Meeting Free

Machine Learning for Carbon Capture & Battery Discovery

4pm – 5pm 11 June 2026
Elastomer networking banner

Machine learning is transforming the discovery of materials critical to the net-zero transition. This webinar explores how artificial intelligence is accelerating the development of carbon capture technologies and next-generation batteries, enabling researchers to identify promising materials, optimise performance and reduce development times for sustainable energy and decarbonisation solutions.

 

Key takeaways:

  1. Understand how machine learning is being applied to accelerate the discovery and optimisation of materials for carbon capture and battery technologies
  2. Explore real-world examples of AI-driven materials research and the opportunities it presents for sustainable innovation
  3. Gain insight into the challenges, limitations and future potential of machine learning in supporting the transition to net zero

 

Agenda

Timing Session Speaker
4-4.05pm

Welcome and introduction

Dr Sergio Gonzalez Sanchez
4.05-4.20pm Small features, big impact: Designing at the microscale with generative AI Dr Sam Cooper, Imperial College London
4.20-4.25pm Q&A  
4.25-4.40pm Auto-assembly of knowledge for battery discovery using AI Prof Jacqueline Cole, University of Cambridge
4.40-4.45pm Q&A  
4.45-5pm Advancing reticular chemistry with generative AI

Dr Nakul Rampal, UC Berkeley
Dr Nakul Rampal is a postdoctoral associate in Prof Omar Yaghi's lab (Nobel Prize in Chemistry 2025)
5-5.05pm Q&A and closing  
     

   

 

You will hear from:

Dr Sergio Gonzalez Sanchez

Dr Sergio Gonzalez Sanchez

Distinguished Researcher Beatriz Galindo Senior Professor, Department of Materials Science and Engineering, Universidad Carlos III de Madrid (UC3M)

Find out more

Dr Sam Cooper

Dr Sam Cooper

Associate Professor in Artificial Intelligence for Materials Design, Imperial College London

Find out more

Prof Jacqueline Cole

Prof Jacqueline Cole

Head of Molecular Engineering, University of Cambridge

Dr Nakul Rampal

Dr Nakul Rampal

BIDMaP Fellow, Bakar Institute of Digital Materials for the Planet University of California, Berkeley

Find out more

 Presenters & Abstracts:

Dr Sam Cooper - Small Features, Big Impact: Designing at the Microscale with Generative AI

Dr Sam Cooper

Dr Sam Cooper

Associate Professor in Artificial Intelligence for Materials Design, Imperial College London

Find out more

The microscale features of porous battery electrodes strongly influence how fast they charge and how long they last. These features are governed by manufacturing parameters such as temperature and pressure, yet the relationship between processing and the resulting microstructure remains extremely difficult to predict using physics-based simulations. In this talk, I will demonstrate how generative AI tools, developed by my team at Imperial College London, have unlocked powerful new capabilities for microstructural characterisation and design, such as the generation of 3D data from a single 2D image. I will also share our work on a new workflow for the characterisation and simulation of graded electrodes, which are common in commercial cells. Finally, I will present my teams most recent work on vision and language foundation models, in particular exploring who they conceptualise scientific concepts and how they can be used in agentic workflows.

Prof Jacqueline Cole - Auto-Assembly of Knowledge for Battery Discovery Using AI

Prof Jacqueline Cole

Prof Jacqueline Cole

Head of Molecular Engineering, University of Cambridge

Artificial intelligence (AI) is developing at a revolutionary pace. There is great interest in deploying AI in science for discovering new materials that are tailored for a given device application. The use of AI to identify a new material that can be incorporated into a battery device to make it perform better is one such example. However, AI algorithms are useless without data to train them. This webinar will showcase our chemistry-aware text-mining tool, ChemDataExtractor, and its use in auto-curating large datasets about materials and their properties; and how the resulting dataset can be used to train large language models that are tailored for a given domain of materials science. I will illustrate the application of ChemDataExtractor to auto-generate a materials database and domain-specific language model about battery devices. I will close by showing how the database and model can be applied to advance battery technology.

Dr Nakul Rampal - Advancing Reticular Chemistry With Generative AI

Dr Nakul Rampal

Dr Nakul Rampal

BIDMaP Fellow, Bakar Institute of Digital Materials for the Planet University of California, Berkeley

Find out more

Generative AI tools, especially LLMs, are being used to accelerate research in reticular chemistry. This talk presents recent work on literature summarisation, synthesis condition extraction, and materials discovery for applications such as carbon capture and water harvesting. It also introduces RetChemQA, a dataset of more than 95,000 Q&A pairs for chemistry-specific LLM tasks, and QAutoEval, an automated agent for evalutating Q&A pairs and synthesis data. Finally, it will cover how robotics and automation are integrated in this workflow to enable LLM-guided experimental planning, synthesis, and feedback. Together, these tools form a closed-loop system for training, evaluating, and deploying LLMs in reticular chemistry. 
 

 

 

In Case You Missed It:

 

Return to listing