1

Computational Chemistry: Basics to Applications.

Introduction-Quantum mechanical calculations

2
Interdisciplinary topic

2 December 2025 3
Year Laureates Contribution
Walter Kohn & John DFT & computational
1998
Pople quantum chemistry
Multiscale modeling—
2013 Karplus, Levitt & Warshel blending quantum and
classical physics
Computational protein
Baker, Hassabis &
2024 design and structure
Jumper
prediction (AI-driven)

2 December 2025 4
 What is Computational Chemistry?
➢The study, prediction, and comprehension of chemical reactions, molecular
structures, and material properties through the use of computers is known as
computational chemistry.

➢Chemists investigate molecular behavior using simulations based on the principles

of quantum mechanics, classical physics, and statistical mechanics rather
than carrying out actual experiments in a lab.

Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011).1-2 5
 Why computational chemistry?
Why Computational Chemistry?

Used by Top Research

Institutions & Industries No Safety Risks: No Explosions, No Toxic
Reduces Experimental Costs
Exposure

Describes Atomic Chemical Reactions

Saves Time: Faster Than Traditional Experiments
in Detail

Works for Systems That Are Difficult to Predicts the Future of Chemistry
6
Study Experimentally Facilitates New Material and Drug Design & Materials Science
 Chemistry on a Computer. Why?

Advantages:
➢Calculations are easy to perform whereas experiments are difficult.
➢Calculations are safe where many experiments are dangerous.
➢Calculations are becoming less costly whole experiments are becoming
more expensive.
➢Calculations can be performed on any chemical system, whereas
experiments are relatively limited.
➢Calculations give direct information whereas there is often uncertain in
interpreting experimental observation.
➢Calculations give fundamental information about isolated molecules
without the complicating solvents

Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011).1-2 7
 What Does Computational Chemistry Calculate?
Energy, structure, and properties

➢What is the energy of the given geometry?

➢How does energy vary when geometry changes?

➢Which geometries are stable?

➢How do atoms rearrange to form new molecules?

➢How do structure, energy, and properties change over time?

8
Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011)1-2
 Properties that can be calculated with computational
chemistry are:
◆Geometry Optimization

◆Stability and Reaction Mechanism

◆Transition-state structures

◆Dipole moments and polarizabilities

◆Electron and Hole Density Distribution Maps

◆Vibrational frequencies, IR and Raman Spectra

◆NMR spectra

◆Electronic excitations and UV spectra

◆Thermochemical data
9
Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011)1-2
 Merits of Computational Chemistry vs Conventional Chemistry

Advantage Computational Chemistry Explanation / Comparison

Can visualize and analyze
Many such details are not directly
Molecular-level insight electronic structures, orbitals,
observable in experiments
reaction paths, transition states
Avoids expensive chemicals, Ideal for screening molecules or
Cost-effectiveness
lab equipment, and consumables reactions before lab testing
Simulations can be faster than Especially for initial screening,
Time efficiency performing complex or slow geometry optimizations, or
experiments property predictions
No physical hazards (e.g., toxic
Important for hazardous or
Safety chemicals, high pressure,
explosive systems
radioactivity)
Can simulate molecules or E.g., drug candidates, materials
Exploration of hypothetical
materials that don't yet exist or are under extreme conditions, reactive
systems
unstable in lab intermediates
Enables high-throughput Used widely in drug discovery,
Systematic screening
virtual screening of compounds materials design, etc.
High reproducibility when using Experimental results can vary with
Data reproducibility
the same inputs and software lab conditions, purity, instruments
Simulate extreme conditions:
Often impossible or difficult in lab
Control over conditions temperature, pressure, electric
setups
field, solvent
Can predict properties (IR,
NMR, UV-Visible spectra, boiling
Predictive power Helps prioritize experimental targets
points, reaction rates) before
synthesis
Efficient modeling of proteins, Lab studies on such systems may
Works well with big
DNA, polymers, and surfaces be costly, time-consuming, or
molecules
using molecular mechanics or DFT ambiguous
 Computational Methods
Computational Methods

Quantum Mechanics Classical Mechanics

Semi-empirical Ab initio Molecular

Method methods Mechanics (MM)
Molecular
Dynamics (MD)
Density Functional
Theory (DFT)

11
 Molecular Mechanics:
◆This method treats molecules as a collection of atoms (balls) connected by bonds
that behave like springs. It uses classical mechanics (the physics of everyday
objects) to calculate the potential energy of a molecule.

◆The energy of a molecule is calculated based on how much its actual geometry
deviates from these ideal values.

◆Geometry optimization involves adjusting the atom positions until the lowest
possible energy is found.

