Electrochemical Concentration of Carbon Dioxide from an Oxygen/Carbon Dioxide Containing Gas Stream
James Landon and John R. Kitchin, “Electrochemical Concentration of Carbon Dioxide from an Oxygen/Carbon Dioxide Containing Gas Stream”, Journal of The Electrochemical Society 157(8), B1149-B1153 (2010). https://doi.org/10.1149/1.3432440
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Question
Electrochemical gas separation can in principle concentrate a gas at very small energy cost: the Nernst equation gives only 0.015 V (about 5.7 kJ/mol O2) to raise O2 from 0.1 to 1.0 atm. Earlier electrochemical CO2 separators, built for cleaning breathing air or using redox carriers or carbonate-soaked asbestos and polyamide supports, needed cell voltages from above 1.23 V up to 3 V, so they split water and consumed too much energy for power plant use. Part of that penalty came from thick supported electrolytes. The paper asks whether a thin anion exchange membrane can lower the ohmic loss enough to concentrate CO2 from a CO2/O2 stream below the water splitting potential, and whether the resulting performance is compatible with capture from coal plants (Landon 2010).
Approach
The separation scheme reduces O2 to hydroxide at the cathode; the hydroxide reacts homogeneously with CO2 in the feed to form bicarbonate (or carbonate), which crosses the membrane to the anode, where hydroxide oxidation evolves O2 and the bicarbonate is neutralized to release CO2. Pure bicarbonate transport would give up to 4 mol CO2 per mol O2, and pure carbonate up to 2. The membrane electrode assembly used a Neosepta AHA anion exchange membrane, soaked in 1 M KOH, between 5 cm2 Toray carbon paper electrodes coated with XC-72R carbon black. The anode carried Ni particles (1 to 5 μm) made by a polyol process from NiCl2 in ethylene glycol at 160 C, and the cathode carried commercial 20% Pt on Vulcan XC-72 bound with Nafion and Teflon (no anion ionomer was available), both at 1 mg/cm2. A humidified 50 cm3/min feed (pure O2, or 49.3% CO2 and 50.7% O2) went to the cathode, an argon sweep to the anode, and the anode gas was analyzed by mass spectrometry. Cell potentials were held at or below 1.2 V, and membrane resistance was measured by current interrupt. As a check on the anode chemistry, KHCO3 solutions (0 to 3 M) were also electrolyzed in a Nafion-based water electrolyzer.
Key findings
Measurable current began near 0.6 V and rose to a maximum of 6 mA/cm2 at 1.2 V. The membrane resistance was 5.06 Ω cm2, contributing only about 30 mV, so nearly all of the cell voltage was electrode overpotential. With pure O2 feed, O2 was pumped to the anode for 19 h. Switching to the CO2/O2 feed left the O2 flux unchanged while the CO2 flux rose sharply, and both fluxes vanished at 0 V, showing the transport was electrochemically driven and that neither gas diffused through the membrane. The average CO2:O2 product ratio at 1.2 V was 3.56:1, implying a mixed carbonate and bicarbonate electrolyte dominated by bicarbonate. The authors ruled out direct electrochemical formation of carbonate from CO2 and O2 (which would give near 2:1) and estimated that electro-osmotic drag could contribute at most about 4% of the observed flux. In the bicarbonate electrolysis tests the cathode H2 rate was unaffected while anode gas evolution grew with bicarbonate concentration, reaching a CO2:O2 ratio of 1.48 at 350 mA in 3 M KHCO3; the lower ratio was attributed to protons crossing the Nafion membrane before neutralizing bicarbonate.
An energy analysis for a 500 MW coal plant at 40% efficiency (about 3250 mol/s CO2) expressed separation power as a fraction of plant output versus cell voltage and CO2:O2 ratio. Staying under the DOE target of less than 35% cost of electricity increase, taken as less than 35% power loss, requires practical rates at about 0.5 V with a ratio of at least 3.5. The ratio was achieved, but the required voltage was about twice too high, and the low current density would make a coal plant separator impractically large.
Significance
This is the first reported use of a thin anion exchange membrane for electrochemical CO2 concentration, and it achieved separation below 1.23 V with a CO2:O2 ratio near the bicarbonate limit (Landon 2010). It identifies electrode kinetics for oxygen reduction and evolution, not membrane resistance, as the main loss, which motivates the search for more active electrocatalysts and thinner, stable membranes. The authors note the CO2/O2 product suits an auxiliary oxy-combustion unit and that the approach may already be more efficient than earlier devices for removing CO2 from breathing air in confined spaces.