2.

2 Quantum Chemistry Principles in Computational Studies

Computational chemistry falls into two main categories: quantum chemical and classical
methods. Classical methods are highly computationally efficient and typically used for very large
molecular systems. They only consider the nuclei of a molecular system, ignoring the electronic
motions. In contrast, quantum chemical (or electronic structure) methods do take the motions of
the electrons into account—they deal with the computation of molecular electronic structures,
and are therefore much more computationally demanding than their classical counterparts. 1 To
predict the chemical properties of a molecular system, the Schrödinger equation must be solved,
yielding the wave function of a molecular system.

HѰ=𝐸Ѱ

𝐻̂ is the Hamiltonian operator; Ψ is the wave function describing the probabilistic behavior of
the particles of the system, and E is the total energy of the system. However, this equation is
unsolvable because the interactions between electrons are so complex that it’s impossible to
figure it out for molecules with more than one electron. To deal with larger systems, the Born-
Oppenheimer approximation is frequently used, which simplifies the treatment of nuclear and
electronic motions separately.

2.3 The Born–Oppenheimer Approximation

The Born–Oppenheimer approximation is a fundamental concept in quantum chemistry that

simplifies the complex interactions in molecular systems and Schrödinger equation, enabling
calculations of electronic structure at fixed nuclear geometries while treating nuclear kinetic
energy separately. In the Born–Oppenheimer approximation, the electrons are forced to remain
in a given eigenstate of the electronic Hamiltonian as the nuclear configuration changes. In other
words, the nuclei constituting the molecule evolve on a single potential energy surface and are
glued by this electronic configuration.2 Key molecular ideas like molecular geometry and
potential energy surface (PES), which explains the energy as a function of nuclei positions, and
are backed up by this approximation. The Born–Oppenheimer Approximation works very well
while managing bigger molecular systems in computational chemistry.

2.4 Electronic Structure Methods

Electronic structure methods are basic quantum mechanical tools to compute the Schrödinger
equation and molecular properties. These methods are classified into ab initio, semi-empirical,
and density functional theory (DFT) approaches. Hartree-Fock and Couple Cluster Theory are ab
initio methods which provide systematic accuracy improvements at significant computational
cost, and are entirely based on quantum principles. Semi empirical methods allow faster
calculations at a cost of loss of precision through use of empirical data. Efficiency in electronic
structure calculations based on electron density is combined with accuracy in DFT which is well
suited for complex systems. If taken together, these methods lay out a general scheme for
understanding molecular behavior.

2.41 The Hartree-Fock Method

The simplest ab initio method is the Hartree–Fock (HF) method. In the HF method each electron
only sees an average field of the other electrons; local distortions of the electron distribution are
ignored. As a result, HF ignores electron correlation, which limits the accuracy achievable with
this method.1 In this method the molecular wave function is taken as a product of combined
atomic orbitals for molecular orbitals, and each electron is assumed to occupy an orbital. It
typically captures more than 99 percent of the electronic energy of a molecule, so it is a solid
starting point for performing quantum mechanical calculations. However, HF uses an
independent-particle model, averaging electron to electron interactions without fully considering
their correlated behavior. The resulting simplification gives higher energy estimates and
compromises the accuracy of molecular property predictions. From this foundation, methods
diverge: semi-empirical techniques includes experimental data to further simplify such
calculations, while advanced efforts relies on refining electron correlation effects in order to
compute approximate exact solutions.

2.42 Electron-Correlation Method

Electron correlation energy is the difference between HF energy and exact energy of a molecule;
the latter is calculated by electron correlation methods which take into account interaction
between the motions of electrons. Now, the drawback is that the Hartree-Fock (HF) method only
accounts for electrons without spatial spin correlation and to deal with this problem several
electron-correlation techniques have been introduced, including coupled cluster (CC) theory,
configuration interaction (CI), and Møller-Plesset (MP) theory. While these methods make a
larger computational effort, they are more accurate than HF. Amongst them, the CI and CC
approaches are also useful because they can be methodically extended towards exact solutions
where Coupled Cluster with Single and Double excitations (CCSD) and perturbative Triple
excitations (T) are regarded by far the best to obtain experimentally accurate results.From the
proof of Hohenberg and Kohn that the ground state electronic energy depends only on three
spatial coordinates and is completely determined by electron density, Density Functional Theory
(DFT) is more computationally efficient and experimentally observable.
2.43 Density Functional Theory

ground state electronic energy is determined completely by the electron density 𝜌 depending
The basis for Density Functional Theory (DFT) is the proof by Hohenberg and Kohn that the

