Ball milling tunes nickel–ceria interfaces for cleaner methane reforming

Chemists have found that a simple, solvent-free ball-milling recipe can reshape the interface between nickel and ceria in ways that make the resulting catalysts both more active and more durable in the dry reforming of methane, a reaction that turns two greenhouse gases into valuable synthesis gas. The study, published in Discover Catalysis, systematically varied milling time and nickel precursor chemistry and then tracked how those choices propagated through catalyst structure, redox behavior and performance under realistic reforming conditions.

Dry reforming of methane, or DRM, converts methane and carbon dioxide into a mixture of hydrogen and carbon monoxide that can be upgraded to synthetic fuels or fed directly into fuel cells. Because the feed mixture mirrors biogas, the reaction is an attractive route for valorizing a renewable gas that would otherwise be burned or vented. The catch is harsh chemistry: temperatures near 800 degrees Celsius are typically required, and side reactions such as the Boudouard reaction and methane cracking deposit coke on the catalyst, gradually poisoning it. Nickel is the pragmatic industrial choice because it is cheap and abundant, but only if carbon buildup can be kept in check.

One established defense is to anchor nickel on ceria, a reducible oxide whose oxygen transfer capacity, especially at the metal–support boundary, helps burn off carbon before it accumulates. The research team, led by Rudy Calligaro, Silvia Mauri, Marta Boaro, Jordi Llorca, Piero Torelli and Alessandro Trovarelli, set out to test whether mechanochemistry, the deliberate use of mechanical force to drive solid-state reactions, could build that critical interface more effectively than conventional wet impregnation, while avoiding solvents altogether.

The preparation was deliberately simple. Commercial high-surface-area ceria, calcined at 800 degrees Celsius to a surface area of 63 square meters per gram, was mixed with a nickel precursor salt, either nickel nitrate hexahydrate or nickel chloride hexahydrate, to give a 4 weight percent nickel loading. The powder was sealed in a 15-milliliter zirconia jar with a single 15-millimeter zirconia ball and milled in a Fritsch Pulverisette 23 mini-mill at a fixed frequency of 15 hertz for times ranging from 5 to 60 minutes. Control materials included loose-contact mixtures prepared without the milling ball, so the components never experienced mechanical energy, and reference catalysts made by standard incipient wetness impregnation from the same salts. All samples were then calcined at 500 degrees Celsius for one hour.

Characterization revealed that the two precursors produced fundamentally different nickel dispersions. X-ray diffraction showed broad nickel oxide peaks for nitrate-derived samples, indicating small, well-spread crystallites, while chloride-derived samples displayed distinct, narrow peaks characteristic of larger particles. The same precursor-dependent difference appeared even in the unmilled loose-contact mixtures, pointing to the different decomposition mechanisms of the two salts rather than to the milling itself. Textural measurements showed that nickel oxide deposition reduced surface area and porosity, more strongly for chlorides and most consistently for the impregnated reference, while longer milling steadily reduced pore volume and diameter at nearly constant surface area, consistent with progressive disaggregation of support agglomerates.

Temperature-programmed reduction added a redox dimension to the picture. All catalysts showed low-temperature reduction features attributed to bulk nickel oxide and to nickel oxide in strong interaction with the support, plus a high-temperature peak from bulk ceria reduction. The nitrate series reduced at slightly lower temperatures with a more intense low-temperature feature, again consistent with smaller crystallites that reduce more easily and interact more broadly with the support. After a first reduction cycle, all components merged into a single lower-temperature event, and quantitative hydrogen uptake dropped, indicating sintering of both ceria and nickel. Chloride-derived catalysts milled for short times lost the most hydrogen consumption, signaling the strongest sintering, which matched observations of large particles that were not fully re-oxidized.

Catalytic testing ran from 550 to 800 degrees Celsius at atmospheric pressure in a fixed-bed reactor, using a methane-rich, biogas-like feed of 44 percent methane, 29 percent carbon dioxide and 4 percent nitrogen in helium, corresponding to a carbon dioxide to methane ratio of 2:3. Nitrate-derived catalysts were consistently more active than their chloride analogues, in line with their smaller nickel particles. Milling time mattered strongly: in the chloride series, the 10-minute milled sample achieved a methane conversion three times that of the impregnated reference, while in the nitrate series activity decreased with milling time and the unmilled loose-contact sample was the most active. Short-milled nitrate catalysts reached hydrogen to carbon monoxide ratios of 0.8 to 0.9.

