Concentrated solar CO₂ reduction in H₂O vapour with >1% energy conversion efficiency

CO₂ reduction involves multiple proton-coupled electron transfers, and while CH₄ is thermodynamically favored over CO, its formation is kinetically challenging. Using H₂O as the proton source is cost-effective, but H₂O dissociation is the rate-limiting step in photo-thermal catalytic CH₄ production from CO₂ and H₂O, typically requiring high temperatures and yielding <1% solar-to-chemical (STC) efficiency. Key goals are efficient H₂O activation, microscopic separation of photo- and thermo-catalysis, and creation of active sites that capture localized phonons and photogenerated carriers. Single-atom Ni is known to promote H₂O dissociation and charge trapping/transfer and can generate hot spots via strong electron-phonon coupling. Motivated by this, the authors investigate a single-atom Ni-based, oxygen-vacancy-rich CeO₂ catalyst for efficient photo-thermal CO₂-to-CH₄ conversion under concentrated solar irradiation, hypothesizing that Ni–V₀ motifs will lower kinetic barriers by enhancing H₂O activation and carrier dynamics.

Catalyst synthesis: CeO₂ nanorods were hydrothermally grown (100 °C, 24 h) on nickel foam supports (cleaned, 1×1 cm², 2 mm thick), then reduced/treated to introduce oxygen vacancies (V₀) and anchor single-atom Ni. NF@0.1%Ni@CeO₂–V₀ was prepared by mixing Ni(NO₃)₂·6H₂O (to 0.1% Ni by ICP-MS) with NaBH₄ in water, adding CeO₂ powder and monolithic NF@NP-CeO₂, and hydrothermally treating at 160 °C for 14 h, yielding porous CeO₂ nanorods with Ni single atoms anchored around surface V₀. NF@0.1%Ni@CeO₂ (without enriched V₀) was prepared similarly at 110 °C for 14 h. NF@CeO₂–V₀ was prepared identically without Ni precursor. Benchmark supports WO₃, ZrO₂, and TiO₂ were synthesized via solvothermal/hydrothermal routes as controls. Characterization: Structure and morphology were examined by TEM, HAADF-STEM (atomically resolved Ni single atoms), XRD (CeO₂ fluorite phase), in-situ XRD (peak shifts under concentrated light), XAS/XANES/EXAFS at Ni K-edge and Ce L₃-edge (Ni oxidation state and Ni–O coordination), EPR (oxygen vacancies, g ≈ 2.002), XPS (valence band, Ce³⁺ content), BET surface area, UV–Vis–NIR DRS (300–2500 nm, V₀ absorption near 1400–1800 nm), TG-DSC (thermal stability), CO₂-TPD and in-situ DRIFTS (intermediates). Surface temperatures were monitored by IR thermography and thermocouples. Photo-thermal catalysis: Reactions were conducted in a 100 mL laterally irradiated, heat-resistant closed reactor using a 300 W Xe lamp (AM 1.5G). Concentrated irradiation of 4200 mW/cm² was achieved with two 30 mm Fresnel lenses; non-concentrated was 420 mW/cm². Reactor setup: catalyst monolith fixed with PTFE clips; 1 mL H₂O added; reactor purged with N₂ (30 min), then fed with 1% CO₂ in N₂ (5 min) and sealed at room temperature. Reaction time: 2 h. Gas analysis every 30 min by GC. Isotope tracing used ¹³CO₂ to confirm carbon source. Thermal-only control experiments were performed in the dark at 200–400 °C. Apparent activation energies were extracted from Arrhenius plots (ln rate vs 1/T) under different light intensities. Opto-electronic measurements: PL and TRPL as a function of temperature, SPV and TPV under varying irradiation intensities to probe carrier lifetimes, recombination (including Auger processes), and extraction efficiencies. Photoelectrochemical EIS, transient photocurrent (I–t), LSV, and OCP used standard three-electrode cells (FTO working electrodes coated with catalyst inks) under a 300 W Xe lamp. IMPS in 1 M KOH with LEDs (325, 560, 780 nm) quantified charge transfer (k_tran, k_rec) and characteristic times (τ_d, τ_a). Computations: DFT and AIMD (VASP, PAW-PBE, E_cut=500 eV; U_eff: Ni 3d=4.5 eV, Ce 4f=5.0 eV; D3 dispersion) on a CeO₂(111) 3×3×3 supercell (108 atoms, 4 layers; bottom two fixed; 15 Å vacuum). Ni substituted Ce to model single-atom Ni. Brillouin sampling: 3×3×1 (relaxation), 9×9×1 (electronic structure). AIMD in NVT; analyses included adsorption energies, carrier states, and H₂O/CO₂ interactions. TDDFT-based nonadiabatic molecular dynamics simulated laser-induced processes (Gaussian-envelope field, 4 eV photon energy, E₀=1.25 V/Å; NVE evolution to 80 fs; catalyst atoms fixed) to decouple thermal vs photonic effects. These simulations assessed CO₂ adsorption/desorption, H₂O adsorption and dissociation pathways, and charge transfer through Ni d-impurity states to adsorbates.

