1 Introduction
Chemical looping technology is a two-step chemical conversion method that enables self-separation of products and has a wide range of applications (
Figure 1), offering unique advantages in combustion, reforming, dehydrogenation, hydrogen production, and hydrogen purification
[1-5].
Take combustion as an example; traditional technologies often use air as the oxidant, resulting in high costs for CO
2 separation and capture
[6-7]. Chemical looping technology utilizes lattice oxygen from oxygen carriers as the oxidant, enabling direct regeneration in high-temperature air through cyclic reactions, thereby eliminating the need for pure oxygen separation and achieving self-separation of CO
2, significantly reducing process costs
[8]. Oxygen carriers supply oxygen and transfer heat between circulating reactors, serving as the core component of chemical looping technology
[9]. Chemical looping technology typically employs metal oxides as oxygen carriers, and its reaction process involves the migration and transformation of lattice oxygen, the diffusion of metal ions, and the evolution of the fuel-oxygen carrier reaction interface. Therefore, exploring the release, reaction, and recovery pathways of lattice oxygen, as well as the dynamic changes of active components, is of great scientific significance for understanding the intrinsic mechanisms of chemical looping reactions, developing methods to regulate reaction efficiency and product selectivity, and designing new high-performance oxygen carriers.
Researchers have delved into the lattice oxygen migration mechanism of different oxygen carriers, the diffusion mechanism of active metals, and the gas-solid reaction characteristics between oxygen carriers and fuels. These studies not only reveal the redox cycling mechanism of oxygen carriers during the reaction process but also provide insights for the design of high-performance oxygen carriers.
This article highlights recent significant advances in the study of oxygen carrier mechanisms. First, it reviews research methods for understanding lattice oxygen transfer mechanisms and the evolution patterns of active components, such as in-situ characterization, visualization techniques, experimental design, and first-principles calculations. These methods have laid the foundation for investigating the oxygen storage and release mechanisms of oxygen carriers. Second, since the types of oxygen species influence product selectivity
[10]and the migration of bulk lattice oxygen affects the reaction characteristics of oxygen carriers
[11], this article also elaborates on the crucial role of oxygen species in the chemical looping conversion process of oxygen carriers, compares the oxygen release characteristics of different types of oxygen carriers, and summarizes the migration pathways and conversion mechanisms of bulk lattice oxygen. Finally, it discusses the points of contention among scholars regarding the oxygen storage and release mechanisms of oxygen carriers, and outlines unresolved issues and future research directions.
2 Research methods for the oxygen storage and release mechanism of oxygen carriers
Oxygen carriers are the cornerstone of chemical looping technology
[12-13]. The lattice oxygen within their bulk phase and the dynamic migration of active components significantly influence the performance of chemical looping conversion. Research into the lattice oxygen migration mechanism helps elucidate the structure-activity relationship and redox mechanism of oxygen carriers, which is crucial for the rational design of oxygen carriers with superior activity, stability, and selectivity. Therefore, it has consistently been a research hotspot in this field.
Currently, the main methods for studying lattice oxygen and active component migration in oxygen carriers are shown in
Figure 2 [14-25].These methods primarily include in-situ characterization, visualization observation, inert labeling experimental design, and first-principles calculations (DFT). Among them, in-situ characterization techniques can capture reaction intermediates and provide a realistic reaction pathway, while visualization characterization techniques can instantly display elemental migration and structural evolution during the oxygen carrier reaction process. Isotope mass spectrometry can track the atomic distribution of reactants in products.
2.1 Advanced Characterization Methods
The evolution of the surface and bulk structures of oxygen carrier materials during chemical looping reactions is a complex, continuous, and dynamic process, and in-situ characterization techniques are effective means to observe these changes.
Table 1summarizes the advantages, disadvantages, and application areas of in-situ X-ray photoelectron spectroscopy (XPS), in-situ X-ray diffraction (XRD), in-situ Raman spectroscopy, as well as other characterization methods such as nuclear magnetic resonance (NMR), atom probe tomography (APT), and synchrotron radiation techniques (XAFS)
[26-33].
2.1.1 Common in-situ characterization methods
In-situ characterization techniques can provide insights into the surface and bulk structural evolution of oxygen carriers under realistic reaction conditions, linking these changes to their catalytic performance. Compared to ex-situ characterization, in-situ methods are more effective at capturing reaction intermediates, identifying active sites, and monitoring the dynamic behavior of the oxygen carrier's crystal structure, thereby helping to clarify the reaction pathways
[34].
In situ XPS can monitor in real time the chemical state of the oxygen carrier surface and the chemical changes in the near-surface region, facilitating the capture of transient reaction products and the tracking of oxygen species migration, thereby providing strong support for studying the reaction mechanisms involving materials. Huang et al.
[14]used in situ XPS technology to capture the formation process of hydroxyl ions on the surface of oxygen carriers during chemical looping reactions, while Wang et al.
[35]employed this technique to identify the key reaction interface of Ni/ZrO
2 oxygen carriers during the reaction process. The results showed that oxygen vacancies, acting as mediators for electron transfer, facilitate the transfer of electrons from Ni
0 to other substances. They suggested that the Ni-O-Zr interface with enhanced charge transfer may play a crucial role in the reaction performance of Ni/ZrO
2. Su et al.
[36]directly tracked the evolution of oxygen vacancies and oxygen species through in situ XPS and near-edge X-ray absorption fine structure characterization, elucidating the fundamental principles underlying the oxygen storage capacity of CeO
2.
Oxygen carriers inevitably undergo crystal structure evolution during oxygen storage and release, which is closely related to changes in their activity. Therefore, some researchers have employed in-situ XRD to track the crystal evolution of oxygen carriers during chemical looping reactions, aiming to understand the stepwise reduction mechanism of metal atoms and capture the existing states of metal components throughout the reaction process. Li et al.
[37]used in-situ synchrotron X-ray powder diffraction (SXRD) to investigate the cyclic stability and structural evolution mechanisms of perovskite-type oxygen carriers (La
1- xSr
xCoO
3- δ) during chemical looping conversion. Miller et al.
[38]also applied this method to observe the evolution of the calcium ferrite phase in spinel-type oxygen carriers during methane reduction (CaFe
2O
4→CaFe
3O
5+Ca
2Fe
2O
5→FeO+Fe+CaO→Fe+CaO). Zhang et al.
[39]utilized in-situ time-resolved X-ray diffraction (TR-XRD) to analyze the structural transformations of 10 wt.% FeO
x/CeO
2, NiO
x/CeO
2, and CuO
x/CeO
2oxygen carriers during the CH
4-TPR reaction. They found that MO
x(M represents a metal element) were gradually reduced and eventually existed as metallic elements, while CeO
2ultimately remained in the cubic Ce
2O
3phase. Furthermore, they correlated the consumption of CH
4at different stages with the phase transformation processes of the oxygen carriers.
Monitoring the changes in chemical bonds and vibrational modes of atoms within oxygen carriers can provide insights into the dynamic evolution of surface components, offering atomic-level evidence for the migration of active species. Therefore, some researchers have employed in-situ Raman spectroscopy (Raman) and in-situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) to investigate changes in the surface chemical environment of metal oxides. Song et al.
[40]confirmed through in-situ Raman that a highly active chemical reaction interface composed of lattice oxygen and corresponding metal atoms exists, and they found that these metal atoms are always present on the surface of the oxygen carrier. Chen et al.
[16]used this method to demonstrate that oxygen anions associated with bridging V―O―Ti bonds in the oxygen carrier are more active than those associated with V―O―V bonds. Additionally, Ren et al.
[41]observed via in-situ DRIFTS spectroscopy that after La doping, CoMn
2O
4 spinel formed more oxygen vacancies, and under conditions of abundant oxygen vacancies, the mobility and reactivity of lattice oxygen were significantly enhanced.
2.1.2 Other characterization methods
With the advancement of characterization techniques, redox behaviors similar to chemical looping processes have been discovered in recent years in fields such as multiphase catalysis
[42]and electrocatalysis
[43-44]. This allows us to draw on characterization techniques from other fields, such as advanced in-situ environmental transmission electron microscopy (ETEM), nuclear magnetic resonance (NMR), time-of-flight secondary ion mass spectrometry (ToF-SIMS), synchrotron radiation techniques (XAFS), and multi-energy photon scanning (MEP), as illustrated in
Figure 3, to study the dynamic migration processes of lattice oxygen and active components in oxygen carriers
[45-47].
Song et al.
