Experimental and computational investigations of ethane and ethylene kinetics with copper oxide particles for Chemical Looping Combustion

Summary¶

Chemical looping combustion routes hydrocarbon fuels through redox cycles of solid oxygen carriers such as metal oxides, demanding elementary insight into how small hydrocarbons oxidize on carrier surfaces at relevant temperatures. This Proceedings of the Combustion Institute article combines fixed-bed flow-reactor experiments with ReaxFF reactive molecular dynamics to study methane, ethane, and ethylene interacting with copper oxide particles as a model oxygen carrier. Experiments span roughly 500–1000 K with time-resolved species detection, while simulations explore 1000–2000 K atomistic trajectories intended to expose bond-making and bond-breaking sequences that feed reduced kinetic descriptions for larger models. Adri C. T. van Duin is among the coauthors, linking the study to the group’s broader combustion-oriented reactive force-field applications.

Methods¶

MD application (ReaxFF, LAMMPS-class)¶

Engine / code: LAMMPS with ReaxFF for reactive molecular dynamics of hydrocarbon fragments on CuO surfaces at elevated temperature (about 1000–2000 K in the abstract framing); QEq cadence and pair_style details N/A—see pdf_path/SI.
System size & composition: CuO surface/supercell models with C₁/C₂ adsorbates; atom counts, stoichiometry, and vacuum thickness N/A in this note—pdf_path tables/figures.
Boundaries / periodicity: PBC slab supercells with lateral periodicity and a vacuum region normal to the surface (standard ReaxFF surface practice; exact choices N/A here—pdf_path).
Ensemble / stages: NVE RMD and/or NVT segments with a thermostat as given in PCI Methods; equilibration then production RMD with 0.1–0.25 fs timestep and ~0.1–2 ns trajectory lengths as tabulated in pdf_path (indicative ranges; confirm there). N/A—NPT barostat on the high-T RMD unless the article lists bulk NPT equilibration of CuO separately in pdf_path.
Temperature (MD): K-series scans (about 1000–2000 K class in the abstract) as in pdf_path (temperature) (ramps) / isotherms. Reactor T for experiments is about 500–1000 K (non-MD).
Barostat / pressure in MD: N/A for isotropic NPT on the RMD slab—fixed-cell NVE/NVT (pressure) control is not the same as the flow reactor (pressure); (reference) (stress)** in pdf_path if reported.
Shear / shock (MSST), external electric field, and enhanced sampling: N/A in the PCI RMD scope summarized here. ReaxFF Coulomb and QEq schedules: pdf_path.

Experiments (chemical-looping, fixed bed)¶

Fixed-bed flow reactor with molecular-beam mass spectrometry (time-resolved species) and gas chromatography (stable oxygenates and C₁/C₂ products). T about 500–1000 K in the work’s experimental window (see pdf_path for τ_residence and feed (composition)). The paper frames MD and reactor data as comparative (not a direct one-to-one rate (fit)).

Force-field training in this work¶

N/A as a new ReaxFF training article—the work applies a published Cu–O–H–C ReaxFF as cited in the PCI text/SI.

Findings¶

Both experiment and simulation report oxygenated and partially oxidized products such as acetaldehyde, formaldehyde, carbon monoxide, and water, supporting a compact set of C1/C2 pathways that can be embedded into larger surrogate kinetic schemes for chemical looping. The authors also document mismatches: for example, some simulated species such as acetylene and methanol appear in MD but are not observed experimentally under the chosen reactor conditions, while C2 experiments near 800 K show rapid CO₂ formation that is not mirrored by short MD trajectories. The article attributes such gaps partly to timescale separation between nanosecond-scale MD and complete oxidation in the flow reactor, and partly to force-field limitations for copper–hydrocarbon oxidation chemistry. Quantitative product branching should be read from the version-of-record PDF at pdf_path, not this summary alone.

Limitations¶

ReaxFF parametrizations for Cu–H–O–C chemistry benefit from iterative refinement; readers should treat quantitative branching ratios as model-dependent. Experimental–computational agreement is qualitative unless the publication provides explicit uncertainty quantification for both sides.

Relevance to group¶

The paper extends ReaxFF validation into oxygen-carrier chemistry relevant to chemical looping, complementing gas-phase and surface pyrolysis studies elsewhere in the corpus.