Unlocking high alcohol product selectivity in methane-to-ethanol conversion at practically relevant current density via dual-site-driven cascade electrocatalysis

Abstract

Electrochemical conversion of methane (CH₄) is a sustainable route for converting greenhouse gases into valuable liquid fuels and chemicals. However, achieving high-yield products at industrially relevant current densities remains a formidable challenge. Here, we report a machine learning-guided Mo–Cu dual-site cascade catalytic strategy, enabling selective modulation of key *CH₃O and achieving ethanol (EtOH) electrosynthesis. This system delivers a current density of 103 mA cm⁻² with an EtOH faradaic efficiency of 55.8% ± 0.2%, establishing new performance benchmarks. Mechanistic and DFT analyses reveal that CH₄ is activated by a three-electron *O₂⁻-mediated oxidation pathway, while *CH₃ spillover from Mo to Mo–Cu active sites facilitates exothermic C–C coupling, leading to high-efficiency EtOH production. Techno-economic analysis suggests that integrating renewable electricity can lower the CH₄-to-EtOH production cost from $2.12 per kg to $1.50 per kg within a decade, offering a 53% energy return. This work establishes a cascade-regulated, dual-site framework for efficient CH₄-to-EtOH conversion and offers a framework for machine learning-assisted catalyst design, contributing to cleaner energy technologies and substantial reductions in greenhouse gas emissions.

---

Broader context

Methane electro-upgrading is a sustainable strategy to valorize the most abundant hydrocarbon while mitigating greenhouse gas emissions. However, the direct and selective conversion of CH₄ to ethanol (EtOH) remains a great scientific challenge. Two central scientific questions have hindered progress: (i) how to efficiently activate the strong C–H bond of CH₄ (439 kJ mol⁻¹) under mild electrochemical conditions and (ii) how to stabilize reactive C₁ intermediates (e.g., *CH₃ and *CH₃O) to enable controlled C–C coupling rather than side reactions such as methanol formation and complete oxidation. Here, we report a machine learning-guided Mo–Cu dual-site cascade strategy that selectively modulates the key *CH₃O intermediate, achieving EtOH electrosynthesis with a faradaic efficiency of 55.8% ± 0.2% at a current density of 103 mA cm⁻²—establishing a new benchmark in CH₄ electroconversion. Mechanistic investigations supported by DFT calculations reveal that CH₄ is activated via a three-electron *O₂⁻-mediated pathway, while *CH₃ spillover from Mo to Cu sites facilitates exothermic C–C coupling toward EtOH. Techno-economic analysis further suggests that integrating renewable electricity can reduce the production cost from $2.12 per kg to $1.50 per kg within a decade, delivering a 53% energy return.

Introduction

Global demand for ethanol (EtOH) is expected to double by 2030, driven by its growing use in automotive and industrial sectors, in line with the broader push for sustainability.^1–3 As a critical renewable resource, EtOH plays a crucial role in advancing the transition to cleaner energy and eco-friendly chemicals. However, conventional EtOH synthesis methods via petrochemical or biomass-derived fermentation processes are energy-intensive and carbon-emissive, underscoring the need for alternative, low-carbon production strategies.⁴ Consequently, prioritizing cleaner and more efficient pathways for sustainable EtOH production is imperative. Methane (CH₄), a potent greenhouse gas, is an attractive feedstock to be readily converted into higher-value products such as EtOH, offering an opportunity for simultaneously mitigating CH₄ emissions and creating a valuable commodity.^5–7

The electrochemical direct partial oxidation of CH₄-to-EtOH (CH₄OR) under ambient conditions is a sustainable and promising alternative to traditional syngas refining and biomass fermentation processes (Fig. 1a and b). However, the dominance of competitive oxidation pathways leading to CO₂ or side reactions presents challenges in electrochemical CH₄-to-EtOH conversion.^8,9 Recent efforts have explored direct CH₄ electrooxidation as a potential route to oxygenates, but achieving high EtOH selectivity via efficient C–C bond coupling remains elusive, largely due to uncontrolled radical pathways and catalyst limitations.^10,11

Fig. 1 CH₄-to-EtOH conversion pathways. (a) and (b) Conventional syngas refining processes and biomass fermentation to produce EtOH. (c) Depiction of electrochemical CH₄-to-EtOH conversion driven by renewable energy sources with a focus on materials regulation.

A critical bottleneck in electrochemical CH₄-to-EtOH conversion is the modulation of the *CH₃O intermediate,^12–16 which is central to optimizing product selectivity and overall energy efficiency. While transition metal-based catalysts have shown potential in selectively electro-synthesizing EtOH via CH₄OR, advanced modification strategies, such as the strong metal–support interaction (SMSI),^17,18 morphology engineering,^19,20 electronic structure regulation,^21,22 and electrode/electrolyte interface optimization,^23,24 have significantly improved the faradaic efficiency (FE) up to 68%. However, currently reported design methods often rely on empirical strategies, which lack the precision needed to optimize multiple interconnected catalytic properties simultaneously. Moreover, they are constrained by low production rates (<4.67 mmol g_cat⁻¹ h⁻¹) and small operational current densities (<20 mA cm⁻²) in flow cell devices,¹⁶ largely hindering their industrial scalability.

To address these challenges, the machine learning method has emerged as a transformative tool in catalyst design, enabling the prediction of structure–activity relationships and accelerating the discovery of high-performance catalysts.²⁵ By employing data-driven algorithms to explore compositional spaces and identify key catalytic descriptors, machine learning offers a systematic approach to unravel the complexity of multi-site catalysis. It is particularly effective in dual-site-driven systems, where it predicts optimal active site configurations and fine-tunes cooperative interactions to enhance intermediate generation, migration, and conversion pathways.^25,26 Recently, Chen et al. demonstrated that dual-site cooperative strategies effectively activated CH₄ and initiated reaction cascades. However, uncontrolled intermediate migration often leads to unforeseeable mixed products,^19,20,27 underscoring the need for precise control over the catalytic behavior of *CH₃O in cascade systems. Specifically, the generation, migration, and transfer of key *CH₃O remain major bottlenecks, which largely limit EtOH selectivity and yield.

