Skip to content

Ion-independent CO2 reduction with boron-doped diamond electrodes for sustained formic acid production and in situ concentration

FieldValue
JournalBulletin of the Chemical Society of Japan
Published2026-04-01
AuthorsSari Araki, Yasuaki Einaga
DOI10.1093/bulcsj/uoag048

A boron-doped diamond electrode sustained formic acid production in a three-compartment CO2 electrolysis cell for 96 h. After an approximately 8 h activation period, the Faradaic efficiency held at 60% to 70% without external supporting-ion replenishment. The static concentrate compartment reached about twice the formate ion concentration of the catholyte. After the main transfer of potassium and sulfate had largely converged, protons, bicarbonate and formate dominated charge compensation.

Sustained production with reduced ion dependence

Section titled “Sustained production with reduced ion dependence”

Conventional two-compartment CO2 electrolysis relies on supporting-ion migration to keep the electrolyte stable. In the authors’ previous system, potassium transfer from the anolyte maintained catholyte pH and supported continued formic acid production. When that ion supply diminished, production efficiency fell.

The three-compartment cell places a concentrate compartment between the anode and cathode compartments. The authors interpret this arrangement as decoupling CO2 reduction from continuous migration of ions supplied by the electrolyte. The result does not mean that ion transport stopped. Potassium and sulfate moved during the early stages, but reaction-derived protons, bicarbonate and formate took over charge compensation as those electrolyte concentrations fell.

The anode and concentrate compartments were separated by a Nafion 424 cation-exchange membrane. The cathode and concentrate compartments were separated by a Neosepta ASE anion-exchange membrane. A 2-inch polycrystalline BDD electrode served as the working electrode, and a 2-inch platinum disk served as the counter electrode. An Ag/AgCl electrode with 3 mol L⁻¹ KCl served as the reference.

The BDD was deposited by plasma-enhanced chemical vapor deposition on 2-inch Si substrates. Its boron concentration was 0.1% relative to carbon. Before use, the electrode was immersed in aqua regia for 30 min and irradiated with UV light for 30 min to oxidize its surface uniformly.

Each compartment used 0.25 mol L⁻¹ K2SO4. The anolyte and catholyte each held 100 mL, while the concentrate held 10 mL. The anolyte and catholyte circulated at 150 mL min⁻¹, but the concentrate remained static. Its open top allowed for ion and water permeation from both electrode compartments.

CO2 was bubbled into the catholyte at 250 mL min⁻¹ for 30 min before electrolysis and at 50 mL min⁻¹ during the run. Electrolysis used constant-current control at 40 mA, corresponding to about 4 mA cm⁻². A 5 V/5 A galvanostatic supply controlled the electrolysis.

K2SO4 was chosen after membrane permeation tests. Under KCl conditions, preferential chloride transport strongly suppressed formate permeation. Under K2SO4 conditions, formate permeated despite competition from sulfate. This transport behavior was the reason for selecting sulfate as the supporting electrolyte.

Product analysis and comparison with the prior cell

Section titled “Product analysis and comparison with the prior cell”

Formic acid was quantified by HPLC with a UV detector. Sulfate concentration was measured by HPLC with a conductivity detector, while potassium concentration and pH were measured with a potassium meter and a pH meter, respectively. The Faradaic efficiency calculation used two electrons per formic acid molecule and a Faraday constant of 96485 C mol⁻¹.

The combined formic acid Faradaic efficiency from the catholyte and concentrate showed no significant variation after the activation period. The authors confirmed reproducibility in Supplementary Table S1. The concentrate formate concentration exceeded the catholyte concentration after 24 h and stayed at approximately twice the catholyte concentration for the remainder of the run.

MetricThree-compartment cellPrevious two-compartment cell
End-of-run formate ion concentrationThe catholyte held 2.07 wt% and the concentrate held 4.01 wt% at 96 h.The catholyte held 1.90 wt% after the same electrolysis time.
Formic acid Faradaic efficiencyIt remained at 60% to 70% from approximately 8 h through the end of the 96 h run.It remained above 80% before production efficiency declined.

The comparison shows that the present cell retained a lower but stable efficiency while producing a more concentrated product. The authors describe its concentration efficiency as approximately twice that of the two-compartment system. It also removed the previous system’s need for continuous potassium-ion replenishment.

Charge compensation in three operating phases

Section titled “Charge compensation in three operating phases”

From 0 to 24 h, electrolyte-derived potassium and sulfate dominated charge compensation. Potassium moved from the anolyte, while sulfate moved from the catholyte into the concentrate. Oxidation in the anode compartment generated oxygen and protons. Cathodic reduction produced formic acid or potassium formate, and cathodic hydroxide reacted with continuously supplied CO2 to form bicarbonate. Formate and bicarbonate could then cross the anion-exchange membrane. The concentrate initially contained potassium sulfate and potassium formate, which kept its pH near neutral.

