Combinatorial Polymer Design: From Flow-Driven Structure to Structure-Driven Flow
ORAL · Invited
Abstract
Polymers are essential to a wide range of technologies, yet designing them with targeted structural and functional properties remains a grand challenge. A major opportunity lies in applying machine learning to help navigate the vast combinatorial design space—spanning sequence, composition, architecture, morphology, processing, and more—to discover new formulations or replace existing ones with more sustainable alternatives. However, this complexity, combined with data scarcity and characterization challenges, limits the effectiveness of purely rational design and/or high-throughput screening
In this talk, I will describe data-driven strategies that integrate molecular simulation, machine learning, and polymer physics to accelerate combinatorial polymer design and extract mechanistic insight. I will specifically feature two recent efforts that additionally contend with the complexity of non-equilibrium characterization of polymers in solution. In the first part, I will discuss the use of polymer chain patterning, as well as the imposition of external shear flow, as a means to design single-chain nanoparticles with specific morphological characteristics. This vignette will illustrate how unsupervised machine learning and data-driven analysis enables extraction of physical design principles. In the second part, I will invert the perspective and instead consider how polymer topology and chemistry can be used to control shear flow. This vignette will be predicated on the use of generative machine learning and Bayesian optimization to identify polymer additives that yield prescribed shear-rate-dependent viscosity profiles. In addition, the results will demonstrate a nice synergy, in which physical principles are used to rectify deficiencies in data-driven design while surrogate-model analysis also helps inform the physical picture of underlying phenomena. The collection of results will highlight how simulation-guided polymer discovery and analysis can be enhanced by machine learning.
In this talk, I will describe data-driven strategies that integrate molecular simulation, machine learning, and polymer physics to accelerate combinatorial polymer design and extract mechanistic insight. I will specifically feature two recent efforts that additionally contend with the complexity of non-equilibrium characterization of polymers in solution. In the first part, I will discuss the use of polymer chain patterning, as well as the imposition of external shear flow, as a means to design single-chain nanoparticles with specific morphological characteristics. This vignette will illustrate how unsupervised machine learning and data-driven analysis enables extraction of physical design principles. In the second part, I will invert the perspective and instead consider how polymer topology and chemistry can be used to control shear flow. This vignette will be predicated on the use of generative machine learning and Bayesian optimization to identify polymer additives that yield prescribed shear-rate-dependent viscosity profiles. In addition, the results will demonstrate a nice synergy, in which physical principles are used to rectify deficiencies in data-driven design while surrogate-model analysis also helps inform the physical picture of underlying phenomena. The collection of results will highlight how simulation-guided polymer discovery and analysis can be enhanced by machine learning.
*The investigation of controlled structure formation in polymers is supported by the National Science Foundation under Grant No. 2237470. The investigation of polymer microarchitecture on solution rheology was supported by ACS Petroleum Research Fund under Doctoral New Investigator Grant 66706-DNI7. Simulations and analyses were performed using resources purchased with support from the National Science Foundation (Grant No. NSF-MRI: OAC-2320649).
–
Publication: https://doi.org/10.1039/D5SM00729A
https://doi.org/10.26434/chemrxiv-2025-td5v6
Presenters
-
Michael A. Webb
- Princeton University