Herein, we report a machine learning-assisted dual-site-driven cascade strategy for CH₄-to-EtOH conversion under ambient conditions. Guided by structure–activity relationship predictions, a Mo-doped Cu₂Se catalyst stabilized by cetyltrimethylammonium chloride (Mo–Cu-CTAC) was identified as the optimal candidate. It achieves an optimal balance between *CH₃O stabilization and migration through its dual-site interface. This cascade system achieved an EtOH FE of 55.8% ± 0.2% at a high current density of 103 mA cm⁻² and a formation rate of 529.1 µmol h⁻¹ cm⁻², exhibiting record-breaking nearly 100% carbon selectivity toward the alcohol product in a gas-fed flow cell (100% alcohol product selectivity refers to all detectable carbon-containing products in the liquid phase, e.g., EtOH, with no detectable gas-phase carbon products). Mechanistic investigations combining density functional theory (DFT) calculations, in situ spectroscopy, isotope labeling, and kinetic isotope effect analysis revealed that CH₄ activation occurs via a three-electron *O₂⁻-mediated oxidation pathway, followed by *CH₃O migration from Mo to Mo–Cu, enabling favorable C–C coupling. The CTAC-induced microenvironment enhanced CH₄ enrichment and suppressed O₂ accumulation, promoting EtOH selectivity. Techno-economic analysis forecasted that integrating renewable energy reduces the cost of CH₄-to-EtOH conversion to $1.50 per kg within a decade, highlighting the dual advantages of economic feasibility and environmental sustainability.

Results and discussion

Dual-site catalyst design for CH₄-to-EtOH conversion

To address the inherent challenges in the CH₄-to-EtOH process consisting of CH₄ activation, *CH₃O migration and C–C coupling, a dual-site-driven cascade catalysis framework is proposed in Fig. 1c, which integrates advanced machine learning and DFT calculations to systematically optimize catalytic performance. Cu₂Se was selected as the model substrate in this study due to its unique electronic structure and adjustable coordination environment, which can provide a robust platform for tailoring bimetallic interactions and facilitating dual-site catalysis behavior.²⁸ This design leverages the synergy between auxiliary and central sites to enhance CH₄ activation, control *CH₃O migration, and facilitate C–C coupling, thereby improving EtOH selectivity.

As depicted in Fig. 2, the proposed framework employs a two-step machine learning assisted screening strategy to identify the optimal dual-site configuration for CH₄-to-EtOH conversion. A computational database was established correlating *CH₃O diffusion barriers and C–C coupling energies across over 300 M-Cu bimetallic catalysts (M = transition metals), as shown in Fig. 2a, revealing quantitative relationships between the local structure and catalytic behaviour. In Fig. 2b, the sure independence screening and sparsifying operator (SISSO) algorithm was applied to extract key descriptors, including d-orbital occupancy, electronegativity differences, and coordination environment, that govern CH₄ activation and *CH₃O generation. These interpretable descriptors retain the physical essence of catalytic mechanisms while reducing computational complexity. The electronic structure analysis (Fig. 2c) shows that the Mo–Cu pair optimally balances oxophilicity and electronic coupling, a configuration critical for highly selective EtOH production: Mo activates CH₄, while Cu promotes *CH₃O transfer and C–C coupling.

Fig. 2 Rational design framework for dual-site catalysts to enhance CH₄-to-EtOH conversion. (a) The energetics of CH₄ activation and C–C coupling across different bimetallic systems. (b) Machine learning-assisted model training identifying central metals with optimal electronic and structural properties for catalytic performance. (c) A physically clarification of the structure–activity relationship via descriptor ϕ^0.5 combined with [small chi, Greek, macron]^−1.5.

Among the screened catalysts, the overall structure–activity relationship exhibits a Mo–Cu dual-site cascade system which emerged as the most promising candidate owing to its balanced electronic coupling, enabling energetically favorable *CH₃O migration (−0.93 eV) from Mo to Cu sites, as well as its optimal C–C coupling energetics. This approach underscores the critical role of electronic and structural tuning in achieving high efficiency, with the Mo–Cu dual-site system identified as optimal for CH₄-to-EtOH conversion. Therefore, subsequent work mainly focuses on the synthesis of the selected Mo–Cu dual-site catalyst and the experimental validation of its electrocatalytic performance in CH₄-to-EtOH conversion.

Structural characterization of dual-site catalysts

The precursor, Cu₂Se nanorod arrays, was synthesized following an established method.²⁹ Tailoring of Cu₂Se with Cu vacancies (V_Cu) was achieved through doping of the Mo atom under reduction conditions (Fig. S1). Field emission-scanning electron microscopy (FESEM) and energy-dispersive X-ray spectroscopy (EDX) showed the preservation of the nanorod structure upon the introduction of Mo (Fig. S1–S4), with an increase in the diameter from 240 to 650 nm correlating with higher Mo content. X-ray diffraction (XRD) patterns of Mo–Cu₂Se aligned with those of cubic Cu₂Se (JCPDS 9-0914), without additional crystalline phases formed after different contents of Mo dopants (Fig. S5), which was corroborated by high-resolution transmission electron microscopy (HRTEM) imaging (Fig. S6). Low-temperature electron paramagnetic resonance (EPR) spectroscopy (Fig. S7) revealed the formation of Cu vacancies (V_Cu) after Mo doping, as indicated by a robust symmetric Lorentzian line with a g value of 2.0023, which is characteristic of typical V_Cu.³⁰ The red shift of the Cu 2p peaks observed in X-ray photoelectron spectroscopy (XPS) (Fig. S8) upon Mo incorporation indicated an increased electron density at the Cu sites, suggesting interfacial electronic charge transfer from Mo to Cu. This electronic interaction correlated with a higher Cu⁺/Cu²⁺ ratio in Table S1 and Fig. S9, stabilized at 1.73 by post-CTAC modification (Fig. S10), and was proposed to stabilize key intermediates by modulating the d-band electronic structure of Cu.

We further employed synchrotron-based X-ray absorption spectroscopy (XAS) to precisely characterize the local coordination environment of the Mo–Cu dual-site catalyst. The Mo K-edge X-ray absorption near-edge structure (XANES) spectrum of Mo–Cu₂Se, positioned between Mo foil and the MoS₂ reference (Fig. S11), in conjunction with Mo 3d XPS results, revealed that doped Mo existed in a state intermediate between Mo(III) and Mo(IV) states.^31,32 The Fourier-transformed extended X-ray absorption fine structure (FT-EXAFS) spectra displayed a prominent peak at 2.16 Å (Fig. S11b), also corroborated by Raman vibrational features at ∼350 and ∼600 cm⁻¹ (Fig. S12), both indicative of Mo–Cu bond formation.¹⁶ Artemis fitting results revealed that one Mo atom is coordinated with three Cu atoms and one Se atom, as shown in Fig. S13, forming a Mo–Cu₃–Se coordination configuration, supported by an R-space fitting factor of 0.004 (Fig. S14). Wavelet transform (WT) analysis of the Mo–Cu dual-site catalyst showed a maximum intensity of approximately 8.5 Å⁻¹ (Fig. S15), a distinctive feature absent in Mo foil, MoS₂ and MoO₃. These results demonstrated that the introduced Mo atoms acted as dopants, establishing a distinctive Mo–Cu₃–Se coordination.

