From cellulose to kerogen: molecular simulation of a geological process
Summary¶
Petroleum genesis from organic matter spans Myr timescales, so brute-force MD alone cannot reach burial chemistry. Atmani et al. (Chem. Sci., DOI 10.1039/C7SC03466K) combine ReaxFF reactive MD with replica-exchange molecular dynamics (REMD) to follow cellulose → kerogen + fluids under maturation-like heating. The abstract reports crystal fragmentation, water release, unsaturated aliphatic condensation, and aromatization, with solid compositions tracking natural type III kerogen and confined artificial maturation benchmarks. After fluid expulsion, the microporous kerogen model is compared to mature type III kerogen and microporous carbon from low-temperature saccharose pyrolysis on structure, texture, density, porosity, and stiffness. Methane dominates hydrocarbon products for this type III precursor class in their narrative.
Methods¶
MD application. LAMMPS-style ReaxFF simulations allow C/H/O bond rearrangements without fixed connectivity; parameter provenance and structural benchmarks are documented in the article and ESI (papers/Atmani_ChemicalScience_2017.pdf). REMD (replica-exchange / parallel-tempering) spans temperature replicas to escape kinetic traps during cellulose breakdown and later aromatization; replica spacing, exchange statistics, and thermostat choices appear in the PDF/ESI, not the short abstract extract. Heating schedules mimic burial/catagenesis-like maturation rather than laboratory flash pyrolysis. Post-processing classifies solid and fluid products to compare density, porosity, and elastic response with natural and sacrificial-carbon references. Atom counts, PBC vectors, NVT/NPT choices, barostat targets, timestep, production duration, and electric fields are not restated in the indexed p1–2 extract—read pdf_path for executable inputs.
Force-field training: N/A — the article uses a ReaxFF description for C/H/O chemistry as referenced in the main text and ESI; it does not center on a new parameterization fit in this publication.
Findings¶
The authors claim a continuous atomistic story from cellulose to kerogen plus volatiles, with semi-quantitative agreement to type III solid composition benchmarks along the maturation path. Confined artificial maturation and sacrificial carbon references anchor comparisons of microporous kerogen density, porosity, and stiffness after fluid loss. Mechanisms emphasize fragmentation, dehydration, aliphatic growth, and aromatization, with methane as the leading hydrocarbon. Temperature ladders accessed through REMD are central to the sampling strategy. Geological time remains compressed; ReaxFF barriers are only QM-accurate within the training scope—take numerical claims from figures/ESI, not this summary.
Limitations¶
Geological time is still compressed via REMD and elevated temperatures—kinetic correspondence to basin burial history requires careful interpretation. ReaxFF organic chemistry is approximate versus QM; quantitative matches should be verified in figures/SI rather than inferred from this summary alone.
Relevance to group¶
Adri C. T. van Duin coauthored this ReaxFF + REMD demonstration that reactive MD can follow cellulose → kerogen chemistry with geoscience-facing validation targets—linking the group’s organic reactive workflows to petroleum and geomaterials audiences beyond classical oxide applications.
Citations and evidence anchors¶
- DOI:
10.1039/C7SC03466K— Chem. Sci. edge article;papers/Atmani_ChemicalScience_2017.pdf; extractnormalized/extracts/2017atmani-chemical-sci-cellulose-kerogen_p1-2.txt.