Operando techniques reveal why nickel–gallium alloy drives selective carbon dioxide hydrogenation

CO₂ can be converted into methanol by reacting it with green hydrogen, thereby turning a greenhouse gas into a product that can serve as a fuel or chemical feedstock. This hydrogenation reaction only proceeds efficiently with a highly active and selective catalyst, but many catalysts produce unwanted by-products, such as carbon monoxide or methane, and can also degrade over time. The challenge is to design a catalyst that is not only active, but also selective and stable.

Catalysts are complex materials that feature specific atomic-scale sites on their surfaces that are active for a given reaction. Typically dispersed on supports such as silica or alumina, these sites are difficult to control or even identify because they are present in small amounts and are often dynamic, sometimes existing only under reaction conditions. Atomic-level insight into how they function is crucial for designing more effective catalysts.

Nickel (Ni), an abundant and active metal for CO₂ hydrogenation, typically favours methane production. Introducing gallium (Ga) to nickel catalysts alters the catalyst’s behaviour, steering the reaction toward methanol. The nanoscale arrangement of Ni and Ga atoms ultimately governs the catalyst’s performance.

With their capability for operando measurements, beamlines ID15A and BM31 enabled researchers to probe the active sites in Ni–Ga catalysts while converting CO₂ with H₂. At ID15A, high-energy X-ray total scattering revealed the atomic structure of Ni–Ga alloy nanoparticles forming in situ. Meanwhile, X-ray absorption spectroscopy (XAS) at BM31 provided element-specific insights into the local environments of Ni and Ga.

Click figure to enlarge

Fig. 1: Structure: Operando total scattering (X-ray diffraction and pair distribution function (PDF) analyses) reveal a nanocrystalline alloy structure. Ga K-edge X-ray absorption spectroscopy (XAS) identifies the formation of gallium oxide species (GaOx). Catalyst performance: These structures correlate with variations in methanol selectivity and productivity, with the highest methanol productivity and selectivity observed for α’-Ni₃Ga nanoparticles supported on SiO₂, with co-existing GaOx. Density functional theory (DFT) calculations: DFT indicates that, relative to pure Ni, Ni-Ga alloy surfaces stabilise key reaction intermediates, helping to elucidate the observed structure–activity relationships.

The catalyst precursors consisted of highly disordered phyllosilicate sheets of Ga and Ni. Upon in situ activation, these sheets transformed into Ni–Ga alloys – either ordered or disordered, depending on the Ni-to-Ga ratio – while some Ga remained as an oxide outside the alloy (Figure 1, top).

By measuring the catalytic performance and structure simultaneously, a direct link between the structure and activity of the catalysts could be established (Figure 1, middle). An ordered alloy, α’-Ni₃Ga, containing well-defined Ni and Ga sites together with a small fraction of gallium oxide, was particularly efficient and selective. Complementary theoretical modelling showed that Ni-rich surface sites of the α’-Ni₃Ga alloy stabilise key intermediates in CO₂ hydrogenation to methanol, while gallium oxide modulates CO* dissociation to enhance methanol selectivity (Figure 1, bottom).

This research reveals how atomic-scale arrangements govern catalyst performance and highlights the unique capabilities of ESRF beamlines to capture the behaviour of elusive active sites under real operating conditions. The findings provide a blueprint for designing more efficient, selective, and durable catalysts for sustainable chemistry.

Principal publications
How Does the Ni–Ga Alloy Structure Tune Methanol Productivity and Selectivity? N.K. Zimmerli et al., ACS Catal. 15(16), 14252 (2025); https://doi.org/10.1021/acscatal.5c02008

Structural Dynamics Behind the Formation of α′-Ni3Ga Alloy Nanoparticles from a Ni–Ga Phyllosilicate Dispersed on Silica Using X-ray Probes, N.K. Zimmerli et al., Chem. Mater. 37(14), 5312 (2025); https://doi.org/10.1021/acs.chemmater.5c01040