The development of electronic structure methods that can accurately describe ground and excited states of molecular systems with manageable computational costs and in a systematically improvable manner continues to be the central theme of quantum chemistry. This dissertation focuses on some of the recent developments in the coupled-cluster (CC) theory and its equation-of-motion (EOM) extension to excited electronic states. One of the key challenges in the development of the CC and EOMCC methodologies is the incorporation of many-electron correlation effects due to higher-rank components of the cluster and EOM excitation operators without incurring significant increase in the computational costs, while avoiding failures of perturbative methods of the CCSD(T) type in multireference situations, such as bond breaking and excited states dominated by two-electron transitions, and in certain weakly bound systems. Among the best ways to address these issues is the CC(P;Q) framework, which provides robust and computationally affordable noniterative energy corrections to lower-order CC/EOMCC calculations. In this dissertation, we discuss the different CC(P;Q) variants relying on both the conventional and unconventional truncations in the cluster and EOM excitation operators. The advantages of the CC(P;Q) hierarchy are illustrated using a few examples ranging from small molecule spectroscopy to photochemistry of large organic species in solution. In particular, we discuss the computational investigations of the novel super photobase FR0-SB, which exhibits a drastic increase in basicity upon photoexcitation, including the energetics and properties of its excited states, the steric effects governing the excited-state proton transfer involving FR0-SB and alcohols, and the enhanced photoreactivity of FR0-SB resulting from two-photon excitations, where the δ-CR-EOMCC(2,3) approach that belongs to the CC(P;Q) hierarchy played a key role. Furthermore, we demonstrate that the relatively inexpensive CC(t;3) and CC(q;4) approaches derived from the CC(P;Q) framework are as accurate in describing the challenging weakly bound magnesium dimer, including its ground-state potential and vibrational levels supported by it, as the much more demanding CCSDT and CCSDTQ parent theories. We also show how the highly accurate ground- and excited-state ab initio potentials obtained in the state-of-the-art CCSDT, CR-EOMCCSD(T), and full configuration interaction (CI) computations allowed us to resolve the existing laser-induced fluorescence and photoabsorption spectra of the magnesium dimer and find the missing high-lying vibrational states of Mg2 that have eluded scientists for half a century. Last, but not least, we discuss our recent extension of the semi-stochastic CC(P;Q) framework, which combines the deterministic CC(P;Q) theory with stochastic CI quantum Monte Carlo (QMC), to excited electronic states, providing rapid convergence to the parent high-level EOMCC methods, such as EOMCCSDT, out of the early stages of QMC propagations. The advantages of the semi-stochastic CC(P;Q) approach targeting EOMCCSDT are illustrated by examining vertical excitations in CH+ and adiabatic excitations in the CH and CNC species.
Read
