Skip to content

Development and application of ReaxFF methodology for understanding the chemical dynamics of metal carbonates in aqueous solutions

Summary

Aqueous carbonate chemistry controls pH, mineral dissolution, and electrolyte speciation in geochemistry and in energy devices where CO₂ interfaces with brines or hydroxides. Dasgupta, Chen, and van Duin extend ReaxFF to metal carbonates by adding Na⁺, Ca²⁺, Mg²⁺, and CO₃²⁻ interactions consistent with existing ReaxFF water and electrolyte building blocks. The parameterization targets DFT data on equations of state, formation and reaction enthalpies, angular distortions, and vibrational benchmarks for condensed carbonates/oxides and selected gas-phase clusters. Subsequent molecular dynamics at 300 K and 700 K examines solvation shells, ion diffusion, and speciation between carbonate and bicarbonate forms in explicit water.

Methods

1 — MD application (atomistic dynamics). - Engine / code: LAMMPS with ReaxFF. - System size & composition: Aqueous carbonate systems with Na/Ca/Mg carbonate chemistry in explicit water; atom counts and exact box compositions are not stated in the indexed extract. - Boundaries / periodicity: N/A — boundary condition details are not stated in the indexed extract/galley text. - Ensemble: NVT is reported for the validation trajectories. - Timestep: N/A — timestep value is not stated in the indexed extract/galley text. - Duration / stages: N/A — equilibration/production durations are not stated in the indexed extract/galley text. - Thermostat: N/A — thermostat type and damping constants are not stated in the indexed extract/galley text. - Barostat: N/A — NVT validation is described; no pressure-coupling/barostat setup is reported in the indexed extract. - Temperature: 300 K and 700 K are the reported simulation temperatures (ambient vs supercritical-condition comparison). - Pressure: N/A — pressure target/control is not stated in the indexed extract/galley text. - Electric field: N/A — no applied electric field is reported. - Replica / enhanced sampling: N/A — no umbrella sampling, metadynamics, or replica-exchange method is reported. - Electrostatics / cutoffs / QEq: ReaxFF charge equilibration and nonbonded settings follow the LAMMPS ReaxFF implementation used in the study; numeric settings are not stated in the indexed extract.

2 — Force-field training. - Parent FF / elements: ReaxFF parameterization extended for carbonate aqueous chemistry with C/H/O/Na/Ca/Mg interactions, built to be compatible with existing ReaxFF water/electrolyte descriptions. - QM reference: VASP with PBE and PAW; plane-wave cutoff reported as 520 eV, with per-structure k-point meshes. - Training set: Condensed-phase carbonate/oxide structures and gas-phase clusters, with targets including equations of state, reaction/formation enthalpies, angular distortions, and vibrational benchmarks. - Optimization: Weighted ReaxFF fitting/optimization against the QM target set; exact optimizer settings/weights are not stated in the indexed extract. - Reference data used: Comparisons to published computational/experimental references are used for validation context; no separate in-house experimental campaign is reported.

3 — Static QM / DFT-only (as training data, not a standalone DFT application study). - Functional: PBE. - Dispersion: N/A — dispersion correction details are not stated in the indexed extract. - Basis: Plane-wave PAW framework (520 eV cutoff reported). - k-sampling: Structure-dependent k meshes are reported. - Structures / pathways: Carbonate/oxide condensed phases and selected gas-phase clusters for force-field fitting targets. - Properties computed: EOS, reaction/formation energetics, angular distortions, and vibrational observables used to train/evaluate the ReaxFF parameterization.

Findings

1 — Outcomes and mechanisms. The reported parameterization reproduces the targeted QM training observables for carbonate-related chemistries and enables aqueous reactive-MD analysis at ambient and supercritical-condition temperatures. Validation analyses emphasize solvation/coordination structure, ionic mobility trends, and carbonate/bicarbonate-sensitive local chemistry for Na/Ca/Mg systems.

2 — Comparisons. The paper compares fitted ReaxFF predictions against the underlying DFT reference set and against literature comparators used for RDF, coordination, and diffusivity-level behavior checks.

3 — Sensitivity and design levers. Temperature (300 K vs 700 K) is the explicit reported lever used to probe changes in aqueous carbonate dynamics and speciation behavior.

4 — Limitations and outlook (as reported). Applicability is bounded by the trained element/chemistry space (C/H/O/Na/Ca/Mg carbonate systems); extension to additional ion chemistries or conditions requires new parameterization/validation beyond this study.

5 — Corpus / KB honesty. The local source is a galley PDF with partial extract coverage, so this page avoids claiming numeric MD durations, pressure controls, or detailed thermostat/barostat constants that are not present in the indexed text; those should be copied from a local version-of-record PDF when available.

Limitations

The corpus includes a galley PDF that may differ cosmetically from the final issue. Scientific scope is limited to chemistries represented in the training set; transfer to transition metals, sulfates, or extreme pH regimes requires additional validation. Car–Parrinello comparisons in the article are useful sanity checks for proton-transfer motifs but do not replace systematic benchmarks for every mineral surface of interest.

Relevance to group

Foundational ReaxFF extension for geochemical and electrolyte carbonate systems led from the van Duin group.

Citations and evidence anchors