Photoemission Spectroscopy of Organic Molecules Using Plane-Wave/Pseudopoential Density Functional Theory: A Comprehensive Computational Protocol for Isolated Molecules, Molecular Aggregates and Periodic Systems

The possibility to perform low-pressure gas-phase photoemission experiments has allowed the exploration at molecular scale of the complex correlation between electronic and structural properties of the matter. A significant advantage in the removal of the interference of the surrounding at low-pressure is also the easier correlation of the experimental results with theoretical ab-initio simulations. In the context of core ionization, the development of accurate methods for calculating inner-shell binding energies (BEs) offers a dual benefit. Firstly, it aids significantly in interpreting experimental data, helping to assign the contributions of all non-equivalent atoms, even in unresolved features arising from a molecular structure. Secondly, it allows for the anticipation of experimental results by accurately predicting spectral lines. In this context, we have developed and extensively tested a computational protocol based on plane- wave/pseudopotential density functional theory (PW-DFT) to predict X-ray photoemission spectra (XPS) in molecules and molecular aggregates. This protocol is based on a ∆SCF approach and has been tested comparing the present theoretical results with experimental XPS spectra collected from several molecular systems in gas phase, ranging from benzene derivates to biomolecules. Our calculations have been performed using semilocal and global/range-separated hybrid density functionals, containing increasing fractions of Hartree-Fock exact exchange (EXX). Specifically, PBE, B3LYP (20 % EXX), HSE (range separated with 25 % EXX at short range) and BH&HLYP (50 % EXX) have been used for the assessment of the computational protocol. Equation-of-motion coupled-cluster with single and double excitations (EOM-CCSD) has been employed as reference theoretical method for comparison. Regarding XPS, our PW-DFT approach demonstrated to be generally accurate and robust even using semilocal DFT, and to be also suitable for application to very large molecular systems and organic thin films deposited on inorganic surfaces. A proof of concept of a robust machine learning (ML) model for the prediction of C1s BEs in isolated organic molecules has been developed and it is discussed. We also present an early-stage verification of the efficiency of the density functionals introduced above for predicting valence-shell ionization spectra. Preliminary data suggest in this case of BH&HLYP as a promising alternative to EOM-CCSD.