%0 Generic %A Mewes, Jan-Michael %D 2015 %F heidok:18465 %K Excited States, Condensed Phase, PCM, xBDSM, Photochemistry, Electronic Structure, ADC, Nitrobenzene %R 10.11588/heidok.00018465 %T Development and Application of Methods for the Description of Photochemical Processes in Condensed Phase %U https://archiv.ub.uni-heidelberg.de/volltextserver/18465/ %X The interaction of complex molecular systems with light has great relevance in nature as well as for many of the latest technological developments. The process of photosynthesis converts light into chemical energy, thereby providing the primary energy source for life on Earth. Photovoltaic devices, as the technological implementation of this principle, constitute one of the most promising sources of electrical energy for the 21st century, whose application has increased tremendously in the past decade. The reverse process, the controlled emission of light from electronically activated (excited) molecules, is central to many modern technologies, the most prominent of which are presumably the ever smaller yet higher resolving display screens of hand-held computers. For all these applications, a fundamental understanding of the processes taking place at the atomistic scale is of key relevance to allow for a rational design and improvement of new technologies. However, due to the ultra-short time-scales on which the elementary steps of most light-induced phenomena occur and their inherent complexity, an exclusively experimental investigation is often tedious, in particular concerning the interpretation of the results. Here, the combination of experimental techniques and theoretical models can help to gain insights into the involved processes. For this purpose, the electronic structure of ground and light-activated (excited) states of the involved molecules as well as the interaction with their environment has to be approximated, which is the central topic of this work. In the first part, namely chapters 2-4, I present applications of the quantum-mechanical methodology introduced in chapter 1 to study light-induced processes in molecular systems. The so-called caged compounds studied in chapters 2 and 3 constitute an attempt to employ the remarkable spatio-temporal light control of modern lasers to control chemical reactions. For this purpose, the investigated, prototypical molecules nitro-phenylacetate (NPA) and ortho-nitrobenzylacetate (oNBA) serve as precursors for the active compounds CO2 and acetate, respectively. Upon irradiation with UV light, the active compound is released within nano- to microseconds, and may e.g. trigger subsequent reactions. In the above-mentioned sense, my theoretical investigation accompanied and guided an experimental study, which allowed to shed light on the molecular processes and to resolve the detail of the mechanism responsible for the light-induced reactivity. The common structural motif of NPA, oNBA and many other photo-active systems is the nitroaromatic moiety in the form of its smallest representative nitrobenzene (NB). Due to this prototypical character, the photochemistry of NB is relevant for many photochemical applications. In chapter 4, I report an extensive theoretical investigation of ground and excited states as well as the non-radiative decay of NB, which due to its small size and high symmetry allows for an application of a hierarchy of state-of-the-art quantum-chemical methods. Surprisingly, I found this small molecule to pose a serious challenge to electronic structure theory and consequently, some rather sophisticated ab initio methods fail to afford an accurate description, e.g. with respect to the photochemically very important ordering of the lowest triplet states. Nevertheless, I determined the mechanism of non-radiative decay in good agreement with experimental findings and, moreover, suggested an experiment to test my hypothesis. Although there exist a number of accurate and reliable quantum chemical methods that allow for an investigation of the ground and excited states of isolated systems with the molecular size of NPA, oNBA and NB, the environment often plays a crucial role and may decisively influence the light-induced processes, as e.g. in NPA. Hence, the approximate modeling of molecular environments for quantum-chemical problems in condensed phase is a very active field of research, which culminated in the 2013 Nobel Price for Chemistry, which was awarded to Karplus, Levitt and Warshel for their pioneering developments in the field of multiscale models for complex chemical systems. To enable a quantum-chemical description of photo-chemical excitation processes in condensed phase, I extended and implemented a quantum-classical polarizable-continuum model (PCM) for calculation of vertical excitation energies, which is described in chapter \ref{part:pcm}. In general, PCMs allow for an efficient computation of the often dominating electrostatic portion of the solute-solvent interaction by means of the macroscopic descriptors epsilon (dielectric constant) and epsilon_opt = n^2 (optical dielectric or squared refractive index, respectively). The implementation of the method was realized in such a way that its application to any quantum-chemical model that affords electron densities for ground- and excited-states is straightforward. For the systematic evaluation of the method, I composed the first set of experimental Benchmark Data for Solvatochromism in Molecules (xBDSM), and part of the data points were measured by myself. Comparing calculated gas phase to solvent shifts to the xBDSM set, I was able to demonstrate the convincing accuracy of my approach in combination with various levels of electronic structure theory and could shed light on the relation of different flavors of excited state PCMs. Moreover, a close examination of the contributions to the calculated shifts revealed general patterns, which are essential regarding any evaluation of calculated solvent shifts by comparison to the experiment. The implemented methodology will be released with one of the next versions of the Q-Chem quantum-chemical software package.