Efficient electrochemical methane coupling enabled by stabilized oxygen species during oxygen evolution in a solid oxide electrolyzer integrated with CO₂ electrolysis

Abstract

The electrochemical oxidative coupling of methane (EOCM), integrated with CO₂ electrolysis enabled by high-temperature electrolysis technology, represents a promising pathway for methane utilization and carbon neutrality. However, progress in methane activation remains hindered by low C₂ product selectivity and limited reaction activity, primarily due to the lack of efficient and stable catalysts and rational design strategies. A critical focus of current research is the development of catalysts capable of stabilizing reactive oxygen species to facilitate C–H bond activation and subsequent C–C bond formation. Herein, an easily fabricated composite electrode consisting of perovskite La_0.6Sr_0.4MnO_3–δ and Ce-Mn-W materials with (Ce_0.90Gd_0.10)O_1.95 as the support was developed, demonstrating efficient activate methane activation. Combined theoretical and experimental investigations reveal that the designed composite electrode stabilizes active oxygen species during the oxygen evolution reaction (OER) while exhibiting superior methane adsorption capability. This design, leveraging oxygen species engineering and interfacial synergy, significantly enhances electrochemical methane coupling efficiency, establishing a strategic framework for advancing high-performance catalyst development.

Graphical abstract

Keywords

electrochemical oxidative coupling of methane (EOCM) / solid oxide electrolysis cell (SOEC) / methane to ethylene / composite electrode / CO₂ electrolysis

1 Introduction

The production of natural gas has been surged due to breakthroughs in shale gas extraction technologies. Among various methane applications, the oxidative coupling of methane (OCM) to produce ethylene has recently emerged as a prominent research focus [1–3]. Compared to traditional olefin production processes [4–7], OCM offers a more streamlined, cost-effective, and environmentally friendly alternative. However, the advancement of OCM technology is hindered by several critical challenges [8,9]: 1) the target products, such as ethylene, are more reactive than methane and prone to over-oxidation, leading to low selectivity; 2) the use of pure oxygen as an oxidant increases costs due to air separation, while using air results in nitrogen accumulation in the recycle stream; and 3) the high-temperature reaction environment poses significant safety risks due to the potential formation of explosive methane-oxygen mixtures.

Solid oxide fuel cell (SOFC) and solid oxide electrolysis cell (SOEC) technologies offer unique advantages in addressing these limitations [10–14]. With an oxygen-ion-conducting electrolyte membrane, these systems can effectively separate methane and oxygen into the anode and cathode compartments, respectively. At the cathode, oxygen molecules are reduced to oxygen ions, which then migrate through the electrolyte membrane to the anode under an electric field. At the anode, these oxygen ions are oxidized to form reactive oxygen species that facilitate methane activation [15–18]. This configuration minimizes explosion risks by preventing direct methane-oxygen mixing and eliminates nitrogen accumulation when air is used as the oxidant. Furthermore, replacing the cathode reaction with CO₂ electrolysis offers significant potential for carbon neutrality [19,20] (Fig.1(a)). The applied potential can be tuned to control oxygen ions flux and oxidation extent, thereby enhancing the efficiency of the electrochemical oxidative coupling of methane (EOCM).

Despite these advantages, the EOCM process still suffers from low C₂ product selectivity and mediocre activity [16,21,22]. Recent research has focused on developing materials that stabilize specific oxygen species to improve the EOCM performance. Pujare et al. [23] reported that rare-earth metal oxides such as Sm₂O₃ activate methane by accommodating populations of surface O⁻ species. White [24] found that perovskite materials containing Cu³⁺ ions stabilize reactive surface O⁻ species, enhancing EOCM reaction. Otsuka et al. [25] demonstrated that LiCl/NiO surfaces can facilitate the generation of O* species, likely competing with more oxidizing oxygen species on the same active sites.

More recently, Zhu et al. [15] and Ye et al. [26] showed that the metal-oxide interfaces are capable of stabilizing the active oxygen species such as O²⁻, {\text{O}}_2^{2-} and {\text{O}}_2^{-}, which benefit both C–H bond activation and C–C bond formation. Catalysts such as redox-reversible Sr₂Fe_1.5Mo_0.5O_6–δ (SFMO) and porous single-crystalline CeO₂ monoliths have shown remarkable EOCM performance. Nevertheless, EOCM development remains constrained by the scarcity of efficient, durable catalysts and the lack of a comprehensive mechanistic understanding of the reaction pathways.

In this work, an easily fabricated composite electrode was developed for efficient EOCM coupled with CO₂ electrolysis. The electrode design meets key requirements for high electronic and ionic conductivity and effective methane activation under EOCM conditions. La_xSr_1‒xMnO₃ (LSM), a commonly used anode material, offers high conductivity and chemical stability but tends to promote overoxidation. To enhance C₂ selectivity, Ce-Mn-W materials, known for their high activity in thermal methane coupling, were incorporated into the LSM framework using Ce_0.90Gd_0.10O_1.95 (GDC) as support.

Theoretical calculations and experimental investigations revealed that the designed composite electrode stabilizes oxygen species during OER and exhibit superior methane adsorption capability, thereby enabling efficient EOCM. These findings provide a new design paradigm for EOCM catalysts, emphasizing oxygen species engineering to enable efficient methane conversion and guide the development of advanced catalytic systems.

