Multimetallic synergies in tandem plasmonic photocatalysis

Plasmonic photocatalysis offers a route to drive demanding chemical transformations using light rather than high temperatures. Given the high energy demand associated with producing chemical feedstocks, it is highly desirable to develop fully electrified approaches that use renewable photons as the driving force and incorporate CO₂ as a reactant in value-added chemistry. Reporting now in PNAS, Sharma et al. demonstrate that visible light alone can power a tandem reaction sequence in which CO₂ is reduced to CO while propane is simultaneously dehydrogenated to propene, using Ga–Ni–Mn active sites supported on plasmonic black gold colloidosomes (1). The system coproduces CO and propene in nearly a one-to-one ratio at substantial rates (~1,600 micromole g⁻¹ h⁻¹) and maintains stable performance for over 500 h on stream under continuous visible-light illumination (1). The combination of high space-time yields, sustained tandem selectivity, and evidence for light-specific pathways enabled by hot carriers makes this study an important step toward industrially relevant plasmonic photochemistry that couples CO₂ utilization with alkane upgrading.

The work by Sharma et al. builds on efforts from the past 15 y to control chemistry with light using plasmonic metal nanostructures (2, 3). Plasmonic nanostructures strongly couple to incident electromagnetic radiation, inducing a resonant oscillation in conduction electrons that concentrates electromagnetic fields, generates energetic charge carriers, and subsequently drives high local heating (4). Early work on “antenna-reactor” nanomaterials, in which a light-harvesting nanoantenna is coupled to catalytically active “reactor” domains, delivered a demonstration that hot carriers and near fields could be directed toward specific catalytic sites to influence reactivity, selectivity, and power dependence in chemoselective hydrogenations (5). Subsequent studies have addressed design principles for the flow and extraction of charge carriers in multicomponent plasmonic nanostructures, developed pathways for CO₂ valorization chemistry, and pushed toward industrially relevant reactions (6–8). Viewed against this backdrop, Sharma et al. took the next step by applying these hot-carrier and antenna–reactor design principles to a fully integrated CO₂-assisted oxidative dehydrogenation of propane demonstration.

In parallel, recent work has shown that light can assist nonoxidative propane dehydrogenation by using plasmonic alloys and single-atom sites to modulate C–H activation barriers. Newmeyer et al. demonstrated that dilute CuPt single-atom alloys can localize hot carriers at isolated Pt sites and enhance nonoxidative propane dehydrogenation under visible light, with a clear connection between dopant concentration, charge localization, and catalytic rates (9). Other plasmonic photocatalysts, such as Al@TiO₂ core–shell particles decorated with Pt single atoms and clusters, have also been shown to drive light-assisted propane dehydrogenation (10). These studies, together with broader advances in plasmonic photocatalysis for chemistry have established a conceptual framework in which hot carriers selectively lower barriers for bond activation, while the underlying metal lattice remains near ambient or at moderately elevated temperatures (11, 12). Sharma et al. extend this framework by showing that hot carrier effects can be harnessed not just for a single reaction such as nonoxidative propane dehydrogenation, but for a coupled tandem sequence in which CO₂ serves as a mild oxidant for propane, coproducing CO and propene together rather than propene and H₂ alone (1).

The careful design of their catalyst helps explain why this is possible. The foundation of their material was a dendritic “black gold” colloidosome, where a fibrous nanosilica framework was decorated with high concentrations of colloidal Au nanoparticles (13–15). The nanoengineered dendritic plasmonic colloidosomes (DPC) enable strong optical absorption across the visible and near-IR spectrum, making efficient use of incident light. Subsequently, Ga, Ni, and Mn were dispersed within this scaffold so that catalytically active sites sat close to plasmonic hotspots, allowing hot carriers generated in the gold to reach Ga–Ni–Mn active sites more efficiently. The roles of these three promoters appear complementary. Gallium promotes oxidative dehydrogenation pathways and helps suppress coking, consistent with the broader literature on Ga-based dehydrogenation and on CO₂-assisted oxidative dehydrogenation of propane catalysts (16). Nickel provides sites for CO₂ activation and stabilizes linearly bound CO intermediates. Manganese, while reported to be less active on its own, was shown to tune the electronic structure of the Ga–Ni ensemble and improve the long-term stability of the DPC/GaNiMn catalyst. Control experiments with mono- and bimetallic variants show that none of the simpler formulations achieve the same near-equimolar CO and propene production or the same stability as Ga–Ni–Mn (1). This progression mirrors catalyst developments in thermal CO₂-assisted oxidative dehydrogenation of propane, where multielement alloys on oxide supports have been used to balance CO₂ activation and olefin selectivity (17), but it now occurs in a plasmonic environment where light injection of charge can be engineered along with composition.

