Skip to content

A reactive force-field for zirconium and hafnium di-boride (Elsevier proof query PDF)

Evidence and attribution

Authority of statements

This corpus PDF is an Elsevier author-query / proof bundle. Full scientific prose is curated against the final article PDF on 2013gouissem-computationa-reactive-force-field-2.

Summary

This wiki page registers an Elsevier author-query / proof PDF (papers/Afif_CompMatSci_2013_galley.pdf) for the Computational Materials Science article that develops a ReaxFF description of zirconium diboride and hafnium diboride (DOI 10.1016/j.commatsci.2012.12.038). The scientific program, co-authored by Adri C. T. van Duin, targets ultra-high-temperature ceramic borides where bond-making and bond-breaking events—including oxidation, defect evolution, and grain-boundary chemistry—are central but difficult to capture with fixed-bond empirical potentials. The manuscript fits bond-order-dependent ReaxFF energy expressions to a combined quantum mechanical training set spanning periodic boride fragments (computed with QuantumWise tools as described in the paper) and cluster targets from Gaussian 09, so that large-scale reactive molecular dynamics of Zr–B and Hf–B chemistry becomes feasible without abandoning the ReaxFF formalism used across the broader corpus. Because the ingested file is a proof bundle, readers should treat pagination, line breaks, and some figure placement as non-final; the version-of-record article PDF curated on 2013gouissem-computationa-reactive-force-field-2 is the preferred locator for section numbers, equations, and edited wording.

Methods

The underlying article follows the standard ReaxFF decomposition of the total energy into bonded (bond-order-dependent), van der Waals, and Coulomb contributions, specialized here to Zr, Hf, and B (see Eq. (3) and surrounding definitions on 2013gouissem-computationa-reactive-force-field-2). Parameter optimization proceeds by weighted least squares minimization against DFT reference data: periodic ZrB₂ and HfB₂ calculations support bulk-like bonding environments, while cluster potential-energy scans and geometries for species such as Zr(BH₂)₂ and Hf(BH₂)₂ (with basis sets and functionals specified in Sections 3–4 of the published paper) anchor molecular fragmentation and coordination trends. Section 5 documents post-fit validation using ReaxFF molecular dynamics on crystal- and defect-like motifs to check consistency with QM or literature references. For this proof path, the wiki does not assert differences in scientific content relative to 2013gouissem-computationa-reactive-force-field-2; the distinction is provenance and citation hygiene (which PDF bytes the manifest tracks) rather than an alternate parameterization.

1 — MD application (atomistic dynamics). Section 5 reports illustrative molecular dynamics with the fitted ReaxFF on crystal- and defect-like ZrB₂/HfB₂ motifs. Engine / code: N/A — explicit MD engine string not transcribed on this proof-ingest page (confirm LAMMPS vs alternatives in pdf_path / 2013gouissem-computationa-reactive-force-field-2). System size & composition: atom counts / supercell definitions for each validation demo—copy from pdf_path. Boundaries / periodicity: PBC details—copy from pdf_path. Ensemble / timestep / duration / thermostat / barostat: N/A — not duplicated here; see canonical page. Temperature: finite-temperature MD is part of the Section 5 narrative in the article (exact K targets in pdf_path). Pressure: N/A — not summarized here. Electric field: N/A —. Replica / enhanced sampling: N/A —.

2 — Force-field training. Parent FF / elements: ReaxFF for Zr–B and Hf–B, using the reduced Eq. (3) energy expression specialized in the manuscript. QM reference: QuantumWise (DFT) for periodic ZrB₂/HfB₂ crystal-phase data; Gaussian 09 for Zr(BH₂)₂ and Hf(BH₂)₂ cluster PES/geometry targets (Section 3/4). Training set: periodic DFT points plus cluster QM scans. Optimization: weighted least squares ReaxFF parameter fit to the QM training data. Reference data / validation: Section 5 MD vs QM/literature checks as reported.

3 — Static QM / DFT. Treated as the QM reference engine for training (not a separate QM-only application paper).

Findings

1 — Outcomes & mechanisms. The optimized ReaxFF reproduces targeted QM energetics and structural metrics for the ZrB₂/HfB₂ training sets, enabling reactive simulations of boride chemistry where simpler fixed-bond models would be inappropriate.

2 — Comparisons. Section 5 compares MD results to QM or literature references for selected motifs (tables on 2013gouissem-computationa-reactive-force-field-2).

3 — Sensitivity & design levers. Which clusters and bulk phases enter the QM training database controls subsequent MD reliability; explicit temperature/pressure sweeps (if any) should be read from pdf_path.

4 — Limitations & outlook. Transferability to complex multicomponent ceramics and long-time high-temperature kinetics remains contingent on expanded training data, as discussed in the article.

5 — Corpus honesty. Cite equations, Section numbers, and validation tables from [[2013gouissem-computationa-reactive-force-field-2]]; this slug documents the Elsevier proof PDF bytes (papers/Afif_CompMatSci_2013_galley.pdf) only.

Limitations

Proof PDF is not authoritative for pagination or final edited wording; transferability to complex ceramics still depends on expanded training (as on the canonical page).

Relevance to group

van Duin collaboration extending ReaxFF to ZrB₂/HfB₂ for extreme-environment materials.

Citations and evidence anchors

Reader notes (navigation)