Electrochemical Methane Conversion

Methane, as an earth‐abundant C1 resource, is a greenhouse gas as well as a key building block in the chemical industry. Electrochemical conversion of methane into fuels and valuable chemicals represents an attractive promise toward carbon neutralization and reducing CO2 emission in industrial methane reforming. To overcome the catalyst degradation and energy cost problems, it is critical to activate the CH bonds in methane effectively to operate under ambient conditions, while without the cost of product selectivity. This review focuses on catalyst structures and system design strategies in recent developments of electrocatalytic methane conversion progresses. The combination of electro‐, thermo‐, and photocatalytic methods can enable complementary and enhanced activities, as well as new insights in reaction conditions and mechanisms.


Introduction
Methane (CH 4 ) is the side product of oil exploitation and widely exists in natural gas, shale gas, and combustible ice. Attributed to its nature of abundance and low price, methane is becoming a critical precursor of chemicals and fuels for replacing coal and oil. [1] The utilization of CH 4 offers attractive potentials to decarbonize current petroleum industries. Although the extent of methane emissions is smaller than carbon dioxide (CO 2 ), CH 4 is more potent as a heat-trap gas, absorbing and emitting infrared radiation, with 86 times per mass unit as much as CO 2 . [2] Thus, compared with direct combustion or emissions of methane, the large-scale methane conversion into value-added chemicals benefits both environment and economy.
The tetrahedral configuration of methane shows a high symmetry and a low polarity. Meanwhile, the dissociation energy of the C─H bond is up to 439.3 kJ mol À1 . Compared with its valuable products (such as CO, CH 3 OH, C 2 H 4 ), the selective conversion of methane under ambient conditions has been attracting substantial interest from both academia and industry. [3] The direct methane conversion in the industry needs harsh reaction conditions (high temperature > 900 K and pressure > 2.5 bar) that are both energy and capital intensive. [4] It is possible to directly oxidize CH 4 into oxygenated products at temperatures below 500 K, whereas expensive oxidants, such as H 2 O 2 , NO x , and oleum, are needed. [5] The in situ generation of active oxidizing species in the electrochemical procedure is an alternative solution. In this Review, we focus on the strategy of electrochemical CH 4 activation and corresponding catalyst designs. The main article includes three sections that introduce current progresses and challenges of the electrochemical CH 4 conversion ( Figure 1). In the first section, we introduce the CH 4 conversion at low temperatures by elaborating catalyst designs and system selection. In the second section, we discuss the CH 4 usage at high temperatures combining thermocatalysis and electrocatalysis together to pursue high conversion rates and large current densities. In the third section, we focus on photochemical methods applied to electrochemical CH 4 oxidation to bring complementary advantages and new insights.

Low-Temperature Methane Electroconversion
Breaking C─H bond(s) requires a large activation energy that does not take place spontaneously. [6] Energy input, as well as oxidants and high-valent metal ions, can facilitate this process. [7] The redox intermediate is required to meet two demands: 1) strong activity to break C─H bond(s) and 2) limited side reactions with solvent and products. The recent representative research progress in methane electroconversion is shown in Table 1. Concentrated sulfuric acid is widely chosen for its broad electrochemical window that is compatible with high-valent metals. For example, Pd 2 III,III complex was generated in concentrated sulfuric acid electrolyte via electrochemical oxidation methods, [8] which presented a low C─H activation energy (25.9 kcal mol À1 ) and a high turnover frequency (TOF) of 2000 h À1 at 140 C and 500 psi of methane. The formation of dinuclear Pd 2 III complex followed an mechanism involving an initial electron transfer, a chemical step, and a subsequent electron transfer (Figure 2a). [8] Pd II was first oxidized to Pd III and then coupled to another Pd II forming the key intermediate of M─M-bonded Pd 2 II,III species, which were further oxidized into Pd 2 III . The structure of Pd 2 III dimer was a Pd─Pd-bonded center with five-coordinated oxygen atoms of H x SO 4 (xÀ2) for each Pd atom. [9] CH 3 OSO 3 H and CH 3 SO 3 H were the major products of methane oxidation, attributed to the polarity effect preventing further oxidation. The H x SO 4 (xÀ2) ligand from Pd 2 III dimer attracted H atom from CH 4 to generate methyl radical (CH 3 •), which acted as the rate-limiting step (Figure 2b). [10] CH 3 • could recombine with Pd 2 II,III center and produce CH 3 OSO 3 H after reductive elimination or CH 3 • reacted with SO 3 to form CH 3 SO 3 H. Liu and coworkers reported that vanadium (V)-oxo dimer reduced the activation energy to 10.8 kcal·mol À1 and achieved a high TOF of 1336 h À1 at 3 bar of methane. [11] The structure of V 2 V,V was a dimer with two terminal O atoms connected by a bridging oxo and two HSO 4 À ligands. In the process of methane activation, the limiting step was the removal of the first electron from O 2p orbital from the HSO 4 À ligand (Figure 2c). CH 3 • was generated by the same ligand from CH 4 , similar to the radical mechanism, but with a lower energy barrier.