Synergistic modulation of the local EEI microenvironment

We systematically investigated the effects of Mo doping and CTAC modification by finite element analysis (FEA) simulations using COMSOL to elucidate the role of the local electrode/electrolyte interface (EEI) in enhancing CH₄OR performance.³³ As shown in Fig. S16, pristine Cu₂Se shows a limited CH₄ adsorption of ∼0.03 mol L⁻¹, while the Mo-doping enhances the adsorption to ∼0.07 mol L⁻¹ by introducing a Mo–Cu dual-site. Further CTAC modification optimizes the EEI microenvironment, improving CH₄ adsorption up to ∼0.09 mol L⁻¹ with a more uniform spatial distribution. This stepwise enhancement highlights the synergistic effects of Mo doping and CTAC modification in promoting CH₄ availability, enabling superior CH₄OR performance. Furthermore, microenvironment analysis also revealed the critical role of Mo-doping and CTAC modification in enhancing CH₄-to-EtOH conversion by optimizing the OH⁻ concentration, charge distribution, and pH.

The Mo–Cu-CTAC system exhibits superior local reactant availability, as evidenced by a higher OH⁻ ion concentration of 0.09 mol L⁻¹ (Fig. S17), compared to the Mo–Cu dual-site (∼0.07 mol L⁻¹) and the Cu only site (∼0.06 mol L⁻¹). Charge distribution analysis also reveals a significantly enhanced surface charge density (20 mC m⁻²), surpassing the Mo–Cu dual-site (∼15 mC m⁻²) and the Cu only site (∼12 mC m⁻²), indicating improved charge transfer dynamics. Moreover, a stable and uniform local pH of 13–14 was maintained, contributing to stabilizing the catalytic microenvironment. COMSOL simulations (Fig. S17 and S18) further demonstrate that the increased local OH⁻ concentration not only promotes the formation of reactive oxygen species (ROS), particularly *O₂⁻ radicals, but also correlates strongly with higher surface EtOH coverage. While the OER is not entirely suppressed under these alkaline anodic conditions, the interfacial environment in Mo–Cu-CTAC kinetically favors the three-electron ROS pathway over the four-electron O₂ evolution route, effectively diverting oxidative equivalents toward CH₄ activation and *CH₃O formation. These optimized local conditions result in a high EtOH production of 0.10 mol L⁻¹, as shown in Fig. S18, outperforming the Mo–Cu dual-site (∼0.07 mol L⁻¹) and the Cu only site (∼0.05 mol L⁻¹). In parallel, the incorporation of CTAC introduces long-chain quaternary ammonium cations that increase surface hydrophobicity, as confirmed by contact angle test results (Fig. S19 and S20a), promoting CH₄ enrichment and limiting O₂ accumulation near active sites¹⁶ (Fig. S21), thereby reinforcing the kinetic selectivity for CH₄ oxidation. Post-electrolysis ¹H nuclear magnetic resonance (¹H NMR) analyses of both the anolyte and the catholyte detected no CTAC, and both the contact angle (Fig. S20b) and XPS N 1s binding energy (Fig. S20c) remained unchanged, indicating that the CTAC layer remains stably anchored to the catalyst surface. Based on the above results, the synergistic combination of Mo–Cu dual-site design and the optimized local microenvironment caused by CTAC enhances the performance of CH₄-to-EtOH conversion.

Electrolysis performance for CH₄-to-EtOH conversion

On the basis of the analysis described above, we systematically evaluated CH₄OR performances in an H-type cell, with products quantified by ¹H NMR spectroscopy and gas chromatography (GC). The optimized Mo–Cu dual-site catalyst exhibited an exceptional CH₄-to-EtOH conversion performance, achieving an EtOH FE of 24.2% at a low potential of 1.9 V (Fig. S22–S24). To exclude morphology-derived surface area effects, the electrochemical surface area was measured and found not to correlate directly with EtOH selectivity (Fig. S25), reinforcing that Mo–Cu electronic cooperation—not geometry—is the dominant performance determinant. To further explore the role of CTAC modification in enhancing CH₄ activation, the Mo–Cu dual-site catalyst was electrodeposited in the presence of different surface modifiers, including tetrabutylammonium chloride (TBAC), tetramethylammonium chloride (TMAC), and ammonium chloride (NH₄Cl) (Fig. S26 and S27). Among these, the 1.5 mM CTAC-modified Mo–Cu dual-site catalyst demonstrated the highest anodic current density in a CH₄-saturated atmosphere. This superior performance can be attributed to its optimal micellar organization and synergistic dual-site coordination,²⁹ which collectively enabled an EtOH production rate of 215.4 µmol h⁻¹ cm⁻² with a high EtOH FE of 54.5% (Fig. 3a and Fig. S28). The CH₄OR performance enhancement aligns with the FEA simulations results, as CTAC modification increases local CH₄ availability and adsorption. Specifically, the hydrophobic surface properties induced by the large quaternary ammonium cations of CTAC promote CH₄ adsorption while suppressing unwanted oxygen desorption associated with the OER (Fig. S29). The contribution from chloride ions is negligible, as confirmed through control experiments.³⁴

Fig. 3 Performance evaluation for CH₄-to-EtOH conversion. (a) EtOH yields and FEs at 1.9 V for Mo–Cu₂Se modified with different CTAC concentrations. (b) FEs and EtOH/O₂ ratio at 1.9 V for different catalysts. (c) FEs of alcohol products at 1.9 V for different transition metal modified Cu₂Se. (d) FEs and EtOH yield of the optimized Mo–Cu₂Se-CTAC in a flow cell with various potentials. (e) Current density and EtOH FE over the 20 h reaction at 1.9 V. (f) Other reported product comparison of CH₄OR with electrocatalysis, photocatalysis, thermal catalysis and coupling reactions; the only EtOH generation is highlighted by oval, the junctions of different colors represent each coupled catalytic reaction; the oval region highlighted in light pink marks the group of catalysts that exhibit EtOH as the major or exclusive carbon-containing product, denoted as the “High EtOH selectivity region”.

