Designing atomic Ni/Cu pairs on a reactive BiOCl surface for efficient photo-chemical HCO₃⁻-to-CO conversion

Abstract

Solar-driven conversion of bicarbonate (HCO₃⁻) to carbonaceous fuels and/or chemicals provides an alternative route for the development of sustainable carbon economies. However, promoting the HCO₃⁻ reduction rate and tuning product selectivity remain significant challenges. This study reports the identification of isolated Ni/Cu atomic pairs dispersed on a BiOCl surface (Ni₁/Cu₁-BOC) as a promising candidate for efficient HCO₃⁻ reduction under UV-vis light irradiation. The optimized photocatalyst exhibits a high CO formation rate of 157.1 μmol g⁻¹ h⁻¹ with nearly 100% selectivity, even in the absence of added proton sources, sacrificial agents, or sensitizers. Experimental and theoretical investigations reveal that the atomically dispersed Ni/Cu pairs facilitate the protonation of HCO₃⁻ to CO₂, which then undergoes a H⁺-assisted reduction pathway to produce CO, with *COOH as the intermediate. The synergistic effects of the Ni/Cu atomic pairs simultaneously promote the HCO₃⁻-to-CO₂ conversion and the subsequent CO₂-to-CO reduction, providing valuable insights for the development of efficient diatomic catalysts for photocatalytic HCO₃⁻ reduction reactions.

---

Tengfei Zhou

Prof. Tengfei Zhou's research is focused on CO2 conversion, metal–CO2 batteries, low-temperature batteries and interfacial catalysis problems encountered in energy storage systems. Zhou started his professional research career as an Australian Research Council (ARC) DECRA Fellow in the Institute for Superconducting and Electronic Materials at the University of Wollongong from 2018 to 2021, under the supervision of Prof. Zaiping Guo. He is currently a professor at the Institutes of Physical Science and Information Technology at Anhui University.

---

Introduction

Direct photocatalytic reduction of bicarbonate (HCO₃⁻R) using naturally abundant carbonate/bicarbonate minerals as the feedstock, without additional CO₂ supplementation in aqueous media, is a challenging yet fascinating approach for the development of economical carbon-resource conversion technologies.¹ The dissolved HCO₃⁻ inevitably undergoes dissociation (HCO₃⁻ → CO₂ (aq) + OH⁻) and/or protonation (HCO₃⁻ + H⁺ → CO₂ (aq) + H₂O) processes, resulting in a near-zero concentration of CO₂, which then needs to be further reduced to carbonaceous products.^2,3 This CO₂-mediated HCO₃⁻ conversion rate is kinetically sluggish due to the ultra-low CO₂ concentration and the unfavourable alkaline pH that restricts CO₂ formation.⁴ Moreover, the excessive presence of HCO₃⁻ anions on the CO₂ reduction (CO₂R) sites prevents the adsorption and activation of CO₂ molecules, further lowering the overall activity.^5,6 Therefore, designing catalysts to enhance local CO₂(aq) concentration with specific HCO₃⁻/CO₂ conversion sites is crucial for the high-performance HCO₃⁻R reaction.

Recent studies have shown that H⁺-driven HCO₃⁻ speciation through long-range proton transport (∼100 nm) can efficiently promote the local CO₂R rate compared to HCO₃⁻ self-dissociation.⁷ The H⁺ generated from nearby redox sites can directly react with HCO₃⁻ and/or OH⁻ to generate in situ CO₂(aq), which can then undergo H⁺-assisted CO₂R to carbonaceous products, thereby facilitating the entire HCO₃⁻R process.^8,9 However, the use of external proton sources, such as inorganic/organic acids (HCl, H₂SO₄, and H₂BQ) or organic additives (EtOH and MeOH) and buffer solution (NaH₂PO₄/Na₂HPO₄), not only adds extra costs to the HCO₃⁻R systems but also introduces undesirable corrosion of the reactors.^10–12 Additionally, the presence of H⁺ can competitively consume photo-induced electrons, further lowering the reaction activity and selectivity.¹³ Therefore, developing cost-effective proton sources (particularly H₂O) that can provide H⁺ under neutral conditions, and novel catalysts that selectively reduce CO₂, are crucial priorities for the practical development of efficient HCO₃⁻R systems.