As the usage of concentrated sulfuric acid can raise problems in equipment corrosion and products separation, more ambient electrolytes may benefit postprocessing and product purification. For instance, Park and coworkers incorporated Co 3 O 4 into ZrO 2 nanotubes for methane electro-oxidation in 0.5 M Na 2 CO 3 at room temperature, and a yield of 2,416 μmol g cat À1 ·h À1 (1-propanol and 2-propanol) was obtained at 1.6 V (vs. reversible hydrogen electrode, RHE) after 12 h. [12] In this system, a ZrO 2 structure with a large surface area was capable of good adsorption ability of carbonate ions, and Co 3 O 4 with rich surface oxygen vacancies could generate carbonate radical (CO 3 À •), which compared with hydroxyl radical (OH•) had a lower energy barrier. Acetaldehyde (CH 3 CHO), although was not detected in the product, was a crucial intermediate for the production of 1-propanol and 2-propanol. CO 3 À • broke the C─H bond to form CH 3 •, which subsequently combined with CH 3 CHO. Nucleophilic addition between CH 4 and the aldehyde group led to 2-propanol, and the radical addition generated 1-propanol ( Figure 2d). [13] Furthermore, Park and coworkers also synthesized a ZrO 2 : NiCo 2 O 4 quasisolid solution, and the methane oxidation efficiency remained at 47.5% after the 20 h test. [14] Compared with CH 4 , 1-propanol and 2-propanol were prone to oxidation and were converted into propionic acid and acetone within 20 h of the electrochemical test ( Figure 2e). Propionic acid was the final main product with the rate of 1,173 μmol g cat À1 ·h À1 . Another approach was using electron-rich organometallic compounds to break C─H bonds. While the low-valent metal centers seemed incompatible with O 2 and prevented catalytic air oxidation of methane, an electrochemical regenerating strategy was developed using silicon nanowire arrays to use O 2 gradient. At the bottom of silicon nanowire arrays, O 2 was consumed and the Rh II complex catalyst was regenerated from Rh III complex by electrochemistry. [15] Rh II complex reacted with dissolved CH 4 and formed Rh III catalyst-CH 3 complex. The Rh III  The direct conversion of CH 4 to CO 2 has been widely investigated in low-temperature methane fuel cells to generate electricity. Pt is an active transition metal catalyst under ambient conditions. The reaction mechanism and active sites of Pt have been studied, [16] and the CH 4 electro-oxidation is sensitive to the applied potentials. Although the CH 4 electro-oxidation is possible to take place at potentials more positive than 0.17 V versus RHE, within the H UPD region (0.05-0.40 V vs. RHE), the adsorbed hydrogen may significantly block active sites on the Pt surface so that CH 4 oxidation cannot proceed. By sweeping potentials to a more positive range, the oxidation peak centered at 0.75 V versus RHE corresponded to the stripping of CO*, an intermediate in CH 4 oxidation ( Figure 3b). CO* was eliminated by the OH* adsorption and CO─OH coupling. Density functional theory (DFT) calculations were carried out to estimate the binding energies of OH*, CO*, CH 3 *, H*, and CH 4 activation energy. The CH 3 * and H* binding energies and the methane activation energy were linear functions of the CO* binding energy. Catalysts with strong CO* binding and weak