Developing Machine Learning Models for Excited States Properties in Molecular Organic Crystals

Project ID: 2531bd1634

(You will need this ID for your application)

Research Theme: Physical Sciences

Research Area(s): Chemistry
Physics
Materials

UCL Lead department: Chemistry

Department Website

Lead Supervisor: Rachel Crespo Otero

Project Summary:

Why this research is important

Light-emitting organic materials are central to technologies such as displays, data storage, sensors, and energy devices. Understanding their excited states is essential for designing better-performing materials. However, modelling these processes in molecular crystals is challenging with standard computational methods. This project will develop machine learning approaches to overcome these limitations, enabling the design of more efficient materials with applications in environmental, medical, and energy-related fields.

Who you will be working with

You will join the Crespo-Otero group at UCL Chemistry, specialising in excited states, nonadiabatic dynamics, and molecular crystals, and will be part of a team developing the fromage platform, an open-source toolkit for modelling molecular crystals created within the group. Prof. Martijn Zwijnenburg, the secondary supervisor, has expertise in the optical properties and photochemistry of materials, including conjugated polymers, self assembled materials, and nanoparticles, as well as the application of ML to materials discovery.

What you will be doing

You will build and apply machine learning models to predict excited-state properties in organic molecular crystals. These models will be trained using high-level quantum chemical reference calculations, including both molecular and periodic boundary condition methods. Your work will focus on weakly bonded organic crystals and porous materials such as MOFs and COFs, integrating your models into the fromage platform.

Who we are looking for

We are seeking a motivated student with a strong interest in theoretical and computational chemistry. A background in physical chemistry, computational chemistry, solid-state physics or materials science is desirable. Experience in programming or machine learning is an advantage. Most importantly, we are looking for curiosity, enthusiasm, and a willingness to develop new skills at the interface of chemistry, physics, and data science.