (CO + H2) and oxidizing (CO2 + H2O) gases. Based on thermodynamic analysis, high conversion of Fe3O4 to metallic Fe can be reached if Rc > 2 and T > 600 °C in the present study (Figure 14c). The stability of the Ni-Fe/MgAl2O4 bifunctional material was determined by examining the changes in morphological structure by SEM (Figure 15a), showing no obvious carbon deposition on the material surface, but an increase in the particle size and a collapse in the pore structure after 25 cycles of CCAR. Therefore, particle sintering results in a decrease of the CO time-averaged spacetime yield, which reaches a stable level after the first 5 cycles (Figure 15b), suggesting that severe sintering of particles only occurs in the initial stage of redox cycling.  The stability of the Ni-Fe/MgAl 2 O 4 bifunctional material was determined by examining the changes in morphological structure by SEM (Figure 15a), showing no obvious carbon deposition on the material surface, but an increase in the particle size and a collapse in the pore structure after 25 cycles of CCAR. Therefore, particle sintering results in a decrease of the CO time-averaged space-time yield, which reaches a stable level after the first 5 cycles (Figure 15b), suggesting that severe sintering of particles only occurs in the initial stage of redox cycling.
Obviously, deactivation of bifunctional materials through particle sintering during chemical looping remains a challenge. As mentioned in Section 3.3.2, a possible countering strategy could be the construction of material with a core-shell structure. Hence, a Ni-Fe bifunctional nanomaterial (Fe/Zr@Zr-Ni@Zr) was designed, specifically for the catalyst-assisted chemical looping process. It consists of a ZrO 2 -coated Ni outer shell, encapsulating a Fe 2 O 3 /ZrO 2 @ZrO 2 core (Figure 16a), showing multiple benefits: (1) the Ni-based shell and the Fe-based core serve as reforming catalyst and oxygen carrier, respectively, and achieve catalytic reforming and oxygen transfer in a single nanoscale unit; (2) the hollow sphere surrounding the core prevents the aggregation of core particles [152], while providing space for gas-solid contact; (3) at the same time, the sphere offers a large specific surface area, leading to fine dispersion of Ni particles and the adequate deposition of the outer ZrO 2 protective layer, both of which contribute to a high catalytic activity [153]. To design such a structure, an extra step is introduced in the synthesis pathway, where SiO 2 is typically used as a template [154]: after synthesis of the Fe 2 O 3 /ZrO 2 @ZrO 2 core, the latter is coated with a dense SiO 2 layer using a modified Stöber process [155]. After loading the Ni@ZrO 2 shell, the SiO 2 template is selectively removed by dissolution with an adequate solvent to free the core, so that it can contact with reactant gases. The result is an eccentric core-shell structure as the core is now no longer supported within the hollow shell (Figure 16b). oxidation, region II: dry and steam reforming, region III: auto-thermal dry reforming; (c) The ratio Rc between reducing gases (CO + H2) and oxidizing gases (CO2 + H2O) as a function of reaction temperature during catalyst-assisted chemical looping auto-thermal dry reforming process, based on the experimental data of (b). Solid circles with dashed line as guide to the eye. Solid lines: Equilibrium lines of the iron/iron oxide system for each Rc at different temperatures.
