Skip to content

Optimization of a New Reactive Force Field for Silver-Based Materials (ChemRxiv preprint PDF)

Summary

This wiki slug tracks the ChemRxiv v1 preprint PDF for AgSCH-ReaxFF, a ReaxFF reparameterization targeting silver clusters, slabs, and thiol/thiolate chemistry relevant to self-assembled monolayers and metal–organosulfur interfaces. The work overlaps substantially with the peer-reviewed J. Chem. Theory Comput. article 2020clement-dulong-j-chem-theor-optimization-new (DOI 10.1021/acs.jctc.0c00480), which should be treated as the primary citable reference for quantitative benchmarks and final supplementary tables. The preprint narrative describes Monte Carlo force-field (MCFF) optimization in ADF against DFT training sets containing >120 QM structures, followed by validation MD in ADF and LAMMPS with 0.25 fs timesteps, Berendsen thermostats (5 fs damping), and ~100 ps segments for SAM-related snapshots (e.g., comparisons near 100 K in cited figures).

Methods

Parallel to the JCTC article: MCFF sampling of ReaxFF parameters against an extended QM training set (>120 structures in the ChemRxiv abstract) spanning silver clusters/slabs and silver–thiolate chemistry. Validation includes Ag₂₀ pyramidal thiolate adsorption benchmarks (compared with gold analogs in the authors’ discussion) and SAM-like Ag(111) assemblies. MD validation comparing Ag vs Au SAM structural motifs appears in the preprint/SI narrative. Because extraction_quality is partial for this ChemRxiv path, operators should prefer the journal PDF for exhaustive parameter listings and final benchmark tables.

MD validation (as in the linked JCTC page): ReaxFF molecular dynamics in ADF and LAMMPS; 0.25 fs timestep; Berendsen thermostat (5 fs damping); ~100 ps NVT-style segments for SAM snapshots; PBC supercells. Barostat: N/A for the quoted NVT validation windows. Hydrostatic pressure control: N/A for those NVT windows. Electric field: N/A. Umbrella/metadynamics: N/A.

Findings

The parameter set improves average cluster and slab energetics vs legacy Ag ReaxFF descriptions; Ag\(_{20}\) pyramidal thiolate adsorption tracks gold benchmarks in the authors’ comparisons. Low-coordination Ag sites still show edge-shortening artifacts, flagged as a known limitation. Ag(111) SAM simulations do not reproduce Au–S–Au “staple” motifs seen for gold, consistent with differing experimental pictures for silver interfaces.

The ChemRxiv abstract also flags a persistent ReaxFF limitation: under-coordinated edge sites can show shortened metal–metal distances that are not fully captured, so cluster-edge geometries should be interpreted cautiously in downstream applications.

Limitations

Preprint formatting may differ from JCTC; residual low-coordination errors; anchor quantitative claims to the journal version when possible. If pdf_sha256 ever diverges between ChemRxiv revisions, update raw/MANIFEST.jsonl in a controlled ingest rather than silent overwrites.

Reader notes (navigation)

Chunk consumers should prefer paper:2020clement-dulong-j-chem-theor-optimization-new for primary scientific text; this page remains valuable for preprint traceability and SI filename alignment during corpus migrations.

Relevance to group

Provenance record for ChemRxiv blob; canonical benchmarks live under 2020clement-dulong-j-chem-theor-optimization-new.

Citations and evidence anchors

Reader notes (navigation)

Curator note (2026-04-20)

Prefer 2020clement-dulong-j-chem-theor-optimization-new for citable science; keep this slug for ChemRxiv provenance and version history when needed.