Supporting Information: Isotope Effects in Water (ReaxFF) — heavy vs light water

Summary¶

This corpus PDF is Supporting Information for a Journal of Physical Chemistry Letters family article on isotope effects in water modeled with ReaxFF, pairing light water (H₂O) against heavy water (D₂O). The opening SI paragraph anchors the discussion to the published ReaxFF energy expression and cites the CHO_2008 and CHON-2017_weak parameter lineage before documenting which bonded, van der Waals, and hydrogen-bond terms were retuned for heavy water. The SI’s role in the knowledge base is documentary: it records how the CHON-2017_weak parameter set is used as the baseline for H₂O, and how a subset of terms is reoptimized to capture the shorter O–D bond length and adjusted hydrogen-bonding balance needed for D₂O structural and dynamical observables within the same overall functional form. Dynamics for D₂O explicitly substitute deuterium mass for hydrogen in the nuclear equations of motion while using the heavy-water-tuned interaction terms. Scientific conclusions, figures, and quantitative performance metrics belong to the primary article page linked below; this entry should not be treated as a standalone publication abstract.

Methods¶

1 — MD application (SI “Simulation details”). ReaxFF molecular dynamics is executed in ADF-2017 (Amsterdam Modeling Suite). Cubic cells with 500 H₂O or D₂O molecules (~1500 oxygen/hydrogen atoms per cell, isotope-dependent): Monte Carlo pack at low density (~0.2 g/cm³), minimize, compress toward ~1.00 g/cm³, then 200 ps NPT MD at room temperature; NPT-equilibrated cells seed 500 ps NVT production (last 250 ps used for bulk property statistics). H₃O⁺/D₃O⁺ inserts follow 500 ps NVT blocks (last 250 ps analyzed). Engine / code: ADF-2017 (the SI states all ReaxFF runs used the Amsterdam Modeling Suite ADF-2017 package). Timestep: 0.25 fs. Thermostat / barostat: Berendsen with pressure and temperature damping 2500 fs and 100 fs for NPT and NVT segments, respectively. PBC cubic boxes. Temperature: room temperature for the equilibration/property blocks tabulated in the SI. Electric field: N/A — not part of the bulk isotope protocol. Enhanced sampling: N/A — standard NVT/NPT MD only. MSD for H₃O⁺/D₃O⁺ averages 50 independent trajectories (SI).

2 — Force-field training (SI). CHON-2017_weak is the H₂O baseline; D₂O uses successive one-parameter ReaxFF reoptimization of O–H bonded, van der Waals, and hydrogen-bond terms (shorter O–D bond, adjusted H-bond balance) while many angular terms track the light-water training unless the SI lists an exception. Nuclear dynamics use deuterium mass. QM/DFT targets and validation tables live in the SI and main letter; copy numeric parameter values from the SI tables.

Findings¶

Mechanism and isotope treatment. The SI argues that CHON-2017_weak already captures the targeted H₂O density, structure, and dynamics observables in their benchmark set, whereas reproducing D₂O requires retuning bonded, van der Waals, and hydrogen-bond contributions—not merely swapping hydrogen for deuterium mass in the equations of motion. That reparameterization narrative ties O–D bond shortening and shifted hydrogen-bond balance to the measured isotope splitting in liquid properties.

Comparisons. Experimental density and related liquid metrics for H₂O vs D₂O are the primary validation anchors; the SI tables should be read beside the main-letter figures on [[20180000-0002-5255-7340-j-phys-chem-isotope-effects]] rather than trusting this wiki alone.

Sensitivity. Temperature and pressure (when NPT blocks appear) enter through the same MD protocols as the parent article; extending the fit to other thermodynamic states requires re-checking diffusion and structure against experiment.

Limitations and corpus honesty. This file is Supporting Information, not the peer-reviewed letter body; pagination and figure numbering follow the SI bundle, and normalized/extracts may remain shorter than the full PDF. Future work implied by the SI—broader state points or additional isotopic species—must be verified against any post-publication corrections on the version-of-record PDF.

Limitations¶

As a supporting-information file, this PDF lacks the full narrative context, peer-reviewed pagination of the main letter, and editorial summary statistics presented in the primary article. The doi field is empty in this wiki stub; operators should align metadata with the version-of-record DOI recorded on [[20180000-0002-5255-7340-j-phys-chem-isotope-effects]].

Relevance to group¶

The document supports reproducibility for isotope-specific water modeling with CHON-family ReaxFF parameters and is linked from the canonical isotope-effects article page for van Duin-group water/electrolyte workflows.

Citations and evidence anchors¶

Primary article: [[20180000-0002-5255-7340-j-phys-chem-isotope-effects]]