Preliminary understanding of initial reaction process for subbituminous coal pyrolysis with molecular dynamics simulation
Summary¶
Subbituminous coal pyrolysis releases a complex mixture of light gases and tars whose early bond-breaking chemistry sets downstream soot and liquid yields. Zhan et al. build a three-dimensional Hatcher-type subbituminous structural model and follow initial thermal decomposition with ReaxFF molecular dynamics, augmented by DFT on representative radicals to clarify homolytic versus heterolytic character where the force field is ambiguous. The Fuel article focuses on first-nanosecond chemistry accessible to reactive MD, emphasizing CO, CO\(_2\), CH\(_4\), and H\(_2\) formation routes rather than full char maturation.
Methods¶
MD application (subbituminous coal pyrolysis onset). The authors build a three-dimensional Hatcher-type subbituminous coal model (composition, atom count, and cell vectors in papers/ReaxFF_others/Preliminary understanding of initial reaction process for subbituminous coal pyrolysis with molecular dynamics simu.pdf) and run ReaxFF molecular dynamics in a PBC bulk cell. After annealing (10 NVT cycles 300–1300 K), they equilibrate with NPT near 300 K to 0.78 g cm\(^{-3}\) at 0.2 GPa (Berendsen barostat, 500 fs damping), relax 10 ps NVT at 300 K, then use a staged heating scan 300–2800 K to bracket bond-cleavage onset before 1 ns NVT production segments at selected high temperatures (velocity Verlet, 0.25 fs; Berendsen thermostat, 100 fs damping). Electric fields, replica sampling, and shock loading are N/A for this protocol.
Force-field training. N/A — a published ReaxFF parametrization for coal-like C/H/O chemistry is applied as cited, not refitted here.
Static QM (spot checks). DFT on representative radicals and fragmentation steps supports interpretation where ReaxFF ambiguity arises; full functional, basis, and k-sampling settings are in the journal Methods (N/A to tabulate on this page).
Findings¶
Initial pyrolysis favors scission of comparatively weak C–C and C–O bridges embedded in the macromolecular model. CO tracks decarbonylation from carbonyl groups, while CO\(_2\) emerges from hydrogen-transfer sequences coupled to carboxyl decarboxylation. CH\(_4\) arises when methyl-like radicals abstract H from hydroxyl sites, and H\(_2\) forms by recombination of H atoms released across functional groups. A C\(_9\)H\(_9\)O radical appears as a hub en route to cresol-type products when H-donor environments persist. The authors relate these pathways to prior experimental signatures cited in their references. Because the Hatcher model is a coarse macromolecular surrogate, absolute yield predictions should be treated as qualitative indicators of early product distribution rather than quantitative kinetic models for industrial reactors.
Limitations¶
Nanosecond trajectories capture onset chemistry but not slow cross-linking or mass-transfer limited release in reactors. ReaxFF errors on barriers should be spot-checked with QM for quantitative Arrhenius parameters.
Relevance to group¶
Provides corpus context for reactive hydrocarbon systems adjacent to van Duin-line combustion and pyrolysis applications, though authorship is external to the host group.