The research described in this dissertation concerns two fields of theoretical chemistry: Part I concerns applications of Density Functional Theory, and part II high accuracy calculations within the Born-Oppenheimer approximation using explicitly correlated Gaussian functions.In the first part, after a brief introduction to Density Functional Theory and Hartree Fock methods, the candidate's research in Density Functional Theory is described in two chapters. One treats the charge transport in B-DNA, specifically (GC)$_N$ oligomers solvated by water. The second chapter treats the charge transfer between the Lithium atom and Fullerene-C$_{60}$ in the endohedral complex Li@C$_{60}$. In both applications Density Functional Theory was the central quantum mechanical technique that allowed the approaching of such large molecular systems.In the second part of this dissertation, the candidate's development of a FORTRAN code using explicitly correlated Gaussian functions within the Born-Oppenheimer approximation is presented.Every item of the author's research during his graduate studies has been published in co-authorship with the author's scientific advisor and other collaborators in peer-reviewed journals. A total of 8 scientific articles and one letter have been published by the author while at The University of Arizona.

This thesis focuses on the construction and role of benchmark quality computations in the area of non-covalent interactions. We have provided a detailed error analysis of focal-point schemes commonly used in benchmark quality computations, as well as provide error and speedup analysis of commonly used approximations to these methods. An analysis of basis set effects on higher-order corrections to MP2/CBS has been carried out, providing the community error bounds on future benchmarks. We demonstrate how these high-level computations can elucidate a better understanding of non-bonded interactions in chemistry as well as provide high-quality reference data to refit existing methods against to increase the overall accuracy of the method.

To follow up on the unexpectedly-good performance of coupled-cluster models with approx- imate inclusion of 3-body clusters [J. Chem. Phys. 151, 064102 (2019)] we performed a more complete assessment of the 3CC method [J. Chem. Phys. 125, 204105 (2006)] for accurate computational thermochemistry in the standard HEAT framework. New spin- integrated implementation of the 3CC method applicable to closed- and open-shell systems utilizes a new automated toolchain for derivation, optimization, and evaluation of operator algebra in many-body electronic structure. We found that with a double-zeta basis set the 3CC correlation energies and their atomization energy contributions are almost always more accurate (with respect to the CCSDTQ reference) than the CCSDT model as well as the standard CCSD(T) model. The mean errors in { 3CC, CCSDT, and CCSD(T) } electronic (per valence electron) and atomization energies were {23, 69, 125} μEh/e and {0.39, 1.92, 2.57} kJ/mol, respectively. The significant and systematic reduction of the error by the 3CC method and its lower cost than CCSDT suggests it as a viable candidate for post-CCSD(T) thermochemistry application. / Doctor of Philosophy / Stepping into the information age, the computing power has rapidly grown over the last half century. Solving chemical problems on computers has improved lives by reducing the cost and time of researching critical technologies. Scientific research is evolving and experimental finding are now supported with a computational model. Doing chemistry on computers requires quantum simulations, which is essentially solving the Schr ̈odinger equation on a computer that simulates a wave function for all the electrons in a system. Different models are built based on how these inter electronic interactions are treated. To predict results with accuracy on par with the experimental findings requires using higher-order wave functions methods.These are computationally expensive and often not practical. The lower-order methods that are easy to implement can be found in all quantum chemistry software packages. On the other hand, the higher-order methods are laborious and error prone to implement manually due to the sheer complexity of theory. Debugging such implementations often requires a lot of effort with the uncertainty in returns. To solve this problem, we implemented a second-quantization toolkit (SeQuant version 2.0) that derives many-body methods, specifically the general-order coupled cluster (CC) model. The CC model is systematically improvable and accurate. One such CC model, the CCSD(T), has been called the gold standard in quantum chemistry. For compactness, these equations are usually derived in their spin-orbital form. The evaluation and storage cost of these methods is reduced up to four-fold by transforming the spin-orbital expressions to a spin-traced form. In this work, the spin-tracing algorithms are described in detail. The general-order coupled cluster approach is used to derive the internally corrected approximate coupled cluster methods. These methods improve the accuracy of a model at a reduced cost. For small molecules, it was observed that the spin-traced evaluation was over three times faster than spin-orbital coupled cluster. To further reduce the cost of calculations, we added explicit correlation to our CC models. These methods improved the quality of our results with a modest increase in the computational cost.