Further optimization of doped Mo contents in CTAC-modified samples revealed that the CTAC-modified catalyst with the Mo–Cu dual site achieved the highest EtOH FE of 54.5% ± 0.3% with an optimal EtOH/O₂ production ratio of 1.23 (Fig. 3b). Potential-dependent experiments demonstrated that maintaining a nearly 1 : 1 EtOH/O₂ production ratio is critical for maximizing CH₄-to-EtOH conversion efficiency (Fig. S30). This observation suggests a synergistic relationship between the OER and CH₄OR, where the controlled release of oxygen facilitates a balanced reaction microenvironment, paving the way for the efficient formation of key intermediates in the CH₄-to-EtOH conversion pathway.

To further validate the superior CH₄-to-EtOH performance, a series of transition metal-doped Cu₂Se samples were examined, among which the Mo–Cu-CTAC system achieved the highest EtOH FE of 54.5% ± 0.3% (Fig. 3c). Control experiments using MoSe_x-CTAC catalysts yielded methanol, formate and only a trace of EtOH (Fig. S31), confirming that only Mo facilitates ROS generation but lacks the ability to drive C–C coupling without Cu. Furthermore, the industrial applicability of Mo–Cu-CTAC at high current density was assessed in a flow cell equipped with a gas diffusion layer (Fig. S32 and S33). The integration of the gas diffusion layer in the flow cell ensured optimized mass transport and reaction kinetics, enabling the cascade system to achieve an impressive EtOH yield of 529.1 µmol h⁻¹ cm⁻² with a FE of 55.8% ± 0.2% at 1.9 V, corresponding to a high current density of 103 mA cm⁻², with stable operation maintained over 20 h (Fig. 3d and e). Compared to other reported transition metal-based CH₄OR catalysts, the Mo–Cu-CTAC system demonstrated an unparalleled activity and selectivity for EtOH production at high current densities (Fig. 3f).^{10,16,18,24,35–45} Although a-KB²⁴ and CoNi₂O_x³⁸ exhibit excellent activity, the Mo–Cu dual-site system operates under oxidant-free, halide-free, and near-neutral conditions to achieve direct C–C coupling and EtOH production. Also, the cascade electrocatalytic system surpassed traditional thermal or photo catalytic processes, which suffer from low selectivity and elevated energy demands.

Experimental mechanistic studies

To elucidate the CH₄OR mechanism, in situ attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy was employed. Under open-circuit conditions, CH₄ adsorption on Mo–Cu-CTAC was confirmed by observing obvious characteristic peaks at 1304, 1447, and 1468 cm⁻¹ (Fig. S34), corresponding to C–H deformation vibrations of CH₄ and *CH₂CH₃ deformation modes.¹⁰ The emergence of a peak at 1540 cm⁻¹ further corroborated the CH₄ adsorption. The progressive intensification of these peaks over time suggested that CTAC modification enhanced CH₄ adsorption efficiency.

The detailed CH₄-to-EtOH conversion process was further investigated at varying potentials and time intervals (Fig. 4a and b). A steady increase in positive peaks within the region of 3200–3600 cm⁻¹, accompanied by broad band signals indicative of O–H stretching vibrations in alcohols, as well as bending vibrations of –OH at 1670 cm⁻¹, signified EtOH formation. Simultaneously, a negative-going peak at 1620 cm⁻¹ reflected continuous water consumption during the reaction.²⁵ In Fig. 4c, positive peaks at 1420, 1390, 1258, 1210, and 1110 cm⁻¹ were attributed to C–H bending vibrations of EtOH, *CH₃O, and *CH₃CH₂O, O–O stretching vibrations of *CH₃OO, and C–O stretching vibrations of *CH₃CH₂O, respectively.^46–48 These intermediates exhibited a sequential trend of *CH₃CH₂O > *CH₃O > *CH₃OO, highlighting a stepwise transformation where *CH₃OO is reduced to *CH₃O and subsequently to *CH₃CH₂O. Time-resolved in situ Raman spectroscopy further revealed the dynamic formation of *CH₃O at ∼1058 cm⁻¹ and its subsequent conversion to EtOH, evidenced by the emergence of a broad band around 1350–1400 cm⁻¹ (Fig. S35). These observations provide direct support for *CH₃O as a key intermediate in the cascade CH₄-to-EtOH pathway.

Fig. 4 Operando spectroscopy analysis and mechanistic insight. In situ ATR-FTIR spectra recorded during the CH₄OR over the Mo–Cu-CTAC catalyst: (a) different potentials from 1.2 V to 2.1 V; (b) different times from 30 s to 300 s at 1.9 V; (c) corresponding contour map; and (d) different catalysts with ¹³CH₄/¹²CH₄ at 1.9 V. (e) EPR spectra of radical adducts trapped by DMPO-O₂ over Mo–Cu dual-site catalysts. (f) A proposed reaction mechanism for electrochemical CH₄-to-EtOH conversion.

Isotope-labeling experiments using ¹³CH₄ (99 atom%) unambiguously verified that both carbon atoms in the EtOH product originated from CH₄, as evidenced by the dominant GC–MS molecular ion at m/z = 48 and characteristic fragment at m/z = 33 (Fig. S36). Control experiments with Na₂¹³CO₃ as the electrolyte and unlabeled CH₄ showed only unlabeled EtOH (m/z = 46 and 31), excluding carbonate as a carbon source under the applied conditions (Fig. S37). To deeply corroborate the proposed CH₄-to-EtOH conversion pathway, isotope-labeling in situ ATR-FTIR experiments were subsequently conducted at 1.9 V using ¹³CH₄ as the feeding gas. As shown in Fig. 4d, the characteristic peaks of ¹³C-labeled intermediates shifted to the lower wavenumbers due to the isotope effect.⁴⁹ The formation of the ¹³EtOH product was confirmed and quantified by ¹³C-NMR spectroscopy (Fig. S38), demonstrating that the carbon in the EtOH product originated from CH₄ oxidation. Combining these findings with free radical trapping results presented in Fig. 4e, we concluded that the *O₂⁻ radical acted as the active species generated during the CH₄OR. This elucidates the catalytic pathway, wherein activated *CH₃ underwent the transformation to *CH₃OO, facilitated by the generation of *O₂⁻, potentially involving a three-electron *O₂⁻-mediated oxidation process. As shown in Fig. S39, the traditional four-electron OER process generates O₂ through a concerted transfer of four-electron pathways. In contrast, the three-electron *O₂⁻mediated oxidation pathway produces ROS such as *O₂⁻, particularly under alkaline and dual-site catalytic conditions. These partially oxidized intermediates exhibit moderate oxidation potential for *CH₃ radicals without overoxidation. On the Cu–Mo dual-site catalyst, this radical-mediated catalytic route is more compatible with cascade coupling pathways, allowing for stabilization and stepwise conversion of *CH₃ to EtOH. Computational calculations of kinetic analysis in Table S2 support that the Mo–Cu-CTAC exhibits the highest formation rates for *CH₃O (7.80 × 10⁻³ s⁻¹), *CH₃CH₂O (6.40 × 10⁻³ s⁻¹), and EtOH (4.20 × 10⁻³ s⁻¹), highlighting the enhanced efficiency enabled by the synergistic interaction between the Mo–Cu dual-site and CTAC. Based on these experimental and kinetic insights, a detailed CH₄-to-EtOH conversion mechanism was proposed (Fig. 4f). The Mo–Cu-CTAC system activated CH₄, enabling its stepwise oxidation via *CH₃, *CH₃OO, and *CH₃O to produce EtOH. Based on these findings, they collectively substantiate that reactive *O₂⁻ generated through a three-electron *O₂⁻-mediated oxidation pathway played a pivotal role in driving CH₄ oxidation and stabilizing key *CH₃O intermediates, where *O₂⁻ radicals initiate C–H activation and enable subsequent C–C coupling, ensuring high selectivity and yield of EtOH production.

