Approximation Methods

and Quantum Chemistry of

Simple Molecular Systems
Physical Chemistry III
Quantum Chemistry
Variational Principle
Variational Principle says that we can
calculate an upper bound on Eo by using
any trial function we wish. The closer Ψ
(assumed value) is to Ψo in some sense,
the closer EΨ will be to Eo
Approximation Methods
A.) Variational Method

 Consider the ground state of some particular arbitrary

system. The ground state wave function Ψo and Eo satisfy the
Schrödinger equation.

H Ψo = Eo Ψo

i.) Multiply by Ψo* to the left and to the right side and integrate over
all space
 Substituting any other function for Ψo , and calculating the new
energy, say Eφ

Eφ = ∫ Ψo* H Ψo / ∫ Ψo* Ψo

 If the calculated Eφ ≥ Eo = Variational Principle

Variational Method
ii.) we can choose φ such that it depends on
some arbitrary parameters α,β,γ….. called
variational parameters. The energy Eφ
will depend on these.

iii.) Now we can minimize Eφ with

respect to each of the variational
parameter and thus approach the exact
ground –state energy Eo.
Variational Method
 A trial function that depends Linearly on the
Variational Parameter leads to a Secular
Determinant.

 Consider a particle in 1- dimensional box, and

symmetric about x = a/2 and go to zero at
the walls.

one of the simplest functions with this symmetry is

xn (x – a)n
Variational Method
 Estimate Eo by using

Φ = c1 x (a –x) + c2x2 (a – x)2

= known as linear combination of function
 Such trial function can be written generally
as

Φ = Σcnfn ( from n =1 to N)
cn = variational parameters
fn = arbitrary known function
Φ = c1f1+ c2f2
Sample Problem
 Consider the case of a particle in a one-
dimensional box where
 Φ = c1x(a –x) + c2x2 (a –x)2

For simplicity let a = 1, which amounts to

measuring all distances in units of a.

Define : H11 = ħ2/6m S11 = 1/30

H12 = ħ2 /30m S12 = S21 = 1/140
H22 = ħ2 /105 m S22 = 1/ 630
Sample Problem
1.) Show explicitly that H12 = H21

f1 = x (1-x) f2 = x2(1- x)2

H12 = ∫(0 to 1) f1Ĥf2dx

= ∫(0 to 1)x(1 –x)[-ħ2/2m d2/dx2] x2(1-
x)2]dx
= ħ2/30 m
Sample Problem
1.) Use the trial wave function
ϕ = cos ax
where a = constant and –π/2a ≤ x ≤ π/2a
to calculate the energy of a simple harmonic
oscillator.
Does the result illustrate the variation theorem?

Ĥ = -ħ2/ 2μ (d2/dx2) + kx2/2

Perturbation Theory
 Applied to a system that cannot be solved exactly
 e.g He atom and an anharmonic oscillator.
 Morse Potential energy

 V = hcDe {1-exp [-a (R – Re)]2}

 where De = depth of the potential minimum
a = (μ/2hcDe)1/2 w

Helium atom

 If the inter electronic repulsion term is not included, the

Helium atomic Ψ would be products of H like atomic wave
functions
Perturbation Theory
 In perturbation theory, the Hamiltonian
consists of two parts one for which the
Schrödinger equation can be solved
exactly (unperturbed Hamiltonian) and
an additional term (perturbation) whose
presence prevents an exact solution.
Sample Problem
 1.) Consider particle in a box with a slanted
bottom
V(x) = ∞ x <0
V(x) = Vo (x/a) 0 ≤ X ≤ a
V(x) = ∞ x >0
Ĥ(1) = Vo (x/a) = perturbed
Ψo = (2/a)1/2 sin nπx/a = flat bottomed as the
unperturbed system
Eo = n2h2/ 8ma2
Ĥ = Ĥ(o) + Ĥ(1) = -ħ2/2m d2/dx2 + Vo (x/a)
Sample Problem
2. The Hamiltonian for a quartic anharmonic
oscillator
Ĥ = -ħ2/2μ d2/dx2 + ½ kx2 + 1/6γx3 +
1/24 bx4
Where : Ĥ(o) = -ħ2/2μ d2/dx2 + ½ kx2
Ψv(o) = [(α/π)1/2 1/2v v!]1/2 Hv
(α1/2x) e-αx2/2
α = (kμ/ħ2)1/2
Ev(o) = (v + ½ ) hv
For ground state : Ψv(o) = [(α/π)1/4 e-αx2/2
Atomic Units
 Atomic and Molecular calculations are expressed
in atomic units

Schrödinger Equation for Helium atom

 One atomic unit of energy is called a hartree (H)

 One atomic unit of length is called a Bohr
Sample Problem
1.)One hartree, the unit of energy in
atomic units is given by
Hartree =me4/ ko2ħ2
Express one hartree in units of joules (j)
wave number and electron volts (eV).

