A New Rulebook Keeps Dilute Alloy Catalysts Stable Under Heat

A Stability Playbook for Catalysts That Use a Pinch of Precious Metal

TL;DR: University of Michigan engineers found that dilute alloy catalysts stay active at high temperature when a strongly binding surface molecule holds the active atoms in place or when the two metals naturally resist mixing.

Researchers at the University of Michigan Engineering have published a practical stability rulebook for dilute alloy catalysts, a class of materials that pairs about 1% or less of an active dopant metal with an inert host metal, according to Michigan Engineering News. The study appeared in the Journal of the American Chemical Society and was funded by the U.S. Department of Energy and the National Science Foundation. The central result is direct: selecting a strongly binding adsorbate or selecting alloying metals that do not mix keeps the emerging material class stable under high temperatures.

The work addresses fuel, plastic and pharmaceutical production, where dilute alloy catalysts could help industrial chemical reactions use less precious metals, extend catalyst lifetime and reduce unwanted byproducts, according to Michigan Engineering News. Until now, instability under heat held back widespread use because surface dopant atoms were prone to dissolving into the interior of the host metal under high temperatures, deactivating the catalyst. The Michigan team, led by corresponding author Suljo Linic, the Martin Lewis Perl Collegiate Professor of Chemical Engineering, with lead author and doctoral student Bill Yan, focused on the variables that govern catalyst stability rather than activity alone.

What a dilute alloy catalyst is

A dilute alloy catalyst contains approximately 1% or less of an active dopant metal dispersed into an inert host metal, according to Michigan Engineering News. The active dopant atoms make up about 1% or less of the metal content and help initiate the reaction by activating the reactants. The resulting intermediate products then spill over to the host metal, where the reaction continues.

That division of labor matters because conventional single-metal catalysts face a constraint known as linear scaling relationships, according to Michigan Engineering News. For conventional single-metal catalysts, strengthening bonds for one reaction step interferes with bonds in the next step, slowing the reaction. Dilute alloy catalysts can break this constraint by distributing the reaction steps onto different active sites.

The broader single-atom alloy literature describes the same architecture in similar terms, according to An Atlas and Design Rules for Single- and Dual-Atom Alloys. In single-atom alloys, isolated atoms of a reactive element are embedded in a more inert host metal. The dopant provides an active site that selectively lowers activation barriers, thereby favoring specific reaction pathways. Upon spillover from the active site, the more inert host provides a less reactive environment that reduces undesired subsequent reactions, including those leading to coke formation and catalyst deactivation.

Why heat has been the breaking point

While highly active and selective, surface dopant atoms are prone to dissolving into the interior of the host metal under high temperatures, deactivating the catalyst, according to Michigan Engineering News. That inward movement removes the initiating sites from the surface where reactants first attach. Once the surface loses those sites, the two-site sequence of activation on the dopant followed by continuation on the host cannot proceed in the same way.

“Industrial application of dilute alloy catalysts requires both good activity and good stability. While other works have mostly focused on the activity part, we tackle the stability aspect by identifying the variables that govern catalyst stability,” said Suljo Linic, the Martin Lewis Perl Collegiate Professor of Chemical Engineering and corresponding author of the study, according to Michigan Engineering News.

The related atlas study notes a parallel limitation for reactive dopants, according to An Atlas and Design Rules for Single- and Dual-Atom Alloys. Many reactive dopants exhibit a thermodynamic preference for bulk segregation, limiting their accessibility as surface active sites at elevated temperatures. The atlas authors also note that in multi-step reactions a single dopant element may not efficiently catalyze every elementary step, and that only one reaction fragment may interact strongly with a given dopant element in heteronuclear bond dissociation or cross-coupling reactions.

The gold-platinum test that revealed rule one

The Michigan research team began its investigation by synthesizing a dilute alloy catalyst made of approximately 18-nanometer-wide gold nanoparticles dotted with platinum atoms, according to Michigan Engineering News. Gold served as the inert host metal and platinum served as the active dopant metal in that starting system. The width of about 18 nanometers defined the particle scale used for the comparative reaction tests.

