Atomic-layer-deposited Pd/Cu₂O/Cu heterojunction as a high-efficiency formate catalyst for boosting aqueous Zn–CO₂ batteries

Abstract

The electrochemical reduction of carbon dioxide (CO₂RR) into value-added chemicals offers a sustainable pathway for mitigating greenhouse gas emissions and advancing carbon-neutral energy technologies. Among all possible CO₂RR products, formate/formic acid is particularly attractive due to its industrial relevance as a promising fuel for fuel cells and its high stability as an energy-storage medium. However, it remains challenging to obtain Cu-based catalysts with high selectivity and stability for CO₂ conversion to formate/formic acid. Herein, we report a metal Pd anchored onto a Cu₂O/Cu heterojunction (Pd–Cu₂O/Cu), synthesized via atomic layer deposition followed by electrochemical reduction. Structural and spectroscopic analyses confirmed that the presence of Pd atoms in the Cu₂O/Cu heterojunction collectively modulated the electronic states, stabilized *OCHO intermediates, suppressed the competing *H pathway, and balanced the *COOH pathway. Electrocatalytic tests of Pd–Cu₂O/Cu in an H-type cell showed a remarkable formate faradaic efficiency of 61.1% with a high partial current density of −24 mA cm⁻² at −0.9 V versus RHE. When integrated into an aqueous Zn–CO₂ battery, the catalyst outperformed most reported Cu-based systems, achieving a maximum power density of 2.54 mW cm⁻² and stable cycling for 130 hours. Overall, this work demonstrates a highly efficient Zn–CO₂ battery technology, highlighting the practical potential of coupling CO₂ utilization with sustainable energy storage.

---

1. Introduction

The efficient electroreduction of carbon dioxide (CO₂) into value-added chemicals and fuels offers a promising approach to mitigate global warming and establish a sustainable, carbon-neutral energy cycle.^1,2 Among various CO₂ reduction reaction (CO₂RR) pathways—such as C₁ products (CO, formate, CH₄, etc.),^3,4 C₂ products (C₂H₄, CH₃CH₂OH, etc.),^5,6 and C₃ products (n-propanol, etc.),^7,8 the conversion of CO₂ to formate (HCOO⁻)/formic acid (HCOOH) has emerged as one of the most promising routes, due to its substantial market potential as an industrial feedstock. In parallel, considering both the thermodynamic and kinetic energy barriers, the energy consumption for electroreduction of CO₂ to formic acid [CO₂ + 2H⁺ + 2e⁻ → HCOOH, −0.19 V versus the reversible hydrogen electrode (RHE)] is relatively less than other products such as CH₃OH and CH₄. Therefore, designing catalysts that can selectively promote the formic acid reaction pathway has become a central focus in CO₂RR research.

Significant progress has been made in developing CO₂RR catalysts, including Pd, Pb, In, Sn, and Cu₂O, as well as their derivatives, which demonstrate enhanced electrocatalytic activity for formic acid production.^9–12 Cu-based catalysts have gained attention in the CO₂RR due to the ease of processing and relatively low cost. For instance, Loiudice et al.¹³ reported that Cu nanocubes exhibited a faradaic efficiency (FE) of 15% (at −1.1 V vs. RHE) and a j_HCOO⁻ (HCOO⁻'s current density) value of ∼1 mA cm⁻² for formate. Despite these advances, Cu-based catalysts still suffer from insufficient selectivity toward formate due to their moderate binding energies with key intermediates. To address this challenge, strategies such as alloying, defect engineering, and single-atom dispersion have been proposed to fine-tune the catalytic sites and improve formate selectivity.^14,15 Recently, Zeng et al.¹⁶ regulated the first protonation step of the CO₂RR by building the single-atom Pb-alloyed Cu catalyst (Pb₁Cu), which improved to ∼96% faradaic efficiency of formate with high activity in excess of 1 A cm⁻².

Beyond elemental modification, the catalytic efficiency of Cu-based systems is also governed by particle size, structural reconstruction, and lattice facet exposure.^16,17 Specific crystallographic orientations can direct product selectivity: the Cu(111) facet favors C₁ products such as CO and formate, while the Cu(110) and Cu(100) facets promote C₂ products including ethylene (C₂H₄).¹⁸ In addition, oxide-derived Cu catalysts have been shown to significantly reduce the overpotential and enhance product selectivity compared with conventional polycrystalline metallic Cu, owing to their unique surface defects and residual oxygen species. For example, Qiao et al.¹⁹ developed a free-standing dendritic Cu heterojunction precatalyst (Cu₂O@Cu) via a facile electrochemical etching method, which underwent extensive structural reconstruction to form stepped Cu sites on a hierarchical dendritic array. The reconstructed Cu₂O@Cu exhibited up to ∼50% formate selectivity at a low potential of −0.6 V (vs. RHE). Similarly, Wang et al.²⁰ reported a Cu₂O/CuS composite catalyst that achieved an FE of 67.6% with a partial current density of 15.3 mA cm⁻² at −0.9 V (vs. RHE) for formate production, and maintained an average FE of 62.9% for at least 30 h under the same conditions, demonstrating both high activity and stability. Taken together, these examples illustrate the great potential of engineered Cu-based catalysts for formate production, while underscoring the ongoing challenge of achieving both high selectivity and long-term stability.