Molecular Mechanics (MM) is a computational method used to model and predict the structure, energy, and
properties of molecules by treating atoms as classical particles (not quantum ones) and using physics-based
force fields to describe the interactions between them.

In Molecular Mechanics, molecules are modeled like a collection of balls (atoms) connected by springs (bonds).
Instead of solving the Schrödinger equation like in quantum chemistry, MM uses simple mathematical equations to
calculate the potential energy of a system based on its geometry.
12
Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011) 2-4
 Molecular Dynamics:
▪ MD simulations apply the laws of classical mechanics to simulate
the movement of atoms and molecules over time.
▪ These calculations apply the laws of motion to molecules, which
change shape or move.
▪ MM is not an electronic structure method; covalent bond-breaking
and bond-making (in contrast to conformational changes) cannot be
studied with molecular dynamics programs that use this method.
▪ For the study of chemical reactions, forces generated with
semiempirical, ab initio, or density functional methods can be used.

Molecular Dynamics (MD) is a computer simulation method used to study the physical movements of atoms and
molecules over time. It allows scientists to observe how a system of particles (like a protein in water) evolves
dynamically under known physical laws, mainly Newton’s laws of motion.

Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011) 2-4 13
 Ab Initio Calculations

◆ These methods attempt to solve the Schrödinger equation from "first principles" (ab initio
means "from the beginning" in Latin) without relying on experimental data. The Schrödinger
equation describes the behavior of electrons in a molecule.

◆Starts with only the fundamental physical constants and the atomic numbers and positions
of the atoms in the molecule.

◆The ab initio method solves the Schrödinger equation for a molecule and gives us an energy
and a wavefunction.

◆The wave function is a mathematical function that can be used to calculate the electron
distribution.

◆From the electron distribution, we can tell things like how polar the molecule is and which
parts of it are likely to be attacked by nucleophiles or by electrophiles.

Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011)2-4 14
 Semi-Empirical Calculations

❖Semiempirical calculations offer a balance between accuracy and computational speed by

simplifying ab initio methods with approximations and experimental parameters. They are
based on the Schrödinger equation but use more approximations to solve it.

❖Semiempirical methods start with the ab initio Hartree-Fock calculations but simplify them by
approximating or omitting certain integrals.

❖This plugging of experimental values into a mathematical procedure to get the best calculated
values is called parameterization (or parametrization).

❖Semiempirical calculations are slower than molecular mechanics (molecular mechanics, too,
is parameterized) but much faster than ab initio calculations.

Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011)2-4 15
 Time-Independent Schrodinger’s Equation

➢The Schrodinger equation is the basis of quantum mechanics and gives

a complete description of the electronic structure of a molecule. If the
equation could be fully solved, all information pertaining to a molecule
Rigid Rotor
could be determined.

➢Schrodinger Equation can only be solved exactly for simple systems.

Rigid Rotor, Harmonic Oscillator, Particle in a box, Hydrogen Atom.

➢For more complex systems (i.e. many electron atoms/ molecules) need
some simplifying assumptions / approximations and solve it numerically.
However, it is still possible to get accurate results.
Harmonic Oscillator
➢In general, the cost of the calculation increases with the accuracy of the
calculation and the size of the system. 16
 Time-Independent Schrodinger’s Equation
The molecular orbital theory explains chemical bonding by treating electrons in molecules as delocalized over the
entire molecule, rather than being localized between individual atoms as in valence bond theory. The bond
formation in a diatomic molecule can be understood by applying the time-independent Schrödinger equation to
the system. By solving the time-independent Schrödinger equation under the Born–Oppenheimer approximation
and using the linear combination of AO approach (leading to bonding and antibonding orbitals), molecular orbital
theory provides a quantum mechanical description of bond formation in diatomic molecules.

It explains the stability, bond order, and electronic configuration in terms of delocalized molecular orbitals, which
is more accurate for describing many molecular properties than localized bonding models.

Time-Independent Schrödinger Equation

The time-independent Schrödinger equation for a molecular system is:
෡ = 𝐸Ψ
𝐻Ψ
Where:
෡ :Hamiltonian operator of the molecule
•𝐻
•Ψ: Molecular wavefunction (describes electron behavior)
•𝐸: Total electronic energy of the molecule
Born–Oppenheimer Approximation
Due to the large difference in mass between electrons and nuclei, we can separate electronic and nuclear motion.
This allows us to fix nuclei at a certain distance and solve the electronic Schrödinger equation:
𝐻෡𝑒𝑙 Ψ𝑒𝑙 = 𝐸𝑒𝑙 Ψ𝑒𝑙
 Density Functional Calculations
➢ Density function calculation is a mathematical entity related to a function, such as ab initio and
semiempirical calculations, based on the Schrödinger equation.

