Abstract

Key quantities for designing new opto-electronic devices that are based on disordered organic semiconductors, such as organic light-emitting diodes (OLEDs), are the frontier orbital energies, exciton energies and exciton binding energies. Obtaining these quantities spectroscopically with a high accuracy is complicated by the non-adiabaticity of such experiments. Here, we present for two isomers of the blue-emitting prototypical OLED material MADN (2-methyl-9,10-bis(naphthalen-2- yl)anthracene) multiscale first-principles simulations of optical absorption spectra and of the electron affinity, EA, a theoretical prediction and analysis of low-energy inverse photoemission spectroscopy (LEIPS) spectra that probe EA and experimental LEIPS spectra. The simulations combine many- body Green’s function theory, polarizable film-embedding, and multimode electron-vibrational cou- pling. We show how the onset energy of LEIPS spectra, which is commonly used to estimate EA, can differ from the actual adiabatic value, depending on material parameters and the instrumental resolution. For the two isomers of MADN, the theoretical and experimental onset energies differ by about 0.2 eV, which is within the expected uncertainty margin. However, the experimental spectra are almost featureless, whereas the theoretical spectra show a clear peak structure. An extensive study is presented of the possible effects of thin film charging on the spectra, including charging of deep trap states. The calculated optical absorption spectra agree excellently with experiment, with a peak energy difference of only about 0.05 eV. For the two isomers of MADN, the calculated adiabatic singlet exciton binding energies are 1.0–1.1 eV.