[19]Using aberration-corrected in-situ ETEM at a high temperature of 850 ℃, they directly observed the dynamic migration of metal atoms and lattice oxygen within the bulk phase of oxygen carriers. Combined with in-situ electron energy loss spectroscopy EELS-Mapping, quasi-in-situ XPS, and first-principles molecular dynamics simulations, they revealed the ion migration mechanism and surface/interface reaction mechanisms of spinel NiFe
2O
4 during chemical-looping CO
2 conversion. They found that during CO
2 decomposition, Ni migrates toward the core of the oxygen carrier to form an Fe-Ni alloy, and upon further oxidation in air, the Fe-Ni alloy migrates outward to form a hollow structure. Moreover, the migration path of metal atoms depends on the rate of oxygen transfer. Based on these findings, they concluded that regulating the oxidizing power of the oxidation medium can effectively suppress phase separation in oxygen carriers.
In situ NMR can provide complementary and quantitative information on catalyst materials, reactants, intermediates, and products, which in turn can facilitate the development of structure-activity relationships. Camp et al.
[17]demonstrated that
13C nuclear magnetic resonance spectroscopy can track the metathesis reaction products of 1-hexene, and by comparing Raman spectra from the same period, they discovered ethylene products not observed in NMR detection. Therefore, they suggested that low-spin-density NMR techniques should be combined with Raman characterization to accurately reveal the relationship between reactions and structural changes within the catalyst. Hunger et al.
[48]combined in situ NMR with in situ UV-Vis spectroscopy to capture the intermediate products of methanol-to-olefins (MTO) over weakly acidic zeolites, successfully detecting the first cyclic compounds and carbon ions at 413 K. They speculated that extra-framework aluminum, acting as a Lewis acid site, facilitates the formation of hydrocarbons and carbon ions at lower reaction temperatures. Thus, they concluded that NMR can more precisely identify signals from major reactants, while UV-Vis spectroscopy can capture the formation of minor cyclic compounds and carbon ions. In fact, NMR technology can also be used to analyze the chemical state and catalytic properties of adsorbates within crystal channels, as well as the nature and distribution of coke deposits. It holds promise for combining this technique with XRD and other methods to study the evolution of the bulk structure of chemical looping oxygen carriers, providing more comprehensive information on material structural changes.
Time-of-flight secondary ion mass spectrometry (ToF-SIMS) has developed into a highly powerful micro-surface analysis technique capable of simultaneously detecting various ions across an almost unlimited mass range, with exceptionally high mass resolution. It can be applied to the precise detection of charged particles on the surface of catalytic materials, allowing observation of changes in their surface structures. Dai et al.
[49]applied ToF-SIMS technology to the study of cerium oxide samples and pure Ce samples, quantifying the variation of the CeO
+/Ce
+ ratio from the surface to the bulk as a function of depth, thereby demonstrating the transition process between oxidized and metallic states. The results indicated that metallic Ce appears when the CeO
+/Ce
+ ratio decreases to around 2. Meanwhile, Tang et al.
[50]combined ToF-SIMS with SEM imaging to investigate the electrochemical activation phenomena at different stages on the surface of gold electrodes in hydrochloric acid solutions. They found that the Au
x- signals detected during various electrochemical cycles continuously changed, and the ion signals of Au
2H
-, Au
2S
-, Au
2CN
-, and Au
2HS
- on the polished gold electrode surface significantly decreased after activation. This suggests that electrochemical activation effectively removes surface contaminants containing sulfur, carbon, nitrogen, and hydrogen, further confirming that ToF-SIMS is a highly surface-sensitive technique capable of providing chemical information from extremely shallow depths of materials. Additionally, Fleig et al.
[46]used a
18O isotope exchange depth profiling method combined with ToF-SIMS to study the influence of grain boundaries on oxygen exchange at the surface and oxygen diffusion within the bulk of La
0.8Sr
0.2MnO
3 thin films. By carefully considering all fitting parameters, including those with different diffusion coefficients and surface exchange coefficients, they measured the
18O concentration at different depths, achieving the first quantitative determination of the surface exchange coefficient for grain boundaries in mixed-conducting oxides. Kubicek et al.
[51]employed this method to explore the impact of lattice stress on the oxygen evolution kinetics of epitaxially grown La
1-xSr
xCoO
3-δ (LSC) thin films. They tested the tracer depth distribution of LSC films under tensile and compressive strains on different substrates and found that the strain state significantly affects oxygen exchange kinetics.
In summary, ToF-SIMS technology is highly attractive for studying the dynamic changes in elemental distribution from the surface to the bulk depth profile of chemical looping oxygen carriers.
With the deepening research on catalytic materials, devices based on advanced synchrotron radiation sources are becoming powerful tools for exploring the microscopic structures of materials, spanning the synthesis, structure, and performance studies, greatly advancing the research progress in material catalytic mechanisms
[52]. Yuan Nini
[53]used XAFS characterization technology to investigate the influence of different A-site element substitutions on the chemical states of Fe in AFeO
3oxygen carriers, finding that typical metal element substitutions at the A-site resulted in significant differences in the electronic properties of the oxygen carriers. Tada et al.
[47]successfully employed this technique to image and visualize the spatial distribution of Ce
3+and Ce
4+in the Ce
2Zr
2O
xsystem during oxygen storage/release cycles, identifying the non-uniform diffusion patterns of oxygen.
2.2 Experimental Design Method
2.2.1 Inert Marker Method
Currently, research on the oxygen storage and release mechanism of oxygen carriers mainly focuses on the types of oxygen species and active components on their surfaces. However, for chemical looping reactions, lattice oxygen within the bulk phase is the primary oxygen source, and the reaction process is often accompanied by the migration of metal ions within the oxygen carrier. Yet, studies on the evolution of internal materials during the reaction process remain scarce. Cutting the oxygen carrier and analyzing the cross-section to obtain static results is currently the only feasible solution. Inert tracer experiments are a simple and effective method commonly used to determine the solid-phase ion transfer mechanism in gas-solid reactions
[54]. Specifically, an inert material Pt marker strip is used as a marker on the surface of the solid reactant, serving as the initial gas-solid interface throughout the reaction process, as shown in
Figure 4a. A Pt marker strip is placed on top of the dense oxygen carrier, and after partial carbonization, the final position of this inert material is determined by the dominant solid-phase ion transfer mechanism (
Figures 4b~d).
The inert labeling method was used to study the ion migration mechanism of basic calcium-based oxygen carriers. Fan et al.
[54]explored the dominant ion diffusion mechanism during the formation of calcite from CaO and CO
2 using inert labeling experiments. They proposed that inward diffusion of CO
3 2-leads to the formation of CaCO
3, consistent with an inward growth pattern (
Figure 4d). Wang Junmin et al.
[55]employed this method to investigate the diffusion of Ca
2+ within the CaSO
4product layer during CaO-based sulfur fixation, with the diffusion model illustrated in
Figure 4c. Additionally, Sun et al.
[56]used this method to study the solid-phase reaction mechanism between CaO and HCl during chemical looping conversion, combining SEM and EDS to elucidate the ion migration behavior during the chlorination process. They determined that the reaction follows an inward growth pattern characterized by inward diffusion of Cl
-and outward diffusion of O
2- (
Figure 4d).
In addition, the inert labeling method has also been applied to study the ion migration mechanism and reaction mechanisms of oxide-based oxygen carriers in chemical-looping combustion. Li et al.
[57]visualized the phenomena of "Fe cation outward diffusion" and "O anion inward diffusion" during the chemical-looping conversion process of ilmenite by combining inert labeling with SEM. They also used inert labeling techniques combined with DFT to reveal the mechanism by which the addition of inert carriers enhances the performance of oxygen carrier particles.
In summary, although these studies did not employ in-situ methods, these static results are crucial for gaining an in-depth and systematic understanding of the spatial evolution characteristics of lattice oxygen migration and active components in the bulk phase.
2.2.2 Experimental verification method
Currently, research on the oxygen storage and release mechanism of oxygen carriers mainly relies on in-situ characterization techniques. However, the reaction temperatures in chemical looping processes are extremely high (generally above 700 ℃), which limits many in-situ characterization methods from achieving their intended goals. Additionally, in-situ characterization is expensive, undoubtedly increasing research costs. Therefore, novel experimental designs have emerged.
Rao et al.
[22]first proposed integrating active sites of oxygen-carrying species through light irradiation, and successfully tracked the anti-coking pathway during the DRM process by combining in-situ diffuse reflectance infrared Fourier transform spectroscopy with steady-state isotope transient kinetic analysis. They found that light irradiation can enhance the targeted pathway from CH
3 *to CH
3O
*on the Ni/CeO
2catalyst, enabling the catalyst to exhibit excellent stability under thermo-photocatalytic conditions at a low temperature of 472 ℃ for up to 230 hours.
Brandon et al.
[58]defined the crucial role of cerium lattice oxygen and, through detailed studies using methane pulse transient experiments, explored the surface reaction pathways of methane reforming over Ni/Gd-doped ceria-based oxygen carriers. They set a specific waiting time after each pulse, ensuring that the sample reached equilibrium before the next pulse. By analyzing the shape of the pulse peaks and the yield of products, they further elucidated the influence of different oxygen species on selectivity and resistance to carbon deposition. They proposed that the oxidation of carbon species involves both surface oxygen and bulk-phase migrating oxygen, with the rate-limiting step for the second oxidation being the migration of oxygen from the bulk phase to the surface. They suggested that if the bulk-phase oxygen has sufficient time to migrate to the surface, no carbon deposition will occur.
