Understanding (photo)electrocatalysis for the conversion of methane to valuable chemicals through partial oxidation processes

Abstract

Methane (CH₄) with sp³ hybridization is known as a main constituent of natural gas, and has been utilized as a fuel and a hydrogen reservoir for power plant and transportation applications. Along with the complete oxidation of CH₄ to CO₂ to retard global warming, the partial oxidation toward the generation of liquid fuels is a big challenge because oxygenated products are more prone to oxidation than CH₄. Therefore, the control of the charge carriers and radical species is critical to prohibit the unwanted reaction, in other words, to maximize the formation of desired products. Photo and electrocatalysis are clean and sustainable techniques for CH₄ conversion, which do not require extreme experimental conditions (i.e., high temperature and pressure), and indeed, the reaction can be driven under sunlight (i.e., charge carriers generated by light absorption or photovoltaic-assisted electricity). In this review, we discuss how to design catalysts to overcome the bottlenecks in the photo and electrocatalytic partial oxidation of CH₄ with focus on (i) an interfacial charge transfer process, (ii) strategies for surface engineering of catalysts, (iii) key factors affecting the reaction mechanism and rate-determining step, and (iv) a systematic investigation of the reaction conditions to lift up the catalytic performance. Finally, we propose the milestone to guide not only the preparation of catalysts that can effectively generate charge carriers and intermediates with a mild oxidation behavior but also the design of reactors for scale-up to prevent undesired reaction pathways.

---

Love Kumar Dhandole

Love Kumar Dhandole received a BE in Electronics & Communication Engineering from RGPV University (India) in 2008, an MTech degree in Green Energy Technology from Pondicherry University (India) in 2014, while working as a Master project student in International Advanced Research Centre for Powder Metallurgy and New Materials R&D (ARCI, Hyderabad, India), and a PhD degree from the Dept of BioTechnology, Chonbuk National University (Iksan campus, Korea) in 2018 under the guidance of Prof. Jum Suk Jang. After brief postdoctoral stints in Chonbuk National University, he joined his second postdoctoral position at the EcoFriendly Catalysis & Energy Lab., under the supervision of Prof. Jae Sung Lee in Dept of Energy Engineering, School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, in 2019, where he worked on visible light photoanode materials for water oxidation and hydrogen evolution applications. After his second postdoctoral, he moved to the Korea Institute of Science and Technology (KIST), Seoul, where he is currently a postdoctoral candidate working on the design of photocatalysts, photoelectrochemical cells, and carbon materials for energy conversion and environmental applications.

Sang Hoon Kim

Dr Sang Hoon Kim received his doctoral degree from Institute of Physics, Humboldt University of Berlin, Germany in 2003 while working as a research associate and doctoral student in Physical Chemistry Department at the Fritz Haber Institute of the Max Planck Society, Berlin, Germany. After brief postdoctoral stints in Materials Sciences Division at Lawrence Berkeley National Laboratory and in Department of Materials Science and Engineering at University of Pennsylvania, he joined the R&D division of Hitachi Global Storage Technologies in San Jose, CA, USA in 2006, working for the development of next-generation hard disk drives. In 2011, he joined Korea Institute of Science and Technology as Senior Research Scientist and in 2016, he was promoted to Principal Research Scientist.

Gun-hee Moon

Gun-hee Moon received a BS degree in Chemical Engineering from Inha University (Incheon, Korea) in 2008, an MS degree in Environmental Engineering from POSTECH in 2011, and a PhD degree in Chemical Engineering from POSTECH (Pohang, Korea) in 2015. During the PhD course, he joined as a visiting student at the Pacific Northwest National Laboratory (PNNL, 2011.06–2012.05). He started his first postdoctoral career at the Institute of Environmental and Energy Technology in POSTECH (2015.03–2016.08) and accumulated research experience at The Institute of Science and Industrial Research (SANKEN), Osaka University (2016.06–2016.07). After his second postdoctoral stay at the Max-Planck-Institut für Kohlenforschung (Germany, 2016.09–2020.10), he moved to the Korea Institute of Science and Technology (KIST), where he is currently employed as a senior scientist working on the design of photocatalysts, electrocatalysts, photoelectrochemical cells, and carbon materials for energy conversion and environmental applications.

---

1. Introduction

Depending on the aspect of climate change and energy crisis, the conversion of methane (CH₄), the largest component of natural gas (70–90%), should lead toward the complete oxidation to carbon dioxide (CO₂) and partial oxidation to valuable oxygenates and (halo)hydrocarbons, respectively. With an environmental perspective, CH₄ is considered as one of the greenhouse gases, whose global warming potential (GWP, a measure of how much heat is trapped by each greenhouse gases in the atmosphere relative to CO₂) is 25 times higher than that of CO₂ over a 100 year period.¹ Indeed, CH₄ is a major source for rising tropospheric ozone level, damaging human health and plant growth.^2,3 Therefore, many countries pledged to cut CH₄ emission by 30% by 2030 under the Paris agreement. CH₄ emission occurs from both nature and human activities, and the main sources in the latter are natural gas harvesting, petroleum industry, enteric fermentation, landfills, manure management, and coal mining (Fig. 1a).⁴ Due to the inevitable emission of CH₄, concerted efforts to cut it down in agriculture and industry has been made for achieving the goals of the climate agreement. From an energy security standpoint, the discovery of a vast storehouse of methane hydrate as well as the advance of gas-well drilling and hydraulic fracturing technologies make CH₄ more readily available in the exothermic combustion of vehicular fuels and thermal power generation.⁵ The production of liquid hydrocarbons and oxygenates through the conversion of CH₄ is of major interest due to it being more useable as a basic building block for manufacturing diverse products, a high volumetric energy density (e.g., 37.8 J L⁻¹ for CH₄versus 17.8 MJ L⁻¹ for methanol at 15 °C under 1 bar), and relatively low transportation/storage cost as well.⁶ Accordingly, it has received attention to develop low-cost, ecofriendly, and highly-efficient methods for the direct conversion of CH₄ to desired compounds.

Fig. 1 (a) Source of methane emission in U.S. in 2019. (b) Enthalpy change of the sequential oxidation of CH₄ at 298 K. Redrawn with permission.⁹ Copyright 1991, Elsevier.

CH₄ with sp³ hybridization is known as a stable molecule in terms of the intrinsic inertness for C–H activation; thus, an energy intensive step requiring a high temperature and pressure is required to convert it to the desired products. In the industry, the steam methane reforming process (CH₄(g) + 0.5O₂(g) → CO(g) + 2H₂(g); CH₄(g) + H₂O(v) → CO(g) + 3H₂(g)) operating at 700–1000 °C and 3–25 bar generates hydrogen (H₂) and carbon monoxide (CO), which are utilized as not only fuel in fuel cell vehicle but also precursors to synthesize a wide range of chemicals and polymers.^7,8 In order to produce value-added chemicals and fuels via the direct conversion of CH₄, reaction with a limited amount of oxygen (i.e., partial oxidation with less than the stoichiometric oxygen content) should be employed for prohibiting the oxidation of CH₄ into CO₂ and H₂O (CH₄(g) + 2O₂(g) → CO₂(g) + 2H₂O(v), ). However, it is unavoidable that a huge amount of CO₂ is released as a byproduct, and the process is not energetically favorable relative to complete oxidation (e.g., CH₄(g) + 0.5O₂(g) → CH₃OH(g), CH₄(g) + O₂(g) → HCHO(g) + H₂O(v), ) (Fig. 1b).⁹

Photo and electricity-assisted conversion of CH₄ to valuable chemicals and liquid fuels is a future-oriented technology since the reaction proceeds under mild conditions (i.e., room temperature and atmospheric pressure) and renewable energy stemming from sunlight, wind, biomass, etc., is likely to replace fossil fuels. Moreover, the system does not require a complex and centralized infrastructure, which makes it suitable for remote location applications such as offshore drilling rigs, shale gas extraction facilities, and landfills. On the other hand, the bottleneck in the partial oxidation process of CH₄ mainly comes from a poor selectivity toward the target products as well as low reactivity owing to the activation energy barrier. The physicochemical property of CH₄, including a high homolytic bond dissociation energy for H-abstraction (440 kJ mol⁻¹), weakly-polarized symmetric sp³ geometry, weak acidity, and poor solubility should be considered to develop active catalysts.^10–12 Given the relatively effective dissociation behavior of CH₄ in thermocatalysis,^13,14 a wide range of transition metal-based catalysts such as zerovalent metals, metal oxides, metal complexes, single metal atoms, and metal-exchanged zeolites have been revisited to be utilized as photo, electro, or cocatalysts. Furthermore, as part of methodologies to overcome such a high energy barrier, the addition of hydrogen peroxide (H₂O₂) enables the formation of hydroxyl radicals (˙OH) with a strong oxidation power (E°(OH/OH⁻) = 1.90–2.70 V_NHE).¹⁵

In photocatalysis, electrons and holes generated under the illumination of UV or visible light can initiate the oxidation of CH₄, which can be controlled by the electronic structure of semiconducting materials and the operation parameters (e.g., light intensity and wavelength, pH, and dissolved oxygen concentration) (Fig. 2a). Without the intervention of photocatalysts, the direct irradiation with wavelength below 140 and 98.3 nm initiates the excitation of CH₄ by the electronic transition (CH₄(g) + hv → CH₄*(g)) and the ionization of CH₄ (CH₄(g) + hv → CH₄⁺(g) + e⁻), respectively.¹⁶ When the valence band potential of photocatalysts is deep enough to oxidize CH₄, the hole-mediated reaction occurs. Gauging the radical-involved process in aqueous media, two main paths produce ˙OH: (i) a reductive pathway mediated by electron transfer, 1st: O_2,dissolved + H⁺ + e⁻ → HO₂(aq), 2nd: HO₂(aq) + H⁺ + e⁻ → H₂O₂(aq), 3rd: H₂O₂(aq) + e⁻ → OH⁻ + ˙OH, and (ii) an oxidative pathway mediated by hole transfer: OH⁻ + h⁺ → ˙OH.^17–19 The non-selective ˙OH directly attacks CH₄ and consequently facilitates its partial and complete oxidation depending on the experimental conditions. In photocatalysis, photogenerated electron–hole pairs are manipulated within one particle, whose conduction/valence band position thermodynamically determines whether an interfacial charge transfer occurs or not. However, in electrocatalysis, the driving force can be controlled by applying the potential, where the reduction and oxidation reactions are separated by a membrane (Fig. 2b). The problems associated with poor mass transfer and low solubility of CH₄ should be avoided to optimize the cell performance. In this regard, a membrane electrode assembly (MEA) with outperforming typical H-type cells is suggested, where each cathode and anode catalyst layer is coated on the front and back side of a separator or membrane, respectively.²⁰ The gas diffusion layer (GDL) coated on the outside of the catalyst layers dramatically improves the mass transport limitation of CH₄ and gas phase products.²¹ Furthermore, the overoxidation can be avoid by the set-up of the flow cell that is able to collect liquid oxygenates from the electrolyte. When designing anode catalysts, it is critical to selectively oxidize CH₄ instead of oxygenate products. Indeed, the competitive oxygen evolution reaction and the performance deterioration caused by ohmic, mass transport, and kinetic losses should be under control, especially when the cell voltage or the current density is high.²² For example, to meet the economic viability for the conversion of CH₄ to methanol, the selectivity should be at least 70% under the conditions (the overpotential <1 V and the current density >100 mA cm⁻² at 100–250 °C).²³ A photoelectrochemical (PEC) cell is a combined system of photocatalysis and electrocatalysis; the operation is possible under bias-free conditions (i.e., without electricity) and the oxidation/reduction reaction can be separated via the proper configuration of cells (Fig. 2c). Because of the deep valence band position of O_2p and N_2p in metal oxides and nitrides, respectively, photogenerated holes with strong oxidation power are feasible to stimulate the oxidation of CH₄. Under an applied external bias or using a tandem PEC cell, a prolonged lifetime of charge carriers is achieved as a result of the band bending of the depletion layer, hence increasing the driving force for interfacial hole transfer to CH₄.

Fig. 2 Schematic diagram of (a) photocatalysis, (b) electrocatalysis, and (c) photoelectrochemical cell for the oxidation of CH₄. Reproduced with permission.^77,114,115 Copyright 2019 and 2021, Elsevier. Copyright 2017, Wiley.

