Skip to content

Development of a Reactive Force Field for Catalytic C/H/O Conversion on Cu and Cu-Oxide Surfaces and Application to Cu/CuO Chemical Looping

Fully interactive Cu/C/H/O ReaxFF development for Cu-metal and Cu-oxide catalysis, combined with new ReaxFF-based transition-state search tools, followed by reactive MD demonstrations including chemical looping scenarios (Cu oxidation with glucose; hydrocarbon oxidation with Cu oxide).

Summary

The work develops a Cu/C/H/O ReaxFF in three stages: re-optimization of the Cu subset against an extended training set including Cu cluster properties; merging with an existing C/H/O ReaxFF; and fitting Cu–C/H/O cross terms using extensive DFT data on binding energies and elementary steps. A transition-state search framework under the ReaxFF formalism is introduced, together with algorithms to build reaction paths and high–degrees-of-freedom coordinate scans. Reactive MD tests probe elementary C/H/O surface chemistry on Cu and two chemical looping–style cases: CuO formation from metallic Cu with glucose, and hydrocarbon oxidation using a copper oxide oxidizer.

Methods

1 — MD application (atomistic dynamics). The abstract reports reactive MD of ≫1000-atom Cu surface and chemical-looping-style Cu/CuO systems using the developed Cu/C/H/O ReaxFF (see J. Phys. Chem. C and SI for slab stoichiometry, PBC, NVT/NVE-style protocols, 0.25–1.0 fs-class timestep choices, equilibration and production lengths in ps/ns, Nose–Hoover-type thermostat use where given, and temperature setpoints in K for isothermal segments not restated in the one-page extract, and whether NPT or NVT is used (many catalysis surfaces are run NVT without barostat hydrostatic pressure control). The indexed excerpt does not restate the MD engine string; the article/SI is authoritative (N/A here for exact LAMMPS line—confirm in full PDF; reactive MD is the stated tier). Shear, shock, replica/metadynamics—N/A in the abstract framing. Electric bias/field in MD—N/A in the excepted summary.

2 — Force-field training (ReaxFF and related). The authors build a Cu/C/H/O ReaxFF in three steps: (1) re-optimization of the Cu subset using an extended training set (including Cu cluster data); (2) merge with an existing C/H/O ReaxFF; (3) fit Cu–C/H/O cross terms to extensive DFT energies for binding and elementary reaction steps. Parametrization uses a ReaxFF optimization workflow against the DFT reference set; EEM-style charges update every iteration. They also report in-ReaxFF transition-state (TS) search and reaction path/coordinate-scan algorithms (new with respect to the paper’s stated ReaxFF tooling goals).

3 — Static QM / DFT (reference calculations). DFT (training/validation data): DMol3 GGA-rPBE, ECPs, unrestricted spin, 0.006 Ha smearing for periodic cells, 4 × 4 × 1 k-mesh, 4.5 Å orbital cutoff, and stated SCF/optimization tolerances. Nonperiodic clusters use integer occupations. Binding energies use the standard surface–adsorbate decomposition in the paper. N/A — stand-alone hybrid DFT beyond this training set for MD production.

4 — Galley corpus note. Ingested PDF is a galley; align pagination to the VOR when citing section numbers.

Findings

Primary results. The fitted ReaxFF matches key DFT trends for Cu/C/H/O interactions and elementary steps on Cu surfaces (authors’ assessment in the article). Test MD on Cu surfaces shows H transfer and H₂/CHO-type dissociation reactions consistent with a complex C/H/O surface network on Cu. Chemical looping-oriented cases include CuO formation from glucose + metallic Cu and hydrocarbon oxidation with copper oxide as oxidant; reaction pathways and redox differ between fuels in the reported scenarios (abstract and discussion). Comparisons are DFT vs ReaxFF and case vs case within the MD suite. Sensitivity to fuel choice and surface redox state is a stated theme. Limitations include transferability to uncovered catalyst morphologies and pressures not in the training set; however much of that remains in the main text. Version-of-record check recommended because the local file is a galley proof.

Limitations

The corpus PDF is a galley; final pagination and minor editorial values may differ from the version of record. TS search and MD numerical details beyond §2.1 require the full article/SI.

Relevance to group

Extends Reaxff parameterization for heterogeneous catalysis and combustion/chemical looping on Cu surfaces with open methodological contributions (TS search on ReaxFF).

Citations and evidence anchors

Reader notes (navigation)

  • paper_keywords includes keyword:galley-or-proof-pdf for this ingest.