Marek et al.
[21]designed temperature-programmed isotope exchange (TPIE) experiments based on Co
3O
4, Co
3O
4-CeO
2, and CeO
2oxygen carriers, and combined these with the isotopic transient kinetic analysis (ITKA) method to investigate oxygen mobility during soot oxidation and obtain insights into the reaction mechanism. The results showed that Co
3O
4-CeO
2exhibited the highest lattice oxygen migration rate, and soot oxidation primarily involved surface-adsorbed oxygen from Co
3O
4and lattice oxygen from CeO
2, while Co
3O
4-CeO
2utilized both surface-adsorbed oxygen and lattice oxygen synergistically.
Researchers elucidated the reaction mechanism of oxygen carriers or catalysts in catalytic systems through logically rigorous experimental design, traced the migration pathways of bulk-phase active species, and thereby revealed the activation mechanism of bulk-phase active species in materials and their participation pathways in reactions.
2.3 First-principles calculation method (DFT)
A comprehensive understanding of ion diffusion in metal oxide composites is crucial for better design and application of chemical looping technology. Although many experimental methods can effectively observe and track changes in lattice oxygen and metal ions within the oxygen carrier bulk phase, research on the microscopic mechanisms of oxygen storage and release by oxygen carriers is severely limited because experimental characterization struggles to capture reactant transformations at extremely rapid timescales
[59-60]. Computational simulations play an important role in studying material surface structures and properties as well as chemical reaction mechanisms. In areas such as solid material structure simulation, property prediction, and in-depth understanding of reaction mechanisms, first-principles calculations can often provide results and insights that are unattainable through experiments
[61-63]. In recent years, DFT has been widely applied to mechanistic studies of oxygen carriers in chemical looping.
Among the various oxygen carriers, Cu-based oxygen carriers have been widely studied as a model due to their strong oxygen storage capacity, high reactivity, and environmental friendliness
[64]. Zhao et al.
[23]used DFT calculations to investigate the oxygen release energy barrier in Cu-based oxygen carriers. The results indicated that the desorption of O
2 molecules formed on the surface of the oxygen carrier is the rate-determining step for all CuO-supported oxygen carriers. The kinetic mechanism of CO oxidation on the CuO (111) surface has also been studied
[24],and energy barrier calculations revealed that the reaction energy barrier between CO and surface-adsorbed oxygen is as low as 0.106 eV, significantly lower than that for reactions involving surface lattice oxygen (0.999 eV). Therefore, they concluded that the presence of surface-adsorbed oxygen on the CuO (111) surface can enhance its catalytic activity for CO oxidation. Additionally, Varghese et al.
[65]reported the activation energy barriers and pathways for methane dissociation on different crystal faces of CuO. They pointed out that methane dissociation primarily relies on the synergistic effect of metal-lattice oxygen sites. In this process, the activation energy barrier for methane on the high-energy CuO (010) surface (60.5 kJ/mol) is much lower than that on the CuO (111) surface (76.6 kJ/mol). They attributed this significant reduction in the activation energy barrier to two factors: first, the stabilization of the transition state and the reduction of strain on the dissociating methane molecule; second, the stabilization of the dissociated co-adsorbed products, which favors the thermodynamic reaction.
Iron-based oxides such as LaFeO
3and Fe
2O
3, due to their simple structure and good chemical stability, are also used as oxygen carriers in DFT calculation models. Mishra et al.
[66]combined DFT calculations with experimental studies to investigate the correlation between the oxygen vacancy formation energy and redox performance of perovskite oxygen carriers. They calculated the oxygen vacancy formation energy of manganese-based oxygen carriers and found that the vacancy formation energy followed the trend: BaMnO
3>CaMnO
3>Ca
0.75Sr
0.25MnO
3>CaMn
0.75Fe
0.25O
3. They suggested that a lower oxygen vacancy formation energy leads to increased oxygen release; specifically, the presence of oxygen vacancies facilitates the adsorption and desorption of oxygen, thereby accelerating the redox reaction. Feng et al.
[67]focused on studying the differences in oxygen vacancy formation energy and oxygen migration barriers between the surface and bulk phases of LaFeO
3(ABO
3-type) perovskite oxygen carriers. They calculated the oxygen vacancy formation energy and oxygen migration barriers at different distances from the surface of the oxygen carrier and found a significant difference between the near-surface and deep-layer oxygen vacancy formation energies, with a deviation of up to 0.95 eV. This is attributed to the strong binding of atoms in the bulk phase by surrounding atoms, making it difficult for vacancies to form or migrate. Cheng et al.
[68]conducted DFT+U calculations and experimental studies on the partial oxidation of methane over Fe
2O
3oxygen carriers to produce syngas. By constructing more oxygen vacancies, they calculated the activation barrier for methane C—H bonds and found that when the oxygen vacancy concentration increased by about 2.5%, the activation barrier for C—H bonds decreased by 0.54 eV.
In recent years, DFT calculations have been increasingly applied to elucidate the oxygen release mechanisms of oxygen carriers with well-defined crystal structures. For example, Gao et al.
[69]elaborated on the reaction mechanism of La
0.8Sr
0.2FeO
3(LSF)@LiBr oxygen carrier in the oxidative dehydrogenation (ODH) of n-butane using chemical looping, based on ab initio molecular dynamics (AIMD) calculations and
18O
2exchange experiments. The shift of the LSF (110) peak observed in in-situ XRD during the ODH reaction corresponds to the transformation between Fe
4+and Fe
3+, which they attribute to the conversion of active lattice oxygen into peroxide species in the form of Li
2O
2. AIMD simulations further indicate that Li
2O
2can oxidize LiBr to atomic bromine, which then acts as a reaction intermediate involved in C—H bond activation. Additionally, through
18O
2exchange experiments, they found that the surface oxygen exchange rate of LSF@LiBr is significantly lower than that of unmodified LSF, suggesting that molten LiBr inhibits the oxygen release capability of LSF, thereby suppressing the formation of non-selective CO
2.
3 Study on the lattice oxygen migration mechanism during oxygen storage and release
As is well known, chemical looping conversion technology using oxygen as an intermediate relies on complex redox reactions, including the adsorption and dissociation of molecules on the oxygen carrier surface, as well as the generation and annihilation of oxygen vacancies. At its core, however, lies the migration and transformation of lattice oxygen
[70].In other words, the intrinsic mechanism of this reaction involves the release and recovery of lattice oxygen. The transformation of lattice oxygen is accompanied by electron migration; the migration processes of lattice oxygen and electrons within the oxygen carrier are illustrated in
Figure 5. During reduction, oxygen carrier particles absorb thermal energy at high temperatures, generating active oxygen species. When these particles are exposed to a reducing gas, oxygen ions are lost from the surface, creating oxygen vacancies. Under the influence of the chemical potential gradient, lattice oxygen migrates from the interior of the oxygen carrier to the surface, further participating in the reaction. At this point, reverse electron flow maintains overall charge balance. During oxidation, oxygen ions are supplied to the surface of the oxygen carrier, subsequently migrating inward to complete the full oxidation process. Meanwhile, internal electrons migrate outward to fill surface electron vacancies, maintaining the charge balance of the oxygen carrier. The supply of lattice oxygen by the oxygen carrier determines the reaction rate and product selectivity of chemical looping. Therefore, studying the release-recovery pathway of lattice oxygen is of great significance for understanding the reaction mechanism of chemical looping
[11,71-72]..
Currently, research on oxygen carriers mainly focuses on single-component metal oxides, composite metal oxides, and non-metal oxides. However, single-metal oxides have certain limitations due to inherent disadvantages of the metals themselves
[73-76]; most non-metal oxides exhibit poor oxygen-carrying capacity and suboptimal cycling performance, resulting in limited variety
[77]and limited recent research interest. Composite metal oxygen carriers generally demonstrate superior reactivity and cycling performance, making them one of the mainstream directions in current oxygen carrier research and development
[78-80]. Composite metal oxygen carriers can be broadly classified into two categories: one includes composite oxygen carriers with special structures achieved through metal doping, such as spinel-structured and perovskite-structured carriers, which have garnered significant attention in recent years due to their unique configurations; the other category involves several composite metals doped in specific proportions, where the synergy between different metals after doping results in excellent oxygen transport, reactivity, and stability. Therefore, this article primarily reviews the research progress on lattice oxygen migration and transformation mechanisms in composite metal oxygen carriers.