OH* binding energies were proposed to be desirable for methane electro-oxidation. Trewyn and coworkers reported that atomically dispersed active Pt organometallic complexes enhanced methane , Pd III , and Pd 2 II,III . Reproduced with permission. [9] Copyright 2020, American Chemical Society. b) The electrophilic C─H activation by a bound bisulfate (HSO 4 À ) ion. Reproduced with permission. [10] Copyright 2020, American Chemical Society. c) Calculated frontier orbitals involved in the turnover-limiting step and the proposed transition state of C─H activation step (left panel) and proposed catalytic cycle (right panel). Reproduced under the terms of the Creative Commons CC-BY license. [11] Copyright 2020, The Authors. Published by Springer Nature. d) Nucleophilic addition reaction of methane and acetaldehyde to form 2-propanol (up) and free radical addition reaction of methane and acetaldehyde to form 1-propanol (down). Reproduced under the terms of the Creative Commons CC-BY license. [13] Copyright 2017, The Authors. Published by Wiley-VCH. e) Production rate of propionic acid, 1-propanol, 2-propanol, acetone, and acetic acid after 5, 12, and 20 h (left panel), and product selectivity after 5, 12, and 20 h of electrochemical CH 4 oxidation by 0.5-ZrO 2 :NiCo 2 O 4 (right panel) in 0.5 M Na 2 CO 3 at 298 K. Reproduced with permission. [14] Copyright 2019, Elsevier B.V. fuel cell performances. [17] Supported by ordered mesoporous carbon, a maximum power of 403 μW mg Pt À1 was obtained at 80 o C. The single Pt site was coordinated by 2,2 0 -bipyridyl ligands and two chlorine (Cl) atoms or phenyl ( Figure 3c). Various 2,2 0 -bipyridyl served as anchors, and the substitution of Cl or phenyl was critical to performance. The phenyl-linked complex provided a power of 101 μW mg Pt

À1
, which rose sharply to 278 μW mg Pt À1 by replacing with Cl. Compared with phenyl, Cl maintained Pt in the þ2 state, and the heterolytic cleavage between Pt and Cl broke the C─H bonds in methane efficiently.
Another strategy is using a potential bias to trigger the surface property change. Sekine and coworkers reported that electric field could facilitate proton hopping on the surface. [18] Using 10 wt% NiÀ10wt%Mg─La 0.1 Zr 0.9 O 2Àx , the incorporation of Mg formed NiO─MgO solid solution on Mg─La 0.1 Zr 0.9 O 2Àx , which suppressed the reduction of NiO at low temperatures by making it more cationic. The reforming process did not occur without the electric field. With an input current of 3.0 mA, CH 4 reforming using CH 4 , H 2 O, CO 2 , and O 2 mixture managed to operate at 473 K, and the apparent activation energy was 8.2 kJ·mol À1 . Reproduced with permission. [15] Copyright 2019, American Chemical Society. b) Argon background-subtracted chronoamperometry during the methane activation phase in 0.5 M perchlroic acid (left panel) and cyclic voltammetry showing the oxidation feature after the electrode potential hold when purging with methane compared with the inert argon purge (right panel). Reproduced with permission. [16] Copyright 2019, American Chemical Society. c) Schematic representation of the general synthetic procedure for ordered mesoporous carbon-tethered single-site catalysts. Reproduced with permission. [17] Copyright 2016, American Chemical Society.