2 Experimental

2.1 Materials synthesis and characterization

The GDC or LSM-supported CeO₂-Mn₂O₃-Na₂WO₄ catalysts were synthesized via a solution impregnation method. Specifically, 0.748 g of Na₂WO₄·2H₂O (Sigma, ACS reagent, ≥ 99.0%) was dissolved into 20 mL of deionized water. Then 5 g of either GDC ((Ce_0.90Gd_0.10)O_1.95, Sofcman, ≥ 99.5%) or LSM (La_0.6Sr_0.4MnO_3‒δ, Sofcman, ≥ 99.5%) powder was added to the solution under vigorous stirring. Separately, 1.495 g of Mn(NO₃)₂ solution (50 wt% (mass fraction) in H₂O, Macklin, ACS reagent) and 1.68 g of Ce(NO₃)₃·6H₂O (Sigma, 99.99% trace metal basis) were each dissolved in 10 mL of deionized water and then slowly added to previous mixture to form a pale-yellow suspension. The suspension was stirred continuously at room temperature for 2 h, then concentrated at 130 °C until the solvent was completely evaporated. The resulting solid was transferred to an oven at 80 °C and dried overnight. The dried material was subsequently sintered at 850 °C for 3 h in air. Finally, the solid was ground into powder and sieved to below 200 mesh. The prepared GDC- and LSM-supported CeO₂-Mn₂O₃-Na₂WO₄ catalysts were denoted as CMW@GDC and CMW@LSM, respectively. The CMW@GDC_LSM composite was obtained by mechanically mixing CMW@GDC and LSM powders at a 1:1 mass ratio using an agate mortar for 1 h.

2.2 Fabrication of planar cell

The NiO (Sigma, −325 mesh, 99%)-YSZ (yttria-stabilized zirconia, 8YS, Tosoh Inc., Tokyo, Japan) cathode substrates were prepared using a conventional tape casting method. NiO and YSZ powders were mixed in a mass ratio of 3:2 and ball-milled for 24 h in ethanol. Additives including triethanlamine (Sigma, GR grade) as a dispersant, starch (Sigma, soluble extra pure) as a pore former, polyvinyl butyral (Alfa, m.w. 90000–120000) as a binder, and dibutyl phthalate (Sigma, 99%) and polyethylene glycol (Alfa, m.w. 200) as the plasticizer were subsequently added to the slurry, followed by ball milling for an additional 2.5 h. Prior to casting, the resulting homogenized slurry was de-gassed in a vacuum vessel under stirring at a vacuum pressure of 300 mmHg for 45 min to remove air bubbles.

The NiO-YSZ slurry was then cast onto a PET substrate using a doctor blade set to a 2 mm gap, producing tapes of 15 cm × 15 cm in size. The tapes were dried at room temperature overnight and sintered at 1000 °C for 2 h to remove organic additives and achieve sufficient mechanical strength for subsequent printing. A YSZ electrolyte layer was formed by screen-printing YSZ ink—prepared by grinding YSZ powder with an organic vehicle in an agate mortar for 1 h—onto the pre-sintered NiO-YSZ substrate. This substrate was then sintered at 1450 °C for 6 h to form a dense YSZ electrolyte.

A GDC barrier layer was applied by spin-grinding GDC ink—prepared by grinding GDC powder with an organic vehicle in an agate mortar for 0.5 h—onto the pre-sintered substrate at a rotation speed of 9000 r/min. The GDC layer was then sintered at 1300 °C for 3 h to serve as a barrier layer. The half-cell substrates were cut into 2 diameter disks using a low-power laser cutter.

Catalyst inks were prepared by mixing the catalysts with an organic vehicle to form a uniform slurry. The organic vehicle was obtained by dissolving ethyl cellulose in terpineol at a concentration of 7 wt% ethyl cellulose. The catalyst inks were then printed onto the YSZ electrolyte surface over a surface area of 0.5 cm² to serve as the anode and then sintered at 1000 °C in air for 3 h. For the CMW@GDC/LSM anode, the CMW@GDC ink was brushed onto the sintered LSM surface and subjected to the same sintering process.

2.3 Electrochemical measurements and product qualification

The prepared cell was positioned between two alumina tubes and sealed gas tight using high-temperature sealing paste. Silver mesh current collectors were affixed to both electrodes using silver paste. All electrochemical measurements were conducted using a Zahner Zennium Pro potentiostat. Gases were supplied independently to the anode and cathode through the respective inlets of the alumina tubes and collected at the outlets. The gas flow rates were controlled using a mass flow controller and calibrated with an ADM flow meter (Agilent Technologies). Inlet and outlet flow rates were measured to confirm the absence of leakage prior to heating. The cell was then heated to 850 °C at a rate of 10 °C/min.

The NiO-YSZ cathode was reduced under a gas mixture of 25% H₂ balanced in Ar, while the anode side was purged with air at a flow rate of 20 mL/min for approximately 30 min. Once the open circuit voltage exceeded 1.0 V, the anode and cathode gas feeds were switched to N₂ and 25% H₂ balanced in CO₂, respectively. After at least 15 min of purging, electrochemical impedance spectroscopy (EIS) was performed. The EIS measurements were conducted over a frequency range of 100 kHz–0.1 Hz with an AC amplitude of 10 mV, after ensuring that the open circuit potential (OCP) had stabilized for at least 10 min following the change in gas atmosphere.

Subsequently, the N₂ feed on the anode side was replaced with pure methane for EIS, linear sweep voltammetry (LSV), and chronoamperometry (CA) measurements. Unless otherwise specified, the methane flow rate was maintained at 20 mL/min and monitored using an ADM flow meter. The gas at the outlet of anode was directly introduced into the sampling loop of a gas chromatograph (Agilent 8890B) for product analysis every 15 min. The gas chromatograph was equipped with a ShinCarbon ST column and a HayeSep Q column, using argon as the carrier gas. Hydrocarbons (CH₄, C₂H₄, C₂H₆, and C₃H₆) were quantified using a flame ionization detector (FID), while H₂, CO, CO₂, and O₂, were analyzed using a thermal conductivity detector (TCD).

The production rate of gaseous product i was calculated using

(1)r_i=(Q_{\text{G}}\times x_i)/S,

where Q_G is the gas flow at the outlet; x_i is the molar fraction of product i of the gas flow at the outlet determined by GC; and S is the electrode surface area.

