1 of 13

AI for safer alternatives

Krishna Rajan

“Charging Forward” Seminar

October 15th 2024

Collaboratory for a Regenerative Economy

2 of 13

!!!!

‘-

2

3 of 13

3

‘-

3

4 of 13

Chemical By-products

Health and Environmental Impact

Synthesis

  • What defines the characteristics of a material?
  • What are the inputs in defining a material that govern both functionality and hazard mitigation
  • Harnessing AI for: substitution, recyclability, process design

‘-

4

5 of 13

​

  • Science challenge:

​

Achieving the dual objectives of Sustainability and Functionality simultaneously in chemicals and materials design is difficult

​

    • Sparse information about the materials and chemicals: their origins, use, and impact
    • Urgency needed in identifying alternatives
    • Status quo in materials discovery is slow and often relies on trial-and-error and heuristic-based interpretation

​

​

​

Scientific Challenges and Needs

‘-

5

6 of 13

Toro et.al. Energies 2023, 16(18), 6571; https://doi.org/10.3390/en16186571

Materials Landscape of Batteries

Drug Discovery Approach to Breakthroughs in Batteries

NSF Workshop (2008)

‘-

6

7 of 13

Correlative analysis of metal organic framework structures through manifold learning of Hirshfeld

Surfaces Xiaozhou Shen, Tianmu Zhang, Scott Broderick and Krishna Rajan*

DOI: 10.1039/c8me00014j

Krishna Rajan

Energy Environ. Sci., 2011,4, 3030-3040

AI aided discovery of new materials

‘-

7

8 of 13

Chemicals

Materials

Chemicals Ecosystem of Materials

‘-

8

9 of 13

  • Science platform requirements:

​

    • Build a trustworthy computational means for estimating hazards in an accelerated manner

​

    • Establish a “rational design’ approach to chemicals and materials --- need to understand why in order to accelerate the process in choosing safer AND functional chemistry/ processes

​

    • Harness techniques to visualize and interpret complex relationships in data to enable decision making for diverse stakeholders

‘-

9

10 of 13

Accelerated Selection of Safer Solvents

Informatics Driven Materials Innovation for a Regenerative Economy: Harnessing NLP for Safer Chemistry in Manufacturing of Solar Cells: D. Giri, A. Mukherjee, and K.Rajan; A. Lazou et al. (eds.), REWAS 2022: Developing Tomorrow’s Technical Cycles Volume 1), The Minerals, Metals & Materials Series,

‘-

10

11 of 13

AI screening of safer chemistries for synthesis of next generation of solar devices that are amenable to ultra low carbon technologies- provides a means utilizing safer chemicals ---multi attribute signature for eco-labeling

Deep Learning Model for Identifying Critical Structural Motifs in Potential Endocrine Disruptors

Arpan Mukherjee, An Su, and Krishna Rajan*

Cite this: J. Chem. Inf. Model. 2021, 61, 5, 2187–2197

AI to trace Chemicals for Materials Synthesis

Artificial intelligence informed toxicity screening of amine chemistries used in the synthesis of hybrid organic–inorganic perovskites A. Su, Y.Cheng. H.Xue, Y.She and K. Rajan AIChE J. 2022;68:e17699

‘-

11

12 of 13

    • Multi attribute database of PFAS family compounds developed by AI methods coupled to computational chemistry

​

    • EPA adopted our AI tools for guiding standards in PFAS screening and classification

Su, A., Rajan, K. A database framework for rapid screening of structure-function relationships in PFAS chemistry. Sci Data 8, 14 (2021). ; Su,A. et.al. An artificial intelligence platform for automated PFAS subgroup classification: A discovery tool for PFAS screening; Science of The Total Environment 921, 171229 (2024)

AI Driven Standards & Screening, Handling and Abatement of PFAS

National PFAS Testing Strategy- October 2021

‘-

12

13 of 13

Benign-by-design

Design with intent

Regenerative Materials Design

‘-

13