High-Temperature Methane Electroconversion
The C─H cleavage is facilitated at high temperatures, in which CH 4 is prone to be fully oxidized to CO 2 . Solid oxide fuel cells (SOFCs) were able to conduct O 2À through oxygen vacancies over 500 C, [19] generating electric power to fully use the combustion energy. Y-stabilized ZrO 2 (YSZ) is commonly used as anode material with excellent performance, [20] whereas the Ni-based materials tend to be deposited with carbon. Material degradation at high temperatures is also a serious problem, which can be improved by forming alloys. For instance, a porous NiO─Sm 0.2 Ce 0.8 O 2Àδ anode was prepared by incorporating SnCl 2 through infiltration. [21] After calcination and reduction, Sn─Ni intermetallic compound was formed. Without the Sn incorporation, the cell had a lifetime of less than 10 h in 3% H 2 O─H 2 . By 1 wt% introduction of Sn, the cell lifetime was increased to over 230 h at 0.8 V and 700 C, with a peak power density of 0.28 W cm À2 . The SOFC operation in dry CH 4 can result in coking on the catalyst, and thus mixing CH 4 with O 2 and H 2 O is critical to prevent the coking issue. Meanwhile, carbon coking can be consumed to refresh the surface of the catalysts. Controlling the supply mode of CH 4 to balance fuel consumption, power can be continuously generated and fuels can be fully utilized. For instance, a cycle with seven steps was applied, namely, anode reduction, Ar purging, supplement of CH 4 , operation in CH 4 , Ar purging, consuming coking, and anode rereduction (Figure 4a). [22] The cell output energy was 394 mW cm À2 in the stage operating in CH 4 , with a high current density of 625 mA cm À2 . The step of consuming coking was a key stage, in which the O 2À flux oxidized carbon as well as the catalysts. After %30 min, the potential reduced to 0 with a constant current density of 41.7 mA cm À2 . The flux of protective gas had a mild influence on the time duration, but the increase in current density dramatically shortened the span. Compared with only 18 min of stable operation with continuous fuel supply, the anode with coking consumption showed much better redox stability and maintained its original activity after six cycles in 90 min.
Metallic state species, such as Ni and Pt, are sensitive to poisoning and coking. Loading Ni on oxides that prefer adsorbing O 2 can alleviate coking. For example, Moon and coworkers reported that NiÀgadolinium-doped ceria presented a stability over 500 h, [23] with 1.35 and 0.74 W cm À2 of energy density at 650 and 550 C, respectively. In addition, enclosing metal catalysts into inorganic shells can prevent coking outside the shell and maintain mass transport as the shell is gas permeable. . a) A successive seven-step process of an SOFC operated on dry methane and deposited carbon at 800 C. Reproduced with permission. [22] Copyright 2016, American Chemical Society. b) The schematic of electrochemical CO 2 /CH 4 reforming process in SOEC to produce syngas. CO 2 electrolysis in the cathode and CH 4 electrochemical oxidation in the anode. Reproduced with permission. [30] Copyright 2018, American Association for the Advancement of Science. c) The product analysis of the electrochemical oxidation of CH 4 with different anodes (1-SFMO, 2-0.025Fe─SFMO, 3-0.050Fe─SFMO, 4-0.075Fe─SFMO, 5-0.100Fe─SFMO). Reproduced under the terms of the Creative Commons CC-BY license. [31] Copyright 2019, The Authors. Published by Springer Nature. d) Catalyst lifetime test at 1000 C, 1 bar. The selectivities of different products were displayed over the reaction time. Reproduced with permission. [32] Copyright 2020, Wiley VCH. For example, Jung and coworkers coated Pt nanoparticles with a 1.5 nm-thick Al 2 O 3 shell, and the electrode activity was raised by 300 times. [24] The use of perovskite-type oxides is an alternative means against coking. Perovskite structures can tolerate large cation distortion and are rich in oxygen vacancies. Yang and coworkers reported that Sr 2 FeNb 0.2 Mo 0.8 O 6Àδ double perovskite exhibited coking resistance and a rigid structure under reducing conditions. [25] The electrocatalytic performance was not compromised at 800 C, achieving a power density of 380 mW cm À2 . Incorporating CH 4 -reforming catalysts can help to alleviate the coking problem. Perovskite La 0.7 Sr 0.3 Fe 0.8 Ni 0.2 O 3Àδ (LSFN) was reported to be sprayed onto 8 mol% NiO─YSZ anode as the catalyst and in situ reduced into Fe 0.64 Ni 0.36 , SrLaFeO 4 and La 2 O 3 with uniformly dispersed FeNi alloys. [26] The catalyst first transformed CH 4 into syngas, which was further oxidized on YSZ. This type of structure of SOFC was able to operate in two types of fuels, that is, 97% CH 4 À3% H 2 O and 30% CH 4 À70% air. At 800 C, the maximum power output was 0.343 W cm À2 , and the stability was obtained over 100 h with the incorporation of catalysts in both types of fuels. Without the LSFN catalysts, the activity decay was observed in 12 h using CH 4 Àair or in 3 h using CH 4 ─H 2 O. By rational design and screening, Liu and coworkers reported Ce 0.90 Ni 0.05 Ru 0.05 O 2 as a catalyst to reform CH 4 to H 2 and CO, showing 99% H 2 selectivity and 97% CO selectivity. [27] The anode was designed as a layer-by-layer structure. A BaZr 0.1 Ce 0.7 Y 0.1 Yb 0.1 O 3Àδ catalyst was prepared and mixed with coking-tolerating material Ce 0.8 Sm 0.2 O 2 , and on the top of it was the reforming catalyst. At a relatively low temperature of 500 C, a power density of 0.37 W cm À2 was generated using 3.5% H 2 O─CH 4 fuel.