In parallel with the exploration of standalone CO₂RR electrocatalysts, researchers have sought to integrate CO₂ reduction into rechargeable battery systems, thereby coupling energy conversion with energy storage. Metal–CO₂ batteries, such as Li–CO₂ and Na–CO₂, have been widely investigated and have demonstrated high energy densities and improved cycling performance.²¹ Nevertheless, the practical application of these systems has been limited by their reliance on toxic organic electrolytes, strict anhydrous and oxygen-free assembly conditions, accumulation of carbonaceous byproducts, and high operational costs. To overcome these drawbacks, aqueous Zn–CO₂ batteries (ZCBs) have been proposed as an attractive alternative. These systems not only provide energy output while consuming CO₂, but also enable the circular generation of valuable products such as CO, hydrocarbons, and HCOOH, which can be harvested as fuels and industrial feedstocks.²² Despite these advantages, ZCBs remain relatively underexplored, with limited reported studies, all of which typically provide a discharge voltage below 0.5 V,^23–25 too low for widespread application. Therefore, further progress hinges on the rational design of cathode catalysts that can efficiently drive the CO₂RR with high activity and selectivity, thereby enhancing the discharge voltage and narrowing the overall voltage gap.

Herein, we report a sustainable and reversible aqueous Zn–CO₂ battery employing a Pd-anchored Cu₂O/Cu heterojunction (Pd–Cu₂O/Cu) cathode paired with a Zn foil anode. The Pd–Cu₂O/Cu heterojunction was prepared by depositing Pd on Cu₂O nanocubes via atomic layer deposition (ALD) followed by electrochemical reduction, yielding Pd-modified Cu₂O/Cu with increased surface area and enhanced catalytic activity. As a result, Pd–Cu₂O/Cu achieved a high formate faradaic efficiency (FE) of 61.1% and a large partial current density of −24 mA cm⁻² at −0.9 V (vs. RHE) in 0.5 M KHCO₃ within an H-type cell. When integrated into a Zn–CO₂ battery, the cathode delivered a maximum power density of 2.54 mW cm⁻² and a high current density of 8.5 mA cm⁻², and a discharge potential of 0.65 V, maintaining stable operation for over 200 cycles (130 h). This work presents a highly efficient Zn–CO₂ battery platform with significant sustainability advantages, demonstrating the practical potential for direct CO₂ utilization in energy storage systems.

2. Results and discussion

The synthesis strategy of the Pd–Cu₂O/Cu catalyst is shown in Fig. 1a. The catalyst was fabricated through a modified wet-chemical reduction of a Cu₂O precursor, followed by atomic layer deposition (ALD) of Pd and the subsequent electrochemical reduction. In a typical wet-chemical reduction process, CuCl₂ aqueous solution and NaOH first reacted to form Cu(OH)₂, which was subsequently reduced to Cu₂O using ascorbic acid.²⁶ 100 ALD cycles of Pd were then deposited onto Cu₂O (denoted as Cu₂O@Pd-100) using Pd(hfac)₂ and plasma H₂ as precursors in a plasma-enhanced ALD system. The Pd content in Cu₂O@Pd-100 was determined to be 0.52 wt% by inductively coupled plasma optical emission spectroscopy (ICP-OES). Prior to electrochemical measurements, the electrodes were reduced to form Pd–Cu₂O/Cu by electrochemical reduction for 5 min in 0.5 M KHCO₃ at −0.5 V (vs. RHE). Cu₂O/Cu was obtained using a Cu₂O precursor by the same electrochemical reduction process.

Fig. 1 Synthesis route and physical characteristics of Pd–Cu₂O/Cu. (a) Schematic illustration of Pd–Cu₂O/Cu; SEM images of (b) Cu₂O, (c) Pd–Cu₂O/Cu; (d) TEM image of Cu₂O/Cu–Pd; (e and f) HR-TEM images of Pd–Cu₂O/Cu; (g) AC-HAADF-STEM image of Pd–Cu₂O/Cu; (h) HAADF-TEM image and corresponding elemental mapping of Pd–Cu₂O/Cu and (i) SAED pattern of Pd–Cu₂O/Cu.

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were employed to investigate the microstructure and lattice characteristics of the as-prepared Cu₂O, Cu₂O/Cu, and Pd–Cu₂O/Cu. As shown in Fig. 1b and S1a, pristine Cu₂O exhibited a well-defined cubic morphology with an average particle size of 200–300 nm. High-resolution TEM (HR-TEM) images of Cu₂O revealed lattice fringe spacings of 0.15 and 0.213 nm, corresponding to the (200) and (220) facets of Cu₂O (PDF #01-77-0199), respectively (Fig. S1b). Raman spectra obtained from Cu₂O powder show four characteristic peaks at 148, 218, 415, and 625 cm⁻¹ (Fig. S2), assignable to Γ⁻₁₅, 2Γ⁻₁₂, 4Γ⁻₁₅, and Γ⁻₁₂ + Γ⁻₁₅ phonon modes of Cu₂O, respectively.²⁷ Together, these indicate the successful production of Cu₂O cubic structures suitable for ALD Pd coating in the next step.

Following ALD of Pd onto Cu₂O, SEM images of Pd-coated Cu₂O (Cu₂O@Pd-100) showed retention of the cubic Cu₂O morphology (Fig. S3). After electrochemical reduction, the parent Cu₂O crystal facets were well-preserved, but with increased surface irregularity, as seen in the SEM and TEM images of Pd–Cu₂O/Cu (Fig. 1c and d). HR-TEM analysis revealed lattice fringes of 0.21, 0.18, and 0.30 nm, corresponding to the Cu (111), Cu (200), and Cu₂O (110) facets (Cu₂O PDF #01-77-0199; Cu PDF #00-004-0836), respectively (Fig. 1e). The coexistence of lattice mismatch and abundant lattice defects in the Pd–Cu₂O/Cu catalyst was further confirmed by HR-TEM (Fig. 1f). Furthermore, the atomic dispersion of Pd in Pd–Cu₂O/Cu was confirmed using aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (AC-HAADF-STEM). The lighter dots, highlighted by blue circles in Fig. 1g, were identified as Pd species. The high-angle annular dark field TEM (HAADF-TEM) image and corresponding elemental mapping (Fig. 1h) demonstrated the uniform distribution of Cu and Pd throughout the Pd–Cu₂O/Cu. The selected area electron diffraction (SAED) pattern of Pd–Cu₂O/Cu (Fig. 1i) displayed concentric diffraction rings indexed to Cu (220)/(111) and Cu₂O (110)/(111), corroborating the coexistence of Cu and Cu₂O phases after electroreduction. Similarly, the Cu₂O/Cu sample maintained its cubic morphology (Fig. S4a). HR-TEM images of Cu₂O/Cu exhibited lattice spacings of 0.21 and 0.25 nm, consistent with the Cu (111) and Cu₂O (111) facets, respectively (Fig. S4b).

