Superbasic CO₂-concentrating layer-protected nickel catalyst for solar CH₄ synthesis via direct air capture

Abstract

Although direct air capture technology shows promise for atmospheric CO₂ reduction, it is hindered by the energy-intensive CO₂ concentration processes and unresolved long-term storage risks. As an alternative approach, direct conversion of atmospheric CO₂ into solar fuels could simultaneously achieve carbon neutrality and energy storage, yet existing conversion technologies predominantly require high-concentration CO₂ streams. Herein, we demonstrate a nickel-encapsulated mesoporous nitrogen-doped carbon (NC) architecture that enables integrated CO₂ capture from air and CH₄ production via in situ catalysis of the captured CO₂ with H₂ under solar irradiation. The engineered mesoporous NC framework with superbasic sites achieves exceptional CO₂ capture capacity (55 cm³ g⁻¹) and ultrafast adsorption–desorption kinetics (equilibrium attained in ∼1 min) under ambient conditions. The Ni nanoparticles and NC layers function as tandem catalytic sites for CH₄ production, where photogenerated electrons drive H₂ dissociation on Ni sites, while the adsorbed CO₂ on NC undergoes photothermal reduction to CH₄ by the spilled hydrogen. This mechanism enables a record CH₄ production rate of 339 mmol g⁻¹ h⁻¹ (nearly identical with that obtained using pure CO₂), with perfect selectivity through atmospheric CO₂ conversion. Furthermore, the hydrophobic NC overlayers effectively prevent Ni sintering via physical confinement effects and inhibit oxidative deactivation by dynamically scavenging the H₂O byproduct, enabling the catalyst to maintain stability for over 100 cycles of atmospheric CO₂ capture and conversion. Our temporal-decoupling strategy for converting atmospheric CO₂ eliminates oxygen interference in ambient air and energy-intensive concentration steps, thereby establishing an innovative paradigm for producing carbon-neutral fuels.

---

Broader context

The solar-powered conversion of atmospheric CO₂ is a promising approach for the production of truly carbon-neutral fuels and mitigation of global warming. Over the past decades, tremendous CO₂ utilization technologies predominantly required a concentrated CO₂ feed or high temperature. However, the efficient conversion of CO₂ in air has been rarely achieved due to the difficulty in adsorption and the activation of ultralow concentrations of CO₂ (∼0.04%). Herein, we developed a catalyst that synergistically integrates atmospheric CO₂ capture with in situ conversion and achieved CH₄ formation with near-perfect selectivity under solar irradiation. The CO₂-enriched Ni@NC-600 system therefore achieves a catalytic enhancement of approximately 100–30 000 times higher than traditional approaches. Notably, the catalyst also achieves exceptional CO₂ capture capacity (55 cm³ g⁻¹) and ultrafast adsorption–desorption kinetics (equilibrium attained in ∼1 min) under ambient conditions. Furthermore, the hydrophobic NC overlayers effectively prevent Ni sintering via physical confinement effects and inhibit oxidative deactivation through dynamically scavenging the H₂O byproduct, enabling the catalyst to maintain stability for over 100 cycles of atmospheric CO₂ capture and conversion. This temporal-decoupling strategy establishes a new framework for solar-powered carbon-neutral fuel production without atmospheric CO₂ separation constraints. Significantly, this work pioneers a solar-driven technology for atmospheric CO₂-to-CH₄ conversion, offering a transformative solution to the critical energy-environment nexus between climate change mitigation and renewable energy storage.

Introduction

The continuous high consumption of fossil fuels has caused annual atmospheric CO₂ emissions of 4.4 Gt,^1,2 intensifying the urgency for carbon removal technologies. Direct air capture (DAC) offers a pathway to mitigate climate change by extracting ambient CO₂, yet its widespread adoption is hindered by energy-intensive CO₂ concentration processes and expensive collection systems.^3,4 Although geological sequestration of captured CO₂ is technically feasible, unresolved concerns about its environmental risks and long-term storage integrity persist.⁴

Direct catalytic conversion of atmospheric CO₂ presents an alternative strategy for both removing excess CO₂ from air and producing fuel. However, this process typically demonstrates reaction rates confined to the micromolar regime (µmol g⁻¹ h⁻¹) due to the ultralow atmospheric CO₂ concentration (0.04%) and sluggish mass-transport kinetics.^5–9 Furthermore, the presence of O₂ in air is generally considered as a threat to CO₂ reduction.⁴ Therefore, catalytic CO₂ reduction is currently based almost exclusively on the use of pure or concentrated CO₂ as the feed gas,^1,10–12 necessitating energy-intensive pre-concentration steps for atmospheric CO₂ (440–600 kJ mol⁻¹, including CO₂ adsorption, adsorbent regeneration, and the movement of large quantities of air).¹³ The massive energy input required for the CO₂ concentration and conversion processes can lead to additional CO₂ emissions and increased costs, thus considerably reducing the benefits of catalytic CO₂ conversion.

To address these challenges, integrated CO₂ capture and utilization has emerged as a transformative strategy that enables the in situ catalytic conversion of captured CO₂ into value-added chemicals. This dual-functional approach eliminates energy-intensive CO₂ concentration and transportation steps compared to conventional decoupled systems. Recently, Reisner and co-authors developed a dual-bed flow reactor for gas-phase direct air capture and utilization.⁴ This system first captures atmospheric CO₂ using a solid silica-amine adsorbent, releasing CO₂-free air. Concentrated light then simultaneously releases the captured CO₂ and converts it directly into syngas (CO + H₂) over a Co-based molecular-semiconductor photocatalyst. Although this dual-bed flow reactor can operate without requiring high temperature or pressure, this system is complex with poor infrared-light (comprises ∼50% of the solar spectrum) utilization and slow capture kinetics (12 h). An alternative approach is to develop dual-functional light-driven catalysts with high capacity for capturing atmospheric CO₂. When powered by renewable energy such as sunlight, such catalysts can produce real carbon-neutral fuels.¹⁴ Despite their potential, dual-functional catalysts remain scarce due to the intrinsic challenge of integrating multifunctional components within a single material.^15–18

For dual-functional catalysts, the strongly adsorbed CO₂ covers the active sites of the catalysts, thus impeding H₂ adsorption and activation and subsequent CO₂ hydrogenation.¹⁹ Therefore, how to balance CO₂ adsorption and H₂ activation is a primary challenge for atmospheric CO₂ hydrogenation on a dual-functional catalyst. Developing catalysts with isolated active sites can provide an effective approach for reducing the co-adsorption coverage effects of CO₂ and H₂ in the hydrogenation reaction. We recently developed a platinum-loaded nickel metal–organic framework (Pt/Ni-MOF) that enables simultaneous CO₂ capture from the atmosphere at the Ni sites and photocatalytic H₂ activated at the Pt sites.² However, Pt/Ni-MOF operates only under near-infrared light (≤940 nm), leaving most of the infrared light (accounting for about 50% of the solar energy) unutilized. Moreover, it requires an external heat source (160 °C) for CO₂ activation, which limits its practical application. Critically, developing dual-functional catalysts with isolated active sites that synergize high atmospheric CO₂ capture capacity, rapid sorption kinetics, and high activity driven by full solar spectrum remains a challenge.