3.1 Lattice Oxygen Migration Mechanism of Spinel-Type Oxygen Carriers
Spinel-type composite metal oxides are widely used oxygen carriers with the general formula AB
2O
4. They exhibit excellent high-temperature stability, and their redox properties can be tuned by adjusting the elements at the A and B sites, making them a subject of extensive research
[81].
Huang et al
[14]To clarify the release-absorption pathway of lattice oxygen in NiFe
2O
4oxygen carriers, we combined XRD analysis to thoroughly discuss the phase evolution of the oxygen carrier during reduction and oxidation processes. The intuitive reduction pathway of the oxygen carrier (i.e., the lattice oxygen release pathway) is shown in
Figure 6a. They suggested that as the reduction degree (
X) increases, lattice oxygen in the oxygen carrier is gradually released, and the Fe/Ni spinel structure progressively disintegrates until an Fe
0.64Ni
0.36alloy phase is formed. This indicates that due to the presence of multivalent metals and the complex spatial structure, the reduction of NiFe
2O
4is a step-by-step process. Subsequently, using XPS characterization, they found that during the reduction of NiFe
2O
4, lattice oxygen at the chemical reaction interface forms hydroxyl ions, as illustrated in
Figure 6b. Specifically, during the reduction of the oxygen carrier with H
2as fuel, H
2first adsorbs onto the surface of active oxygen carrier particles and then dissociates into two hydrogen radicals (H•). Surface lattice oxygen (O
2-) readily transforms into highly reactive chemisorbed oxygen species (O
-/O
2 2-) at the chemical reaction interface
[82]. Consequently, hydrogen radicals combine with chemisorbed oxygen at the reaction interface to form hydroxyl ions (OH
-). In other words, during the oxygen release process, the oxygen carrier undergoes a series of transformations involving the formation and conversion of hydroxyl ions, ultimately releasing water (H
2O) and its dissociated H
2.
Song et al.
[40]thoroughly investigated the oxygen properties near the surface of NiFe
2O
4 oxygen carriers by precisely adjusting their reduction and oxidation states, using XPS combined with XRD. The results indicated that, in the early stage of the reduction reaction, the lattice oxygen content on the carrier surface remained relatively stable. Due to the abundant lattice oxygen within the bulk phase and the concentration gradient between the surface and the bulk, lattice oxygen from the bulk migrated toward the surface. This process was accompanied by the partial reduction of certain Ni species in the carrier into free metallic nickel, which became separated from the spinel lattice of NiFe
2O
4 (
Figure 7a), simultaneously fixing the gas-solid reaction interface at the carrier surface. They also found that even when the carrier surface was fully reduced and contained free metallic Fe and Ni, a small amount of lattice oxygen always remained (
Figure 7b). During CO
2 oxidation, the metallic Fe and Ni phases more readily adsorbed CO
2, converting it into lattice oxygen. Therefore, throughout the entire redox cycle, an active interface composed of lattice oxygen and corresponding metal ions consistently existed (
Figure 7c). Later, they
[19] employed in-situ electron microscopy to further visually reveal the dynamic evolution of lattice oxygen migration in NiFe
2O
4 during chemical looping conversion (
Figure 7d). They observed that the migration rate of lattice oxygen was rapid in the early stage of reduction, but slowed down as the reaction progressed and Fe—O bonds began to break. Notably, at the end of the reduction process, a stable oxide layer composed of lattice oxygen and metal cations (Fe) formed on the carrier surface. During CO
2 oxidation, driven by the lattice oxygen concentration gradient, oxygen anions migrated from the surface into the fully reduced bulk phase of the carrier. Experimental results combined with DFT theoretical calculations demonstrated that the migration pathway of metal atoms in the carrier depended on the rate of lattice oxygen release and recovery. These findings underscore the importance of regulating the lattice oxygen release-recovery rate to maintain the structural integrity and reactivity of the oxygen carrier.
Zhao et al.
[83]used the CH
4temperature-programmed reduction method to test the reactivity of MnCo
2O
4/SiC oxygen carriers with CH
4,and investigated the changes in various oxygen species during the reaction. They found that surface active oxygen first activates the C—H bonds in CH
4,resulting in the oxidation of a small amount of CH
4into CO
2and H
2O, indicating that surface oxygen largely determines the ignition temperature for CH
4combustion, while lattice oxygen directly influences the reactivity of manganese-based oxygen carriers. Zhou et al.
[84]took MnFe
2O
4oxygen carriers as the research object and studied their lattice oxygen migration behavior in the CH
4-CO
2reaction system. Changes in surface elemental composition, metal oxidation states, and adsorbed oxygen types of a series of oxygen carriers with different degrees of reduction revealed the lattice oxygen migration mechanism: lattice oxygen in the bulk phase of the oxygen carrier gradually dissociates and migrates to the surface during the reaction, directly participating in CH
4activation or transforming into adsorbed oxygen.
Jin et al.
[85]studied the combination of NiO with different support materials at 600 ℃ and found that NiO/NiAl
2O
4exhibited outstanding regeneration capability. To investigate its underlying mechanism, Readman et al.
[86]used H
2and CH
4as fuel gases and O
2as the oxidizing gas, employing thermogravimetric analysis to study oxygen migration and in-situ XRD analysis to examine phase transformations of the oxygen carrier. The results indicated that when oxygen transport to the particle surface is not rate-limited, rapid reduction occurs on the surface of the oxygen carrier, with approximately 60% of NiO being reduced. This is followed by a slower reduction process, which is caused by the limitation of oxygen species transport through the bulk particles.
Miller et al.
[15]studied the mechanism of CaFe
2O
4oxygen carrier in the coal chemical-looping gasification process. As shown in
Figure 8, the results indicated that the initial reduction of CaFe
2O
4occurred at the Fe
(8)O
6and Fe
(7)O
6octahedral sites, which was associated with morphological changes on the (400) and (212) crystal planes. After the initial reduction, the remaining oxygen at the Ca-containing Fe-O sites (such as Fe-O-Ca) participated in the reduction reaction, leading to a decrease in Raman intensity of the corresponding substances. The appearance of CaO and Fe
0indicated that deep reduction resulted in the removal of oxygen from the Fe-O-Ca bonds. They suggested that Ca not only controlled the movement of oxygen ions within the solid but also regulated the oxidation rate of carbon in the char, thereby causing incomplete oxidation and the production of CO.
The migration energy barrier of lattice oxygen is crucial for its migration mechanism. Liu et al.
[87]studied the migration of lattice oxygen on the CuFe
2O
4(100) and CuO (111) surfaces. The computational results indicated that bulk lattice oxygen tends to migrate toward the surface, and this migration process involves a certain energy barrier that requires external energy input. Zhang et al.
[88]investigated the oxygen decoupling mechanism of copper-based oxygen carriers using DFT. Their computational results showed that oxygen formation and desorption are the rate-determining steps. The migration energy barrier of lattice oxygen in CuAl
2O
4is significantly higher than that in CuO, and the process is endothermic. They suggested that this is the reason for the reduced oxygen release rate of the oxygen carrier, thus requiring a higher reaction temperature for oxygen release.
Filip et al.
[89]explored the roles of surface active oxygen (surface mechanism) and lattice oxygen (intra-interface mechanism) in the catalytic combustion of CH
4 on cobalt spinel nanocubes using steady-state isotope transient kinetic analysis (SITTKA), temperature-programmed surface reaction (TPSR), and first-principles thermodynamics combined with molecular simulations. They found that from 300 ℃ to 450 ℃, the surface Langmuir-Hinshelwood mechanism dominates, with the catalyst's stoichiometry remaining unchanged and CH
4 activated by single-atom oxygen O—Co species. In the temperature range of 450 ℃ <
T < 650 ℃, both the Langmuir-Hinshelwood and Mars-van Krevelen mechanisms may operate simultaneously due to the coexistence of stoichiometric and oxygen vacancy defect sites at equilibrium within this temperature range. Above 650 ℃, vacancy formation extends throughout most of the cobalt spinel structure, and the significant involvement of lattice oxygen indicates that the Mars-van Krevelen mechanism becomes dominant.
It is not difficult to find that the above studies generally agree that surface oxygen in spinel-type oxygen carriers is activated first, followed by the gradual dissociation and migration of bulk lattice oxygen to the carrier surface to participate in the reaction. During the oxygen release process, the spinel structure gradually collapses. However, the migration of bulk lattice oxygen is subject to transport limitations, thus requiring external energy input, such as high temperature. The mechanisms of different oxygen species have also been elucidated: surface active oxygen primarily determines the ignition temperature of reactants, while bulk lattice oxygen mainly influences the selectivity of the reaction.