In addition to the full oxidation into CO 2 , CH 4 is also a critical C 1 resource. Thus, great efforts have been made toward the CH 4 partial (or selective) oxidation to value-added products. Industrial hydrogen is mainly obtained from methane steam reforming. For example, Ni─CeO 2 /γ-Al 2 O 3 ─MgO was reported to lower the operating temperature, and 96.4% CH 4 conversion and 75.3% H 2 selectivity were achieved at 600 C and 4.5 A. [28] As CO 2 is another important C 1 resource, the solid oxide electrolyzer cells (SOECs) provide feedstocks for Fischer-Tropsch's synthesis by converting both CH 4 and CO 2 into syngas, thus eliminating these two kinds of greenhouse gases simultaneously. For example, Sr 2 Fe 1.5 Mo 0.5 O 6Àδ ─Sm 0.2 Ce 0.8 O 1.9 was reported as both the anode and cathode, assembling a symmetrical cell. [29] At 850 C and 0.3 V, the current density reached 242 mA cm À2 . Xie and coworkers chose redox-stable perovskite-type La 0.75 Sr 0.25 Cr 0.5 Mn 0.5 O 3Àδ in a symmetrical cell (Figure 4b), and CuNi was doped into the perovskite lattice and reduced to metal nanoparticles. [30] H 2 O in the stream dissociated at the metal-oxide interfaces, and *OH reacted with carbon to form *COH that subsequently decomposed to *CO and *H. The *OH group also helped to remove carbon.
However, the conversion of CH 4 to hydrocarbon in SOECs has rarely been reported, due to its low selectivity. The catalysts need to efficiently break C─H bond(s) and favor the C─C coupling, as well as inhibit the CO 2 formation. A Fe-doped Sr 2 Fe 1.5 Mo 0.5 O 6Àδ anode was reported to convert methane to 11.5% C 2 H 4 and 5.2% C 2 H 6 under nonoxidative conditions. [31] The conversion ratio of CH 4 was 40.5% and the combined C 2 H 4 þ C 2 H 6 selectivity was 82.2% (Figure 4c). The CH 4 conversion rate and C 2 H 4 concentrations were dependent on the applied potentials, suggesting that C 2 H 4 further transformed from C 2 H 6 . Tang and coworkers achieved C 6 ─C 8 through CH 4 reforming on Fe─Al/SiO 2 . [32] Incorporating that material in SOEC fostered equilibrium to CH 4 conversion, as the byproduct H 2 was exhausted. At 1050 o C, the CH 4 conversion rate reached 15.62%, and the selectivities were 40.08%, 27.37%, and 10.72% for C 2 , benzene-toluene-xylene, and naphthalene, respectively (Figure 4d).

