Transferring the available fused cyclic scaffolds for high—throughput combinatorial design of highly energetic materials via database mining

Recently, the fused cyclic compounds have been the object of an increased interest in the field of energetic materials (EMs) due to the trade-off between energy and safety. Compared with the fused cyclic EMs consisting of the azoles or azines, the oxazole-based fused EMs which possibly possess higher energy–density are very lacking. Here, we proposed an efficient approach to design the highly energetic oxazole-based fused materials. The domain-related knowledge promoted an advanced database search for the aromatic oxazole-based scaffolds from the buyable subset of the ZINC20 database, ensuring scaffolds are available for purchase. Then, 171 target scaffolds were transferred into the EM field and cooperated with combinatorial design to construct a chemical space containing over 10⁷ potential energetic molecules. The high-throughput screening was performed in four aspects, namely, density, difficulty of synthesis, detonation performance, and sensitivity, to accelerate the search for candidates. Meanwhile, the statistical analysis through the hierarchical filtrations clarified the potential of 2r-3s and 2r-4s scaffold types for creating highly energetic molecules. Finally, several candidates stood out owing to nearly 10000 m/s detonation velocity and acceptable predicted sensitivity, elucidating the effectiveness of our approach. We anticipate this investigation could not only be a vital point for subsequent fused cyclic EM research, but also offered a new avenue for material design in other fields.