Molecular dynamics simulations of asphaltene aggregation in heavy oil system for the application to solvent deasphalting

This study evaluated the application of molecular dynamics (MD) simulations to solvent deasphalting (SDA) by comparing the extent of asphaltene aggregation with the experimental results under near-critical operating conditions. In the MD simulations, a large-scale system containing 288 oil molecules (12 oil molecular models) was established to mimic a real heavy oil system, with n -pentane adopted as the extraction solvent. The same operating conditions were employed for both MD simulations using an all-atomic force field and SDA extraction experiments. Several properties, including density, asphaltene aggregation extent, molecule type in aggregates, quality of separated oil phases, radial distribution function, and dimer interaction energy, were analyzed using the MD trajectories. The extent of asphaltene aggregation and the quality of the two oil phases (solvent-rich and asphaltene-rich) simulated using MD under several operating conditions were qualitatively consistent with the experimental SDA results. The calculation of radial distribution functions and dimer interaction energies of asphaltenes revealed that the size and shape of the cyclic ring sheet and steric hindrance determined the aggregation tendency. Interestingly, each type of asphaltene favorably aggregated with the same type over the other types, in accordance with the rationale of like attracts like . These findings are promising for the practical applications of MD simulations for developing SDA processes.

Automated summary

This study evaluated the application of molecular dynamics (MD) simulations to solvent deasphalting (SDA) by comparing the extent of asphaltene aggregation with experimental results under near-critical operating conditions. The results showed that MD simulations could accurately predict asphaltene aggregation and separated oil phases, demonstrating promising practical applications in developing SDA processes.