Skip to content

Large-Scale Reactive Molecular Dynamics Simulation and Kinetic Modeling of High-Temperature Pyrolysis of the Gloeocapsomorphaprisca Microfossils

Summary

ReaxFF reactive MD on multi-thousand-atom models of Guttenberg and Kukersite microfossil fragments quantifies pyrolysis products, radical initiation sites, and lumped kinetics, comparing isolated fragments with multi-component mixtures. The J. Phys. Chem. B study targets kerogen-like macromolecular motifs relevant to oil shale retorting and thermal maturation proxies drawn from microfossil chemistry literature.

Methods

Force-field training. N/A — this application paper employs the established C/H/O ReaxFF parameterization cited as reference 12 in the article rather than refitting the potential within this study.

MD application (LAMMPS ReaxFF RMD). Fragment geometries are pre-optimized with B3LYP/6-31G in Gaussian 09, then solvated into periodic supercells with PACKMOL at the compositions and mass densities listed in Table 1 (including duplicated microfossil fragments up to order 10⁴–10⁵ atoms per abstract). Equilibration follows an NVT minimization at 2 K for 1.5 ps, a 2 K → 300 K heat ramp over 25 ps, a short 300 K hold, compression toward ~1.38 g cm⁻³, and NPT relaxation at 1 bar and 300 K before launching high-temperature NVT pyrolysis trajectories (1800–2500 K for Guttenberg fragments A/B, 1500–2200 K for fragment D and mixture cases, with additional 0.6–1.38 g cm⁻³ sets tabulated in Table 1). Reported reactive production segments span 250 ps per the manuscript. Thermostat/barostat brands, output cadence for species tracking, and the numeric integration step are specified in papers/Zou_Raman_JPCB_2014_Pyrolysis.pdf; N/A — the integration timestep is not recovered reliably from the PDF text layer in this workspace and should be read directly from the Methods/SI tables.

Static QM / DFT-only. B3LYP/6-31G relaxations in Gaussian 09 supply gas-phase fragment geometries prior to packing; the reactive production trajectories themselves are ReaxFF-based rather than ab initio MD.

Replica sampling, electric fields, and pressure during pyrolysis. No replica-exchange or electric-bias protocols are reported. After the 300 K, 1 bar NPT relaxation used to set density, high-temperature chemistry is advanced in constant-volume NVT windows at the pyrolysis temperatures listed above.

Findings

Radical inventories and carbon-number distributions highlight preferred bond-scission initiation motifs (benzylic cleavage in fragment A; phenoxy-ether cleavage in fragments B/D) consistent with the bond-strength arguments developed in the paper. Guttenberg mixture simulations show radical cross-talk: radicals generated on one fragment accelerate chemistry on another—especially component A—relative to isolated fragment runs. A lumped kinetic model fit to the RMD trajectories yields an Arrhenius barrier the authors describe as reasonable for overall cracking, while noting that running at T > 1500 K within ~1 ns windows biases product distributions toward entropically favored routes such as ethylene, so low-temperature selectivity predictions require caution or accelerated-sampling follow-on (as stated in the abstract).

Limitations

High temperature MD windows and short trajectories limit direct extrapolation to industrial retort timescales; lumped kinetics are illustrative rather than drop-in process models without calibration.

Relevance to group

Adri C. T. van Duin coauthors; ReaxFF pyrolysis of complex hydrocarbon feedstocks complements other organics/combustion entries in the corpus.

Citations and evidence anchors