Single atomic catalysts (SACs) with highly exposed reactive sites, unsaturated coordinate geometries, and unique chemical/physical properties have garnered extensive research interest in the fields of CO₂ reduction, N₂ fixation and H₂O splitting.^14–16 The isolated, dispersed metal sites in SACs may outperform metal clusters/alloys in selective chemical conversion to target products, as they are less susceptible to competing reactions due to their single-type atomic geometry, particularly in H⁺-assisted molecular catalysis that prevents undesirable H₂ formation.¹⁷ For instance, manipulating Ni single atoms on carbon nitride/ZrO₂ substrates to form Ni-based SACs has favoured electro/photo-chemical CO₂ reduction to CO with >99% CO Faraday efficiency and/or ∼100% CO selectivity, even in aqueous media.^18,19 However, while known SACs comprising noble metals (Ru, Ir and Re) and earth-abundant metals (Co, Ni and Fe) can reduce CO₂ to CO, relatively few of them can realize HCO₃⁻R.^20–26 The CO₂-to-CO undergoes a 2e⁻/2H⁺ transfer process, while HCO₃⁻-to-CO requires an additional H⁺ to pre-generate CO₂(aq). Incorporating metals capable of accomplishing multi-e⁻/H⁺ transfer/formation into SACs as catalytic pairs could theoretically drive the HCO₃⁻ reduction reaction.²⁷ Previous work has demonstrated a self-adaptive dual-metal-site catalyst (DMSC) with flexible Ni–Cu pairs that can reduce CO₂ to CH₄,^28,29 motivating the design of photocatalysts with Ni/Cu catalytic pairs for HCO₃⁻R attempts.

Herein, we present a PVP-mediated solvothermal method for fabricating a diatomic catalyst by immobilizing Ni and Cu single atoms on the photoreactive (001) surface of BiOCl (abbreviated as Ni₁/Cu₁-BOC). The single-site Ni and Cu atoms in their respective forms enabled photocatalytic CO₂R and HCO₃⁻R under mild conditions without the assistance of sacrificial agents or photosensitizers. Ni₁/Cu₁-BOC exhibited an impressive HCO₃⁻-to-CO (in aqueous media) reduction rate of around 157.1 μmol g⁻¹ h⁻¹, with nearly 100% CO selectivity, outperforming the individual Ni₁-BOC (115.2 μmol g⁻¹ h⁻¹), Cu₁-BOC (89.1 μmol g⁻¹ h⁻¹) and pristine BOC (34.6 μmol g⁻¹ h⁻¹) catalysts. Detailed characterization revealed that the CO formation was closely correlated with the local CO₂(aq) from the protonation of HCO₃⁻R, resulting in a CO₂-intermediated HCO₃⁻R mechanism. In situ Fourier-transform infrared spectroscopy (in situ FTIR) and density functional theory (DFT) calculations confirmed that the Ni/Cu catalytic pairs controlled the reaction path to CO by significantly promoting *COOH as the intermediate and reducing the energy barriers for CO formation.

Results and discussion

The synthetic procedure for the Ni₁/Cu₁-BOC photocatalyst is described in Fig. 1a. NiCl₂·6H₂O, CuCl₂·2H₂O and Bi(NO₃)₃·5H₂O respectively served as Ni²⁺, Cu²⁺ and Bi³⁺/Cl⁻ sources with ethylene glycol (EG) as the solvent. During the solvothermal stage, PVP (K30) was added as a surface-mediating agent to create more sites for the immobilization of singly dispersed Cu and Ni on the BiOCl substrate. PVP can also prevent the agglomeration of BiOCl nanoparticles, favoring the crystallization of BiOCl in well-defined morphologies with large specific surface areas.³⁰ Other control photocatalysts, namely Ni₁-BOC, Cu₁-BOC and BOC, were prepared using a similar synthetic method. The Ni and Cu loadings in the different photocatalysts were quantified by inductively coupled plasma optical emission spectroscopy (ICP-OES) (Table S1†).

Fig. 1 (a) Scheme depicting the synthesis of Ni₁/Cu₁-BOC. (b and c) XRD patterns and Raman spectra of Ni₁/Cu₁-BOC, Ni₁-BOC, Cu₁-BOC and BOC. (d–f) TEM, HRTEM and EDS mapping images of Ni₁/Cu₁-BOC.

The XRD patterns of the Ni₁/Cu₁-BOC, Ni₁-BOC and Cu₁-BOC photocatalysts showed only the typical reflection peaks of tetragonal BiOCl, indicating that any Ni and Cu were highly dispersed on the support (Fig. 1b). Raman spectra of these photocatalysts also showed identical vibration peaks A_1g² internal Bi–Cl phonon mode (145 cm⁻¹) and E_g internal Bi–Cl phonon mode (190 cm⁻¹) to that of BiOCl,³¹ suggesting that the Ni and Cu were not incorporated by interlaminar Cl atoms (Fig. 1c). TEM images revealed that the Ni₁/Cu₁-BOC photocatalyst had a square-like morphology and was free of any irregular impurities or clusters (Fig. 1d). High-resolution TEM (HRTEM) images for the selected regions showed sets of interlaced lattice fringes, with interplanar spacings of around 0.27 nm which were well-crystallized along the [001] direction (Fig. 1e and S1†). Energy dispersive spectroscopy (EDS) elemental mapping images for Bi, O, Cl, Ni and Cu revealed a uniform distribution of these elements in the entire Ni₁/Cu₁-BOC nanosheets (Fig. 1f).