➢ DFT is based on the idea that the electronic structure of a molecule is completely determined by its
electron density (the probability of finding an electron at a given point in space).

➢ Instead of calculating the wavefunction (as in ab initio methods), DFT directly calculates the electron
density.

➢ DFT uses functionals (mathematical functions of functions) to approximate the relationship between
the electron density and the energy of the molecule.

➢ Generally more accurate than Hartree-Fock theory and often comparable to more advanced ab
initio methods, but at a significantly lower computational cost.

➢ Can be applied to a wide range of molecules and properties.

Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011)2-4 18
Approaches and cost

Friction, 2018, 1–38. Chem. Rev. 2021, 121, 10, 5633–5670

2 December 2025 19
 Overview
❑Computational chemistry allows one to calculate molecular
geometries, reactivities, spectra, and other properties. It employs:
❑Molecular mechanics – based on a ball-and-springs model of
molecules
❑Ab initio methods – based on approximate solutions of the
Schrödinger equation without appeal to fitting to experiment.
❑ Semiempirical methods – based on approximate solutions of the
Schrödinger equation with appeal to fitting to experiment (i.e., using
parameterization).
❑Density functional theory (DFT) methods – based on approximate
solutions of the Schrödinger equation, bypassing the wavefunction that
is a central feature of ab initio and semiempirical methods.
❑Molecular dynamics methods study molecules in motion
21
Lewars, Errol G. "Computational chemistry." Introduction to the theory and applications of molecular and quantum mechanics 318 (2011) 5.
 Atomic and Molecular Orbitals
Reference Book: Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons,
2007. Page No. 5 – 21.
 Atomic Orbitals
Atomic orbital theory is a branch of quantum mechanics that describes the behavior of electrons within individual
atoms. It defines various energy levels and orbitals within atoms.
Properties of Atomic Orbital
Individual Atom Focus

In atomic Orbital Theory, the behavior of electrons within individual atoms. It describes the arrangement of electrons
in various atomic orbitals.

Quantum Numbers

The theory introduces quantum numbers (n, l, m, s) to describe the energy levels, shapes, orientations, and spin of
atomic orbitals.

Energy Levels

Electrons occupy these levels in accordance with the Aufbau principle, Pauli Exclusion Principle, and Hund’s Rule,
23
which govern the filling of orbitals.
Atomic Orbitals

24
 Atomic Orbitals
AOs are occupancy is determined by:
Aufbau principle

Goudsmit, S. A., Richards, Paul I Proc. Natl. Acad. Sci. 1964, 51, 664–671. 25
 Types of Orbitals
• s & p orbital

Type of s and p-orbital

26
Book: JD Lee
 d & f-Orbitals

Type of d and f-orbital

27
Book: JD Lee
 Molecular Orbital Theory
❖ The molecular orbital theory is a theory on chemical bonding developed at the beginning of
the twentieth century by F. Hund and R. S. Mulliken to describe the structure and properties
of different molecules.
❖ In Molecular Orbital Theory, the bonding between atoms is described as a combination of
their atomic orbitals.
❖ The Molecular Orbital Theory allows one to predict the distribution of electrons in a molecule,
which in turn can help predict molecular properties such as shape, magnetism, and bond
order.
❖ Molecular orbital theory revolutionized the study of chemical bonding by approximating the
states of linear combination of atomic orbitals (LCAO).
❖ These approximations are made by applying the density functional theory (DFT) or Hartree-
Fock (HF) models to the Schrödinger equation. 28
 Molecular Orbital Theory
Linear Combination of Atomic Orbitals (LCAO)
Figure. Molecular
Orbitals for the
H2 Molecule. (a) This
diagram shows the
formation of a bonding
σ1s molecular orbital for
H2 as the sum of the
wave functions (Ψ) (b)
This plot of the square of
the wave function (Ψ2)
for the bonding
σ1s molecular orbital (c)
This diagram shows the
formation of an
antibonding σ∗1s molecul
ar orbital (d) This plot of
the square of the wave
function (Ψ2)

29
 Molecular Orbital Theory
Linear Combination of Atomic Orbitals (LCAO)
❖ Molecular orbitals can generally be expressed through a linear combination of atomic orbitals.
These LCAOs are useful in the estimation of the formation of these orbitals in the bonding between
the atoms that make up a molecule.
❖ The Schrödinger equation is used to describe the electron behavior for molecular orbitals
❖ It is an approximate method for representing molecular orbitals. It’s more of a superimposition
method where constructive interference of two atomic wave functions produces a bonding molecular
orbital, whereas destructive interference produces a anti-bonding molecular orbital.
Conditions for Linear Combination of Atomic Orbitals

1. Same Energy of Combining Orbitals

The atomic orbitals combining to form molecular orbitals should have comparable energy. This
means that the 2p orbital of an atom can combine with another 2p orbital of another atom.
30
 Molecular Orbital Theory
2. Same Symmetry about Molecular Axis
❖ The combining atoms should have the same symmetry around the molecular axis for proper
combination; otherwise, the electron density will be sparse.

