BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ZContent.net//ZapCalLib 1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
SUMMARY:Design Principles for Sustainable Chemistry: A Theoretical and Mach
 ine Learning Approach
LOCATION:Chemistry A101
TZID:America/Denver
DTSTART:20241010T160000
UID:2026-04-24-07-37-24@natsci.colostate.edu
DTSTAMP:20260424T073724
Description:About the Seminar: \n\nOur group\\'s overarching goal is to de
 velop new methods to extract sustainable fuels\, polymers\, and chemicals 
 from plants. Our approach has been to develop and apply computational tool
 s to biological and chemical conversion processes as part of an iterative 
 ‘model-validate-predict’ design process for de novo catalysts.\n\nWi
 th its high carbon and hydrogen content\, lignocellulosic biomass presents
  an alternative to petroleum as a nearly carbon-neutral precursor to upgra
 ded liquid fuels\, renewable polymers\, and chemicals. I will show some re
 presentative results in designing new catalysts for biological and chemoca
 talytic processes of biomass using theoretical and machine learning approa
 ches.\n\nSolubility is a crucial chemical property that affects the reacti
 vity in chemical synthesis and the separation ability between solvents and
  solutes during the extraction and purification of products. Its importanc
 e has been considered a key factor when designing chemical processes in va
 rious fields\, such as electrochemistry\, organic\, pharmaceutical\, petro
 leum\, and sustainable polymer chemistry. Here\, we first introduce solubi
 lity prediction of organic compounds via self-evolving solubility database
 s and graph neural networks (GNNs). Prediction of polymer and protein solu
 bility followed by experimental validation are other practical application
 s. This project resulted in extensive\, self-evolving\, and reliable solub
 ility databases and accurate GNNs for various solutes\, including polymers
  and proteins dissolved in single and multicomponent solvents.\n\nThen\, w
 e will go over our traditional “Fuel property first” design approach t
 o reduce emissions and increase performance for biofuel candidates. We hav
 e developed chemistry and physics-informed GNN models to predict a few fue
 l properties\, including yield sooting index (YSI)\, cetane number (CN)\, 
 critical temperature (Tc)\, and heat of vaporization (HoV). These properti
 es are key factors in determining fuel performance\, emissions\, and safet
 y and can vary substantially between bioblendstock candidates under consid
 eration. The model and methodology used in this work can be applied to oth
 er fuel properties\, leading to rational principles for designing high-per
 formance fuels. 4:00 pm
END:VEVENT
END:VCALENDAR