measurements.

Working across the University of Manchester and Peking University, the student will combine world-leading atmospheric-chamber and mass-spectrometry facilities at Peking with Manchester’s expertise in aerosol modelling, computational chemistry and trustworthy AI. At Peking, chamber experiments on secondary organic aerosol formation, new particle formation and black carbon ageing will generate well-characterised, mechanism-constrained training data, alongside ambient measurements from highly polluted Beijing. At Manchester, the student will develop foundation and transformer models for mass spectra, embedded in a retrieval-augmented, literature-grounded reasoning pipeline that reports calibrated confidence. The framework will be tested on contrasting ambient datasets from Beijing and the Manchester Air Quality Supersite.

The student will gain interdisciplinary training spanning experimental atmospheric chemistry, mass spectrometry, machine learning and scientific software, and will graduate equipped to develop the next generation of AI tools for environmental science. The project suits a candidate with a background in atmospheric/environmental science, chemistry, physics, or a computational discipline, and an interest in applying AI to real measurement data.

measurements.

Working across the University of Manchester and Peking University, the student will combine world-leading atmospheric-chamber and mass-spectrometry facilities at Peking with Manchester’s expertise in aerosol modelling, computational chemistry and trustworthy AI. At Peking, chamber experiments on secondary organic aerosol formation, new particle formation and black carbon ageing will generate well-characterised, mechanism-constrained training data, alongside ambient measurements from highly polluted Beijing. At Manchester, the student will develop foundation and transformer models for mass spectra, embedded in a retrieval-augmented, literature-grounded reasoning pipeline that reports calibrated confidence. The framework will be tested on contrasting ambient datasets from Beijing and the Manchester Air Quality Supersite.

The student will gain interdisciplinary training spanning experimental atmospheric chemistry, mass spectrometry, machine learning and scientific software, and will graduate equipped to develop the next generation of AI tools for environmental science. The project suits a candidate with a background in atmospheric/environmental science, chemistry, physics, or a computational discipline, and an interest in applying AI to real measurement data.