The stability of the Ni-Fe/MgAl2O4 bifunctional material was determined by examining the changes in morphological structure by SEM (Figure 15a), showing no obvious carbon deposition on the material surface, but an increase in the particle size and a collapse in the pore structure after 25 cycles of CCAR. Therefore, particle sintering results in a decrease of the CO time-averaged spacetime yield, which reaches a stable level after the first 5 cycles (Figure 15b), suggesting that severe sintering of particles only occurs in the initial stage of redox cycling.  Obviously, deactivation of bifunctional materials through particle sintering during chemical looping remains a challenge. As mentioned in Section 3.3.2, a possible countering strategy could be the construction of material with a core-shell structure. Hence, a Ni-Fe bifunctional nanomaterial (Fe/Zr@Zr-Ni@Zr) was designed, specifically for the catalyst-assisted chemical looping process. It consists of a ZrO2-coated Ni outer shell, encapsulating a Fe2O3/ZrO2@ZrO2 core (Figure 16a), showing multiple benefits: (1) the Ni-based shell and the Fe-based core serve as reforming catalyst and oxygen carrier, respectively, and achieve catalytic reforming and oxygen transfer in a single nanoscale unit; (2) the hollow sphere surrounding the core prevents the aggregation of core particles [152], while providing space for gas-solid contact; (3) at the same time, the sphere offers a large specific surface area, leading to fine dispersion of Ni particles and the adequate deposition of the outer ZrO2 protective layer, both of which contribute to a high catalytic activity [153]. To design such a structure, an extra step is introduced in the synthesis pathway, where SiO2 is typically used as a template [154]: after synthesis of the Fe2O3/ZrO2@ZrO2 core, the latter is coated with a dense SiO2 layer using a modified Stöber process [155]. After loading the Ni@ZrO2 shell, the SiO2 template is selectively removed by dissolution with an adequate solvent to free the core, so that it can contact with reactant gases. The result is an eccentric core-shell structure as the core is now no longer supported within the hollow shell (Figure 16b).  The bifunctional material Fe/Zr@Zr-Ni@Zr, made according to this recipe, is active for reforming of a feed mixture of CH 4 , CO 2 and O 2 (molar ratio of 1:1:0.2) as well as for conversion of CO 2 to CO in a CCAR process (Figure 17a). O 2 conversion during reduction is always 100% in these conditions. Methane dry reforming is the dominant reaction, yielding syngas with a H 2 :CO ratio of unity, together with some methane combustion into CO 2 and H 2 O. In the second half-cycle, the re-oxidation of Fe into Fe 3 O 4 rapidly converts CO 2 to CO. Hence, the CO space-time yield shows a sharp peak in the first 20 s of the re-oxidation half-cycle and then steadily decreases towards zero, indicating the majority of Fe 3 O 4 has been regenerated. Moreover, when averaged over 25 CCAR cycles, a stable CO time-averaged space-time yield is obtained over Fe/Zr@Zr-Ni@Zr (Figure 17b). Characterizations (XPS, STEM-EDX mapping and XRD) of the spent sample after 25 cycles indicate no obvious carbon deposits on the core-shell bifunctional material and no sintering of the Ni and Fe particles. The bifunctional material Fe/Zr@Zr-Ni@Zr, made according to this recipe, is active for reforming of a feed mixture of CH4, CO2 and O2 (molar ratio of 1:1:0.2) as well as for conversion of CO2 to CO in a CCAR process (Figure 17a). O2 conversion during reduction is always 100% in these conditions. Methane dry reforming is the dominant reaction, yielding syngas with a H2:CO ratio of unity, together with some methane combustion into CO2 and H2O. In the second half-cycle, the re-oxidation of Fe into Fe3O4 rapidly converts CO2 to CO. Hence, the CO space-time yield shows a sharp peak in the first 20 s of the re-oxidation half-cycle and then steadily decreases towards zero, indicating the majority of Fe3O4 has been regenerated. Moreover, when averaged over 25 CCAR cycles, a stable CO time-averaged space-time yield is obtained over Fe/Zr@Zr-Ni@Zr (Figure 17b). Characterizations (XPS, STEM-EDX mapping and XRD) of the spent sample after 25 cycles indicate no obvious carbon deposits on the core-shell bifunctional material and no sintering of the Ni and Fe particles. The CO2 utilization during this process can be further enhanced through the construction of a double-zone reactor bed configuration [126], composed of a core-shell structured Fe/Zr@Zr-Ni@Zr bifunctional material (the first zone) followed by a core-shell structured Fe2O3/ZrO2@ZrO2 oxygen carrier (the second zone). In this bed configuration, it is expected that much more CO2 can be converted to CO due to the additional oxygen storage capacity provided by the second oxygen carrier zone.