In this review, we overview the state-of-the-art strategies that have been well established for the partial oxidation of CH₄ by means of photocatalysis, electrocatalysis, PEC cells, and advanced oxidation process (AOP)-related radical reaction, and discuss the geometric and electronic parameters of catalysts, reactor engineering, and the effect of the surroundings contributing to the activity and selectivity. Understanding the mechanistic pathway of electron- and hole-mediated reactions correlated with the formation of intermediates and products will offer a rational solution to overcome the bottleneck and optimize the catalytic performance as well. Rather than describing the synthesis of new materials and their characterizations, we focus on the surface engineering of catalysts affecting the interfacial charge transfer processes in the microenvironment and key factors changing the reaction mechanism and rate-determining step. At the end, we outline the milestones to lead researchers into the right path for not only the design of promising catalysts but also the engineering of reactors for scale-up.

2. Mechanistic view for methane conversion to desired products

As CH₄ is pyrolyzed above 700 °C, hydrocarbons can be produced by autocatalysis following the sequential processes (CH₄ → C₂H₆ → C₂H₄ → C₂H₂ → aromatic hydrocarbons and coke), which is initiated by the radical species (CH₄ → CH₃˙ + H˙).²⁴ Likewise, C₂ and C₃ products are obtained by the chain propagation (e.g., 2CH₃˙ → C₂H₆; CH₃˙ + C₂H₆ → C₂H₅˙ + CH₄; C₂H₅˙ → C₂H₄ + H˙; CH₃˙ + C₂H₄ → C₂H₃˙ + CH₄; C₂H₃˙ → C₂H₂ + H˙ for C₂ hydrocarbons and CH₃˙ + C₂H₄ → n-C₃H₇˙; n-C₃H₇˙ → C₃H₆ + H˙; CH₃˙ + C₃H₆ → C₃H₅˙ + CH₄; C₃H₅˙ → C₃H₄ + H˙ for C₃ hydrocarbons). In the presence of O₂, diverse oxygenates can also be generated by both CH₃˙ and OH₂˙ radicals (CH₄ + O₂ → CH₃˙ + OH₂˙), which are involved in the formation of acetic acid, methanol, formaldehyde, etc. (e.g., CH₃˙ + O₂ ↔ CH₃O₂˙; CH₃O₂˙ + CH₄ → CH₃COOH + CH₃˙; CH₃O₂˙ + CH₄ → CH₃OH + CH₃O; CH₃O₂˙ → HCHO + OH˙).²⁵ However, the problem is that the yield is extremely low.

The introduction of oxidants (e.g., nitrogen oxide, H₂O₂, O₂, and water) or/and heterogeneous catalysts sheds light on the solution to enhance the yield and selectivity since the activation energy can be effectively reduced. For example, the reactive radical species (CH₃˙, CH₃O˙, etc.) can appear through the attack of nitrogen dioxide (NO₂) (e.g., CH₄ + NO₂ → CH₃˙ + HNO₂; CH₃˙ + NO₂ → CH₃O˙ + NO). Over the surface of heterogeneous catalysts at low temperature, it is proposed that the C–H bond is cleaved by two mechanistic pathways. One is dehydrogenation involving electrophilic oxygen atoms (M–O sites) as a result of the formation of CH₃˙ by H-abstraction (Fig. 3a). The other is deprotonation stemming from the dissociative adsorption of CH₄ (i.e., CH₄ → CH₃⁻ + H⁺) following an instant coordination onto unsaturated metal centers to form the M–C σ-bond (Fig. 3b).²⁶ The former is a radical-mediated oxidation process with a formal oxidation state (FOS) of C^−III for CH₄ (i.e., 1st: M + 0.5O₂ → M–O for the formation of active site; 2nd: M–O + CH₄ → M–O⋯CH₄ with sp² trigonal geometry → M–OH + CH₃˙ for the activation of CH₄). In the latter, the direct cleavage of CH₄ occurs from the complexation with atomic metals accompanied by an FOS of C^−IV for CH₄ (i.e., M + CH₄ → H⁺ + M–CH₃ with sp³ tetragonal geometry). As for the dehydrogenation process, the formation of the M–O active site and the C–H activation should be kinetically balanced to optimize the performance of the catalysts. Photo and electrocatalysis over transition metal oxides can assist the (re)generation of active sites on the catalyst surface. In regard to the deprotonation process, the presence of proton acceptors on the surface of catalysts facilitates the reaction kinetics, and surface oxidizing M–O species that are able to be formed by photo and electrocatalysis may play an important role as a proton acceptor.

Fig. 3 (a) Partial oxidation of CH₄ into formaldehyde by electrophilic oxygen atoms (M–O sites) over molybdenum oxide. (b) Adsorption and transition states of CH₄ on Pd-doped CeO₂ and electron density map (cyan = Pd; red = O; off-white = Ce; gray = C; white = H). Reproduced with permission.¹¹⁶ Copyright 2018, American Chemical Society.

The study of CH₄ conversion in heterogeneous photocatalysis has mainly focused on gas phase systems, where the overoxidation of CH₄ was caused by the lattice and surface adsorbed oxygens, which consequently led to the formation of CO and CO₂. The cleavage of the C–H bond initiated by the trapped hole (M^z+–O²⁻ + h⁺ → M^z+–O⁻˙ in metal oxides), ˙OH, or other electrophilic oxygen species is crucial for the partial oxidation. In contrast, the formation of reactive oxygen species (ROS) should be minimized in order to increase the selectivity for the non-oxidative coupling of CH₄.²⁶ As such, the manipulation of proper oxidants and radical species determines whether to synthesize oxygenates or hydrocarbons as the final products. In an aqueous system, the activation of CH₄ may be more favorable by mobile ˙OH rather than by the trapped hole (or surface bound ˙OH) in that the solubility of CH₄ is extremely low and the adsorption of CH₄ on the surface of catalysts is difficult under mild conditions (Fig. 4a).^27–31 The opposite outcome was also reported, i.e., mobile ˙OH stimulates the formation of C₂H₆, whereas the reaction of surface-bound ˙OH with CH₃˙ selectively produces methanol (Fig. 4b).³² The true mechanism of photocatalytic CH₄ conversion is still a mystery. Nevertheless, it should be noted that the handling of time-dependent ˙OH concentration is important to improve the catalytic performance. The design of flow-type reactors is one of the strategies for not accumulating the oxidative radical species (i.e., to prevent overoxidation).³³ In addition, the photo-Fenton process is another approach to supply ˙OH in terms of the activation of H₂O₂ (Fe³⁺ + e⁻ → Fe²⁺; Fe²⁺ + H₂O₂ → Fe³⁺ + ˙OH + OH⁻) (Fig. 4c).^34,35 In spite of much efforts, the issue that the relatively high quantum yield and selectivity are obtained when the conversion of CH₄ is extremely low is still unsolved in photocatalysis.

Fig. 4 (a) Mobile ˙OH-mediated CH₄ oxidation pathway over the 0.30% RuO_x/ZnO/CeO₂ composite. Reproduced with permission.³¹ Copyright 2022, American Chemical Society. (b) Change of the reaction pathway for CH₄ oxidation before and after the surface fluorination of WO₃. While the surface bound ˙OH produces methanol, mobile ˙OH prefers to generate ethane. Reproduced with permission.³² Copyright 2015, Elsevier. (c) Partial oxidation of CH₄ by reactive oxygen species including mobile ˙OH formed through the photo-Fenton process. Reproduced with permission.³⁴ Copyright 2020, Royal Society of Chemistry.

The electrocatalytic conversion of CH₄ to value-added chemicals is driven by an anodic pathway, where active sites generated by applying positive bias (M–O²⁻ → M–O + 2e⁻ or MO_x + H₂O → MO_x(˙OH) + H⁺ + e⁻; MO_x(˙OH) → MO_x–O + H⁺ + e⁻) oxidizes CH₄. Along with the applied electrode potential, the local chemical environment influenced by adsorbates, supports, electrolyte constitutes, 3D chemical surroundings, temperature, pressure, etc., would also change the CH₄ oxidation process.³⁶ The oxygen evolution reaction (OER) is a predominant process when a high voltage is applied for transition metal oxide catalysts such as Co₃O₄, NiO, and RuO₂,^37–39 which is known as a competitive reaction for the oxidation of CH₄ in aqueous media. Fig. 5a compares the Gibbs free energy of transition metal oxides for water oxidation (O* + H₂O → *OOH + H⁺ + e⁻) and CH₄ activation (O* + CH₄ → *O–CH₄ → *(HO–CH₃)).²² Through the reaction steps, molecular oxygen (*OOH → * + O₂ + H⁺ + e⁻) and methanol (*(HO–CH₃) → *CH₃OH → * + CH₃OH) are produced as the final product. In both the reactions, the dehydrogenation of *OH (*OH → *O + H⁺ + e⁻) that is formed by single electron transfer process (* + H₂O → *OH + H⁺ + e⁻) creates the adsorbed active oxygen atom (O*) as a transient species. In order to overcome the OER, the design of catalysts with a high Gibbs free energy for water oxidation but with a low one for CH₄ activation is highly required, and V₂O₅, TiO₂, SnO₂, PtO₂, and RhO₂ meet the conditions well. Theoretically, the operation in potentials close to 1.23 V_NHE is amenable to suppress the OER. The emergence of high-valent metal oxidation states and the change in the surface free energy would increase the coverage of active M–O sites and modify the reaction pathway. Regarding TiO₂/RuO₂ composites, the introduction of V₂O₅ switches the reaction pathway as follows.⁴⁰ Two ruthenium redox couples, Ru^4+/6+ and Ru^3/+4+, in RuO₂ participate in the production of oxygenates and affect the selectivity of the final products. For the formation of methanol, Ru⁶⁺ is reduced to Ru⁴⁺ sequentially following (i) CH₄ + RuO₂²⁺ → CH₃–RuO₂⁺ + H⁺ and (ii) CH₃–RuO₂⁺ + H⁺ → CH₃OH + RuO²⁺. On the other hand, formaldehyde is formed by the conversion of ruthenium (Ru⁶⁺ → Ru⁴⁺ → Ru³⁺) following the reactions CH₄ + 2RuO₂²⁺ → RuO⁺–O–CH₂–O–RuO⁺ → HCOOH + 2RuO⁺. However, in the presence of V₂O₅ on RuO₂, the redox couple of vanadium (V^5+/4+) may not donate enough electrons to form a double bond for formaldehyde and formic acid; thus, the selectivity toward methanol increases (CH₄ + V₂O₄²⁺ → CH₃–V₂O₄⁺ + H⁺; CH₃–V₂O₄⁺ + H⁺ → CH₃OH + V₂O₃²⁺). Transition metal ions also can be utilized as a redox couple in a homogenous system, where metal centers of organometallic complexes dissolved in the electrolyte turn into a high-valent state by applying anodic bias. As shown in Fig. 5b, the Pt^II of chloroplatinate accepts CH₄ by electrophilic addition reaction, and deprotonation produces the Pt^II–CH₃ complex, which is further oxidized into the Pt^IV–CH₃ complex by virtue of the oxidation by the electrochemically-formed [Pt^IVCl₆]²⁻ (i.e., the Shilov cycle catalyzed by the redox cycle of [Pt^IVCl₆]²⁻ and [Pt^IICl₆]²⁻).⁴¹ When H₂O or HCl exists, the Pt^IV–CH₃ complex leads to the formation of methanol or methyl chloride, respectively. The mechanistic pathway based on the PEC oxidation of CH₄ is quite analogous to the electrochemical system. The surface-bound ˙OH on the (010) facet of the WO₃ photoanode can oxidize in situ formed methanol, resulting in the selective formation of ethylene glycol (Fig. 5c).⁴² A wide range of strategies to improve the active site and the activation of CH₄ toward the production of desired products will be discussed in the following section.

Fig. 5 (a) Comparison of the Gibbs free energy of transition metal oxide catalysts calculated from water oxidation (O* + H₂O → *OOH + H⁺ + e⁻) and CH₄ activation (O* + CH₄ → *O–CH₄ → *(HO–CH₃)). Reproduced with permission.²⁰ Copyright 2020, American Chemical Society. (b) Catalytic cycle of CH₄ conversion to methanol or methyl chloride mediated by electrochemically regenerated [Pt^IVCl₆]²⁻. Reproduced with permission.⁴¹ Copyright 2020, American Chemical Society. (c) Reaction pathway to form ethylene glycol on the WO₃ photoanode. Reproduced with permission.⁴² Copyright 2021, Wiley.