The selectivity of C₂ products is based on carbon atoms and was calculated using

(2)\begin{array}{rl}{S}_{{\text{C}}_{{}^2}}=& (2\times r_{{\text{C}}_{\text{2}}{\text{H}}_{\text{4}}}+2\times r_{{\text{C}}_{\text{2}}{\text{H}}_6})/(2\times r_{{\text{C}}_{\text{2}}{\text{H}}_{\text{4}}}+2\times r_{{\text{C}}_{\text{2}}{\text{H}}_6}+\\ & 3r_{{\text{C}}_3{\text{H}}_6}+r_{\text{C}{\text{O}}_2}+r_{\text{CO}}).\end{array}

The conversion of CH₄ can be calculated as

(3)X_{\text{C}{\text{H}}_{\text{4}}}=\frac{(Q_{\text{C}{\text{H}}_{\text{4}},\text{inlet}}-Q_{\text{C}{\text{H}}_{\text{4}},\mathrm{o}\mathrm{u}\mathrm{t}\mathrm{l}\mathrm{e}\mathrm{t}})}{Q_{\text{C}{\text{H}}_{\text{4}},\text{inlet}}}\times 100\mathrm{\%}.

2.4 Computational details

All DFT calculations were performed using the projector augmented wave pseudopotentials provided in the Vienna ab initio simulation package (VASP) [27,28]. The generalized gradient approximation (GGA) with the spin-polarized Perdew-Wang 1991 (PW91) functional was employed for the electronic exchange-correlation potential [29]. A DFT + U approach, as proposed by Dudarev et al. [30], was applied with on-site Coulomb interaction parameter (U–J) of 3.9, 5.0, and 5.0 eV for Mn, Ce, and Gd, respectively, consistent with values reported in previous DFT + U studies [31–33]. The plane-wave cutoff energy was set to 660 eV, and Brillouin zone was sampled using the Monkhorst‒Pack scheme with a k-point spacing of 0.05 Å⁻¹ [34]. The energy convergence criteria of electronic self-consistent was set to 10⁻⁵ eV, and the force convergence criteria on each atom of geometry optimization were set to 0.02 eV/Å. Data postprocessing was conducted using the VASPKIT package from VASP [35].

The free energy, G, of each step was calculated using G = E + ZPE − TS, where E is the total DFT calculation energy, ZPE is the zero-point energy correction, and TS is entropy contribution. ZPE and TS corrections for several surface-adsorbed intermediates were calculated at T = 850 °C. The bulk LSM system was modeled by substituting 3 of every 8 La atoms with Sr atoms in a 2 × 2 × 2 supercell of LaMnO₃. A (001) surface was established, consisting of four alternating La/Sr–O and MnO₂ layers, with the bottom La/Sr–O and MnO₂ layers fixed during relaxation, to investigate oxygen migration and surface adsorption behavior [36]. The bulk 12.5% Gd-doped ceria, Ce_0.875Gd_0.125O₂ system was constructed by substituting Ce atoms with Gd in a 2 × 2 × 2 CeO₂ supercell. The (111) surface, composed of two layers of O-Ce-O, was selected for catalytic mechanism studies because the (111) surface is energetically the most stable among the low-index CeO₂ (111), (110), and (100) surfaces [37–39].

For Na₂WO₄ models supported on GDC or LSM surfaces, the WO₄ tetrahedron was anchored to the GDC (111) or LSM (001) surface through a W-O bond, based on previous structural models of supported WO_x catalysts [40–42]. Sodium atoms were introduced to maintain charge neutrality. All computational slab models include a vacuum layer of approximately 15 Å perpendicular to the surface to reduces periodic image interactions.

3 Results and discussion

The EOCM was conducted in the anode of a SOEC, which incorporates an oxygen-ion-conducting electrolyte that facilitates O²⁻ transport, reduced from oxygen-containing gases such as CO₂ in the cathode (Fig.1(a)). CH₄ was oxidized to produce C₂ products in the anode (4CH₄ + 2O²⁻→ C₂H₄ + C₂H₆ + H₂ + 2H₂O + 4e⁻), while CO₂ was reduced to CO in the cathode (2CO₂ + 4e⁻→ 2CO + 2O²⁻). The overall reaction is given by 4CH₄ + 2CO₂→ C₂H₄ + C₂H₆ + H₂ + 2H₂O + 2CO.

A porous NiO-YSZ cathode was used in this study due to its excellent electrochemical performance and high stability for CO₂ electrolysis [43,44]. The electrolyte consisted of a yttria-stabilized zirconia (YSZ) layer with a thickness of less than 20 μm, chosen for its excellent oxygen-ion conductivity [45] (Fig.1(b)).

A composite electrode was employed to meet the essential requirements of high electronic and ionic conductivity as well as effective methane activation in the EOCM process. LSM was selected as one component of the composite electrode due to its good conductivity and chemical stability at high temperatures [46,47]. In addition, the relatively moderate OER capability of LSM enhances the potential for utilizing the oxygen species generated during OER for methane activation [47,48]. The DFT calculations revealed that the OER process on LSM proceeds via the diffusion of O²⁻ from the bulk to the surface, the formation of reactive oxygen species, and the subsequent generation of gaseous oxygen (Fig. S1). These steps are all exothermic, suggesting that LSM is indeed capable of forming active oxygen species and generating gaseous oxygen. The {\mathrm{O}}_2^{2-} species were identified as the key intermediate, while the formation of single oxygen species is unfavorable during the OER process on LSM (Fig. S2).

The {\mathrm{O}}_2^{2-} species was hypothesized to be capable of activating methane when introduced through Ce-Mn-W materials, which are recognized for their high methane oxidation activity [49,50], in combination with GDC supports that offer high oxygen ion mobility and oxygen storage/release capabilities [51,52] (Fig.2(a)). GDC was selected as the support material for Ce-Mn-W not only because it matches the composition of the GDC barrier layer—thereby reducing contact resistance between the anode and electrolyte layers, but also because DFT calculation suggests that the {\text{O}}_2^{2-} species generated can be further stabilized on the GDC surface (Fig.2(b)).

The Gibbs free energy of {\text{O}}_2^{2-} absorbed on GDC is 3.08 eV lower than that on LSM, and the Gibbs free energy of {\text{O}}_2^{2-} generation is also lower than that for the formation of gaseous oxygen (Fig.2(b)). These findings suggests that GDC is a favorable support for methane activation via the {\mathrm{O}}_2^{2-}⁻ species when used as a support. In contrast, the use of SiO₂ as a support for Ce-Mn-W materials was excluded due to the risk of Si impurities migrating to the cathode and poisoning the triple-phase boundaries [53,54]. This assessment is in agreement with the result that the electrolysis was unstable when using SiO₂ as a support at the anode (Fig. S3).

