Machine learning enabled quantification of the hydrogen bonds inside the polyelectrolyte brush layer probed using...

Area of Science:

• Soft Matter Physics
• Computational Chemistry
• Materials Science

Background:

• Polyelectrolyte (PE) brush configuration depends on counterions and solvent molecules.
• Atomistic studies of PE brushes reveal insights into ion and water behavior.
• Generic definitions for water-water hydrogen bonds (HBs) are inadequate within PE brushes due to nanoconfinement.

Purpose of the Study:

• To address the limitations of generic HB definitions in PE brushes.
• To employ machine learning (ML) for predicting water-water HBs inside cationic PE brushes.
• To identify structural modifications of water-water HBs induced by soft nanoconfinement.

Main Methods:

• All-atom molecular dynamics (MD) simulations of a cationic PE brush.
• Unsupervised machine learning (ML) approach utilizing water molecule coordinates.
• Clustering analysis to identify and characterize water-water HBs.

Main Results:

• Structurally similar water-water HB clusters inside and outside the PE brush.
• Shorter cluster margins inside the PE brush, indicating HB disruption.
• Reduced average "hydrogen-acceptor-oxygen-donor-oxygen" angle for HBs within the brush.
• Generic HB definitions overpredict the number of HBs inside the PE brush.

Conclusions:

• ML accurately identifies structural modifications of water-water HBs in PE brushes.
• Soft nanoconfinement significantly impacts water-water HB characteristics.
• A specialized approach is necessary for accurate HB quantification within PE brushes.