The phase composition and surface chemical states of the catalysts were further investigated. In the XRD patterns of Cu₂O/Cu and Pd–Cu₂O/Cu (Fig. 2a), diffraction peaks at 29.6°, 36.5°, 42.4°, and 61.5° correspond to the (110), (111), (200), and (220) planes of Cu₂O (PDF #01-77-0199), while the characteristic peaks at 44.4° and 54.6° are indexed to the (101) and (004) planes of carbon (PDF #00-023-0064) from the gas diffusion electrode (GDE). As shown in Fig. 2b, new diffraction peaks at 43.3°, 50.4°, and 74.1° emerged in both Cu₂O/Cu and Pd–Cu₂O/Cu nanocubes, and correspond to the (111), (200), and (220) planes of metallic Cu (PDF #00-004-0836). The persistence of the Cu₂O (111) peak at 36.5° indicates that Cu₂O was only partially reduced to metallic Cu during the electrochemical reduction.

Fig. 2 Physical and chemical characteristics of Cu₂O, Cu₂O@Pd100, and Pd–Cu₂O/Cu. (a) XRD pattern of Cu₂O and Cu₂O@Pd100; (b) XRD pattern of Cu₂O/Cu and Pd–Cu₂O/Cu; (c) Cu 2p XPS spectra of Cu₂O/Cu and Pd–Cu₂O/Cu; (d) Cu LMM Auger spectra of Cu₂O/Cu and Pd–Cu₂O/Cu; (e) O 1s and (f) Pd 3d XPS spectra of Cu₂O/Cu and Pd–Cu₂O/Cu.

X-ray photoelectron spectroscopy (XPS) was performed to reveal the surface valence states and chemical bonding of Cu₂O/Cu and Pd–Cu₂O/Cu. In the Cu 2p spectra (Fig. 2c), the peaks at 932.1 and 952.0 eV correspond to Cu 2p_3/2 and Cu 2p_1/2, respectively, consistent with Cu⁰ or Cu⁺ species.²⁸ The additional peaks at 934.2 and 954.4 eV, together with shake-up satellites near 942 eV, reveal the presence of Cu²⁺, which is attributed to Cu(II) carbonate dihydroxide formed by air oxidation after electrochemical reduction in 0.5 M KHCO₃.²⁹ Cu LMM Auger electron spectroscopy (AES, Fig. 2d) showed dominant peaks near 919.2 eV and 917.6 eV, characteristic of Cu⁰ and Cu⁺, further confirming the partial reduction of Cu₂O to Cu₂O/Cu after electroreduction.³⁰ The O 1s spectrum displayed three distinct components: lattice oxygen (Cu–O, 530.5 eV), oxygen vacancies (O_v, 531.5 eV), and chemisorbed oxygen species (O_c, 532.8 eV). These results are consistent with the vacancy defects observed in HR-TEM (Fig. 2e).³¹ Notably, the content of oxygen vacancies in Pd–Cu₂O/Cu (47%) was higher than in Cu₂O/Cu (43%), calculated from the O 1s fitting XPS spectra. According to a previous report, the synergistic effect between facet-oriented Cu₂O and enriched oxygen vacancies has been proposed to facilitate CO₂ electroreduction to formate.²⁹ In the Pd 3d XPS spectrum of Pd–Cu₂O/Cu (Fig. 2f), the peaks at 336.8 and 343.4 eV are assigned to the peak of Pd²⁺.^32,33 These findings confirm the absence of Pd nanoparticles and the successful formation of isolated Pd single atoms on the surface of the Pd–Cu₂O/Cu catalyst.

We studied the CO₂RR properties of the GDE-supported Cu₂O/Cu and Pd–Cu₂O/Cu catalysts (active GDE area, 1 cm²; catalyst loading, 0.5 mg cm⁻²) in an H-cell setup with a CO₂-and N₂-saturated 0.5 M KHCO₃ electrolyte. All data were recorded after electrochemical reduction and 20 cycles of potential scans over the range of 0 to −0.4 V (vs. RHE) at a scan rate of 20 mV s⁻¹, to reduce Cu₂O to Cu⁰ and a small quantity of metastable Cu⁺. Linear sweep voltammetry (LSV) measurements were performed in the electrolyte at a scan rate of 10 mV s⁻¹ both in the N₂- and CO₂-saturated 0.5 M KHCO₃ aqueous solutions. As shown in Fig. 2a, the onset potential of both catalysts in CO₂-saturated KHCO₃ solution shifted to the positive direction compared with that in the N₂-saturated KHCO₃ solution. In the CO₂-saturated 0.5 M KHCO₃ aqueous solutions, Pd–Cu₂O/Cu provides a more positive onset potential (−0.49 V vs. RHE at j = 10 mA cm⁻²) than Cu₂O/Cu (−0.58 V vs. RHE at j = 10 mA cm⁻²). The current density of the Pd–Cu₂O/Cu catalyst increased rapidly as the potential became increasingly negative and reached approximately 32 and 43 mA cm⁻² at −0.9 and −1.1 V (vs. RHE), respectively. In Fig. 3b and c, the electrochemical surface area (ECSA) of the as-obtained samples is evaluated using the double-layer capacitance (C_dl) method. The Pd–Cu₂O/Cu catalyst exhibited a C_dl value of 10.06 mF cm⁻² (Fig. 3d), which is almost double that of Cu₂O/Cu (5.83 mF cm⁻²), indicating a greater exposure of active sites in Pd–Cu₂O/Cu. In Fig. S5, the Tafel slope of Pd–Cu₂O/Cu (412 mV dec⁻¹) is higher than that of Cu₂O/Cu (505 mV dec⁻¹), implying that Pd–Cu₂O/Cu has a high reaction rate, which was significantly improved by CO₂ activation.

