Grants and Contributions:

Deep Generative AI for the Design and Synthesis of Perovskite with Targeted Properties

1014941

$319,000.00

Mar 21, 2024 - Mar 31, 2026

The field of novel materials discovery has witnessed several interesting techniques that span across traditional synthetic chemistry and computational methods. Synthetic chemistry methods are normally Edisonian-based (i.e., trial-and-error) and often will rely on experiential knowledge to synthesize chemically stable materials. Machine Learning (ML) approaches, that are based on Deep Generative Modeling (DGM), are commendable alternatives for novel materials discovery due to their impressive ability in analyzing robust chemical design space. Famous for their speed, reliability, and low cost, DGM techniques can intelligently identify hidden patterns and correlations in a training dataset by solving an inverse design scheme. However, deep generative ML (DGML) approaches face challenges related to lattice reconstruction at the decoding phase, potentially leading to two major shortcomings. To address the highlighted challenges, the project will develop progressive deep learning approaches for novel materials discovery in two stages. The first stage will leverage on the technical strengths of both a semi-supervisory VAE (i.e. SS-VAE) model and an auxiliary GAN (i.e. A-GAN) model. The model architecture is referred to as Lattice-Constrained Materials Generative Model (LCMGM) and will be used to screen stable inorganic perovskite materials with crystal lattice conformities that are consistent with predefined symmetrical constraints at the encoding phase. Furthermore, the second stage will involve the novel design and development of a Thermodynamics Guided Diffusive Generative Model (TGDGM) for discovering Hybrid Organic-Inorganic Perovskite (HOIP). The TGDGM shall build on the LCMGM for facilitating autonomous materials search with high scalability and reliability. The AI generated perovskite crystal structures will be synthesized using highly-scalable experimental means for meeting targeted applications in many engineering domains. The synthetization processes will be developed for the most promising AI discovered compounds at the University of Ottawa. The focus will be on producing powders or particles with optimized properties for future research on novel perovskite structures produced by coating or layering. Two approaches will be investigated simultaneously, starting primarily from mixed oxide particles: (1) solid state processing and (2) sol gel synthesis. In addition, extensive materials characterization will provide important knowledge and materials data, which will serve as crucial closeloop feedback for further optimizations of both the Machine Learning models and synthetization processes, with primary focus on applications such as solar cells, energy storage materials and catalysts.

National Research Council Canada

In the short term, anticipated outcomes will be strengthened collaborations across industry, academia, and government to support research excellence. In the medium term, anticipated outcomes will be the development of new and potentially disruptive technologies with collaborators. In the long term, find collaborative solutions to public policy challenges and create stronger innovation systems.

Ottawa, Ontario, CA K1N 6N5

172-2023-2024-Q4-1014941

Grant

Grants and Contributions

119278877

Academia

This agreement has been amended 1 time(s) the total amended value is $319,000.The end date of this agreement has been modified by -1 days.

May 26, 2025

University of Ottawa

Ottawa--Vanier--Gloucester

35081

Collaborative Science, Technology and Innovation Program - Collaborative R&D Initiatives

Collaborate on multiparty research and development programs to catalyze transformative, high-risk, high-reward research with the potential for game-changing scientific discoveries and technological breakthroughs in priority areas.

541710