Photoelectrochemical Methane Conversion
Photocatalysts can absorb ultraviolet-visible (UV-vis) light to produce electrons and holes. Meanwhile, the bandgaps need to match the redox potentials of CH 4 oxidation. [33] The challenge of photoconversion of CH 4 is the kinetic sluggish of breaking C─H bonds. When both thermo-and photomethods are combined, the inert C─H activation step can be managed to operate at enhanced rates. Steam reforming of methane into H 2 and CO 2 is thermodynamically preferred at lower temperatures, whereas at higher temperatures, only CO/H 2 mixture can be obtained. For example, under air mass (AM) 1.5 G illumination, a Pt/blackTiO 2 catalyst on a light-diffuse reflection surface achieved a H 2 yield of 185 mmol·h À1 g À1 and 60% quantum efficiency at 500 C, which was 1,000 times larger than state-of-art performance at room temperature. [34] Efforts have also been made to reduce the catalyst deactivation and high energy consumption at high temperatures. Using a Ag-decorated ZnO nanostructure, a quantum yield of 8% (<400 nm) and over 0.1% (%470 nm) were achieved to quantitatively convert CH 4 to CO 2 at room temperature. [35] Due to its wide bandgap, pure ZnO only showed UV absorption. With 0.1 wt% Ag decoration, the absorption was extended to visible light region, attributed to the surface plasmon resonance of Ag nanoparticles (Figure 5a). In situ electron paramagnetic resonance (EPR) indicated the existence of electron trapping sites (Zn þ ) and surface defects (O À ) under illumination. CH 4 was converted through a two-step mechanism. Initially, surface O À radical ions attracted H from CH 4 , generating •CH 3 . Meanwhile, •CH 3 was further oxidized into CO 2 by radicals, such as •OH (O À reacting with OH À ) and O 2 À (Zn þ reacting with O 2 ). Instead of converting CH 4 to CO 2 , Zn species could also oxidize CH 4 to CO through methyl carbonate intermediates. Khodakov and coworkers reported that a ternary composite of Zn, H 3 PW 12 O 40 , and TiO 2 showed a quantum efficiency of 7.1% at 362 nm with 84% CO selectivity. [36] The CO formation rate was 0.02 mmol·g À1 ·h À1 without Zn and increased by 20 times to 0.429 mmol·g À1 ·h À1 with Zn incorporation. After exposure to pure methane, Zn 2þ changed to Zn 0 , which could be reoxidized to Zn 2þ by air. The in situ Fourier-transform infrared (FTIR) spectroscopy study presented signals of bidentate carbonate (CO 3 ), H 2 O (O─H), and carbonate ester (C─O), suggesting a mechanism of carbonate and its ester formation (Figure 5b). Applying 12 CH 4 , in O 2 and 13 CO 2 atmosphere, both 13 CO and 12 CO were detected, implying that a part of CO 2 was converted to CO. The formation of CO was ascribed to the decomposition of surface methyl carbonates under illumination. In the presence of water, ZnO exhibited selectivity toward methanol and formaldehyde under mild light irradiation at room temperature. Ye and coworkers reported that ZnO loaded with 0.1 wt% of Au catalyst presented 125 μmol oxygenates ·h À1 rate with %95% selectivity. [37] www.advancedsciencenews.com www.small-structures.com Small Struct. 2021, 2, 2100037 Surface plasmon resonance had little contribution to photocatalytic performance, because no products were detected under visible light. Without adding water, the overoxidized product (i.e., CO 2 ) was detected. By increasing water amount, both the yield and selectivity of oxygenates were promoted. Loading Au onto ZnO boosted •CH 3 and •OH formation, indicating that Au functioned as cocatalyst. As described earlier, •CH 3 was produced by ZnO and subsequently combined with •OOH to form CH 3 OOH in the pathway to oxygenates (Figure 5c). The existence of water helped to protonate O 2 to •OOH on the Au cocatalyst and facilitated single-electron-coupled proton reduction of CH 3 OO•. CH 3 OOH was the precursor of CH 3 OH through the photoreduction procedure and might directly decompose into HCHO (Figure 5d). Similarly, Yi et al. investigated a >CuO/ ZnO nanocomposite. In spite of a smaller bandgap of CuO than ZnO, the photocurrent response of CuO was negligible compared with ZnO, suggesting the fast recombination of electrons and holes in CuO. [38] The electrons enriched in the conduction band of CuO were further transferred to the conduction band of ZnO, thus fostering oxygen activation and leading to an enhanced activity. Another potential photocatalyst is BiVO 4 with a bandgap 2.4 eV, which can typically convert CH 4 directly to CO 2 . With a low concentration (0.5-5 mM) of nitrite ions, the undesired CO 2 formation was inhibited and the production of CH 3 OH increased, with a high selectivity of CH 3 OH over 90%. [39] In the UV range, NO 2 À had three separated optical absorption bands, which worked as UV filters slightly preventing water photolysis. More importantly, NO 2 À acted as an •OH scavenger balancing •OH concentration, thus cutting off the pathway of CH 3 OH over-oxidation toward CO 2 . Meanwhile, •CH 3 was excited from CH 4 and •OH, and the reduction of •OH decreased •CH 3 , blocking the generation of ethane from two •CH 3 radicals coupling. Andreu and coworkers reported a BiVO 4 /V 2 O 5 codispersed beta zeolite as a visible light photocatalyst for transforming CH 4 into CH 3 OH. [40] The introduction of Bi and V decreased the acidity of OH groups controlling the surface oxidation and extended the absorption spectrum to longer wavelengths. Compared with pure beta zeolites that produced CO 2 with a yield of 59.8 μmol h À1 , the incorporation of both Bi and V allowed to reduce the CO 2 yield to 23.0 μmol h À1 . The yield of CH 3 OH remained unchanged at around 10 μmol·h À1 g À1 . Because of the blocking of OH group, the amount of •CH 3 reduced, leading to the C 2 H 6 /CH 3 OH ratio from 1.43 to 0.23.