Beyond the conventional photocatalysis that is dependent on photoinduced charge carriers, solar energy can drive CO₂ reduction not only through the isolated photothermal effect but also via synergistic photochemical–photothermal activation pathways, offering enhanced efficiency and tailored reaction control. The photothermal catalysis operates via synergistic coupling of photon-to-heat conversion and thermally activated CO₂ reduction, providing an opportunity for converting underutilized infrared photons in traditional photocatalysis. For example, a series of Ni-,²⁰ Co-,²¹ Cu-,²² Fe-,²³ Ir-,²⁴ and Ru²⁵-based catalysts have recently been reported for CO₂ reduction driven by light-induced heating. Under light irradiation, the metal nanoparticles (NPs) can generate the so-called “hot electrons” by localized surface plasmon resonance (LSPR), which can catalyse CO₂ conversion reactions and/or the localized heating of nanoparticles from electron-lattice collisions. By coupling the photothermal and photochemical effects, the metal NPs effects can achieve an outstanding enhancement of the catalytic CO₂ reduction activity. Recently, catalytic methanation of CO₂ has been achieved over Ru@Ni₂V₂O₇²⁶ and Au/Ce_0.95Ru_0.05O₂,²⁷ which effectively converts high-energy photons into electronic excitations while transforming low-energy infrared photons into localized heat, demonstrating a unique advantage in harnessing the full solar spectrum. Consequently, this synergistic photochemical–photothermal activation enables efficient CO₂ reduction at lower temperatures²⁸ and reduced activation barriers compared to thermocatalysis,^26,27,29,30 while delivering much higher activity compared to single-mode photocatalytic systems.

Beyond metal-based photothermal catalysts, carbon-coated metal NPs have emerged as promising candidates owing to their broadband light absorption, high photothermal conversion efficiency, large surface area, and suitable porosity.²¹ The application of a carbon coating to metal sites is an effective strategy for mitigating metal oxidation by the H₂O formed during CO₂ hydrogenation, thereby ensuring long-term stability under reaction conditions. Such a core–shell structure can also provide two distinct active sites for CO₂ adsorption (on the carbon layer) and H₂ activation (on the metal site). Despite these advantages, the existing systems still fail to address the critical challenge of reducing ultralow-concentration atmospheric CO₂ (0.04%) under full solar spectrum irradiation.

Herein, we develop a Ni-encapsulated mesoporous nitrogen-doped carbon (NC) architecture that synergistically combines atmospheric CO₂ capture with solar-driven catalytic conversion. This design achieves exceptional CO₂ capture kinetics (55 cm³ g⁻¹ capacity, about 1 min equilibrium) and enables direct methane synthesis from captured CO₂ under sunlight irradiation. The CH₄ production occurs via a dual-site mechanism: the NC sites adsorb atmospheric CO₂, while the Ni sites activate H₂. Under ultraviolet light irradiation, H₂ dissociates into H atoms. These atoms then spill over to the CO₂ adsorption sites, where they facilitate CH₄ generation via photothermal reduction. Under synergetic photochemical–photothermal activation, the system attains a record CH₄ production rate of 339 mmol g⁻¹ h⁻¹ with perfect selectivity, matching pure CO₂ feedstock performance while eliminating energy-intensive concentration requirements. Furthermore, the catalyst maintains 100-cycle operational stability, overcoming a critical durability bottleneck in direct atmospheric CO₂ conversion systems.

Results and discussion

Synthesis and characterization of the Ni@NC catalysts

The two-step process for synthesizing the Ni@NC catalyst is schematically depicted in Fig. 1A. First, Ni-MOF was synthesized according to a previously reported method.² Thermogravimetric analysis of Ni-MOF in nitrogen gas indicated that decomposition occurred above 500 °C (Fig. S1). Therefore, Ni-MOF was pyrolyzed in Ar gas at temperatures of 500, 600 and 700 °C for 30 minutes to produce Ni@NC-X (X = 500, 600 and 700). In this process, the reduction of the Ni²⁺ species in the MOF was achieved along with the decomposition and carbonization of the organic linker, as observed in previous reports.^31–33 Simultaneously, N was doped in the carbon shells by using coordinated N,N-dimethylformamide (DMF) on the Ni-MOF as the N source under carbonization conditions.

Fig. 1 Structural characterization of the Ni@NC-X (X = 500, 600 and 700) catalysts. (A) Schematic of the two-step process for the formation of Ni@NC-X (X = 500, 600 and 700) catalysts. (B) SEM image of the Ni@NC-600 catalyst. (C) HRTEM image of Ni@NC-600 showing the carbon shells and encapsulated metal NPs. Enlarged image shows the (111) interplanar spacing of Ni and (002) interplanar spacing of C. (D) XRD patterns of the Ni@NC-X (X = 500, 600 and 700) catalysts. (E) Raman spectra of the Ni@NC-X (X = 500, 600 and 700) catalysts. (F) Pore size distribution plots of the Ni@NC-X (X = 500, 600 and 700) catalysts. (G) HAADF-STEM image and the corresponding elemental maps of C, Ni and N. High-resolution XPS spectra of (H) Ni 2p and (I) N 1s for the Ni@NC-X (X = 500, 600 and 700) catalysts.

Scanning electron microscopy (SEM) images of Ni@NC-X (X = 500, 600 and 700) are shown in Fig. 1B and Fig. S2; the samples exhibit a distinct prism-shaped morphology, consistent with that of the Ni-MOF precursor (Fig. S3). However, compared to Ni-MOF with a smooth surface, the surfaces of the pyrolyzed samples were relatively rough. The transmission electron microscopy (TEM) images (Fig. S4), high-resolution TEM (HRTEM) images (Fig. 1C) and X-ray diffraction (XRD) patterns (Fig. 1D) indicate that carbon-coated Ni NPs with an average size of approximately 10 nm were produced over the N-doped carbon support during heat treatment at different temperatures (Fig. S4). The structure of the NC layers was also confirmed by Raman spectroscopy (Fig. 1E).

The HRTEM image (Fig. 1C) reveals that the thickness of the NC shell was 1–3 nm and that irregular nanopores were formed among the core–shell Ni@NC NPs. N₂ adsorption–desorption isotherms reveal the mesoporous structure of the pyrolyzed samples (Fig. S5). The pore size distribution ranges from 3 to 12 nm, while the pore volumes of Ni@NC-X (X = 500, 600 and 700) are 0.55, 0.67 and 0.48 cm³ g⁻¹, respectively (Fig. 1F). High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) imaging coupled with energy-dispersive X-ray (EDX) mapping confirmed that C, N and Ni are distributed throughout the NC matrix (Fig. 1G). The Ni contents of Ni@NC-X (X = 500, 600 and 700) determined via inductively coupled plasma-mass spectrometry are almost the same, with values of ∼18.9%, 18.5%, and 18.4%, respectively.

As shown in Fig. 1H, the X-ray photoelectron spectroscopy (XPS) spectrum of Ni 2p reveals that the Ni species exist in the metallic state (at 852.6 and 870.2 eV),^31–33 which is consistent with the results of the HRTEM and XRD analyses. The shoulder peaks in Fig. 1H are attributed to the satellite peaks (at 858.5 and 874.3 eV) and the formation of Ni–C bonding (at 854.6 eV).³³ As shown in Fig. 1I, the N 1s spectrum of Ni@NC-500 exhibits two distinct peaks corresponding to pyridinic–N (398.2 eV) and pyrrolic–N (400.1 eV),³⁴ among which pyrrolic–N (86%) represents the majority. In contrast, peaks attributed to pyridinic–N (398.2 eV) and graphitic–N (401.2 eV) are observed for Ni@NC-600 and Ni@NC-700, respectively.^35,36 The N 1s spectra confirm that the N species doped in the C layers were effectively tuned by adjusting the pyrolysis temperature.

