Improvement of the ReaxFF Description for Functionalized Hydrocarbon/Water Weak Interactions in the Condensed Phase (proof PDF)
PDF variant
ACS line-numbered proof. Canonical note: 2018zhang-j-phys-chem-improvement-reaxff with non-proof PDF when available.
Summary¶
The proof PDF papers/Zhang_JPCB_Weak_Interactions_2018_proof.pdf corresponds to J. Phys. Chem. B DOI 10.1021/acs.jpcb.8b01127, “Improvement of the ReaxFF Description for Functionalized Hydrocarbon/Water Weak Interactions in the Condensed Phase.” The abstract in the extract states that the protein-2013 aqueous ReaxFF branch simulates biomolecules and membrane fuel cells but inaccurately describes weak interactions of functionalized hydrocarbons with water in condensed phases, especially densities. It introduces CHON-2017_weak built on protein-2013 to improve C/H/O/N weak interactions, validated with MD showing experimental density trends for pure and mixed systems, hexane–water phase separation, and ethanol or tetramethylammonium (TMA) dissolution. For alkaline membrane fuel cells, it reports structural properties and TMA degradation chemistry in alkaline water, plus small-molecule reaction benchmarks, and states an additional degradation pathway for TMA appears more energetically favorable than the main mechanism from prior QM studies.
Methods¶
This slug indexes an ACS line-numbered proof of the same DOI as 2018zhang-j-phys-chem-improvement-reaxff; use that version-of-record page for definitive tables.
Force-field training (proof text). The Introduction on the proof PDF explains QM/DFT cost limits versus ReaxFF, contrasts generic MM force fields (AMBER, CHARMM, OPLS, etc.), and traces protein-2013 lineage: first-row carbon from C/H/O-2008 combustion training, oxygen/hydrogen from the first-generation water line, nitrogen from glycine-in-water extensions. It records protein-2013 weaknesses—liquid water density underestimated, hydrocarbon density overestimated—that motivate CHON-2017_weak while preserving prior AEM-related successes. Training targets, optimization steps, and reference data tables match the issue article.
MD application (proof text mirrors the issue article). Engine / code: LAMMPS + standalone ReaxFF development path as in §2.2 of the JPCB article. System / composition: validation supercells include ~500-molecule liquid water boxes, ~200-molecule n-alkane periodic cells, hexane–water mixtures, TMA/OH⁻/water compositions, and crystal amino acid cells as enumerated in §2.2—see the supercell atom counts in the issue PDF for each figure. Ensemble / protocol: NPT equilibration of liquid densities from low-density starts, then 500 ps NPT at room temperature (density averaged last 400 ps); 500 ps NVT segments for RDF/MSD and TMA/OH⁻/water mixtures; amino acid crystals use 100 K NPT to limit thermal noise. Timestep 0.25 fs; Berendsen thermostat with 100 fs damping for NVT and 2500 fs pressure damping for NPT. PBC in all directions. For line-level tables and any proof-vs-issue deltas, prefer [[2018zhang-j-phys-chem-improvement-reaxff]].
Findings¶
Qualitative conclusions follow the abstract above; numerical density deviations and reaction energetics should be read from the version-of-record PDF linked on the canonical page. The proof extract also notes prior AEM studies of functionalized poly(phenylene oxide) membranes where ReaxFF predicted hydroxide and water diffusion constants successfully and captured degradation trends for large hydrophilic cations—context for why TMA chemistry is used as a validation case for CHON-2017_weak. Proof line numbers and placeholder page fields (“XXXX”) in the extract header should not be used as citation locators; migrate to the issue PDF when available.
Limitations¶
Proof PDFs can differ in pagination from the final issue. Operators should migrate citations to the non-proof file when ingested.
Relevance to group¶
Van Duin-group aqueous-branch ReaxFF refinement for biomolecular and fuel-cell-adjacent electrolyte modeling.
CHON-2017_weak is explicitly positioned as fixing condensed-phase densities and mixing behavior where protein-2013 struggled; applications that rely on accurate hydration free energies should cite validation cases in the JPCB article (pure liquids, mixtures, and TMA chemistry) rather than assuming transfer to arbitrary organics without testing.
Alkaline fuel-cell degradation pathways for TMA involve multiple protonation states; reproducers should verify that their simulation pH and hydroxide count match the cases enumerated in the published article’s Methods.
Reader notes (navigation)¶
Citations and evidence anchors¶
DOI: 10.1021/acs.jpcb.8b01127.