BIO:
Elaheh Sedghamiz is a Senior Scientist in Schrödinger's Materials Science team, having earned her Ph.D. in computational chemistry from the University of Isfahan, Iran. Her Ph.D. research focused on molecular dynamics simulation of materials, highlighting her expertise in this area. During her three-year postdoctoral position at KIT, Karlsruhe, she developed a model to study the mechanical properties of crosslinked polymer networks obtained from 3D direct laser writing. Since joining Schrödinger in 2022, her research has centered around all-atom and coarse-grained MD simulation for different material groups.
You will be speaking on “Synergy of machine learning and molecular modeling in the design of elastomer formulations” – can you give us an insight into what your presentation will cover?
I'll be talking about how we can use a combination of machine learning (ML) and molecular dynamics (MD) simulations to design more advanced elastomer formulations. Case studies will show how we can screen thermoplastic elastomers for key properties like the glass transition temperature (Tg), highlighting the advantages of using ML and MD together. I'll also dive into the simulation of silicon-based polymers at low temperatures to demonstrate how digital tools can make the design process more efficient. Overall, the aim is to showcase how these approaches can help optimize elastomer chemistry and performance for specific applications.
With machine learning optimizing elastomer properties, from durability to flexibility, how might this impact the product development cycle in industry?
Integrating machine learning (ML) into optimizing elastomer properties like durability, flexibility, and glass transition temperature can greatly speed up the product development cycle. ML allows for rapid screening and prediction of materials behavior, reducing the need for lengthy experimental efforts. This leads to quicker identification of optimal formulations that meet performance goals. ML can also reveal relationships between chemistry and properties that may not be immediately apparent, enabling more informed decision-making processes. Overall, this approach streamlines the R&D process, lowers costs, and shortens the time to bring new elastomer products to market.
What are the benefits and challenges you foresee when incorporating these advanced simulations?
Incorporating advanced simulations offers several advantages, such as faster development by reducing the need for physical experiments and cost savings through early identification of optimal formulations. Machine learning improves predictive accuracy by uncovering complex relationships between chemistry and properties, while enabling the custom design of elastomers for specific applications. Main challenges to incorporating these approaches include high-quality data and reliable algorithms for machine learning, as well as the large-scale computer resource and automation technology for physics-based simulations. In the end, the key is to integrate these technologies into cost-effective and user-friendly workflow solutions.
This is your first time at the event, what are you most looking forward to?
I’m most excited about engaging with fellow experts and learning about the latest advancements in elastomer design and digital tools. I'm looking forward to exchanging ideas, gaining new insights from different perspectives, and discussing how innovative technologies like machine learning and molecular simulations are shaping the future of material development. It’s a great opportunity to connect with others in the field and explore potential collaborations.
