Hydrogen Activation via Dihydride Formation on a Rh1/Fe3O4(001) Single-Atom Catalyst

Technical Summary: Hydrogen Activation via Dihydride Formation on a Rh1/Fe3O4(001) Single-Atom Catalyst

Problem Statement
Hydrogen activation is a fundamental step in catalytic hydrogenation, yet the mechanisms differ significantly between homogeneous and heterogeneous systems. Homogeneous catalysts typically activate H₂ through the formation of dihydride or dihydrogen intermediates at a single metal center, enabling high selectivity and precise control. In contrast, heterogeneous catalysts usually rely on dissociative adsorption on metal nanoparticles followed by atomic hydrogen diffusion or spillover onto the support. While spillover can enhance activity, it often creates a mobile hydrogen reservoir that compromises selectivity for partially hydrogenated products. Single-atom catalysts (SACs) have been proposed as a bridge between these regimes due to their structural similarity to homogeneous complexes. However, direct experimental evidence that SACs operate via dihydride pathways—rather than through metal-hydride formation or spillover—has remained elusive.

Methodology
The study employs a combined experimental and theoretical approach to investigate hydrogen activation on isolated Rh adatoms supported on a reducible Fe₃O₄(001) surface.

• Experimental Techniques:
  • Scanning Tunneling Microscopy (STM): Time-lapse STM was conducted at room temperature under H₂ partial pressure (4 × 10⁻⁹ mbar) to visualize reversible adsorption and desorption events at individual Rh₁ sites.
  • X-ray Photoelectron Spectroscopy (XPS): Used to analyze the Rh 3d and O 1s regions after D₂ exposure to detect electronic state changes and rule out the formation of surface hydroxyls (OD species).
  • Temperature-Programmed Desorption (TPD): Performed using D₂ and H₂/D₂ mixtures to quantify adsorption energies and test for isotope exchange (scrambling) or spillover products (e.g., D₂O, DH).
• Theoretical Methods:
  • Density Functional Theory (DFT+U): Calculated reaction pathways, adsorption energies, and atomic structures for D₂ adsorption, dissociation, and spillover.
  • Advanced Benchmarks: Results were validated using r²SCAN+U, HSE06 functionals, and Random Phase Approximation (RPA) calculations to ensure robustness against self-interaction errors.
  • Thermodynamic Analysis: Gibbs free energy profiles were constructed at 300 K, incorporating zero-point energy (ZPE) corrections and entropic contributions to compare desorption versus spillover barriers.

Key Results

1. Localized Adsorption without Spillover: STM observations revealed reversible height fluctuations at isolated Rh₁ sites upon H₂ exposure, with no new features appearing on the oxide lattice. XPS confirmed the absence of OD species (hydroxyls) on the support, even after prolonged D₂ exposure, indicating that hydrogen remains bound to the Rh atom and does not migrate to the oxide.
2. Strong, Reversible Binding: TPD experiments showed a distinct D₂ desorption peak at 295 K. Analysis yielded a strong adsorption energy of approximately 0.98 ± 0.06 eV. Crucially, exposure to equimolar H₂/D₂ mixtures resulted in competitive adsorption but no isotope scrambling (no DH signal), confirming that hydrogen remains localized at the Rh₁ site without exchanging with the support or other sites.
3. Dihydride Formation Mechanism: DFT calculations identified two local minima: a weakly bound dihydrogen-like complex (D–D distance 0.93 Å) and a more stable dihydride configuration (D–D distance 1.8 Å). The dihydride state is thermodynamically favored by 0.22 eV. The conversion from molecular D₂ to the dihydride is calculated to be barrierless.
4. Kinetic Suppression of Spillover: While the final state of hydrogen spillover (migration to the oxide) is thermodynamically accessible, the activation barrier for spillover (0.86 eV) is higher than the free-energy barrier for recombinative H₂ desorption (0.71 eV) at 300 K. This difference arises because desorption releases translational entropy, whereas spillover keeps hydrogen surface-bound. Consequently, desorption is kinetically preferred, preventing the formation of a mobile hydrogen reservoir.
5. Transient Heterolytic Cleavage: The calculations revealed a transient "split-hydride" configuration where one hydrogen atom transfers to the oxygen atom directly bonded to Rh. While this state is sparsely populated (~9 × 10⁻⁴ at 300 K), it suggests a mechanism of local metal-support cooperation that mimics bifunctional activation in molecular complexes without long-range diffusion.

Significance and Claims
The authors claim that this work provides direct experimental evidence that isolated Rh atoms on Fe₃O₄(001) activate hydrogen through a dihydride pathway analogous to homogeneous catalysis. By forming a stable dihydride species without atomic hydrogen spillover, these single-atom sites replicate the controlled reactivity of molecular complexes. This finding establishes a mechanistic bridge between homogeneous and heterogeneous catalysis, demonstrating that SACs can host discrete metal-hydride species at finite temperatures. The localization of hydrogen prevents the formation of a mobile reservoir, which the authors suggest is a key factor in enabling selective hydrogenation reactions (such as acetylene to ethylene) that are difficult to achieve on traditional nanoparticle catalysts. The study concludes that geometric confinement and covalent metal-oxide interactions can stabilize molecular-like hydride species, offering a foundation for designing SACs that combine molecular precision with the robustness of solid surfaces.