Conventional correlation methods suffer from the slow convergence of electron correlation energies with respect to the size of orbital expansions. This problem is due to the fact that orbital products alone cannot describe the behavior of the exact wave function at short inter-electronic distances. Explicitly correlated methods overcome this basis set problem by including the inter-electronic distances (rij) explicitly in wave function expansions. Here, the origin of the basis set problem of conventional wave function methods is reviewed, and a short history of explicitly correlated methods is presented. The F12 methods are the focus herein, as they are the most practical explicitly correlated methods to date. Moreover, some of the key developments in modern F12 technology, which have significantly improved the efficiency and accuracy of these methods, are also reviewed. In this work, the extension of the perturbative coupled-cluster F12 method, CCSD(T)F12, developed in our group for the treatment of high-spin open-shell molecules (J. Zhang and E. F. Valeev, J. Chem. Theory Comput., 2012, 8, 3175.), is also documented. Its performance is assessed for accurate prediction of chemical reactivity. The reference data include reaction barrier heights, electronic reaction energies, atomization energies, and enthalpies of formation from the following sources: (1) the DBH24/08 database of 22 reaction barriers (Truhlar et al., J. Chem. Theory Comput., 2007, 3, 569.), (2) the HJO12 set of isogyric reaction energies (Helgaker et al., Modern Electronic Structure Theory, Wiley, Chichester, first ed., 2000.), and (3) the HEAT set of atomization energies and heats of formation (Stanton et al., J. Chem. Phys., 2004, 121, 11599.). Two types of analyses were performed, which target the two distinct uses of explicitly correlated CCSD(T) models: as a replacement for the basis-set-extrapolated CCSD(T) in highly accurate composite methods like HEAT and as a distinct model chemistry for standalone applications. Hence, (1) the basis set error of each component of the CCSD(T)F12 contribution to the chemical energy difference in question and (2) the total error of the CCSD(T)F12 model chemistry relative to the benchmark values are analyzed in detail. Two basis set families were utilized in the calculations: the standard aug-cc-p(C)VXZ (X = D, T, Q) basis sets for the conventional correlation methods and the cc-p(C)VXZ-F12 (X = D, T, Q) basis sets of Peterson and co-workers that are specifically designed for explicitly correlated methods. The conclusion is that the performance of the two families for CCSD correlation contributions (which are the only components affected by the explicitly correlated terms in our formulation) are nearly identical with triple- and quadruple-ζ quality basis sets, with some differences at the double-ζ level. Chemical accuracy (~4.18 kJ/mol) for reaction barrier heights, electronic reaction energies, atomization energies, and enthalpies of formation is attained, on average, with the aug-cc-pVDZ, aug-cc-pVTZ, cc- pCVTZ-F12/aug-cc-pCVTZ, and cc-pCVDZ-F12 basis sets, respectively, at the CCSD(T)F12 level of theory. The corresponding mean unsigned errors are 1.72 kJ/ mol, 1.5 kJ/mol, ~ 2 kJ/mol, and 2.17 kJ/mol, and the corresponding maximum unsigned errors are 4.44 kJ/mol, 3.6 kJ/mol, ~ 5 kJ/mol, and 5.75 kJ/mol. In addition to accurate energy calculations, our studies were extended to the computation of molecular properties with the MP2-F12 method, and its performance was assessed for prediction of the electric dipole and quadrupole moments of the BH, CO, H2O, and HF molecules (J. Zhang and E. F. Valeev, in preparation for submission). First, various MP2- F12 contributions to the electric dipole and quadrupole moments were analyzed. It was found that the unrelaxed one-electron density contribution is much larger than the orbital response contribution in the CABS singles correction, while both contributions are important in the MP2 correlation contribution. In contrast, the majority of the F12 correction originates from orbital response effects. In the calculations, the two basis set families, the aug-cc-pVXZ (X = D, T, Q) and cc-pVXZ-F12 (X = D, T, Q) basis sets, were also employed. The two basis set series show noticeably different performances at the double-ζ level, though the difference is smaller at triple- and quadruple-ζ levels. In general, the F12 calculations with the aug-cc- pVXZ series give better results than those with the cc-pVXZ-F12 family. In addition, the contribution of the coupling from the MP2 and F12 corrections was investigated. Although the computational cost of the F12 calculations can be significantly reduced by neglecting the coupling terms, this does increase the errors in most cases. With the MP2-F12C/aug-cc-pVDZ calculations, dipole moments close to the basis set limits can be obtained; the errors are around 0.001 a.u. For quadrupole moments, the MP2-F12C/aug-cc-pVTZ calculations can accurately approximate the MP2 basis set limits (within 0.001 a.u.). / Ph. D.

