AI in CHEM Seminar Series: Lab Automation for Scattering Analysis with Andy Anker

This talk is part of the “AI and Machine Learning in Chemical Research and Industry” Seminar Series organized by the Aalto University School of Chemical Engineering. It is open to all members of the public. Registered students in course CHEM-E4190 can also obtain 1cr by attending the seminars and completing the assignments.

Date and location

• Wednesday 15 January 2025 @ 13:15-14:00
• A304 Ke2 lecture hall in the main building of the School of Chemical Engineering, Kemistintie 1, 02150 Espoo.

Agenda

• 13:00-13:15. Setup and brief info for the registered students.
• 13:15-14:00. Seminar by Andy Anker, lecture hall A304.
• 14:00-onwards. Coffee, netwoking and mingling in the lobby adjacent to the lecture hall.

Seminar info

Machine learning experimental scattering data analysis: concept, practice, and a future with automated laboratories

Andy S. Anker^1,2

1. Department of Energy, Danish Technical University, Denmark, ansoan@dtu.dk
2. Department of Chemistry, University of Oxford, United Kingdom, andy.anker@chem.ox.ac.uk

The rapid growth of materials chemistry data has outpaced conventional data analysis and modelling methods, which can require enormous manual effort. To effectively analyse this wealth of information, we are using machine learning (ML) models trained on extensive datasets of physics-based simulations for analysis of experimental scattering data [1,2]. Yet, relying solely on a single experimental technique often fails to provide sufficient information for resolving complex material structures. To overcome these limitations, we are integrating diverse datasets into unified ML pipelines. Building on these methodological advances, we look towards developing automated laboratories capable of accelerating materials synthesis.

References

1. Andy S. Anker, Keith T. Butler, Raghavendra Selvan, Kirsten M. Ø. Jensen, Chemical Science 2024, 48, 14003–14019.
2. Emil T. S. Kjær, Andy S. Anker, Marcus N. Weng, Simon J. L. Billinge, Raghavendra Selvan, Kirsten M. Ø. Jensen, Digital Discovery 2023, 1, 69–80.

About the speaker

See Andy’s Github page for more info

I have recently been awarded a 4 000 000 DKK (~ £500 000) postdoctoral grant to pursue an academic career in the interface of materials chemistry, machine learning and robotics. Here, I am building a self-driving laboratory for controlled synthesis of inorganic nanomaterials in collaboration with Prof. Tejs Vegge and the CAPeX center at Technical University of Denmark, Assoc. Prof. Volker Deringer’s group at Oxford University and Prof. Kasper Støy’s group at the IT University of Copenhagen. In 2024+2025, I am physically working from Oxford.

I obtained my PhD in materials chemistry from the Nanostructure Group UPCH, University of Copenhagen, supervised by Assoc. Prof. Kirsten Marie Ørnsbjerg Jensen, where my main interest was to study nanoparticles and structures in solution with Total X-ray Scattering with Pair Distribution Function (PDF) and Small-Angle X-ray Scattering (SAXS). I applied advanced computer modelling, in Python, to combine information of both the local order from PDF and the particle order from SAXS, which overcome problems that the methods cannot overcome individually. During my career, the research focus has converged towards developing machine learning (ML) methods to analyse chemical data; especially PDF & SAXS, after I met Assistant Professor Raghavendra Selvan who I had collaborated with since 2019. I have furthermore spent 6 months during my PhD working at Rutherford Appleton Laboratory with Senior Lecturer Keith Tobias Butler and the Scientific Machine Learning Group to develop an general approach to match simulated and experimental data in materials chemistry. During the last period of my PhD, I have especially focused on using generative models to analyse scattering-, and spectroscopy data.