6533b85ffe1ef96bd12c281b

RESEARCH PRODUCT

Similar chemical structures, dissimilar triplet quantum yields: a CASPT2 model rationalizing the trend of triplet quantum yields in nitroaromatic systems

subject

description

The photophysics of nitroaromatics compounds stand out for being characterized by an ultrafast decay into the triplet manifold and by a significant value of the triplet quantum yield. The latter quantity can change dramatically from one system to another, as proven for the molecules 2- nitronaphthalene, 1-nitronaphthalene, and 2methyl-1nitronaphthale, whose triplet quantum yield have been previously measured to be 0.93 ± 0.15, 0.64 ± 0.12, and 0.33 ± 0.05, respectively (J. Phys. Chem. A 2013, 117, 14100). In the present contribution we rationalize the reported trend for the triplet quantum yield on the basis of the different ability that the excited S 1 state has in the three molecules to reach a non-previously characterized conical intersection with the ground state. Such a path is in competition with the one leading to triplet states population, which, on the basis of the present static description, appear to be equally favorable in the three systems. Performing high-level ab-initio computations, the energy barrier from the S 1 CASPT2//CASSCF minimum to a CASPT2 minimum-energy-crossing-point of the mentioned S 1 /S 0 conical intersection have been computed to follow the same trend than the values of triplet quantum yield in the three nitroaromatics system here under analysis. The CASPT2 minimum-energy-crossing-point have been obtained using the projected constrained optimization method as recently implemented in the Molcas code. The same path has been characterized also for nitrobenzene, obtaining a value for the mentioned energy barrier that nicely fit in the model derived for the three nitro-naphthalene systems, and in agreement with its high value of the triplet quantum yield (greater than 0.8). The ability of the present model to not only rationalize the experimental data of a single molecule but to reproduce a trend for four slightly different systems speaks in favor of its reliability.