BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:SimPoly: Simulation of Polymers with Machine Learning Force Fields
  Derived from First Principles + the Challenge of Long-Range Interactions 
 (arXiv:2510.13696) - Dr Lixin Sun and Dr Gregor Simm\, Microsoft
DTSTART:20260304T143000Z
DTEND:20260304T153000Z
UID:TALK241648@talks.cam.ac.uk
CONTACT:Lisa Masters
DESCRIPTION:Polymers are a versatile class of materials with widespread in
 dustrial applications. Advanced computational tools could revolutionize th
 eir design\, but their complex\, multi-scale nature  poses significant mod
 eling challenges. Conventional force fields often lack the accuracy and tr
 ansferability required to capture the intricate interactions governing pol
 ymer behavior.  Conversely\, quantum-chemical methods are computationally 
 prohibitive for the large systems and long timescales required to simulate
  relevant polymer phenomena. Here\, we overcome these limitations with a m
 achine learning force field (MLFF) approach. We demonstrate that macroscop
 ic properties for a broad range of polymers can be predicted ab initio\, w
 ithout fitting to experimental data. Specifically\, we develop a fast and 
 scalable MLFF to accurately predict polymer densities\, outperforming esta
 blished classical force fields. Our MLFF also captures  second-order phase
  transitions\, enabling the prediction of glass transition temperatures. T
 o accelerate progress in this domain\, we introduce a benchmark of experim
 ental bulk properties  for 130 polymers and an accompanying quantum-chemic
 al dataset. In this talk\, we also discuss the persistent of learning long
 -range non-covalent interactions  in MLFF\, best practices\, and potential
  future development.
LOCATION:Unilever Lecture Theatre\, Yusuf Hamied Department of Chemistry
END:VEVENT
END:VCALENDAR