Superhydrophobicity-enabled triphase interfacial microenvironment for high-temperature CO₂-to-formate electrocatalysis

INTRODUCTION

Electrocatalytic CO₂ reduction (eCO₂RR) represents a promising strategy for converting waste CO₂ into useful chemicals, while advancing carbon neutrality and sustainable energy cycles^[1-4]. Unlike many multi-carbon hydrocarbon products from eCO₂RR that exhibit low faradaic efficiencies, formate represents one of the most commercially viable targets^[5], owing to its broad utility in industrial synthesis and energy-related applications. Despite significant advances in catalyst design, the practical application of this technology remains constrained not only by sluggish reaction kinetics^[6-10], but also by detrimental formate accumulation at active sites, which degrades the interfacial microenvironment and promotes the hydrogen evolution reaction (HER)^[11-13]. Unfortunately, conventional optimization strategies centered on catalyst or electrolyte modification offer limited leverage over the inherent reaction-diffusion constraints^[14-17] and are often insufficient to simultaneously enhance the intrinsic catalytic turnover and mitigate the mass transport bottleneck imposed by formate accumulation.

Recent advances highlight thermal activation as an attractive strategy for enhancing catalytic activity and alleviating product accumulation^[18,19], thereby addressing the dual challenges of conventional eCO₂RR systems. Nevertheless, the thermodynamic benefit of heating is compromised by the concomitant reduction in CO₂ solubility, leading to severe mass transport limitations due to reactant depletion at the electrode interface^[20,21]. This transport bottleneck is further aggravated by the Arrhenius-type acceleration of the competing HER, which excessively consumes protons and electrons^[22-24]. Although this thermally driven desorption is beneficial for formate removal, it occurs alongside the solubility-driven CO₂ deficit and HER acceleration mentioned above, creating a complex trade-off that thus constrains its performance^[25-28]. Consequently, this inherent conflict imposes both thermodynamic and kinetic constraints, effectively limiting high-performance electrocatalysis to a narrow near-ambient temperature window^[29-31].

As an interface-dominated process, eCO₂RR suffers from severe mass transport limitations arising from the low solubility and slow diffusion of CO₂ in aqueous electrolytes. To address this, constructing solid-liquid-gas triphase boundaries (TPBs) has emerged as an effective strategy for achieving localized CO₂ enrichment at the catalyst surface, thereby significantly enhancing the CO₂ supply capability at the catalyst/electrolyte interface^[32-35]. By stabilizing a persistent gas film atop superhydrophobic scaffolds, TPBs decouple the interfacial reactant supply from bulk solubility constraints and concurrently modulate the local hydration environment to mitigate parasitic reactions. While previous TPB-based studies have predominantly focused on mitigating reactant (CO₂) starvation at ambient temperatures, the impact of thermal activation on product expulsion dynamics remains largely unexplored. When applied to thermally activated systems, this architecture uniquely enables management of the complex interplay between thermal kinetics and mass transport^[36,37]. The inherent hydrophobic confinement preserves gaseous reactant (CO₂) channels at elevated temperatures, allowing the thermally enhanced diffusion kinetics to be specifically leveraged for accelerating product expulsion from the interface^[38,39].

Inspired by these advances, herein we propose a catalyst-integrated, superhydrophobic TPB architecture that resolves the conflict inherent in thermal activation for high-efficiency CO₂-to-formate conversion. As illustrated in Figure 1, the superhydrophobic substrate establishes a gas/liquid dual-transport pathway. This design utilizes the gas channel (red arrow) to compensate for the solubility-driven CO₂ depletion in the liquid phase (blue arrow) and leverages elevated temperatures to accelerate formate desorption (black arrow), thereby ensuring optimal utilization of thermal intensification strategies. Electrochemical evaluation reveals that, in stark contrast to conventional diphase systems where HER dominates upon heating, our TPB electrode achieves a 1.7-fold increase in formate production rate at 50 °C relative to 20 °C, while maintaining a Faradaic efficiency above 80%. Steady-state multiphysics simulations were employed to quantify the interfacial CO₂ and formate distributions within different microenvironments under thermal activation. This work addresses this gap by elucidating how elevated temperatures actively leverage the triphase boundary to accelerate product desorption, thereby presenting a dual-regulation strategy for thermally intensified CO₂ electroreduction. The results confirm that the gas channel sustains near-saturated CO₂ concentrations precisely at the reaction interface, effectively offsetting the thermally induced solubility loss in the bulk liquid. This dual-regulation mechanism simultaneously addresses both the CO₂ supply deficiency and the adverse effect of product accumulation. By maintaining robust CO₂ delivery and rapid product removal, this design overcomes the inherent mass-transport bottlenecks of thermal intensification, ensuring high-efficiency formate production.