The GDC-supported Ce-Mn-W catalysts were synthesized using a wet impregnation method with a nominal composition of 10%Ce-5%Mn-10%Na₂WO₄@GDC. Scanning electron microscopy (SEM) analysis revealed micron-scale particle agglomeration, while energy-dispersive X-ray spectroscopy (EDS) elemental mapping confirmed the homogeneous distribution of Ce, Mn, and W species across the GDC support (Fig.3(a)). Powder X-ray diffraction patterns exhibited peaks at the expected positions for the corresponding phases of the GDC-supported Ce-Mn-W and LSM catalysts (Fig. S4). The XPS spectra of O 1s, Mn 2p, Ce 3d, and W 4f are shown in Fig.3(b), with all binding energies referenced to the C 1s spectrum (284.8 eV) to account for the charging effects during the measurement. The peaks at 529.3 and 530.95 eV were assigned to lattice oxygen and surface oxygen species, respectively. The percentage of surface oxygen species was found to be as high as 58.3%, which is beneficial for the methane activation process [16].

The XPS spectra of Mn 2p showed peaks for both Mn⁴⁺ and Mn³⁺, indicating the coexistence of these two species on the catalyst surface. Since Mn₂O₃ was identified as the bulk phase in the XRD results, the presence of Mn⁴⁺ likely results from the interaction between Mn₂O₃ and CeO₂, facilitating electron transport. This is further supported by the observation of a small portion of Ce³⁺ in the Ce 3d spectra. The interaction between Mn³⁺ and Ce⁴⁺ is expected to facilitate the oxygen cycling in the catalyst [49]. The W 4f spectra show doublet peaks at 35.1 and 37.1 eV, corresponding to the standard tetrahedral geometry of the {\mathrm{W}\mathrm{O}}_4^{2-}structure (Fig. S5).

TEM images revealed that the composite electrode consisted of small grains, with sizes ranging from 100 to 200 nm (Fig.3(c)). The interplanar spacings of 0.32, 0.31, and 0.38 nm were attributed to the (111) planes of Na₂WO₄, (111) planes of CeO₂, and (211) planes of Mn₂O₃, respectively, highlighting their high dispersion on the catalyst.

The composite electrodes, 10%Ce-5%Mn-10%Na₂WO₄@GDC and LSM, were prepared via two methods: physical mixing and layered coating. In the physical mixing approach, the two components were homogeneously mixed by grinding and then printed onto the GDC barrier layer. In the layered coating method, the 10%Ce-5%Mn-10%Na₂WO₄@GDC and LSM components were sequentially coated onto the GDC barrier layer. These two samples are referred to as CMW@GDC_LSM (physically mixed) and CMW@GDC/LSM (layered), respectively.

The electrochemical performance of the composite electrodes was investigated at 850 °C using a Zahner Zennium Pro potentiostat. EIS was first recorded with CO₂ and H₂ at the cathode and N₂ at the anode. All impedance spectra were measured at the OCP to ensure zero electrolysis current during the tests. Since the cathode conditions for all the measurements were the same, the differences observed in the Nyquist plots can be attributed to the anode.

The ohmic impedance measured at OCP for the CMW@GDC_LSM, CMW@GDC/LSM, CMW@LSM, and LSM catalysts are 0.15, 0.16, 0.14, and 0.13 Ω cm², respectively (Fig.4(a)). The slight increase in ohmic impedance with the addition of non-conductive components to LSM indicates that the coating and mixing methods for combining the Ce-Mn-W materials with LSM were effective. The total polarization impedance, which includes CO₂ electroreduction on the cathode and OER on the anode, is around 2.25, 2.4, 1.5, and 0.75 Ω·cm² for CMW@GDC_LSM, CMW@GDC/LSM, CMW@LSM, and LSM, respectively. The differences in polarization impedance are likely attributed to the reaction occurring at the anode, particularly the OER. The increased polarization impedance in the Ce-Mn-W containing catalysts is likely due to their poor OER activity.

Interestingly, all Nyquist plots show a mass transfer control region, indicating one of the reactions at either the cathode or anode was confined by the mass transport of reactants or products. When the input gas was switched to pure methane, the ohmic impedance of all the catalysts remained nearly unchanged, while the polarization impedance significantly decreased. This reduction in polarization impedance after the introduction of CH₄ is likely due to the assisting role of methane in CO₂ electrolysis in SOECs, as evidenced by previous studies [55–57], when current density increased at a certain applied potential with elevated CH₄ concentration [58]. This observation is further supported by the LSV polarization curves recorded with or without CH₄ in the anode (Fig. S6).

Surprisingly, the mass transport control characteristic at CMW@GDC_LSM and CMW@GDC/LSM anodes disappeared, while it remained at the CMW@LSM and LSM catalysts. The disappearance of the mass transfer control region when switching the gas atmosphere in the anode suggested that this phenomenon was ascribed to the anode, specifically the diffusion of O₂ generated during OER. Hence, the absence of mass transfer control at CMW@GDC_LSM and CMW@GDC/LSM is likely because the oxygen species generated during the OER process were consumed by CH₄, benefiting from the presence of Ce-Mn-W materials that favor CH₄ activation. Notably, the LSM-supported Ce-Mn-W catalyst still showed O₂ diffusion control, highlighting the important role of GDC in CH₄ activation using oxygen species generated during OER.

The LSV results also demonstrated the superior activities of the CMW@GDC_LSM and CMW@GDC/LSM catalysts, achieving current densities of around 700 and 600 mA/cm at an applied potential of 2 V, respectively (Fig.4(b)). In contrast, the current densities of LSM and CMW@LSM anodes are only 400 and 500 mA/cm², respectively. It is important to note that the onset potential for CH₄-assited electrolysis in this study is higher than that in previous studies [57,59], primarily due to the higher OCP and the sluggish kinetics of EOCM process. The higher OCP value in this study can be attributed to the high 25% H₂ concentration introduced at the cathode, whereas H₂O was used in most CH₄-assisted electrolysis studies [57,59]. Replacing H₂ with H₂O significantly decreases the oxygen partial pressure at the cathode, thereby increasing the OCP [60]. In addition, the sluggish kinetics of EOCM, compared to the faster oxidation of CH₄ to CO_x in traditional CH₄-assisted electrolysis, also contributes to the higher OCP value. This is because the theoretical Nernst potential is calculated under the assumption of sufficient reaction rates on the electrode surface [59].