Conclusions
Chemical looping technology has recently emerged as a promising "clean energy conversion technology" due to the advantages of manageable inherent CO2 capture. In a chemical looping scheme, a single reaction is broken down into two, spatially or temporally separated half-cycle reactions, coupled through the cyclic transformation of a solid looping material. Depending on their functions, the looping materials can be divided into two main types, correspondingly defining two groups of chemical looping processes: As an oxygen carrier, to transport oxygen through cyclic reduction and re-oxidation during chemical looping combustion or reforming; as a CO2 carrier, to remove CO2 in-situ through cyclic carbonation and calcination during sorption-enhanced chemical looping. The flexibility of chemical looping processes allows the utilization of various feedstocks (fossil fuels, CO2 and renewables) and the production of various value-added chemicals (H2, CO, syngas and liquid fuels) via tailoring of reactor feed and looping materials. The CO 2 utilization during this process can be further enhanced through the construction of a double-zone reactor bed configuration [126], composed of a core-shell structured Fe/Zr@Zr-Ni@Zr bifunctional material (the first zone) followed by a core-shell structured Fe 2 O 3 /ZrO 2 @ZrO 2 oxygen carrier (the second zone). In this bed configuration, it is expected that much more CO 2 can be converted to CO due to the additional oxygen storage capacity provided by the second oxygen carrier zone.

Conclusions
Chemical looping technology has recently emerged as a promising "clean energy conversion technology" due to the advantages of manageable inherent CO 2 capture. In a chemical looping scheme, a single reaction is broken down into two, spatially or temporally separated half-cycle reactions, coupled through the cyclic transformation of a solid looping material. Depending on their functions, the looping materials can be divided into two main types, correspondingly defining two groups of chemical looping processes: As an oxygen carrier, to transport oxygen through cyclic reduction and re-oxidation during chemical looping combustion or reforming; as a CO 2 carrier, to remove CO 2 in-situ through cyclic carbonation and calcination during sorption-enhanced chemical looping.
The flexibility of chemical looping processes allows the utilization of various feedstocks (fossil fuels, CO 2 and renewables) and the production of various value-added chemicals (H 2 , CO, syngas and liquid fuels) via tailoring of reactor feed and looping materials.
At the heart of chemical looping technology is the selection of looping materials. The employment of a suitable looping material determines the sustainability and economics of the technology. From a practical point of view, some commonly used looping materials were compared. CaO is an ideal material to serve as CO 2 carrier owing to its low cost, non-toxicity, acceptable CO 2 capacity, regenerability and heat conduction. Further efforts for a CaO-based carrier should focus on increasing its CO 2 capacity and adsorption rate, as well as improving its thermal stability and mechanical strength. Transition metals, such as Ni, Fe, Cu, Mn and Co-based materials, are commonly used as oxygen carriers, among which Fe-and Ni-based materials are most employed, especially for carbonaceous feedstock-driven chemical looping processes. Iron oxide is highly tolerant to carbon deposition, environmentally friendly, cheap, and has a high oxygen storage capacity from CO 2 . A future task in the development of Fe-based oxygen carriers is the enhancement of its ability to resist sintering. Ni then again has excellent reactivity towards hydrocarbon fuels, making Ni-based material a promising redox catalyst for chemical looping reforming, but the drawbacks with regard to particle sintering, carbon deposition and sulfur poisoning of the material need to be further addressed.
Based on different purposes of improvement of oxygen carriers, approaches have been investigated, mainly focusing on compositional and structural optimization. The stabilization of active components with promoter materials is a widely used method. Some physical promoters (e.g., Al 2 O 3 ) can react with the active component to form spinel phases (such as NiAl 2 O 4 and FeAl 2 O 4 ), which can continue to act as inert binders but resulting in the partial loss of reactivity. Adding an appropriate amount of MgO into NiO/Al 2 O 3 or Fe 2 O 3 /Al 2 O 3 systems can mitigate the formation of Ni-or Fe-Al 2 O 4 spinel through the competition of MgAl 2 O 4 formation. Owing to the thermal and chemical stability, both MgAl 2 O 4 and ZrO 2 are considered as a good physical promoter materials for oxygen carriers. In addition, reducible CeO 2 can serve not only as a physical, but also a chemically active promoter. The CeO 2 -modified oxygen carriers show high reactivity, good reducibility, high oxygen transport capacity as well as strong carbon resistance.
