Quantifying Pore Width Effects on Diffusivity via a Novel 3D Stochastic Approach with Input from Atomistic Molecular...

Area of Science:

• Computational materials science
• Chemical engineering
• Petroleum engineering

Background:

• Understanding fluid transport in nanopores (1-50 nm) is crucial for unconventional hydrocarbon production and materials science.
• Current methods struggle to quantitatively predict diffusion coefficients as a function of pore width in this critical range.
• Such pores are prevalent in shale formations and engineered materials like catalysts.

Purpose of the Study:

• To develop a computationally efficient methodology for predicting fluid diffusion coefficients in narrow pores.
• To enable accurate predictions across various pore widths and material compositions (silica, alumina, etc.).
• To create a digital library of gas diffusivity based on pore chemistry and width.

Main Methods:

• Atomistic molecular dynamics (MD) simulations to analyze fluid structure and transport in 5 nm pores.
• Kinetic Monte Carlo (KMC) modeling, using MD data as input, to predict fluid transport in mesopores.
• Extrapolation of KMC results to wider pores and validation against MD simulations.

Main Results:

• The combined MD-KMC methodology accurately predicts diffusion coefficients in slit-shaped pores.
• Results for supercritical methane show within 10% agreement with atomistic MD simulations.
• Significant computational time savings are achieved compared to pure atomistic simulations.

Conclusions:

• The novel MD-KMC approach offers an efficient and accurate way to predict fluid diffusivity in nanoporous materials.
• This methodology is applicable to diverse materials and relevant for hydrocarbon transport in shale and catalyst optimization.
• The generated digital library provides valuable data for engineering applications involving fluid flow in confined spaces.