3. Selective oxidation of methane into oxygenates via light- and electricity-driven processes over heterogeneous catalysts

3.1. Photocatalytic CH₄ conversion

Photocatalysis is a tool to utilize photonic energy as a driving force for photocatalytic (ΔG° < 0: downhill process) and photosynthetic (ΔG° > 0: uphill process) reactions.⁴³ As mentioned before, the oxidation of CH₄ is a downhill process; thus, the Gibbs free energy for partial oxidation to obtain oxygenates is more unfavorable than that for the complete oxidation to CO₂ (e.g., ΔG°(CH₃OH) = −115 kJ mol⁻¹; ΔG°(CH₂OH) = −296 kJ mol⁻¹; ΔG°(CO₂) = −818 kJ mol⁻¹). Although the schematic diagram representing the generation of electron–hole pairs and the interfacial charge transfer is simply depicted (Fig. 6a), understanding the detailed photophysical and photochemical processes is very difficult. Indeed, various factors should be considered to design highly active photocatalysts (Fig. 6b). Takanabe articulates the mechanistic pathway of photogenerated electron–hole pairs with considering time scale in his review article.⁴⁴ The photocatalytic performance is governed by (i) photon absorption, (ii) exciton separation, (iii) carrier diffusion, (iv) carrier transport, (v) catalytic efficiency, and (vi) mass transfer of reactants and products. The excitation of the electron from the valence band to the conduction band (cf., from HOMO to LUMO in homogeneous photocatalysts) and the relaxation of the electron and hole to the band edge level occur in femtoseconds and in the range of femto to picosecond time scale, respectively.⁴⁵ Free charge carriers separated from excitons in nano to microseconds are delivered to the surface, which participate in the redox reaction within milliseconds to seconds. Therefore, the optimization of the electronic structure of the photocatalyst can improve the photon absorption, exciton separation, carrier diffusion, and carrier transport, which is a descriptor of how efficiently generation and transfer of the charge carriers to the surface take places. In particular, the introduction of catalytically active sites on the surface of photocatalysts such as defect sites, functional groups, and cocatalysts would help to facilitate C–H activation in terms of the dehydrogenation and deprotonation of CH₄. The control of the materials' morphologies and the design of reactors can solve the mass transfer problems of CH₄ and gas phase products.

Fig. 6 (a) Simplified photocatalytic process. (b) Diverse factors influencing the photocatalytic performance. IPCE and QE refer to ‘incident-photon-to-electron conversion efficiency’ and ‘quantum efficiency’, respectively. Reproduced with permission.⁴⁴ Copyright 2017, American Chemical Society.

Some factors influencing the redox reaction such as experimental conditions, nature of the oxidants, sacrificial reagents, morphology, and surface properties of the photocatalysts should be considered for the effective conversion of CH₄.^46,47 Photocatalytic partial oxidation of CH₄ can be achieved under relatively low temperature and pressure.^20,48 Contrarty to thermocatalysis, the mild experimental conditions make it the systematic build-up of reactors much more easy. On the other hand, the addition of oxidants including H₂O₂, nitric oxide (NO), and O₂ could accelerate the photocatalytic activation of CH₄ by means of the emergence of highly active radical species such as ˙OH or ˙O₂⁻.^49–51 The introduction of suitable sacrificial reagents can also boost up the reaction kinetics with scavenging either photogenerated electrons or holes. Accordingly, more charge carriers survived against the recombination possibly accelerate the activation of CH₄.⁵² The dehydrogenation of CH₄ can be achieved by photogenerated holes directly, and the addition of electron acceptors such as Cu²⁺, Ag⁺, Fe³⁺, and methyl viologen (MV²⁺) may facilitate the kinetics since they are endowed with prolonged lifetime of holes.^20,22,53 The concept is analogous to a three-electrode system in electrochemical cells. Consistent with many other studies, the morphology and surface functionality of photocatalysts have significant impact on the photocatalytic oxidation of CH₄.^20,46

A systematic investigation on the parameters influencing the partial oxidation of CH₄ to CH₃OH was conducted using bismuth-based photocatalysts (e.g., Bi₂WO₆, Bi₂WO₆/TiO₂, and BiVO₄).²⁹ The experiment conditions were as follows: medium-pressure mercury lamp (450 W), 55 °C, CH₄ : He (20 : 80) mixture with continuous purging (22.4 mL min⁻¹), and 1 g L⁻¹ catalyst. Under catalyst-free conditions, the direct photolysis of water converted CH₄ into CH₃OH, CO₂, and C₂H₆. The selectivity for these products was 22.3, 59.7, and 18.1%, respectively. The CH₄ conversion and CH₃OH production rate was 0.8% and 2.1 μmol h⁻¹, respectively. The higher selectivity to CH₃OH (27.6%) and production rate (4.6 μmol h⁻¹) were achieved by the photocatalytic process with Bi₂WO₄, whereas the lower selectivity toward C₂H₆ (9.0%) was observed. When the temperature increased from 55 to 90 °C, the formation of C₂H₆ was markedly enhanced (almost 3-fold) with a negligible amount of CH₃OH. This could be ascribed by a high concentration of ˙CH₃ in terms of the dehydrogenation of CH₄ by ˙OH in the presence of an excess amount of water vapor. In order to check the effect of dissolved oxygen as an electron acceptor, O₂ was continuously purged. While the production of CO₂ sharply rose (almost 20-fold), the formation of C₂H₆ was almost negligible and the CH₃OH production rate was barely affected. Introducing Fe³⁺ as a more effective electron scavenger greatly accelerated the evolution of CO₂, which was probably due to the enhanced charge separation and the overoxidation of CH₄ by ˙OH (Fig. 7a). Interestingly, the higher selectivity toward CH₃OH (below 10%) was obtained in spite of the sharp increase in the total conversion (∼7%). The blank test with Fe³⁺ solution showed the oxidation of CH₄ to CH₃OH, CO₂, and C₂H₆, and was attributed to hydroxyl radicals produced in the photo-Fenton process under UV light irradiation. The fabrication of Bi₂WO₆/TiO₂ and BiVO₄ was carried out to investigate the effect of the ability of charge separation and the conduction band energy level on CH₄ oxidation, respectively (Fig. 7b). Among the three samples, Bi₂WO₆/TiO₂ showed the highest CO₂ production rate and lowest CH₃OH production rate, which might be due to the production of more ˙OH. Despite such a low CH₄ conversion, the selectivity (i.e., higher CH₃OH, lower CO₂, and C₂H₆ production) makes BiVO₄ a promising material.

Fig. 7 (a) Photocatalytic oxidation of CH₄ using Bi₂WO₆, Bi₂WO₆ with Fe³⁺ (1 mM), and a solution of Fe³⁺. (b) Different coversion rate and selectivity of CH₃OH, CO₂, and C₂H₆ for Bi₂WO₆, Bi₂WO₆/TiO₂, and BiVO₄ photocatalysts. Reproduced with permission.²⁹ Copyright 2014, American Chemical Society. (c) Proposed mechanism mediated MV^2+/˙⁺ for the partial oxidation of CH₄ in La-doped WO₃. (d) Time-dependent profile of products including CH₃OH, CO, H₂, and O₂ under dark and light irradiation for La-doped WO₃. Reproduced with permission.²⁷ Copyright 1997, Elsevier.

Noceti et al. synthesized La- and Pt-doped WO₃ and carried out the photocatalytic activity test in the presence of MV²⁺ as an electron acceptor. This allowed photogenerated holes to oxidize H₂O to ˙OH effectively instead of the recombination with electrons (Fig. 7c).²⁷ ˙OH induced the dehydrogenation of CH₄ to ˙CH₃, and CH₃OH was generated by the reaction between ˙CH₃ and water (or ˙OH). The electron transfer from the conduction band to protons mediated by MV^2+/˙⁺ led to the production of H₂. The La-doped WO₃ showed a relatively good catalytic activity (cf., about 4% of CH₄ conversion) under mild conditions (at 94 °C and under a constant gas flow of CH₄), where by-products including CO, H₂, and O₂ appeared (Fig. 7d). In the presence of H₂O₂ as a source to make ˙OH, the loading of FeOOH on WO₃ significantly improved the photocatalytic conversion of CH₄. However, an excess amount of H₂O₂ significantly reduced the production of the desired products (i.e., a poor selectivity) owing to overoxidation.⁵⁴ The introduction of NO₂⁻ as an electron scavenger in the BiVO₄-dispersed solution lifted up the formation of ˙OH in terms of the concomitant water oxidation by holes. Consequently, the selectivity of CH₄ to CH₃OH reached up to 93.0% but the low conversion problem still remained.⁵⁵

BiVO₄ with unique morphologies was also investigated for CH₄ conversion. Zhu et al. compared the photocatalytic activity of BiVO₄ with different morphologies, and concluded that the bipyramid BiVO₄ with (102) and (012) surface facets was favorable for the production of CH₃OH relative to the platelet-type one with (001) facets.³⁰ The selectivity toward CH₃OH reached up to 85% and the mass activity was recorded as 112–134 μmol h⁻¹ g⁻¹. The high selectivity is probably originated from not only the effective interfacial hole transfer but also the adjusted surface energy for the activation of the C–H bond.

Carbon nitride (C₃N₄), consisting of only carbon and nitrogen, is one of cost-effective visible light-active photocatalysts, and its preparation is also quite easy.^56–58 Li et al. observed that CO and CO₂ were the main products when bare g-C₃N₄ was utilized, indicating that the overoxidation of CH₄ was severe.⁵⁹ However, it could be overcome by the fabrication of mace-like g–C₃N₄–decorated Cs_0.33WO₃ (g-C₃N₄@Cs_0.33WO₃) nanocomposites. As seen in Fig. 8a, the selectivity of CH₄ (1000 ppm) to CH₃OH, CO, and CO₂ was estimated to be 51.59%, 12.08%, and 36.33%, respectively, where the concentration of CH₃OH increased linearly with the irradiation time. Cs_0.33WO₃ was likely to protect the overoxidation of CH₃OH. According to the screening test result (Fig. 8b), it was concluded that both ˙O₂⁻ and e⁻ are a main source to activate CH₄ (i.e., the C–H bond activation via the reductive pathway mediated by ˙O₂⁻) (Fig. 8c), which is in contrast to the result of other studies. The in situ EPR measurement should be helpful to determine the ROS mainly involved as well as to monitor the structure change of the catalysts during the reaction.

Fig. 8 (a) Time-dependent formation of CO₂, CO, and CH₃OH for g-C₃N₄@Cs_0.33WO₃ sample under irradiation. (b) Trapping of the radical species; herein, K₂Cr₂O₇, para-quinone, salicylic acid, and Na₂C₂O₄ were used as sacrificial reagents for electrons, O₂˙⁻, ˙OH, and holes, respectively. (c) Proposed mechanism for the conversion of CH₄. Reproduced with permission.⁵⁹ Copyright 2019, American Chemical Society.

In the presence of H₂O₂, the selective conversion of CH₄ to CH₃OH was accomplished by virtue of the fabrication of highly dispersed iron species on TiO₂ photocatalysts.⁶⁰ In pristine TiO₂, CH₄ conversion and its selectivity toward CH₃OH during 3 h irradiation was 10.9% and 36%, respectively. As seen in Fig. 9a and b, the loading of noble metals on TiO₂ adversely affected the photocatalytic performance because zerovalent metals are known as a catalyst for the oxidation of CH₄ to CO and CO₂. Notably, the loading of 0.33 wt% iron species on TiO₂ enhanced the CH₄ conversion and the selectivity toward CH₃OH. By counting the formation of C₂H₅OH as the by-product, the total conversion to alcohols was estimated to be 97%. The control experiment clearly supports that the conversion of CH₄ mainly proceeded by the photocatalytic activation of H₂O₂ (Fig. 9c). Indeed, the labelling experiment using ¹³CH₄ revealed a new peak at m/z = 33, which indicates the formation of ¹³CH₃OH⁺ from the photocatalytic process (Fig. 9d). Considering the X-ray absorption spectroscopy data, it was confirmed that FeO_x consisted of FeOOH and Fe₂O₃, whose oxidation state was close to +3. Along with the enhanced electron transfer process from the TiO₂ conduction band to FeO_x, the effective activation of H₂O₂ through a concomitant electron transfer improved the photocatalytic performance (Fig. 9e). As a result, it was proposed that the reaction between ˙CH₃ and ˙OH produced CH₃OH; the formation of O₂˙⁻ should be prohibited to avoid the overoxidation of CH₃OH to CO₂ (Fig. 9f).