3.2 Lattice Oxygen Migration Mechanism of Perovskite-Type Oxygen Carriers
Perovskite-type oxygen carriers
[90-91]possess a stable crystal structure and high redox activity, making them an excellent class of oxygen carriers. Their general formula is ABO
3,where the A site is occupied by rare earth or alkaline earth metals, and the B site by transition metals. In their structure, the A-site metal is coordinated with 12 oxygen atoms, while the B-site metal is coordinated with 6 oxygen atoms
[92]. The lattice oxygen in perovskites can participate in chemical looping reactions, exhibiting outstanding oxygen supply capability and thermal stability. Compared to single-metal oxides that provide and replenish lattice oxygen through phase transitions, perovskites demonstrate a unique oxygen supply capacity
[93].
Zhao et al.
[94]used in-situ XPS to analyze the changes in the surface oxygen concentration and active component content of La
1.6Sr
0.4FeCoO
6oxygen carrier during the reduction reaction. The results showed that chemical adsorbed oxygen gradually increased, while surface lattice oxygen gradually decreased as the reduction progressed. Therefore, they concluded that oxygen migrated from the bulk phase to the surface, with most of it becoming activated and participating in the reduction reaction of CH
4, as illustrated in
Figure 9. Meanwhile, some lattice oxygen transformed into chemically adsorbed oxygen, which is more stable, leading to a continuous decrease in surface lattice oxygen and an increase in chemically adsorbed oxygen.
Galinsky et al.
[95]studied the mechanism by which lanthanum strontium ferrite perovskite (LSF) as a support enhances the redox performance of Fe
2O
3, where Fe
2O
3serves as the primary oxygen-carrying material for lattice oxygen storage, and LSF, acting as a mixed ionic-electronic conductor (MIEC), facilitates the conduction of O
2-and electrons/holes. The oxygen release mechanism is illustrated in
Figure 10: active O
2-is transferred from the bulk phase of Fe
2O
3to the surface of the oxygen carrier via the support. Simultaneously, electrons released from O
2-on the surface are guided back to the primary oxygen-carrying material (Fe
2O
3) to maintain charge balance, thereby promoting efficient oxygen consumption by the primary oxygen-carrying material during redox reactions.
Yuan et al.
[96]synthesized a series of strontium-doped perovskite-structured oxygen carriers (designated as Sr
xCa
1- xFeO
3). Through experiments and DFT calculations, they investigated the performance changes of CaFeO
3 with different Sr doping levels and the oxygen release mechanism of Sr
xCa
1- xFeO
3. They concluded that the reduction process of Sr
0.4Ca
0.6FeO
3can be divided into two stages: first, it is reduced to SrFeO
3- δat a high reaction rate around 350 ℃; subsequently, at 400~900 ℃, it proceeds at a relatively lower reaction rate to form the CaO phase. As the reduction temperature and extent increase, the intensity of the CaO diffraction peak also increases, indicating an increase in the amount of the reduced phase and a gradual collapse of the perovskite structure. Additionally, they found that with increasing Sr doping, the types of oxygen on the surface of the oxygen carrier transition from lattice oxygen (O
1) to defect oxygen (O
d), as shown in
Figure 11. This suggests that Sr doping affects the distribution of oxygen and the content of Fe
3+, thereby promoting the release of lattice oxygen.
Surface oxygen content and lattice oxygen mobility are key factors in determining the reactivity of oxygen carriers. Li et al.
[71]used H
2-TPR to study the oxygen release behavior of LaFeO
3doped with different elements, as shown in
Figure 12a. The results indicate that the H
2-TPR curve of LaFeO
3is nearly flat, suggesting minimal surface oxygen adsorption and limited lattice oxygen mobility. In contrast, the H
2-TPR curves of LaCu
0.5Fe
0.5O
3and LaCo
0.5Fe
0.5O
3exhibit a distinct reduction peak in the low-temperature region, which is attributed to the simultaneous reduction of surface-adsorbed oxygen and some bulk lattice oxygen. This indicates that the bulk lattice oxygen within the oxygen carrier particles can rapidly migrate to the surface to fill oxygen vacancies when surface oxygen is consumed, demonstrating high lattice oxygen mobility
[97-98]. The same phenomenon was observed in the study by He et al.
[99](
Figure 12b), where La
1- xSr
xFeO
3- δ(
x=0.5 and 0.9) exhibited a single reduction peak. They attributed this to the rapid migration of bulk lattice oxygen from within the oxygen carrier particles to the surface to replenish oxygen vacancies as surface oxygen was consumed (
Figure 12c), further confirming high lattice oxygen mobility. Additionally, they pointed out that doping significantly affects the valence states of B-site ions and the distribution of oxygen.
Zhang et al.
[100]demonstrated the excellent oxygen release capability of iron-nickel-based (LFN) perovskite oxygen carriers in the partial oxidation of CH
4, with the oxygen release mechanism illustrated in
Figure 13. They pointed out that during the reaction, lattice oxygen from the bulk phase migrates outward, and as the perovskite oxide is reduced, Ni gradually separates from the bulk phase and forms metallic aggregates (such as clusters or particles) on the surface. These metallic aggregates act as active sites, accelerating methane dissociation. Meanwhile, the generated H radicals spill over onto the surface of the oxygen carrier and combine with lattice oxygen to form H
2O. Subsequently, H
2O diffuses via the gas phase to the metal surface, where it reacts with deposited carbon to produce CO and H
2. The selectivity of the products depends on the real-time lattice oxygen concentration and oxygen release capability of the oxygen carrier.
Lee et al.
[101]prepared LaCoO
3and B-site substituted LaCo
0.6B
0.4O
3(B=Fe, Mn, Ni), investigating the oxygen cycling mechanism of transition metal-substituted perovskite materials in methane chemical looping steam reforming. The results of cyclic performance tests indicated that Fe- and Mn-substituted perovskites exhibited superior regeneration performance, whereas LaCoO
3was not completely oxidized by steam. O
2-TPD test results also showed that the oxygen mobility increased in Fe- and Mn-substituted samples; thus, during oxidation of the oxygen carrier, surface oxygen species could readily transfer into the lattice to regenerate lattice oxygen. Additionally, they used in-situ XPS (
Figure 14a, b)and O
2-TPD to find that Fe substitution enhanced the adsorption and decomposition of OH
‒on the perovskite material surface, as well as the oxygen mobility from the surface to the lattice, which facilitated steam decomposition and regeneration of lattice oxygen. H
2-TPR test results revealed that LaCo
0.6Fe
0.4O
3had a slower lattice oxygen migration rate. They suggested that the outward migration of lattice oxygen due to Fe substitution was insufficient to fill surface oxygen vacancies, thereby promoting partial oxidation of CH
4(
Figure 14c).
Chang et al.
[102]used Al-element B-site doping in LaFeO
3to construct asymmetric Fe-O-Al motifs, and revealed at the electronic level the mechanism by which this mismatch enhances methane activation and inhibits peroxidation. Through calculations of the oxygen vacancy formation energy (
E v), they found that Al doping can increase the oxygen migration rate within the Fe-O-Fe motif, while the Al—O interaction effectively stabilizes lattice oxygen. Bader charge analysis and charge transfer energy calculations showed that the introduction of Al provides more electrons to neighboring lattice oxygen, while simultaneously enhancing Fe—O covalency.
Xia et al.
[103]synthesized LaFe
0.8M
0.2O
3(M=Al, Ga, Fe, Sc) oxygen carriers via B-site doping to investigate the correlation between structural distortion and lattice oxygen activity. They found through
57Fe-Mössbauer spectroscopy that the isomer shift (IS) of Fe cations gradually increased as the B-site cation radius decreased, indicating an increase in the d-electron density of Fe
3+. Therefore, doping with elements having smaller ionic radii at the B site can enhance the Fe—O interaction. Additionally, their calculations of the tolerance factor (
t) showed that Al doping, which has the smallest ionic radius, resulted in the least tilting of the FeO
6octahedra, significantly enhancing the covalency of the Fe—O bond and reducing the formation energy of oxygen vacancies, thereby facilitating reactant activation and accelerating oxygen migration.
Compared to the chemically inert A-site atoms, most researchers generally attribute the excellent oxygen activity of perovskite structures to the B-site transition metals, believing that A-site atoms can only indirectly regulate oxygen activity by altering the crystal structure or adjusting the valence states of B-site atoms. For a long time, the potential role of A-site atoms in directly controlling oxygen activity has been overlooked. However, He et al.
[104]found that reducing the La/Fe ratio at the A-site or surface reconstruction of LaFeO
3 through redox treatment can both create surface La defects. Experimental and theoretical calculations indicate that the absence of subsurface La
sub-O interactions can reduce the electron density of surface oxygen, increase oxygen mobility, and lower the activation energy barrier for methane. Additionally, Gao et al.
