Exploiting the unseen to identify the unknown: Machine learning-assisted platforms for the detection of toxic ‘forever chemicals’ (PFAS) in water

Identifying toxic agents is important in all areas of our society, from ensuring the quality of drinking water and food supplies to the detection of disease. Systems that can reliably report such threats and be adapted to detect new emerging threats would address many important grand challenges. This project will develop portable chip-based systems that can be interpreted using machine learning to detect toxic PFAS “forever chemicals” important in environmental and manufacturing contexts.

The research team will blend machine learning-assisted computation and experimental methods to exploit the properties of liquid crystalline materials to enable rapid, inexpensive, and portable detection of PFAS in water at vanishingly low levels relevant in U.S. regulatory contexts. This approach can also advance strategies for the sensitive and selective identification of other important analytes. This project maps directly to current RISE initiatives on campus and to themes increasingly central to calls for large-scale sources of federal funding.

PRINCIPAL INVESTIGATOR

David Lynn, professor of chemical and biological engineering

CO-INVESTIGATORS

Helen Blackwell, professor of chemistry

Christina Remucal, professor of civil and environmental engineering

Reid Van Lehn, associate professor of chemical and biological engineering

Victor Zavala, professor of chemical and biological engineering