Atomically dispersed nickel as coke-resistant active sites for methane dry reforming

Let's start with the promise. Dry reforming of methane takes two problematic gases—carbon dioxide and methane—and turns them into synthesis gas, a mix of carbon monoxide and hydrogen that feeds chemicals, fuels, and materials. It's a neat bit of alchemy with a climate twist. The catch has always been the same: nickel, the workhorse metal you'd like to use because it's cheap and active, tends to smother itself in carbon. Coke builds up, particles sinter, and the catalyst dies young. Akri, Zhao, Li, and their colleagues approached that problem from a different angle. Instead of making better nickel nanoparticles, they stabilized nickel as single atoms on a cerium-doped hydroxyapatite support. One nickel atom at a time, anchored by a calcium phosphate scaffold tuned with a little cerium. The claim is bold and simple: an isolated nickel atom will split the first carbon-hydrogen bond in methane, but it won't keep ripping hydrogen away to leave carbon behind. It hands off the methyl fragment to an oxygen-assisted pathway that ends in carbon monoxide. That choice—first bond, then detour—starves coke at the source. You can see the logic in the numbers. On a single nickel bound to a ceria surface, the first carbon-hydrogen cleavage comes with a barrier of about 0.63 electron volts, comfortably lower than the roughly 0.9 you need on a flat nickel one hundred eleven surface. Go one step deeper, and the chemistry pushes back: turning methyl into methylene plus hydrogen is uphill in energy and faces a wall of 1.54 electron volts. There's an escape hatch, though. Oxidize methyl to methoxy and the door opens with a 0.69 electron volt push; from there, the route to carbon monoxide is downhill, with a modest final carbon-hydrogen activation later on. In other words, the single atom is eager to start the job and reluctant to finish it the usual way. It takes the side road that avoids carbon. If you're wondering how they held nickel to one atom apiece at temperatures where methane reforming occurs, this is where the support does the heavy lifting. The team doped hydroxyapatite—a calcium phosphate familiar to anyone who studies biominerals—with five weight percent cerium, then used strong electrostatic adsorption to deposit nickel at loadings from the sub-percent range up to ten percent. Cerium turns out to accomplish two things at once: it reduces the base strength of the surface, which can help tamp down coke chemistry, and it builds a local environment that grips nickel hard enough to resist reduction to metal clusters when you heat it. The microscopy and spectroscopy read like a checklist for single atoms. High-angle annular dark-field images show mostly isolated bright spots for the low-loading samples, with only a sliver—about seven percent—aggregated into sub-nanometre clusters after a five hundred degrees Celsius reduction. Extended X-ray absorption fine structure fills in the distances: nickel to oxygen bonds around two angstroms, a distinct nickel to cerium scattering path near two and a half, and almost no nickel to nickel contribution, with a nickel to nickel coordination number well under one. In plain language, each nickel "sees" oxygen and cerium, but not other nickel. X-ray photoemission puts the nickel in hydroxide-like environments before reaction and, crucially, shows that these sites refuse to reduce easily in hydrogen. That stubbornness is a feature, not a bug. When reduction is easy, sintering is easy. They also checked the broader support chemistry. Cerium oxide is less basic than bare hydroxyapatite—the carbon dioxide temperature-programmed desorption traces are clear on that—and the cerium-doped support alone is nearly inert for dry reforming. That's good: it means the activity you see isn't a trick of the scaffold. The X-ray diffraction patterns don't show big ceria crystals either, which suggests that cerium is highly dispersed or folded into the apatite lattice. All signs point to strong metal-support interactions holding nickel as single atoms in a ceria-like neighborhood. Then comes the test that matters. They flowed a methane and carbon dioxide mixture over these materials at seven hundred fifty degrees Celsius. Under those harsh conditions, the nickel single-atom catalysts on the cerium-doped support behaved like they were built for it. The half-percent sample, with predominantly isolated nickel, held around ninety percent carbon dioxide conversion for more than two and a half days on stream, and the hydrogen to carbon monoxide ratio sat near one. Increase the nickel to one or two percent, and the picture stays bright: conversions around ninety-five percent for sixty-five to about one hundred hours, still with negligible coke. That's a long time for a nickel catalyst to stay clean. Put conventional nanoparticles into the same flame, and the difference is immediate. A ten percent nickel sample on the cerium-doped support deactivated within fifteen hours as coke piled on. The same high loading on undoped hydroxyapatite