[105]developed an electronic engineering strategy induced by A-site vacancies and synthesized oxygen carriers with different La vacancies (V
La), namely LaFe
0.5Al
0.5O
3 (1.5M-LFAO, 3M-LFAO). Through characterization methods such as XRD refinement and FT-EXAFS spectroscopy, they found that 3M-LFAO had the highest V
Lacontent but the lowest V
Ocontent. They suggested that V
Lapromotes the formation of the Fe
4+-O
2--Fe
3+configuration. Subsequent temperature-programmed oxygen isotope exchange experiments confirmed that the Fe
4+-O
2--Fe
3+configuration facilitates the activation of lattice oxygen within the bulk phase. Furthermore, they pointed out that this configuration promotes the hybridization of Fe 3d-O 2p orbitals and the DE mechanism in perovskites. Jiang et al.
[106]reported research on how changes in A-site lanthanide elements affect the electronic structure and subsequently impact syngas production rates. By performing DFT calculations on charge transfer energies of perovskites in k-space, they found that GdFeO
3>SmFeO
3>PrFeO
3>LaFeO
3. Therefore, a decrease in the A-site cation radius leads to severe geometric tilting of the FeO
6octahedra, weakening the Fe-O orbital hybridization in real space and also reducing the Fe—O covalency in k-space, resulting in decreased oxygen mobility and surface oxygen activity.
In summary, the current consensus on the lattice oxygen migration pathway in perovskite-type oxygen carriers is that it migrates from the bulk phase to the surface. Specifically, oxygen species on the surface of the oxygen carrier are preferentially consumed, creating numerous oxygen vacancies, which in turn promote the migration of lattice oxygen from the bulk phase to the surface to replenish them, enabling continuous redox reactions. Most scholars believe that lattice oxygen reacts only after migrating to the surface. For perovskite-type oxygen carriers, the B-site elements have a greater influence on the carrier's activity; therefore, researchers mostly focus on doping the B-site with various elements to regulate the migration rate of lattice oxygen and the distribution of oxygen species, thereby improving product selectivity. Carbon deposition in perovskite oxygen carriers does not depend on the electronegativity of the B-site elements but rather on whether there is sufficient mobile lattice oxygen available. Additionally, the discovery that directly regulating the A-site atoms can modulate oxygen activity provides new insights into the study of lattice oxygen mechanisms in oxygen carriers and offers fresh perspectives for designing highly efficient oxygen carriers.
3.3 Other Metal-Based Oxygen Carriers: Lattice Oxygen Migration Mechanisms
In addition to oxygen carriers with specific structures such as spinel and perovskite types, many other types of oxygen carriers also exhibit excellent oxygen supply capabilities in chemical looping applications. It is equally important to study the mechanisms of their oxygen components and the migration of bulk-phase oxygen species.
Chen et al.
[11]used CH
3SH as a model reactant to investigate the relationship between cerium-based oxygen carriers' activity and surface lattice oxygen as well as bulk lattice oxygen. They performed linear fitting of CH
3SH conversion rate against the relative content change of surface lattice oxygen (O
S-L) species, and found that O
S-Lplays a major role in the model reaction and is not adsorbed oxygen. In addition, their designed H
2-TPR provided direct evidence for the dynamic migration of bulk lattice oxygen (O
B-L). Furthermore, by employing H
2-TPR, XPS, and analysis of reaction products, they confirmed that the migration of O
B-Lto replenish O
S-Land subsequently participate in the decomposition of CH
3SH can enhance catalytic stability.
Penkala et al.
[107]explored the oxygen absorption/release behavior of
18O-doped CeO
2under reaction conditions using Raman spectroscopy experiments and isotope-labeled pulse temperature-programmed oxidation combined with mass spectrometry analysis. The results demonstrated that quantitative exchange of
18O lattice oxygen occurs only above 400 ℃, while below 400 ℃, significant amounts of loosely bound water are released. Wang et al.
[108]investigated the contribution of oxygen vacancies to CO
2activation during methane dry reforming over CeO
2-based solid solutions by combining transient response experiments with numerical simulations (
Figure 15). Using a combination of transient response experiments and numerical simulations, they calculated the concentration and reactivity of surface or bulk oxygen vacancies. They suggested that trivalent ion substitution for Ce
4+increases the number of oxygen vacancies, which serve as active sites for the direct dissociation of CO
2, thereby facilitating the replenishment of oxygen vacancies in doped CeO
2with more gas-phase oxygen atoms from CO
2. This leads to high CO
2activation efficiency, with oxygen atoms derived from CO
2participating both in the oxidation of deposited carbon and in the activation of CH
4and H
2.
Kang et al.
[109]first reported and compared the lattice oxygen evolution behavior of cubic Cu
2O model catalysts in chemical looping combustion (CLC) reactions. Through in-situ infrared spectroscopy, isotopic (
18O
2) transient exchange experiments, and DFT simulations, they found that bulk lattice oxygen (
16O
2) from Cu
2O was replaced and activated by gaseous
18O
2, forming C
16O
18O and C
18O
2. This effectively verified that the lattice
16O
2-in the oxygen vacancies of Cu
2O was efficiently replaced by
18O
2-, and the rapid appearance followed by complete disappearance of
16O
2during the oxidation of reduced Cu
2O by
18O
2demonstrated the stable migration of bulk lattice oxygen to the surface.
Gong et al.
[110]suggested that the type of oxygen on the oxygen carrier determines the performance of the chemical-looping partial oxidation of methane, and described the dynamic migration and reaction of oxygen on the nickel-cerium oxygen carrier. Experimental results showed that in the early stage of the reaction, Ni-O species were rapidly consumed, leading to the complete oxidation of methane to CO
2. Subsequently, the active Ni-O-Ce species dominated the partial oxidation of methane, enhancing syngas production. They also found that the CO generation rate was linearly correlated with the surface oxygen content of Ni-O-Ce. Furthermore, lattice oxygen from CeO
2would migrate through the bulk phase to replenish the consumed Ni-O-Ce oxygen species.
Chen et al
[111]studied the lattice oxygen migration of Ca
2Fe
2O
5/Zr
0.5Ce
0.5O
2during redox processes. During reduction, lattice oxygen reacts with hydrogen radicals, leading to a decrease in lattice oxygen content and an increase in oxygen vacancies. As lattice oxygen is released, the spatial structure of Ca
2Fe
2O
5is disrupted and transforms into Fe
8Ca
8O
20with oxygen defects due to lattice distortion. The exothermic reaction between H
2and Ca
2Fe
2O
5promotes the formation of Fe
H, which has a hexagonal close-packed structure. After 160 minutes of reduction, the unstable Fe
Hcompletely transforms into structurally stable Fe
C, achieving complete reduction. During oxidation, Fe and CaO react with oxygen radicals to form Ca
2Fe
2O
5, and lattice oxygen is fully restored within 40 minutes. However, after multiple redox cycles, crystal phase transformations lead to instability in the spatial structure and sintering.
Li Wanying et al.
[112]used molecular dynamics simulations to investigate the interfacial reaction pathways and mechanisms of lattice oxygen transport in Mn
2O
3. They suggested that surface lattice oxygen, as the active oxygen species for methane activation, is transferred into the gas phase in the form of H
2O or CH
3OH, resulting in a concentration gradient of lattice oxygen on the surface of the oxygen carrier. Under the driving force of this concentration difference, lattice oxygen from the bulk phase migrates toward the surface and transforms into new surface lattice oxygen, which then participates in the interfacial reactions.
In summary, for other types of oxygen carriers, whether it involves the decomposition of CH3SH, the adsorption and oxidation of CO, or the dissociation reactions of H2 and CH4, surface oxygen species are preferentially consumed. Under the driving force of the oxygen concentration gradient, deep lattice oxygen migrates to fill surface oxygen vacancies and participates in chemical looping reactions. This conclusion is almost identical to that observed for spinel- and perovskite-type oxygen carriers.
4 Study on the Migration Mechanism of Metal Ions during the Oxygen Storage and Release Process
During the oxygen storage and release process in oxygen carriers, in addition to the migration of oxygen species, active metal cations also exhibit dynamic migration. Moreover, the diffusion and aggregation of cations can affect the lattice oxygen migration process
[113].Currently, studies on metal ion migration typically employ inert labeling techniques to mark the initial gas-solid or solid-solid interfaces
[20,54,56]. By observing the relative positional changes between the inert label and the product layer after the reaction, the dominant ion transfer mechanism can be identified. Based on the differences in migration rates between reactants from the gas phase migrating inward and those from the solid phase migrating outward, the "outward growth mode," "mixed growth mode," and "inward growth mode" of ion migration have been proposed
[55,114].
Studying the migration behavior of active ions in oxygen carriers is crucial for regulating the bulk structure and enhancing reaction rates. In recent years, numerous researchers have conducted extensive studies in the field of ion diffusion.