As shown in Fig. 2A, the CO₂ adsorption capacity of Ni@NC-600 (191 cm³ g⁻¹) at 25 °C is greater than those of Ni@NC-500 (135 cm³ g⁻¹) and Ni@NC-700 (49 cm³ g⁻¹). Impressively, Ni@NC-600 exhibits a sharp increase in the low-pressure region, indicating its remarkable capacity to adsorb CO₂ at low concentrations. As shown in Fig. 2B and Fig. S6, Ni@NC-600 demonstrated a superior ability for direct CO₂ capture from air, achieving an adsorption capacity of 55 cm³ g⁻¹ at 25 °C and 52.8 cm³ g⁻¹ at 400 °C under 0.4 mbar, respectively, which is approximately equivalent to the partial pressure of atmospheric CO₂. This CO₂ adsorption capacity is much larger than that of most solid sorbents (Table S1), suggesting the superior activity of Ni@NC-600 for direct capture of CO₂ from air. In N-doped carbon materials, pyridinic–N, as a Lewis base site, has much stronger alkalinity than pyrrolic–N and graphitic–N and is therefore responsible for capturing low-concentration CO₂ molecules and stabilizing *CO₂ intermediates (where * represents the active site on the surface).^37–39 These results are consistent with the calculated adsorption energies of CO₂ on different types of N (Fig. 2C). Furthermore, as shown in Fig. S7, Ni-MOF undergoes solvent replacement for 12 h and 48 h with dichloromethane (CH₂Cl₂) to gradually remove DMF molecules.² Then, the obtained Ni-MOF without DMF was pyrolyzed at 600 °C, which produced the carbon-coated Ni NPs (denoted as Ni@C-600). Negligible CO₂ adsorption capacity and nearly no activity for photothermal or thermal conversion of CO₂ were observed on Ni@C-600 (Fig. S8), indicating that the presence of N in Ni@NC-X (X = 500, 600 and 700) plays a crucial role in CO₂ adsorption.

Fig. 2 Characterization of CO₂ adsorption and surface temperature over the Ni@NC-X catalysts. (A) CO₂ adsorption isotherms for the Ni@NC-X (X = 500, 600 and 700) catalysts at 25 °C. (B) CO₂ uptakes of Ni@NC-X (X = 500, 600 and 700) at 0.4 mbar (pressures relevant to direct air capture). (C) Optimized geometries of CO₂ adsorption and the adsorption free energies (E_ads) for different types of N in the NC layers. From left to right: pyrrolic–N, pyridinic–N and graphitic–N. The adsorption energy of CO₂ on pyridinic–N of Ni@NC-600 (−4.44 eV) is much lower than those on pyrrolic–N of Ni@NC-500 (−2.56 eV) and graphitic–N of Ni@NC-700 (−0.58 eV), indicating that CO₂ adsorption on Ni@NC-600 is more energetically favorable. (D) CO₂-TPD profiles of the Ni@NC-X (X = 500, 600 and 700) catalysts. (E) UV-vis-IR absorption spectra of the Ni@NC-X (X = 500, 600 and 700) catalysts. (F) Temperature over the Ni@NC-X (X = 500, 600 and 700) catalysts obtained by an in situ infrared thermal imager under full-spectrum irradiation with a xenon lamp intensity of 1.8 W cm⁻².

CO₂ temperature-programmed desorption (CO₂-TPD) of Ni@NC-500 catalyst shows a small desorption peak at 505 °C (Fig. 2D), indicating a low ability to adsorb CO₂. For Ni@NC-600, the material displayed two evident CO₂ desorption peaks. The peak at 312 °C is assigned to medium adsorption sites on pyridinic–N, which stems from the weak Lewis basicity of the N lone pair electrons interacting with CO₂ to form the weakly bound surface species.⁴⁰ The higher temperature peak, observed at approximately 600 °C, is attributed to chemisorbed CO₂ molecules. Furthermore, as shown in Fig. S9, Ni@NC-600 demonstrates a significant capacity for CO₂ adsorption, even at the elevated temperature of 400 °C. This phenomenon can be explained by the electron-withdrawing inductive effect of pyridinic–N, which substantially alters the electron density of its neighboring carbon atoms.³⁷ This result confirms that the pyridinic–N in Ni@NC-600 results in the formation of superbasic sites on the NC layers.⁴¹ However, for Ni@NC-700, only a small CO₂ desorption peak at 195 °C is observed, which may be induced by physical adsorption of CO₂. Although the CO₂ adsorbed on the sample prepared at 700 °C desorbs at 195 °C, the desorbed CO₂ can still react with hydrogen on the catalyst surface due to the confined nature of the reactor. Therefore, this sample retains a certain level of methanation activity even when the photothermal temperature reaches 400 °C. However, the CH₄ production rate remains low (0.06 mmol g⁻¹ h⁻¹), which is attributed to the low concentration of released CO₂ in the gas phase. This observation further confirms the important role of strong CO₂ adsorption in enhancing its conversion performance. Compared to the other two samples, Ni@NC-600 displays a much stronger CO₂ desorption peak, suggesting that Ni@NC-600 has a much higher CO₂ adsorption capacity. These results suggest that the pyridinic–N species promote the generation of solid superbasic sites,⁴² which characteristically adsorb and activate CO₂, especially for low concentrations of CO₂.

As shown in Fig. 2E, the Ni@NC-X (X = 500, 600 and 700) samples exhibit broad light absorption from 200 to 2500 nm, implying their potential for high solar energy utilization efficiency. This characteristic is essential for converting absorbed solar energy to simultaneously stimulate photochemical and thermochemical reactions. As depicted in Fig. 2F, the temperatures of the catalysts (measured in situ with a thermal imaging camera) increase rapidly and reach the equilibrium temperature (∼400 °C) within the first minute under 300 W Xe lamp irradiation. To gain further insight into the role of carbon coating, finite-difference time-domain (FDTD) simulations (Fig. S10) were performed. The results reveal that the NC-encapsulated structure of Ni@NC-600 triples the electric field intensity compared to that of the bare metallic Ni nanoparticles under resonant photon excitation. This substantial enhancement of the localized surface plasmon resonance (LSPR) intensity significantly promotes the plasmon–photon coupling effect of the metallic nanoparticles,⁴³ thereby efficiently converting light into heat to locally elevate the catalyst's temperature for photothermal CO₂ conversion (Fig. S11). Considering the ability of Ni@NC-600 to capture trace CO₂ (Fig. 2B), we carried out solar-driven CO₂ conversion by using atmospheric-level CO₂ (0.04%) as the feed gas in the next experiment.

Direct capture and conversion of atmospheric CO₂ on Ni@NC-600

Ni@NC-600 demonstrates exceptional CO₂ capture performance, achieving an uptake capacity of 55 cm³ g⁻¹ at 25 °C and 0.4 mbar (Fig. 2B) with pronounced selectivity for CO₂ over O₂/N₂ in ambient air (CO₂/N₂ selectivity = 41; CO₂/O₂ selectivity = 36; Fig. S12). This dual functionality prompted its application as an integrated capture-conversion catalyst in a batch reactor equipped with a top quartz window for light irradiation (Fig. S13). Under Xe lamp irradiation, the catalyst surface temperature was stabilized at approximately 400 °C across the entire sample area. This uniform temperature profile confirms the spatial homogeneity of the incident light (Fig. S14).

Initially, the CO₂ conversion performance of Ni@NC-600 was evaluated under irradiation intensities ranging from 1 to 21 suns. The temperatures of the Ni@NC-600 sample increase from 25.6 °C (the ambient temperature) to maximum values between 93.3 and 419.8 °C within 1 minute, depending on the used Xe lamp with different intensities (Fig. S15). As shown in Fig. S16, the highest CH₄ formation rate was achieved at 18 suns (399.6 °C), and similar activity is exhibited when the irradiation intensity is further increased to 21 suns (419.8 °C). This result proves that the photothermal effect of the Xe lamp play a vital role on the CO₂ reduction on Ni@NC-600. Under a reaction temperature of approximately 400 °C, the adsorbed CO₂ is effectively and nearly completely converted, and further increasing the temperature beyond this point does not improve the conversion but instead decreases the overall energy efficiency. Moreover, the selectivity toward CH₄ remained nearly unchanged (close to 100%) across all irradiation intensities (Fig. S16). Therefore, from an energy-efficiency perspective, an intensity of 18 suns was selected for all subsequent CO₂ conversion experiments.

