Grammar of Impact Sensitivity: An Incremental Theory.

In this paper, we report the first attempt to quantify impact sensitivity using the second-order incremental approach based on the structural features of explosives. It has been found that impact height (h50) can be expressed via a multiplicative incremental exponential form, in which the exponents are characteristic coefficients of structural increments multiplied by their numbers in the molecule. The method was developed on a large array of experimental data (450 molecules and salts) of different energetic materials, namely, nitro compounds, peroxides, nitrogen-rich salts, heterocycles, etc., while testing of the model was performed for 170 compounds. The results demonstrate a noticeable correlation with the experimental h50 values. Thus, the corresponding R2 and RMSE for the training and test sets are 0.56 (12.5 J) and 0.63 (18.8 J), respectively. In this work, we use 53 individual structural increments, but their number can be extended, and the corresponding coefficients can be refined; this allows for increasing the prediction accuracy on-the-fly. The calculation algorithm is discussed, and the corresponding examples are presented. The performed machine-based regression analysis using genetic function approximation, multiple linear regression, and artificial neural network has proven the reasonability and informativity of the proposed incremental theory. Thus, the developed approach significantly extends our understanding of the impact sensitivity phenomenon and translates it into the category of one that can be calculated by a pocket calculator.

Accelerating the Design of High-Energy-Density Hydrocarbon Fuels by Learning from the Data.

In the ZINC20 database, with the aid of maximum substructure searches, common substructures were obtained from molecules with high-strain-energy and combustion heat values, and further provided domain knowledge on how to design high-energy-density hydrocarbon (HEDH) fuels. Notably, quadricyclane and syntin could be topologically assembled through these substructures, and the corresponding assembled schemes guided the design of 20 fuel molecules (ZD-1 to ZD-20). The fuel properties of the molecules were evaluated by using group-contribution methods and density functional theory (DFT) calculations, where ZD-6 stood out due to the high volumetric net heat of combustion, high specific impulse, low melting point, and acceptable flash point. Based on the neural network model for evaluating the synthetic complexity (SCScore), the estimated value of ZD-6 was close to that of syntin, indicating that the synthetic complexity of ZD-6 was comparable to that of syntin. This work not only provides ZD-6 as a potential HEDH fuel, but also illustrates the superiority of learning design strategies from the data in increasing the understanding of structure and performance relationships and accelerating the development of novel HEDH fuels.

Rational design of covalent heptazine framework photocatalysts with high oxidation ability through reaction-dependent strategy.

Due to structural tunability, high surface area, abundant pore structures, and abundant active sites, covalent heptazine frameworks (CHFs) constructed from heptazine and other molecular blocks are especially prominent. Here, we proposed a reaction-dependent strategy for designing two dimensional CHFs including high-throughput precursors screening, structure generation, and performance evaluation. Assuming that oxamide-like precursors can undergo the same thermal polymerization reaction as producing C6N7, seven precursors were screened from more than 109 molecules in the ZINC20 database in terms of molecular weight, number of substructures, shape index, and symmetry. Furthermore, CHF-L1 to CHF-L7 were constructed from urea and the seven precursors according to the topologically assembling scheme in thermal polymerization. The designed CHFs had band gaps ranging from 1.89 to 3.10 eV. Among them, CHF-L3 assembled structurally by urea and 1,2,4,5-tetrazine-3,6-dicarboxamide with the smallest bandgap and an oxidative potential bias of 1.38 V for oxygen evolution reaction was screened as the candidate with high oxidative ability. The negative formation energy based on the synthesis route indicated the synthetic feasibility of CHF-L3, and negative cohesive energy as well as the stable structure under ab initio molecular dynamics simulations confirmed the stability of CHF-L3. The present work is expected to provide a powerful design strategy for two-dimensional CHFs design and is broadly applicable to various computational covalent organic framework design systems and experimental studies.

Accelerating Molecular Design of Cage Energetic Materials with Zero Oxygen Balance through Large-Scale Database Search.

Domain-related knowledge promoted high-throughput cage scaffold screening from the ZINC15 database containing over 130 000 scaffolds and cooperated with combinatorial design to alleviate the lack of cage energetic materials. A dozen candidates were discovered that show excellent energy and safety performance, confirming the effectiveness of our strategy.