Skip to content

ReaxFF Molecular Dynamics Study on the Influence of Temperature on Adsorption, Desorption, and Decomposition at the Acetic Acid/Water/ZnO(1010) Interface Enabling Cold Sintering

Evidence and attribution

Authority of statements

Prose below summarizes the ACS article identified by doi, title, and pdf_path in the front matter.

Summary

Reactive molecular dynamics with a combined ReaxFF potential (O/H/Zn for ZnO–water from prior work plus O/H/C for organics, with merged optimization for acetic acid deprotonation in water) is applied to a ZnO(101̅0) slab contacting a concentrated acetic acid/water mixture (800 H\(_2\)O + 200 CH\(_3\)COOH, intended to mimic low-pH cold-sintering environments). Simulations span effective 300–1200 K with staged heating and long equilibration; species populations and coverages are analyzed in the surface layer versus gas-like region. The work connects adsorption/desorption trends to Langmuir-type behavior, tracks acetate decomposition toward CO\(_2\) and formaldehyde, and discusses decomposition pathways and barrier estimates from constrained ReaxFF runs.

Methods

  • Force field: H/O/Zn potential for ZnO–water (literature parameterization for water on ZnO surfaces) combined with C/H/O parameters for biomolecular solutions; cross-terms optimized so that dissociative adsorption energies of acetic acid on ZnO(101̅0) match DFT trends; Zn–C bonding terms disabled as nonphysical for this system.
  • Energy minimizations on 12-layer ZnO slabs (360 Zn + 360 O per layer stack) with controlled acetic acid coverage \(\theta\) (e.g. 30 molecules for \(\theta=1\)) to obtain adsorption energies via an adsorption-energy formula in the paper.
  • Dynamics: ADF ReaxFF MD; 0.25 fs integration (velocity Verlet); Berendsen thermostat with 0.1 ps damping; simulation cell about 60 × 65 × 300 Å\(^3\) including vacuum; liquid slab placed on the oxide after density-based equilibration (~1 g/cm\(^3\) water, ~1.05 g/cm\(^3\) acetic acid).
  • Protocol: Temperature raised in 100 K steps with 50 ps equilibration per step, then 1 ns at the target temperature; production statistics from the last 0.5 ns (25 fs sampling) with surface species defined by ≤ 3.0 Å from surface Zn.
  • Decomposition path analysis: Restrained ReaxFF runs at very low temperature (0.25 K) with distance-based restraints (exponential restraint form in the paper) to step formaldehyde oxidation toward CO\(_2\) on the surface; barrier heights reported in kcal/mol for sequential steps.

Production MD (interface, consolidated). Same protocol as the bullets above: NVT-style Berendsen-thermostatted ReaxFF molecular dynamics in AMS/ADF on the 60 × 65 × 300 Å\(^3\) periodic supercell with PBC; 0.25 fs velocity Verlet; staged 100 K ramps with 50 ps equilibration per step and 1 ns at target T; statistics from the last 0.5 ns (25 fs output interval). Barostat / pressure: N/A — constant-volume (no NPT). Electric field: N/A — not used. Enhanced sampling: N/A — umbrella / metadynamics / replica exchange not used for production trajectories (low-T restrained scans only).

Findings

At 300 K, roughly one-third of water molecules adsorb on the surface (mix of molecular and dissociated forms, including half-dissociated (2×1)-like patterns consistent with prior DFT/ReaxFF); about 58% of acetic acid molecules adsorb, with chelating, bridging, and monodentate acetate modes and two deprotonation routes (to lattice oxygen vs bridging hydroxyl/water formation). Overall surface coverage by water- and acid-derived species is high (~80% of sites), with acetates competitively stabilized relative to molecular water.

With heating, adsorption/desorption and decomposition evolve: acetate surface fraction can rise to a maximum near ~800 K in the simulated protocol, then CO\(_2\) appears as decomposition sets in; trends align qualitatively with experiments that coupled gas-phase acid and CO\(_2\) evolution, with the paper noting temperature offsets versus experiment due to faster simulated heating. Surface coverage decreases with temperature in a manner consistent with a Frumkin–Fowler–Guggenheim (modified Langmuir) picture with positive lateral interaction energies from fits. Acetate decomposition to H\(_2\)CO and CO\(_2\) proceeds via nucleophilic attack on the methyl group and radical intermediates as illustrated in the paper; formaldehyde is reported to desorb rather than further oxidize on the simulated timescales. Restrained MD along the CH\(_2\)O → CO\(_2\) pathway yields small step barriers (order ~0.5–6 kcal/mol for the documented stages), interpreted as consistent with observed decomposition at elevated temperature.

Limitations

  • Effective temperatures and rapid heating in MD shift quantitative temperatures relative to experiment; the authors caution that trends are more reliable than absolute T values.
  • CH\(_2\)O further oxidation on ZnO is not observed within the simulated timescales.

Reader notes (navigation)