A photoelectrochemical strategy has also emerged for CH 4 transformation at room temperatures. The energy input for CH 4 activation is supplied by light, and the intrinsic activity can be regulated by applied potentials, resulting in various partial oxidized products. For example, a TiO 2 photoelectrode was reported as a photoelectrochemical CH 4 catalyst in 1.0 M Figure 5. a) Metallic nanostructures functioning as both a cocatalyst and a light-harvesting medium. Reproduced under the terms of the Creative Commons CC-BY license. [35] Copyright 2016, The Authors. Published by Springer Nature. b) Reaction steps in methane photo-oxidation over Zn, H 3 PW 12 O 40 , and TiO 2 ternary composites. Reproduced under the terms of the Creative Commons CC-BY license. [36] Copyright 2019, The Authors. Published by Springer Nature. c) Yields and selectivities of liquid products under simulated sunlight irradiation. d) The proposed reaction mechanism for photocatalytic CH 4 oxidation to CH 3 OOH, CH 3 OH, and HCHO. Reproduced with permission. [37] Copyright 2019, American Chemical Society.
www.advancedsciencenews.com www.small-structures.com NaOH under UV light with CH 4 bubbling. [41] The applied potential bias was in the range of 0.4-1.2 V versus RHE. The combined CO and O 2 production added up to 80-90% of Faradaic efficiency (FE), and the CO selectivity varied at different potentials (Figure 6a). At an applied potential of 0.4 V versus RHE, the highest FE for producing CO was reported as 81.9%, which dramatically decreased to 24.7% at 1.2 V versus RHE. The mechanism of TiO 2 activating CH 4 was proposed as the charge separation forming Ti 3þ ─O• À and the subsequent attacking CH 4 by O• À . After yielding Ti─O─CH 3 and subsequent Ti─O═CH 2 , the CO pathway diverged from carbonate. The Ti 3þ site was crucial in the CO production pathway, in which enriched Ti 3þ sites tended to form Ti 3þ ─C bond to promote CO formation. Otherwise, the Ti─O─C structure generated carbonate.
To overcome the low solubility of CH 4 in water, triple-phase boundary has also been adopted in photoelectrochemical process (Figure 6b). For example, Takenaka and coworkers reported using WO 3 , with a bandgap of 2.7 eV and absorbing blue light, to covert CH 4 to C 2 H 6 . [42] With an applied voltage, the electron-hole recombination was reduced. The porous n-type semiconductor WO 3 was dispersed on Ti microfiber to conduct excited electrons, Figure 6. a) Schematic illustration of selective CH 4 oxidation to CO on a TiO 2 photoelectrode (left panel) and dependence of the CO efficiency and selectivity on the applied potentials (right panel). Left-axis: efficiency; right-axis: selectivity of CO over all carbonaceous products. Reproduced with permission. [41] Copyright 2018, American Chemical Society. b) Photoelectrochemical system for gas-phase CH 4 activation. c) Time course of product formation in the photoanode compartment. Reproduced with permission. [42] Copyright 2019, American Chemical Society. and the photogenerated positive holes on WO 3 oxidized CH 4 to •CH 3 , accelerating coupling to C 2 H 6 . At 1.2 V applied voltage, a photon-to-current conversion efficiency of 11% and a C 2 H 6 selectivity of 54% (Figure 6c) were recorded. The ratio of CH 4 /H 2 O feeding stream needed to be strictly controlled; otherwise, oxygen evolution reaction could dominate.

