Taking trace compounds into account in Chemoinformatics

With the announced drastic reduction of the amount of fossil energy resources in the European energy mix, it is important to define viable alternatives to petrochemicals. The transformation of lignocellulosic biomass and the various synthesis processes make it possible to provide a wide variety of families of compounds with high added value. Many fluids (lubricants, fuels, etc.) are used in the energy, transport and environment sectors, for various applications ranging from the production of renewable energies to the transport sector.
Whatever the application is, the fluid properties should be constant over time and therefore, the fluids stability represents a major issue. Degradation due to thermal or oxidation effects leads to altering product quality which can thus limit systems efficiency or even in certain cases lead to damages. This stability can be significantly improved or altered with the presence of additives or impurities. In recent years, we have implemented numerous techniques based on artificial intelligence applied to chemistry databases (Chemoinformatics methods) to develop predictive models. We have demonstrated their potential for various applications in the fields of energy, transport and environment. The prediction of mixture properties, and more particularly the case of complex mixtures (fuels, lubricants, etc.), still requires numerous developments, especially for the evaluation of impurities or additives effects. Indeed, these fluids are currently simplified, assimilated to simple mixtures and trace compounds are not considered.
The challenge of this thesis work is therefore to improve the representation of complex fluids to increase the robustness of models based on machine learning.

Keywords: Chemoinformatics, Machine Learning, Fluids, Additives, Oxidative stability

• Academic supervisor    Dr Gilles MARCOU, Laboratoire de Chemoinformatique, ORCID: 0000-0003-1676-6708
• Doctoral School    ED222, Ecole Doctorale des Sciences Chimiques, Université de Strasbourg.
• IFPEN supervisor    Dr Benoît CRETON, benoit.creton@ifpen.fr, ORCID: 0000-0002-3287-877X
• PhD location    IFPEN, Rueil-Malmaison, France.  
• Duration and start date    3 years, starting in the fourth quarter 2024
• Employer    IFPEN
• Academic requirements    University Master degree in Chemical sciences.  
• Language requirements    Fluency in English, and in French or willingness to learn French
• Other requirements    Chemoinformatics, computer science

This thesis offer is valid subject to funding agreement.

To apply, please send your cover letter and CV to the IFPEN supervisor indicated here below.