Theoretical mechanistic investigations

To further elucidate the CH₄-to-EtOH conversion mechanism, DFT calculations were performed. The free-energy diagrams for oxygen evolution on the Cu site and the Mo–Cu dual-site (Fig. S40) reveal key differences in the reaction pathways. On Cu₂Se, the OER pathway progresses through the sequential formation of *OH, *O, *OOH, and O₂. However, the Mo doping alters this pathway by facilitating *OH deprotonation to form *O and enhancing *OO formation via *OOH deprotonation. Importantly, the Mo–Cu dual-site catalyst exhibits a three-electron *O₂⁻-mediated oxidation process that generates the *OO intermediate with a high energy barrier for O₂ desorption (+5.6 eV). This effectively prevents the traditional four-electron OER, instead stabilizing *OO. This stabilization is critical for subsequent CH₄ activation and conversion.

Theoretical exploration of CH₄OR identifies two potential mechanisms: non-radical and free-radical pathways, on both Mo–Cu and Mo sites (Fig. S41–S43). Both sites show comparable activity for CH₄ activation, with low activation energies of +0.22 eV (Mo–Cu) and +0.24 eV (Mo). The intermediate *OO plays a central role, reacting with *CH₃ to produce *CH₃OO, which subsequently transforms into *CH₃O. These steps are exothermic on Mo sites, with energy changes of −4.97 and −4.88 eV. In contrast, the critical C–C coupling step to form *CH₃CH₂O is exothermic on Mo–Cu sites (−0.93 eV) but endothermic on Mo sites (+0.42 eV), highlighting the superior role of Mo–Cu sites in enabling efficient *CH₃O transfer and final C–C bond formation.⁵⁰ These insights collectively emphasize the importance of the synergistic interactions between Mo and Mo–Cu sites in optimizing the CH₄OR processes, particularly in the formation and transfer of *CH₃O.

To deeply elucidate the dynamic behavior of the *CH₃O during the CH₄OR, molecular dynamics (MD) simulations were conducted. For comparative analysis, the Zn-doped system was included as a benchmark due to its previously reported high mobility for *CH₃O (Fig. S44). Mean square displacement (MSD) analysis (Fig. 5a) shows that the Mo–Cu dual-site exhibits a remarkable *CH₃O mobility, achieving a favorable diffusion coefficient of 0.017 Ų ps⁻¹, surpassing those of other dopants such as Co, Ni, and Fe, and nearly matching the performance of the Zn-doped system. However, for driving force (F_drive) of *CH₃O transfer, the Mo dopant offers additional advantages in catalytic efficiency and energy favorability, as evidenced by minimal energy requirement of −0.70 eV, among all other metal doped models in Table S3. This enhanced *CH₃O migration facilitates subsequent C–C coupling steps, underscoring the critical role of Mo as the preferred dopant in optimizing CH₄-to-EtOH conversion.

Fig. 5 Theoretical calculations for EtOH electrosynthesis. (a) Driving force and diffusion efficiency of *CH₃O on different metal-doped Cu₂Se surfaces in CH₄OR. (b) Electrochemical CH₄-to-EtOH mechanism highlighting the pivotal role of *CH₃O transfer. (c) Calculated free-energy diagrams for CH₄-to-EtOH conversion on the Mo–Cu₂Se model. (d) Energy barriers of the *CH₃O transfer pathway from the Mo to Cu atom; left inset: illustration of *CH₃O desorbing from the Mo site; right inset: illustration of *CH₃O absorbing onto the Mo–Cu site; the yellow region represents the electron accumulation area, and the blue region represents the electron loss area. (e) Corresponding atomic structures of the transition state of the *CH₃O transfer pathway (side view, isovalue = 0.003). (f) Possible pathway illustration of CH₄ to EtOH on the Mo–Cu₂Se model.

Based on DFT and MD simulations, the Mo–Cu dual-site exhibits enhanced *CH₃O migration and C–C coupling efficiency during the CH₄OR, due to distinct electronic complementarity. The Mo center, with stronger oxophilicity, stabilizes *CH₃O via enhanced electron withdrawal, while the adjacent Cu site offers moderate adsorption strength that facilitates *CH₃O desorption and subsequent C–C coupling. This dual-site configuration enables interfacial charge redistribution and orbital interactions between Mo d-states and Cu conduction bands, collectively lowering the energy barriers for intermediate migration and favoring cascade conversion toward EtOH (Fig. 5b). Notably, the unique role of the Mo dopant, vividly depicted as a fast-running athlete, ensures rapid *CH₃O migration to the Cu sites with precisely matched speed and efficiency, outperforming other metal dopants such as Zn, Fe, and Co. This synergy between Mo and Cu not only accelerates the transfer process but also enhances the selectivity and yield of EtOH production.