failed even faster. Even at low loading, the undoped support couldn't keep nickel isolated; post-reaction diffraction picked up the fingerprint of face-centered cubic nickel, evidence of sintering. Carbon analysis told the same story in a different language: for the single-atom catalysts, the carbon balance closed and the post-mortem spectra were quiet; for the nanoparticles, you saw the messy signature of disordered and graphitic carbon. Coke, in other words, is the default unless you change the site. If you like to translate stability into rates, the numbers hold up there too. At seven hundred fifty degrees, the single-atom sample delivered a methane consumption rate of roughly three hundred seventy-three moles per gram of nickel per hour and a carbon dioxide rate near seven hundred twenty-nine under the same conditions. The turnover frequency—how many times per second a site does its job—sat around six per second for methane on that catalyst. The two percent sample roughly halved those specific rates, and the ten percent nanoparticle catalyst dropped to the tens and low hundreds. It's a simple pattern: keep nickel isolated, and the activity per nickel soars; let it cluster, and the specific numbers slide while coke creeps in. Let's stitch the mechanism back into this performance. On the calculated single-atom site, methane doesn't latch on strongly—it's more of a gentle physisorption than a deep embrace—and the carbon sits about two point fifty-six angstroms above the nickel. That geometry and the nearby oxygen on the ceria-like surface give you that easy first carbon-hydrogen break. The key step on the detour route, the carbon-hydrogen activation within methoxy that tips it toward carbon monoxide, carries a barrier just under one electron volt. Together, those features route the chemistry toward carbon monoxide and hydrogen without asking nickel to host the bulky carbon-hydrogen fragments that seed carbon growth. There's an important contrast to keep in mind here. People have known for years that precious metals can shrug off coke in dry reforming, and elegant studies, like the ruthenium one nickel one ensembles reported by Tang and co-workers, show how carefully designed sites can bend reaction pathways. What's new in Akri's work is doing this with earth-abundant nickel and a support that you can make by co-precipitation, not a bespoke alloy. It's a different design philosophy: don't fight coke on a nanoparticle; prevent the steps that make it by changing the site to a single atom bound to the right neighborhood. One might ask whether the support itself is doing unseen magic. The team looked for that and didn't find it. Cerium-doped hydroxyapatite by itself barely touches methane and carbon dioxide under these conditions. Its main role is architectural and electronic: create fewer basic sites that might otherwise help carbon nucleate and pin nickel in an oxidic, ceria-coordinated state that’s stable enough to keep atoms apart but flexible enough to shuttle oxygen for that methoxy route. Hydrogen temperature-programmed reduction maps the line: the single-atom materials resist reduction, while the ten percent systems slip to metal below six hundred degrees and then sinter. For all the success, there are limits. Even atomically dispersed nickel can drift under the kind of heat and reactive environment that dry reforming demands. The authors acknowledge that sintering remains a risk under industrially relevant conditions. While the catalysts ran clean for up to about one hundred hours in the lab, scaling those runs and cycling them will test how robust the nickel, cerium, and phosphate bond really is. That said, holding nickel steady for days while keeping conversion high and coke at bay is a bar most nickel catalysts don't clear. Step back, and you can see the design rule emerging. If the site only lowers the barrier for the first carbon-hydrogen activation and offers oxygen to catch the methyl before it dehydrogenates further, coke loses its foothold. Cerium-doped hydroxyapatite gives nickel that site. The strongest signals back it up: atom-by-atom images, scattering paths that see cerium and not nickel neighbors, and a reaction profile whose bottlenecks match what density functional theory predicts for a single-atom route to carbon monoxide. So here's the takeaway to carry with you. Akri and colleagues didn't just report a better catalyst; they showed why a different kind of nickel—isolated, oxygen-coupled, and support-stabilized—changes the dry reforming game. Under a twenty to twenty methane and carbon dioxide feed and a brisk gas hourly space velocity, they held high conversion, near-unity hydrogen to carbon monoxide, and negligible coke for tens to about one hundred hours, while nanoparticle nickel failed early. You can argue about which support is best or how to push the stability further. But the center of gravity has shifted. For nickel in dry reforming, single atoms on the right oxide neighborhood aren't a curiosity; they're a path around coke that the data, and the chemistry, both support.