Integrating hydrogen utilization in CO₂ electrolysis with reduced energy loss

The study addresses the high energy input and low efficiency barriers in low-temperature electrochemical CO₂ reduction (CO₂RR), especially the kinetically sluggish and energy-intensive oxygen evolution reaction (OER) at the anode that also induces severe carbon loss via CO₂ crossover and re-release at the anode. In typical anion-exchange membrane CO₂ electrolyzers, carbon loss can reach ~70%, and CO₂ recovery by amine scrubbing incurs 3–7 GJ per tonne CO₂, estimated to exceed the electrolysis energy by ~1.6×. Paired electrolysis strategies that replace OER with organic oxidations can reduce energy but suffer from market-size mismatch between CO₂-derived products and anodic coproducts and pose separation challenges. Given hydrogen’s central role as a renewable energy carrier and the fact that green hydrogen production via water electrolysis performs OER under more favorable thermodynamic and kinetic conditions (e.g., very low anodic overpotentials in SOEC/AWE), the authors hypothesize that integrating CO₂RR with hydrogen oxidation reaction (HOR) can lower overall energy consumption while mitigating carbon loss and catalyst poisoning. The purpose is to design and demonstrate a single-cell architecture coupling CO₂RR with HOR, using a Ni(OH)₂/NiOOH redox mediator to decouple reactions, minimize anodic losses, and quantify energy and polarization reductions relative to conventional CO₂RR.

Background literature highlights advances in CO₂RR catalysts and reactors achieving high-rate, selective conversion to C1–C3 products but constrained by energy efficiency due to OER. Prior paired electrolysis approaches oxidizing low-value organics (e.g., glycerol, alcohols, aldehydes) reduce anodic overpotentials and co-produce value-added chemicals, yet are limited by small market sizes compared to gigatonne-scale CO₂ utilization needs and by product separation complexity. Water electrolysis technologies (alkaline electrolyzers and solid oxide electrolysis cells) exhibit very low OER overpotentials under favorable conditions (e.g., ~0.01 V in SOEC and ~0.2 V in AWE at 50 mA cm⁻²), contrasted with >0.52 V in neutral CO₂RR cells. Techno-economic assessments indicate high energy penalties for CO₂ recovery from crossover and significant overall energy burdens in conventional systems. These insights motivate transferring the OER function from CO₂RR cells to dedicated water electrolyzers and pairing CO₂RR with HOR to achieve better kinetics and reduced system energy consumption.

System design: A single electrochemical cell integrates CO₂RR at the cathode with a Ni(OH)₂/NiOOH redox mediator and HOR at the anode. The mediator decouples the cathodic CO₂RR from anodic HOR to avoid OER, carbon loss, and HOR catalyst poisoning. Operating principle: Two alternating steps enable continuous operation.

