Design of silicon-containing arylacetylene resins aided by machine learning enhanced materials genome approach

Silicon-containing acetylene resins have a broad application prospect as a type of organic-inorganic hybrid high-temperature resistant resins. However, its processability still needs further improvement to meet processing requirements for low viscosity. We proposed a materials genome approach to design and screen silicon-containing acetylene resins with excellent processing properties and heat resistance. To high-throughput screen the promising resin, we established machine learning models for predicting the properties of processing and heat resistance. Ten latent resins were screened, and one easy-to-synthesize resin was prepared by the Grignard reagent method to verify the materials genome approach. The results showed that the processing properties of the screened resin are improved evidently, with excellent heat resistance maintained. This work generates a fresh way for cost-effective data-driven designs of silicon-containing acetylene resins.

Automated summary

This study proposed a new approach using a materials genome method to design and screen silicon-containing acetylene resins with excellent processing properties and heat resistance. Through high-throughput screening and machine learning model prediction, a promising silicon-containing acetylene resin was successfully synthesized, confirming improved processing properties.