To probe the coordination structures of Ni and Cu atoms in the Ni₁/Cu₁-BOC photocatalyst, Ni and Cu K-edge X-ray absorption near-edge structure (XANES), extended X-ray absorption fine structure (EXAFS) and wavelet transform (WT) data were collected for Ni₁/Cu₁-BOC and related reference materials. As shown in Fig. 2a, the Ni K-edge of Ni₁/Cu₁-BOC and Ni₁-BOC exhibited higher pre-edge energies than that of Ni foil and NiO, indicating that the Ni center in Ni₁/Cu₁-BOC and Ni₁-BOC had a higher valence state than NiO. The Ni K-edge EXAFS spectra of Ni₁/Cu₁-BOC and Ni₁-BOC showed one intense peak at approximately 1.3 Å, closely related to the first shell of Ni–O coordination possibly due to the embedded Ni in the [Bi₂–O₂]²⁺ matrix (Fig. 2b). No Ni–Ni bond (2.17 Å for Ni foil and 2.34 Å for NiO) was observable in Ni₁/Cu₁-BOC and Ni₁-BOC, indicating that Ni atoms were singly dispersed. The atomic dispersion of Ni species in Ni₁/Cu₁-BOC and Ni₁-BOC was further confirmed by WT analysis of Ni K-edge EXAFS oscillations. The WT contour plots of Ni₁/Cu₁-BOC and Ni₁-BOC showed maximum intensities attributable to Ni–O bonding (at 4.9 and 4.8 Å⁻¹, respectively), whereas the Ni–Ni bonding was absent (Fig. 2c).³²Fig. 2d shows the Cu K-edge XANES spectra of Ni₁/Cu₁-BOC, Cu₁-BOC, Cu foil and CuO. The Cu K-edge of Ni₁/Cu₁-BOC and Cu₁-BOC was close to that of CuO, suggesting that Cu atoms in Ni₁/Cu₁-BOC and Cu₁-BOC were positively charged with valence states near +2. The Cu K-edge EXAFS spectra of Ni₁/Cu₁-BOC and Cu₁-BOC showed peaks only assigned to the Cu–O bond (1.5 Å) while Cu–Cu interactions (2.24 Å for Cu foil and 2.46 Å for CuO) were undetectable (Fig. 2e). This indicated that Cu atoms in Ni₁/Cu₁-BOC and Cu₁-BOC were also isolated and dispersed, for which the WT contour plots of Ni₁/Cu₁-BOC demonstrated different Cu features than that of Cu foil and CuO (Fig. 2f). To better understand the electronic and spatial structures of Ni and Cu atoms in the Ni₁/Cu₁-BOC photocatalyst, Ar ion sputtering and high-resolution X-ray photoelectron spectroscopy (XPS) were performed over the Ni 2p, Cu 2p, Bi 4f, O 1s, and Cl 2p regions of Ni₁/Cu₁-BOC along with selected reference samples including Ni₁-BOC, Cu₁-BOC and BOC. As shown in Fig. 2g, the Ni 2p spectra of Ni₁/Cu₁-BOC was deconvoluted into five peaks including Ni²⁺ (2p_3/2 855.4 eV and 2p_1/2 873 eV), Ni^δ+,δ>2 (2p_3/2 857.3 eV) and Ni²⁺ shake up satellites (880.2 eV and 861.4 eV).³³ This evidenced the presence of over-oxidized Ni, consistent with the afore-analyzed Ni K-edge XANES results. The Ni 2p signals almost vanished after Ar ion sputtering, implying that Ni single atoms were mainly dispersed on the outermost surfaces of Ni₁/Cu₁-BOC. Fig. 2h shows the Cu 2p spectra of Ni₁/Cu₁-BOC. The intense peak centered at 940.5 eV was ascribed to Bi 4s orbitals while peaks at 932.1 eV and 951.0 eV were assigned to Cu²⁺ cations.³⁴ The Cu atoms were also localized on the outermost layer of Ni₁/Cu₁-BOC, and its oxidative states were rarely influenced by Ni loading (Fig. S2†). The Bi 4f spectra of Ni₁/Cu₁-BOC are depicted in Fig. 2i. Two peaks at 158.4 eV and 164.4 eV were clearly resolved, related to partially reduced Bi^3−x states. The surface Bi atoms were relatively negatively charged in contrast to the inner form, indicating that Bi atoms were unconventionally coordinated. This might be attributed to the additives Ni and Cu in the BiOCl substrate, which significantly modified the delocalized electrons near Bi, O and Cl atoms, thus altering the redox capacity (Fig. S3†).