From 24 to 56 h, protons competed with potassium as the main cation crossing the cation-exchange membrane. Protons became the dominant cation, while formate and bicarbonate competed with continuing sulfate transport. The concentrate became acidic as sulfuric acid and formic acid formed alongside the neutral salts.

From 56 to 96 h, migration of the main electrolyte-derived ions had largely diminished. Proton transport from the anolyte and formate and bicarbonate transport from the catholyte became predominant. The anolyte and catholyte pH values stabilized, while the concentrate remained acidic because it continued to accumulate formic acid, a weak acid.

Sulfate migration continued for about 60 h, and the catholyte sulfate concentration fell to around 0.5 wt%. Potassium migrated rapidly during the first 24 h and then more gradually until 48 h, when the anolyte potassium concentration reached about 0.2 wt%. The authors attribute the different migration rates to competition between sulfate and formate and to sulfate’s larger molecular size.

Implications for electrode and cell design

Section titled “Implications for electrode and cell design”

The 96 h result shows that the cell can sustain operation with a boron-doped diamond working electrode under the tested conditions. BDD offers a wide potential window, low background current, and physical and chemical stability. Boron doping also controls the hydrogen evolution overpotential, which can suppress competing reactions and affect formic acid selectivity.

Earlier BDD studies cited by the authors reported approximately 100% formic acid Faradaic efficiency under mild conditions. The present lower efficiency occurred with K2SO4, an anion-exchange membrane and a three-compartment flow cell. The paper used 0.1% boron relative to carbon, but it did not compare doping levels. The experiment therefore does not establish 0.1% as an optimum boron concentration.

Electrode performance must be assessed with the electrolyte and membrane because the anion-exchange membrane controls formate crossover and the electrolyte anion affects the interfacial reaction environment. The 10 mL static concentrate accumulated about twice the formate concentration of the 100 mL catholyte. This avoided a separate distillation, extraction, or membrane adsorption step, although water transport can still increase the concentrate volume.

Unresolved side reactions and transport losses

Section titled “Unresolved side reactions and transport losses”

The study did not quantitatively analyze gaseous products, so the Faradaic efficiency below 100% cannot be assigned to individual side reactions. The authors identify hydrogen evolution as the most likely major competing reaction. Under K2SO4, the authors considered the interfacial promotion of formic acid production relatively weaker than in halide electrolytes. They also note that hydrogen overpotential differs between these electrolyte conditions. The paper does not establish these effects as the sole cause. Electrode structure, surface condition and flow conditions can also affect the competition.

The authors call for quantitative gas analysis to evaluate the side-reaction balance. AEM selectivity should be tuned between formate and competing anions. Water transfer through the membranes should be reduced to limit concentrate volume growth. Electrolyte anion selection and electrolysis conditions should then be optimized together to suppress competing reactions while retaining product concentration.

The paper’s material requirement is a boron-doped diamond cathode for CO2 reduction. The demonstrated working electrode was 2-inch polycrystalline BDD with 0.1% boron relative to carbon. 6CCVD’s diamond materials catalog lists boron-doped diamond, providing a material option for evaluation against those electrode requirements. The engineering support page can be used to discuss custom fabrication requirements for electrode geometry and cell integration.

View Original Abstract

Abstract While the electrochemical reduction of CO2 to formic acid has been intensively investigated, conventional 2-compartment electrolysis systems inherently rely on continuous ion migration to stabilize the electrolyte. Once supporting ions are depleted, production efficiency deteriorates, resulting in a fundamental limitation for long-term operation. Herein, we report a 3-compartment CO2 electrolysis system that overcomes this intrinsic constraint and enables sustained formic acid production with reduced dependence on electrolyte-derived ion migration. By introducing a concentrate compartment between the anode and cathode compartments and employing boron-doped diamond (BDD) electrodes, stable electrolysis was achieved without external ion replenishment or continuous ionic flux. A Faradaic efficiency of 60% to 70% was maintained for more than 96 h under constant-current conditions, demonstrating the robustness of the system during prolonged operation. In addition to stable production, the proposed configuration simultaneously enabled in situ concentration of the product, yielding formic acid at approximately twice the concentration of that in the catholyte. Detailed analyses of ion transport behavior and pH evolution revealed a transition from charge compensation dominated by electrolyte-derived ion migration to that dominated by proton and hydrogen carbonate ion transport, without compromising production efficiency. These results demonstrate that sustained formic acid generation is feasible even after the completion of major ion transfer. The present 3-compartment design provides a new strategy for decoupling CO2 reduction from electrolyte-derived ion migration, offering a scalable and energy-efficient approach for long-term carbon resource utilization.