The performance of EOCM on different catalysts was studied using CA methods at 850 °C. Pure methane was fed into the anode at a flow rate of 20 mL/min, while a gas mixture of CO₂ and H₂ (molar ratio 3:1) was supplied to the cathode at the same total flow rate. The gas products at the outlet were analyzed using gas chromatography (GC) equipped with both a TCD and an FID. In the absence of an external applied potential, nearly no products were detected, indicating that the thermal splitting of methane is unlikely at this temperature (Fig. S7). However, after applying a constant current, gas products including C₂H₄, C₂H₆, C₃H₆, CO₂, CO, H₂, and O₂ were detected at the anode (Fig. S8), with CO also being detected at the cathode (Fig. S9). The total production rate increased with increasing applied current (Fig.4(c)), confirming the successful activation of EOCM on these catalysts.

At a current density of 100 mA/cm², the production rates of C₂ products (C₂H₄ and C₂H₆) on CMW@GDC_LSM and CMW@GDC/LSM catalysts were 135 and 152 μL·min⁻¹·cm⁻², respectively, significantly higher than that on LSM and CMW@LSM catalysts, which were only 77 and 85 μL·min−1·cm−2 , respectively, highlighting the outstanding CH₄ activation performance of the composite electrodes. The production rates of C₂ products further increased at elevated current densities, reaching 313 and 268 μL·min⁻¹·cm⁻² on the CMW@GDC_LSM and CMW@GDC/LSM catalysts at a current density of 600 mA/cm². However, the selectivity of C₂ products decreased with increasing current densities. The highest selectivity of 80.3% was achieved at 100 mA/cm² on the CMW@GDC/LSM catalyst, but it decreased to 52.89% at 600 mA/cm² (Fig.4(d)). The reduced selectivity at higher current densities is likely due to an increased oxygen flux, leading to a higher ratio of oxygen species to methane, which favors overoxidation of CH₄ [16,17].

The highest C₂ products production rate and selectivity on the composite anode are among the best reported in the literature (see Table S1 in Supplementary Material). Future work could focus on increasing the specific surface area of the composite anode to decrease the local oxygen concentration at each contact point along the methane path, thus lowering the tendency for deep oxidation [61]. On the other hand, the selectivity of C₂ products on the LSM and CMW@LSM catalyst was much lower, with values of only 43.6% and 49.7%, respectively, at a current density of 100 mA/cm², further demonstrating the superior performance of the composite electrodes.

When electrolysis was directly conducted on Ce-Mn-W or GDC mixed LSM anodes, the C₂ production rates were only 54 and 92 μL·min⁻¹·cm⁻², with C₂ product selectivity of 38% and 44% respectively, at a current density of 200 mA/cm² (Fig. S10). These results suggest that the components in the composite electrode play complementary roles and work synergistically during the EOCM process. Replacing GDC with other supports such as TiO₂ or TS-1, ZSM zeolite led to decreased production rates and selectivity for C₂ products (Fig. S11). The absence of GDC or substituting it with other supports diminished the catalyst performance, indicating the crucial role of GDC in electrochemical oxidative of methane. This is likely due to the synergy between GDC, Ce-Mn-W, and LSM components. These mechanistic aspects will be discussed in the following sections.

No carbon deposition was observed on either the anode or cathode surfaces after electrolysis, as evidenced by Raman spectra in Fig. S12. The potential issue of Ni oxidation can be ruled out since the OCP could reach values above 1.0 V when the gas was switched to air on the anode and 25% H₂ in Ar on the cathode. The CMW@GDC/LSM anode exhibited stable electrolysis with C₂ selectivity of around 70% at a current density of 200 mA/cm² over 5500 s with no deactivation (Fig. S13). This highlights the good stability of the composite electrode and its potential for future industrial applications.

To further explore the performance, the EOCM was investigated at various partial pressures methane on the CMW@GDC/LSM catalyst due to its superior performance. Inert Ar gas was introduced to maintain a constant total gas flow rate while adjusting the partial pressure of methane. The dependence of the reaction rate on methane partial pressure at a fixed current density of 200 mA/cm² resembled a Langmuir chemisorption model (Fig.5(a)). The chemisorption behavior of methane is commonly observed in OCM on Mn-Na₂WO₄@SiO₂ catalysts [57,58], in which CH₄ first adsorbs on the catalyst surface and then reacts with either gaseous oxygen or surface oxygen species. Interestingly, the production rate of C₂ products showed a linear correlation with CH₄ partial pressure, which was attributed to the increased C₂ products selectivity at high CH₄ concentrations.

The conversion of methane decreased with increasing CH₄ partial pressure, as the reaction order of methane is lower than 1 (Fig.5(b)). The influence of residence time was also studied by varying the total gas flow rates. Reducing the gas flow rate led to higher conversions, although it decreased the selectivity of C₂ products (Fig.5(c)). This can be attributed to longer residence times, which results in further oxidation of C₂ products. Moreover, higher residence times favor a higher C₂H₄ to C₂H₆ ratio. The C₂H₄ to C₂H₆ ratio increased from 0.6 at a 20 mL/min gas flow rate to 1.7 at a 5 mL/min gas flow rate, while the selectivity of C₂ products decreased from 72.1% to 46.6%. Finally, increasing the methane gas flow rate led to higher production rates of C₂ products (Fig.5(d)), as the methane conversion rates increased slightly with larger methane input at higher gas flow rates (Fig. S14).

