Details

We invite applications for a four year EPSRC ICASE to work under the supervision of Prof Joan Cordiner at the University of Sheffield in collaboration with our industry partner, Syngenta.

Help feed the world! With a growing population, climate change and water scarcity it’s important that we can design better formulations for agrochemicals.

Being able to predict what formulation will be stable will speed up formulation and product design allowing us to improve efficacy, efficiency, sustainability and safety.

The proposal is to model mixtures that can form stable Emulsion Concentrate (EC) formulations using machine learning from a dataset produced by Syngenta using Artemus (Robot). The ability to model this complex multi-dimensional problem cannot be currently completely solely using mechanistic models alone.

This proposal would take our machine learning methodologies, expertise and state of the art from the literature and adapt them to model the data of a designed experimental dataset of formulation mixtures that cover the breadth of EC mixtures of interest to Syngenta.

This currently has not been done to date however we have had success with for example metal organic frameworks in eutectic mixtures and many other datasets there is scope to discover how we can use machine learning in development of formulation design.

The plan is to start with EC's as a simple formulation type to demonstrate the technology and to learn how best to adapt the methodology. You would need to be able to learn programming and be interested in learning machine learning techniques.

Contact Prof Joan Cordiner  j.cordiner@sheffield.ac.uk for more details.  You can find information on our group at: https://sites.google.com/sheffield.ac.uk/the-cordiner-group/home.

Please see this link for information on how to apply: https://www.sheffield.ac.uk/cbe/postgraduate/phd/how-apply. Please include the name of your proposed supervisor and the title of the PhD project within your application.

Applicants should have, or expect to achieve, an MEng in Engineering/Computer Science, MSc in Chemistry or similar. Applicants should also have a keen interest in machine learning and modelling.