Fig. 2 (a) Ni K-edge XANES spectra, (b) k³-weighted Fourier transforms of the Ni K-edge EXAFS spectra and (c) relative WT plots of Ni foil, Ni₁/Cu₁-BOC, Ni₁-BOC and a NiO standard. (d) Cu K-edge XANES spectra, (e) k³-weighted Fourier transforms of the Cu K-edge EXAFS spectra and (f) relative WT plots of Cu foil, Ni₁/Cu₁-BOC, Cu₁-BOC and a CuO standard. (g–i) Ar ion sputtering X-ray photoelectron spectra of Ni 2p, Cu 2p and Bi 4f orbitals in Ni₁/Cu₁-BOC.

The photocatalytic HCO₃⁻R experiments were performed under UV-vis light irradiation without any assistance of photosensitizers and sacrificial agents. Prior to each test, the custom-built photoreactor was purged using high-purity Ar to eliminate gaseous impurities. Fig. 3a shows that Ni₁-BOC delivered a HCO₃⁻-to-CO production rate of 115.2 μmol g⁻¹ h⁻¹, much higher than that of BOC (34.6 μmol g⁻¹ h⁻¹). After the introduction of Cu single atoms, the photocatalytic activity was further improved. Ni₁/Cu₁-BOC offered the highest CO formation rate of 157.1 μmol g⁻¹ h⁻¹, which was 1.4, 1.8 and 4.5 times higher than the activities of Ni₁-BOC, Cu₁-BOC and the BOC reference photocatalyst, respectively. The H₂ formation was negligible during CO production, probably due to the alkaline HCO₃⁻R conditions that prevent H⁺-coupling. This superb HCO₃⁻R rate demonstrated by Ni₁/Cu₁-BOC could also be extended to the CO₂R reaction, where humid CO₂ gas replaced aqueous HCO₃⁻ solution as the catalytic feed. As shown in Fig. 3b, Ni₁/Cu₁-BOC afforded the optimal CO₂R performance amongst Ni₁-BOC, Cu₁-BOC and BOC catalysts. The CO produced from CO₂ was less than that from HCO₃⁻, which might be because of the lower solubility of CO₂ in aqueous media. Fig. 3c shows the HCO₃⁻R results over Ni₁/Cu₁-BOC at different pH values. The CO formation rate was dramatically increased by H⁺ addition, and suppressed under excessively alkaline conditions. Based on the Bjerrum plot, the fraction of water-soluble CO₂ was nearly 0 at pH > 8.5.⁷ This could be utilized to explain the almost 0 μmol g⁻¹ h⁻¹ CO formation rate at pH ≈ 9, because HCO₃⁻ might not evolve into CO₂ thus suspending the overall HCO₃⁻-to-CO conversion. Lowering the pH value to a neutral value and/or increasing the local CO₂ contents could promote CO production, further evidencing that CO₂ was the key intermediate during the HCO₃⁻R reaction. Cycling experiments of Ni₁/Cu₁-BOC showed no decrease in the CO formation after 5 runs, indicating that the photocatalyst possessed good photostability (Fig. S4†). XRD patterns and TEM images of Ni₁/Cu₁-BOC before and after the reaction were almost identical, suggesting that Ni/Cu catalytic pairs were stable against aggregating during the photocatalytic tests (Fig. S5†). To gain deeper insights into the mechanism underpinning the fast HCO₃⁻R conversion on Ni₁/Cu₁-BOC as compared with BOC, in situ FTIR spectra were deliberately employed to monitor the reaction details. As shown in Fig. 3d and e, IR spectra of Ni₁/Cu₁-BOC and BOC showed peak positions ranging from 1200 to 1700 cm⁻¹, respectively assigned to chemisorption CO₂ (˙CO₂⁻, 1248 cm⁻¹, 1700 cm⁻¹), bicarbonates (HCO₃⁻, 1400–1435 cm⁻¹), monodentate carbonates (m-CO₃²⁻, 1381 cm⁻¹ and 1451 cm⁻¹) and bidentate carbonates (b-CO₃²⁻, 1357 cm⁻¹ and 1556 cm⁻¹).^35–37 The intensity of these peaks intensified on prolonging the irradiation time, indicating that carbonaceous species accumulated on the catalyst surfaces. The obvious differences between Ni₁/Cu₁-BOC and BOC were the symmetric IR peaks centered at 1650 and 1540 cm⁻¹, which can be readily indexed as the *OH (–OH in H₂O) and *COOH groups.^38,39 The *OH signal gradually increased on BOC and almost vanished on Ni₁/Cu₁-BOC, implying that Ni₁/Cu₁-BOC was unfavourable for *OH or H₂O adsorption in contrast to BOC. The *COOH peak of Ni₁/Cu₁-BOC was strengthened compared to that of BOC, evidencing that Ni₁/Cu₁-BOC facilitated protonation during the HCO₃⁻R process with *COOH as the intermediate. Fig. 3f directly provides IR peak intensities of *OH and *CO collected over Ni₁/Cu₁-BOC and BOC. The *CO formation slope showed negative correlation with the surface *OH uptake, clearly emphasizing that excess *OH impeded HCO₃⁻-to-CO conversion. This phenomenon can be explained as *OH consumed additional H⁺ involved in HCO₃⁻-to-CO₂ and CO₂-to-CO paths, thus weakening CO₂/*COOH evolution as well as lowering the CO production. The afore-mentioned analysis depicted a typical HCO₃⁻R route through HCO₃⁻-CO₂-*COOH–CO, during which Ni₁/Cu₁-BOC outperformed BOC in CO formation due to its efficiency in H⁺ utilization (Fig. 3g).