They tested the alloy under two different reactions — ethylene hydrogenation, used in plastics manufacturing, and carbon monoxide oxidation, used in car emission control, according to Michigan Engineering News. Each reaction was operated under temperatures ranging from 50 C to 250 C while a spectroscopy technique tracked surface platinum atoms in real time. The temperature window therefore covered 50 C through 250 C for both reactions, with continuous observation of platinum at the surface.

During ethylene hydrogenation, the reaction rate sharply dropped off when temperatures rose past 100 C but continued to increase with temperature during carbon monoxide oxidation, according to Michigan Engineering News. The same gold-platinum material behaved differently in the two chemical environments despite experiencing the same broad heating range. That contrast pointed to the role of the molecules present during reaction, not temperature alone.

Strongly binding molecules hold the surface together

This suggested that adsorbates — molecules that stick to a solid surface — that bind more strongly to active dopant metals would improve stability, according to Michigan Engineering News. Weak-binding ethylene could not hold platinum to its surface, allowing entropy to drive platinum atoms inside the gold particle. Carbon monoxide instead firmly pinned platinum in place, overriding the tendency to sink into the gold.

In other words, the choice of reacting molecule changed whether platinum remained available where activation begins, according to Michigan Engineering News. Ethylene left platinum free to move inward as temperature increased. Carbon monoxide opposed that inward movement and retained platinum at the surface across the tested range.

“This is somewhat like fishing. A fish on the line wants to go deep into the water, and you need a strong grip to pull the fish to the surface. Here, you need a strongly binding adsorbate to hold the dopant metal at the alloy surface,” said Bill Yan, a doctoral student of chemical engineering at U-M and lead author of the study, according to Michigan Engineering News. The analogy describes entropy as the pull toward the interior and strong adsorption as the counteracting grip at the surface.

Rule two: pick metals that do not want to mix

Follow-up atomic simulations tested how reactions would change with other combinations of host metals including gold, silver and copper and dopant metals including iridium, palladium and platinum, according to Michigan Engineering News. The simulations expanded the comparison beyond the initial gold-platinum pair to nine host-dopant combinations across those listed elements. The goal was to test whether the observed stability pattern extended to other pairings.

Calculations showed that alloying metal mixing behavior — called miscibility — controls whether the surface dopant dissolves into the host metal, according to Michigan Engineering News. Metals like platinum or palladium favor mixing, causing a tendency to deactivate the catalyst, while iridium resists it, staying active on the host metal’s surface. Miscibility therefore functions as a selection criterion before any reaction begins.

To validate the simulation, the research team synthesized gold-iridium dilute alloy catalysts, according to Michigan Engineering News. Experiments using a gold-iridium dilute alloy catalyst backed up the calculations, with zero deactivation up to 250 C for both ethylene hydrogenation and carbon monoxide oxidation. Unlike the gold-platinum system, which lost rate past 100 C in ethylene hydrogenation, the gold-iridium system maintained performance through 250 C in both tested reactions.

What zero deactivation up to 250 C means in practice

Zero deactivation up to 250 C for both ethylene hydrogenation and carbon monoxide oxidation means the gold-iridium system did not show the sharp rate loss seen for gold-platinum in ethylene hydrogenation above 100 C, according to Michigan Engineering News. The validation covered the same two application-linked reactions: ethylene hydrogenation linked to plastics manufacturing and carbon monoxide oxidation linked to car emission control. Both were examined across the elevated-temperature regime that previously caused dopant loss.

Together, the results provide researchers with practical rules for selecting the best metal combination and operating conditions to stabilize dilute alloy catalysts, according to Michigan Engineering News. The metal combination addresses miscibility, while operating conditions address the identity and binding strength of adsorbates present during reaction. The two rules can be considered together when planning a dilute alloy system.

“Our proposed strategies for enhancing stability can be readily applied to most current dilute alloy systems,” said Linic, according to Michigan Engineering News. The statement describes applicability of the strategies, not completion of industrial deployment. The catalysts were studied at the Michigan Center for Materials Characterization which is operated and maintained with support from indirect cost allocations in federal grants.