Under light illumination, the maximum CH₄ formation (∼100% selectivity) on Ni@NC-600 in a H₂ : Ar gas mixture (4 : 1 ratio) reaches 5.6 mmol g⁻¹ in 1 min (Fig. 3A). Notably, this high activity was maintained even under humid conditions of 50% and 75% RH (Fig. S17). Moreover, O 1s XPS analysis revealed no detectable changes in the oxygen species following CO₂ capture-conversion across varying humidity levels, further confirming the stability of the Ni@NC-600 surface (Fig. S18). The superior performance of this photothermal catalyst surpasses that of many previously reported photocatalytic systems for CO₂ methanation around 400 °C (Table S2). As shown in Fig. 3A, no decrease in CH₄ generation was observed after 100 CO₂ capture-convention cycles, demonstrating the unprecedented stability for the atmospheric CO₂ utilization systems based on Ni@NC-600. Moreover, XRD, SEM, TEM, and XPS analysis revealed negligible changes in the Ni@NC-600 sample after 100 cycles (Fig. S19). Online gas chromatography (GC) analysis (Fig. S20) and an isotope labelling experiment (Fig. S21) verified the high selectivity of CH₄, further confirming the stability of Ni@NC-600. The excellent stability of Ni@NC-600 can be attributed to the existence of NC layers that can suppress the sintering of Ni NPs.

Fig. 3 Direct capture and conversion of atmospheric CO₂ on Ni@NC-600. (A) Time-dependent CH₄ yield for Ni@NC-600 to capture and convert atmospheric CO₂ in a batch reactor under 300 W Xe light irradiation. After being exposed to the air for 1 min, the CH₄ yield was regenerated. (B) The kinetics of Ni@NC-600 for directly capturing CO₂ from the atmosphere (monitored by online mass spectrometry). Before this test, the adsorbed CO₂ on Ni@NC-600 was exhausted by performing CO₂ methanation. Then, the reaction system was purged by Ar gas for 10 min to remove the H₂ gas. Finally, the feed gas was quickly switched to clean air for the CO₂ capture kinetics measurements. (C) In situ FT-IR spectra of the concentration of CO₂ during the light-driven CO₂ reduction and CO₂ capture from air processes on Ni@NC-600. (D) CH₄ formation rates of Ni@NC-600 under high purity CO₂ (test condition: H₂ : CO₂ = 4 : 1) and atmospheric-concentration CO₂ (direct air capture, H₂ : Ar = 4 : 1), respectively. (E) Photograph of the reaction system under irradiation from natural sunlight. (F) Catalytic performance of Ni@NC-600 with captured CO₂ under the illumination of a Xe lamp and concentrated sunlight with an equivalent light intensity of 1.8 W cm⁻².

Remarkably, the Ni@NC-600 catalyst achieves CO₂ adsorption–desorption equilibrium within 1 minute after the exhaustion of the captured CO₂ (Fig. 3B), surpassing conventional benchmarks by 30–300 times in equilibration kinetics (Table S3). The kinetics for CO₂ capture on Ni@NC-600 was also studied by in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) spectroscopy. As shown in Fig. 3C, the intensity of the CO₂ absorption peak gradually decreases under light irradiation, and then disappears after 1 min reaction. Upon dark cycling, rapid *CO₂ signal recovery occurs within 1 min of air exposure, which is in accordance with the result shown in Fig. 3B.

As shown in Fig. 3D, the Ni@NC-600 catalyst demonstrates an exceptional ability to convert atmospheric CO₂. In batch reactor tests, it achieved a high CH₄ production rate of 339 mmol g⁻¹ h⁻¹ using CO₂ captured directly from air, a value nearly identical to the 342 mmol g⁻¹ h⁻¹ obtained with high-purity CO₂ (Fig. 3D). This parity confirms that Ni@NC-600 has the remarkable ability to effectively capture and convert highly diluted CO₂ from the atmosphere. Furthermore, Ni@NC-600 exhibits significantly higher CH₄ production activity compared to Ni@NC-500 (0.13 mmol g⁻¹ h⁻¹) and Ni@NC-700 (0.06 mmol g⁻¹ h⁻¹) under identical experimental conditions, which may be attributed to the superior atmospheric CO₂ adsorption capacity of Ni@NC-600 (Fig. 2B). Strikingly, conventional solar-driven catalysts for atmospheric-concentration CO₂ reduction typically exhibit activity levels in the micromolar range (µmol g⁻¹ h⁻¹; Table S4). The CO₂-enriched Ni@NC-600 system therefore achieves a catalytic enhancement of approximately 100–30 000 times higher than that achieved using traditional approaches. These results establish that Ni@NC-600 enables efficient atmospheric CO₂ conversion at ultralow concentrations (∼0.04%). The integrated CO₂ capture-conversion approach circumvents the need for transportation and storage infrastructure, bypasses the energy-intensive regeneration of the capture media, and avoids molecular CO₂ release steps, thereby significantly enhancing the energy efficiency of atmospheric CO₂ conversion processes.

The ultimate goal of artificial photosynthesis is to harness natural solar energy as the exclusive driving force. Given the remarkable performance of the Ni@NC-600 catalyst for light-driven CH₄ production, the catalytic CO₂ methanation reaction was evaluated under natural sunlight utilizing a Fresnel lens as concentrator. The experiment was carried out on May 6th, 2023 from 11 : 30 to 12 : 30 under natural sunlight in Harbin, China (Fig. 3E). Under these conditions, the temperature of the Ni@NC-600 catalyst rapidly increased from ambient condition (29 °C) to 402.2 °C upon continuous irradiation for 1 min (Fig. S22), achieving a remarkable CH₄ production rate of 338.1 mmol g⁻¹ h⁻¹ under concentrated sunlight, comparable to Xe lamp-driven conditions (Fig. 3F). As depicted in Fig. S23, outdoor solar-driven photoreforming maintained 100% CH₄ selectivity throughout the day. To evaluate the practical application of Ni@NC-600, the device based on this catalyst for the large-scale methanation of CO₂ in air is designed and depicted in Fig. S24. Based on this device, we calculated the surface area required to process 1 kg of CO₂. This calculation was based on a cyclic procedure comprising CO₂ adsorption (1 min), N₂ purging (3 min), and photocatalytic conversion under illumination (1 min). Under these conditions, the catalyst exhibited a CO₂ capture capacity of 55 cm³ g⁻¹ and achieved a high CO₂-to-CH₄ conversion efficiency of 96%. When normalized to 8 hours of daily operation, the required catalyst area for methanation of 1 kg atmospheric CO₂ is calculated to be approximately 2.83 m². This performance starkly contrasts with that obtained through natural photosynthesis. For instance, the most productive terrestrial ecosystems, such as tropical forests, exhibit a maximum carbon fixation rate of about 2200 g C m⁻² year⁻¹, equivalent to ∼8074 g CO₂ m⁻² year⁻¹.⁴⁴ On an annualized and per-unit-area basis, our catalyst demonstrates a CO₂ fixation rate that is approximately 16 times greater. Furthermore, a critical advantage of our technology lies in the product outcome: it selectively transforms captured CO₂ into readily usable CH₄ fuel, whereas natural photosynthesis produces recalcitrant biomass that is challenging to utilize efficiently.

The photochemical–photothermal effect on Ni@NC-600 for atmospheric CO₂ methanation

To investigate the electron transfer pathway on Ni@NC-600 under light irradiation, quasi in situ XPS measurements were conducted to examine the charge transfer between the Ni core and carbon shell in Ni@NC-600. Under light irradiation, the Ni 2p spectra show a positive binding energy shift of roughly 0.42 eV (Fig. S25a), and the pyridinic–N peak in the N 1s spectra shows a negative shift of about 0.41 eV (Fig. S25b). These shifts confirm that photogenerated electrons are transferred from the Ni core to the NC surface, which is beneficial for CO₂ photoreduction. Furthermore, the prolonged fluorescence lifetime of Ni@NC-600 (28.2 ns), as shown by the time-resolved PL spectra in Fig. S26, is notably longer than the lifetime observed for Ni/C (14.6 ns), which endows photogenerated carriers with adequate time to migrate to the NC surface and participate in the photocatalytic CO₂ reduction.