To better understand the cascade mechanism in facilitating *CH₃O transfer and subsequent C–C coupling during CH₄-to-EtOH conversion, Fig. 5c shows the critical synergy between Mo and Cu sites in driving selective CH₄-to-EtOH conversion. At 1.9 V, the free energy for *CH₃O migration from Mo to the Mo–Cu interface is minimized, enabling an efficient cascade process. The stepwise energy profile reveals that the subsequent C–C coupling step is highly exothermic, releasing −1.01 eV upon formation of *CH₃CH₂O. To assess the competing C₁ pathways, the thermodynamics of *CH₃ evolution toward CH₃OH and HCOOH were evaluated. Both routes exhibit higher relative free energy, indicating that these by-products are thermodynamically disfavored under the electrochemical CH₄-to-EtOH conversion conditions. Besides, in situ ATR-FTIR spectra reveal *CH₃O formation on Mo-based catalysts and EtOH on Cu₂Se, while both signals are observed on the Mo–Cu₂Se (Fig. S45). This spectral evolution supports an intermediate spillover process,⁵¹ in which *CH₃O is generated on Mo sites and migrates to adjacent Cu domains for subsequent C–C coupling. Fig. 5d further illustrates the energy profile for *CH₃O migration within the Mo–Cu interface, identifying key intermediates and a transition state corresponding to a migration barrier of +0.86 eV. This step represents the rate-determining step of the cascade mechanism and highlights the importance of spatially separated sites.

As shown in Fig. 5e, the subsequent *CH₃ and *CH₃O coupling is highly exergonic (−1.01 eV), reflecting the strong thermodynamic driving force toward EtOH formation. A comprehensive cascade mechanism is summarized in Fig. 5f and Fig. S46, where CH₄ activation initiates at Mo sites through a three-electron *O₂⁻-mediated oxidation pathway, generating *O₂⁻ that abstracts H from CH₄ to form *CH₃. This is followed by *CH₃O spillover from Mo to Cu, where it undergoes C–C coupling with *CH₃ to yield *CH₃CH₂O. This mechanistic sequence is supported by kinetic isotope effect (KIE) experiments. In Fig. S47, a significant primary KIE with CD₄ (k_CH4/k_CD4 of ∼2.88) confirms that C–H bond cleavage is rate-determining, while the minor effect with D₂O (k_H2O/k_D2O of ∼1.15) suggests that protonation steps are not involved in the kinetic bottleneck. To confirm the robustness of the dual-site system under CH₄OR conditions, DFT calculations (Fig. S48) revealed the thermodynamic stability of the Mo–Cu interface during *CH₃O binding and migration, while the post-reaction XPS analysis (Fig. S49 and Table S4) showed negligible changes in Mo and Cu oxidation states, demonstrating excellent structural and chemical durability.

Techno-economic analysis

To assess the economic feasibility of electrochemical CH₄-to-EtOH conversion, a preliminary techno-economic analysis (TEA) was performed to evaluate the levelized cost of the Mo–Cu-CTAC system. The analysis focuses on a hypothetical flow cell technology with a production capacity of 100 tons per day, considering scalability for larger operations (Fig. S50). The TEA encompasses a thorough breakdown of capital expenditure (CAPEX) and operational expenditure (OPEX) under optimal conditions. As illustrated in Fig. 6a, OPEX is primarily driven by the costs associated with electrolyzer electricity and CH₄ purchase, while CAPEX is dominated by the electrolyzer and balance of plant investments. This highlights the critical role of electrocatalysis in shaping both the efficiency and economic feasibility of CH₄-to-EtOH conversion.

Fig. 6 Economic and efficiency analysis of CH₄-to-EtOH conversion. (a) Production cost contribution to the CH₄OR (100 tons day⁻¹ EtOH), CAPEX: capital expenditure; OPEX: operating expenditure; PSA: pressure swing adsorption. (b) Production cost and energy return on investment were obtained as a function of the generation of EtOH.

The economic feasibility of the electrochemical process was evaluated against conventional thermal catalytic methods, including steam reforming methanol (SRM) and chemical looping. Based on the economic and technological assumptions outlined in Table S5 and the SI, the analysis reveals that the electrochemical approach is more cost-competitive for EtOH production capacities below 500 tonnes per day (Fig. 6b).^52–54 At a scale of 400 tonnes per day, the electrocatalysis system achieves a competitive production cost of ∼$2.12 per kg, compared to $2.59 per kg for SRM and $1.71 per kg for chemical looping. Although the electrochemical process currently involves a slightly higher cost than chemical looping, it benefits from ambient-condition operation, modular scalability, and low CO₂ emissions. Importantly, techno-economic projections further suggest that integrating renewable electricity could lower the CH₄-to-EtOH cost to ∼$1.50 per kg within a decade, achieving a 53% energy return. Besides, the modest improvements in FE, cell voltage, CAPEX, or OPEX, etc., in Table S6, could yield economic parity, underscoring the route's promising scalability and technological headroom.

Moreover, process-level performance metrics, including single-pass CH₄ conversion, energy efficiency, specific energy consumption, and electron economy, are summarized in Table S7, further substantiating the strong scalability and energy return potential of this electrocatalytic platform. Looking ahead, the integration of renewable natural gas and electricity is expected to further transform the economics of CH₄ electro-upgrading, offering a sustainable and flexible platform for future energy and carbon utilization infrastructures.

Conclusions

In summary, we have developed a machine learning-assisted dual-site-driven cascade strategy for selective CH₄-to-EtOH conversion under ambient conditions. This approach achieved nearly 100% alcohol selectivity, with an EtOH FE of 55.8% ± 0.2% and a formation rate of 529.1 µmol h⁻¹ cm⁻², at 103 mA cm⁻², setting a new performance benchmark. DFT calculations and in situ ATR-FTIR spectroscopy revealed that the Mo–Cu dual-site interface promotes CH₄ oxidation via a three-electron *O₂⁻-mediated oxidation pathway, followed by precise *CH₃O migration and exothermic C–C coupling. Isotope labeling and KIE studies validated *CH₃O as a key intermediate and confirmed that homolytic C–H bond cleavage is rate-limiting. TEA results demonstrated the economic viability of the dual-site-driven cascade system for production capacities at a scale of 400 tonnes per day, with an EtOH production cost of $2.12 per kg alongside a 53% energy return on investment. This work not only deepens the mechanistic understanding of CH₄OR by emphasizing the central role of *CH₃O spillover and dual-site cascade cooperation but also establishes a scalable, cost-effective and machine learning-driven pathway for EtOH production. Importantly, this integrative strategy—combining data-driven catalyst design with cascade-regulated active sites—offers a generalizable framework that may be extended to other small-molecule transformations such as CO₂ reduction and N₂ fixation, paving the way for future innovations in sustainable electrocatalysis.

