A1 Journal article (refereed)
Machine Learning Model to Predict Saturation Vapor Pressures of Atmospheric Aerosol Constituents (2024)
Hyttinen, N., Li, L., Hallquist, M., & Wu, C. (2024). Machine Learning Model to Predict Saturation Vapor Pressures of Atmospheric Aerosol Constituents. ACS - ES & T Air, 1(9), 1156-1163. https://doi.org/10.1021/acsestair.4c00113
JYU authors or editors
Publication details
All authors or editors: Hyttinen, Noora; Li, Linjie; Hallquist, Mattias; Wu, Cheng
Journal or series: ACS - ES & T Air
ISSN: 2837-1402
eISSN: 2837-1402
Publication year: 2024
Publication date: 22/07/2024
Volume: 1
Issue number: 9
Pages range: 1156-1163
Publisher: American Chemical Society
Publication country: United States
Publication language: English
DOI: https://doi.org/10.1021/acsestair.4c00113
Publication open access: Openly available
Publication channel open access: Partially open access channel
Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/96632
Additional information: Published as part of ACS ES&T Air virtual special issue “Elevating Atmospheric Chemistry Measurements and Modeling with Artificial Intelligence”
Abstract
We present a novel machine learning (ML) model for predicting saturation vapor pressures (psat), a physical property of use to describe transport, distribution, mass transfer, and fate of environmental toxins and contaminants. The ML model uses σ-profiles from the conductor-like screening model (COSMO) as molecular descriptors. The main advantages in using σ-profiles instead of other types of molecular representations are the relatively small size of the descriptor and the fact that the addition of new elements does not affect the size of the descriptor. The ML model was trained separately for liquid and solid compounds using experimental vapor pressures at various temperatures. The 95% confidence intervals of the error in the liquid- and solid-phase log10(psat/Pa) are 1.02 and 1.4, respectively. Especially our solid-phase model outperforms all group-contribution models in predicting experimental sublimation pressures of solid compounds. To demonstrate its applicability, the model was used to predict psat of atmospherically relevant species, and the values were compared with those obtained from a new experimental method. Here, our model provided a tool for a better description of this critical property and gave a higher confidence in the measurements.
Keywords: atmosphere (earth); air impurities and contaminants; aerosols; machine learning; modelling (representation)
Free keywords: COSMO; extreme minimal learning machine; σ-profile; liquid; solid; volatility
Contributing organizations
Related projects
- Improving thermodynamic property estimates of SOA constituents using machine learning
- Hyttinen, Noora
- Research Council of Finland
Related research datasets
Ministry reporting: Yes
VIRTA submission year: 2024
Preliminary JUFO rating: 1