Figure 1. Schematic illustration of the solid-liquid-gas triphase interface for thermally intensified CO₂ electroreduction to formate. With the gas channel (red arrow) providing a sufficient supply to counteract the solubility-driven CO₂ depletion in the liquid phase (blue arrow), the elevated temperature accelerates formate desorption (black arrow), synergistically ensuring the maximal utilization of thermal intensification strategies.

EXPERIMENTAL

RESULTS AND DISCUSSION

To realize the triphase interface concept depicted in Figure 1, we engineered an integrated electrode architecture wherein the catalyst is embedded within a stabilized triphase microenvironment [Figure 2A]. The fabrication commenced with a commercial hydrophobic carbon fiber paper pre-loaded with hydrophobic carbon particles, which served as a conductive backbone and imparted the necessary surface roughness to ensure superhydrophobicity. This hierarchical roughness enables the substrate to trap a stable gas film, thereby establishing the triphase interface essential for the efficient catalysis. Field emission scanning electron microscope (FE-SEM) imaging [Figure 2B] reveals the hierarchical micro/nano-structure, while the accompanying water contact angle measurement of 146 ± 2° (Figure 2B, inset) confirms the surface’s exceptional hydrophobicity, guaranteeing the formation of a stable gas film upon immersion^[40,41].

Figure 2. (A) Schematic illustration of the catalyst-integrated, superhydrophobic carbon-based TPB architecture; (B) Field emission scanning electron microscope (FE-SEM) of the superhydrophobic carbon substrate. The inset shows a water droplet placed on the substrate with a water contact angle of approximately 146 ± 2°; (C) FE-SEM of the In nanoparticles immobilized on the carbon substrate; (D) EDS elemental maps of In element; (E) XRD patterns of carbon substrate and In@substrate. EDS: Energy-dispersive X-ray spectroscopy; TPB: triphase boundary; XRD: X-ray diffraction; SEM: scanning electron microscope.

Indium (In) nanoparticles, serving as a model electrocatalyst for CO₂ conversion into formate, were loaded onto the superhydrophobic carbon substrate via a thermal decomposition procedure, followed by in situ electrochemical reduction to yield the triphase electrode. Following the immobilization of In nanoparticles, the electrode surface exhibited a water contact angle (CA) of 142 ± 2° [Supplementary Figure 1], confirming its robust hydrophobic nature. As shown in Figures 2C and Supplementary Figure 2, these In nanoparticles (10~30 nm) are uniformly dispersed across the carbon particle surfaces. This uniform distribution is further substantiated by energy-dispersive X-ray spectroscopy (EDS) elemental mapping [Figure 2D], which verifies the homogeneous spatial distribution of In signals across the electrode area, corroborating consistent catalyst loading. XRD pattern in Figure 2E exhibits distinct peaks that correspond precisely to the standard reference pattern for body-centered tetragonal metallic indium (PDF#85-1409), confirming the successful synthesis and crystallinity of the In catalyst. These characterizations confirm the successful fabrication of the targeted triphase electrode architecture, which maintains a stable solid-liquid-gas microenvironment during electrolysis and serves as a robust platform for high-temperature CO₂ electroreduction studies. For comparison, a diphase electrode (In-2P) was prepared by subjecting the triphase In electrode (In-3P) to O₂ plasma treatment. As shown in Supplementary Figure 3, this treated surface became hydrophilic, exhibiting a water contact angle of 21 ± 2°.

Following structural characterizations of the as-prepared electrode, we evaluated its thermally activated electrocatalytic activity via potential-resolved chronoamperometric tests [Supplementary Figures 4 and 5]. The electrocatalytic CO₂ reduction performance was evaluated at different temperatures in CO₂-saturated 0.5 M KHCO₃. Prior to each test, the entire assembly - including the electrode, electrolyte, and cell - was preheated to the target temperature in a water bath under a continuous CO₂ flow. Steady-state current densities were extracted from the acquired i-t profiles at different applied potentials [Figure 3]. Figure 3A illustrates that the In-3P electrode yields steady, monotonically increasing current densities as temperature rises throughout the measured potential range. At -0.8 V vs. Reversible Hydrogen Electrode (RHE) as a representative point, current density rises substantially from -15 mA cm^-2 at 20 °C up to -26 mA cm^-2 at 50 °C. This observation confirms that the triphase boundary suppresses local CO₂ starvation, enabling thermal energy to continuously boost catalytic currents without encountering mass-transport-limited saturation^[42]. By comparison, the conventional diphase In-2P electrode exhibits markedly diminished performance enhancements upon heating [Figure 3B]. Notably, improvements remain negligible at potentials anodic of -0.8 V vs. RHE; within this potential window, raising the temperature from 20 °C to 50 °C elicits little to no increase in measured current densities. Such behavior demonstrates that mass-transport constraints impose an insurmountable upper limit on activity, as the decline in CO₂ solubility at higher temperatures negates kinetic advantages from thermal activation.

