Conference

Global Polymer Summit 2025

CalendarDate & Time
  • September 8th-11th, 2025
LocationLocation
  • Cleveland, Ohio

Schrödinger is excited to be participating in the Global Polymer Summit 2025 conference taking place on September 8th – 11th in Cleveland, Ohio. Join us for a presentation by Croix Laconsay, Senior Scientist I at Schrödinger, titled “Automated Discovery of Polymer Elementary Reaction Networks with the Nanoreactor.” Stop by booth 1314 to speak with Schrödinger scientists.

icon time SEPT 9 | 4:00PM
Automated Discovery of Polymer Elementary Reaction Networks with the Nanoreactor

Speaker:
Croix Laconsay, Senior Scientist I, Schrödinger

Abstract:
In molecular modeling of polymer materials, autonomous reaction network exploration algorithms offer a systematic framework for uncovering genuine mechanisms of chemical reactions. Understanding elementary reaction networks of filled polymer composites in tires, for example, can be of great importance in optimizing these materials. At the core of these networks are elementary reaction steps, which could serve as the fundamental building blocks in this endeavor. Various automated methods have been proposed to discover elementary reaction steps. Graph-based approaches efficiently generate elementary reaction pathways but suMer from exponential computational scaling as system size increases, often producing many reactions irrelevant to the conditions of interest. Alternatively, first-principles enhanced-sampling molecular dynamics can capture relevant reactions but demands thousands to millions of energy and force evaluations, making largescale applications computationally prohibitive. Inspired by the work of Jensen, we introduce a fully automated approach for Elementary Reaction Network exploration. Nanoreactor utilizes GFN2-xTB-based metadynamics within a confined reaction sphere to efficiently sample elementary reaction steps. Subsequent AutoTS computations—an automated transition-state search workflow—further refines the process through automated transition state optimization. While conceptually like the method of Jensen, our approach differs fundamentally at the algorithmic level, offering a unique strategy for efficient elementary reaction network generation and subsequent iterative growths towards large reaction networks. Nanoreactor accelerates the chemical network exploration of molecular degradation mechanisms. In this talk, I will demonstrate use cases that showcase the value of the Nanoreactor in polymer and carbon black materials development.