Fig. 3 (a) CO production rates from HCO₃⁻R collected over Ni₁/Cu₁-BOC, Ni₁-BOC, Cu₁-BOC and BOC. Reaction conditions: 1 mol per L HCO₃⁻ aqueous solution, full spectrum irradiation and 10 mg of the catalyst. (b) CO production rates from CO₂R collected over Ni₁/Cu₁-BOC, Ni₁-BOC, Cu₁-BOC and BOC. (c) CO formation rates of Ni₁/Cu₁-BOC under different pH conditions. (d and e) In situ FTIR spectra of BOC and Ni₁/Cu₁-BOC during photocatalytic HCO₃⁻R reactions. (f) *OH and *CO signals of BOC and Ni₁/Cu₁-BOC collected from the in situ FTIR spectra. (g) Schematic diagrams showing the HCO₃⁻R mechanism over Ni₁/Cu₁-BOC.

Note that the multi-protons/electron transfer process was involved in HCO₃⁻-to-CO conversion, and the photoelectric properties and electron-migration capacities of Ni₁/Cu₁-BOC and BOC were then respectively investigated. Fig. 4a displays the light absorption characteristics of Ni₁/Cu₁-BOC and BOC. The BOC sample had an adsorption edge near 380 nm, arising from the O 2p to Bi 6p transitions. Ni₁/Cu₁-BOC showed additional peaks ranging from 400–700 nm, involving O 2p to Ni/Cu 3d transitions and d–d transitions in Ni/Cu 3d orbitals, respectively. The band gap values of Ni₁/Cu₁-BOC and BOC were calculated to be 3.31 eV and 3.46 eV based on the Kubelka–Munk function, clearly emphasizing that Ni/Cu incorporation changed the intrinsic band alignment of BiOCl (Fig. 4b). The conduction band (CB) levels of Ni₁/Cu₁-BOC and BOC were determined to be −0.85 V and −0.7 V (vs. NHE) from the Mott–Schottky plots, all satisfying the thermodynamic requirements for CO production from CO₂ (Fig. S6†).⁴⁰ Additionally, the valence band (VB) positions of Ni₁/Cu₁-BOC and BOC were estimated to be about 2.46 eV and 2.76 eV, all above the redox potential for H₂O oxidation to O₂ (Fig. 4c).⁴¹ This observation suggested that CO and O₂ possibly served as the products from reduction/oxidation-half reactions, with pre-analyzed surface *OH probably being the oxygen evolution intermediate.⁴² The dynamics of photo-induced charge carriers were then employed by photoluminescence (PL) and electrochemical impedance spectroscopy (EIS). As shown in Fig. 4d and e, Ni₁/Cu₁-BOC displayed weakened PL peaks with the smallest EIS semicircles as compared with BOC. Note that the quenched PL signals reflected the suppressed electron–hole recombination and the EIS radius empirically correlated with the inner-resistance; the afore-mentioned results thus collectively indicated that Ni₁/Cu₁-BOC possessed the best charge separation efficiency due to its high conductivity.^43–48Fig. 4f shows the photocurrent signals collected over Ni₁/Cu₁-BOC and BOC during the same time interval. The photovoltaic peaks demonstrated by Ni₁/Cu₁-BOC surpassed that of BOC, further evidencing that Ni₁/Cu₁-BOC facilitated photoelectric conversion and charge carrier migration.^49–51

Fig. 4 (a and b) Diffuse reflectance spectra and Tauc plots of Ni₁/Cu₁-BOC and BOC. (c) Band alignments of Ni₁/Cu₁-BOC and BOC. (d–f) Steady-state PL spectra, EIS and photocurrent patterns of Ni₁/Cu₁-BOC and BOC.