Fig. 3 CO₂RR performance of Cu₂O/Cu and Pd–Cu₂O/Cu. (a) LSV curves of Cu₂O/Cu and Pd–Cu₂O/Cu in a H-type cell with CO₂-saturated and N₂-saturated, respectively. CV curves measured at 20, 50, 100, 150 and 200 mV s⁻¹ in a H-type cell; (b): Cu₂O/Cu and (c): Pd–Cu₂O/Cu; (d) the C_dl values of Cu₂O/Cu and Pd–Cu₂O/Cu. Product distribution of Cu₂O/Cu and Pd–Cu₂O/Cu at (e) −0.9 V (vs. RHE) and (f) −1.1 V (vs. RHE). (g) Comparison of FE_formate for Cu₂O/Cu and Pd–Cu₂O/Cu catalysts. (h) Stability test of Pd–Cu₂O/Cu at −0.9 V (vs. RHE).

The distribution of gaseous (i.e., CO, CH₄, and C₂H₂) and liquid products (i.e., formate (HCOO⁻), ethane-diol (CH₃CH(OH)₂), acetic acid (CH₃COOH), and acetaldehyde (CH₃CHO)) obtained during the CO₂RR process was assessed using gas chromatography (GC) and ex situ nuclear magnetic resonance (NMR) in 0.5 M KHCO₃ electrolyte. Fig. 3e and f present the obtained faradaic efficiencies (FEs) of products with Cu₂O/Cu and Pd–Cu₂O/Cu at −0.9 and −1.1 V (vs. RHE), respectively. As the applied potential varied, the Cu₂O/Cu catalyst became more negative, the FE for formate decreased from about 48.2% to 40.1% under −0.9 and −1.1 V (vs. RHE) with a slight increase in H₂. The FE for formate production with different catalysts is compared in Fig. 3g. Compared with Cu₂O/Cu, the FE for formate on Pd–Cu₂O/Cu increases, reaching 61.1% at −0.9 V (vs. RHE). The improvement may result from the structural reconstruction of the catalyst and a synergistic effect between Pd modification and the catalytically active Cu sites during the CO₂RR.^19,34,35 As the applied potential becomes increasingly negative, the FE for formate decreases and remains over 40% when the potential is more negative than −1.1 V (vs. RHE), whereas that for H₂ increases to 54.8%. The chronoamperometry (CA) curves and ex situ ¹H NMR spectra are shown in Fig. S6 and S7. Although Cu₂O or Cu catalysts have been reported to promote the formation of C₂H₄ during the CO₂ reduction reaction, the generation of C₂₊ products typically requires highly negative potentials under alkaline conditions, where the elevated pH can lower the C–C coupling energy barrier while simultaneously suppressing competing reactions such as H₂ and CH₄ evolution.^36,37 In published reports, Cu-derived heterojunction electrocatalysts have been shown to produce liquid products, particularly formate.^38,39 The Pd–Cu₂O/Cu catalyst favors the production of formic acid, probably due to the incorporation of Pd, which enhances interfacial charge redistribution within the catalyst and modulates its d-band electronic structure.⁴⁰ Furthermore, during continuous electrocatalysis at −0.9 V (vs. RHE) (Fig. 3h), Pd–Cu₂O/Cu maintains a high stability with minimal potential degradation for 12 h, which was better than most reported Cu-based electrocatalysts, as summarized in Table S1. As shown in Fig. S8, after the stability test for 12 h, the FE for formate decreased from 61.1% to 54.2%, and 88.7% FE retention was achieved. The typical cubic morphology of Pd–Cu₂O/Cu had been retained after the long-term electrolysis (Fig. S9). This indicated the superiority of Pd–Cu₂O/Cu for the selective production of formic acid.