Fig. 9 (a) Photocatalytic conversion of CH₄ and (b) production of CH₃OH for metals or metal oxides loaded on TiO₂ after 3 h of full arc irradiation. (c) CH₄ conversion and product yield in control experiment. (d) Labelling experiment of liquid products using ¹³CH₄. (e) Comparison of LSV curves for FTO, TiO₂, and FeO_x/TiO₂. (f) Proposed reaction pathway on the production of radical species. Reproduced with permission.⁶⁰ Copyright 2018 Springer Nature.

ZnO is an UV light active photocatalyst, whose bandgap size is almost identical to TiO₂ (about 3.2 eV).⁶¹ Without adding oxidants or cocatalysts, the photocatalytic conversion of CH₄ was almost negligible, indicating that the proper modification of ZnO is highly required.^50,62 The loading of cocatalysts (e.g., Pt, Pd, Au, or Ag) on ZnO showed the good photocatalytic conversion of CH₄ to CH₃OH and CH₃OOH under aerobic conditions.⁶² For instance, 0.1 wt% Au/ZnO resulted in a high photocatalytic activity of 125 μmol h⁻¹ for oxygenates with selectivity over 95%. The isotope experiment exhibited that the oxygen in CH₃OH and CH₃OOH stemmed from dissolved O₂ in water. The result supports the proposed mechanism that ˙CH₃ formed by the dehydrogenation of CH₄ with holes and ˙OOH generated by the oxygen reduction with electrons are combined together and then turned into CH₃OH and CH₃OOH. It should be noted that water oxidation was not considered in the overall reaction. Therefore, the balance between the reductive and oxidative pathway is critical to increase the selectivity. The different mechanism was also proposed in the Au/ZnO photocatalyst.⁵⁰ The dehydrogenation of CH₄ to ˙CH₃ by ˙OH can proceeded through both reductive and oxidative pathways; consequently, the reaction of ˙CH₃ with either ˙OH or O₂˙⁻ produced CH₃OH. In here, both H₂O and O₂ were invloved in the reaction.

3.2. Electrocatalytic CH₄ conversion

The electrocatalytic activation of CH₄ is able to be driven by either active oxygen on the surface electrode directly or free radicals formed at the electrode/electrolyte interface indirectly. In general, the cell performance is governed by the overpotential, surroundings, and the collision frequency, which are closely related to the activation energy, the ease of forming an active surface oxygen, and mass transfer of CH₄ and products, respectively.⁶³ In order to optimize the partial oxidation of CH₄ to the desired products, efforts on catalyst discovery, electrolyte engineering, and reactor design should be made considering the three factors shown in Fig. 10. To date, not only diverse electrocatalysts but also the engineering of electrolytes and reactors have been investigated for the partial oxidation of CH₄. The electrochemical system can be classified as (i) heterogeneous and homogeneous reaction, (ii) high and low temperature process, (iii) cathodic and anodic reaction, (iv) solid oxide and liquid electrolyte, and (v) fuel cell and electrolyzer type cell. In previous review articles, the effects of electrocatalysts, reaction conditions, and cell configurations on the formation of final products have already been well discussed and summarized in tables.^{20,26,63–73} In this review, we mainly focused on the theoretical approach to interpret the CH₄ partial oxidation process on active sites in mild conditions, and overviewed the recent research results.

Fig. 10 Parameters consisting of catalyst discovery, electrolyte engineering, and reactor design interconnected with three factors including energetic barrier, frequency of rare events, and techno-economic analysis to be considered for the development of effective electrocatalytic CH₄ partial oxidation systems. Reproduced with permission.⁶³ Copyright 2019, Elsevier.

Transition metal oxides⁷⁴. The electrocatalytic oxidation of CH₄ on transition metal oxides proceeds through the physical adsorption of CH₄, followed by the activation of the C–H bond. Fig. 11a shows the products obtained from the oxidation of CH₄ over various transition metal oxide catalysts. CO₂ evolution was only observed both for TiO₂, IrO₂, and PbO₂ in phosphate buffer and for PtO₂ in KCl electrolyte. The inactivity of PtO₂ in phosphate buffer was due to the poisoning of Pt covered by phosphate ions. The physical adsorption of CH₄ transformed the tetrahedral symmetry of CH₄ to the distorted structure (D_2d) with the H–C–H bond angle of about 120°, which scaled with the electrostatic interaction between CH₄ and transition metal oxides. Therefore, the experimentally measured CH₄ binding energy linearly increased with the Madelung potential of the metal in transition metal oxides (Fig. 11b). TiO₂, IrO₂, PbO₂, and PtO₂, classified as active catalysts, have a higher CH₄ binding energy (>0.23 V) and a lower Madelung potential (<−40 V). The inactivity of ZrO₂ and SnO₂, despite satisfying the condition, was ascribed to the poor electrical conductivity and a small number of active sites, respectively. The introduction of Cu₂O₃ on TiO₂ resulted in 6% faradaic efficiency at 0.35 mA cm⁻² current density for the production of CH₃OH, in which *OH was provided from Cu to *CH₃-populated TiO₂. Cu has a lower energy barrier for the reaction of *CH₃ with *OH and is resistant to the overoxidation of CH₃OH (Fig. 11c). The DFT-predicted reaction profile for OER is depicted for the fully reduced state (Fig. 11d) and fully oxidized one (Fig. 11e) according to the reaction pathway (*H₂O → *OH + 0.5H₂; *OH → *O + 0.5H₂; O* + H₂O → * + O₂ + H₂). OER was favorable for the undercoordinated metal sites of SnO₂, TiO₂, and IrO₂ but on the bridging O site of PbO₂. Fig. 11f shows the reaction profile for CH₄ oxidation following five steps (1st: * + CH₄ → *CH₃ + 0.5H₂; 2nd: *CH₃ → *CH₂ + 0.5H₂; 3rd: CH₂* → *CH + 0.5H₂; 4th: *CH + H₂O → *CO + 1.5H₂; 5th: *CO + H₂O → * + CO₂ + H₂). The C–H bond activation was energetically advantageous for TiO₂, IrO₂, and PbO₂; however, SnO₂, with a lower M–O site, encountered a higher energy barrier. According to reactant-impulse chronoamperometry (RIC) and open circuit potential (OCP) changes along with DFT calculation, the reaction mechanism proposed that CH₄ oxidation proceeded through the formation of *CH_x and *CH_xO_y intermediates at lower and higher potentials, respectively (Fig. 11g).

Fig. 11 (a) CH₄ oxidation products and faradaic efficiency over diverse transition metal oxides in 0.1 M potassium phosphate buffer (pH = 7). The result of PtO₂ was obtained in 1 M KCl (pH = 7) because of inactivity in 0.1 M potassium phosphate buffer. (b) Scaling relationship between the measured binding energy of *CH₄ and the Madelung potential of metal in transition metal oxides. (c) Faradaic efficiency of methanol oxidation to CO₂ over TiO₂ with applied potential in the presence/absence of Cu₂O₃. Reaction profile (d) for OER on the undercoordinated metal sites, (e) for OER on the bridging oxygen binding site, and (f) for CH₄ oxidation on the undercoordinated metal sites in transition metal oxides. (g) Possible reaction pathways for CH₄ oxidation and OER in an aqueous electrolyte. Reproduced with permission.⁷⁴ Copyright 2021, National Academy of Sciences.

2D transition metal carbides (MXenes)⁷⁵. The DFT calculation of the electrocatalytic CH₄ oxidation on various 2D carbides (MXenes) offers important clues to understand the relation among the stability of active oxygen, the reactivity of C–H bond activation, and the selectivity toward specific oxygenates. Under applied anodic potential, the O-terminated surface was predominant rather than both metal-terminated and OH-terminated surface. The C–H bond was activated by an active oxygen (O*) derived from water oxidation (i.e., hydroxyl adsorption: H₂O + * → OH* + H⁺ + e⁻; deprotonation: OH* → O* + H⁺ + e⁻). As the lower limit of electrochemical potential to form active oxygen (U_form, derived from _max[ΔG_O* − ΔG_O*]/e⁻) is lower than the dissociation potential of active oxygen (U_diss, calculated by (ΔG_OOH* − ΔG_O*)/e⁻), O* can be accumulated on the surface of the catalysts. As shown in Fig. 12a, the OER follows a volcano curve; thus, the MXenes at the right and left side of the volcano curve are suitable for CH₄ oxidation and OER, respectively. The change in C–H bond activation (ΔE_a,TS = E_a,TS − E_CH4*) for the affinity energy of proton (ΔE_OH* − ΔE_O*) indicates that MXenes with a high proton affinity tend to stimulate the activation of the C–H bond (Fig. 12b). A Sabatier-like relation was observed between the stability of active oxygen and the reactivity of C–H bond activation. The active oxygen in Hf₂CO₂ and Zr₂CO₂ is too active but the stabilized O* in Ta₂CO₂, NbC₂O₂, and V₂CO₂ is too inert, consequently lowering the CH₄ conversion yield. Contrary to MXenes composed of single metals, CrHf₂CO₂ and TaHfCO₂ are energetically favorable for CH₄ activation. Depending on the applied potential for TaHfC₂O₂, the formation of intermediates, known as relatively stable species, was changed (i.e., CH₃O*, CO*, and CO₂* at 1.4, 1.8, and 2.4 V, respectively) (Fig. 12c). On TaHf₂C₂O₂, the production of oxygenates was preferred to that of hydrocarbons (inset in Fig. 12c), which is supported by the calculation based on the projected density of states (PDOS) and projected crystal orbital Hamilton population (pCOHP). For the Hf (5d²) with a few d electrons in TaHf₂C₂O₂, the Hf–O bond is stronger than the Hf–C bond; therefore, the formation of oxygenates is predominant (Fig. 12d). The observation that only hydrocarbons, CO, and CO₂ were produced on metallic Pt in a previous study⁷⁶ would be due to a strong interaction between Pt and C. The electrons in Pt–C that are able to occupy the anti-bonding orbital are fewer than those in Pt–O (Fig. 12e).

Fig. 12 (a) Free energies of the OH*, O*, and OOH* formation steps in OER as functions of ΔE_OH* − ΔE_O*. (b) The energy barrier of C–H bond activation (ΔE_a,_TS; TS means transition state) as a function of ΔE_OH* − ΔE_O*. (c) Free energy diagram of stable products in methane oxidation at different electrode potentials (1.40, 1.80, and 2.40 V_RHE). Inset: schematic diagram of the electrocatalytic conversion of CH₄ on TaHf₂C₂O₂ and blue arrows mean the energetically favorable route. Band filling of oxygen and carbon adsorbed on (d) TaHf₂C₂O₂ and (e) metal Pt. Reproduced with permission.⁷⁵ Copyright 2021, Elsevier.

Only few research articles have reported the electrocatalytic partial oxidation of CH₄ at room temperature.^40,77–81 The introduction of V₂O₅ in TiO₂/RuO₂ composites increased the current density and selectivity toward CH₃OH when a gas diffusion electrode (GDE) cell was utilized in 0.1 M Na₂SO₄.⁴⁰ The redox couple of V^5+/4+ would not give electrons to form a double bond; thus, the selectivity toward CH₃OH increased against HCHO and HCOOH. Park's group reported the partial oxidation of CH₄ to oxygenates using Co₃O₄/ZrO₂ nanocomposites,⁷⁷ Co₃O₄-incorporated ZrO₂ nanotubes,⁷⁹ and ZrO₂/NiCo₂O₄ nanowires⁸¹ using the typical single cell reactor. The Co₃O₄/ZrO₂ nanocomposites led to the formation of acetaldehyde as a major intermediate in the first 3 h, and finally turned into 1- and 2-propanol as the main products after 12 h in 0.5 M Na₂CO₃ (Fig. 13a). It was proposed that the electrochemically formed carbonate radicals would initiate the dehydrogenation of CH₄ and the addition reaction of ˙CH₃ to acetaldehyde generated C3 products. Herein, the role of ZrO₂ was assigned as a promoter to adsorb carbonates. In the control experiment, no increase in the LSV curve was observed for pristine Co₃O₄ regardless of CH₄ purging. With increasing ZrO₂ content, the current density increased and then decreased, supporting that the optimal ratio of ZrO₂ to Co₃O₄ should be considered to design highly-efficient electrocatalysts (Fig. 13b). The electrocatalytic performance was further enhanced either by the morphology control of Co₃O₄/ZrO₂ with Co₃O₄ nanoparticles embedded on the surface of ZrO₂ nanotubes⁷⁹ or ZrO₂/NiCo₂O₄ quasi-solid solution nanowires,⁸¹ which improved the mass transfer problem of CH₄.