To further elucidate the photocatalytic HCO₃⁻R dynamics at the atomic level, DFT calculations of elementary HCO₃⁻R steps over Ni₁/Cu₁-BOC and BOC were separately simulated. Based on the XAFS studies, we constructed surface models as BOC (001) with/without Ni/Cu catalytic pairs (Ni/Cu-BOC and BOC) to represent the likely coordination geometries of Ni₁/Cu₁-BOC and BOC samples (Fig. 5a). Fig. 5b shows Gibbs energy barriers involved in HCO₃⁻-to-CO₂ conversion. The solvated HCO₃⁻ anions were first chemically adsorbed on the catalytic surfaces, followed by C–OH cleavage (by dehydration) to form *CO₂. Ni/Cu-BOC offered a lower energy barrier for *HCO₃ protonation/dehydration to *CO₂, evidencing that the Ni₁/Cu₁-BOC catalyst favoured H⁺ utilization. The subsequent transformation of CO₂ into CO was predicted to occur endothermically over the Ni/Cu-BOC model, and the formation of adsorbed *COOH was the rate limiting step (Fig. 5c). The summed-up energy barriers required for CO production within the Ni/Cu-BOC model (ΔG = 0.57 eV) also lower than that of the BOC model (ΔG = 0.63 eV), indicating the advantages of Ni/Cu catalytic pairs in HCO₃⁻R/CO₂R reactions. The dissociative H⁺ tended to be captured by BOC as adsorbed *H, while Ni/Cu-BOC unfavoured *H formation thus preferred H⁺-driven HCO₃⁻ speciation/H⁺-assisted CO₂ reduction (Fig. 5d).

Fig. 5 (a) DFT models representing BOC and Ni₁/Cu₁-BOC samples. (b–d) Energy profiles for the elementary steps of HCO₃⁻-to-CO₂, CO₂-to-CO and H⁺-to-H₂ pathways over BOC and the Ni/Cu-BOC model.

Conclusions

In conclusion, a Ni₁/Cu₁-BOC photocatalyst comprising singly dispersed Ni/Cu catalytic pairs over a BiOCl substrate afforded excellent performance for the photocatalytic reduction of bicarbonate (HCO₃⁻) to carbon monoxide (CO) under UV-vis irradiation. Benefiting from the surface-immobilized Ni/Cu catalytic pairs, a remarkable CO formation rate of 157.1 μmol g⁻¹ h⁻¹ was realized in the HCO₃⁻ aqueous solution, surpassing the activities of the individual Ni₁-BOC, Cu₁-BOC and BOC catalysts. In situ FTIR results and photoelectric tests showed that the protons and electrons required for the HCO₃⁻-to-CO₂ conversion were dramatically increased and promoted on the Ni/Cu-involved catalyst, facilitating the subsequent CO₂ reduction to CO with *COOH as the reactive intermediate. DFT calculations also demonstrated that the Ni/Cu-BOC model lowers the energy barriers involved in the HCO₃⁻ reduction process by significantly promoting the utilization of H⁺ through a series of protonation and dehydration pathways leading to CO production. These findings shed light on the rational design of diatomic photocatalysts for sustainable HCO₃⁻/CO₂ reduction reactions. The Ni₁/Cu₁-BOC photocatalyst developed in this study delivers an unusually high CO formation rate during the photocatalytic HCO₃⁻ reduction tests, showcasing its potential for practical applications in the development of economical carbon-resource conversion technologies.

Data availability

All the data supporting this article have been included in the main text and the ESI.†

Author contributions

D. Ke, B. Sun and Y. Zhang: investigation, visualization and writing – original draft. F. Tian and H. Yang: supervision and conceptualization. Y. Chen: resources and software. Q. Meng, Y. Zhang, Z. Hu and C. Yang: data curation, formal analysis and visualization. X. Xiong and T. Zhou: supervision, writing – review & editing, resources and funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors are grateful for the financial support of the Anhui Provincial Natural Science Foundation for Outstanding Young Scholar (2208085Y05), Anhui Provincial Scientific Reuter Foundation for Returned Scholars (2022LCX030), and the National Natural Science Foundation of China (52202198 and 22272207). The authors also thank Dr Yu Chen from the Shanghai Synchrotron Radiation Facility for his help with the XAS data analysis.