To gain further insights into the underlying mechanisms, density functional theory (DFT) calculations were employed. The optimized models of Cu₂O/Cu and Pd–Cu₂O/Cu are constructed in Fig. S10. As depicted in Fig. 4a and b, the partial density of states (PDOS) analyses for Cu 3d orbitals revealed that Pd–Cu₂O/Cu had an elevated d-band center −1.87 eV, being closer to the Fermi level than that of the Cu site in Cu₂O/Cu. The higher d-band center of the Cu led to reinforced adsorption behaviors of the Pd–Cu₂O/Cu surface.^41,42 The differential charge density and Bader charge were simulated (Fig. 4c, d and S11). The positive Δρ (yellow) and negative Δρ (blue) represent electron accumulation and depletion regions, respectively. The Bader charge of Pd–Cu₂O/Cu is higher than that of Cu₂O/Cu, in which the Pd atom doping on Pd–Cu₂O/Cu promotes electron transfer from the Pd–Cu₂O/Cu surface to the intermediates, leading to high-density electron depletion regions on the Pd side and indicating a strong electronic interaction with the adsorbates. During the CO₂RR process, the active sites and electrons on the surface might be occupied and consumed by intermediates of multiple reactions, leading to a decline in the catalytic efficiency of the CO₂RR. Hence, the adsorption energy of *H, *OCHO and *COOH on different catalyst surfaces was calculated. As shown in Fig. 4e–g and S12, S13, the adsorption energy of *H on Pd–Cu₂O/Cu is 0.33 eV more than 0 eV (Fig. 4e), and lower than that of Cu₂O/Cu, suggesting that Pd atom doping is beneficial for effectively restraining the HER through a stronger interaction with adsorbed *H. In Fig. 4f, the adsorption energy of *OCHO intermediates on Pd–Cu₂O/Cu is −2.12 eV, lower than that of Cu₂O/Cu (−0.71 eV), implying that Pd–Cu₂O/Cu could effectively promote the selectivity of formate through stronger interaction with adsorbed *OCHO intermediates. In addition, considering the intermediate to produce CO, the adsorption energy of the *COOH intermediate on Pd–Cu₂O/Cu is −2.39 eV, lower than that of Cu₂O/Cu (−1.93 eV), suggesting that Pd–Cu₂O/Cu could effectively enhance the efficiency of CO (Fig. 4g). In other words, the Pd–Cu₂O/Cu catalyst enhances the adsorption process of *OCHO and *COOH intermediates. It could be contradictory that the Pd–Cu₂O/Cu could increase the efficiency of formate similar to CO in a thermodynamical process. Fig. 4i displays the Gibbs free energy profiles during the CO₂RR on the surface of Pd–Cu₂O/Cu, accompanied by the profiles for Cu₂O/Cu for purposes of comparison. The ΔG value associated with the rate-determining step (RDS) is positive and highest throughout the entire CO₂RR process.^43,44 Notably, the transition from CO₂ to *OCHO, from *OCHO to *HCOOH represents the RDS for Pd–Cu₂O/Cu and Cu₂O/Cu during the CO₂RR, respectively. It is evident that the reaction free energy for the RDS is lower on Pd–Cu₂O/Cu (0.39 eV) compared to Cu₂O/Cu (0.47 eV). This suggests that the incorporation of Pd atoms into Cu₂O/Cu reduces the reaction energy barrier of the RDS, thereby facilitating the formate production process during the CO₂RR. Moreover, the ΔG *H on Pd–Cu₂O/Cu is 0.43 eV, which is lower than that of Cu₂O/Cu (1.03 eV), but more than 0 eV, confirming that Pd–Cu₂O/Cu significantly suppresses its competing HER across a broad potential window to optimize the CO₂RR process. Although *COOH is thermodynamically more stable (−2.39 eV), due to the energy barrier from CO₂ to *OCHO being only 0.39 eV and the rapid conversion of *OCHO to HCOOH, the reaction kinetics effectively bypass the *COOH pathway, thereby avoiding competition from the CO pathway. Thus, combining theoretical and experimental analyses, it is strongly supported that the inhibition of parallel reactions on the Pd–Cu₂O/Cu surface plays a crucial role in achieving efficient CO₂ reduction to formate.

Fig. 4 Theoretical calculations. PDOS of Cu 3d orbitals for (a) Cu₂O/Cu and (b) Pd–Cu₂O/Cu; differential charge density and Bader charge simulations of (c) Cu₂O/Cu and (d) Pd–Cu₂O/Cu; the calculated adsorption energy of (e) *H, (f) *OCHO and (g) *COOH on Cu₂O/Cu and Pd–Cu₂O/Cu, respectively; the free energy diagram of producing (h) H₂ and (i) formic acid on Cu₂O/Cu and Pd–Cu₂O/Cu, respectively.

Given the excellent catalytic performance of Pd–Cu₂O/Cu toward the CO₂RR, a Zn–CO₂ battery (ZCB) was further investigated as a promising energy conversion system for CO₂ utilization. The Zn–CO₂ battery was assembled in a flow cell with a GDE, and the Pd–Cu₂O/Cu catalyst and a Zn plate were applied as the cathode and anode, respectively. The cell configuration and the working principle of a Zn–CO₂ battery are schematically illustrated in Fig. 5a. During discharge, Zn is oxidized to Zn²⁺ at the anode with the release of electrons, while CO₂ reduction occurs at the cathode. Conversely, during charging, Zn²⁺ at the anode is reduced back to metallic Zn, accompanied by the oxygen evolution reaction (OER) at the cathode. As shown in Fig. 5b, the open-circuit voltage (OCV) of the ZCB was 1.41 V and remained stable for 10 min. The discharge polarization and corresponding power density curves are presented in Fig. 5c and S14. The peak power density of the ZCB with the Pd–Cu₂O/Cu cathode was determined to be 2.54 mW cm⁻² at a high current density of 8.5 mA cm⁻², which was much higher than that for the Cu₂O/Cu (1.32 mW cm⁻²) cathodic catalyst, outperforming several recently reported catalysts (Fig. 5d and Table S2).^45–53 Furthermore, at discharge current densities of 0.5, 1.0, 2.0, and 3.0 mA cm⁻², the ZCB employing the Pd–Cu₂O/Cu cathode exhibited stable voltage plateaus of approximately 1.17, 1.07, 0.78, and 0.62 V, respectively, over a period of one hour. When the discharge current density was reduced to 0.5 mA cm⁻², the stable voltage plateau of the ZCB with the Pd–Cu₂O/Cu cathode reached approximately 1.21 V, indicating favorable discharge characteristics of this battery (Fig. S15). At a constant current density of 0.5 mA cm⁻² (Fig. 5e and S16), the charge–discharge cycles and corresponding discharge efficiency of ZCB with the Pd–Cu₂O/Cu cathode were each set to 20 min (total 40 min per cycle). The ZCB device exhibits a charge potential of 2.62 V and a discharge potential of 0.65 V, while delivering excellent cycling stability for over 200 cycles and maintaining robust operational performance for 135 h without noticeable degradation. No significant change in the typical morphology or composition of Pd–Cu₂O/Cu was observed after the long-term cycling of the ZCB (Fig. S17), indicating the favorable structural stability of the Pd–Cu₂O/Cu catalyst. Furthermore, as shown in the inset of Fig. 5b, a single aqueous rechargeable ZCB employing the Pd–Cu₂O/Cu cathode successfully powered an electronic timer, suggesting the practicability of this aqueous Zn–CO₂ battery.

