Abstract
In recent years, the synthesis of monomer sequence-defined polymers has expanded into broad-spectrum applications in biomedical, chemical, and materials science fields. Pursuing the characterization and inverse design of these polymer systems requires our fundamental understanding not only at the individual monomer level, but also considering the chain scales, such as polymer configuration, self-assembly, and phase separation. However, our accessibility to this field is still rudimentary due to the limitations of traditional design approaches, the complexity of chemical space along with the burdened cost and time issues that prevent us from unveiling the underlying monomer sequence-structure-property relationships. Fortunately, thanks to the recent advancements in molecular dynamics simulations and machine learning (ML) algorithms, the bottlenecks in the tasks of establishing the structure-function correlation of the polymer chains can be overcome. In this review, we will discuss the applications of the integration between ML techniques and coarse-grained molecular dynamics (CGMD) simulations to solve the current issues in polymer science at the chain level. In particular, we focus on the case studies in three important topics—polymeric configuration characterization, feed-forward property prediction, and inverse design—in which CGMD simulations are leveraged to generate training datasets to develop ML-based surrogate models for specific polymer systems and designs. By doing so, this computational hybridization allows us to well establish the monomer sequence-functional behavior relationship of the polymers as well as guide us toward the best polymer chain candidates for the inverse design in undiscovered chemical space with reasonable computational cost and time. Even though there are still limitations and challenges ahead in this field, we finally conclude that this CGMD/ML integration is very promising, not only in the attempt of bridging the monomeric and macroscopic characterizations of polymer materials, but also enabling further tailored designs for sequence-specific polymers with superior properties in many practical applications.
Highlights
Polymer materials, a class of natural or synthetic substances composed of long-chain molecules, are prevalent, ranging from proteins, cellulose, nucleic acids in a living organism to familiar man-made materials such as concrete, glass, paper, plastics, and rubbers (Council, 1994; Brinson and Catherine Brinson, 2008; Sawyer et al, 2008; Namazi, 2017)
Classical density functional theory (DFT) shares a number of similarities with the polymer self-consistent-field theory (SCFT) and can reproduce the morphologies of block copolymer thin films predicted by SCFT as well as resolve the structural properties near the interface, but needs more development for complex systems with multidimensional density profiles (Frischknecht et al, 2002)
We are going to discuss the most recent studies of machine learning (ML) applications in polymer chain characterization and inverse design by answering the following four key questions: 1) What is the bottleneck in polymer chain characterization or inverse design? 2) What is the ML strategy? 3) How can ML solve the challenging problem? and 4) How can we leverage the model in future applications?
Summary
A class of natural or synthetic substances composed of long-chain molecules, are prevalent, ranging from proteins, cellulose, nucleic acids in a living organism to familiar man-made materials such as concrete, glass, paper, plastics, and rubbers (Council, 1994; Brinson and Catherine Brinson, 2008; Sawyer et al, 2008; Namazi, 2017). Classical DFT shares a number of similarities with the polymer SCFT and can reproduce the morphologies of block copolymer thin films predicted by SCFT as well as resolve the structural properties near the interface, but needs more development for complex systems with multidimensional density profiles (Frischknecht et al, 2002) Another approach is the DMFT which proved to be effective to simulate processes with length and time scales for the polymer systems currently inaccessible by MD simulations (Knoll et al, 2004), but inefficient for nonlocal coupling effect due to the huge computational expense to acquire the chemical potential (Zhang et al, 2017).
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have