- A single-atom Ni-loaded, oxygen-vacancy-rich CeO₂ catalyst (NF@0.1%Ni@CeO₂–V₀) achieved CH₄ yields of 192.75 µmol·cm⁻²·h⁻¹ under concentrated solar irradiation (4200 mW·cm⁻²), with ~100% CH₄ selectivity and 1.14% solar-to-chemical (STC) energy conversion efficiency. - Performance comparisons: yield was 78× higher than under non-concentrated irradiation (2.47 µmol·cm⁻²·h⁻¹) and 42× higher than NF@0.1%Ni@CeO₂ without enriched V₀ under concentrated irradiation (4.62 µmol·cm⁻²·h⁻¹). A V₀-free Ni catalyst stabilized at a lower surface temperature (306.5 °C, 12.6 s) versus Ni@CeO₂–V₀ (362.1 °C, 7.5 s) under the same flux. - Apparent activation energy for CH₄ formation decreased from 19.54 kJ·mol⁻¹ (non-concentrated) to 5.49 kJ·mol⁻¹ (concentrated), evidencing thermally assisted photocatalysis; CH₄ yield increased with temperature at constant flux. - Structural/chemical state: XANES showed Ni²⁺ in Ni@CeO₂–V₀ (vs Ni³⁺ in Ni@CeO₂). EXAFS-derived Ni–O coordination numbers were 4.3 (V₀-rich, NiO₄-like) vs 5.1 (V₀-poor, NiO₅-like), with no Ni–Ni coordination (isolated single atoms). EPR and in-situ XPS (Ce³⁺ increase) confirmed higher V₀ content under concentrated light, indicating vacancy regeneration. - Optical/thermal coupling: UV–Vis–NIR DRS showed enhanced full-spectrum absorption in V₀-rich samples (V₀-related bands near 1400 and 1800 nm). In-situ XRD under concentrated flux showed peak shifts consistent with high photon flux inducing defect-related lattice changes. Concentrated irradiation increased carrier density (Fermi level upshift), promoted Auger processes, and generated local hot spots via Ni d-orbital impurity states capturing carriers and converting their energy into high-frequency phonons. - Mechanism and intermediates: In-situ DRIFTS identified m-CO₃²⁻ (CO₂ adsorbed on lattice O of CeO₂), and -CH₂O, -CH₃O, -CH₃ intermediates en route to CH₄. AIMD/TDDFT indicated CO₂ weakly interacts with Ni and desorbs (~100 fs), while H₂O adsorbs strongly at Ni (E_ads ≈ −1.47 eV vs −0.67 eV at V₀/Ce sites), hybridizes with Ni d-states near VBM, captures photogenerated holes, and undergoes thermally assisted photo-dissociation to provide H for CO₂ hydrogenation. Released O participates in O₂ formation; measured O₂ reached 762.5 µmol·cm⁻² under concentrated irradiation. - Charge dynamics: Rising temperature decreased PL intensity and lifetime (enhanced mobility and recombination), yet overall reaction benefited as migration outpaced recombination; TPV showed reduced carrier lifetimes at higher intensities (Auger recombination) and improved charge extraction. IMPS revealed minimized τ_a under UV–visible irradiation for Ni@CeO₂–V₀, indicating accelerated charge separation/transfer under photo-thermal coupling. - Stability: Catalyst retained structure and Ni coordination state after reaction; activity showed only a slight decline after six cycles and remained stable after 30 days in Ar. Hydrogen introduction restored/enhanced activity in subsequent cycles, consistent with vacancy regeneration under concentrated light. - Isotope tracing with ¹³CO₂ confirmed CO and CH₄ originated from feed CO₂, excluding carbon contamination.

The study addresses the kinetic bottleneck of H₂O activation in CO₂ methanation by engineering Ni single atoms anchored adjacent to oxygen vacancies on CeO₂. Concentrated solar irradiation creates a high photon flux that (i) reduces the apparent activation energy via photogenerated carriers and Auger-assisted phonon generation, (ii) elevates surface temperatures through localized hot-spot formation at Ni d-impurity states, and (iii) regenerates surface oxygen vacancies, maintaining active site density. DFT/TDDFT/AIMD and in-situ spectroscopy converge on a mechanism where CO₂ adsorbs on lattice oxygen of CeO₂ while H₂O preferentially adsorbs at Ni single atoms; photogenerated holes transfer through Ni d-states to H₂O, enabling thermally assisted photo-dissociation. The resulting H atoms hydrogenate adsorbed CO₂-derived intermediates to CH₄ with near-unity selectivity, while O recombines to O₂ without depleting V₀. Enhanced carrier density and improved charge separation/transfer under concentrated light further accelerate reaction kinetics. Collectively, the Ni–V₀ motif decouples and synergizes photo- and thermo-effects to overcome the H₂O dissociation barrier, delivering >1% STC efficiency with high stability.

A Ni single-atom, oxygen-vacancy-rich CeO₂ catalyst enables efficient, selective photo-thermal CO₂-to-CH₄ conversion under concentrated solar irradiation, achieving 192.75 µmol·cm⁻²·h⁻¹ CH₄ yield, ~100% selectivity, and 1.14% STC efficiency. Mechanistically, Ni adjacent to V₀ captures carriers, generates localized phonons, and promotes thermally assisted photo-dissociation of H₂O; concentrated light lowers apparent activation energy and regenerates vacancies, sustaining active sites. The integration of concentrated solar photonics with defect-engineered single-atom catalysis provides a viable route to practical solar-to-chemical conversion. Future research should optimize catalyst architectures for enhanced vacancy stability and phonon management, scale reactor designs for field conditions, and explore analogous single-atom/defect motifs to activate H₂O (or other proton donors) for multi-carbon products.

- A slight decrease in catalytic activity was observed after six cycles, indicating some performance loss over extended operation, though activity remained high and could be restored/enhanced with hydrogen pretreatment. - In-situ XPS could not resolve changes in Ni valence due to the very low Ni content (0.1%), limiting direct observation of Ni redox dynamics during operation. - Under purely thermal conditions (200–400 °C, dark), no CH₄ was produced, confirming reliance on photo-thermal coupling; performance under different gas compositions or higher CO₂ concentrations was not detailed in the main text. - Potential Ni deactivation can occur if O fills V₀ (structural shift from planar quadrilateral to octahedral coordination), although concentrated light conditions favor vacancy regeneration.