Fig. 13 (a) Production efficiencies of 1-propanol, 2-propanol, and acetaldehyde with the reaction time for ZrO₂/Co₃O₄. (b) LSV curves of ZrO₂/Co₃O₄ with a different amount of ZrO₂ loaded. Reproduced with permission.⁷⁷ Copyright 2017, Wiley. (c) Yields of ethanol and methanol produced over NiO/Ni catalysts with different NiO contents with increasing potential. (d) Energy profiles for CH₄ oxidation to CH₃OH and C₂H₅OH at the NiO(200)/Ni(111) interface. Reproduced with the permission.⁷⁸ Copyright 2020, Elsevier.

The electrocatalytic conversion of CH₄ to C₂H₅OH was investigated on NiO/Ni and Rh/ZnO catalysts. The NiO/Ni electrode in the potential range from 1.37 to 1.43 V_RHE produced C₂H₅OH, whose selectivity was recorded to be about 81–89% in 0.1 M NaOH.⁷⁸ Counting the formation of CH₃OH as a byproduct, the selectivity toward oxygenates was estimated to be over 90%. The yield of C₂H₅OH and CH₃OH increased up to 1.40 V, and then decreased with applied potential (Fig. 13c). The DFT calculation revealed that the formation of C₂H₅OH is more favorable than that of CH₃OH, and the proposed reaction pathway follows the sequential steps (i.e., CH₄* → CH₃* + H*; CH₃* → CH₂* + H*; CH₂* + OH* → CH₂OH*; CH₃* + CH₂OH* → CH₃CH₂OH*) (Fig. 13d). Regarding Rh/ZnO nanosheets, the conversion, faradaic efficiency, and selectivity toward C₂H₅OH was measured as 789 μmol g_cat⁻¹ h⁻¹, 22.5%, and 85%, respectively, at 2.2 V_RHE in 0.1 M KOH.⁸⁰ It was suggested that Rh nanoparticles on ZnO nanosheets accelerated the adsorption of CH₄ and the formation of active oxygen species. The operando analysis is a powerful tool to understand the real-time process involved in the activation and conversion of CH₄. Hahn et al. investigated the activation of CH₄ on the Pt catalyst at room temperature using attenuated total reflectance-surface enhanced infrared absorption spectroscopy (ATR-SEIRAS), and it was confirmed that the intermediates –CHO (aldehyde) and –COOH (acid) were formed during the oxidation of CH₄.⁸² Furthermore, the intermediates were detected by in situ linear potential sweep-Fourier transform infrared spectroscopy (LPS-FTIRS). Together with CO₂ (2345 cm⁻¹), CO (2020 cm⁻¹), and H₂O (1650 cm⁻¹) peaks, a broad peak (1750 cm⁻¹) was observed due to the formation of aldehydic and ketonic species with the –CO moiety. At 1.46 V, the intermediate peaks for –CHO (1719 cm⁻¹) and –COOH (2027 cm⁻¹) were measured on the Pt electrode surface. The electrochemical oxidation of CH₄ over Pt/C, Pt/C-ATO, Pd/C, and Pd/C-ATO electrocatalysts was also confirmed by in situ ATR-Fourier transform infrared (ATR-FTIR) spectroscopy.⁸³ Herein, the intermediates –CHO and CO were observed in the potential range from 0.05 to 1.2 V, which were clarified as intermediates for the production of CO₂.

3.3. Photoelectrochemical CH₄ conversion

When the Fermi level of the semiconductor and the chemical potential of the electrolyte (chemical species) are at equilibrium, ‘band-bending’ occurs depending on n-type and p-type materials. Under applied external bias or by the configuration of the tandem cell, photogenerated electron–hole pairs are separated, and the recombination is significantly retarded. In the n-type photoanode, the photogenerated holes transferred to the electrolyte/semiconductor interface initiate the oxidation. At the same time, the electrons collected on the counter electrode begin the reduction reaction. Ideally, the photogenerated holes oxidize CH₄ to oxygenates, while the electrons transferred to the cathode produce H₂. As mentioned before, ROS formed by water oxidation can attack and activate the stubborn C–H bond in terms of lowering the energy barrier, accompanying the dehydrogenation of CH₄ through a radical process. The technical advantages of the PEC cells including the controllable selectivity and production rate by applied potential, tunability of the electrode properties with the cocatalyst, and the facile fabrication of the cells are visible. However, it has failed to overcome a low efficiency and selectivity problem for the oxidation of CH₄. Due to limited studies, we discuss the PEC conversion of CH₄ to oxygenates and hydrocarbons as well.

Amano et al. reported the activation of CH₄ on WO₃ photoanode with a high quantum efficiency under visible light irradiation at room temperature.⁸⁴ Applying external bias enabled the use of photogenerated electrons for H₂ production and the deeper depletion layer formed in WO₃ suppressed the recombination of charge carriers. In order to solve the low solubility and mass transport problems of CH₄, a bimodal (meso and macro) porous WO₃ photoanode was fabricated via the deposition of WO₃ nanoparticles on the microporous 3D Ti microfiber. The concentration of protons between the photoanode and cathode chambers was balanced by the assembly with Nafion membrane (Fig. 14a and b). Since the valence band maximum of WO₃ is more positive than the standard electrode potential of CH₄ oxidation (CH₄ + h⁺ → ˙CH₃ + H⁺ + e⁻; E°(˙CH₃/CH₄) = 2.06 V_SHE), the photogenerated holes can dehydrogenate CH₄ into a methyl radical (˙CH₃). Consequently, the homogeneous coupling of ˙CH₃ brought about the production of C₂H₆ (˙CH₃ + ˙CH₃ → C₂H₆). The gas–electrolyte–solid triple-phase boundary significantly improved the oxidation of the inert C–H bond of CH₄ and led to the formation of CO₂ and C₂H₆. When the concentration of CH₄ was reduced in the feed (i.e., the stream of Ar with 10% CH₄ and 3 vol% H₂O), the production of C₂H₆ was negligible (Fig. 14c). In reverse, as the concentration of CH₄ fed increased, C₂H₆ production steadily increased. Under 97/3 vol% of CH₄/H₂O stream, the selectivity and the faradaic efficiency toward C₂H₆ reached up to 53.5% and 12%, respectively. These results indicate that the access of gaseous CH₄ to the surface of the WO₃ photoanode is important for preferential C–H bond activation by photogenerated holes. The selectivity toward C₂H₆ could be improved by making the surface of WO₃ hydrophobic for the effective adsorption of CH₄ instead of water.

Fig. 14 (a) PEC system (left) and its expansion to MEA (right) for CH₄ activation. (b) SEM image of WO₃ deposited on Ti microfibers. Inset: picture of the PEC electrode (left lower) and MEA (right upper). (c) Time-dependent production of O₂, CO₂, and C₂H₆ in the photoanode compartment under 453 nm blue LED light irradiation at an applied voltage of 1.2 V. Reproduced with permission.⁸⁴ Copyright 2019, American Chemical Society.

Ma et al. optimized the reactivity of ˙OH on WO₃via facet tuning to produce ethylene glycol (EG) under mild conditions without adding any oxidants.⁴² Three WO₃ photoanodes with different (010) facet ratios including WO₃ nanobar arrays (WO₃NB), WO₃ nanoplate arrays (WO₃NP), and WO₃ nanoflake arrays (WO₃NF) were fabricated for comparison (Fig. 15a). Among them, WO₃NB with a predominant (010) facet revealed a superior CH₄ conversion performance. To confirm the production of ˙OH on the catalysts with different facets, electron paramagnetic resonance (EPR) spectroscopy was measured using 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as a spin-trapping agent. The signal for DMPO–˙OH on WO₃NB was stronger than that of other WO₃ structures. Based on the result that no H₂O₂ formation was observed during the experiment, it was expected that the oxidation of water was the only source to provide ˙OH. Consequently, the EG production rate, selectivity, and solar-to-fuel efficiency in WO₃NB were recorded to be 0.47 μmol cm⁻² h⁻¹, 66%, and 0.12% at 1.3 V_RHE, respectively (Fig. 15b). The in situ diffuse reflectance infrared Fourier-transform spectroscopy (DRIFTS) data support that EG was steadily produced on the PEC cell (Fig. 15c). The typical peaks of EG at 2853 and 2924 cm⁻¹ (cf., assigned to methylene (CH₂) symmetric C–H stretching and antisymmetric C–H stretching, respectively) were measured, which increased with the irradiation time. Moreover, the emergence of methyl (CH₃) symmetric C–H stretching at 2879 cm⁻¹ and antisymmetric C–H stretching of CH₃OH at 2957 cm⁻¹ indicate that CH₃OH was formed as an intermediate during the PEC process. Therefore, the mechanism can be explained as follows: CH₃OH was formed by the reaction between ˙OH and ˙CH₃, which was further attacked by ˙OH, finally producing the EG (Fig. 15d).

Fig. 15 (a) WO₃ with a different (010) facet ratio grown on FTO glasses. (b) EG production rate in term of PEC CH₄ conversion on WO₃ photoanodes under 100 mW cm⁻² irradiation. (c) In situ DRIFTS for WO₃NB during the reaction. (d) Proposed mechanism for the conversion of CH₄ to EG. Reproduced with permission.⁴² Copyright 2021, Wiley.

It should be worth highlighting other metal oxide candidates for PEC cell applications. Fe₂O₃ (hematite) has a suitable valence band minimum for CH₄ oxidation and an ideal visible light absorption with n-type property, whose band gap energy was about 2.1–2.2 eV.^85,86 Iron oxide is abundant in nature, cheap, and non-toxic, and the hematite-based photoanode is very stable during OER in an alkali electrolyte.⁸⁷ BiVO₄ with a bandgap of 2.4 eV shows an excellent PEC OER performance and is relatively stable in neutral pH.^88,89 Metal chalcogenide, C₃N₄, and metal carbide are also good candidates for visible light PEC CH₄ oxidation because the outstanding photocatalytic activity was already reported for the partial oxidation of CH₄ in slurry-type systems.^90,91 The modification of the bulk and surface properties of semiconducting materials and cocatalysts or the introduction of highly active photocatalysts and electrocatalysts in PEC systems should be helpful to advance the design of the PEC device with high performance under sunlight.

3.4. Activation of oxidants for radical-assisted CH₄ conversion

While the direct interaction between the surface-active site of heterogeneous (photo)electrocatalysts and CH₄ stimulates the activation of CH₄ in terms of the dehydrogenation and deprotonation processes, free-standing radical species (especially, ˙OH and ˙Cl) floating in the solution enable the extraction of hydrogen from CH₄. Like Shilov cycle (Fig. 5b), the supply of ˙OH can be maintained by the Fenton process unless H₂O₂ is completely consumed. The ferric/ferrous redox cycle is regenerated by the electron transfer induced by photo and electrocatalysis. Under anaerobic conditions, a hole-mediated oxidation or an anodic reaction of water or chloride is the only way to produce active radical species. However, under aerobic conditions, the reductive pathway (i.e., electron-mediated reactions) produces H₂O₂ and further electron transfer to in situ formed H₂O₂ generates ˙OH. Since an excessive amount of ˙OH leads to the overoxidation of CH₄, it is essential to establish the optimized experimental condition in order to increase the selectivity of the desired products.