❖ For e.g., all the sub-orbitals of 2p have the same energy, but still, the 2pz orbital of an atom can
only combine with a 2pz orbital of another atom, but cannot combine with 2px and 2py orbitals, as
they have a different axis of symmetry.
3. Proper Overlap between Atomic Orbitals

For proper energy, the two atomic orbitals should have the same energy, and the appropriate
orientation, the atomic orbitals should have proper overlap, and the same molecular axis of symmetry.

31
 Molecular Orbitals
❖ The molecular orbitals (MOs) of a given compound are the solutions of the Schrödinger equation for only
one electron.
❖ The electrons are distributed among them according to the Aufbau and Pauli principles and, eventually,
Hund’s rules.
The MOs of a Homonuclear Diatomic Molecule
Molecules formed by the bonding of two identical elements are known as homonuclear diatomic
molecules. For example, dihydrogen (H2).
Bond order
(bonding electrons)−(antibonding electrons)
Bond order = 2

Example H2,
2−0
Bond order = 2

=1
32
1. Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No. 9,10.
 Molecular Orbitals
The following procedure is used to calculate the resulting MOs:
❖ The two nuclei are held at a certain fixed distance from each other.
❖ The time-independent Schrödinger equation (1) is written for the molecule, and integrated over all space
[equation (2)]:
ĤΨ=EΨ (1)

<Ψ|Ĥ|Ψ>=E<Ψ|Ψ> (2)
It's a way to calculate the energy (E) of the molecule.
❖ Each MO is expressed as a linear combination of atomic orbitals (LCAOs):
Ψ=c1ᵩ1+c2ᵩ2 (3)
c₁ and c₂ is coefficients that determine the contribution of each atomic orbital to the molecular orbital.
φ₁ and φ₂ is the atomic orbitals of the atoms that make up the molecule.
33
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No. 9,10.
 Molecular Orbitals
The solutions are,
𝜶+𝜷 𝜶−𝜷
E1 = and E2 = (4)
𝟏+𝑺 𝟏−𝑺
E1 and E2 are the only energies which an electron belonging to the diatomic molecule.
(ᵩ1 + ᵩ2), and Ψ2 = (ᵩ1 - ᵩ2)
1 1
Ψ1 = (5)
√2(1+S) √2(1+𝑆)

❖ Ψ1 , lies at lower energy than the isolated AOs whereas the other,
Ψ2 , is at higher energy.
❖ Ψ2 with respect to the parent atomic orbitals is greater than the
stabilization of Ψ1 , so the stability of the product will depend on the
number of its electrons.
❖ If the system contains three electrons, the two occupying Ψ1 will be
stabilized, and the other one, localized in Ψ2 , destabilized. Figure 3. The MOs of the homonuclear
❖ Ψ1 are termed bonding orbitals. diatomic A2.
❖ Ψ2 are termed antibonding orbitals. 34
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No. 10,11,12.
 Molecular Orbitals
The Coulomb Integral α
❖ To a first approximation, the Coulomb integral αA gives the
energy of an electron occupying the orbital Ψ2pA in the
isolated atom A.
❖ αA is always negative and increases in absolute value
with the atom's electronegativity.

The Resonance Integral β

❖ The absolute value of the resonance integral gives a measure of
the A1 A2 bond strength.
❖ Is a measure the strength of the bonding interaction as a result
of the overlap of orbitals Ψ2pA and Ψ2pB.

35
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No.12,13.
 Molecular Orbitals
The Overlap Integral S
Consider two overlapping orbitals ϕi and ϕj . They define four regions in space:
❖ Region 1 lies outside ϕi and ϕj , where both orbitals have small values. The product ϕi ϕj is negligible.
❖ Region 2 (enclosed by ϕi but outside ϕj ) and region 3 (enclosed by ϕj but outside ϕi ) also have
negligible values for ϕi ϕj : one component is appreciable, but the other is very small.
❖ Region 4, where both ϕi and ϕj are finite. The value of Sij comes almost exclusively from this region
where the two orbitals overlap.