Li et al
[115]earlier studied the ionic diffusion behavior of Fe
2O
3using Al
2O
3and TiO
2as support materials, and found that the addition of the support significantly improved the reactivity and recyclability of the oxygen carrier particles
[116]. They argued that gas diffusion had little impact on the performance of the oxygen carrier, but rather the presence of the support affected the migration of ions within the bulk phase, with the diffusion of Fe
3+cations and O
2-anions playing a major role in the reaction performance of the oxygen carrier. Later, they combined inert tracer experiments with theoretical calculations to conduct a detailed study of the ionic diffusion behavior of pure Fe
2O
3and Fe
2O
3supported on TiO
2[57]. The study revealed that compared to pure Fe
2O
3, the addition of the inert support TiO
2significantly increased the pore volume of the oxygen carrier; however, after the first reduction reaction, its pore volume became almost identical to that of pure Fe
2O
3. Unlike pure Fe
2O
3particles, which rapidly deactivated, the reactivity of Fe
2O
3/TiO
2particles was not adversely affected by the reduction in pore volume caused by prior redox cycles. Therefore, they concluded that during the redox reactions of the oxygen carrier, the support could enhance the rate of ionic diffusion, and solid-state ionic diffusion might play a more important role than gas diffusion within the particles. By comparing SEM and EDS images, they observed that the product layer above the Pt marker on pure Fe
2O
3was significantly thicker and denser, leading them to believe that outward diffusion of iron cations appeared to be the primary mechanism for iron oxidation. Earlier literature
[117]also reported similar results, suggesting that the migration of iron cations was the sole factor driving iron oxidation. In contrast, Fe
2O
3/TiO
2exhibited an inward growth pattern throughout the oxidation process. They attributed this to the fact that the addition of the support created more oxygen vacancies, significantly enhancing the ionic diffusion rate of oxygen anions. At this point, the inward migration rate of oxygen anions far exceeded the outward migration rate of iron cations (
Figure 16), thus shifting the primary ionic transfer mechanism for iron oxidation from "outward diffusion of Fe cations" to "inward diffusion of O anions."
Based on the above research results, Zeng et al
[20]delved into the intrinsic mechanism by which TiO
2supports enhanced diffusion of oxygen anions. They suggested that the formation of defects in the TiO
2crystal structure due to Fe
3+and Fe
2+ions facilitates oxygen ion diffusion. Specifically, loading introduces lattice substitutional defects or interstitial additive defects in the TiO
2lattice; in other words, for every two Ti
4+ions replaced by two Fe
3+ions, one oxygen ion (O
2-) is vacated to maintain electrical neutrality within the crystal structure. If the vacancy concentration is below the critical level, the resulting oxygen vacancies will create point defects within the crystal structure. Wang et al
[118]determined the critical atomic concentration of Fe
3+in TiO
2to be 2%. In summary, it is precisely the presence of oxygen vacancies that enables rapid diffusion of oxygen ions within the crystal structure. However, such vacancies are absent in the pure Fe
2O
3lattice, severely restricting diffusion. Therefore, they concluded that pure Fe
2O
3follows an outward diffusion mechanism of iron cations, whereas iron-loaded materials follow an inward diffusion mechanism of oxygen anions
[57].
Knutsson et al.
[119]also investigated the ion migration mechanism of ilmenite oxygen carriers. They used epoxy resin to fix and polish particle cross-sections, obtaining elemental distribution maps of oxygen carriers with varying porosities. They found that as porosity increased, iron segregation in the particles became more pronounced. They concluded that iron migration results from diffusion processes within the particles, and the formation of iron-rich shells around the particles is beneficial to the chemical looping combustion process.
Sun et al.
[56]reported the ion migration mechanism of CaO-based oxygen carriers as adsorbents for dry removal of HCl from syngas or flue gas. To understand the dominant ion diffusion mechanism in the chlorination reaction under non-porous conditions, they conducted inert labeling experiments. The experimental results showed that a product layer of CaCl
2 formed between the inert Pt layer and the solid-phase oxygen carrier CaO bulk, indicating that the dominant ion diffusion mechanism is an "inward growth mode." In other words, the primary diffusion mode involves inward diffusion of Cl
- anions, where at the CaCl
2-CaO interface, Cl
- anions replace O
2- anions in CaO to form CaCl
2. To maintain local mass and charge balance, the replaced O
2- anions diffuse outward through the CaCl
2 layer and react with H
+ cations at the gas-solid interface to produce H
2O.
Su et al
[120]used thermogravimetric analysis for isothermal oxidation and DFT calculations to investigate the ionic migration characteristics of Cu-based oxygen carriers during the oxidation process. The results showed that the initial interfacial reaction energy barrier was significantly lower than the ionic diffusion energy barrier within the oxygen carrier, resulting in a surface reaction rate faster than the ionic diffusion rate. This led to the rapid formation of an oxide film (CuO) on the outer surface of the grains (
Figure 17 (P1)). Subsequently (
Figure 17 (P2)), due to the dominant outward diffusion of Cu cations, Cu
+ions in the subsurface layer migrated toward the outer surface to balance the copper concentration gradient, while Cu cations from deeper layers migrated upward to fill the copper vacancies in the subsurface layer (
Figure 17 (P3)). Meanwhile, the slightly excess Cu
1+ xO cations in the external layer continued to be rapidly oxidized by O
2molecules, leading to the formation of a new product layer on the exterior (
Figure 17 (P4)). As diffusion progressed, the reaction Cu
2O+V
Cu→CuO caused the Cu/O ratio in the "inner layer" to approach 1∶1, resulting in the formation of the bulk-phase CuO and completing the bulk-phase reconstruction of the oxygen carrier.
Currently, all identified metal ion migration mechanisms follow an outward growth pattern, which results in the formation of hollow structures in oxygen carriers, facilitating an increased reaction surface area and accelerating the rapid release of lattice oxygen
[121-122]. However, some studies also indicate that metal ion segregation may lead to accumulation and phase separation, which can cause sintering and deactivation of the oxygen carrier
[123].
Cuadrat et al
[124]studied the activity of ilmenite in chemical-looping combustion reactions. Their research found that although the pore structure on the surface of ilmenite became more developed with increasing cycle numbers, the reaction activity of ilmenite did not improve. They attributed this primarily to the enrichment of iron elements on the surface of the oxygen carrier. Chung et al
[125]addressed the issue of deactivation of Fe-Ti-based oxygen carriers due to elemental segregation by proposing the use of an aluminum-based framework for encapsulation, thereby preparing a core-shell structured composite oxygen carrier. After encapsulating the Fe-Ti-based oxygen carrier with an aluminum-based framework, although iron and titanium elements still separated, the confinement provided by the aluminum-based framework disrupted the long-distance migration pathways of iron, inhibiting its migration and enrichment toward the surface, thus effectively solving the problem of easy deactivation of the oxygen carrier.
The same phenomenon was also observed in the study by Tseng et al.
[126], who investigated Fe
2O
3-NiO oxygen carriers. They found that during oxidation, Fe ions tended to migrate to the surface of the oxygen carrier, resulting in a higher concentration of Fe ions on the oxidized surface compared to the fresh carrier. In contrast, during reduction, Ni ions exhibited stronger mobility, and after the reduction reaction, only Ni and O elements could be detected on the surface of the oxygen carrier. After 9 redox cycles, the oxidation/reduction capacity of the oxygen carrier continued to decline, which can be attributed to severe particle agglomeration at high temperatures. When conducting 20 cycles of redox reactions with Fe
2O
3-Al
2O
3, it was found that the reactivity of the oxygen carrier gradually decreased with increasing cycle number. Characterization analysis of the oxygen carrier after reactions revealed that the Fe content on its surface gradually increased with the number of cycles, and the sintering of Fe on the surface led to a decrease in the oxygen carrier's activity
[127].
Blas et al
[128]investigated the redox performance evolution in chemical looping combustion (CLC) reactions using CO as fuel and NiO/NiAl
2O
4 as the oxygen carrier. Characterization results showed that a nickel-rich layer (20 μm thick) formed on the surface of the oxygen carrier after the reduction reaction, and the particles partially sintered during regeneration in a 20% oxygen atmosphere; however, these phenomena were not observed on particles regenerated under low oxygen concentrations. During cycling, NiO could not return to its initial state, causing the surface NiO layer to become progressively thicker. The higher the temperature, the faster the diffusion rate of Ni
2+ cations, resulting in a thicker layer and gradual deactivation of the oxygen carrier.
5 Limitations of Research in the Process of Stored Oxygen Release
5.1 Limitations of the research methodology
In the study of oxygen storage and release mechanisms of oxygen carriers, characterization techniques and experimental design play an important role.
Although characterization techniques have achieved remarkable success in studying the chemical looping mechanism of oxygen carriers, several factors still hinder our understanding of the oxygen storage and release mechanisms of these carriers. For instance, the high-temperature operating environment of chemical looping systems limits the implementation of in-situ experiments. In-situ Raman and in-situ DRIFTS exhibit baseline drift and loss of peak signals at temperatures above 700 ℃; in-situ EPR cannot detect changes in material oxygen vacancies at temperatures exceeding 300 ℃; in-situ TEM requires extremely high thermal resistance from its heat-resistant gratings for high-temperature detection; NMR suffers from poor signal resolution at elevated temperatures; and nearly all detection techniques experience varying degrees of reduced signal accuracy under high-temperature conditions, placing higher demands on the thermal resistance and heat dissipation capabilities of characterization instruments.
