METECH 2026

Accelerating the Petrochemical Value Chain: Physics-Informed AI for Catalysis, Polymers, and Formulations

Speaker:
Michael Rauch, Director, Materials Science Product and Discovery, Schrödinger

Abstract:
As the global downstream sector pivots toward high-value chemicals and circularity, traditional R&D must transition from trial-and-error to a predictive, digital-first model. This session explores how a robust digital chemistry strategy—combining physics-based simulation and ML/AI—is reshaping the fundamental workflows of R&D.

We will present case studies and methodologies demonstrating this synergy in action:
• Next-Gen Catalyst Discovery: Utilizing atomistic simulations and ML to enable the optimization of both homogeneous and heterogeneous catalysts for higher selectivity and energy efficiency.
• Polymer Informatics at Scale: Demonstrating structure-to-property mapping to discover novel polymer architectures that meet stringent thermal and mechanical constraints, as evidenced in collaborative work with global leaders like SABIC.
• Intelligent Formulation Optimization: Applying Bayesian optimization and predictive modeling to navigate multi-parameter design spaces for fuels and additives, achieving peak performance with a fraction of the traditional experimental waste.
This presentation offers a technical deep-dive into how the integration of physics and AI enables downstream players to accelerate innovation while lowering the carbon footprint of R&D.