How the findings fit the wider alloy-design effort

The Michigan result arrives alongside broader efforts to map which dopant structures form across large compositional spaces, according to An Atlas and Design Rules for Single- and Dual-Atom Alloys. Using density functional theory screening, that team constructed an atlas spanning transition metal single- and dual-atom alloys in copper and silver surfaces, validated by scanning tunneling microscopy. The Michigan simulations examined gold, silver and copper hosts with iridium, palladium and platinum dopants, while the atlas examined copper and silver surfaces across transition-metal dopants.

The atlas reports that stability follows an electron count, with dopants pairing most strongly when their combined d-electron count approaches ten, according to An Atlas and Design Rules for Single- and Dual-Atom Alloys. Host metal and surface facet can be used to tune the resulting active-site motifs, while size mismatch and spin explain variations around this trend. The Michigan study reports that miscibility controls whether the surface dopant dissolves into the host, with iridium resisting mixing and platinum or palladium favoring it.

The atlas further introduces a reactor-anchor concept, in which one dopant anchors a second, otherwise bulk-segregating dopant at the surface, according to An Atlas and Design Rules for Single- and Dual-Atom Alloys. That concept addresses the same bulk-segregation challenge highlighted in the Michigan work, where entropy drives platinum inside gold unless opposed by strong binding or by selection of an immiscible pairing. Together, the studies describe complementary levers: adsorbate choice, miscibility choice, electron-count pairing, host choice and facet choice.

Why cheaper, longer-lasting, selective catalysts matter

Dilute alloy catalysts could help industrial chemical reactions use less precious metals, extend catalyst lifetime and reduce unwanted byproducts, according to Michigan Engineering News. Using about 1% or less of active dopant metal reduces the amount of precious component required relative to a conventional single-metal design. Extending catalyst lifetime addresses the deactivation pathway in which surface atoms dissolve inward under heat. Reducing unwanted byproducts connects to selectivity, where intermediates move from the dopant to the host and continue under less reactive conditions.

The single-atom alloy literature describes related selectivity effects in detail, according to An Atlas and Design Rules for Single- and Dual-Atom Alloys. Reaction intermediates generally bind more weakly to dopant sites in single-atom alloys than on the corresponding pure dopant metal surface. Weaker binding on the host also facilitates product desorption. The dopant can additionally bring reactants into proximity and stabilize their co-adsorbed configuration, promoting bond formation in coupling reactions.

Catalysis underpins 85% of industrial chemical processes and contributes to roughly 35% of global gross domestic product, according to An Atlas and Design Rules for Single- and Dual-Atom Alloys, citing prior literature. Even modest improvements in catalyst performance therefore reduce material and energy demand, with economic and environmental benefits. The Michigan team positions its contribution as movement toward use in fuel, plastic or pharmaceutical production by resolving the heat-stability barrier.

What was actually demonstrated by Sept. 16, 2026

By Sept. 16, 2026, the demonstrated items were the gold-platinum comparison across 50 C to 250 C, the adsorbate-binding explanation, the multi-metal simulations and the gold-iridium validation with zero deactivation up to 250 C in both reactions, according to Michigan Engineering News. The study outlining the rulebook was published in the Journal of the American Chemical Society. Research support came from the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, and the National Science Foundation, with additional characterization support through the Michigan Center for Materials Characterization.

What was not demonstrated was full industrial adoption, according to the same account in Michigan Engineering News. The reported advance is a design strategy for making cheaper, longer-lasting and more selective industrial catalysts, framed as movement toward use in fuel, plastic or pharmaceutical production. The practical guidance covers selection of metal combination and operating conditions, applicable to most current dilute alloy systems as proposed by the authors, rather than a completed retrofit of existing plants.

For readers tracking next steps, the two operational questions are which adsorbates will be present and how strongly they bind the chosen dopant, and whether the chosen host-dopant pair favors or resists mixing, according to Michigan Engineering News. The two research questions correspond to the two rules. Strong binding retains dopants at the surface during heating. Low miscibility retains dopants at the surface by compositional choice, as shown by gold-iridium through 250 C.