To study the effect of light irradiation, we calculated the E_a of Ni@NC-600 for CO₂ methanation using the Arrhenius equation (Fig. 4A) based on the CH₄ formation rates and surface temperatures (T_s) obtained with light irradiation and/or electrical heating. As shown in Fig. 4A, the E_a calculated under full-spectrum irradiation with an intensity of 1.8 W cm⁻² is 23.19 kJ mol⁻¹, which is significantly lower than that in the dark (51.98 kJ mol⁻¹, grey line in Fig. 4A). This finding reveals that CO₂ methanation is kinetically favorable under irradiation, which is a key factor for the greater increase in the reaction rate compared to that of thermocatalysis on Ni@NC-600. This result also proves that the photoelectric (nonthermal) effect plays a critical role in CO₂ methanation over Ni@NC-600, indicating the distinct advantage of light-driven catalysis over thermocatalysis.

Fig. 4 The photochemical–photothermal effect on Ni@NC-600 for atmospheric CO₂ methanation. (A) Arrhenius plots of the apparent activation barriers for CH₄ formation over Ni@NC-600 in the dark and under light irradiation (1.8 W cm⁻²). T_s was determined by an in situ thermal imaging camera to account for light-induced heating. (B) CH₄ formation rate over the Ni@NC-600 catalyst under irradiation with different optical cut-off filters under 300 W Xe lamp illumination and UV light + 400 °C (electrical heating). The test conditions were denoted as 1 to 3, 1: no filter (UV-vis-IR light), 2: λ ≥ 420 nm (vis-IR light) and 3: λ < 420 nm (UV light). Error bars represent the standard deviation based on three measurements. (C) Comparison of the CH₄ formation rates over the Ni@NC-600 and NH₄Cl treated catalysts. (D) H₂-TPD profiles of the Ni@NC-600 and NH₄Cl treated catalysts. (E) H₂ decomposition activities of the Ni@NC-600 and NH₄Cl treated catalysts as a function of time under full-spectrum irradiation. (F) H₂ decomposition activities of Ni@NC-600 as a function of time under light irradiation with different wavebands (the irradiation intensity of the Xe lamp was fixed at 1.8 W cm⁻², regardless of the utilization of different filters). (G) Photocurrent response curves over Ni@NC-600 under irradiation with different optical cut-off filters under 300 W Xe lamp illumination. (H) H₂ decomposition activities of Ni NPs and Ni@C and Ni/C under irradiation by a full-spectrum Xe lamp at 1.8 W cm⁻². (I) Photographs of the mixtures of WO₃ and Ni@NC-600 under UV, vis-IR, and full-spectrum Xe lamp irradiation in H₂ : Ar (with a 1 : 20 ratio) for 30 min.

To clarify the contribution of light to CH₄ formation kinetics, the E_a values in the ultraviolet (UV) and visible-infrared (vis-IR) ranges at a fixed intensity of the full-spectrum (1.8 W cm⁻²) were measured.²⁹ As shown in Fig. 4A, the E_a under vis-IR light was 49.62 kJ mol⁻¹ (orange line), which is similar to that under thermocatalysis (51.98 kJ mol⁻¹, gray line). In sharp contrast, UV light illumination causes a reduction in E_a to 23.64 kJ mol⁻¹ (purple line), which is similar to that under full-spectrum irradiation (23.19 kJ mol⁻¹, blue line). The wavelength dependence of E_a indicates that UV and vis-IR light illumination are responsible for electronic and thermal excitation, respectively, in the CO₂ methanation reaction on Ni@NC-600.²⁹ This result is also proved by the comparable CO₂ conversion activities of Ni@NC-600 under full-spectrum illumination and UV light + 400 °C (electrical heating) conditions (Fig. 4B). Furthermore, the CH₄ formation rate on Ni@NC-600 driven by UV or vis-IR light is much lower than that driven by full-spectrum irradiation (Fig. 4B), implying that the CO₂ reduction kinetics over Ni@NC-600 is determined primarily by the synergistic effect of photochemical and photothermal processes.

Generally, CO₂ activation is considered as the rate-determining step in the methanation reaction.⁴⁵ Considering the E_a values in different conditions (Fig. 4A), we can speculate that photogenerated electrons under UV light play a critical role in activating CO₂. In addition to CO₂ adsorption/activation, H₂ activation is a key factor determining the CO₂ methanation performance. In the CO₂ methanation reaction, Ni NPs are usually regarded as the active sites for dissociating H₂ molecules (H₂ → 2H*).^46–48 To investigate the role of Ni NPs in CO₂ methanation, the Ni@NC-600 catalyst was treated with ammonium chloride (NH₄Cl) at 350 °C for 6 h and 12 h to gradually remove the embedded Ni NPs.⁴⁹ ICP analysis confirmed that the Ni content in the Ni@NC-600 catalyst was 8.8 wt% and 0.27 wt% after treatment times of 6 h and 12 h, respectively (Fig. S27). Under the same conditions, the light-powered CO₂ reduction performance of Ni@NC-600 was approximately 2.2 and 33.6 times higher than that of the catalysts treated with NH₄Cl for 6 h and 12 h, respectively (Fig. 4C). This confirms the crucial role of the Ni content in CO₂ methanation. Combined H₂-TPD and decomposition analyses of Ni@NC-600 and the NH₄Cl-treated catalysts (Fig. 4D and E) reveal that the H₂ activation step on the catalyst is governed by its Ni content, a conclusion consistent with the observed methanation performance in Fig. 4C. XRD, SEM, TEM, and XPS analyses revealed that the NH₄Cl treatment selectively removed the encapsulated Ni NPs while preserving the outer NC layers (Fig. S28), a finding consistent with prior reports that this etching process does not alter the porous structure of the carbon support.⁴⁹ Under this constant pore-size condition, the H₂ dissociation activity showed a direct correlation with the amount of remaining Ni NPs (Fig. 4E), thus playing a decisive role in the CO₂ methanation performance of Ni@NC-600 (Fig. 4C).

The wavelength-dependent H₂ decomposition abilities of Ni@NC-600 suggest that the H₂ dissociation on Ni sites is driven mainly by UV light illumination, whereas vis-IR light has a negligible effect (Fig. 4F). This result is in good agreement with the wavelength-dependent photocurrent measurement of Ni@NC-600 (Fig. 4G). Furthermore, as shown in Fig. 4H, the hydrogen dissociation activities of Ni nanoparticles (Ni NPs, Fig. S29), Ni@C (Ni NPs encapsulated in carbon, Fig. S29), and Ni/C (Ni NPs loaded on carbon support, Fig. S30) with an average size of approximately 10 nm are independent of the carbon support under Xe lamp irradiation. This finding indicates that hydrogen dissociation is governed by the Ni NPs themselves, not by their loading configuration or the support material. Therefore, to isolate the size effect of Ni on CO₂ methanation, we investigated the correlation between the size of bare Ni NPs and their H₂ dissociation activity. Ni nanoparticles with mean sizes of 5.6, 10.2, and 20.8 nm were synthesized via the acid–base-mediated alcohol reduction method (Fig. S31).⁵⁰ Ni NPs with a middle size of approximately 10.2 nm exhibit optimal activity for H₂ dissociation (Fig. S32). Smaller Ni NPs are rich in low-coordinated edge/corner sites that facilitate H₂ adsorption, while larger NPs offer more high-coordinated terrace sites that are active for H–H bond cleavage.⁵¹ Hence, Ni NPs of intermediate size (10.2 nm) achieve peak activity by providing a balanced ensemble of sites for both essential steps—adsorption and dissociation.