• Step 1 (CO₂RR + Ni(OH)₂ oxidation, NIOR): At the CO₂RR gas-diffusion electrode (GDE), CO₂ is reduced to CO (CO₂ + H₂O + 2e⁻ → CO + 2OH⁻) or to formate (CO₂ + 2H₂O + 2e⁻ → HCOOH + 2OH⁻). At the mediator, Ni(OH)₂ is oxidized (Ni(OH)₂ + OH⁻ → NiOOH + H₂O + e⁻). Overall: CO₂ + 2Ni(OH)₂ → 2NiOOH + CO + H₂O or → 2NiOOH + HCOOH.
• Step 2 (NiOOH reduction + HOR): At the mediator, NiOOH is reduced (NiOOH + H₂O + e⁻ → Ni(OH)₂ + OH⁻), while the anode performs HOR (H₂ + 2OH⁻ → 2H₂O + 2e⁻). Overall: 2NiOOH + H₂ → 2Ni(OH)₂. This acts as a Ni–H₂ battery segment to harvest energy and offset Step 1 power. Catalysts and electrodes: For CO₂-to-CO, Zn nanosheets were electrodeposited on Cu foam (samples Zn-Cu-100/500/1000 by deposition time) to form a high-surface-area CO-selective catalyst. For CO₂-to-formate, porous Bi₂O₃ nanospheres were synthesized via a carbon-templated hydrothermal route and calcination. A gradient functional layer was developed to transform the Zn-Cu-500 foam into a robust GDE: a layer-by-layer drop-cast carbon/PTFE composite (varying carbon:PTFE ratios) formed gas diffusion channels and ionic/electronic pathways, improving active site accessibility, mass transport, contact, and mechanical integrity. For OER controls, nanostructured Co₃O₄ on Ni foam was prepared by solvothermal growth and calcination. HOR used commercial Pt/C GDE. Cell assembly: The custom cell comprised a CO₂RR GDE (1×1 cm²), a Ni(OH)₂/NiOOH mediator on Ni foam (2×2 cm²), and a Pt/C HOR GDE (1×1 cm²). A 1.5 cm PEEK frame separated CO₂RR GDE and mediator with a Hg/HgO reference. The mediator and HOR GDE were separated by a 130 μm porous separator pre-soaked in 1 M KOH. Both membrane-free and AEM-separated configurations were tested for gas collection and crossover studies. Electrolytes and operating conditions: Step 1 used CO₂ feed (~35 mL min⁻¹) with circulating 1 M KOH or 1 M KHCO₃ at ~20 mL min⁻¹. Step 2 used 6 M KOH and H₂ feed (~10 mL min⁻¹) at the anode. Polarization curves were recorded after stabilization; no iR compensation in full-cell tests. Product analysis: Gas products quantified by GC; liquid products by NMR with DMSO internal standard; FEs computed from flow and concentration. Electrode kinetics and mechanisms were probed by LSV, CV, Tafel analysis, and EIS (charge-transfer and Warburg components). DEMS monitored O₂ evolution to confirm OER suppression during Step 1. Control experiments replaced NIOR with OER (Co₃O₄) and tested direct CO₂RR-HOR coupling without mediator to assess HOR catalyst poisoning via FTIR (COad on Pt). Benchmarking hydrogen production: To incorporate upstream H₂ production energy, an alkaline water electrolyzer (AWE) and an electrode-supported SOEC were built and characterized to obtain their polarization behavior at 50 mA cm⁻² for system-level voltage and energy analyses. Data analysis: Overpotential partitions for half-reactions were compiled to compare conventional CO₂RR (with OER) versus H₂-integrated CO₂RR + AWE or + SOEC. Energy consumption per tonne product was estimated with and without accounting for CO₂ recovery energy due to crossover in neutral media. Preliminary techno-economic assessment considered a 100 t d⁻¹ CO pilot plant, comparing capital costs and operational benefits (energy storage behavior and H₂ co-production).

• The H₂-integrated CO₂RR cell using a Ni(OH)₂/NiOOH mediator operates at <0.9 V at 50 mA cm⁻² and achieves high selectivity and stability for both gaseous and soluble products.
• CO₂-to-CO: In the H-cell, Zn-Cu-500 catalyst delivered CO Faradaic efficiency (FE) up to 85.8% at −1.0 V (vs RHE) with j_CO ≈ 11 mA cm⁻². In the integrated cell, CO FE reached 81.9% at 150 mA cm⁻². The cathodic performance was unaffected by replacing OER with NIOR, but overall cell voltage dropped by 0.15–0.20 V across 20–250 mA cm⁻² due to lower anodic overpotentials.
• CO₂-to-formate: Bi₂O₃ catalyst achieved formate FE ≈ 89.0% (H-cell) and ≈95.3% at 150 mA cm⁻² in the integrated cell. Cell voltages were 0.17–0.23 V lower than with OER at the same currents.
• Anode kinetics advantage: At 20 mA cm⁻², OER overpotential was ~330 mV versus ~40 mV for NIOR. NIOR showed lower onset potential and Tafel slope than OER (including Co₃O₄ benchmark). EIS indicated significantly smaller charge-transfer resistance for NIOR than OER (over an order of magnitude lower). OER initiated only after full conversion of Ni(OH)₂ to NiOOH (Depth-of-Charge 100%), evidenced by a sharp cell voltage rise and O₂ detection.
• OER suppression and carbon loss mitigation: Online GC and in situ DEMS detected no O₂ during Step 1 within specified current density ranges and mediator charge states, indicating effective OER suppression. The configuration allows easy retrieval of any crossed-over CO₂ without mixing with O₂.
• HOR catalyst protection: Direct CO₂RR-HOR coupling without the mediator initially reduced voltage but rapidly degraded due to Pt/C poisoning by CO and other CO₂RR products (formate, methanol, ethanol), confirmed by FTIR COad bands after ~300 s. The mediator decoupling eliminated HOR poisoning; no CO contamination on Pt/C after longevity testing.
• Step 2 performance: Ni–H₂ battery-like operation achieved maximum power density of 221 mW cm⁻²; steady galvanostatic voltage about −1.28 V at 20–50 mA cm⁻² with >95% voltage efficiency. Mediator Coulombic efficiency exceeded 99%.
• Stability: Multi-swap cycling at 50 mA cm⁻² for CO production showed no degradation over 10 cycles (CO FE 72 ± 1%). 100 h tests for both CO and formate showed trivial voltage decay (0.11–0.12 V) primarily in Step 1; GDE microstructures remained intact.
• System-level energy comparison at 50 mA cm⁻²: Equivalent operating voltages for CO₂-to-CO were 1.81 V (H₂-integrated + SOEC) and 2.30 V (H₂-integrated + AWE) versus 2.32 V for conventional CO₂RR. Summed overpotential losses for H₂-integrated + SOEC (ηsum ≈ 0.26 V) and + AWE (≈0.44 V) were lower than OER alone in conventional CO₂RR (≈0.47 V). Total polarization loss reduction up to ~22% and total energy consumption reduction up to ~42% when including upstream H₂ generation and CO₂ recovery energy in neutral conditions.
• Energy consumption per tonne CO: Conventional CO₂RR ~22.4 GJ t⁻¹; H₂-integrated + AWE ~22.2 GJ t⁻¹; H₂-integrated + SOEC ~17.5 GJ t⁻¹. Accounting for CO₂ recovery energy, both integrated systems show 27–42% lower energy consumption. The approach offers flexibility to generate CO or formate with high selectivity at practical current densities and prolonged stability.