1 hartree = 4.3595 x 10-18J

= 2.195 x 105 cm-1
= 27.21eV
Approximate Orbitals
(For many electron System)
Slater Orbitals
J.C. Slater (1930)
 proposed a set of rules for taking into account the
influence of shielding
Assumption
 the angular wave functions derived from the exact
solution of the H –atom Schrödinger equation were
preserved, but the radial wave function were replaced by
a new set
Rs (r) = Nnrn-1e-ζr/ao
where: Nn = normalization constant for each orbital
ζ = quantity which takes screening into account
Slater Orbitals
 Limitations

i.) STO are not good approximations close to the nucleus, but
they improve as r increases
ii.) the difference in the orbitals is the replacement of the
exponential factor e-Zr/nao by the Slater factor e-ζr/ao. If the
electron is shielded by the other electrons, it feels only a
fraction of the total charge at the nucleus.

ζ = Zeff/n (Slater’s zeta value)

where: n = principal quantum number
 The computer method of calculating zeta value (ζ) is the
method of the Self –Consistent Field (SCF)- Hartree -
Fock
Hartree – Fock SCF Orbitals
Iterative method for solving the many-electron
Schrödinger equation
 developed by D.R. Hartree (1928)
General Strategy of SCF
1. Form an initial guess for the molecular orbital
coefficients.
2. Construct the density matrix and Fock matrix
(represents the average effects of the field of all the
electrons on each orbital).
3. Solve for the density matrix and test for
convergence.
At convergence, the energy is at a minimum,
and the orbitals generate a field which
produces the same orbitals.
General Strategy of SCF
 Four electrons in a system a,b,c,and d

i.) guess their initial wavefunctions

Ψ1(a), Ψ1(b), Ψ1(c), Ψ1(d)
ii.) single out electron a and calculate the effective
field of b,c and d
iii.) after solving the Schrödinger equation by
numerical methods we obtain an improved
wavefunction, Ψ2(a)
iv.) choose electron b and calculate the effective
field Ψ2(a),, Ψ1(c), Ψ1(d)
v.) Solve the Schrodinger equation to obtain an
improved wavefunction Ψ2(b)
Electron has an Intrinsic Spin Angular
Momentum
 Electrons were spinning on their axes

it possess spin angular momentum or just

spin for short

The magnitude of the spin was determined

by the spin quantum number, ms, whose
value was restricted to be ±1/2.

 For the orbital angular momentum, L,

the total
L = ħ[(l(l+1)]1/2
Electron has an Intrinsic spin angular
momentum
 Spin angular momentum, S is given by
S = ħ [(S (S+1)]1/2
 Spin is an important quantity, so the wave
function of an electron govern by spin is
called spin eigenfunction.
α = spin wavefunction for an electron with
ms = +1/2
β = spin wavefunction for an electron with
ms = -1/2
Antisymmetric wavefunction
 ∫α* α ds = ∫β* β ds = 1
 ∫α* β ds = ∫ α β* ds =0

where ds = spin variable

Spin wave function must be anti

symmetric in the interchange of any two
electrons
Pauli Exclusion Principle
 No two electrons in an atom can have
the same values of all four quantum
numbers n,l, ml, and ms.

 The antisymmetric wave function

e.g. For He atom
Ψ
(1,2) = 1sα(1) 1sβ(2)
 where: 1sα = (Ψ
1001/2)

1sβ = (Ψ100-1/2)
Pauli Exclusion Principle
Because electrons are indistinguishable
from each other
Ψ
(2,1) = 1sα(2) 1sβ(1)

Linear combination of these gives

Ψ
(1) = Ψ1(1,2) + Ψ1(2,1) = 1sα(1) 1sβ(2) +
1sα(2)1sβ(1)
Ψ
(2) = Ψ2(1,2) – Ψ2(2,1) = 1sα(1) 1sβ(2) -
1sα(2) 1sβ(1)
Pauli Exclusion Principle
 Both the wave functions Ψ1 and Ψ2 appear
to be acceptable wave functions for the
ground state of helium atom, but
experimentally only the Ψ2 describes the
ground state of He.
 Ψ2 has the property that it changes sign
when the two electrons are interchanged
Pauli Exclusion Principle
 Ψ2(1,2) = Ψ(1,2) – Ψ(2,1)