ϕi ϕj

Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No.13.
36
 Molecular Orbitals
Mulliken Analysis
❖ Mulliken population analysis is a method used to calculate atomic charges, which give a rough idea of
the electronic charge distribution in the molecule.

MOs of a Heteronuclear Diatomic Molecule

❖ A heteronuclear diatomic molecule is comprised of two different atoms A and B.
❖ The two are relevant AOs. Here 𝜑A of energy αA and 𝜑B of energy αB .
(αA-E) (αB-E)- (β-ES)2 = 0 (6)
Using this approximation, it is possible to rewrite Equation (6) in the form,

37
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No.13,14,15.
 Molecular Orbitals
Combination of the two AOs ϕA and ϕB (energies αA < αB )
produces two MOs:
❖ Ψ1 has lower energy than αA.
❖ Ψ2 has higher energy than αB.
❖ The destabilization of Ψ2 with respect to αB is always larger
than the stabilization of Ψ1 with respect to αA.
❖ The bonding MO Ψ1 mainly contributes ϕA, with a small
contribution from an in-phase mixing with ϕB
❖ The antibonding orbital Ψ2 is mainly contributed by ϕB, with
Figure 2 MOs of a heteronuclear diatomic
a small out-of-phase contribution from ϕA. molecule. 𝜑A and 𝜑B are shown as s orbitals.
❖ The stabilization of Ψ1 with respect to αA and the
destabilization of Ψ2 with respect to αB increase
❖ As the αA-αB energy gap is decreases.
38
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No.15,16.
 Molecular Orbitals

How to Calculate Hückel MOs

❖ An approximation introduced by Hückel in 1931 considers only the delocalized p electrons moving in a
framework of π-bonds.
❖ There are two reasons. First, experimentalists often need only rapid `back of an envelope’ solutions,
which can be readily obtained with Hückel calculations.
❖ Second, there is a close analogy between the formalisms of Hückel and perturbation methods.
❖ Most modern Hückel programs will accept the molecular structure as the input.

39
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No. 16,21.
 Molecular Orbitals
Choosing the Parameters α and β Heteroatoms
Heteroatoms
❖ Theoreticians call any non-hydrogen atom a heavy atom, and any heavy atom other than carbon a
heteroatom.
❖ In the Hückel model, all carbon atoms are assumed to be the same. Consequently, their Coulomb and
resonance integrals never change from α and β, respectively.
αx = α + kβ (9)
βcx = hβ (10)

40
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007. Page No.21.
 Comparison Atomic Orbital Molecular Orbital
Scope Individual atoms Molecules and compounds

Formation of orbital Localized atomic orbital Linear combination of atomic

orbital

Energy Levels Discrete levels Varying energy levels

Electron Configuration Within atoms Within molecules

Chemical Bonding Atom level Molecular level

Wave Particle Duality Electron behavior Extended to molecules

Application Atomic physics, quantum Molecular chemistry, materials

chemistry science, quantum chemistry

Quantum Chemistry Less common Fundamental

Materials science Limited Significant

41
Orbital Interaction and Chemical Bonding
 p-p Orbital Overlap

Fig. p-p orbital overlap showing bonding and antibonding Book: JD Lee
Reference: [Link]
43
Science_(Brown_et_al.)/09%3A_Molecular_Geometry_and_Bonding_Theories/9.07%3A_Molecular_Orbitals
 p-p Orbital Overlap

Fig. 5. p-p orbital overlap showing bonding and antibonding Book: JD Lee
Reference: [Link]
(Brown_et_al.)/09%3A_Molecular_Geometry_and_Bonding_Theories/9.07%3A_Molecular_Orbitals 44
 s-p Orbital Overlap

Fig. 6. s-p orbital overlap showing bonding and antibonding Book: JD Lee
Reference: [Link] 45
Central_Science_(Brown_et_al.)/09%3A_Molecular_Geometry_and_Bonding_Theories/9.07%3A_Molecular_Orbitals
 Diatomic molecule

Fig. 8. Two hydrogen approaching each other

Fig. 9. Two hydrogen bond made from electron pair

46
Book: Electrons and Chemical Bonding | Page: 37-39
 Triatomic molecule

Fig. 10. Overlap of 1s orbital of hydrogen and 2s orbital of Beryllium

47
Book: Electrons and Chemical Bonding
 Trigonal planar molecule

Fig. 11. Overlap between 2p and 2s of Boron and fluorine respectively

48
Book: Electrons and Chemical Bonding
 Tetrahedral molecule

Fig. 12. Tetrahedra structure methane molecule Fig. 13. Overlap between 2s and 1s orbital of carbon
and hydrogen respectively
49
Book: Electrons and Chemical Bonding | Page: 122
 Trigonal pyramidal molecule

