Activity, Selectivity, and Durability of Ruthenium Nanoparticle Catalysts for Ammonia Synthesis by Reactive Molecular Dynamics Simulation: The Size Effect
Summary¶
Develops a ReaxFF parameterization for the Ru–N–H ternary system (trained against DFT data summarized in the paper and SI), then uses LAMMPS ReaxFF MD to study NH₃ formation from N₂ and H₂ over Ru nanoparticles (diameters 3–10 nm). The study reports size-dependent patterns in activity, selectivity, and mechanical durability metrics derived from surface-site distributions and stress measures. The introduction recalls that NH₃ synthesis remains difficult because N₂ is strongly triple-bonded, that industrial Haber–Bosch conditions are hot and pressurized, and that Ru, Os, and Fe rank highly on volcano-style activity plots—with Ru often outperforming Fe in both theory and experiment. The authors motivate ReaxFF MD as a route to durability and selectivity metrics that are hard to reach from static DFT alone because of system size and time-scale limits.
Methods¶
- Force field: ReaxFF parameters for Ru–N–H optimized with a successive one-parameter search against first-principles training sets (equations of state, surface energies, adsorption and reaction pathways, bond dissociations); parameter tables referenced in Table S4 (SI).
- MD: LAMMPS; Verlet integration with 0.5 fs timestep; NVT at 1500 K with a Nosé–Hoover thermostat (damping 0.01 fs⁻¹); N₂/H₂ pressures explored in 300–1000 atm as described in the article; spherical Ru NP on a Ru slab support model (see Fig. 1). The abstract states simulations predict activities, selectivities, and durabilities jointly and discuss temperature and pressure effects alongside size.
Additional controls. PBC: three-dimensional PBC for the supported NP supercells. Barostat: N/A — NPT not used for the summarized constant-volume NVT production. Gas pressure: 300–1000 atm N₂/H₂ ratios set gas density in the fixed simulation cell rather than via fluctuating volume. Electric field: N/A — bias not applied. Enhanced sampling: N/A — umbrella / metadynamics / replica exchange not reported. Equilibration / production lengths in ps/ns—see pdf_path/SI for full staging.
Findings¶
Outcomes. NH₃ formation activity peaks near 4 nm Ru particles in the explored grid, while 10 nm particles show the strongest selectivity among reported cases; hcp/fcc/top site populations shift with size and help rationalize reactivity.
Comparisons. Results are discussed relative to Haber–Bosch literature and volcano trends for Ru/Fe catalysts, with explicit caveats about the elevated 1500 K acceleration strategy.
Sensitivity. H₂ pressure raises selectivity more steeply than N₂ pressure in the scanned window; mean stress vs stress concentration trades off with diameter, yielding 5 nm as the best durability compromise in the study.
Limitations and PDF grounding. High temperature/pressure MD is a kinetic accelerator, not a literal industrial reactor replica—see article barrier checks. All quantitative tables: pdf_path.
Limitations¶
- Operating conditions (1500 K, high gas pressures) are chosen to accelerate chemistry in MD and are not direct Haber–Bosch process replicas; the paper discusses barrier checks to motivate the elevated temperature approach.
Relevance to group¶
Illustrates ReaxFF training from DFT plus large-scale catalytic MD on metal NPs.