A central question in plasmonic catalysis is whether light primarily serves as a heat source or whether hot carriers directly participate in chemical steps (18). Sharma et al. present several levels of evidence for hot carrier involvement (1). Under dark conditions, the apparent activation energy for CO formation is substantially higher than under illumination, indicating a change in the rate-determining step rather than a simple temperature offset from photothermal heating. The rate of CO formation was found to scale approximately with the fourth power of light intensity, an unusually strong dependence not expected for a reaction driven solely by heating. When the catalyst was heated in the dark to temperatures similar to those measured under Xe lamp irradiation (350 °C), the products shifted significantly toward CO and H₂ relative to propene, suggesting that C─C bond cleavage and dry reforming mechanisms were dominant at high temperatures in the dark, distinct from light illumination (1). These observations support their conclusions that hot carriers generated upon visible light excitation open reaction channels that are not accessible under purely thermal conditions, in line with prior observations that nonequilibrium surface dynamics in plasmon-assisted reactions are significant (7, 9, 10). Importantly, the illumination-dependent shift in product selectivity at corresponding reactor temperatures suggests that hot carriers can govern elementary step pathways in tandem reaction networks, distinctly from simply accelerating thermally favorable steps. These results reframe how plasmonic effects should be evaluated in complex catalytic systems.

The work by Sharma et al. builds on efforts from the past fifteen years to control chemistry with light using plasmonic metal nanostructures.

In situ spectroscopy and structural probes further support the interpretation of critical contributions of hot carriers. Diffuse reflectance Fourier transform infrared spectroscopy (DRIFTS) under illumination revealed linearly bound CO on Ni sites, as well as formate and carbonate intermediates, consistent with both direct CO₂ dissociation and hydrogen-assisted reduction pathways (1). X-ray absorption spectroscopy revealed reversible changes in Ni and Ga oxidation states under reaction conditions, indicative of redox cycling associated with CO₂ activation and propane dehydrogenation (1). Transient absorption spectroscopy revealed that the DPC/GaNiMn catalyst exhibits a rapid 200 fs rise time in the excited-state absorption, followed by a notably slower (2.8 ps) decay in the multimetallic system, indicating longer-lived charge carriers compared to monometallic controls, possibly due to interfacial charge transfer. These in situ measurements help connect macroscopic reaction rates to microscopic charge and structural dynamics at the heart of the observed photocatalytic performance. The combination of kinetic data, intensity dependence, and in situ spectroscopy therefore raises new questions about how to incorporate hot carrier–driven steps into microkinetic models of tandem reactions and how to distinguish these contributions from local heating in complex reactors beyond lab scale (19).

Beyond mechanistic insights, the long-term performance of the DPC/GaNiMn photocatalyst is notable. The catalyst maintains a nearly constant CO-to-propene ratio near unity over 500 h of operation, with minimal deactivation (1). In thermal CO₂-assisted oxidative dehydrogenation of propane, the deactivation of the catalyst via nanoparticle sintering and coke deposition is a persistent challenge, especially at the high temperatures required to activate propane and CO₂ (17). In the DPC/GaNiMn system, the combination of moderate bed temperatures, a thermally robust silica scaffold, and carefully tuned Ga–Ni–Mn chemistry appears to mitigate these degradation pathways, but exactly how remains unclear and invites closer examination of how plasmon resonances may mitigate classic deactivation pathways.

More broadly, the recently published PNAS article sits within a growing literature exploring how the highly localized electromagnetic fields that emerge on the surface of plasmonic nanoparticles can reshape the energy landscape for heterogeneous reactions. Sharma et al. make this design space tangible for CO₂-assisted oxidative propane dehydrogenation (1). Coupling CO₂ reduction to propene formation through oxidative dehydrogenation effectively links carbon management and commodity chemical production in a single reactor powered by light and electricity. DPC/GaNiMn catalysts demonstrate how a plasmonic scaffold can effectively harness radiant electromagnetic energy, localize hot carriers near active sites, and stabilize critical surface intermediates. By clearly showing that hot carriers can be exploited to manage both CO₂ activation and C–H activation steps in a single multicomponent material, this work raises new questions about how to integrate optical excitation and multimetallic design into reactor engineering and process design for more complex light-driven chemical reactions.