Accessible and automated computational catalyst discovery and reactivity optimization

Designing high-performing homogeneous catalysts and optimizing non-catalytic transformations often relies on trial-and-error iteration. On the other hand, computational tools often remain inaccessible to those without specialized expertise.

In this webinar, we will demonstrate how an end-user physics–AI platform removes barriers to entry, making this process accessible to both experts and non-experts while enabling seamless scalability. By integrating quantum chemistry, fast semiempirical tight-binding methods, and machine learning interatomic potentials, the platform enables automated exploration of reaction networks and simultaneous optimization of multiple performance metrics. Using real-world catalytic and non-catalytic examples, we will show how mechanistic insight can be translated into actionable design decisions—quickly, accurately, and without coding.

Key highlights:

• Lower the barrier to catalyst design
  Discover how a Physics–AI end-user platform enables both experts and non-experts to perform advanced in silico catalyst design and reactions optimization
• Multi-level modeling at scale
  Leverage quantum chemistry, semiempirical tight-binding methods, and machine learning interatomic potentials (MLIPs) within a single, scalable framework
• No coding required
  Intuitive, GUI-driven environment designed for accessibility and productivity

Who should attend: Anyone interested in non-catalytic and catalytic reactivity optimization, or homogeneous catalyst design