Figure 3. (A and B) Total current densities for the CO₂RR achieved using the triphase and diphase electrodes, respectively, at various potentials and temperatures; (C) FE of CO₂RR products on In-3P electrode at a potential of -0.9 V vs. RHE and various temperatures; (D) FE of CO₂RR products on In-2P electrode at a potential of -0.9 V vs. RHE and various temperatures. CO₂RR: CO₂ reduction reaction; FE: Faradaic efficiency; RHE: reversible hydrogen electrode.

To decouple contributions to the measured current between target CO₂ reduction and parasitic hydrogen evolution (HER), product speciation was quantified via in-line gas chromatography and post-reaction NMR characterization [Supplementary Figures 6 and 7]. Over the full range of measured temperatures and potentials, the In-3P catalyst sustains remarkably high formate selectivity, with Faradaic efficiencies for formate (FE-HCOOH) consistently above 80% [Figure 3C]. Quantitatively, the formate partial current density (J-HCOOH) increases by a factor of ~1.7 as the temperature rises from 20 °C to 50 °C, corresponding to an ~1.7-fold increase in formate production [Supplementary Figure 8], a performance enhancement ranking among the competitive results for high-temperature CO₂-to-formate conversion^[43-46]. This firmly validates that the thermally amplified currents plotted in Figure 3A originate from faster CO₂-to-formate conversion rather than HER. Though heating increases local proton availability and typically favors HER, the triphase boundary modulates the near-electrode microenvironment to effectively suppress competing hydrogen evolution. Conversely, the In-2P electrode suffered a precipitous loss in selectivity upon heating [Figure 3D]. While a reasonable FE-HCOOH (~70%) was sustained at 20 °C, it declined with increasing temperature. At 50 °C and -0.9 V vs. RHE, hydrogen evolution nearly dominates the entire electrolysis process, yielding FE-HCOOH below 12%. Such results confirm that the thermal current gain in typical diphase systems primarily originates from parasitic HER activity. To further decouple the role of the gaseous CO₂ pathway from potential substrate modifications introduced during electrode fabrication, we conducted a control experiment using the pristine In-3P electrode (In-3P-N₂). In this configuration, the gas chamber was purged with inert N₂ while CO₂ was supplied exclusively through the liquid electrolyte. This approach preserves the native surface chemistry of the catalyst and carbon scaffold while solely removing the gas-phase reactant supply. As demonstrated in Supplementary Figures 9 and 10, the In-3P-N₂ electrode mirrors the In-2P performance trajectory, exhibiting increased current density with temperature but a sharp decline in formate selectivity as HER dominates at elevated temperatures. This convergence confirms that the absence of a gaseous CO₂ pathway is the determining factor governing high-temperature selectivity.

The markedly different temperature-dependent behaviors in the diphase and triphase systems using the same catalyst can be attributed to their distinct interfacial microenvironments, which modulate the local availability of CO₂ and H₂O at the reaction interface. To explore this difference further, the electrochemical responses of both electrodes were examined at various temperatures. As shown in Figure 4A, the linear sweep voltammetry (LSV) profiles were recorded in CO₂-saturated electrolyte at varying temperatures. In line with the trends shown in Figure 3, heating boosts current densities for both electrodes, demonstrating that thermal promotion of reaction kinetics applies to both systems. Notably, the In-3P electrode delivers substantially higher catalytic currents than the In-2P counterpart under identical thermal conditions, highlighting the mass-transport advantages of the triphase design.

Figure 4. (A) LSV curves for triphase and diphase electrodes at 20 °C and 50 °C; (B) Tafel analyses of the partial current densities for formate production; (C) Apparent turnover frequency (TOF) of HCOOH production at -0.9 V vs. RHE under various temperatures; (D) Stability of In-3P electrode at -0.9 V vs. RHE and 40 °C. LSV: Linear sweep voltammetry; FE: Faradaic efficiency; RHE: reversible hydrogen electrode.