Notes and references

1. C. Hepburn, E. Adlen, J. Beddington, E. Carter, S. Fuss, N. Mac Dowell, J. Minx, P. Smith and C. Williams, Nature, 2019, 575, 87–97.
2. S. Zhang, C. Chen, K. Li, H. Yu and F. Li, J. Mater. Chem. A, 2021, 9, 18785–18792.
3. D. Pimlott, A. Jewlal, Y. Kim and C. Berlinguette, J. Am. Chem. Soc., 2023, 145, 25933–25937.
4. M. Dunwell, Q. Lu, J. Heyes, J. Rosen, J. Chen, Y. Yan, F. Jiao and B. Xu, J. Am. Chem. Soc., 2017, 139, 3774–3783.
5. C. Stalder, S. Chao and M. Wrighton, J. Am. Chem. Soc., 1984, 106, 3673–3675.
6. N. Queyriaux, ACS Catal., 2021, 11, 4024–4035.
7. R. Yanagi, T. Zhao, M. Cheng, B. Liu, H. Su, C. He, J. Heinlein, S. Mukhopadhyay, H. Tan and D. Solanki, J. Am. Chem. Soc., 2023, 145, 15381–15392.
8. J. Schneider, H. Jia, J. Muckerman and E. Fujita, Chem. Soc. Rev., 2012, 41, 2036–2051.
9. Q. Dong, X. Zhang, D. He, C. Lang and D. Wang, ACS Cent. Sci., 2019, 5, 1461–1467.
10. M. Nielsen, E. Alberico, W. Baumann, H. Drexler, H. Junge, S. Gladiali and M. Beller, Nature, 2013, 495, 85–89.
11. M. Symes and L. Cronin, Nat. Chem., 2013, 5, 403–409.
12. K. Joya, Y. Joya and H. De Groot, Adv. Energy Mater., 2014, 4, 1301929.
13. N. Elgrishi, M. Chambers and M. Fontecave, Chem. Sci., 2015, 6, 2522–2531.
14. A. Wang, J. Li and T. Zhang, Nat. Rev. Chem, 2018, 2, 65–81.
15. S. Ding, M. Hülsey, J. Pérez-Ramírez and N. Yan, Joule, 2019, 3, 2897–2929.
16. Q. Wu and C. Wu, J. Mater. Chem. A, 2023, 11, 4876–4906.
17. X. Yang, A. Wang, B. Qiao, J. Li, J. Liu and T. Zhang, Acc. Chem. Res., 2013, 46, 1740–1748.
18. H. Yang, S. Hung, S. Liu, K. Yuan, S. Miao, L. Zhang, X. Huang, H. Wang, W. Cai and R. Chen, Nat. Energy, 2018, 3, 140–147.
19. X. Xiong, C. Mao, Z. Yang, Q. Zhang, G. Waterhouse, L. Gu and T. Zhang, Adv. Energy Mater., 2020, 10, 2002928.
20. S. Takizawa, T. Okuyama, S. Yamazaki, K. Sato, H. Masai, T. Iwai, S. Murata and J. Terao, J. Am. Chem. Soc., 2023, 145, 15049–15053.
21. S. Sato, T. Morikawa, T. Kajino and O. Ishitani, Angew. Chem., Int. Ed., 2013, 52, 988–992.
22. B. Su, Y. Kong, S. Wang, S. Zuo, W. Lin, Y. Fang, Y. Hou, G. Zhang, H. Zhang and X. Wang, J. Am. Chem. Soc., 2023, 145, 27415–27423.
23. L. Chen, Z. Guo, X. Wei, C. Gallenkamp, J. Bonin, E. Anxolabéhère-Mallart, K. Lau, T. Lau and M. Robert, J. Am. Chem. Soc., 2015, 137, 10918–10921.
24. H. Rao, L. Schmidt, J. Bonin and M. Robert, Nature, 2017, 548, 74–77.
25. K. Sun, Y. Huang, Q. Wang, W. Zhao, X. Zheng, J. Jiang and H.-L. Jiang, J. Am. Chem. Soc., 2024, 146, 3241–3249.
26. X. Yu, M. Sun, T. Yan, L. Jia, M. Chu, L. Zhang, W. Huang, B. Huang and Y. Li, Energy Environ. Sci., 2024, 17, 2260–2268.
27. B. Mondal, A. Rana, P. Sen and A. Dey, J. Am. Chem. Soc., 2015, 137, 11214–11217.
28. J. Li, H. Huang, W. Xue, K. Sun, X. Song, C. Wu, L. Nie, Y. Li, C. Liu and Y. Pan, Nat. Catal., 2021, 4, 719–729.