To investigate the origin of the high performance of the composite electrodes, DFT calculations were conducted to explore the process of methane activation on different catalysts surfaces. Given that Na₂WO₄ is considered the actual active site for methane coupling, while the Ce- and Mn- components primarily serve as promoters [49,62], Na₂WO₄ was treated as the active site in the calculation model, consistent with previous studies [63,64]. The first C–H bond cleavage of methane, which is widely recognized as the rate-determining step of methane activation, was investigated on the Na₂WO₄@GDC, LSM and Na₂WO₄@LSM surfaces using the {\mathrm{O}}_2^{2-} species (Fig.6).

Methane was readily adsorbed on Na₂WO₄ with a reaction free energy of −0.49 eV, confirming that Na₂WO₄ served as the active sites for methane activation. The adsorbed methane then reacted with the surface {\mathrm{O}}_2^{2-} species to form an O–H bond and a methyl species, with a reaction free energy of −0.61 eV, indicating that the first C–H bond cleavage using {\mathrm{O}}_2^{2-} on Na₂WO₄@GDC is favorable. This result aligns with the experimental observations. In contrast, Na₂WO₄@GDC is unable to activate methane without the presence of {\mathrm{O}}_2^{2-} species, as the reaction free energy for this step is positive (Fig. S15). This result corresponds to the fact that no products were detected without an applied potential.

Although methane adsorption on the Na₂WO₄@LSM surface is also favorable, the reaction free energy for the next step of C–H bond cleavage is uphill, suggesting that this catalyst is not suitable for methane activation. In addition to the formation of gaseous oxygen, the {\mathrm{O}}_2^{2-} species on the LSM surface could also activate methane. However, the adsorption of methane on LSM is unfavorable, meaning that methane likely occurs in the gas phase rather than on the surface. This behavior would promote the overoxidation of methane to form CO_x [12,65,66], which is consistent with the observed low selectivity for C₂ products in experiment.

Overall, the DFT calculation results suggest that the ability of the composite electrodes to stabilize oxygen species during OER and the strong adsorption of methane of the designed composite electrodes facilitate the efficient conversion of methane. These findings, which are in agree with the experimental results, highlight an effective strategy for designing advanced catalysts and reaction schemes to achieve efficient electrochemical methane coupling.

4 Conclusions

In this study, a composite electrode comprising perovskite LSM and Ce-Mn-W materials supported by GDC was designed, significantly enhancing the reactivity and C₂ product selectivity of the EOCM process through oxygen species engineering and interfacial synergy. Compared to the traditional LSM anode material, the incorporation of Ce-Mn-W and GDC stabilizes oxygen species during the OER and enhances methane adsorption, thereby promoting the surface reactions involved in methane oxidation, which is beneficial for C₂ products formation.

Electrochemical evaluations demonstrated that the CMW@GDC/LSM composite electrode achieves a remarkable C₂ production rate of 626 μL/min^/cm² and a selectivity of 80.3% at 850 °C. The influence of partial pressure and the flow rate of methane on reaction rate was also studied, providing deeper insights into the underlying mechanisms of EOCM. DFT theoretical investigations elucidate the critical role of the composite electrode in stabilizing reactive oxygen species {\mathrm{O}}_2^{2-} during the OER and strengthening methane adsorption.

While methane conversion in this system remains moderate, effective strategies to overcome this limitation include the development of a large-scale stack cells with larger catalyst specific areas. This work not only establishes a paradigm for designing high-activity and durable EOCM catalysts, but also advances the foundational framework for SOEC applications in low-carbon energy conversion, especially when incorporating CO₂ electrolysis at the cathode to contribute to carbon neutrality.

References

[1] Schwach P, Pan X, Bao X. Direct conversion of methane to value-added chemicals over heterogeneous catalysts: challenges and prospects. Chemical Reviews, 2017, 117(13): 8497–8520

[2] Arinaga A M, Ziegelski M C, Marks T J. Alternative oxidants for the catalytic oxidative coupling of methane. Angewandte Chemie International Edition, 2021, 60(19): 10502–10515

[3] Gambo Y, Jalil A A, Triwahyono S. . Recent advances and future prospect in catalysts for oxidative coupling of methane to ethylene: A review. Journal of Industrial and Engineering Chemistry, 2018, 59: 218–229

[4] Zhong J, Han J, Wei Y. . Catalysts and shape selective catalysis in the methanol-to-olefin (MTO) reaction. Journal of Catalysis, 2021, 396: 23–31

[5] Yang M, Fan D, Wei Y. . Recent progress in methanol-to-olefins (MTO) catalysts. Advanced Materials, 2019, 31(50): 1902181

[6] Wang S, Wang P, Shi D. . Direct conversion of syngas into light olefins with low CO₂ emission. ACS Catalysis, 2020, 10(3): 2046–2059

[7] Jiao F, Li J, Pan X. . Selective conversion of syngas to light olefins. Science, 2016, 351(6277): 1065–1068

[8] Liu J, Yue J, Lv M. . From fundamentals to chemical engineering on oxidative coupling of methane for ethylene production: A review. Carbon Resources Conversion, 2022, 5(1): 1–14

[9] Ortiz-Bravo C A, Chagas C A, Toniolo F S. Oxidative coupling of methane (OCM): An overview of the challenges and opportunities for developing new technologies. Journal of Natural Gas Science and Engineering, 2021, 96: 104254

[10] Thyssen V V, Vilela V B, de Florio D Z. . Direct conversion of methane to C₂ hydrocarbons in solid-state membrane reactors at high temperatures. Chemical Reviews, 2022, 122(3): 3966–3995

[11] Otsuka K, Suga K, Yamanaka I. Oxidative coupling of methane applying a solid oxide fuel cell system. Catalysis Today, 1990, 6(4): 587–592

[12] Appamana W, Khammona K, Wiyaratn W. . Oxidative coupling of methane over YSZ support catalysts for application in C₂ hydrocarbon production. Engineering Journal, 2015, 19: 1–11

[13] Li F, Li Y, Chen H. . Impact of strain-induced changes in defect chemistry on catalytic activity of Nd₂NiO_4+δ electrodes. ACS Applied Materials & Interfaces, 2018, 10(43): 36926–36932