To further understand the distinct activity and selectivity at elevated temperatures, Tafel analysis was performed on the formate partial current densities of both electrodes at 20 °C and 50 °C [Figure 4B], derived from the LSV profiles [Figure 4A] and corresponding Faradaic efficiencies. The In-3P electrode exhibited a marked shift toward less negative Tafel slopes, from -108 to -54 mV dec^-1, as the temperature increased from 20 °C to 50 °C. This shift confirms that the triphase design mitigates the reactant depletion typically exacerbated at high temperatures, effectively accelerating the intrinsic kinetics of formate formation. This behavior is attributed to the ability of the triphase boundary to sustain near-saturated interfacial CO₂ while simultaneously alleviating local product accumulation. By decoupling the reaction zone from bulk mass-transport constraints, thermal energy is channeled into accelerating the intrinsic catalytic turnover rather than being dissipated by kinetic bottlenecks. In contrast, the In-2P electrode exhibited a marked shift toward more negative Tafel slopes, from -165 mV dec^-1 at 20 °C to -185 mV dec^-1 at 50 °C. This change indicates that the kinetic acceleration from thermal activation is largely negated by intensified mass-transport limitations and parasitic HER. The increasing slope magnitude quantitatively reflects the intensified competition between HER and CO₂ reduction, highlighting the system’s inability to utilize thermal energy for the target reaction without a gaseous CO₂ supply. These results demonstrate that the rationally designed triphase interface successfully transforms thermal energy from a performance-limiting factor in conventional aqueous electrocatalysis into a powerful amplifier for boosting target CO₂ reduction kinetics.

The intrinsic advantage of the triphase architecture emerges clearly when examining the turnover frequency (TOF) for formate production shown in Figure 4C. Notably, the TOF values reported herein represent apparent TOF, calculated based on the total moles of In precursor loaded during fabrication and assuming quantitative conversion to metallic In. While the In-3P electrode exhibits a monotonic TOF increase from 0.079 s^-1 at 20 °C to 0.108 s^-1 at 50 °C, the In-2P electrode follows an opposite trend, with its TOF decreasing from 0.022 s^-1 to 0.0064 s^-1. This difference reveals a fundamental mechanistic shift that aligns with the observations in the selectivity data in Figure 3. Conversely, the diphase electrode undergoes a thermally driven decay in apparent intrinsic activity, countering the acceleration in reaction kinetics and resulting in only marginal current increase. More critically, the loss of intrinsic activity correlates with the significant deterioration in selectivity, where thermally enhanced HER outcompetes CO₂ reduction under reactant-starved conditions. Therefore, the triphase design overcomes the typical activity-selectivity trade-off in high-temperature CO₂ electroreduction by directing thermal energy specifically into the target reaction pathway. Chronoamperometry at -0.9 V vs. RHE and 40 °C shows a steady current density over 24 h, with the Faradaic efficiency for formate remaining above 80% [Figure 4D]. Post-test characterization [Supplementary Figure 11] confirms the intact chemical state and structural robustness of the catalyst, with only a marginal decline in surface hydrophobicity after the 24 h test. This operational durability highlights the thermostability of the triphase design for continuous, high-efficiency CO₂ electroreduction. While these results are highly encouraging for industrial translation, future long-term operation warrants monitoring for potential hydrophobic scaffold evolution and possible trace In leaching under extended cathodic biases.

To further understand the effect of interfacial architecture on reactant and product distributions, a continuum-scale multiphysics model was constructed to simulate the steady-state concentration profiles within the diphase and triphase systems [Figure 5A, Supplementary Figure 12 and Supplementary Table 1]. It is important to note that this simulation framework was constructed to deconvolute experimental observations rather than to predict unknown phenomena. The computational domain for the triphase system comprises a gas diffusion layer (GDL), a catalyst layer (CL), a liquid diffusion layer (LDL), and the underlying substrate. The model incorporates species diffusion (including CO₂), CO₂ dissolution, and formate production. Species transport is described by Fick’s law. CO₂ dissolution from the gas phase to the liquid phase is described by Henry’s law [Supplementary Table 2]. The catalytic reaction for formate production is modeled as first-order kinetics with respect to the interfacial formate concentration. This formulation is dictated by the pronounced shift toward less negative Tafel slopes at elevated temperatures, which signifies a transition to a rate-determining step no longer limited by the accumulation of reaction products. This rate law ensures the model captures the decoupling of kinetics from bulk solubility constraints under thermal activation.