only on three spatial coordinates, making it more computationally efficient and experimentally
observable. DFT is strong because only the total density matters, and its computational cost is
similar to that of HF theory, but the results are much better. Implementation strategy has a strong
impact on the computational cost of a DFT calculation. The computational cost of a DFT
calculation depends strongly on the implementation strategy. DFT methods are also one-
dimensional just like HF methods, and increasing the size of the basis set allows a better and
better description of the KS orbitals.3 The real forte of DFT is its favorable price/performance
ratio compared with electron-correlated wave function-based methods such as Møller–Plesset
perturbation theory or coupled cluster. Thus, larger (and often more relevant) molecular systems
can be studied with sufficient accuracy, thereby expanding the predictive power inherent in
electronic structure theory. As a result, DFT is now by far the most widely used electronic
structure method.4

Density Functional Theory (DFT) functionals, central to describing the electronic structure, are
categorized based on their complexity and ability to capture electronic interactions, often
demonstrated by the concept of "Jacob's Ladder." DFT works via many types of methods, and
different type of methods are used for specific systems and problems. The simplest form is still
the Local Spin Density Approximation (LSDA) and is normally used for simple crystalline
structures and metals, assuming a uniform electron density. For more complicated systems with
non-uniform electron densities, Generalized Gradient Approximations such as BP86 and PBE are
used having advantage in predicting accurate reaction energy and other molecular properties.
B3LYP and MPW1PW91 are hybrid functionals that combine DFT methods with a percentage
of Hartree–Fock exchange that improves performance for these metals complexes as well as
reaction mechanisms. Further extension of this includes kinetic energy density terms in the Meta-
GGA methods, such as TPSS, which are suitable for systems with complicated electron
correlation. Furthermore, range separated hybrid functionals, such as CAM-B3LYP and ωB97X-
D, handles long range interactions and are therefore particularly applicable to systems involving
dispersion forces, charge transfer, or excited states. These different methods allows DFT to
properly model molecular geometries, electronic structures, reaction pathways, and material
properties, making it essential in modern computational chemistry.
2.5 Basis Sets

Basis sets are defined by the number of primitive Mathematical functions and the way these
functions are contracted to approximate the shape of atomic orbitals effectively. The molecular
orbital shapes are created by executing a linear combination of atomic orbitals (LCAO), in which
the distribution of electrons in an atom is efficiently described by these basis functions. The goal
of a basis set is to provide the best representation of the unknown molecular orbitals (or electron
density), with as small a computational cost as possible. Because different theoretical methods
and molecular properties have different basis set demands, different computer architectures and
algorithms have different efficiency requirements, and the desired accuracy varies with the
application.5

Nuclear-centered Slater or Gaussian-type functions have dominated for molecular systems, with
Gaussian functions being preferred due to the better computational efficiency. Plane-wave basis
functions are often used for extended systems, as they are naturally suited to periodic boundary
conditions; but for any reasonable number of plane waves, this necessitates the use of pseudo-
potential for representing the atomic core potential. Nuclear-centered basis functions can also be
used for periodic systems, and this treats all electrons on an equal footing. Recent developments
have investigated the use of finite-element methods where piecewise polynomials are used for
representing the orbitals.5 The kind and size of various basis sets greatly affects the degree of
computation accuracy as smaller basis sets may result in representations of the electron
distribution that are somewhat less accurate, whereas bigger basis sets allow for more accurate
results since the electron distribution can be represented with fewer limitations. However, the
expansion of basis sets comes at high computational cost, and the compromise is that we need to
balance the expansion.

2.6 Calculation Types

2.61 Geometry Optimization

Geometry optimization is a computational method with atomic positions iteratively changed to

find a geometry which minimizes the total energy of the system, therefore generating a local or
global minimum on the potential energy surface (PES). A stable arrangement is indicated by a
point on the potential energy surface (PES) that is stationary, either a minimum energy
configuration or a Transition State. During the course of the optimization, the system energy, and
gradients (Atomic Forces), are used to change the atomic coordinates at each stage of the
optimization. Then we continue this iterative process until the forces on all the atoms are below a
pre- defined level which shows convergence to a stable structure. Geometry optimization is an
important part of computational chemistry, bringing in critical information about bond lengths,
angles, and overall molecular stability which will serve as a basis for further studies in electronic
properties and reactivity.
2.62 Vibrational Frequency

Vibrational frequency analysis is a vital tool to study the behavior of molecular systems and is
used to determine the nature of the stationary point which is found using geometry optimization.
The frequencies are derived from the second derivative of the potential energy surface which
displays the stability of molecular geometry, with real frequencies showing stable minimums and
imaginary frequencies indicating transition state. In this thesis, vibrational frequency analysis
was used to confirm optimized geometries of beryllium complexes as true energy minima, and to
find potential transition states. Thermochemical properties such as zero point energy (ZPE),
enthalpy, and Gibb's free energy are also determined in this analysis.