In the presence of H₂O₂ under mild conditions (below 100 °C), (i) iron-based catalysts (e.g., single iron atoms confined in graphene,⁹² ferric chloride (FeCl₃),⁹³ Fe–Cu-ZSM-5,⁹⁴ Fe/ZSM-5,^95,96 Fe/Fe₃C,³⁵ N-doped carbon supported iron species,⁹⁷etc.), (ii) atomic metals or metal nanoparticles embedded/loaded on metal oxides (e.g., single atomic Rh/ZrO₂,⁹⁸ AuPd/TiO₂,⁹⁹ single atomic Cr/TiO₂,¹⁰⁰ Rh/CeO₂,¹⁰¹ single atomic Cu/C₃N₄,¹⁰²etc.), and (iii) bimetallic materials (e.g. Pd@Pt core–shell nanoparticles,¹⁰³ Au–Pd alloys,¹⁰⁴ AuPd_x colloids,¹⁰⁵etc.) have been investigated for the partial oxidation of CH₄. For example, graphene-confined single Fe sites (FeN₄/graphene nanosheets (GNs)) (Fig. 16a) effectively oxidized CH₄ to C1 oxygenated products including CH₃OH, CH₃OOH, HOCH₂OOH, and HCOOH in the liquid phase, whose turnover frequency (TOF) reached to 0.47 h⁻¹ at room temperature.⁹² The labelling experiment using ¹³CH₄ revealed that C1 oxygenated products were originated from the partial oxidation process instead of the self-oxidation of graphene (Fig. 16b). As the Fe content increased from 1.5 to 4.0 wt%, the yield of C1-oxygenated products increased from 27 to 100 μmol; however, the turnover number (TON) peaked at 2.7 wt% Fe and then decreased with Fe wt%. While more active sites emerged with the increase in the Fe content, the agglomeration among excess Fe species adversely affected the catalytic performance. Besides, with longer reaction time, the C1-oxygenated products were prone to mineralization into CO₂. Under the optimized conditions, the selectivity of C1-oxygenated products and CO₂ was 94 and 6%, respectively. Based on in operando time-of-flight mass spectrometry (TOF-MS) data, it was confirmed that the formation of CH₃OH and CH₃OOH was predominant in the first 300 min, whereas HOCH₂OOH and HCOOH markedly increased in the last 300 min (Fig. 16c). The DFT calculation supports that the active oxygen atom can be easily formed by the adsorption and decomposition of H₂O₂ on the Fe site (Fig. 16d). The methyl radical generated through CH₄ activation over O–FeN₄–O would attack the hydroxide and hydroperoxide groups, which produced CH₃OH and CH₃OOH, and the further oxidation of CH₃OH allowed to form HOCH₂OOH and HCOOH (Fig. 16e). As FeCl₃ was utilized as a homogenous catalyst in the presence of H₂O₂, the yield of CH₃OH and HCO₂H was recorded as 1.97 and 33.3 mmol g_cat⁻¹ with a TOF of 5.7 h⁻¹, respectively.⁹³ It was proposed that the hydrated ferryl ion, [(H₂O)₅Fe^IVO]²⁺, that originated from the reaction with H₂O₂ (i.e., Fe³⁺ + H₂O₂ → [Fe^IVO]²⁺ + ˙OH + H⁺ or Fe²⁺ + H₂O₂ → [Fe^IVO]²⁺ + H₂O) led to the activation of CH₄ and consequently the formation of CH₃OH. On the other hand, ˙OH provided by the Fenton process brought about the production of HCO₂H, CO, and CO₂.

Fig. 16 (a) HAADF-STEM image of FeN₄/GN. Single iron atoms are marked by red circles. (b) TOF-MS data for the labelling experiment using ¹³CH₄. The W denotes water molecule. (c) In operando TOF-MS data collected under different pressure of ¹³CH₄. (d) Energy change of FeN₄ center by the adsorption of H₂O₂. (e) Proposed mechanism on the partial oxidation of CH₄ over FeN₄/GN catalysts. Reproduced with the permission.⁹² Copyright 2018, Elsevier.

On irradiating UV light with a wavelength below 380 nm, the direct photolysis of H₂O₂ drove the formation of ˙OH, which promoted the oxidation of CH₄ to CH₃OH and HCO₂H.³³ With an excess amount of H₂O₂, the consumption of ˙OH by H₂O₂ (H₂O₂ + ˙OH → ˙OH₂ + H₂O; ˙OH₂ + ˙OH → O₂ + H₂O) decreased the CH₄ conversion. Although the overoxidation of CH₄ was inevitable, the complete mineralization to CO₂ was significantly retarded by establishing the steady state condition that H₂O₂ was added at a constant and relatively low rate. To overcome the drawback, especially a need for energy-intensive UV light and low quantum yield, the activation of H₂O₂ on diverse photocatalysts (e.g., WO₃,^49,53 g-C₃N₄,⁴⁷ TiO₂,³⁴ iron clusters anchored on carbon aerogel,¹⁰⁶ and BiVO₄ ⁵⁵) and its correlation with the oxidation of CH₄ has been carried out. Under simulated solar light irradiation, the system containing TiO₂, Fe²⁺, and H₂O₂ transformed CH₄ to CH₃OH, whose yield and selectivity reached up to 471 μmol g⁻¹ h⁻¹ and 83%, respectively.³⁴ The conduction and valence band energy levels of TiO₂ are suitable for the redox reaction of ferric/ferrous ions and the formation of ˙OH by virtue of either water oxidation or H₂O₂ activation (Fig. 17a). Compared with the cases (e.g., TiO₂ + Fe²⁺ + light; TiO₂ + H₂O₂ + light; Fe²⁺ + H₂O₂), a higher yield and selectivity toward CH₃OH was achieved (Fig. 17b). The screening test using K₂Cr₂O₇, para-quinone, salicylic acid, and Na₂C₂O₄ as scavengers of e⁻, O₂˙⁻, ˙OH, and h⁺, respectively, supports (i) the regeneration of Fe²⁺ is induced by the electron transfer, (ii) the activation of H₂O₂ produces ROS, and (iii) the oxidation of CH₄ proceeds through photogenerated holes and ROS (Fig. 4c and 17c).

Fig. 17 (a) TiO₂ conduction and valence band position and redox potentials of the active species. (b) Yield of CH₄ to CH₃OH under AM 1.5 at room temperature for different systems. (c) Scavenger experiments of ROS (h⁺, e⁻, ˙OH, and ˙O₂⁻). Reproduced with permission.³⁴ Copyright 2020, Royal Society of Chemistry. (d) HAADF-STEM image and elemental mapping of carbon and iron for 0.75FeCA800-4. (e) Photocatalytic conversion kinetics and selectivity for the partial oxidation of CH₄ to CH₃OOH, depending on the H₂O₂ concentration. (f) In situ DRIFTS spectra during photocatalysis. (g) In situ EPR spectra of the Fe^δ+ contents with the irradiation time. (h) The proposed mechanism of the conversion of CH₄ to CH₃OOH over 0.75FeCA800-4. Reproduced with the permission.¹⁰⁶ Copyright 2021, American Chemical Society.

The positively charged iron clusters anchored on 3D porous carbon aerogel, denoted as 0.75FeCA800-4, has a good activity for the partial oxidation of CH₄ to CH₃OOH. The 3D porous carbon aerogel with a high electrical conductivity is a good support; therefore, iron clusters with size from 0.25 to 0.45 nm were uniformly dispersed (Fig. 17d). Under simulated solar light irradiation, the production rate and selectivity toward CH₃OOH were 13.2 mmol g_Fe⁻¹ h⁻¹ and almost 100% in the presence of H₂O₂ (2 mmol), respectively (Fig. 17e). The trend that an excess amount of H₂O₂ significantly cut down the selectivity owing to the overoxidation of CH₃OOH to CO₂ was also observed. The in situ DRIFTS and in situ EPR are powerful tools to understand the real-time reaction mechanism during photocatalysis. As seen in Fig. 17f, new peaks observed at 1306/1348, 1453, and 1550 cm⁻¹, assigned to *C–H, *C–O, and *CH₃–O groups, respectively, increased with the irradiation time. Indeed, Fig. 17g demonstrates that the intensity of both the high spin state Fe^δ+ at 200 G and the low spin state Fe^δ+ at 3750 G was markedly enhanced with the irradiation time, which indicates that Fe^δ+ gave electrons to the carbon atoms. The proposed mechanism is as follows: (i) photoexcitation of carbon atoms in 0.75FeCA800-4, (ii) electron transfer from Fe^δ+ to C*, (iii) formation of C*⁻, (iv) production of CH₄⁻via the electron transfer from C*⁻ to CH₄, (v) generation of Fe^δ+–CH₃* intermediates, (vi) reaction of Fe^δ+–CH₃* with ˙OH formed by the activation of H₂O₂, (vii) formation of Fe^δ+–CH₃O* intermediates, (viii) reaction of Fe^δ+–CH₃O* with ˙OH, and finally, (ix) the production of CH₃OOH (Fig. 17h).

The introduction of a relatively excess amount of oxidants and their activation to ROS with high oxidation power may damage the membrane in electrochemical cells. Therefore, oxygen reduction reaction (ORR) catalysts should promote 4e⁻ transfer (O₂ + 4H⁺ + 4e⁻ → 2H₂O) with suppressed 2e⁻ transfer (O₂ + 2H⁺ + 2e⁻ → H₂O₂) to be utilized for a proton exchange membrane fuel cell (PEMFC).¹⁰⁷ Although no studies have been reported so far except for one paper, the cathodic activation of H₂O₂ for CH₃OH production in a PEMFC,¹⁰⁸ the recent research work provides a profound insight into the feasibility of producing and handling reactive species for electrocatalytic CH₄ conversion.¹⁰⁹ The cobalt–nickel mixed spinels of Co₂NiO₄ consisting of tetrahedral Ni²⁺ and octahedral Co³⁺ and CoNi₂O₄ composed of octahedral Ni²⁺ and tetrahedral Co³⁺ (Fig. 18a) were favorable for generating active chlorine species (*Cl), which promoted the production of CH₃Cl under an applied potential. The surface adsorbed *Cl as an intermediate of the chlorine evolution reaction (CIER) stimulated the cleavage of the C–H bond and overoxidation could be avoided by the release of gaseous CH₃Cl from the electrolyte. In 0.5 M Na₂SO₄ electrolyte, the current density in the LSV curve was quite low due to a high energy barrier, and only CO₂ was evolved (Fig. 18b). However, when an excess amount of Cl⁻ was present (5.4 M NaCl), the current density at 2.3 V rose up to 19.5 mA cm⁻² and the CIER-assisted pathway prevailed on the OER-assisted one. The in situ EPR experiment confirmed the formation of *Cl and subsequently ˙Cl, where DMPO–Cl˙, DMPO–OH˙, and DMPO–OCl˙ were simultaneously observed (Fig. 18c). Regarding the conversion of CH₄ to CH₃Cl, a higher overpotential to make active oxygen atoms (Ni–O*) is required; thus, the overoxidation of liquid oxygenates is inevitable (Fig. 18d).

Fig. 18 (a) Co–Ni mixed spinel structures consisting of CoNi₂O₄ (left) and Co₂NiO₄ (right). (b) LSV curve (upper panel) and partial current density for products in 0.5 M Na₂SO₄ and saturated NaCl (lower panel). (c) In situ EPR spectrum collected using DMPO as a radical capture reagent in electrocatalysis. (d) Different mechanism of CH₄ oxidation in terms of ClER (upper panel) and OER (lower panel). Reproduced with the permission.¹⁰⁹ Copyright 2021, Wiley.

4. Conclusions and outlook

Photocatalysis, electrocatalysis, PEC cell, and radical-mediated AOP techniques are capable of bypassing energy-intensive processes (i.e., high temperature and pressure) but are available to utilize the photonic energy (cf., ideally, direct absorption of solar light or PV-powered electrolyzer) as a driving force for the activation of the C–H bond (Scheme 1). The major hurdle of CH₄ partial oxidation refers to (i) low conversion yield of CH₄, (ii) a competitive OER process, (iii) poor selectivity induced by the overoxidation, (iv) mass transfer problem of CH₄ and the final products, (v) separation of the desired products from the electrolytes, (vi) scale-up issue, and (vii) a mysterious reaction mechanism, which should be overcome and verified for practical applications.

Scheme 1 A schematic representation of photocatalytic (PC), electrocatalytic (EC), photoelectrochemical (PEC), and advanced oxidation process (AOP)-based activation of methane as a tool to detour the energy-intensive process. Partial oxidation processes driven by clean and sustainable technologies are one of the most promising ways to obtain valuable products directly from methane.