Fig. 5 The electrochemical performance of the Zn–CO₂ battery (ZCBs). (a) Schematic illustration of the designed ZCBs with Pd–Cu₂O/Cu as the cathode; (b) open circuit voltage over time with the inset showing the digital photo of an electronic timer powered by one ZCB; (c) discharge curves and power density; (d) comparison of the obtained current and power density of our work with those of others, and (e) galvanostatic discharge–charge cycling curves at 0.5 mA cm⁻².

3. Conclusion

We developed a Pd anchored on a Cu₂O/Cu heterojunction (Pd–Cu₂O/Cu), which exhibited outstanding activity and selectivity for CO₂ electroreduction. It was found that the introduction of Pd induced electron localization and lattice strain at the Cu₂O/Cu interface, favoring the stabilization of key intermediates (e.g., *OCHO) while suppressing competitive *H and *COOH adsorption pathways. As a result, the Pd–Cu₂O/Cu catalyst achieved a high faradaic efficiency for CO₂ reduction to formate of 61.1% at −0.9 V (vs. RHE), along with a large partial current density of −24 mA cm⁻² in H-type cell tests. When integrated into an aqueous Zn–CO₂ battery, Pd–Cu₂O/Cu delivered superior performance with an ultralow loading of 0.5 mg cm⁻², sustaining high current (8.5 mA cm⁻²), power densities (2.54 mW cm⁻²), and discharge voltage (0.65 V) under continuous CO₂ purging-values, surpassing the performance of most reported catalysts. This work not only demonstrates a highly efficient route to valorize excess atmospheric CO₂, but also establishes a sustainable pathway toward advanced energy-storage devices that couple CO₂ utilization with renewable energy integration, simultaneously mitigating CO₂ emissions and advancing electrification.

Author contributions

Jia Xu: writing – original draft, synthesis, methodology, investigation, formal analysis, data curation, conceptualization. Li Tao, Yue Zhang and Evan J. Hansen: validation, writing – review and editing, methodology, investigation. Robert Godin: writing – review and editing, supervision, validation, investigation. Jian Liu: writing – review and editing, supervision, project administration, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: TEM images and Raman spectrum of Cu₂O; SEM images of Cu₂O@Pd-100; TEM and HR-TEM images of Cu₂O/Cu; Tafel slopes and chronoamperometry (CA) curves of Cu₂O/Cu and Pd-Cu2O/Cu; ¹H NMR spectra of Pd-Cu₂O/Cu before and after stability test; TEM, HRTEM and AC-HAADF-STEM images of Pd-Cu₂O/Cu after stability test; top view, side view, and differential charge density of Cu₂O/Cu and Pd-Cu₂O/Cu models; schematic diagram of *H, *OCHO and *COOH adsorbed on Cu₂O/Cu and Pd-Cu₂O/Cu; galvanostatic discharge curves at different current densities and discharge efficiency of ZCBs; TEM, HRTEM and AC-HAADF-STEM images of Pd-Cu₂O/Cu as the cathode after ZCBs cycling. See DOI: https://doi.org/10.1039/d6ta00814c.

Acknowledgements

This research was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC), Canada Foundation for Innovation (CFI), BC Knowledge Development Fund (BCKDF), the University of British Columbia (UBC), and UBC Eminence Program (Net Zero Research Excellence Cluster and Battery Innovation Research Excellence Cluster). The authors are grateful to Dr Carmen Andrei for HRTEM characterization at the Canadian Centre for Electron Microscopy (CCEM) at McMaster University, and Dr Shihong Xu and Dr Peng Li for XPS characterization at NanoFAB at the University of Alberta.