2.63 Natural Bond Orbitals (NBOs)

Natural Bond Orbital (NBO) analysis is a multiple step method that establishes the electron
density of lone electrons, bonding orbitals, and anti-bonding orbitals on an atom. Core electrons
and lone pairs are localized on an atom in atomic orbitals (AOs).These AOs are then used to
localize the bonding and anti-bonding molecular orbitals (MOs) between the atoms, which are
hybridized orbitals. The NBO calculation shows the contribution of each atomic orbital to each
of the MOs present in the molecule.6 Additionally, the NBO analysis helps in computing electron
delocalization, veil transfer effects, and strength of donor-acceptor interactions, explaining a
better understanding of molecular stability and reactivity. In this thesis, NBO analysis was
applied to the investigation of the electronic structures of beryllium complexes with special
attention to their bonding patterns, orbital contributions and charge distributions.

2.64 TD-DFT/ UV-Vis Absorption Analysis

Time dependent density functional theory (TD-DFT) is the extended form of DFT, used in the
study of electronic excitation of molecules and to compute the UV vis absorption spectra of
beryllium metal complexes. The following technique is critical for understanding the
photophysical properties of Beryllium complexes and their applications in materials science or
catalysis.
2.65 Benchmarking Methodology

Benchmarking is a necessary measure to learn what computational methods are best and most
efficient to properly model the properties of molecular systems. In order to find the most optimal
method several techniques and basis sets are evaluated to predict electronic properties and
structural characteristics of beryllium complexes with different ligands. We tested various DFT
functionals, including B3LYP, PBE0, and M06-2X in combination with 6-31G*, ccpVTZ and
aug-cc-pVTZ basis sets, known for their applicability with light elements such as beryllium.
Considering the environment factor, polarizable continuum model (PCM) was used to model
solvent interactions. The effect of solvents with different dielectric constant on the electronic and
spectroscopic properties was analyzed for water, ethanol and tetrahydrofuran (THF).

2.7 Isotropic and Anisotropic Effects in Ligand Interactions

The isotropic and anisotropic effects are important properties that give insight into the detail of
the molecule electron distribution and directional bonding. The isotropic effect is a direction
independent interaction that can be used to understand the Beryllium complexes overall stability
and symmetric geometry. The main cause of this is the static force existing between the
surrounding ligand and the atoms of the beryllium nucleus itself. In cases where ligands have
similar electronegativity’s and are evenly distributed have a more predictable stable geometry
and bond strength. On the contrary, anisotropic effects depend on direction and are crucial to
understand pattern of reactivity and transition states of chemical reactions. They arise from the
spatial orientation of the ligands relative to the small, highly charged beryllium center, and these
effects are more significant in complexes where the ligands are asymmetrically arranged. The
characterization of these effects is important for beryllium complexes because for the first time
these effects are related, which allows a more balanced view on how to control and predict the
behavior of molecules in different chemical environments.

2.8 Metal Complexes

REFERENCES:
(1) Tanja van Mourik. First-Principles Quantum Chemistry in the Life Sciences. R. Soc. 2004, 1
(Department of Chemistry, University College London, 20 Gordon Street, London WC1H
0AJ, UK ([email protected])), 1–2. https://doi.org/10.1098/rsta.2004.1469.
(2) Basile F. E. Curchod, F. A. Chemistry without the Born–Oppenheimer Approximation.
royalsocietypublishing 2022, 1–2.
(3) Frank Jensen. Introduction to Computational Chemistry, 3rd ed.; Wiley, 2017.
(4) Marie-Pierre Gaigeo, T. van M. Density Functional Theoryacross Chemistry, Physicsand
Biology. royalsocietypublishing 2014, 1–2. http://dx.doi.org/10.1098/rsta.2012.0488.
(5) Frank Jensen. Atomic Orbital Basis Sets. In Computational Molecular Science; Wiley-
Blackwell, 2012; pp 1–2.
(6) Erica Hess. Implementation of Computational Chemistry as Part of the Physical Chemistry
Curriculum. Bridgew. State Univ. 2019, 18–19.