Photocatalysis follows a series of processes including (i) photon absorption, (ii) exciton separation, (iii) carrier diffusion, (iv) carrier transport, (v) catalytic efficiency, and (vi) mass transfer of the reactants and products. Therefore, the overall reaction kinetics are governed by the most sluggish step, which means that all the six factors should be optimized in order to increase the photocatalytic activity. In addition, the electronic structure of the photocatalyst (i.e., the bandgap, conduction and valence band position, energy level of the trapping sites for charge carriers, and Fermi level equilibrium) is associated with the thermodynamics that determine whether the reaction proceeds or not. When focusing on not only interfacial charge transfer but also the activation of the C–H bond, the direct dehydrogenation of CH₄ likely proceeds by photogenerated holes over the surface of the photocatalysts. On the other hand, ˙OH provided from the cleavage of in situ formed H₂O₂ by electron transfer or hole-mediated water oxidation enables to abstract hydrogen from CH₄ in the proximity of the catalysts. The formation of ˙OH and O₂⁻˙ is adjusted by the addition of H₂O₂ or by the reaction under aerobic conditions. The combination of ˙CH₃ with ˙OH produces oxygenates, and the selectivity toward the desired products can be improved by controlling the concentration of ˙OH by suppressing the overoxidation. One of the strategies will be the introduction of a proper scavenger of ˙OH; for example, carbonate (CO₃²⁻) effectively quenches ˙OH (k(HCO₃⁻) = 8.5 × 10⁶ M⁻¹ s⁻¹ and k(CO₃²⁻) = 3.9 × 10⁸ M⁻¹ s⁻¹),¹¹⁰ resulting in the formation of CO₃˙⁻ with a higher standard reduction potential (E°(CO₃˙⁻/CO₃²⁻) = 1.63 V).¹¹¹ Likewise, such a high oxidation power of ˙OH would be systematically reduced by electron transfer from the halide ions (E°(Cl˙/Cl⁻) = 2.5 V; E°(Cl₂˙⁻/Cl⁻) = 2.2 V; E°(Br˙/Br⁻) = 2.0 V; E°(Br₂˙⁻/Br⁻) = 1.7 V).¹¹² Another approach is the control of the HOMO and LUMO energy levels in terms of the synthesis and modification of transition metal complexes. Not only the change in the oxidation potential of ˙OH by redox couples but also the emergence of high-valent transition metal ions may assist C–H bond activation and increase the yield and selectivity. Beyond the utilization of H₂O₂, persulfate (peroxymonosulfate and peroxydisulfate), periodate, hypochlorous acid, etc., are also good candidates to supply the radical species through the photocatalytic process. As a non-radical reaction pathway, the influence of singlet oxygen (¹O₂) formed via the conversion of the singlet excited state to the triplet excited state on the photocatalytic oxidation of CH₄ needs to be proven.

The electrocatalytic partial oxidation of CH₄ to oxygenates under mild conditions is still at quite an early stage, and very limited studies have been undertaken so far. The design of appropriate electrocatalysts and electrochemical cells in consideration of factors such as an electrode configuration, a membrane, an electrolyte, a surrounding condition, etc., is critical to increase the conversion of CH₄ and its selectivity to desired products. In general, high selectivity is achieved at a low current density (i.e., a low conversion yield due to a small number of electrons involved), and as the current density rises, the selectivity significantly decreases due to not only the overoxidation of the target products but also the competitive reaction with OER. To overcome the competitive OER, the development of catalysts with a high Gibbs free energy for water oxidation but with a low one for CH₄ activation is highly required. Under an applied bias, active M–O sites on transition metal oxides promote the dehydrogenation of CH₄, in which the catalyst with a high CH₄ binding energy but a low Madelung potential leads to a higher performance. In the presence of proton acceptors, the deprotonation pathway drives the activation of the C–H bond as well. Therefore, materials that can simultaneously stimulate the dehydrogenation and deprotonation will be favorable for the conversion of CH₄, where more precise control of the active sites is essential to optimize the selectivity. To better understand the reaction mechanism, real-time spectroscopy is necessary to monitor the change of (i) the oxidation states and electronic structures of transition metal-based catalysts, (ii) reorientation of surface atoms and defects, (iii) electrode surface pH and interaction with electrolyte ions confined in the microenvironment, (iv) the formation of transient species and intermediates, and their correlation with DFT calculation. Through either by the fabrication of MEAs or the modulation of cells by stacking, the size of the device can be scaled-up; thus, the productivity increases to meet the demand of industries. The development of membranes that enable the desired products to be transported from the anode to the cathode chamber, and electrolyte engineering that may help to stabilize oxygenates by solvation (e.g., ionic liquid and H₂SO₄) may effectively retard the overoxidation process. As proposed in photocatalysis and electrocatalysis, the same strategy may be valid to optimize the performance of PEC cells.

The radical-mediated AOP technique looks similar to a double-edged sword because ROS formed by the activation of oxidants accelerate the oxidation of CH₄ but the excess amount drives the complete mineralization into CO₂. Indeed, the prolonged reaction time overoxidizes the desired products. Therefore, the addition of H₂O₂ at regular time intervals to maintain a steady state condition may be more advantageous to reach a higher selectivity. The activation of oxidants into ROS occurs via heterogeneous catalysis, whereas the dehydrogenation of CH₄ by ROS is proceeded in the homogeneous phase. Ideally, if ROS are instantaneously produced (i.e., 100% conversion of H₂O₂ to ˙OH within a few seconds) and the concentration of dissolved CH₄ is enough to consume all ˙OH, overoxidation will be suppressed due to the termination of the radical reaction in the second scale. The concept is quite similar to burst nucleation to synthesize uniform-sized metal nanoparticles.¹¹³ Under atmospheric pressure, micro or nanobubbling of CH₄ would increase the solubility of CH₄ as well as the collision probability of ROS. In order to commercialize the electrocatalytic CH₃OH production directly from CH₄, the conversion efficiency should reach up to 70% by satisfying the operation condition (i.e., cell area ≥200 cm², current density ≥100 mA cm⁻², and applied bias ≤1 V).²³ Although it may seem like a distinct future when it comes to realization under mild conditions, it is a promising research topic and rapid scientific progress will advance the deployment of self-standing reactors at commercial scales.

Author contributions

Love Kumar Dhandole: investigation, visualization, writing – original draft. Sang Hoon Kim: investigation, writing – original draft. Gun-hee Moon: conceptualization, supervision, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was financially supported by the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (NRF-2020M3H4A3106354), by Korea Environment Industry & Technology Institute (KEITI) through Ecological Imitation-based Environmental Pollution Management Technology Development Project, funded by Korea Ministry of Environment (MOE) (2021002800005), and the KIST internal projects.