Author contributions

Y. H. supervised the project. L. Z. conceptualized and conducted the majority of the experiments, and drafted and revised the manuscript. X. W, X. P., Z. L., and K. W. conducted part of the synthesis of catalysts and their characterization studies. D.W. performed theoretical calculations. N. W. carried out the technoeconomic analysis. Z. L., B. Y., Q. Z., L. L., and P. S. were involved in data analysis, discussions, and manuscript revision. All authors actively contributed to the interpretation of results and provided critical feedback on the manuscript.

Conflicts of interest

The authors declare no competing interest.

Data availability

The data supporting the findings of this study are available in the article and its supplementary information (SI). Supplementary information: additional control experiments verifying product specificity and electrolyte purity (Fig. S51–S53), which confirm that no EtOH or other carbon-containing liquid products were detected in the absence of CH₄, ensuring the robustness and specificity of the reported catalytic performance. See DOI: https://doi.org/10.1039/d5ee05379j.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (U22A20432, 22425805, 22408325, 22278364, 22211530045, 22178308, 21878270, 21961160742, 22578393, and 22238008), the National Key Research and Development Program of China (2022YFB4002100), the Development Project of Zhejiang Province's “Jianbing” and “Lingyan” (2023C01226), the Baima Lake Laboratory Joint Fund of Zhejiang Provincial Natural Science Foundation (LBMHZ25B030007), the Fundamental Research Funds for the Central Universities (226-2024-00060, 226-2025-00224), the Research Funds of Institute of Zhejiang University-Quzhou (IZQ2021RCZX022), the Key Technology Breakthrough Program of Ningbo “Science and Innovation Yongjiang 2035” (2024H024), and the Fundamental Research Funds for the Zhejiang Provincial Universities (226-2025-00224).