Another strategy is the use of an oxygen carrier with mixed active metals, combining the complementary advantages of individual metals. The mixing of Co into a Ni-based oxygen carrier can improve its reducibility and carbon resistance. The addition of Fe into a Cu-based oxygen carrier increases its melting point to resist agglomeration. Combining advantages of both Ni and iron oxide, the bimetallic Ni/Fe-based material-containing an optimized molar ratio of Ni/Fe and supported by a stable promoter material-gradually becomes a highly promising oxygen carrier or redox catalyst for chemical looping reforming, showing high reactivity towards hydrocarbon fuels, good thermal stability, high ability to resist carbon deposition and low cost.
Structural optimization is of equal importance for an ideal oxygen carrier. Perovskite-structured oxide (ABO 3 -type) materials show remarkable results. Owing to the unique structure of perovskite, its lattice oxygen is readily utilized for syngas production with high selectivity, and phase transformation can be avoided due to its thermal stability. The partial substitution of A-and B-site metals by an aliovalent metal can improve the stability and reactivity, respectively. One optimum composition of perovskite-structured oxygen carriers is La 0.7 Sr 0.3 Cr 0.1 Fe 0.9 O 3 . Future work must focus on improvement of the oxygen storage and transport capacities of perovskite-structured oxygen carriers. Furthermore, a core-shell structure, i.e., an active core encapsulated by a thin and porous shell, has become the most competitive structure, due to its strong adjustability: Control of the particle size of the active core enables to control the reaction kinetics; Tailoring of the composition, thickness and pore properties of the shell can change the stability of the material and its selectivity towards specific products. An ideal core-shell structured oxygen carrier applied in chemical looping should show high oxygen transport capacity and strong thermal stability under the harsh operation conditions. In this sense, a nano-size active metal core encapsulated by a chemically inert and porous shell is required to construct a core-shell structured oxygen carrier. A proposed Fe 2 O 3 /ZrO 2 @ZrO 2 oxygen carrier, composed of a stable Fe 2 O 3 /ZrO 2 core and a thin and mesoporous ZrO 2 shell, proved to possess excellent redox activity and stability for CO 2 utilization during a long-term chemical looping process.
Efficient conversion of hydrocarbon fuels to syngas accompanied by a goal of maximum CO 2 utilization has been achieved through a novel process-catalyst-assisted chemical looping dry or auto-thermal dry reforming-implemented over a bifunctional material. The latter must show functions of both reforming catalyst and oxygen carrier. The performances of series of Ni-Fe bifunctional materials were investigated in terms of syngas production, CO 2 conversion and material stability. A core-shell structured Fe/Zr@Zr-Ni@Zr bifunctional material, providing catalyst and oxygen storage functions from a Ni-based shell and a Fe-based core, respectively, shows a stable CO yield during prolonged cycling without significant carbon deposition and particle sintering. On the reactor scale, careful distribution of the Ni-based catalyst and Fe-based oxygen carrier in the reactor bed could further enhance CO 2 utilization.

Outlook
Although chemical looping materials have now been developed as core-shell structured bifunctional nanomaterial, applicable in the new process of chemical looping technology-catalyst-assisted chemical looping, the research on the industrially-applicable looping materials has not yet reached its final destination. Further outlooks could be as follows: (1) Considering the economic effectiveness, the establishment of low-cost and facile synthesis methods is crucial for the development of the industrial looping materials, especially for the core-shell structured nanomaterials.
(2) High-pressure operation of the chemical looping reactors is essential to maintain the circulation efficiency of an industrial platform. Hence, holistic evaluation of the effect of pressure on the behavior of the core-shell structured materials is necessary for further improvement of the materials.
(3) Research towards the enhancement of the long-term stability, for example maintaining reactivity and structural integrity for thousands of redox cycles, of the looping materials is crucial. The choice of the developed bifunctional looping materials should be screened by taking into account of their lifecycle and cost of production.
(4) Most of the fuel feedstocks contain sulfur-based contaminates, which easily bind to the active metal sites to form stable surface oxy-sulfides and/or bulk metal sulfides, resulting in the material deactivation. Therefore, study on the sulfur-poisoning mechanism of the materials by using feed compositions that are close to real feedstocks is indispensable.