Accurate correlated electronic structure methods involve a significant amount of computations and can be only employed to small molecular systems. For example, the coupled-cluster singles, doubles, and perturbative triples model (CCSD(T)), which is known as the ``gold standard" of quantum chemistry for its accuracy, usually can treat molecules with 20-30 atoms. To extend the reach of accurate correlated electronic structure methods to larger molecular systems, we work towards two directions: parallel computing and reduced-cost/scaling approaches. Parallel computing can utilize more computational resources to handle systems that demand more substantial computational efforts. Reduced-cost/scaling approaches, which introduce approximations to the existing electronic structure methods, can significantly reduce the amount of computation and storage requirements. In this work, we introduce a new distributed-memory massively parallel implementation of standard and explicitly correlated (F12) coupled-cluster singles and doubles (CCSD) with canonical bigO{N^6} computational complexity ( C. Peng, J. A. Calvin, F. Pavov{s}evi'c, J. Zhang, and E. F. Valeev, textit{J. Phys. Chem. A} 2016, textbf{120}, 10231.), based on the TiledArray tensor framework. Excellent strong scaling is demonstrated on a multi-core shared-memory computer, a commodity distributed-memory computer, and a national-scale supercomputer. We also present a distributed-memory implementation of the density-fitting (DF) based CCSD(T) method. (C. Peng, J. A. Calvin, and E. F. Valeev, textit{in preparation for submission}) An improved parallel DF-CCSD is presented utilizing lazy evaluation for tensors with more than two unoccupied indices, which makes the DF-CCSD storage requirements always smaller than those of the non-iterative triples correction (T). Excellent strong scaling is observed on both shared-memory and distributed-memory computers equipped with conventional Intel Xeon processors and the Intel Xeon Phi (Knights Landing) processors. With the new implementation, the CCSD(T) energies can be evaluated for systems containing 200 electrons and 1000 basis functions in a few days using a small size commodity cluster, with even more massive computations possible on leadership-class computing resources. The inclusion of F12 correction to the CCSD(T) method makes it converge to basis set limit much more rapidly. The large-scale parallel explicitly correlated coupled-cluster program makes the accurate estimation of the coupled-cluster basis set limit for molecules with 20 or more atoms a routine. Thus, it can be used rigorously to test the emerging reduced-scaling coupled-cluster approaches. Moreover, we extend the pair natural orbital (PNO) approach to excited states through the equation-of-motion coupled cluster singles and doubles (EOM-CCSD) method. (C. Peng, M. C. Clement, and E. F. Valeev, textit{submitted}) We simulate the PNO-EOM-CCSD method using an existing massively parallel canonical EOM-CCSD program. We propose the use of state-averaged PNOs, which are generated from the average of the pair density of excited states, to span the PNO space of all the excited states. The doubles amplitudes in the CIS(D) method are used to compute the state-averaged pair density of excited states. The issue of incorrect states in the state-averaged pair density, caused by an energy reordering of excited states between the CIS(D) and EOM-CCSD, is resolved by simply computing more states than desired. We find that with a truncation threshold of $10^{-7}$, the truncation error for the excitation energy is already below 0.02 eV for the systems tested, while the average number of PNOs is reduced to 50-70 per pair. The accuracy of the PNO-EOM-CCSD method on local, Rydberg and charge transfer states is also investigated. / Ph. D.

Title: Study of Diradicals by Explicitly Correlated Multireference Coupled Cluster Methods Author: Matej Švaňa Department: Department of Physical and Macromelecular Chemistry Supervisor: Mgr. Jiří Pittner, Dr. rer. nat., J. Heyrovský Institute of Physical Chemistry Abstract: Total energies of cyclopropane, trimethylene, and propylidene were calculated with conventional post-HF CCSD(T), BWCCSD(T), MkCCSD(T) methods and their explicitly correlated alternatives. Main aims of the the- sis were to compare the basis set convergence of total energies and relative energies between cyclopropane and trimethylene/propylidene, both at the conventional and the explicitly correlated levels. It was shown that use of explicit correlation accelerates the convergence of the total energy by one or- der of basis set quality, resulting in considerable savings in computational times. Also, the MkCCSD(T)-F12/QZ and the BWCCSD(T)-F12/QZ calcula- tions belong to the most sophisticated approaches employed for estimation of the relative energies of cyclopropane and trimethylene/propylidene to date. Keywords: explicitly correlated, coupled cluster, multi-reference, cyclopropane isomerisation, trimethylene, propylidene 1

In the exploration of chemical systems through quantum mechanics, accurate treatment of the electron wavefunction, and the related electron density, is fundamental to extracting information concerning properties of a system. This work examines challenges in achieving accurate chemical information through manipulation of the one-electron basis set.