Notes and references

1. L. C. King and J. van den Bergh, Nat. Sustainable, 2022, 5(10), 899–905.
2. N. Aggarwal, H. L. Pham, B. Ranjan, M. Saini, Y. Liang, G. S. Hossain, H. Ling, J. L. Foo and M. W. Chang, Nat. Rev. Bioeng., 2024, 2, 155–174.
3. E. Kocak, F. Bilgili, U. Bulut and S. Kuskaya, Renewable Energy, 2022, 199, 689–696.
4. P. De Luna, C. Hahn, D. Higgins, S. A. Jaffer, T. F. Jaramillo and E. H. Sargent, Science, 2019, 364, eaav3506.
5. X. Meng, X. Cui, N. P. Rajan, L. Yu, D. Deng and X. Bao, Chem, 2019, 5(9), 2296–2325.
6. K. Shen, S. Kumari, Y. C. Huang, J. Jang, P. Sautet and C. G. Morales-Guio, J. Am. Chem. Soc., 2023, 145(12), 6927–6943.
7. S. Wang, V. Fung, M. J. Hülsey, X. Liang, Z. Yu, J. Chang, A. Folli, R. J. Lewis, G. J. Hutchings, Q. He and N. Yan, Nat. Catal., 2023, 6, 895–905.
8. S. Yuan, Y. Li, J. Peng, Y. M. Questell-Santiago, K. Akkiraju, L. Giordano, D. J. Zheng, S. Bagi, Y. Román-Leshkov and Y. Shao-Horn, Adv. Energy Mater., 2020, 10(40), 2002154.
9. Z. Guo, B. Liu, Q. Zhang, W. Deng, Y. Wang and Y. Yang, Chem. Soc. Rev., 2014, 43(10), 3480.
10. W. Jiang, J. Low, K. Mao, D. Duan, S. Chen, W. Liu, C. W. Pao, J. Ma, S. Sang, C. Shu, X. Zhan, Z. Qi, H. Zhang, Z. Liu, X. Wu, R. Long, L. Song and Y. Xiong, J. Am. Chem. Soc., 2021, 143(1), 269–278.
11. 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(11), 889–896.
12. Y.-F. Tsai, T. Natarajan, Z.-H. Lin, I.-K. Tsai, D. Janmanchi, S. I. Chan and S. S.-F. Yu, J. Am. Chem. Soc., 2022, 144, 9695–9706.
13. X. Peng, L. Zeng, D. Wang, Z. Liu, Y. Li, Z. Li, B. Yang, L. Lei, L. Dai and Y. Hou, Chem. Soc. Rev., 2023, 52, 2193–2237.
14. S. Nie, L. Wu, Q. Zhang, Y. Huang, Q. Liu and X. Wang, Nat. Commun., 2024, 15, 6669.
15. W. Li, Y. Liu, Z. Wang, Y. Liu, X. Yang, X. Wang, Y. Wang, A. Wang and B. Su, Proc. Natl. Acad. Sci. U. S. A., 2023, 120, e2206619120.
16. C. Kim, H. Min, J. Kim and J. H. Moon, Energy Environ. Sci., 2023, 16(7), 3158–3165.
17. Z. Luo, G. Zhao, H. Pan and W. Sun, Adv. Energy Mater., 2022, 12(37), 2201395.
18. 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.
19. L. Zeng, A. R. Thiruppathi, J. van der Zalm, T. Shi and A. Chen, J. Mater. Chem. A, 2022, 10, 15012–15025.
20. N. Xu, C. A. Coco, Y. Wang, T. Su, Y. Wang, L. Peng, Y. Zhang, Y. Liu, J. Qiao and X.-D. Zhou, Appl. Catal., B, 2021, 282, 119572.
21. J. Deng, S. C. Lin, J. T. Fuller III, B. Zandkarimi, H. M. Chen, A. N. Alexandrova and C. Liu, Angew. Chem., Int. Ed., 2021, 60(51), 26630–26638.
22. Y. Liu, J.-C. Liu, T.-H. Li, Z.-H. Duan, T.-Y. Zhang, M. Yan, W.-L. Li, H. Xiao, Y.-G. Wang, C.-R. Chang and J. Li, Angew. Chem., Int. Ed., 2020, 59, 18586–18590.
23. 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(9), 2954–2963.
24. J. Kim, J. H. Kim, C. Oh, H. Yun, E. Lee, H. S. Oh, J. H. Park and Y. J. Hwang, Nat. Commun., 2023, 14(1), 4704.
25. X. Y. Lin, Y. T. Wang, X. Chang, S. Y. Zhen, Z. J. Zhao and J. L. Gong, Angew. Chem., Int. Ed., 2023, 62, e202300122.
26. J. Chen, D. Wang, X. Yang, W. Cui, X. Sang, Z. Zhao, L. Wang, Z. Li, B. Yang, L. Lei, J. Zheng, L. Dai and Y. Hou, Angew. Chem., Int. Ed., 2023, 62(10), e202215406.
27. A. Prajapati, B. A. Collins, J. D. Goodpaster and M. R. Singh, Proc. Natl. Acad. Sci. U. S. A., 2021, 118(8), e2023233118.
28. J. Gong, H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang and X. Ma, J. Am. Chem. Soc., 2012, 134(34), 13922–13925.
29. W. Sun, P. Wang, Y. Jiang, Z. Jiang, R. Long, Z. Chen, P. Song, T. Sheng, Z. Wu and Y. Xiong, Adv. Mater., 2022, 34(50), e2207691.
30. L. Zeng, X. Li, S. Fan, Z. Yin, J. Mu, M. Qin and A. Chen, J. Power Sources, 2020, 478, 228755.
31. W. Chen, J. Pei, C. T. He, J. Wan, H. Ren, Y. Zhu, Y. Wang, J. Dong, S. Tian, W. C. Cheong, S. Lu, L. Zheng, X. Zheng, W. Yan, Z. Zhuang, C. Chen, Q. Peng, D. Wang and Y. Li, Angew. Chem., Int. Ed., 2017, 56(50), 16086–16090.
32. M. A. Lukowski, A. S. Daniel, F. Meng, A. Forticaux, L. Li and S. Jin, J. Am. Chem. Soc., 2013, 135(28), 10274–10277.
33. B. Ren, Z. Zhang, G. Wen, X. Zhang, M. Xu, Y. Weng, Y. Nie, H. Dou, Y. Jiang, Y. P. Deng, G. Sun, D. Luo, L. Shui, X. Wang, M. Feng, A. Yu and Z. Chen, Adv. Mater., 2022, 34(38), e2204637.
34. Y. Zhong, Y. Xu, J. Ma, C. Wang, S. Sheng, C. Cheng, M. Li, L. Han, L. Zhou, Z. Cai, Y. Kuang, Z. Liang and X. Sun, Angew. Chem., Int. Ed., 2020, 59(43), 19095–19101.
35. J. Li, L. Yao, D. Wu, J. King, S. S. C. Chuang, B. Liu and Z. Peng, Appl. Catal., B, 2022, 316, 121657.
36. Z. Xie, M. Chen, Y. Chen, A. Guan, Q. Han and G. Zheng, J. Phys. Chem. C, 2021, 125(24), 13324–13330.
37. J. Lee, J. Yang and J. H. Moon, ACS Energy Lett., 2021, 6(3), 893–899.
38. 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.
39. Z. Jin, L. Wang, E. Zuidema, K. Mondal, M. Zhang, J. Zhang, C. Wang, X. Meng, H. Yang, C. Mesters and F. S. Xiao, Science, 2020, 367(6474), 193–197.
40. H. Zhou, T. Liu, X. Zhao, Y. Zhao, H. Lv, S. Fang, X. Wang, F. Zhou, Q. Xu, J. Xu, C. Xiong, Z. Xue, K. Wang, W. C. Cheong, W. Xi, L. Gu, T. Yao, S. Wei, X. Hong, J. Luo, Y. Li and Y. Wu, Angew. Chem., Int. Ed., 2019, 58(51), 18388–18393.
41. B. Han, W. Wei, M. Li, K. Sun and Y. H. Hu, Chem. Commun., 2019, 55(54), 7816–7819.
42. D. L. McRose, J. Li and D. K. Newman, Proc. Natl. Acad. Sci. U. S. A., 2023, 120(14), e2217951120.
43. B. Yu, L. Cheng, S. Dai, Y. Jiang, B. Yang, H. Li, Y. Zhao, J. Xu, Y. Zhang, C. Pan, X. M. Cao, Y. Zhu and Y. Lou, Adv. Sci., 2023, 10(26), e2302143.
44. N. Agarwal, S. J. Freakley, R. U. McVicker, S. M. Althahban, N. Dimitratos, Q. He, D. J. Morgan, R. L. Jenkins, D. J. Willock, S. H. Taylor, C. J. Kiely and G. J. Hutchings, Science, 2017, 358, 223–227.
45. B. An, Z. Li, Z. Wang, X. Zeng, X. Han, Y. Cheng, A. M. Sheveleva, Z. Zhang, F. Tuna, E. J. L. McInnes, M. D. Frogley, A. J. Ramirez-Cuesta, L. S. Natrajan, C. Wang, W. Lin, S. Yang and M. Schroder, Nat. Mater., 2022, 21, 932–938.
46. S. A. Khan, S. B. Khan, L. U. Khan, A. Farooq, K. Akhtar and A. M. Asiri, Handbook of Materials Characterization, Springer International Publishing, Cham, 2018, pp. 317–344.
47. K. C. Wong, Review of spectrometric identification of organic compounds, ACS Publications, 2015.
48. B. C. Smith, Infrared spectral interpretation: a systematic approach, CRC press, 2018.
49. C. Chen, X. Zhu, X. Wen, Y. Zhou, L. Zhou, H. Li, L. Tao, Q. Li, S. Du, T. Liu, D. Yang, C. Xie, Y. Zou, Y. Wang, R. Chen, J. Huo, Y. Li, J. Cheng, H. Su, X. Zhao, W. Cheng, Q. Liu, H. Lin, J. Luo, J. Chen, M. Dong, K. Cheng, C. Li and S. Wang, Nat. Chem., 2020, 12(8), 717–724.
50. W. Ma, S. Xie, T. Liu, Q. Fan, J. Ye, F. Sun, Z. Jiang, Q. Zhang, J. Cheng and Y. Wang, Nat. Catal., 2020, 3(6), 478–487.
51. W. Ye, Y. Yao, X. Wei, M. Xu, S. Zhao, W. Wang, G. Jia, F. Dai, P. Gao, X. Lu, X. Li, B. Xi, N. Wang and S. Xiong, Angew. Chem., Int. Ed., 2025, e202509303.
52. D. Lim, B. Lee, H. Lee, M. Byun and H. Lim, Renewable Sustainable Energy Rev., 2022, 155, 111876.
53. A. Blankenship, M. Artsiusheuski, V. Sushkevich and J. A. van Bokhoven, Nat. Catal., 2023, 6(9), 748–762.
54. J.-P. Lange, V. L. Sushkevich, A. J. Knorpp and J. A. van Bokhoven, Ind. Eng. Chem. Res., 2019, 58(20), 8674–8680.

---