Project Description
A PhD studentship is available in the group of Dr Ephrath Solel Moroshko, Dr Antonia Mey (University of Edinburgh; School of Chemistry), and Dr Kartic Subr (University of Edinburgh; School of Informatics).
This project will combine quantum chemistry and machine learning to understand what features govern the reactivity of catalysts for CO2 epoxidation reactions, and use generative AI to design better ones.
CO2 is the most significant contributor to global warming, emitted in large amounts from fossil fuel combustion, industrial processes, and land-use changes. With climate change posing an existential threat, innovative solutions are needed to reduce CO2 emissions and repurpose existing CO2. Carbon capture and utilisation (CCU) aims to turn CO2 from a waste product into a chemical feedstock. One of the most promising CCU reactions is the addition of CO2 to epoxides to form cyclic carbonates. It is 100% atom-economical, and its products are used as battery electrolytes, green solvents and precursors to polymers and pharmaceuticals.
This reaction needs a catalyst. Metal-free organocatalysts are an attractive alternative to metal-based catalysts. They are cheap, have low toxicity and are easy to tune. However, most still need high temperatures or pressures, high catalyst loadings or a halide co-catalyst, which limits their industrial use. Designing better organocatalysts requires understanding why some outperform others. Computational chemistry can provide mechanistic insights and quantify catalytic activity, but general models capable of predicting reactivity across different catalyst families are still lacking.
This project will combine computational chemistry and machine learning (ML) in an iterative approach to catalyst discovery. First, quantum chemical calculations will be used to investigate reaction mechanisms and calculate turnover frequencies (TOFs) for a diverse set of organocatalysts. These data will be used to train ML models that learn structure-activity relationships and predict the reactivity of unexplored catalysts. The resulting models will guide the generation and optimisation of new catalyst structures using approaches such as genetic algorithms and active learning. Promising candidates will then be evaluated using computational chemistry, validating ML predictions and providing new data to improve the models. This feedback loop between chemical modelling and ML will enable efficient exploration of chemical space and the discovery of highly active organocatalysts for CO2 conversion.
Methodology
The project combines quantum chemistry methods (Solel Moroshko group) with machine learning and generative design (Mey and Subr groups).
Training
The student will be jointly supervised, with regular one-to-one meetings and fortnightly joint meetings with Dr. Solel Moroshko and Dr. Mey, and will be a full member of both research groups. Additionally, there will be monthly meetings with Dr Subr for machine-learning input and an informatics perspective. Alongside the E5 cohort training programme, a training plan tailored to the student’s background will cover quantum chemistry, machine learning for molecules, generative models and high-performance computing, with an emphasis on writing reproducible, open-source research software. Further opportunities include the CCP5 Summer School and training from the AIchemy Hub (https://aichemy.ac.uk), and the student will present their work at national and international conferences.
For further information on the project – Project | E5 Doctoral Training Partnership | E5 Doctoral Training Partnership
Funding Information
Funding Notes
The studentship is fully funded by the E5 Doctoral Training Partnership and the University of Edinburgh for 48 months and covers tuition fees and an annual stipend at the UKRI rate, for 2026-27 this is £21,805 per annum, for a candidate satisfying UKRI residency criteria:
Eligibility Requirements
The entry and eligibility criteria can be found at:
Entry and Eligibility criteria | E5 Doctoral Training Partnership | E5 Doctoral Training Partnership
Application Process
Application Process
More information on the application process is available on the E5 DTP website – How to Apply | E5 Doctoral Training Partnership | E5 Doctoral Training Partnership
Closing date: Wednesday 2nd December 2026 (12 Noon GMT)
In the first instance, the initial application of cover letter and CV should be directed to: [email protected]
The University reserves the right, in its sole discretion, to amend or withdraw any of the advertised scholarships, without further notice.
Important
Before Submitting your cover letter and CV, please complete the online School of Chemistry Equality, Diversity & Inclusion Form for 27/28 Entry
The form will automatically generate a unique ‘Receipt Number’ that you must include in your cover letter.
References
References
1. E. Solel, N. Tarannam and S. Kozuch, Chem. Commun., 2019, 55, 5306-5322.
2. M. D. Wodrich, B. Sawatlon, E. Solel, S. Kozuch and C. Corminboeuf, ACS Catal., 2019, 9, 5719-5725.
3. N. T. Runcie and A. S. J. S. Mey, J. Chem. Inf. Model., 2023, 63, 5996-6005.
4. Q.-W. Song, R. Ma, P. Liu, K. Zhang and L.-N. He, Green Chem., 2023, 25, 6538-6560.
5. M. Alves, B. Grignard, R. Mereau, C. Jerome, T. Tassaing and C. Detrembleur, Catal. Sci. Technol., 2017, 7, 2651-2684.
6. M. Ferrer, I. Iribarren, T. Renningholtz, I. Alkorta and C. Trujillo, Beilstein J. Org. Chem., 2024, 20, 2668-2681.
Supplementary Information
Equality and Diversity
The School of Chemistry holds a Silver Athena SWAN award in recognition of our commitment to advance gender equality in higher education. The University is a member of the Race Equality Charter and is a Stonewall Scotland Diversity Champion, actively promoting LGBT equality. The University has a range of initiatives to support a family friendly working environment. See our University Initiatives website for further information. University Initiatives website: https://equality-diversity.ed.ac.uk/inclusion/family-and-carer