The inability to quantify is also a major pain point limiting research into the mechanism of oxygen carriers. Many scholars have attempted to quantitatively explain the mechanism of oxygen carriers using elemental detection techniques
[129],thermogravimetric analysis
[130],and kinetic calculations
[131];however, it has been proven that existing elemental detection techniques can only measure elemental content at the material's surface or subsurface, failing to reveal deep bulk-phase information. Thermogravimetric analysis is not accurate for complex, multi-component reactions, and if there is any deviation in selecting the kinetic calculation model, it will be difficult to reflect the true reaction conditions. Capturing reaction intermediates remains challenging, especially under high-temperature conditions. Some reaction intermediates are unstable, prone to decomposition or reacting with the surrounding environment, resulting in a short lifespan. Current advanced characterization techniques lack the temporal resolution and sensitivity required to capture these types of reaction intermediates, causing us to miss much valuable information in mechanistic studies.
Experimental design is indispensable in the study of oxygen carrier mechanisms. Techniques such as isotopic labeling, inert tagging, and DFT calculations have provided new research perspectives on lattice oxygen migration and can effectively complement characterization methods. However, many challenges still remain. For instance, experimental design for isotopic labeling is relatively difficult, requiring high sensitivity and accuracy from detection equipment. Inert tagging demands high density of the oxygen carrier, often necessitating high-temperature calcination to eliminate most pores and prevent gas diffusion from affecting experimental results, thus limiting its widespread application. DFT calculations rely on the rational selection of models and methods; different choices may lead to variations in computational results, and interpreting and validating these results also pose significant challenges.
5.2 Limitations of Mechanism Studies
Regarding the study of oxygen storage and release mechanisms in oxygen carriers, numerous scholars have already contributed abundant research findings. However, due to limitations in characterization techniques and the complexity of chemical reactions, challenges still remain in this field.
Currently, research mainly focuses on changes in surface oxygen species and active components of oxygen carriers. Studies on the activation, migration, and transformation of bulk lattice oxygen during chemical looping reactions are extremely limited, and there is still no direct evidence to reveal the reaction pathways of bulk lattice oxygen in oxygen carriers or the mechanism of action of active metal ions.
Research on the correlation between lattice oxygen activity and the selective regulation of target products is still in its early stages, and there is no consensus yet on the transformation and roles of bulk lattice oxygen, surface lattice oxygen, and adsorbed oxygen in chemical looping reactions. Moreover, the impact of inevitable reactant diffusion on the activation and migration of oxygen species has not yet been systematically studied, and there is still a long way to go for an in-depth understanding of oxygen species.
The role of oxygen vacancies during the adsorption and dissociation of reactants remains controversial. The concentration of oxygen vacancies is often used to determine the extent of oxygen species consumption, and a gradient in oxygen vacancy concentration may drive the migration of oxygen species. Some studies have also shown that oxygen vacancies can serve as active sites involved in reactant activation. However, there is still a lack of systematic understanding of oxygen vacancies, and insufficient evidence to clarify their critical role in chemical looping reactions.
The migration mechanism of metal cations in oxygen carriers remains inconclusive. Some research results
[121-122]suggest that cation diffusion can increase the reaction surface area and facilitate the rapid supply of lattice oxygen. However, other studies
[124-127]indicate that metal ion diffusion may accelerate the sintering-induced deactivation of oxygen carriers. In summary, research on the cation migration mechanism in oxygen carriers is still insufficient, and the interaction mechanism between oxygen species and metal ions remains unclear.
In summary, there is still a long way to go in studying the mechanisms involved in the oxygen storage and release process of oxygen carriers.
6 Conclusion and Outlook
Oxygen carriers are at the core of chemical looping technology, serving as carriers for oxygen and heat transfer and playing a crucial role in the chemical looping conversion process. The redox capability of the oxygen carrier itself determines the direction and extent of the cyclic reaction; for instance, its oxygen transfer capacity directly affects the efficiency of the chemical looping reaction and the degree of fuel conversion. Additionally, the oxygen carrier plays a dominant role in selectively controlling the products formed. The type and properties of the oxygen carrier directly influence the formation and distribution of products during the chemical looping reaction. By regulating the lattice oxygen characteristics or migration rate of the oxygen carrier and creating active sites, product selectivity can also be influenced. A deep understanding of the oxygen storage and release mechanism of oxygen carriers is essential for their design and preparation. In recent years, significant research findings regarding the active characteristics and oxygen storage and release mechanisms of oxygen carriers have been summarized as follows.
(1) Composite metal oxygen carriers, primarily based on spinel and perovskite structures, exhibit advantages such as high oxygen storage capacity, good reactivity, and strong stability. Currently, it is widely accepted that the surface oxygen species are consumed first, creating numerous oxygen vacancies. This results in an oxygen concentration gradient between the surface and the bulk phase, driving the migration of lattice oxygen from the bulk phase to the surface to replenish the oxygen vacancies, thus enabling the continuous progression of chemical looping reactions.
(2) The release rate of surface lattice oxygen and the migration rate of bulk lattice oxygen determine the location of the reaction interface between the fuel and the oxygen carrier. The surface of a deeply reduced oxygen carrier consists of an active reaction interface composed of oxygen and active components. It is generally believed that the gas-solid reaction interface is fixed at the surface of the oxygen carrier.
(3) Bulk lattice oxygen migration requires overcoming a certain energy barrier and is replenished by migrating to surface oxygen vacancies under external energy supply. Some of this oxygen directly participates in the activation and oxidation of fuels as surface lattice oxygen, while another portion is converted into surface-adsorbed oxygen.
(4) Solid-state ion diffusion within oxygen carriers is typically studied using the inert tracer method. The migration mechanism of metal ions during reduction follows an outward growth pattern, which leads to the formation of hollow structures in the oxygen carrier, thereby increasing the reaction surface area and accelerating the rapid release of lattice oxygen. However, after multiple redox cycles, the segregation of metal elements can result in deactivation of the oxygen carrier.
Although extensive research has been conducted on the oxygen storage and release mechanism of oxygen carriers, the mechanisms involving bulk lattice oxygen and active component migration still require further confirmation. Additionally, basic research needs to be strengthened in the following five areas.
(1) Improving characterization techniques. The limitations of characterization techniques are key factors restricting the study of oxygen carrier mechanisms. To capture intermediates in rapid reaction processes, it is necessary to develop characterization techniques with high temporal resolution, high spatial resolution, and high sensitivity, such as ultrafast laser spectroscopy and X-ray free-electron lasers. Secondly, developing high-temperature in-situ characterization techniques is also crucial. In-situ environmental transmission electron microscopy may be the most suitable method for studying the bulk migration of lattice oxygen and metal elements, allowing visualization of microscopic migration phenomena. In the future, it can be combined with pulsed isotope transient response, thermogravimetric analysis, mass spectrometry, and other techniques to provide direct experimental evidence for the migration mechanism of lattice oxygen.
(2) Focus on the bulk-phase research of oxygen carriers. Currently, there is limited research on the mechanisms underlying the spatiotemporal evolution of oxygen carriers in their bulk phase. Future research should focus on exploring the spatiotemporal evolution characteristics within oxygen carriers, providing direct evidence for the dynamic evolution of lattice oxygen and active ions in the bulk phase.
(3) Clarify the role of oxygen vacancies. Currently, research on oxygen vacancies is relatively abundant; however, it remains unclear whether oxygen vacancies serve as active sites or influence the migration of lattice oxygen and metal cations. Future studies should aim to clarify the role of oxygen vacancies.
(4) Elucidate the correlation between lattice oxygen and metal ions. Currently, the mutual influence mechanism of bulk lattice oxygen and metal cations during the oxygen storage and release process has not been systematically studied. Future research should focus on exploring the dynamic evolution patterns of these two components in chemical looping reactions and their mechanisms affecting product selectivity and reaction rates.
(5) Innovative experimental methods. Design novel oxygen-carrying structures, such as core-shell structures, carbon nanocage-encapsulated structures, and skeletal structures, to investigate their impact on lattice oxygen migration through structural design. Introduce plasma-assisted technologies to reduce the reaction energy barriers of oxygen carriers, and conduct in-situ characterization at moderate and low temperatures. Incorporate isotope labeling experiments, using characterization techniques to track the evolution of the bulk structure of isotope-labeled oxygen carriers; isotopic labeling can also be applied during chemical looping reactions, with mass spectrometry employed to trace the atomic distribution mechanism of reactants entering products.
the National Natural Science Foundation of China(52174279)
PDF
