Development and implementation of theoretical methodologies for the study of luminescent systems

Abstract

Accurate prediction of excited state energies is extremely important for studies in various fields. In lanthanide complexes, calculations for determining excited states are generally performed using ab initio methods, TDDFT, or the semi-empirical INDO/S-CIS method. The first two cases are limited to systems containing a few dozen atoms. Although capable of handling large molecules, the semi-empirical quantum INDO/S-CIS method exhibits high errors when excitation window adjustment is not performed. Thus, we propose two parameterizations of CIS methods based on the INDO and NDDO approximations, named INDO/S-CIS2 and NDDO/S-CIS, for calculating the singlet and triplet excited state energies of systems containing lanthanide ions. To validate the methods, we compared the results obtained against TDDFT methodologies, based on a benchmark study, and INDO/S-CIS. Both developed methods presented errors smaller than or equivalent to those of the TDDFT methodologies. The error ranges for INDO/S-CIS2 and NDDO/SCIS were 10.43 ± 9.53 and 9.52 ± 7.13, respectively, for the singlet energy, in a set of 41 structures. For the triplet energy, in a set of 87 structures, the error ranges were 9.46 ± 6.88 and 7.68 ± 5.65, respectively. Furthermore, we also present the second version of the LUMPAC software with the incorporation of new methods and resources. Methods for calculating Judd-Ofelt intensity parameters from the excitation spectrum were implemented, as well as a methodology for calculating energy transfer rates for Tb3+ compounds. A novel module for visualizing molecular orbitals was also introduced, facilitating the characterization of excited states centered on the ligands involved in the energy transfer process and the interpretation of the theoretical absorption spectrum. The molecule viewer was optimized. The corresponding Jablonski diagram is generated automatically. In addition, code improvements have been made.