Dissociated H atoms transfer from the Ni core to the NC shell to induce the CO₂ methanation reaction because CO₂ molecules are adsorbed mainly on the NC shell of Ni@NC-600. The diffusion of H atoms can be improved by the photothermal heating effect of vis-IR light, which was proven by the adoption of a WO₃ chromogenic agent. As shown in Fig. 4I, the colours of Mixtures 1 and 2 were nearly unchanged under only UV or vis-IR light irradiation, suggesting that H spill over did not occur under those conditions. Considering that the UV light of the Xe-lamp dominates the H₂ dissociation on Ni NPs (Fig. 4F), the unchanged colour of Mixture 1 indicates that the H* spill over can be hardly induced with UV light input alone. The unchanged colour of Mixture 2 is understandable because H₂ cannot be decomposed on Ni@NC-600 under vis-IR light irradiation (Fig. 4F). In sharp contrast, the colour of Mixture 3 under full spectrum irradiation turns dark, indicating that the vis-IR light-induced heat input can drive the diffusion of the dissociated H* species by UV light to react with WO₃ to form H_xWO₃. Based on the observations in Fig. 4I, we can conclude that the photothermal effect of vis-IR light helps the dissociated H* species on Ni cores to spill toward the NC shell on Ni@NC-600 for reducing the adsorbed CO₂. This result is consistent with the fact that the CO₂ methanation on Ni@NC-600 was induced by the synergistic effect of photochemical and photothermal processes. Therefore, the enhanced kinetics was achieved on Ni@NC-600 under combined thermal and electronic excitation. Furthermore, the isolated active sites limit the coadsorption of the H₂ and CO₂ reactants, which also boosts the overall kinetics of solar-driven CH₄ formation on Ni@NC-600.

The roles of hydrophobicity NC layers on the stability of CO₂ methanation

Ni@NC-600 has excellent hydrophobicity with a contact angle of 133.3° (Fig. 5A), which might be caused by its high surface roughness (Fig. 1B).⁵² Hydrophobicity is conducive to preventing direct contact with water and protecting catalysts from corrosion because H₂O severely affects the deactivation of Ni and other metal catalysts for hydrogenation reactions by changing the surface structure of the catalyst or blocking active sites.^53–57

Fig. 5 Catalytic stability for solar-driven CO₂ methanation over Ni@NC-600. (A) Water-droplet contact angle test of the Ni@NC-600 catalyst. (B) Top: Schematic of the photothermal reactor. Bottom: Infrared thermal image showing the temperature of the batch reactor and Ni@NC-600 catalyst under Xe lamp irradiation. The insert yellow square represents the average temperature of the corresponding region. (C) Scheme showing CO₂ enrichment in the nanopores from the photothermal reactor, and removal of the H₂O byproducts by the local photothermal heating effect. Moreover, the hydrophobic surface prevents water readsorption. (D) Enlarged in situ FTIR spectra at 4000–3000 cm⁻¹ over the Ni@NC-600 catalyst during the CO₂ methanation reaction (4 h). (E) Data showing the influence of water on CO₂ methanation with the Ni@NC-600 catalyst. Reaction conditions: 10 mg of the evenly dispersed Ni@NC-600 catalyst was used. The flow rates of CO₂ (400 ppm, diluted with Ar), 4 vol% H₂ with a space velocity of ∼60 000 mL g_cat h⁻¹ at ambient pressure. The water feed rate (via infusion needle) was ∼0.01 mL min⁻¹. (F) Enlarged FTIR spectra at 4000–3000 cm⁻¹ over the Ni@NC-600 catalyst before and after the CO₂ methanation reaction with water (12 h).

The infrared image clearly shows that under Xe lamp irradiation (1.8 W cm⁻²), the temperature on the surface of Ni@NC-600 reached ∼400 °C due to the localized photothermal effect, while the temperature of the reaction chamber was only 65 °C (Fig. 5B). This large spatial temperature difference (335 °C) provides a strong driving force for pumping water byproducts out of the nanopores and promoting its removal from the Ni@NC-600 surface (denoted as *H₂O ⇌ * + H₂O), which facilitates the concentration and localization of the produced H₂O in the low-temperature zone of the reactor (Fig. 5B). Unfortunately, this type of temperature gradient cannot be induced in traditional thermocatalytic systems with a uniform heat distribution. Moreover, the hydrophobic surface helps prevent the readsorption of H₂O on Ni@NC-600 and recover surface active sites (Fig. 5C).⁵⁵ As a result of the combined effects of local photothermal heating and hydrophobicity, the local H₂O concentration on the Ni@NC-600 surface should be significantly decreased. The Ni@NC-600 samples before and after 4 h of photoreaction (Fig. 5D) show similar FT-IR spectra at 4000–3000 cm⁻¹ (belong to –OH groups) after the CO₂ methanation reaction, indicating that the produced water can be readily removed once it is formed on the catalyst.

To further understand this mechanism, we performed a solar-powered CO₂ methanation reaction on Ni@NC-600 in a flow reactor with a volume of 100 mL by adding water at a rate of 0.01 mL min⁻¹, which is approximately 22 times greater than the rate of water formation in the CO₂ methanation reaction if all the CO₂ is converted into CH₄. During the 12-hour test, no decline in CH₄ formation was detected compared with that in the control experiment without the addition of H₂O (Fig. 5E). After this reaction, no obvious changes in the shape of the FT-IR spectrum of Ni@NC-600 were observed in the wavenumber range of 4000–3000 cm⁻¹ (Fig. 5F). These results indicate that the removed H₂O byproducts can be readily confined to the low-temperature zone of the reactor and cannot reach the catalyst surface; therefore, the local H₂O concentration on Ni@NC-600 can be greatly reduced during the solar-driven CO₂ methanation reaction.

The mechanism for solar-driven catalytic CO₂ methanation on Ni@NC-600

To study the catalytic mechanism, in situ DRIFTS measurement was employed to monitor the formed intermediates on Ni@NC-600 during atmospheric-level CO₂ methanation powered by solar light. As shown in Fig. 6A, several new infrared peaks were detected under light irradiation. The peaks around 1253, 1446 and 1509 and 1562 cm⁻¹ are attributed to the stretching of active *CO₂⁻, *HCO₃⁻, m-CO₃²⁻ (monodentate carbonate), and b-CO₃²⁻ (bidentate carbonate),⁵⁴ respectively, indicating that CO₂ was efficiently absorbed and activated on the surface of Ni@NC-600. The peaks at 1649 cm⁻¹ and 2101 cm⁻¹ can be ascribed to the formation of the COOH* and *CO species,^58–61 respectively. Therefore, we speculate that CO* was formed by the dehydration of COOH* due to the fact that no peaks belonging to the formate species were observed in Fig. 5B. Furthermore, the peaks at 1838, 1760, and 1098 cm⁻¹ correspond to the formation of the *CHO, *CH₂O, and *CH₃O species, respectively.^59,61 Based on the in situ DRIFTS test, we propose that the solar-driven CO₂ methanation on Ni@NC-600 was achieved by a stepwise hydrogenation process.

Fig. 6 Catalytic mechanism for solar-driven CO₂ methanation over Ni@NC-600. (A) Time-dependent in situ FTIR spectra for CO₂ conversion on Ni@NC-600. (B) Calculated Gibbs free energy of the CO₂ methanation reaction over Ni@NC-600. (C) The proposed pathway of CO₂ capture and methanation on the Ni@NC-600 catalyst under full-spectrum irradiation.

Next, we carried out density functional theory (DFT) calculations of the CO₂ methanation on Ni@NC-600 based on the in situ DRIFTS results. As shown in Fig. 6B, the conversion of CO₂ to COOH* is the rate-determining step of the CO₂ methanation on the catalysts. The formation of COOH* on pyridinic–N of Ni@NC-600 had a free-energy barrier of 1.35 eV, which is much lower than that on pyrrolic–N of Ni@NC-500 (1.66 eV) and graphitic–N of Ni@NC-700 (1.84 eV). This result indicates that the types of nitrogen in the NC layers primarily influence the rate-determining step of the CO₂ methanation. The DFT study is in good agreement with the experimental observation on the CO₂ adsorption and methanation properties of the three catalysts. Moreover, the energy barriers of *CO hydrogenation on the three catalysts are exothermic (Fig. 6B), which are much lower than that for CO desorption (*CO → * + CO). This phenomenon indicates that *CO is energetically favourable to hydrogenation, further supporting the catalyst's selectivity for CH₄. Based on this mechanistic insight, further product selectivity modulation can be achieved by precisely tuning the chemisorption energy of crucial reaction intermediates. Specifically, a weak adsorption energy of *CO will favour CO product formation.⁶² Based on the above discussions, the proposed pathway for CO₂ methanation on Ni@NC-600 is illustrated in Fig. 6C.

