Why dibenzo[b,f][1,4,5]oxadiazepine fails for molecular solar thermal energy storage: Mechanistic insights and a...

Azobenzene derivatives, especially seven-membered cyclic compounds such as dibenzo[b,f][1,4,5]thiadiazepine (DBTD), are considered promising candidates for molecular solar thermal (MOST) energy storage systems due to their significant energy storage capabilities. In contrast, its structural analogue, dibenzo[b,f][1,4,5]oxadiazepine (DBOD), fails to exhibit substantial Z → E photoisomerization, despite possessing a higher isomerization energy. In this study, we investigated the underlying mechanistic causes of this paradox by comparing the photoisomerization dynamics of DBTD and DBOD through electronic structure calculations and nonadiabatic molecular dynamics (NAMD) simulations. Our results revealed that, while photoexcitation leads to the formation of the E-isomer in DBOD, its metastable E-configuration rapidly relaxes back to the Z-form within 400 fs at room temperature. This rapid back-conversion arises from three synergistic factors: (i) a lower thermal isomerization barrier, (ii) substantial kinetic energy retained upon de-excitation, and (iii) preferential allocation of kinetic energy into azo-group torsional modes, which together accelerate the reverse reaction on the ground state. To mitigate this issue, we proposed a synergistic strategy combining fluorine substitution with low-temperature regulation. Fluorination yields perfluorodibenzo[b,f][1,4,5]oxadiazepine (PDOD), which increases the thermal E → Z isomerization barrier while preserving a high isomerization energy of 164.3 kJ/mol. NAMD simulations at 100 K show that the E-isomer of PDOD remained stable for over 400 fs, with a Z → E quantum yield of 26%, significantly higher than the 19% quantum yield of DBOD at 300 K. This study provides mechanistic insights into the "high isomerization energy yet poor stability" paradox observed in DBOD and establishes a practical strategy for developing high-performance MOST photoswitches.