In this thesis, the implementation of the incremental scheme for open-shell systems with unrestricted Hartree-Fock reference wave functions is described. The implemented scheme is tested within robustness and performance with respect to the accuracy in the energy and the computation times. New approaches are discussed to implement a fully automated incremental scheme in combination with the domain-specific basis set approximation. The alpha Domain Partitioning and Template Equalization are presented to handle unrestricted wave functions for the local correlation treatment. Both orbital schemes are analyzed with a test set of structures and reactions. As a further goal, the DSBSenv orbital basis sets and auxiliary basis sets are optimized to be used as environmental basis in the domain-specific basis set approach. The performance with respect to the accuracy and computation times is analyzed with a test set of structures and reactions. In another project, a scheme for the optimization of auxiliary basis sets for uranium is presented. This scheme was used to optimize the MP2Fit auxiliary basis sets for uranium. These auxiliary basis enable density fitting in quantum chemical methods and the application of the incremental scheme for systems containing uranium. Another project was the systematical analysis of the binding energies of four water dodecamers. The incremental scheme in combination with the CCSD(T) and CCSD(T)(F12*) method were used to calculate benchmark energies for these large clusters.

In this thesis, the implementation of the incremental scheme for open-shell systems with unrestricted Hartree-Fock reference wave functions is described. The implemented scheme is tested within robustness and performance with respect to the accuracy in the energy and the computation times. New approaches are discussed to implement a fully automated incremental scheme in combination with the domain-specific basis set approximation. The alpha Domain Partitioning and Template Equalization are presented to handle unrestricted wave functions for the local correlation treatment. Both orbital schemes are analyzed with a test set of structures and reactions. As a further goal, the DSBSenv orbital basis sets and auxiliary basis sets are optimized to be used as environmental basis in the domain-specific basis set approach. The performance with respect to the accuracy and computation times is analyzed with a test set of structures and reactions. In another project, a scheme for the optimization of auxiliary basis sets for uranium is presented. This scheme was used to optimize the MP2Fit auxiliary basis sets for uranium. These auxiliary basis enable density fitting in quantum chemical methods and the application of the incremental scheme for systems containing uranium. Another project was the systematical analysis of the binding energies of four water dodecamers. The incremental scheme in combination with the CCSD(T) and CCSD(T)(F12*) method were used to calculate benchmark energies for these large clusters.

In dieser Arbeit werden Algorithmen für die Berechnung elektronischer Korrelations- und Anregungsenergien mittels der Coupled-Cluster Methode auf adaptiven Gittern entwickelt und implementiert. Die jeweiligen Funktionen und Operatoren werden adaptiv durch Multiskalenanalyse dargestellt, was eine Basissatz unabängige Beschreibung mit kontrollierter numerischer Genauigkeit ermöglicht. Gleichungen für die Coupled-Cluster Methode werden in einem verallgemeinerten Rahmen, unabhängig von virtuellen Orbitalen und globalen Basissätzen, neu formuliert. Hierzu werden die amplitudengewichteten Anregungen in virtuelle Orbitale ersetzt durch Anregungen in n-Elektronenfunktionen, welche durch Gleichungen im n-Elektronen Ortsraum bestimmt sind. Die erhaltenen Gleichungen können, analog zur Basissatz abh¨angigen Form, mit leicht angepasster Interpretation diagrammatisch dargestellt werden. Aufgrund des singulären Coulomb Potentials werden die Arbeitsgleichungen mit einem explizit korrelierten Ansatz regularisiert. Coupled-Cluster singles mit genäherten doubles (CC2) und ähnliche Modelle werden, für geschlossenschalige Systeme und in regularisierter Form, in die MADNESS Bibliothek (eine allgemeine Bibliothek zur Darstellung von Funktionen und Operatoren mittels Multiskalenanalyse) implementiert. Mit der vorgestellten Methode können elektronische CC2 Paarkorrelationsenergien und Anregungsenergien mit bestimmter numerischer Genauigkeit unabhängig von globalen Basissätzen berechnet werden, was anhand von kleinen Molekülen verifiziert wird / In this work algorithms for the computation of electronic correlation and excitation energies with the Coupled-Cluster method on adaptive grids are developed and implemented. The corresponding functions and operators are adaptively represented with multiresolution analysis allowing a basis-set independent description with controlled numerical accuracy. Equations for the coupled-cluster model are reformulated in a generalized framework independent of virtual orbitals and global basis-sets. For this, the amplitude weighted excitations into virtuals are replaced by excitations into n-electron functions which are determined by projected equations in the n-electron position space. The resulting equations can be represented diagrammatically analogous to basis-set dependent approaches with slightly adjusted rules of interpretation. Due to the singular Coulomb potential, the working equations are regularized with an explicitly correlated ansatz. Coupled-cluster singles with approximate doubles (CC2) and similar models are implemented for closed-shell systems and in regularized form into the MADNESS library (a general library for the representation of functions and operators with multiresolution analysis). With the presented approach electronic CC2 pair-correlation energies and excitation energies can be computed with definite numerical accuracy and without dependence on global basis sets, which is verified on small molecules.