Fig. 15. Overlap between 2spand 1s orbital of nitrogen and hydrogen respectively
Fig. 14. Pyramidal structure NH3 molecule
50
Book: Electrons and Chemical Bonding | Page: 130-131
 Bonding in an organic molecule

Fig. 16. Structure of ethane Fig. [Link] between C-C and C-H in ethane

51
Book: Electrons and Chemical Bonding | Page: 159
 Potential energy curve/surface,
Electrostatic potentials, Geometry optimization,
Frequency
Reference book – Computational chemistry: Introduction to the theory and applications of molecular and
quantum mechanics, Fourth edition by Errol G. Lewars

52
 Potential energy curve/surface
➢ The energy an object has due to ➢ PEC = relationship between
its relative position molecule energy & geometry
➢ When the object position (position of atoms)
changes = Potential energy also ➢ Mathematical / graphical
change representation

Potential
energy curve
➢ Chemical potential energy – ➢ Uses – understanding molecular
energy stored in bonds interactions & chemical reactions
➢ In quantum chemistry context – ➢ Predicting reaction rate
energy associated with position ➢ Optimizing molecular structure
of atoms and molecules ➢ Molecular dynamics simulations 53
 Potential energy curve
Potential energy Harmonic oscillator model
curve ➢ Potential energy is parabolic function of
displacement of R from equilibrium position
➢ V (vibrational energy) = ½ k (R – Re)2
➢ Re = equilibrium position
Anharmonic oscillator model
➢ Diatomic molecule (AB) ➢ Real molecule deviate from perfect parabola
➢ Molecule = balls & bond = spring ➢ V = De (1 – e
-a(R-Re))2

➢ Bond stretch's & compress – Exceptions

potential energy change
➢ As bond length changes = PE
decrease
➢ qe = equilibrium value
➢ A molecule can never be stationary at zero potential energy 54
 Potential energy surface
➢ Surface = top layer of material / object
Why we need ?
➢ Understand how energy changes based
on atom position
➢ Identifying transition state and
➢ Potential energy surface = based on intermediates of reaction
atomic position and energy
➢ Predicting reaction rate
➢ Graph created with hills and valleys
➢ Finding most stable configuration
➢ Energy pathway

55
 Potential energy surface
2D -Potential energy
curve What if I take a water molecule?

➢ One bond length (both O-H

distance same) + 1 bond
angle = 2 parameter
➢ 2D object but needs 3D
graph (including energy)

➢ Represented by 2D graph
➢ Example – H2 gas
➢ only one bond length is available ➢ A 3D graph is a potential energy surface (PES) 56
PEC for H2
PEC for HCl
 Potential energy hypersurface
What if I take HOF molecule ?

Can not be used!

➢ Parameter
1. O-H bond length ➢ 3 parameter needs 4D graph
2. O-F bond length ➢ Called potential energy hypersurface
3. H-O-F bond angle ➢ For higher dimensions E = f(q1, q2,
➢ Potential energy q3….qn)
59
 Potential energy surface
Determining transition state
energy

60
 Potential energy surface
Determining transition state
energy ➢ Lowest energy path connecting
ozone, isoozone & transition state
➢ Stationary point = point where
surface is flat

➢ Two types
1. Energy minima
2. Saddle point

61
 Electrostatic potentials
What it is ? Why we need ?
➢ Electrostatic potential (ESP) – a ➢ Understand molecule interaction
measure of charge distribution
➢ Definition – amount of energy towards each other
needed to bring a unit positive ➢ Attraction/ repulsive nature of particular
charge (eg, proton) from infinity
to a point (P). atom
➢ Overall charge distribution in the
molecules
➢ Identification of reactive sites
➢ Visualizing electron rich and poor
ESP map of ammonia. Red – negative
charge, blue- positive charge. regions in a molecule using ESP maps 62
 Electrostatic potentials
➢ V = potential
➢ Negative ESP value – 𝑣 𝑃 𝑡𝑜𝑡 = 𝑣 𝑃 𝑛𝑢𝑐 + 𝑣 𝑃 𝑒𝑙 ➢ P = point
positive charge (eg – proton)
➢ Z = point charge
attracted to point (P) 𝑧𝐴 𝜌 𝑃 ➢ r = distance
➢ Positive ESP value - positive = ෎ −඲ ⅆ𝑟
𝑟𝑃 − 𝑟𝐴 𝑟𝑃 − 𝑟 ➢ A = nucleus
charge (eg – proton)
𝐴 ➢ ρ = electron/charge
repulsion to point (P)
density
➢ ESP of a point = effect
positive nuclei + negative
electrons
➢ Table – Electrostatic
potential is not
changing based on the
theory and basis set
63
 Geometry optimization
What it is ? Why we need ?
➢ Characterization of stationary point in PES
is geometry optimization
➢ Find minimum potential energy
➢ Stationary point can be minimum,
transition state or saddle point structure
Input structure (closer to possible stationary ➢ Accurate prediction of bond length,
structure)
bond angles and dipole moment
➢ Reaction mechanism
Computer ➢ Predict reaction energy, activation
barrier
➢ Interpreting spectroscopic data
Changes the geometry until it finds stationary point
64
 Geometry optimization
Does the initial input structure always lead to the same stationary point?