Outlook and Perspectives
In this Review, we summarized the development of electrochemical CH 4 conversion in both mild conditions and SOFC/SOEC systems. The combination of thermo-, electro-, and photo-strategies has provided new insights in the activation of C─H bonds as well as the stability control of intermediates. Despite great efforts made to pursue high CH 4 conversion rates and selectivities, several substantial challenges still exist in these fields toward practical CH 4 utilizations. Low-temperature CH 4 electroconversion. To avoid the competing reactions of oxygen evolution, the electrolyte requires a larger stability window. Strategies applying high-valent metal catalysts (Pt, Pd, and V complex) focus on the use of concentrated sulfuric acid and SO 3 /H 2 SO 4 . Although high TOFs, yields, and stabilities have been reported, substantial challenges existed in the separation of products from electrolytes. Meanwhile, the usage of concentrated sulfuric acid is also capital costing and adds strict requirement on the equipment. Electrolysis in the less-corrosive electrolyte is more attractive, such as using CO 3 • À /CO 3 2À as the redox pair to break C─H bonds. Non-noble metal (Co, Ni, V) oxides and RuO 2 have emerged as electrocatalysts in water-based electrolyte. It needs further exploration of more suitable redox pairs that can break C─H bonds and are compatible with water. In addition, although alcohols and acids with multiple carbon atoms can be generated, the yields have generally remained low for practical applications, and the over-oxidation of products is still severe.
High-temperature CH 4 electroconversion. As the cleavage of C─H bond is easier at higher temperatures, the current density can reach 200-400 mA·cm À2 with reasonably good CH 4 conversion rates. Simultaneous utilization of both anode and cathode reactions to convert CH 4 and CO 2 to syngas and hydrocarbon is an attractive strategy, applying Ni, Pt, Cu, Fe, and their alloys as catalysts. Meanwhile, the coking and deactivation of catalysts still remain major problems impacting the cell's lifetime. High-performance coking-resistant catalysts that are metal free or with metal/oxide interfaces are urgently needed. SrFeMoO xand SmCeO x -based coking-resistant materials are under development. The operating time region of cells can be around hundreds of hours, because coking and deactivation are prominent at high temperatures (800 C). The operating temperature is needed to be reduced but not compromise electroconductivity. Solid electrolytes with rich O vacancies can conduct O 2À at lower temperature (%300 C) to reach sufficient ionic conductivity, which can be further explored.
Photoelectrochemical methane conversion. The photogenerated radicals and holes can help to break C─H bonds at ambient conditions. The major products generally reported are CO and CO 2 , suggesting that the selective oxidation of CH 4 into alcohols and hydrocarbons remains a big challenge. In addition, UV light is often needed to initiate the methane activation for photocatalysts (e.g., TiO 2 , ZnO, and BiVO 4 ), also limiting the applications. The potential of using longer wavelengths and lower photoenergies can help to prevent product over-oxidation, whereas the yields and quantum efficiencies still remain at a low level. Thus, substantial efforts in developing and adopting smaller-bandgap materials and hierarchical structures to improve light absorption, as well as incorporation of cocatalysts (such as Au, Ag, Pt, and Pd) to reduce energy barriers, are needed. Furthermore, the rational design and regulating the electrolyte compositions to increase selectivity are also worth exploring.
Qihao Wang obtained his Bachelor of Science in Chemistry at Fudan University in 2019, and is currently conducting graduate research with Prof. Gengfeng Zheng at Fudan University. His research focuses on the fabrication of nanomaterials and their applications in water splitting, CH 4 electrocatalytic oxidation, and CO 2 electrocatalytic reduction. www.advancedsciencenews.com www.small-structures.com