Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Intermolecular Forces03:13

Intermolecular Forces

67.5K
Atoms and molecules interact through bonds (or forces): intramolecular and intermolecular. The forces are electrostatic as they arise from interactions (attractive or repulsive) between charged species (permanent, partial, or temporary charges) and exist with varying strengths between ions, polar, nonpolar, and neutral molecules. The different types of intermolecular forces are ion–dipole, dipole–dipole, hydrogen bonds, and dispersion; among these, dipole–dipole, hydrogen...
67.5K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Interplay of structure and dynamics in solid polymer electrolytes: a molecular dynamics study of LiPF<sub>6</sub>/polypropylene carbonate.

Physical chemistry chemical physics : PCCP·2026
Same author

Length scales in electrolytes.

The Journal of chemical physics·2025
Same author

Managing Expectations and Imbalanced Training Data in Reactive Force Field Development: An Application to Water Adsorption on Alumina.

Journal of chemical theory and computation·2024
Same author

Modelling structure and ionic diffusion in a class of ionic liquid crystal-based solid electrolytes.

Physical chemistry chemical physics : PCCP·2024
Same author

Water sub-diffusion in membranes for fuel cells.

Scientific reports·2017
Same author

Acoustic excitations and elastic heterogeneities in disordered solids.

Proceedings of the National Academy of Sciences of the United States of America·2014

Related Experiment Video

Updated: Dec 8, 2025

Hydrophobic Salt-modified Nafion for Enzyme Immobilization and Stabilization
11:16

Hydrophobic Salt-modified Nafion for Enzyme Immobilization and Stabilization

Published on: July 11, 2012

16.6K

From Ionic Surfactants to Nafion through Convolutional Neural Networks.

Loïc Dumortier1, Stefano Mossa2,3

  • 1X-ray Microspectroscopy and Imaging Group, Department of Chemistry, Ghent University, Krijgslaan 281/S12, B-9000 Gent, Belgium.

The Journal of Physical Chemistry. B
|September 18, 2020
PubMed
Summary

Deep convolutional neural networks analyze soft matter data, identifying nanostructures in ionic surfactants. This machine learning approach aids in understanding polymer ionomer properties without prior models.

More Related Videos

Merging Ion Concentration Polarization between Juxtaposed Ion Exchange Membranes to Block the Propagation of the Polarization Zone
08:06

Merging Ion Concentration Polarization between Juxtaposed Ion Exchange Membranes to Block the Propagation of the Polarization Zone

Published on: February 23, 2017

8.8K
Extraction and Characterization of Surfactants from Atmospheric Aerosols
09:34

Extraction and Characterization of Surfactants from Atmospheric Aerosols

Published on: April 21, 2017

17.2K

Related Experiment Videos

Last Updated: Dec 8, 2025

Hydrophobic Salt-modified Nafion for Enzyme Immobilization and Stabilization
11:16

Hydrophobic Salt-modified Nafion for Enzyme Immobilization and Stabilization

Published on: July 11, 2012

16.6K
Merging Ion Concentration Polarization between Juxtaposed Ion Exchange Membranes to Block the Propagation of the Polarization Zone
08:06

Merging Ion Concentration Polarization between Juxtaposed Ion Exchange Membranes to Block the Propagation of the Polarization Zone

Published on: February 23, 2017

8.8K
Extraction and Characterization of Surfactants from Atmospheric Aerosols
09:34

Extraction and Characterization of Surfactants from Atmospheric Aerosols

Published on: April 21, 2017

17.2K

Area of Science:

  • Materials Science
  • Computational Chemistry
  • Machine Learning

Background:

  • Soft matter systems, like hydrated ionic surfactants, exhibit complex structural and phase behaviors.
  • Understanding the relationship between hydration levels and nanostructure is crucial for materials science applications.
  • Traditional methods for analyzing such complex systems can be limited by model assumptions.

Purpose of the Study:

  • To apply deep convolutional neural networks (CNNs) to analyze three-dimensional soft matter data from molecular dynamics simulations.
  • To develop a machine learning classifier for automatically detecting water content and associating it with specific nanostructures in ionic surfactants.
  • To investigate the transferability of the trained CNN to related materials, such as the polymer ionomer Nafion, and quantify structural similarities.

Main Methods:

  • Utilized deep convolutional neural networks (CNNs) for analyzing three-dimensional voxel data of soft matter systems.
  • Trained a CNN classifier to identify hydration levels and corresponding nanostructures in coarse-grained models of ionic surfactants.
  • Employed transfer learning by applying the trained CNN to Nafion data to assess configuration similarity.

Main Results:

  • Successfully trained a CNN classifier to accurately detect water quantity and identify representative nanostructures at different hydration levels.
  • Demonstrated the transferability of the CNN to Nafion, enabling a quantitative measure of structural similarity between surfactants and the polymer.
  • Showed that the static structure factor of Nafion can be represented as a superposition of surfactant structure factors at various hydration levels.

Conclusions:

  • Machine learning, specifically CNNs, offers a powerful, data-driven approach for analyzing the multiscale structure of disordered materials.
  • The proposed method provides an agnostic and precise description of material structure without relying on a priori models.
  • This approach has the potential to advance the understanding and design of complex soft matter and polymer systems.