Multi-objective optimization strategy for green solvent design via a deep generative model learned from pre-set molecule pairs

Abstract

Green solvent design is usually a multi-objective optimization problem that requires identification of a set of solvent molecules to balance multiple, often trade-off, properties.

An interpretable 3D multi-hierarchical representation-based deep neural network for environmental, health and safety properties prediction of organic solvents

“…Recently, the idea of sustainable development 1 has gained widespread attention in many chemical processes, such as solvent selection 2,3 and molecule design, 4,5 to avoid negative consequences as efficiently as possible. 6,7 This is because the products and process activities of the chemical processes have a potential influence on the environmental, health, and safety (EH&S) performance. 8…”

Comprehensive Framework for the Design of High-Performing and Sustainable Polyamides

“…This strategy, alongside other MCTS-based studies elsewhere, , has leveraged controlled search spaces to uncover new organic materials. Moreover, ML methods have proved applicable in learning, predicting, and optimization of environmental, health, and safety properties in prospective environmental sustainability studies. − In our earlier work, we demonstrated a computational framework for application-specific metal–organic frameworks − that was based on generative models and demonstrated adaptability across different materials, such as ionic liquids. These techniques are pivotal in unlocking the vast potential of materials design and finding better applications tailored to a particular target.…”