Prediction of Molecular Structure of Asphaltenes in Heavy Oils

Asphaltene is a crucial component of petroleum. The molecular structure is difficult to determine, posing challenges for its analysis and application. In this study, a program named Prediction of Molecular Structures of Asphaltenes (PMSA) was developed to predict asphaltene structures with its validity rigorously verified through molecular dynamics simulations. Using the modified Brown-Ladner method, key structural parameters (with corresponding probabilities) were first derived from basic experimental data, including molecular weight, elemental composition, and ¹H NMR spectra. Based on these parameters, 5000 candidate asphaltene structures were constructed by integrating Monte Carlo sampling with a pre-established library of aromatic building blocks. Nonlinear optimization with a least-squares objective function was then employed to screen out 4-6 dominant molecular structures and determine their respective molar fractions in the sample. To validate the predicted structures, asphaltene-toluene systems were established, and analyses of radial distribution functions (RDF) and dihedral angle distributions between aromatic planes were conducted. The results showed that PMSA-generated molecules can form stable π-π stacked aggregates, exhibiting an aggregation free energy of approximately 4-6 kcal/mol, which aligns well with the characteristic structural features of asphaltenes and provides robust theoretical support for the reliability and applicability of the PMSA program.