Activity, however, came with a carbon bill. Thermogravimetric analysis showed more than 25 weight percent residual carbon on the most active nitrate samples, falling below 10 weight percent at longer milling times, while chloride catalysts accumulated little carbon and the unmilled chloride sample none at all. The decisive result came from 70-hour stability runs at 750 degrees Celsius, a temperature chosen to discriminate deactivation behavior without operating at the very high conversion regime of 800 degrees. The impregnated nitrate catalyst lost 85 percent of its methane conversion and 75 percent of its carbon dioxide conversion over the test. The 30-minute milled sample lost 77 and 56 percent respectively, but the 10-minute milled nitrate catalyst, despite forming the most carbon, was the most stable of all, retaining activity with only a 49 percent drop in methane conversion and a 36 percent drop in carbon dioxide conversion. In the chloride series, the 10-minute milled sample deactivated rapidly, losing roughly 80 percent of performance in 70 hours, which the authors attribute to nickel sintering.

The structural origin of this resilience emerged from electron microscopy and synchrotron spectroscopy. High-resolution transmission electron microscopy revealed sub-nanometric nickel-bearing corrugations on ceria crystallite surfaces, most abundant at intermediate milling times, alongside partial nickel encapsulation within the ceria lattice. After reduction at 800 degrees Celsius, which sintered ceria particles to 40 to 60 nanometers and nickel to about 5 to 6 nanometers in the nitrate sample versus 8 to 15 nanometers in the chloride sample, these surface features survived, showing that the interfacial morphology is thermally robust. Nickel L-edge NEXAFS spectra, measured at the Elettra synchrotron in Trieste, showed a progressive attenuation of the octahedral nickel(II) multiplet feature from the loose-contact to the milled samples, without any detectable metallic nickel signature. The authors interpret this as a modified ligand field and increased nickel–oxygen covalency, consistent with nickel embedded in a strongly interacting interfacial structure resembling Ni–O–Ce arrangements, motifs that the literature associates with highly active sites for methane C–H bond activation.

The authors are careful about the limits of the evidence. X-ray photoelectron spectra of the milled catalysts showed severe, milling-dependent distortions of the cerium 3d lines, attributed to differential charging in resistive, defect-rich ceria, so oxidation states were assigned only qualitatively. The NEXAFS interpretation of Ni–O–Ce-like motifs remains a hypothesis that would require dedicated in situ or operando spectroscopy to confirm, and thermogravimetry measured only total carbon, not its speciation, so the link between carbon amount and deactivation cannot be read one-to-one. A turnover-frequency analysis was also impossible without independent counts of accessible nickel sites. Even so, the comparative trends hold because all catalysts were tested identically at the same loading, and mass-normalized conversion rates at 550 degrees Celsius preserved the same ranking. Within those constraints, the study demonstrates that a mild, minutes-long, solvent-free milling protocol, tuned for precursor chemistry and milling time, can engineer a durable nickel–ceria interface that outperforms impregnation on the balance of activity and stability, positioning mechanochemistry as a scalable route toward next-generation dry reforming catalysts.

Subject of Research: Mechanochemical synthesis of Ni/CeO2 catalysts for methane dry reforming

Article Title: Medium-energy mechanosynthesis parameters enhancing the dry reforming of methane over Ni/CeO2 catalysts

Article References: Calligaro, R., Mauri, S., Boaro, M., Llorca, J., Torelli, P., & Trovarelli, A. (2026). Medium-energy mechanosynthesis parameters enhancing the dry reforming of methane over Ni/CeO2 catalysts. Discover Catalysis, 3(1), Article 11. https://doi.org/10.1007/s44344-026-00039-6

Image Credits: AI Generated

DOI: 10.1007/s44344-026-00039-6

Keywords: dry reforming of methane, Ni/CeO2 catalysts, mechanochemistry, ball milling, ceria support, nickel–ceria interface, syngas, biogas valorization, catalyst deactivation, NEXAFS, HRTEM, coking resistance