29. Y. Cai, J. Fu, Y. Zhou, Y. Chang, Q. Min, J. Zhu, Y. Lin and W. Zhu, Nat. Commun., 2021, 12, 586.
30. Y. Lei, G. Wang, S. Song, W. Fan and H. Zhang, CrystEngComm, 2009, 11, 1857–1862.
31. A. Phuruangrat, S. Thongtem and T. Thongtem, Appl. Phys. A, 2020, 126, 245.
32. T. Zhang, X. Han, H. Yang, A. Han, E. Hu, Y. Li, X. Yang, L. Wang, J. Liu and B. Liu, Angew. Chem., Int. Ed., 2020, 59, 12055–12061.
33. S. Uhlenbrock, C. Scharfschwerdt, M. Neumann, G. Illing and H. Freund, J. Phys.: Condens. Matter, 1992, 4, 7973–7978.
34. S. Poulston, P. Parlett, P. Stone and M. Bowker, Surf. Interface Anal., 1996, 24, 811–820.
35. Y. Zhang, Z. Xu, Q. Wang, W. Hao, X. Zhai, X. Fei, X. Huang and Y. Bi, Appl. Catal., B, 2021, 299, 120679.
36. E. Köck, M. Kogler, T. Bielz, B. Klötzer and S. Penner, J. Phys. Chem. C, 2013, 117, 17666–17673.
37. S. Collins, M. Baltanás and A. Bonivardi, J. Phys. Chem. B, 2006, 110, 5498–5507.
38. J. Wu, X. Li, W. Shi, P. Ling, Y. Sun, X. Jiao, S. Gao, L. Liang, J. Xu, W. Yan, C. Wang and Y. Xie, Angew. Chem., Int. Ed., 2018, 57, 8719–8723.
39. X. Li, Y. Sun, J. Xu, Y. Shao, J. Wu, X. Xu, Y. Pan, H. Ju, J. Zhu and Y. Xie, Nat. Energy, 2019, 4, 690–699.
40. S. Nitopi, E. Bertheussen, S. Scott, X. Liu, A. Engstfeld, S. Horch, B. Seger, I. Stephens, K. Chan and C. Hahn, Chem. Rev., 2019, 119, 7610–7672.
41. M. Yagi and M. Kaneko, Chem. Rev., 2001, 101, 21–36.
42. X. Shi, S. Back, T. Gill, S. Siahrostami and X. Zheng, Chem, 2021, 7, 38–63.
43. H. Zhang, C. Wang, H. Luo, J. Chen, M. Kuang and J. Yang, Angew. Chem., Int. Ed., 2023, 62, e202217071.
44. C. Ban, Y. Wang, Y. Feng, Z. Zhu, Y. Duan, J. Ma, X. Zhang, X. Liu, K. Zhou and H. Zou, Energy Environ. Sci., 2024, 17, 518–530.
45. J. Ma, D. Wu, Y. Feng, C. Ban, L. Xia, L. Ruan, J. Guan, Y. Wang, J. Meng, J. Dai, L. Gan and X. Zhou, Nano Energy, 2023, 115, 108719.
46. J. Chen, F. Zhang, M. Kuang, L. Wang, H. Wang, W. Li and J. Yang, Proc. Natl. Acad. Sci. U. S. A., 2024, 121, e2318853121.
47. Y. Shi, G. Zhan, H. Li, X. Wang, X. Liu, L. Shi, K. Wei, C. Ling, Z. Li and H. Wang, Adv. Mater., 2021, 33, 2100143.
48. Y. Shi, J. Li, C. Mao, S. Liu, X. Wang, X. Liu, S. Zhao, X. Liu, Y. Huang and L. Zhang, Nat. Commun., 2021, 12, 5923.
49. J. Zhao, Y. Mu, L. Wu, Z. Luo, L. Velasco, M. Sauvan, D. Moonshiram, J. Wang, M. Zhang and T. Lu, Angew. Chem., Int. Ed., 2024, 136, e202401344.
50. R. Fang, Z. Yang, J. Sun, C. Zhu, Y. Chen, Z. Wang and C. Xue, J. Mater. Chem. A, 2024, 12, 3398–3410.
51. C. Ban, Y. Wang, J. Ma, Y. Feng, X. Wang, S. Qin, S. Jing, Y. Duan, M. Zhang, X. Tao, L. Gan and X. Zhou, Chem. Eng. J., 2024, 488, 150845.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta02199a

‡ These authors contributed equally.

---