Accurate Thermochemistry of Novel Energetic Fused Tricyclic 1,2,3,4-Tetrazine Nitro Derivatives from Local Coupled Cluster Methods.

Highly accurate theoretical values of formation enthalpies and bond energies are crucial for reliable predictions of performance and detonation-related phenomena of energetic materials (EM). However, high-level ab initio calculations even for medium-sized important EMs still remain a demanding challenge. In the present work, we studied in detail the gas-phase thermochemistry of novel high-energy polynitro derivatives of 5/6/5 structural frameworks comprised of fused 1,2,3,4,-tetrazine and two 1,2,4-triazole or pyrazole rings. To this end, we proposed and benchmarked a "bottom-up" approach. First, highly accurate multi-level procedures W2-F12 and/or W1-F12 in conjunction with the atomization energy approach were utilized for smaller species. In turn, for medium-sized species (up to 24 non-H atoms), these values were complemented with the enthalpies of isodesmic reactions calculated using the recently proposed domain-based local pair natural orbitals (DLPNO) modifications of coupled cluster techniques. The benchmarks on a number of atomization energies and enthalpies of isodesmic reactions reveal that the DLPNO-CCSD(T)/aVQZ approach does not deteriorate the quality of the W1-F12 and W2-F12 procedures and exhibits overall accuracy close to "chemical" (~1 kcal mol-1). We obtained a set of accurate and mutually consistent gas-phase formation enthalpies for 12 energetic heterocyclic species. Among them, the gas-phase formation enthalpy of 1,2,9,10-tetranitrodipyrazolo[1,5-d:5',1'-f][1,2,3,4]tetrazine, a novel promising EM, turned out to be 214.5 kcal mol-1, which is ~12 kcal mol 1 higher than the best theoretical estimates available in the literature. The formation enthalpy of another novel EM, a fused tricyclic 1,2,3,4-tetrazine with two nitro-1,2,4-triazole moieties, was predicted to be 213.5 kcal mol 1, which is also ~4 kcal mol 1 higher than the reported value. Apart from this, we considered the thermodynamics of radical reactions (viz., C-NO2 bond scission) and the thermochemistry of the corresponding radicals. The difference between DLPNO-CCSD(T)/aVQZ and CCSD(T)-F12/VTZ-F12 benchmark values did not exceed 1 kcal mol-1. In a more general sense, the use of DLPNO-CCSD(T) in conjunction with the "bottom-up" approach is promising for quantitative thermochemical calculations of energetic materials comprised of the species up to several dozens of CHNO-atoms.