2025 AIChE Annual Meeting
(33f) Data-Driven Design of Polymer Chemistry and Architecture for Tuning Rheological Properties
In this talk, I will present our recent work on developing data-driven methodologies to design soluble polymers as additives that tune rheological responses. Our tiered workflow begins with low-cost simulations that capture hydrodynamic interactions to efficiently probe polymer dynamics under shear. These simulations guide generative machine learning models [5] that propose candidate microarchitectures, which are iteratively refined through active learning to balance chemical exploration with fidelity to rheological targets. I will discuss key insights derived from these campaigns, highlighting design principles that lead to specific shear-thinning behaviors and exploring implications of synthetic constraints. By integrating advanced modeling and data-driven optimization with physical insight, this work provides a foundation to pursue development of sustainable, high-performance water-soluble polymers tailored to both industrial and emerging applications.
[1] Martini, A., Ramasamy, U. S. & Len, M. Review of Viscosity Modifier Lubricant Additives. Tribol. Lett. 66, 58 (2018).
[2] Gbadamosi, A. et al. Application of Polymers for Chemical Enhanced Oil Recovery: A Review. Polymers 14, 1433 (2022).
[3] Kästner, U. The impact of rheological modifiers on water-borne coatings. Colloids Surf. A: Physicochem. Eng. Asp. 183, 805–821 (2001).
[4] Dickinson, E. Hydrocolloids at interfaces and the influence on the properties of dispersed systems. Food Hydrocoll. 17, 25–39 (2003).
[5] Jiang, S., Dieng, A. B. & Webb, M. A. Property-guided generation of complex polymer topologies using variational autoencoders. npj Comput. Mater. 10, 139 (2024).