Deep learning-based quasi-continuum theory for structure of confined fluids

Area of Science:

• Computational physics and chemistry
• Materials science
• Nanotechnology

Background:

• Accurate prediction of fluid behavior in nanoscale confined systems is crucial for applications like energy storage and biomolecular systems.
• Classical continuum theories struggle to precisely model the interfacial structure of confined fluids.
• Understanding nanoscale fluid dynamics is key to advancing various scientific and technological fields.

Purpose of the Study:

• To develop a novel deep learning-based quasi-continuum theory (DL-QT) for predicting structural properties of confined fluids.
• To accurately model the concentration and potential profiles of Lennard-Jones (LJ) fluids and water in nanochannels.
• To bridge the gap between molecular-level physics and continuum theory using artificial intelligence.

Main Methods:

• Development of a deep learning model using a convolutional encoder-decoder network (CED) for high-dimensional surrogate modeling.
• Integration of the CED model with interatomic potential-based continuum theory.
• Application of the DL-QT model to simulate LJ fluids and water confined in nanochannels of varying widths and thermodynamic states.

Main Results:

• The DL-QT model demonstrates robust predictive performance for confined LJ fluids across diverse thermodynamic conditions.
• Accurate prediction of concentration profiles for water confined within nanochannels of different dimensions was achieved.
• The study validates the effectiveness of the DL-QT approach in capturing nanoscale fluid behavior.

Conclusions:

• The developed DL-QT model successfully predicts the structural properties of confined fluids, overcoming limitations of classical theories.
• This deep learning approach effectively integrates molecular physics with continuum theory for nanoscale simulations.
• The DL-QT model offers a powerful tool for research in energy storage, biomolecular systems, and other fields involving confined fluids.