➢ No
➢ Initial structure
closer to the isomer
will give the isomer 65
 Geometry optimization
Conformer analysis –
Propane

➢ 3 stationary points
➢ Hilltop – at most maximum energy
➢ Transition state
➢ Minimum – local minimum energy
66
 Geometry optimization
➢ Creating an input structure is important to
achieve the nearest global minimum
➢ Can not be achieved in one step
➢ Each time bond length & angle changed a little
to achieve the minimum energy geometry

67
 Geometry optimization application

➢ Predict most stable structures of

molecules

➢ Helps in identifying transition state

and intermediates of chemical

reaction

➢ Designing stable drug molecules

68
 Frequency
What is it ? Why we need ?
➢ Occurrence of same event in ➢ Understand vibrational frequency of
repetitive motion molecules
➢ Measurement of the repetitive
motion = frequency ➢ Compute thermodynamic properties –
➢ Chemistry = Number of times a entropy and enthalpy calculation
molecule vibrate per second
➢ Structure validation
➢ Identification of transition state, minima and
hill top
➢ IR and Raman spectra
➢ Developing chemical reaction mechanism
69
 Frequency
➢ Geometry optimization – finds ➢ Linear molecule = 3N – 5 modes
stationary point ➢ Non-linear molecule = 3N – 6 modes
➢ Frequency – check whether it’s a ➢ N = number of atoms
transition state, minimum or
hilltop
➢ Normal vibrational modes – all
molecule move together in a
synchronized manner
➢ It reach maximum and minimum
displacements and pass through
equilibrium at the same time.
➢ Particular frequency will be given
for all the atoms
➢ Infrared spectroscopy
19.3: The Molecular Interpretation of Entropy - Chemistry LibreTexts
70
 Frequency
➢ Non-linear molecule = 3N – 6
modes
➢ N = number of atoms
➢ 3(3) – 6 = 3 normal modes

71
19.3: The Molecular Interpretation of Entropy - Chemistry LibreTexts
 Frequency calculation
Diatomic molecule Polyatomic molecule
𝟏/𝟐
𝟏 𝒌 ➢ Method of calculating vibrational
෥=
𝒗 modes from force constants and
𝟐𝝅𝒄 𝝁 atom masses
➢ C = velocity of light ➢ Force constant = second derivative of
➢ K = vibration force constant energy with respect to geometric
➢ μ = reduced mass of molecule = position
(mAmB)/(mA + mB) ⅆ𝟐 𝑬Τⅆ 𝒒𝟐 = 𝟐𝒌 Force constant
➢ mA & mB = masses of A & B
➢ Approximating harmonic ➢ High K value = more stiff the bond
oscillator + solving
Schrodinger equation
➢ But we have polyatomic
molecules 72
 Frequency calculation
Normal modes of vibrations in ammonia

1 Experimental
1 Calculated IR
IR

2
3
2

Ref - Ammonia 73
 Frequency application
How find energy minimum, transition state
and hilltop from frequency calculation ?
➢ Minimum = vibrates forever + zero negative
frequency
➢ Transition state = only one negative frequency
➢ Hill top = more than one negative frequency

Criteria for transition state…

➢ Model should be in-between reactant and
product
➢ Only one imaginary frequency
➢ Animation should be clear
➢ Energy of transition state should be higher
than reactant and product 74
 Reaction mechanisms
What is a Reaction Mechanism? Why Study Reaction Mechanisms?
•A step-by-step sequence of elementary reactions that describe Helps in predicting reactivity & selectivity.
how reactants convert into products. Useful for designing catalysts & optimizing reactions.
•Provides insight into transition states, intermediates, and Computational methods (DFT, frontier orbital theory)
energy barriers. provide quantitative insights.

What is Absolute Reactivity? Bimolecular Reactions

Definition: A chemical system must overcome an energetic Definition: A reaction involving two molecules colliding to
barrier before it can react. form products.