Figure 5. (A) The schematic diagram of the computational domain with boundary indicators; (B) Simulated CO₂ concentration profiles as a function of distance from the triphase interface into the solution for the triphase (red) and diphase (black) systems at 50 °C. Diagram on the right is CO₂ concentration diffusion model in the triphase (the upper) and diphase (the lower) systems; (C) Simulated local formate concentration at the reaction interface as a function of temperature in the triphase system. Diagram on the right is the formate concentration diffusion model.

Figure 5B shows the simulated results of CO₂ concentration (c_CO2) distribution in the catalytic zone during the reaction for diphase and triphase systems at 50 °C, where the abscissa denotes the distance from the substrate/liquid interface. For the diphase system, the CO₂ concentration is critically low at the interface (~0.22 mmol L^-1) and gradually increases with distance into the electrolyte, reaching ~0.37 mmol L^-1. This profile arises from the combined effects of reduced CO₂ solubility at 50 °C and continuous reactant consumption at the electrode interface. The diminished bulk concentration establishes a low baseline, while interfacial consumption generates a concentration gradient. Consequently, regions distal to the interface retain relatively higher CO₂ levels due to the attenuated depletion away from the catalytic surface. The entire reaction zone therefore remains in a CO₂-deficient state. In contrast, for the triphase system with direct gas-phase CO₂ supply, the concentration decreases with distance but remains substantially higher than that in the diphase system. CO₂ at the solid-liquid-gas interface (x = 0) approaches saturation (~17.21 mmol L^-1). Therefore, even under the elevated reaction rates induced by high temperature (50 °C), the CO₂ concentration remains above 10 mmol L^-1 throughout the entire reaction zone, sustaining a robust supply consistent with the high catalytic activity measured in the TOF tests.

Furthermore, we investigated the local formate concentration (c_HCOOH) within the triphase microenvironment as a function of temperature [Figure 5C]. While the formate production rate increases with temperature, the simulated formate concentration at the interface decreases systematically from 20 °C to 50 °C. This counterintuitive trend arises because the thermally enhanced diffusion coefficient exceeds the increased formate generation rate, thereby facilitating rapid product evacuation from the interface. This reduction in interfacial accumulation mitigates the inhibitory effect of formate accumulation on the active sites, thereby sustaining the high intrinsic activity of CO₂ reduction. Figure 5C shows the simulated interfacial formate concentration drops sharply from 2.86 mol L^-1 at 20 °C to 0.53 mol L^-1 at 50 °C. This severe depletion at the catalytic interface implies that product removal is no longer the bottleneck; rather, the reaction rate is now dictated by the intrinsic surface kinetics.

Crucially, this simulation prediction is experimentally validated by the flow-rate analysis shown in Supplementary Figure 13. The current response exhibited a distinct evolution with temperature. At 20~30 °C, the steady-state current decreased with increasing flow rate, attributed to shear-induced compression of hydrophobic gas cavities that reduces the effective triple-phase boundary area. At 40 °C, the trend reversed, with current increasing until reaching a plateau above 15 mL min^-1. Most significantly, at 50 °C, the current plateaued at an even lower flow rate of 10 mL min^-1. This progression indicates that mass-transport limitations are progressively alleviated at elevated temperatures. The observation that the current becomes insensitive to flow rates above 10 mL min^-1 at 50 °C aligns with simulations showing a marked depression of interfacial formate concentration. This experimental plateau serves as empirical evidence that thermally accelerated diffusion effectively evacuates interfacial products, thereby decoupling the reaction kinetics from convective supply and underpinning the high efficiency for formate production. Consequently, these results confirm that the triphase architecture maintains a favorable microenvironment at elevated temperatures, characterized by both high reactant supply and low product retention, which underpins the high efficiency for formate production.

CONCLUSIONS

In this work, we constructed a solid-liquid-gas triphase boundary (TPB) architecture to address the inherent conflict between kinetic acceleration and mass-transport limitations in thermally intensified CO₂ electroreduction to formate. A catalyst-integrated, superhydrophobic carbon scaffold was fabricated and deployed as the TPB electrode. The triphase design established a unique solid-liquid-gas microenvironment that decoupled the interfacial CO₂ supply from bulk solubility constraints. Multiphysics simulations demonstrate that the TPB sustains near-saturated CO₂ concentrations at the reaction interface, effectively counteracting the solubility-driven reactant depletion observed in conventional diphase systems, thereby achieving a ~1.36-fold enhancement in apparent TOF at 50 °C relative to 20 °C. Moreover, the triphase system harnesses thermal energy to accelerate formate expulsion, mitigating product accumulation and further boosting the overall reaction efficiency. These results highlight the critical role of interfacial microstructure engineering in reconciling mass-transport constraints with thermal kinetic enhancement for selective CO₂ electroreduction.