References

1. S. Hao, A. Elgazzar, S. K. Zhang, T. U. Wi, F. Y. Chen, Y. Feng, P. Zhu and H. Wang, Science, 2025, 288, 1182.
2. W. Ma, J. Morales-Vidal, J. Tian, M. T. Liu, S. Jin, W. Ren, J. Taubmann, C. Chatzichristodoulou, J. Luterbacher, H. M. Chen, N. López and X. Hu, Nature, 2025, 641, 1156–1161.
3. W. Zhang, A. Mehmood, G. Ali, H. Liu, L. Chai, J. Wu and M. Liu, Angew. Chem., Int. Ed., 2025, 64, e202424552.
4. C. Wang, Z. Lv, Y. Liu, R. Liu, C. Sun, J. Wang, L. Li, X. Liu, X. Feng, W. Yang and B. Wang, Angew. Chem., Int. Ed., 2024, 63, e202404015.
5. D. Zhou, C. Chen, Y. Zhang, M. Wang, S. Han, X. Dong, T. Yao, S. Jia, M. He, H. Wu and B. Han, Angew. Chem., Int. Ed., 2024, 63, e202400439.
6. W. Choi, Y. Chae, E. Liu, D. Kim, W. S. Drisdell, H. S. Oh, J. H. Koh, D. K. Lee, U. Lee and D. H. Won, Nat. Commun., 2024, 15, 8345.
7. M. Choi, S. Bae, Y. Kim, Y. Lee, M. Cho, S. Kang and J. Lee, Nat. Catal., 2025, 8, 476–486.
8. K. Qi, Y. Zhang, N. Onofrio, E. Petit, X. Cui, J. Ma, J. Fan, H. Wu, W. Wang, J. Li, J. Liu, Y. Zhang, Y. Wang, G. Jia, J. Wu, L. Lajaunie, C. Salameh and D. Voiry, Nat. Catal., 2023, 6, 319–331.
9. Y. Wei, X. Wang, J. Mao, Y. Song, H. Zhu, X. Liu, C. Luo, S. Li, A. Chen, G. Li, X. Dong, W. Wei and W. Chen, Angew. Chem., Int. Ed., 2025, 137, e202423370.
10. Y. Shi, Y. Ji, J. Long, Y. Liang, Y. Liu, Y. Yu, J. Xiao and B. Zhang, Nat. Commun., 2020, 11, 3415.
11. B. Jia, Z. Chen, C. Li, Z. Li, X. Zhou, T. Wang, W. Yang, L. Sun and B. Zhang, J. Am. Chem. Soc., 2023, 145, 14101–14111.
12. M. Wang, J. Jia, J. Xia, C. K. Peng, J. He, Y. Qiu, Y. He, L. Gao, F. Xue, Y. G. Lin, G. Zhan, Y. Guo, X. Huang and Y. Xu, Nat. Commun., 2025, 16, 6454.
13. A. Loiudice, P. Lobaccaro, E. A. Kamali, T. Thao, B. H. Huang, J. W. Ager and R. Buonsanti, Angew. Chem., Int. Ed., 2016, 55, 5613–5868.
14. S. Nitopi, E. Bertheussen, S. B. Scott, X. Liu, A. K. Engstfeld, S. Horch, B. Seger, I. E. L. Stephens, K. Chan, C. Hahn, J. K. Norskov, T. F. Jaramillo and I. Chorkendorff, Chem. Rev., 2019, 119, 7610–7672.
15. J. Dong, Y. Liu, J. Pei, H. Li, S. Ji, L. Shi, Y. Zhang, C. Li, C. Tang, J. Liao, S. Xu, H. Zhang, Q. Li and S. Zhao, Nat. Commun., 2023, 14, 6849.
16. T. Zheng, C. Liu, C. Guo, M. Zhang, X. Li, Q. Jiang, W. Xue, H. Li, A. Li, C. W. Pao, J. Xiao, C. Xia and J. Zeng, Nat. Nanotechnol., 2021, 16, 1386–1393.
17. S. Liang, J. Xiao, T. Zhang, Y. Zheng, Q. Wang and B. Liu, Angew. Chem., Int. Ed., 2023, 135, e202310740.
18. Z. Z. Wu, X. L. Zhang, Z. Z. Niu, F. Y. Gao, P. P. Yang, L. P. Chi, L. Shi, W. S. Wei, R. Liu, Z. Chen, S. Hu, X. Zheng and M. R. Gao, J. Am. Chem. Soc., 2022, 144, 259–269.
19. S. Qiao, G. Zhang, D. Tian, W. Xu, W. Jiang, Y. Cao, J. Qian, J. Zhang, Q. He and L. Song, Energy Environ. Sci., 2024, 17, 6779–6786.
20. S. Wang, T. Kou, J. B. Varley, S. A. Akhade, S. E. Weitzner, S. E. Baker, E. B. Duoss and Y. Li, ACS Mater. Lett., 2020, 3, 100–109.
21. X. Zhang, N. Zhao, H. Zhang, Y. Fan, F. Jin, C. Li, Y. Sun, J. Wang, M. Chen and X. Hu, Nanomicro Lett, 2024, 17, 99.
22. J. Xu, M. Galib, Z. Wu, L. Tao, Y. Shao, Y. Zhang, X. Guo, E. J. Hansen, Y. Chen, Z. Wang, C. Liu, M. Ponga and J. Liu, Nano Energy, 2024, 132, 110373.
23. Z. Zeng, L. Y. Gan, H. Bin Yang, X. Su, J. Gao, W. Liu, H. Matsumoto, J. Gong, J. Zhang, W. Cai, Z. Zhang, Y. Yan, B. Liu and P. Chen, Nat. Commun., 2021, 12, 4088.
24. L. L. Ling, L. Jiao, X. Liu, Y. Dong, W. Yang, H. Zhang, B. Ye, J. Chen and H. L. Jiang, Adv. Mater., 2022, 34, e2205933.
25. J. Chen, Z. Li, X. Wang, X. Sang, S. Zheng, S. Liu, B. Yang, Q. Zhang, L. Lei, L. Dai and Y. Hou, Angew. Chem., Int. Ed., 2021, 61, e202111683.