Factors Affecting Reactivity:

1️⃣ Energy Barrier – Required to break bonds & overcome
electron repulsions.
2️⃣ Frontier Orbital Interactions – Stabilizing interactions
lower the energy barrier.
3️⃣ Symmetry Considerations – Determines if reaction
pathways are allowed or forbidden. Example: Cycloadditions
If frontier orbital overlap is zero, the reaction is forbidden. Energy required to bring two reactants together.
Reaction is allowed if frontier orbitals (HOMO-LUMO) have
the same symmetry.
Anh, Nguyen Trong. Frontier orbitals: a practical manual. John Wiley & Sons, 2007, 48
Computational Insights On Nucleophilic Substitution
Reaction
Reference book – Bachrach, Steven M. Computational organic chemistry. John Wiley &
Sons, 2014.

76
 Nucleophilic Reaction Involving Anion
Nucleophilic reactions involving anions are organic reactions where negatively charged
species (anions) donate electron pairs to electrophilic(+) carbon atoms.

Types of Nucleophilic Reactions

SN1 & SN2 (Substitution): Addition Reactions: Addition–Elimination (at carbonyls):

→ Nucleophile attacks sp³ carbon → Nucleophile attacks sp²/sp carbon → Common in carboxylic acid
→ SN1: Bond breaks first, then forms → Seen in 1,2- & 1,4-additions derivatives
→ SN2: Bond breaks & forms → Nucleophile adds, then electrophile → Bond forms first, then leaving group
simultaneously follows departs
Nucleophilic Substitution Reaction: SN1 &SN2
o First reactions taught in organic chemistry
Why important ? o Key models for understanding reaction
mechanisms
o Predict regioselectivity, stereoselectivity, What do we learn from
reaction rates, and solvent effects them?
o Computational models give deep insight,
especially for SN2
o Most quantum calculations are in the gas phase
o SN1 is rare in gas phase due to ion pair separation
Gas phase Vs Solution:
barrier
C–X bond breaks → carbocation + anion →
strong electrostatic attraction = high energy barrier
o SN2 is preferred in computational studies
 Gas Phase SN2 Reaction
What Happens in SN2 (Gas Phase)?
o Nucleophile (Y⁻) and substrate (RX) come together to form a stable ion–dipole complex (ID1).
o The nucleophile attacks the central carbon from the backside → leaving group (X⁻) is pushed out.
o This occurs through a single transition state (TS).
o An exit ion–dipole complex (ID2) forms before final separation into products.

Potential Energy Surface (PES):

❑ Has a double-well shape:
❑ Two valleys = ID1 and ID2
❑ One peak = Transition State
❑ This shape is universal for SN2
reactions in the gas phase.

Figure. Potential energy surface for the gas phase SN2 reaction.
 Effect of Solvent on SN2
In Gas Phase:
•Reactions are very fast due to no solvent
stabilization.
•But hard to control/selectivity.

In Solution (e.g., water):

•Reaction rate can drop by up to 20 orders of
magnitude
•Reason: solvation stabilizes nucleophile
(e.g., Cl⁻), but not TS as well → higher
barrier.

Figure. Calculated HF/6-31G* gas-phase energies (dashed

curve) and potential of mean force in aqueous solution (solid
curve) for Reaction 2.
 AI-Driven Screening of
Semiconductor Materials

81
 Introduction to AI and ML
➢ Artificial intelligence - Simulation of human intelligence in machines
that are programmed
➢ Machine learning – subset of AI that enables systems to learn from
data and improve over time without being explicitly programmed
➢ Make predictions or decisions based on data
➢ Types of machine learning
1. Supervised learning – Labeled data
2. Unsupervised learning – Unlabeled data
3. Reinforcement learning – learning by rewards and penalties 82
 Application
➢ Prediction – material properties – conductivity and stability
➢ Classification – classifies huge number of materials based on the
requirement
➢ Accelerate identification of drug design
➢ Generative models to design novel compounds
➢ Reaction prediction and optimization
➢ Spectroscopy and structural analysis
➢ Predict pollutant behavior and degradation
83
 AI-Driven Screening of Semiconductor
Materials
➢ Screening and identification of optimal semiconductor material
➢ Data – Formation energy, lattice constants, density, and total
magnetization
➢ Target – Band gap (eV)
➢ Model is trained for classification, regression and clustering
➢ Hyperparameter is used to optimize the best model
➢ Top semiconductor with band gap is predicted
84
 AI-Driven Screening of Semiconductor
Materials

Figure 1. a) Feature importance plot for XGBoost model and b) Predicted top 10 materials 85