Conclusions

In conclusion, we developed a Ni@NC-600 core–shell catalyst via the pyrolysis of a Ni-MOF precursor for light-driven methanation of atmospheric CO₂. The Ni core is mainly responsible for H₂ dissociation when excited by UV light under a Xe lamp. The dissociated H₂ spills to the NC shell, driven by light-induced heating, and enables methanation of the adsorbed CO₂via the coupled photo/photothermal effect. In this process, the isolated active sites limit the coadsorption of the H₂ and CO₂ reactants, and the NC shell plays four critical roles. First, the mesoporous structure and super-basicity of NC allow easy adsorption of the diluted CO₂ in the nanopores of Ni@NC-600. Second, the local photothermal heating effect and the hydrophobic properties of NC facilitate the removal of H₂O byproducts in the CO₂ methanation reaction. Third, the photothermal heating effect of NC energizes the active sites to boost CH₄ production with the help of photogenerated electrons from Ni, which reduce the E_a of the CO₂ methanation reaction. Fourth, the hydrophobic NC shell not only inhibit the sintering of the Ni catalyst but also prevent it from reacting with H₂O byproducts during the methanation reaction. As a result, a 339 mmol g⁻¹ h⁻¹ CH₄ production rate with ∼100% CH₄ selectivity is achieved on Ni@NC-600 under light irradiation for over 100 capture-convention cycles, representing a promising approach for producing truly carbon-neutral fuels.

Author contributions

Y. L. and W. M. conceived the project and designed the experiments. S. G., W. M., and H. Y. synthesized the catalysts and conducted the structure analysis and photocatalytic CO₂ reduction tests. H. Y. performed the TPD measurements. J. Z. performed the theoretical calculations. J. P., J. S., and Y. L. co-wrote the paper. All authors commented on the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article, titled “Superbase CO₂-concentrating layers protected nickel catalyst for solar CH4 synthesis via direct air capture,” have been included as part of the supplementary information (SI). The supplementary information file includes detailed experimental procedures, catalysts characterization, catalytic performance, density functional theory (DFT) calculations and catalytic comparison tables. See DOI: https://doi.org/10.1039/d5ee06482a.

Acknowledgements

This work is jointly supported by the National Natural Science Foundation of China (22372050), the Fundamental Research Funds for the Central Universities (HIT.OCEF.2023012) and the Doctor Research Start Foundation of Shaanxi University of Technology (SLGRCQD038).

