Skip to content

Development of a ReaxFF reactive force field for Fe/Cr/O/S and application to oxidation of butane over a pyrite-covered Cr2O3 catalyst

Evidence and attribution

Authority of statements

Prose below summarizes the publication identified by doi and pdf_path. Definitive protocol tables and numbers remain in the PDF / 2015shin-venue-microsoft-word SI package.

Summary

Shin et al. develop a ReaxFF parametrization for Fe/Cr/O/S against QM data and apply it to reactive MD of butane oxidation on Cr oxide surfaces, comparing clean chromia with pyrite (FeS₂)-modified models motivated by coal-derived fuels and slagging mineralogy. At 1600 K (abstract), simulations highlight surface oxygen species that drive dehydrogenation to radicals and subsequent C–O coupling, producing CH₂O as a major partial-oxidation product on clean surfaces while FeS₂ accelerates complete oxidation to CO and CO₂ with surface reconstruction and SOH release pathways.

Methods

Force-field training. Scope: new Fe/Cr/O/S cross-terms within ReaxFF, parametrized against QM energies/reaction data for oxide, sulfide, and hydrocarbon motifs relevant to Cr–O, Fe–S, and S–O chemistry (Computational Methods + SI). QM reference level, training set composition, optimizer, and weighting: pdf_path and SI tables.

MD application (surface oxidation). System chemistry: butane + O₂ reacting on Cr₂O₃ slabs with optional FeS₂ patches (initial layouts summarized in SI captions; main text defines slab stoichiometries). Temperature: headline oxidation trajectories at 1600 K (abstract). Engine, timestep, thermostat, duration (ps/ns), PBC, and any NPT segments are specified in Computational Methods on pdf_path (not enumerated on this page from the short abstract excerpt). Electric field / enhanced sampling: N/A in the abstract framing. Barostat: N/A unless Methods explicitly add NPT for these combustion-style runs (see pdf_path).

Static QM / DFT: QM enters as training/validation for ReaxFF, not as standalone AIMD production for the headline 1600 K catalytic trajectories.

Findings

Clean Cr oxide: Dehydrogenation initiated by surface oxygen yields butane radicals and surface OH; radical intermediates form C–O bonds or C=C when neighboring carbons dehydrogenate, producing light alkenes; CH₂O is the major partial oxidation product on clean surfaces under the simulated conditions (abstract).

Pyrite-modified oxide: FeS₂ accelerates complete oxidation to CO₂ and CO; surface reconstruction by pyrite is proposed as the origin of the selectivity shift (abstract). SOH release appears on the modified surface whereas clean surfaces favor reoxidation via H₂O desorption and O₂ adsorption at vacancies in the abstract’s mechanistic sketch.

Sensitivity: The abstract frames results at 1600 K “combustion-like” conditions; lower-temperature ODH relevance is not claimed there.

Limitations

Single high-temperature window in the abstract is far from many industrial ODH conditions. ReaxFF cannot subsume all electronic redox subtleties of sulfide–oxide interfaces without targeted validation.

Relevance to group

Illustrates multi-element ReaxFF parameterization for sulfur-bearing fossil-catalyst chemistries central to Shin/van Duin collaborations.

Citations and evidence anchors

  • DOI 10.1021/acscatal.5b01766; papers/Shin_ACS_Catalysis_2015.pdf.
  • normalized/extracts/2015shin-venue-research_p1-2.txt.