Transferring the anodic function from OER in CO₂RR to NIOR within a mediator and to OER in a separate water electrolyzer leverages more favorable reaction kinetics and conditions, thereby reducing anodic overpotentials and overall cell voltage. The Ni(OH)₂/NiOOH mediator decouples cathodic CO₂RR from anodic HOR, effectively suppressing OER and associated carbon loss while preventing HOR catalyst poisoning by migrated CO₂RR products. The approach maintains high CO₂RR selectivity and stability for both gaseous (CO) and soluble (formate) targets and achieves sub-0.9 V operating voltages at 50 mA cm⁻². System-level analyses, including upstream H₂ generation via AWE or SOEC, show lower operating voltages and energy consumption than conventional CO₂RR, with up to 42% energy savings when CO₂ recovery penalties are included. These findings confirm the hypothesis that coupling CO₂RR with HOR and transferring OER to a dedicated electrolyzer improves energy efficiency and operational robustness, offering a path to integrate CO₂ utilization with the hydrogen economy and grid-scale energy storage.

The work introduces a single-cell CO₂RR architecture that integrates HOR via a Ni(OH)₂/NiOOH redox mediator, achieving high selectivity (up to ~95% for formate; ~82% for CO at 150 mA cm⁻²), durable operation (>100 h), and markedly reduced operating voltages (<0.9 V at 50 mA cm⁻²). By suppressing OER at the anode and avoiding HOR catalyst poisoning, the system mitigates carbon loss and enhances efficiency. When accounting for hydrogen generation via AWE or SOEC, the H₂-integrated CO₂RR reduces total polarization and energy consumption relative to conventional CO₂RR, with up to 42% energy savings when CO₂ recovery is considered. Future research should target direct CO₂RR–HOR coupling without mediators by developing HOR catalysts resistant to CO₂RR product poisoning and by mitigating CO₂ crossover in neutral electrolytes, as well as extending the approach to other CO₂RR products and optimizing system integration for cost and performance.

• The demonstrated system relies on a Ni(OH)₂/NiOOH mediator and alternating operation (Step 1/Step 2), adding complexity and periodicity; OER begins once the mediator is fully oxidized, necessitating timely swapping.
• Direct CO₂RR–HOR coupling without a mediator currently suffers from rapid HOR catalyst poisoning (e.g., Pt/C deactivation by CO and liquid products), indicating a need for poisoning-resistant HOR catalysts.
• CO₂ crossover in neutral/near-neutral conditions remains a challenge; although OER is suppressed and crossed CO₂ can be recovered more easily, comprehensive carbon balance and separation steps for soluble products were not fully addressed.
• Voltage efficiencies decrease at higher current densities; mass transport in complex nanostructured electrodes (e.g., Zn-Cu-500) may limit performance at high rates.
• Energy and techno-economic analyses are preliminary and scenario-dependent; capital costs increase due to the mediator and advanced water electrolyzer components.
• The study focuses on CO and formate as model products; generalization to multi-carbon products requires further validation.