[14] Fan L, Luo W, Fan Q. . Status and outlook of solid electrolyte membrane reactors for energy, chemical, and environmental applications. Chemical Science, 2025, 16(16): 6620–6687

[15] Zhu C, Hou S, Hu X. . Electrochemical conversion of methane to ethylene in a solid oxide electrolyzer. Nature Communications, 2019, 10(1): 1173

[16] Xu X, Li H, Han X. . Enhancing electrochemical methane coupling in solid oxide cells by tuning oxygen species in the catalyst. Journal of Materials Chemistry. A, Materials for Energy and Sustainability, 2024, 12(9): 5115–5123

[17] CaravacaA, de Lucas-Consuegra A, González-CobosJ, et al. Simultaneous production of H₂ and C₂ hydrocarbons by gas phase electrocatalysis. Applied Catalysis B: Environmental, 2012, 113–114: 192–200

[18] Zheng Y, Zhao C, Wu T. . Enhanced oxygen reduction kinetics by a porous heterostructured cathode for intermediate temperature solid oxide fuel cells. Energy and AI, 2020, 2: 100027

[19] Liu D, Shang H, Zhou C. . Active Cu and Fe nanoparticles codecorated Ruddlesden–Popper-type perovskite as solid oxide electrolysis cells cathode for CO₂ splitting. Energy & Environmental Materials, 2024, 7(5): e12717

[20] Wang W, Li H, Park K Y. . Enhancing direct electrochemical CO₂ electrolysis by introducing A-site deficiency for the dual-phase Pr(Ca)Fe(Ni)O_3−δ cathode. Energy & Environmental Materials, 2024, 7(5): e12715

[21] Denoyer L H, Benavidez A, Garzon F H. . Highly stable doped barium niobate based electrocatalysts for effective electrochemical coupling of methane to ethylene. Advanced Materials Interfaces, 2022, 9(27): 2200796

[22] Kodama S, Kikuchi R, Fujiwara N. . Oxidative coupling of methane in solid oxide electrolysis cell. ECS Transactions, 2019, 91(1): 2697–2705

[23] Pujare N U, Sammells A F. Methane activation to C₂ hydrocarbon species in solid oxide fuel cell. Journal of the Electrochemical Society, 1988, 135(10): 2544–2545

[24] WhiteJ. The electrochemical oxidative dimerization of methane. Solid State Ionics, 1992, 53–56: 149–161

[25] Otsuka K, Suga K, Yamanaka I. Electrochemical enhancement of oxidative coupling of methane over LiCl-doped NiO using stabilized zirconia electrolyte. Catalysis Letters, 1988, 1(12): 423–428

[26] Ye L, Shang Z, Xie K. Selective oxidative coupling of methane to ethylene in a solid oxide electrolyser based on porous single-crystalline CeO₂ monoliths. Angewandte Chemie International Edition, 2022, 61(32): e202207211

[27] Kresse G, Furthmuller J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Physical Review B: Condensed Matter, 1996, 54(16): 11169–11186

[28] Kresse G, Furthmüller J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Computational Materials Science, 1996, 6(1): 15–50

[29] Perdew J P, Burke K, Ernzerhof M. Generalized gradient approximation made simple. Physical Review Letters, 1996, 77(18): 3865–3868

[30] Dudarev S L, Botton G A, Savrasov S Y. . Electron-energy-loss spectra and the structural stability of nickel oxide: An LSDA+U study. Physical Review B: Condensed Matter, 1998, 57(3): 1505–1509

[31] Chen H T, Raghunath P, Lin M C. Computational investigation of O₂ reduction and diffusion on 25% Sr-doped LaMnO₃ cathodes in solid oxide fuel cells. Langmuir, 2011, 27(11): 6787–6793

[32] Polfus J M, Yildiz B, Tuller H L. Origin of fast oxide ion diffusion along grain boundaries in Sr-doped LaMnO₃. Physical Chemistry Chemical Physics, 2018, 20(28): 19142–19150

[33] Grimme S, Antony J, Ehrlich S. . A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. Journal of Chemical Physics, 2010, 132(15): 154104

[34] Monkhorst H J, Pack J D. Special points for Brillouin-zone integrations. Physical Review. B, Solid State, 1976, 13(12): 5188–5192

[35] Wang V, Xu N, Liu J C. . VASPKIT: A user-friendly interface facilitating high-throughput computing and analysis using VASP code. Computer Physics Communications, 2021, 267: 108033

[36] Piskunov S, Jacob T, Spohr E. Oxygen adsorption atLa_1−xSr_xMnO₃(001) surfaces: Predictions from first principles. Physical Review B: Condensed Matter and Materials Physics, 2011, 83(7): 073402

[37] Yang Z, Woo T K, Baudin M. . Atomic and electronic structure of unreduced and reduced CeO₂ surfaces: A first-principles study. Journal of Chemical Physics, 2004, 120(16): 7741–7749

[38] Lyons D M, Ryan K M, Morris M A. Preparation of ordered mesoporous ceria with enhanced thermal stability. Journal of Materials Chemistry, 2002, 12(4): 1207–1212

[39] Lyons D M, McGrath J P, Morris M A. Surface studies of ceria and mesoporous ceria powders by solid-state 1H MAS NMR. Journal of Physical Chemistry B, 2003, 107(19): 4607–4617

[40] Kiani D, Sourav S, Wachs I E. . A combined computational and experimental study of methane activation during oxidative coupling of methane (OCM) by surface metal oxide catalysts. Chemical Science, 2021, 12(42): 14143–14158

[41] Kiani D, Sourav S, Taifan W. . Existence and properties of isolated catalytic sites on the surface of β-cristobalite-supported, doped tungsten oxide catalysts (WO_x/β-SiO₂, Na-WO_x/β-SiO₂, Mn-WO_x/β-SiO₂) for oxidative coupling of methane (OCM): A combined periodic DFT and experimental study. ACS Catalysis, 2020, 10(8): 4580–4592

[42] Zou S, Li Z, Zhou Q. . Surface coupling of methyl radicals for efficient low-temperature oxidative coupling of methane. Chinese Journal of Catalysis, 2021, 42(7): 1117–1125