Notes and references

1. I. Pérez-Domínguez, A. del Prado, K. Mittenzwei, J. Hristov, S. Frank, A. Tabeau, P. Witzke, P. Havlik, H. van Meijl, J. Lynch, E. Stehfest, G. Pardo, J. Barreiro-Hurle, J. F. L. Koopman and M. J. Sanz-Sánchez, Nat. Food, 2021, 2, 970–980.
2. T. Vandyck, K. Keramidas, A. Kitous, J. V. Spadaro, R. Van Dingenen, M. Holland and B. Saveyn, Nat. Commun., 2018, 9, 4939.
3. J. J. West, A. M. Fiore, L. W. Horowitz and D. L. Mauzerall, PNAS, 2006, 103, 3988–3993.
4. M. Saunois, A. R. Stavert, B. Poulter, P. Bousquet, J. G. Canadell, R. B. Jackson and e. al, Earth Syst. Sci. Data, 2020, 12, 1561–1623.
5. E. McFarland, Science, 2012, 338, 340–342.
6. R. Horn and R. Schlögl, Catal. Lett., 2015, 145, 23–39.
7. M. Wang, X. Tan, J. Motuzas, J. Li and S. Liu, J. Membr. Sci., 2021, 620, 118909.
8. S. D. Angeli, G. Monteleone, A. Giaconia and A. A. Lemonidou, Int. J. Hydrogen Energy, 2014, 39, 1979–1997.
9. M. J. Brown and N. D. Parkyns, Catal. Today, 1991, 8, 305–335.
10. N. J. Gunsalus, A. Koppaka, S. H. Park, S. M. Bischof, B. G. Hashiguchi and R. A. Periana, Chem. Rev., 2017, 117, 8521–8573.
11. V. N. Cavaliere and D. J. Mindiola, Chem. Sci., 2012, 3, 3356–3365.
12. Z. Duan and S. Mao, Geochim. Cosmochim. Acta, 2006, 70, 3369–3386.
13. Z. Liang, T. Li, M. Kim, A. Asthagiri and J. F. Weaver, Science, 2017, 356, 299–303.
14. G. Fang, J. Lin and X. Wang, Chin. J. Chem. Eng., 2021, 38, 18–29.
15. G.-h. Moon, S. Kim, Y.-J. Cho, J. Lim, D.-h. Kim and W. Choi, Appl. Catal., B, 2017, 218, 819–824.
16. P. A. Cook, M. N. R. Ashfold, Y.-J. Jee, K.-H. Jung, S. Harich and X. Yang, Phys. Chem. Chem. Phys., 2001, 3, 1848–1860.
17. H. Park, H.-i. Kim, G.-h. Moon and W. Choi, Energy Environ. Sci., 2016, 9, 411–433.
18. W. Kim, T. Tachikawa, G.-h. Moon, T. Majima and W. Choi, Angew. Chem., Int. Ed., 2014, 53, 14036–14041.
19. J. Y. Hwang, G.-h. Moon, B. Kim, T. Tachikawa, T. Majima, S. Hong, K. Cho, W. Kim and W. Choi, Appl. Catal., B, 2021, 286, 119905.
20. J. C. Fornaciari, D. Primc, K. Kawashima, B. R. Wygant, S. Verma, L. Spanu, C. B. Mullins, A. T. Bell and A. Z. Weber, ACS Energy Lett., 2020, 5, 2954–2963.
21. K. Liu, W. A. Smith and T. Burdyny, ACS Energy Lett., 2019, 4, 639–643.
22. S. Yuan, Y. D. Li, J. Y. Peng, Y. M. Questell-Santiago, K. Akkiraju, L. Giordano, D. J. Zheng, S. Bagi, Y. Roman-Leshkov and Y. Shao-Horn, Adv. Energy Mater., 2020, 10, 2002154.
23. P. Promoppatum and V. Viswanathan, ACS Sustainable Chem. Eng., 2016, 4, 1736–1745.
24. J. M. Roscoe and M. J. Thompson, 1985, 17, 967–990.
25. A. V. de Vekki and S. T. Marakaev, Russ. J. Appl. Chem., 2009, 82, 521–536.
26. X. Meng, X. Cui, N. P. Rajan, L. Yu, D. Deng and X. Bao, Chem, 2019, 5, 2296–2325.
27. R. P. Noceti, C. E. Taylor and J. R. D'Este, Catal. Today, 1997, 33, 199–204.
28. C. E. Taylor, Catal. Today, 2003, 84, 9–15.
29. S. Murcia-López, K. Villa, T. Andreu and J. R. Morante, ACS Catal., 2014, 4, 3013–3019.
30. W. Zhu, M. Shen, G. Fan, A. Yang, J. R. Meyer, Y. Ou, B. Yin, J. Fortner, M. Foston, Z. Li, Z. Zou and B. Sadtler, ACS Appl. Nano Mater., 2018, 1, 6683–6691.
31. D. Yu, Y. Jia, Z. Yang, H. Zhang, J. Zhao, Y. Zhao, B. Weng, W. Dai, Z. Li, P. Wang, J. A. Steele, M. B. J. Roeffaers, S. Dai, H. Huang and J. Long, ACS Sustainable Chem. Eng., 2022, 10, 16–22.
32. K. Villa, S. Murcia-López, T. Andreu and J. R. Morante, Catal. Commun., 2015, 58, 200–203.
33. Á. López-Martín, A. Caballero and G. Colón, J. Photochem. Photobiol., A, 2017, 349, 216–223.
34. Y. Zeng, H. C. Liu, J. S. Wang, X. Y. Wu and S. L. Wang, Catal. Sci. Technol., 2020, 10, 2329–2332.
35. Y. Xing, Z. Yao, W. Li, W. Wu, X. Lu, J. Tian, Z. Li, H. Hu and M. Wu, Angew. Chem., Int. Ed., 2021, 60, 8889–8895.
36. A. Wagner, C. D. Sahm and E. Reisner, Nat. Catal., 2020, 3, 775–786.
37. G.-h. Moon, M. Yu, C. K. Chan and H. Tüysüz, Angew. Chem., Int. Ed., 2019, 58, 3491–3495.
38. M. Yu, G. Moon, E. Bill and H. Tüysüz, ACS Appl. Energy Mater., 2019, 2, 1199–1209.
39. G.-h. Moon, Y. Wang, S. Kim, E. Budiyanto and H. Tüysüz, ChemSusChem, 2022, 15, e202102114.
40. R. S. Rocha, R. M. Reis, M. R. V. Lanza and R. Bertazzoli, Electrochim. Acta, 2013, 87, 606–610.
41. R. S. Kim and Y. Surendranath, ACS Cent. Sci., 2019, 5, 1179–1186.
42. J. Ma, K. Mao, J. Low, Z. Wang, D. Xi, W. Zhang, H. Ju, Z. Qi, R. Long, X. Wu, L. Song and Y. Xiong, Angew. Chem., Int. Ed., 2021, 60, 9357–9361.
43. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253–278.
44. K. Takanabe, ACS Catal., 2017, 7, 8006–8022.
45. N. Turro, V. Ramamurthy and A. Scaiano, Modern Molecular Photochemistry of Organic Molecules, University Science Books, CA, 2010.
46. K. Villa, S. Murcia-López, J. R. Morante and T. Andreu, Appl. Catal., B, 2016, 187, 30–36.
47. S. Shi, Z. Sun, C. Bao, T. Gao and Y. H. Hu, Int. J. Energy Res., 2020, 44, 2740–2753.
48. Y. Hu, S. Higashimoto, S. Takahashi, Y. Nagai and M. Anpo, Catal. Lett., 2005, 100, 35–37.
49. M. A. Gondal, A. Hameed and A. Suwaiyan, Appl. Catal., A, 2003, 243, 165–174.
50. W. Zhou, X. Qiu, Y. Jiang, Y. Fan, S. Wei, D. Han, L. Niu and Z. Tang, J. Mater. Chem. A, 2020, 8, 13277–13284.
51. Y. Hu, Y. Nagai, D. Rahmawaty, C. Wei and M. Anpo, Catal. Lett., 2008, 124, 80.
52. H.-B. Huang, Z.-B. Fang, K. Yu, J. Lü and R. Cao, J. Mater. Chem. A, 2020, 8, 3882–3891.
53. K. Villa, S. Murcia-López, T. Andreu and J. R. Morante, Appl. Catal., B, 2015, 163, 150–155.
54. S. Wang, L. Wang and W. Huang, J. Mater. Chem. A, 2020, 8, 24307–24352.
55. S. Murcia-López, K. Villa, T. Andreu and J. R. Morante, Chem. Commun., 2015, 51, 7249–7252.
56. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76–80.
57. F. K. Kessler, Y. Zheng, D. Schwarz, C. Merschjann, W. Schnick, X. Wang and M. J. Bojdys, Nat. Rev. Mater., 2017, 2, 17030.
58. S. Cao, J. Low, J. Yu and M. Jaroniec, Adv. Mater., 2015, 27, 2150–2176.
59. Y. Li, J. Li, G. Zhang, K. Wang and X. Wu, ACS Sustainable Chem. Eng., 2019, 7, 4382–4389.
60. J. Xie, R. Jin, A. Li, Y. Bi, Q. Ruan, Y. Deng, Y. Zhang, S. Yao, G. Sankar, D. Ma and J. Tang, Nat. Catal., 2018, 1, 889–896.
61. X. Chen, Y. Li, X. Pan, D. Cortie, X. Huang and Z. Yi, Nat. Commun., 2016, 7, 12273.
62. H. Song, X. Meng, S. Wang, W. Zhou, X. Wang, T. Kako and J. Ye, J. Am. Chem. Soc., 2019, 141, 20507–20515.
63. J. Jang, K. Shen and C. G. Morales-Guio, Joule, 2019, 3, 2589–2593.
64. Q. Wang, M. Kan, Q. Han and G. Zheng, Small Struct., 2021, 2, 2100037.
65. S. Xie, S. Lin, Q. Zhang, Z. Tian and Y. Wang, J. Energy Chem., 2018, 27, 1629–1636.
66. M. S. A. Sher Shah, C. Oh, H. Park, Y. J. Hwang, M. Ma and J. H. Park, Adv. Sci., 2020, 7, 2001946.
67. J. Baltrusaitis, I. Jansen and J. D. Schuttlefield Christus, Catal. Sci. Technol., 2014, 4, 2397–2411.
68. T. Shi, D. Sridhar, L. Zeng and A. Chen, Electrochem. Commun., 2022, 135, 107220.
69. Y. Zhang, J. Li and N. Kornienko, Chem. Commun., 2021, 57, 4230–4238.
70. H. J. Yin, Y. H. Dou, S. Chen, Z. J. Zhu, P. R. Liu and H. J. Zhao, Adv. Mater., 2020, 32, 1904870.
71. R. F. B. de Souza, D. Z. Florio, E. Antolini and A. O. Neto, Catalysts, 2022, 12, 217.
72. D. Richard, Y.-C. Huang and C. G. Morales-Guio, Curr. Opin. Electrochem., 2021, 30, 100793.
73. A. G. Abdelkader Mohamed, S. A. Zahra Naqviab and Y. Wang, ChemCatChem, 2021, 13, 787–805.
74. A. Prajapati, B. A. Collins, J. D. Goodpaster and M. R. Singh, PNAS, 2021, 118, e2023233118.
75. Y. Kang, Z. Li, X. Lv, W. Song, Y. Wei, X. Zhang, J. Liu and Z. Zhao, J. Catal., 2021, 393, 20–29.
76. M. J. Boyd, A. A. Latimer, C. F. Dickens, A. C. Nielander, C. Hahn, J. K. Nørskov, D. C. Higgins and T. F. Jaramillo, ACS Catal., 2019, 9, 7578–7587.
77. M. Ma, B. J. Jin, P. Li, M. S. Jung, J. I. Kim, Y. Cho, S. Kim, J. H. Moon and J. H. Park, Adv. Sci., 2017, 4, 1700379.
78. Y. Song, Y. Zhao, G. Nan, W. Chen, Z. Guo, S. Li, Z. Tang, W. Wei and Y. Sun, Appl. Catal., B, 2020, 270, 118888.
79. C. Oh, J. Kim, Y. J. Hwang, M. Ma and J. H. Park, Appl. Catal., B, 2021, 283, 119653.
80. Z. Xie, M. Chen, Y. Chen, A. Guan, Q. Han and G. Zheng, J. Phys. Chem. C, 2021, 125, 13324–13330.
81. M. Ma, C. Oh, J. Kim, J. H. Moon and J. H. Park, Appl. Catal., B, 2019, 259, 118095.
82. F. Hahn and C. A. Melendres, Electrochim. Acta, 2001, 46, 3525–3534.
83. J. Nandenha, E. H. Fontes, R. M. Piasentin, F. C. Fonseca and A. O. Neto, J. Fuel Chem. Technol., 2018, 46, 1137–1145.
84. F. Amano, A. Shintani, K. Tsurui, H. Mukohara, T. Ohno and S. Takenaka, ACS Energy Lett., 2019, 4, 502–507.
85. Y. W. Phuan, W.-J. Ong, M. N. Chong and J. D. Ocon, J. Photochem. Photobiol., C, 2017, 33, 54–82.
86. L. Kumar Dhandole, H. Hwi Lee, W.-S. Chae, J. Suk Jang and J. Sung Lee, J. Energy Chem., 2022, 65, 415–423.
87. L. K. Dhandole, P. Anushkkaran, J. B. Hwang, W.-S. Chae, M. Kumar, H.-H. Lee, S. H. Choi, J. S. Jang and J. S. Lee, Renewable Energy, 2022, 189, 694–703.
88. J. H. Kim, D. Hansora, P. Sharma, J.-W. Jang and J. S. Lee, Chem. Soc. Rev., 2019, 48, 1908–1971.
89. Y.-C. Wu, R.-J. Song and J.-H. Li, Org. Chem. Front., 2020, 7, 1895–1902.
90. Y. Miao and M. Shao, Chin. J. Catal., 2022, 43, 595–610.
91. A. Mehmood, S. Y. Chae and E. D. Park, Catalysts, 2021, 11, 1387.
92. X. Cui, H. Li, Y. Wang, Y. Hu, L. Hua, H. Li, X. Han, Q. Liu, F. Yang, L. He, X. Chen, Q. Li, J. Xiao, D. Deng and X. Bao, Chem, 2018, 4, 1902–1910.
93. L. Zhang, Z. Sun, J. Lang and Y. H. Hu, Int. J. Energy Res., 2021, 45, 2581–2592.
94. C. Hammond, M. M. Forde, M. H. Ab Rahim, A. Thetford, Q. He, R. L. Jenkins, N. Dimitratos, J. A. Lopez-Sanchez, N. F. Dummer, D. M. Murphy, A. F. Carley, S. H. Taylor, D. J. Willock, E. E. Stangland, J. Kang, H. Hagen, C. J. Kiely and G. J. Hutchings, Angew. Chem., Int. Ed., 2012, 51, 5129–5133.
95. T. Yu, Z. Li, W. Jones, Y. Liu, Q. He, W. Song, P. Du, B. Yang, H. An, D. M. Farmer, C. Qiu, A. Wang, B. M. Weckhuysen, A. M. Beale and W. Luo, Chem. Sci., 2021, 12, 3152–3160.
96. Á. Szécsényi, G. Li, J. Gascon and E. A. Pidko, ACS Catal., 2018, 8, 7961–7972.
97. L. Zhang and Y. Lin, Appl. Catal., A, 2021, 615, 118052.
98. Y. Kwon, T. Y. Kim, G. Kwon, J. Yi and H. Lee, J. Am. Chem. Soc., 2017, 139, 17694–17699.
99. C. Williams, J. H. Carter, N. F. Dummer, Y. K. Chow, D. J. Morgan, S. Yacob, P. Serna, D. J. Willock, R. J. Meyer, S. H. Taylor and G. J. Hutchings, ACS Catal., 2018, 8, 2567–2576.
100. Q. Shen, C. Cao, R. Huang, L. Zhu, X. Zhou, Q. Zhang, L. Gu and W. Song, Angew. Chem., Int. Ed., 2020, 59, 1216–1219.
101. S. Bai, F. Liu, B. Huang, F. Li, H. Lin, T. Wu, M. Sun, J. Wu, Q. Shao, Y. Xu and X. Huang, Nat. Commun., 2020, 11, 954.
102. B. Wu, R. Yang, L. Shi, T. Lin, X. Yu, M. Huang, K. Gong, F. Sun, Z. Jiang, S. Li, L. Zhong and Y. Sun, Chem. Commun., 2020, 56, 14677–14680.
103. J. Chen, S. Wang, L. Peres, V. Collière, K. Philippot, P. Lecante, Y. Chen and N. Yan, Catal. Sci. Technol., 2021, 11, 3493–3500.
104. P. K. Sajith, A. Staykov, M. Yoshida, Y. Shiota and K. Yoshizawa, J. Phys. Chem. C, 2020, 124, 13231–13239.
105. Y. Yan, C. Chen, S. Zou, J. Liu, L. Xiao and J. Fan, Front. Chem., 2020, 8, 252.
106. K. Zheng, Y. Wu, Z. Hu, X. Jiao, L. Li, Y. Zhao, S. Wang, S. Zhu, W. Liu, W. Yan, Y. Sun and Y. Xie, Nano Lett., 2021, 21, 10368–10376.
107. X. Ren, Y. Wang, A. Liu, Z. Zhang, Q. Lv and B. Liu, J. Mater. Chem. A, 2020, 8, 24284–24306.
108. J. Nandenha, R. M. Piasentin, L. M. G. Silva, E. H. Fontes, A. O. Neto and R. F. B. de Souza, Ionics, 2019, 25, 5077–5082.
109. Q. Wang, T. Li, C. Yang, M. Chen, A. Guan, L. Yang, S. Li, X. Lv, Y. Wang and G. Zheng, Angew. Chem., Int. Ed., 2021, 60, 17398–17403.
110. G. V. Buxton, C. L. Greenstock, W. P. Helman and A. B. Ross, J. Phys. Chem. Ref. Data, 1988, 17, 513–886.
111. Z. Zuo, Z. Cai, Y. Katsumura, N. Chitose and Y. Muroya, Radiat. Phys. Chem., 1999, 55, 15–23.
112. K. Zhang and K. M. Parker, Environ. Sci. Technol., 2018, 52, 9579–9594.
113. J. Park, J. Joo, S. G. Kwon, Y. Jang and T. Hyeon, Angew. Chem., Int. Ed., 2007, 46, 4630–4660.
114. X.-Y. Lin, J.-Y. Li, M.-Y. Qi, Z.-R. Tang and Y.-J. Xu, Catal. Commun., 2021, 159, 106346.
115. H. Song, X. Meng, Z.-j. Wang, H. Liu and J. Ye, Joule, 2019, 3, 1606–1636.
116. Y.-Q. Su, J.-X. Liu, I. A. W. Filot, L. Zhang and E. J. M. Hensen, ACS Catal., 2018, 8, 6552–6559.

---