26. D.-F. Zhang, H. Zhang, L. Guo, K. Zheng, X.-D. Han and Z. Zhang, J. Mater. Chem., 2009, 19, 5220.
27. K. Zhang, B. Li, F. Guo, N. Graham, W. He and W. Yu, Angew. Chem., Int. Ed., 2024, e202411796.
28. L. Wan, Q. Zhou, X. Wang, T. E. Wood, L. Wang, P. N. Duchesne, J. Guo, X. Yan, M. Xia, Y. F. Li, F. M. Ali, U. Ulmer, J. Jia, T. Li, W. Sun and G. A. Ozin, Nat. Catal., 2019, 2, 889–898.
29. G. Li, Y. Song, C. Zhu, X. Dong, W. Chen, G. Wu, G. Feng, S. Li and W. Wei, J. CO2 Util., 2023, 70, 102446.
30. Y. Gao, S. Yu, P. Zhou, X. Ren, Z. Wang, Z. Zheng, P. Wang, H. Cheng, Y. Liu, W. Wei, Y. Dai and B. Huang, Small, 2022, 18, e2105212.
31. J. Fu, F. M. Hassan, C. Zhong, J. Lu, H. Liu, A. Yu and Z. Chen, Adv. Mater., 2017, 29, 1702526.
32. Q. Wei, C. Yu, X. Song, Y. Zhong, L. Ni, Y. Ren, W. Guo, J. Yu and J. Qiu, J. Am. Chem. Soc., 2021, 143, 6071–6078.
33. G. Li, X. Wang, L. Yan, Y. Wang, Z. Zhang and J. Xu, ACS Appl. Mater. Interfaces, 2019, 11, 26116–26126.
34. R. Zhou, X. Fan, X. Ke, J. Xu, X. Zhao, L. Jia, B. Pan, N. Han, L. Li, X. Liu, J. Luo, H. Lin and Y. Li, Nano Lett., 2021, 21, 4092–4098.
35. J. Li, S. X. Guo, F. Li, F. Li, X. Zhang, J. Ma, D. R. MacFarlane, A. M. Bond and J. Zhang, ChemSusChem, 2019, 12, 4471–4479.
36. C. Lu, Q. He, S. Huang, P. Shi, C. Yang, J. Zhang, J. Zhu, J. Zhang, T. Wang and X. Zhuang, Adv. Mater., 2025, 37, e2415092.
37. S. Chen, C. Ye, Z. Wang, P. Li, W. Jiang, Z. Zhuang, J. Zhu, X. Zheng, S. Zaman, H. Ou, L. Lv, L. Tan, Y. Su, J. Ouyang and D. Wang, Angew. Chem., Int. Ed., 2023, 62, e202315621.
38. S. Wu, S. Li, Z. Hou, Y. Hu, Z. Zhang, J. Zhu, S. Xu, R. Wang, N. Zhang, L. An, P. Xi and C. H. Yan, Adv. Mater., 2025, 37, e2503772.
39. Y. Shi, Y. Wang, J. Yu, Y. Chen, C. Fang, D. Jiang, Q. Zhang, L. Gu, X. Yu, X. Li, H. Liu and W. Zhou, Adv. Energy Mater., 2023, 13, 2203506.
40. X. Zhang, L. Hui, F. He and Y. Li, J. Am. Chem. Soc., 2025, 147, 436–445.
41. F. Wang, H. Tian, B. Nan, G. Chen, K. Song, S. Wang, S. Chen, S. Yin, Z. Lu, W. Tu, W. Jiang, S. Wang and G. Liu, J. Am. Chem. Soc., 2026, 148(18), 19283–19292.
42. P. Chen, X. Liu, D. He, Y. Liao, H. Li, J. Cheng, Y. Lu, Z. Yang, Y. Yang and F. Jin, Nat. Commun., 2026, 17, 2395.
43. H. G. Qin, Y. F. Du, Y. Y. Bai, F. Z. Li, X. Yue, H. Wang, J. Z. Peng and J. Gu, Nat. Commun., 2023, 14, 5640.
44. X. Zhu, J. Huang and M. Eikerling, ACS Catal., 2021, 11, 14521–14532.
45. Y. Wang, Z. Huang, Y. Lei, J. Wu, Y. Bai, X. Zhao, M. Liu, L. Zhan, S. Tang, X. Zhang, F. Luo and X. Xiong, Chem. Commun. (Camb), 2022, 58, 3621–3624.
46. W. Zheng, J. Yang, H. Chen, Y. Hou, Q. Wang, M. Gu, F. He, Y. Xia, Z. Xia, Z. Li, B. Yang, L. Lei, C. Yuan, Q. He, M. Qiu and X. Feng, Adv. Funct. Mater., 2019, 30, 1907658.
47. T. Wang, X. Sang, W. Zheng, B. Yang, S. Yao, C. Lei, Z. Li, Q. He, J. Lu, L. Lei, L. Dai and Y. Hou, Adv. Mater., 2020, 32, e2002430.
48. Y. Bao, J. Xiao, Y. Huang, Y. Li, S. Yao, M. Qiu, X. Yang, L. Lei, Z. Li, Y. Hou, G. Wu and B. Yang, Angew. Chem., Int. Ed., 2024, 63, e202406030.
49. S. Gong, W. Wang, C. Zhang, M. Zhu, R. Lu, J. Ye, H. Yang, C. Wu, J. Liu, D. Rao, S. Shao and X. Lv, Adv. Funct. Mater., 2022, 32(17), 2110649.
50. J. Xie, X. Wang, J. Lv, Y. Huang, M. Wu, Y. Wang and J. Yao, Angew. Chem., Int. Ed., 2018, 57, 16996–17001.
51. P. Rao, X. Han, H. Sun, F. Wang, Y. Liang, J. Li, D. Wu, X. Shi, Z. Kang, Z. Miao, P. Deng and X. Tian, Angew. Chem., Int. Ed., 2024, 137, e202415223.
52. Y. Wang, L. Xu, L. Zhan, P. Yang, S. Tang, M. Liu, X. Zhao, Y. Xiong, Z. Chen and Y. Lei, Nano Energy, 2022, 92, 106780.
53. S. Yan, C. Peng, C. Yang, Y. Chen, J. Zhang, A. Guan, X. Lv, H. Wang, Z. Wang, T.-K. Sham, Q. Han and G. Zheng, Angew. Chem., Int. Ed., 2021, 133, 25789–26197.

---