[43] Song Y, Zhou Z, Zhang X. . Pure CO₂ electrolysis over an Ni/YSZ cathode in a solid oxide electrolysis cell. Journal of Materials Chemistry. A, Materials for Energy and Sustainability, 2018, 6(28): 13661–13667

[44] Chen M, Liu Y L, Bentzen J J. . Microstructural degradation of Ni/YSZ electrodes in solid oxide electrolysis cells under high current. Journal of the Electrochemical Society, 2013, 160(8): F883–F891

[45] Yu B, Zhang W, Xu J. . Preparation and electrochemical behavior of dense YSZ film for SOEC. International Journal of Hydrogen Energy, 2012, 37(17): 12074–12080

[46] Carda M, Budáč D, Paidar M. . Current trends in the description of lanthanum strontium manganite oxygen electrode reaction mechanism in a high-temperature solid oxide cell. Current Opinion in Electrochemistry, 2022, 31: 100852

[47] Lin Q, Bian L, Liu C. . Improved La_0.8Sr_0.2MnO_3‒δ oxygen electrode activity by introducing high oxygen ion conductor oxide for solid oxide steam electrolysis. International Journal of Hydrogen Energy, 2024, 49: 616–624

[48] Zhao Z, Wang X, Tang S. . High-performance oxygen electrode Ce_0.9Co_0.1O_2‒δ-LSM-YSZ for hydrogen production by solid oxide electrolysis cells. International Journal of Hydrogen Energy, 2021, 46(50): 25332–25340

[49] Song J, Ren Y, Gao X. . Ce-Driven Ce-MnO_x/Na₂WO₄/SiO₂ composite catalysts for low-temperature oxidative coupling of methane. ACS Catalysis, 2024, 14(7): 5116–5131

[50] Gholipour Z, Malekzadeh A, Hatami R. . Oxidative coupling of methane over (Na₂WO₄ + Mn or Ce)/SiO₂ catalysts: In situ measurement of electrical conductivity. Journal of Natural Gas Chemistry, 2010, 19(1): 35–42

[51] Chen K, Ai N, Jiang S P. Performance and structural stability of Gd_0.2Ce_0.8O_1.9 infiltrated La_0.8Sr_0.2MnO₃ nano-structured oxygen electrodes of solid oxide electrolysis cells. International Journal of Hydrogen Energy, 2014, 39(20): 10349–10358

[52] Kim H Y, Lee S W, Lee S H. . Accelerating the electrochemical performance of solid oxide fuel cells using a Ce(Gd, Bi, Yb)O_2−δ diffusion barrier layer acting as an oxygen reservoir at high-current loading conditions. Journal of Materials Chemistry. A, Materials for Energy and Sustainability, 2025, 13(5): 3474–3483

[53] Tao Y, Shao J, Cheng S. Electrochemically scavenging the silica impurities at the Ni-YSZ triple phase boundary of solid oxide cells. ACS Applied Materials & Interfaces, 2016, 8(27): 17023–17027

[54] Jensen K. Effect of impurities on structural and electrochemical properties of the Ni–YSZ interface. Solid State Ionics, 2003, 160(1-2): 27–37

[55] Liu F, Chen Z, Zhou H. . Highly efficient CH₄-assisted CO₂ electrolysis for syngas production in a quasi-symmetric Ni-ceramic electrolyzer. Journal of Power Sources, 2024, 609: 234703

[56] Lu J, Zhu C, Pan C. . Highly efficient electrochemical reforming of CH₄/CO₂ in a solid oxide electrolyser. Science Advances, 2018, 4(3): 5100

[57] Cui C, Wang Y, Tong Y. . Syngas production through CH₄-assisted co-electrolysis of H₂O and CO₂ in La_0.8Sr_0.2Cr_0.5Fe_0.5O_3‒δ-Zr_0.84Y_0.16O_2‒δ electrode-supported solid oxide electrolysis cells. International Journal of Hydrogen Energy, 2021, 46(39): 20305–20312

[58] Xu H, Chen B, Irvine J. . Modeling of CH₄-assisted SOEC for H₂O/CO₂ co-electrolysis. International Journal of Hydrogen Energy, 2016, 41(47): 21839–21849

[59] Wang Y, Liu T, Lei L. . Methane assisted solid oxide co-electrolysis process for syngas production. Journal of Power Sources, 2017, 344: 119–127

[60] Stempien J P, Liu Q, Ni M. . Physical principles for the calculation of equilibrium potential for co-electrolysis of steam and carbon dioxide in a Solid Oxide Electrolyzer Cell (SOEC). Electrochimica Acta, 2014, 147: 490–497

[61] Liu K, Zhao J, Zhu D. . Oxidative coupling of methane in solid oxide fuel cell tubular membrane reactor with high ethylene yield. Catalysis Communications, 2017, 96: 23–27

[62] Jodaian V, Mirzaei M. Ce-promoted Na₂WO₄/TiO₂ catalysts for the oxidative coupling of methane. Inorganic Chemistry Communications, 2019, 100: 97–100

[63] Gu S, Kang J, Lee T. . Na₂WO₄/Mn supported on all-silica delaminated zeolite for the optimal oxidative coupling of methane via the effective stabilization of tetrahedral WO₄: Elucidating effects of support precursors with different crystal structures, Al-addition, and morphologies. Chemical Engineering Journal, 2023, 457: 141057

[64] Werny M J, Wang Y, Girgsdies F. . Fluctuating storage of the active phase in a Mn-Na₂WO₄/SiO₂ catalyst for the oxidative coupling of methane. Angewandte Chemie International Edition, 2020, 59(35): 14921–14926

[65] TagawaT, Kyaw Moe K, HiramatsuT, et al. Design of electrode for solid oxide fuel cells reactor. Solid State Ionics, 1998, 106(3‒4): 227–235

[66] Ramaiyan K P, Denoyer L H, Benavidez A. . Selective electrochemical oxidative coupling of methane mediated by Sr₂Fe_1.5Mo_0.5O_6‒δ and its chemical stability. Communications Chemistry, 2021, 4(1): 139

RIGHTS & PERMISSIONS

Higher Education Press 2025