Mesoscale Simulation and Machine Learning of Asphaltene Aggregation Phase Behavior and Molecular Assembly Landscapes

Area of Science:

• Petroleum Chemistry
• Materials Science
• Computational Chemistry

Background:

• Asphaltenes, the heaviest aromatic fraction in crude oil, cause significant economic losses in the petroleum industry due to aggregation and precipitation.
• The Yen-Mullins model describes asphaltene aggregation, but microscopic details of cluster morphology and stability under varying conditions are poorly understood.

Purpose of the Study:

• To establish a phase diagram for asphaltene self-assembly based on temperature, pressure, and solvent composition.
• To understand how external processing conditions influence asphaltene aggregation.
• To determine the low-dimensional free energy surfaces governing asphaltene self-assembly.

Main Methods:

• Coarse-grained molecular dynamics simulations of a prototypical asphaltene molecule.
• Graph matching and nonlinear manifold learning, including a variant of diffusion maps for data with large local density variations.
• Application of many-body diffusion maps to molecular self-assembly.

Main Results:

• A phase diagram was established, mapping self-assembled morphologies against temperature, pressure, and n-heptane:toluene solvent ratio.
• Increasing pressure minimally affects the free energy landscape, primarily destabilizing larger aggregates.
• Elevated temperature and higher toluene solvent fractions stabilize smaller clusters and looser bonding arrangements.

Conclusions:

• Toluene acts as an effective temperature in asphaltene self-assembly, stabilizing smaller clusters and looser arrangements.
• Understanding these self-assembly dynamics allows for better control over asphaltene aggregation during petroleum processing.
• The study provides insights into controlling external processing conditions to mitigate asphaltene-related inefficiencies.