Notes and references

1. I. Sullivan, A. Goryachev, I. A. Digdaya, X. Li, H. A. Atwater, D. A. Vermaas and C. Xiang, Nat. Catal., 2021, 4, 952–958.
2. W. Ma, J. Sun, S. Yao, Y. Wang, G. Chen, G. Fan and Y. Li, Angew. Chem., Int. Ed., 2023, 62, e202313784.
3. B. Parkinson, ACS Energy Lett., 2016, 1, 1057–1059.
4. S. Kar, D. Kim, A. Bin Mohamad Annuar, B. B. Sarma, M. Stanton, E. Lam, S. Bhattacharjee, S. Karak, H. F. Greer and E. Reisner, Nat. Energy, 2025, 10, 448–459.
5. J. Ding, P. Du, P. Li, W. Liu, J. Xu, W. Yan, Y. Pan, J. Hu, J. Zhu, Q. Chen, X. Jiao and Y. Xie, Angew. Chem., Int. Ed., 2025, 64, e202414453.
6. P. Du, J. Ding, C. Liu, P. Li, W. Liu, W. Yan, Y. Pan, J. Hu, J. Zhu, X. Li, Q. Chen and X. Jiao, Angew. Chem., Int. Ed., 2025, 64, e202421353.
7. Y. Ma, X. Yi, S. Wang, T. Li, B. Tan, C. Chen, T. Majima, E. R. Waclawik, H. Zhu and J. Wang, Nat. Commun., 2022, 13, 1400.
8. Y. Wei, J. Li, D. Zhao, Y. Zhao, Q. Zhang, L. Gu, J. Wan and D. Wang, CCS Chem., 2024, 6, 3065–3076.
9. X. Wu, Y. Li, G. Zhang, H. Chen, J. Li, K. Wang, Y. Pan, Y. Zhao, Y. Sun and Y. Xie, J. Am. Chem. Soc., 2019, 141, 5267–5274.
10. F. Wang, S. He, H. Chen, B. Wang, L. Zheng, M. Wei, D. G. Evans and X. Duan, J. Am. Chem. Soc., 2016, 138, 6298–6305.
11. M. Ghoussoub, M. Xia, P. N. Duchesne, D. Segal and G. Ozin, Energy Environ. Sci., 2019, 12, 1122–1142.
12. W. Gao, Y. Li, D. Xiao and D. Ma, J. Energy Chem., 2023, 83, 62–78.
13. F. E. Osterloh, ACS Energy Lett., 2016, 1, 1060–1061.
14. R. Schäppi, D. Rutz, F. Dähler, A. Muroyama, P. Haueter, J. Lilliestam, A. Patt, P. Furler and A. Steinfeld, Nature, 2022, 601, 63–68.
15. D. Sengupta, S. Bose, X. Wang, N. M. Schweitzer, C. D. Malliakas, H. Xie, J. Duncan, K. O. Kirlikovali, T. Yildirim and O. K. Farha, J. Am. Chem. Soc., 2024, 146, 27006–27013.
16. Y. Wang, J.-X. Wei, H.-L. Tang, L.-H. Shao, L.-Z. Dong, X.-Y. Chu, Y.-X. Jiang, G.-L. Zhang, F.-M. Zhang and Y.-Q. Lan, Nat. Commun., 2024, 15, 8818.
17. Y. Tian, R. Wang, S. Deng, Y. Tao, W. Dai, Q. Zheng, C. Huang, C. Xie, Q. Zeng, J. Lin and H. Chen, Nano Lett., 2023, 23, 10914–10921.
18. Y. Yang, H.-Y. Zhang, Y. Wang, L.-H. Shao, L. Fang, H. Dong, M. Lu, L.-Z. Dong, Y.-Q. Lan and F.-M. Zhang, Adv. Mater., 2023, 35, 2304170.
19. H. Jia, X. Feng, X. Du, L. Lin, R. Mu and Q. Fu, Angew. Chem., Int. Ed., 2025, 64, e202503319.
20. X. Zhu, H. Zong, C. J. V. Pérez, H. Miao, W. Sun, Z. Yuan, S. Wang, G. Zeng, H. Xu, Z. Jiang and G. A. Ozin, Angew. Chem., Int. Ed., 2023, 62, e202218694.
21. J. Ma, J. Yu, G. Chen, Y. Bai, S. Liu, Y. Hu, M. Al-Mamun, Y. Wang, W. Gong, D. Liu, Y. Li, R. Long, H. Zhao and Y. Xiong, Adv. Mater., 2023, 35, 2302537.
22. Y. Li, X. Bai, D. Yuan, C. Yu, X. San, Y. Guo, L. Zhang and J. Ye, Nat. Commun., 2023, 14, 3171.
23. H. Zhang, T. Wang, J. Wang, H. Liu, T. D. Dao, M. Li, G. Liu, X. Meng, K. Chang, L. Shi, T. Nagao and J. Ye, Adv. Mater., 2016, 28, 3703–3710.
24. Y. Tang, T. Zhao, H. Han, Z. Yang, J. Liu, X. Wen and F. Wang, Adv. Sci., 2023, 10, 2300122.
25. Z. Chen, X. Dong, Z.-X. Sun, X. An, C. Li, S. Liu, J. Shen, C. Wu, J. Wang, Z. Wang, Z. Zhu, Y. Zhou, K. Yu, Y. Ma, J. He, K. Feng, L. He and Z. Hu, ACS Nano, 2024, 18, 19672–19681.
26. Y. Chen, Y. Zhang, G. Fan, L. Song, G. Jia, H. Huang, S. Ouyang, J. Ye, Z. Li and Z. Zou, Joule, 2021, 5, 3235–3251.
27. H. Jiang, L. Wang, H. Kaneko, R. Gu, G. Su, L. Li, J. Zhang, H. Song, F. Zhu, A. Yamaguchi, J. Xu, F. Liu, M. Miyauchi, W. Ding and M. Zhong, Nat. Catal., 2023, 6, 519–530.
28. K. Feng, S. Wang, D. Zhang, L. Wang, Y. Yu, K. Feng, Z. Li, Z. Zhu, C. Li, M. Cai, Z. Wu, N. Kong, B. Yan, J. Zhong, X. Zhang, G. A. Ozin and L. He, Adv. Mater., 2020, 32, 2000014.
29. L. Zhou, D. F. Swearer, C. Zhang, H. Robatjazi, H. Zhao, L. Henderson, L. Dong, P. Christopher, E. A. Carter, P. Nordlander and N. J. Halas, Science, 2018, 362, 69–72.
30. L. Kang, X. Y. Liu, A. Wang, L. Li, Y. Ren, X. Li, X. Pan, Y. Li, X. Zong, H. Liu, A. I. Frenkel and T. Zhang, Angew. Chem., Int. Ed., 2020, 59, 12909–12916.
31. I. S. Khan, D. Mateo, G. Shterk, T. Shoinkhorova, D. Poloneeva, L. Garzón-Tovar and J. Gascon, Angew. Chem., Int. Ed., 2021, 60, 26476–26482.
32. M. Mukoyoshi, H. Kobayashi, K. Kusada, M. Hayashi, T. Yamada, M. Maesato, J. M. Taylor, Y. Kubota, K. Kato, M. Takata, T. Yamamoto, S. Matsumura and H. Kitagawa, Chem. Commun., 2015, 51, 12463–12466.
33. X. Lin, S. Wang, W. Tu, H. Wang, Y. Hou, W. Dai and R. Xu, ACS Appl. Energy Mater., 2019, 2, 7670–7678.
34. M. Zhang, Y.-G. Wang, W. Chen, J. Dong, L. Zheng, J. Luo, J. Wan, S. Tian, W.-C. Cheong, D. Wang and Y. Li, J. Am. Chem. Soc., 2017, 139, 10976–10979.
35. A. Aijaz, N. Fujiwara and Q. Xu, J. Am. Chem. Soc., 2014, 136, 6790–6793.
36. Z. Zhang, L. Yu, Y. Tu, R. Chen, L. Wu, J. Zhu and D. Deng, Cell Rep. Phys. Sci., 2020, 1, 100145.
37. D. Guo, R. Shibuya, C. Akiba, S. Saji, T. Kondo and J. Nakamura, Science, 2016, 351, 361–365.
38. T. Muthusamy, S. Sethuram Markandaraj and S. Shanmugam, J. Mater. Chem. A, 2022, 10, 6470–6474.
39. Y. Dong, Q. Zhang, Z. Tian, B. Li, W. Yan, S. Wang, K. Jiang, J. Su, C. W. Oloman, E. L. Gyenge, R. Ge, Z. Lu, X. Ji and L. Chen, Adv. Mater., 2020, 32, 2001300.
40. Q. Lu, C. Chen, Q. Di, W. Liu, X. Sun, Y. Tuo, Y. Zhou, Y. Pan, X. Feng, L. Li, D. Chen and J. Zhang, ACS Catal., 2022, 12, 1364–1374.
41. L. B. Sun, J. Yang, J. H. Kou, F. N. Gu, Y. Chun, Y. Wang, J. H. Zhu and Z. G. Zou, Angew. Chem., Int. Ed., 2008, 47, 3418–3421.
42. B. Li, X. Sun and D. Su, Phys. Chem. Chem. Phys., 2015, 17, 6691–6694.
43. J. Ma, X. Mao, C. Hu, X. Wang, W. Gong, D. Liu, R. Long, A. Du, H. Zhao and Y. Xiong, J. Am. Chem. Soc., 2024, 146, 970–978.
44. F. Stuart Chapin, P. A. Matson and P. M. Vitousek, Principles of Terrestrial Ecosystem Ecology, Springer Science & Business Media, 2011, pp. 157–181.
45. C. Vogt, M. Monai, G. J. Kramer and B. M. Weckhuysen, Nat. Catal., 2019, 2, 188–197.
46. L. Ma, R. Ye, Y. Huang, T. R. Reina, X. Wang, C. Li, X. L. Zhang, M. Fan, R. Zhang and J. Liu, Chem. Eng. J., 2022, 446, 137031.
47. P. A. U. Aldana, F. Ocampo, K. Kobl, B. Louis, F. Thibault-Starzyk, M. Daturi, P. Bazin, S. Thomas and A. C. Roger, Catal. Today, 2013, 215, 201–207.
48. V. Golovanova, M. C. Spadaro, J. Arbiol, V. Golovanov, T. T. Rantala, T. Andreu and J. R. Morante, Appl. Catal., B, 2021, 291, 120038.
49. M.-X. Chen, M. Zhu, M. Zuo, S.-Q. Chu, J. Zhang, Y. Wu, H.-W. Liang and X. Feng, Angew. Chem., Int. Ed., 2020, 59, 1627–1633.
50. C. Dong, C. Lian, S. Hu, Z. Deng, J. Gong, M. Li, H. Liu, M. Xing and J. Zhang, Nat. Commun., 2018, 9, 1252.
51. Z. Zou, Y. Shen, C. Chen, W. Li, Y. Zhang, H. Zhang, Z. Yu, H. Zhao and G. Wang, Adv. Funct. Mater., 2025, 35, 2417584.
52. Z. Zhao, L. Duan, Y. Zhao, L. Wang, J. Zhang, F. Bu, Z. Sun, T. Zhang, M. Liu, H. Chen, Y. Yang, K. Lan, Z. Lv, L. Zu, P. Zhang, R. Che, Y. Tang, D. Chao, W. Li and D. Zhao, J. Am. Chem. Soc., 2022, 144, 11767–11777.
53. L. C. Buelens, V. V. Galvita, H. Poelman, C. Detavernier and G. B. Marin, Science, 2016, 354, 449–452.
54. D. Mateo, N. Morlanes, P. Maity, G. Shterk, O. F. Mohammed and J. Gascon, Adv. Funct. Mater., 2021, 31, 2008244.
55. W. Fang, C. Wang, Z. Liu, L. Wang, L. Liu, H. Li, S. Xu, A. Zheng, X. Qin, L. Liu and F.-S. Xiao, Science, 2022, 377, 406–410.
56. M. Wolf, N. Fischer and M. Claeys, Nat. Catal., 2020, 3, 962–965.
57. H. Li, C. Qiu, S. Ren, Q. Dong, S. Zhang, F. Zhou, X. Liang, J. Wang, S. Li and M. Yu, Science, 2020, 367, 667–671.
58. X. Zu, Y. Zhao, X. Li, R. Chen, W. Shao, Z. Wang, J. Hu, J. Zhu, Y. Pan, Y. Sun and Y. Xie, Angew. Chem., Int. Ed., 2021, 60, 13840–13846.
59. W. Zhang, C. Deng, W. Wang, H. Sheng and J. Zhao, Adv. Mater., 2024, 36, 2405825.
60. L. Ran, Z. Li, B. Ran, J. Cao, Y. Zhao, T. Shao, Y. Song, M. K. H. Leung, L. Sun and J. Hou, J. Am. Chem. Soc., 2022, 144, 17097–17109.
61. S. Karmakar, S. Barman, F. A. Rahimi, D. Rambabu, S. Nath and T. K. Maji, Nat. Commun., 2023, 14, 4508.
62. K. Lin, P. Qiao, Q. Liu, Y. He, X. Kang, D. Wang, C. Tian, A. Wu and H. Fu, Adv. Funct. Mater., 2025, e15276.

---