Copper compound converts carbon dioxide to methane by reshaping itself

Researchers tested three different copper-based molecular catalysts to see how well they could convert carbon dioxide into methane using electricity, a process called electrochemical CO2 reduction. The three materials were copper(II) phthalocyanine (CuPc), a copper-based metal-organic framework called HKUST-1, and a compound called [Cu(cyclam)]Cl2. All three were mixed with carbon nanotubes and tested as electrodes in a carbon-dioxide-saturated water-based electrolyte.

CuPc performed best by a clear margin. At a working voltage of -1.06 V, it converted 66% of the electrical charge passing through it into methane, with a partial current density of 13 milliamps per square centimeter dedicated to methane production. That corresponds to a methane formation rate of 0.36 millimoles per second for every gram of CuPc catalyst used. The other two catalysts needed 100 to 200 millivolts more negative voltage just to start working, and topped out lower: HKUST-1 reached a maximum efficiency of 27% at -1.16 V, while [Cu(cyclam)]Cl2 reached only 15% at -1.26 V.

The product mix from CuPc also changed with voltage. At lower overpotentials, the catalyst mainly produced formic acid, ethylene, and carbon monoxide. Methane became the dominant product only once the voltage reached -0.96 V, and peaked at -1.06 V. The team, including researchers from Yale University, Oregon State University, and Northwestern University, published the work in Nature Communications.

To understand why CuPc worked better than the other two catalysts, the researchers used a technique called operando X-ray absorption spectroscopy, which lets scientists watch a material's atomic structure and the electron-counting state of its atoms (called oxidation state) change in real time, while the reaction is actually happening rather than only before and after. One part of this technique, XANES, reveals how many electrons each copper atom is holding onto. A second part, EXAFS, reveals what atoms are sitting next to each copper atom and how far away they are.

As the voltage on the CuPc electrode was made more negative, the copper atoms, which started out in a state called Cu(II), first shifted to an intermediate state called Cu(I), then to a fully metallic state called Cu(0). By the time the voltage reached -1.06 V, where methane production peaked, the spectrum was dominated by this metallic Cu(0) signal, and the EXAFS data showed direct copper-to-copper bonds forming, the signature of small clusters of metallic copper about 2 nanometers across. Not every copper atom made this change, though: a signal from unchanged Cu(II) persisted at every voltage tested, meaning some of the original CuPc structure never restructured at all.

The most notable finding came when the voltage was switched back toward positive values. The metallic Cu(0) signal disappeared, and the spectrum returned to nearly what it had looked like before the reaction started. In other words, the copper clusters that formed during the reaction dissolved back into the original CuPc molecular structure once the driving voltage was removed.

HKUST-1 and [Cu(cyclam)]Cl2 also converted to metallic copper at -1.06 V, and their EXAFS spectra at that voltage looked similar to a standard sample of pure metallic copper, suggesting they had formed bulk copper metal rather than small clusters. But neither material returned to its starting structure when the voltage was switched back to positive values. The metallic copper from HKUST-1 instead oxidized into a different compound, copper(I) oxide, rather than reforming the original metal-organic framework. The cyclam-derived catalyst formed a visible reddish metallic film on the electrode surface and stayed dominated by metallic copper even after the voltage was reversed.

The researchers concluded that CuPc's advantage lies in this reversibility: it can form small copper clusters during the reaction and then rebuild its original molecular structure afterward, while the other two catalysts undergo one-way, irreversible restructuring into larger, less controlled copper structures. They describe this as a case where identifying the true catalytically active structure required watching the material while it worked, since the structure the catalyst held onto after the reaction ended was not necessarily the structure doing the useful chemistry during the reaction.

Electricity-driven reduction is not the only route researchers are exploring to turn carbon dioxide into usable fuel. A separate review of photocatalytic CO2 reduction, which uses light rather than electricity to drive the chemistry, notes that copper oxides and cuprous oxide are also widely studied catalysts, in this case for producing methanol rather than methane. The review describes how copper's ability to shift between electron-counting states helps keep the light-generated charges in these materials from recombining before they can react, which is part of why copper-based semiconductors tend to favor methanol as a product.

That photocatalytic route faces its own separate challenges, according to the review: low efficiency at converting available light into product, difficulty moving reactants and products in and out of the reaction zone, and instability from photo-corrosion in materials like copper oxide and cuprous oxide. The review notes ongoing work using structures like Z-scheme heterojunctions and core-shell nanostructures to address these issues, reporting methanol yields and efficiencies in different experimental systems, though it frames this as still an active area of engineering rather than a settled technology.

The electrochemical results for CuPc, HKUST-1, and [Cu(cyclam)]Cl2 were measured under one specific set of conditions: a potassium bicarbonate electrolyte and a range of voltages tested between -0.76 and -1.36 V. Even the best-performing catalyst, CuPc, still required substantial negative voltage to reach its peak methane efficiency, and produced a mix of other products such as formic acid, ethylene, and carbon monoxide at less negative voltages rather than methane alone. The paper focuses on explaining why CuPc outperforms the other two materials rather than on how far its efficiency or voltage requirements could still be improved.

The separate photocatalytic methanol review makes clear that, for that light-driven approach at least, the transition from laboratory-scale results to industrial application remains hindered by low quantum efficiency and mass transfer limitations. Taken together, the sources describe copper as a recurring and useful element across different strategies for converting carbon dioxide into fuel, but neither source suggests that any of the systems described are yet ready to move beyond laboratory-scale testing.