 Ψ2(2,1) = Ψ(2,1) – Ψ(1,2)

= - Ψ2(1,2)
Ψ2(1,2) = is antisymmetric under the
influence of two electrons
Pauli Exclusion principle
 Symmetric wavefunctions are not acceptable as
electronic wave function = Pauli Exclusion
Principle

 Anti symmetric wave functions can be

represented by Slater Determinants
Slater Determinants
 can also be written in determinants to construct
antisymmetric wave function

Ψ(1,2) = /1sα (1) 1s β (1) /

/ 1sα (2) 1s β(2) /
Called a Slater determinant wave function

Ψ2 = 1sα (1) 1sβ(2) – 1sα (2) 1sβ(1)

 can be factored into a spatial part and a spin part

Ψ2 = 1s(r1) 1s(r2) [α (σ1)β(σ2) – α (σ2) β(σ1)]

Normalized N -electron Determinantal
wave function
Ψ = 1/ (2)1/2 /1sα (1) 1s β (1) /
/1sα (2) 1s β(2) /

For N –electron determinantal wave function

Ψ(1,2….N) = 1/(N!)1/2 / u1(1)u2(1)…. uN(1)/

/ u1 (2) u2(2)…uN(2)/
/ u1 (3) u2(3)…uN(3)/
Normalized N –electron Determinantal
wave function
 For Lithium atom
 Ψ = 1/(3!)1/2/1sα(1)1sβ(1)2sα(1)/

/1sα(2)1sβ(2)2sα(2)/
/1sα(3)1sβ(3)2sα(3)/
First excited State of Helium
 The wavefunction for the first excited state of helium
may be approximated by the product of 1s and 2s.(2e-)

 Ψa = 2-1/2 [1s(1) 2s(2) + 1s (2)2s(1)] symmetrric

 Ψb = 2-1/2 [1s(1) 2s(2) - 1s (2)2s(1)] antisymmetric

 Now incorporate spin into the spatial wavefunction.

 Since the spatial wavefunction (for Ψb) is
antisymmetric, it must be multiplied by symmetric
spins.
First excited State of Helium
 The wavefunctions of the first excited state
of Helium are

Ψb1 = 2-1/2 [1s(1)2s(2) - 1s(2)2s (1)] α(1) α (2)

Ψb2 = 2-1/2 [1s(1)2s(2) - 1s(2)2s (1)] β(1) β(2)
Ψb3 = 2-1/2 [1s(1)2s(2) - 1s(2)2s (1)] α(1) β(2)
+ α(2)β(1)]
 First excited State of Helium
 Since Ψa is symmetric, it must be multiplied by an
antisymmetric spin function to obtain an
antisymmetric total wavefunction

 Ψa1 = 2-1/2 [1s(1)2s(2) + 1s(2) 2s(1)] α(1) β(2) - α (2) β(1)

 The first excited state Ψb is a triplet state because it has

a degeneracy of 3 due to spin. In the presence of magnetic
field the first excited state is split into three energy levels.
 The second excited state Ψa of helium is a singlet state.
Hydrogen Molecule-Ion, H2+
 Simplest diatomic molecule (H2+)
Hydrogen Molecule-Ion, H2+
 Illustration

Let
RAB = inter nuclear
distance
 rA = electronic distance
from nucleus A
 rB = electronic distance
from nucleus B
Hydrogen Molecule-Ion, H2+

The complete Hamiltonian operator is

Born-Oppenheimer Approximation
 Max Born (1882 – 1970) and J. Robert
Oppenheimer (1904 – 1967)
 The nuclei can be assumed to be
stationary relative to electronic
motion
a proton has 1838 times the mass of an
electron, thus the nuclei move much more
slowly than the electrons
Born-Oppenheimer approximation
 The Schrödinger equation for H2+
can be approximated by 2 equations;
 Variation Method
(Energy Minimization Method)

 Used to calculate electronic Ψ and the allowed energy

levels of H2+

Does not calculate a direct solution to the SE but rather

uses an approximate wavefunction called “trial
wavefunction”

Steps:
1. Guess several functions (trial wavefunctions).
Start with a trial function containing arbitrary
parameters.
2. Calculate the expectation value of the energy.
3. Minimize the energy with respect to the arbitrary
parameters.
 Variation Method
 Apply the Variation Method for H2+
Step 1: Determine the trial wave
function

 A molecular orbital (ΨMO) is a linear

combination of atomic orbitals (LCAO)
 Trial wave function for molecule → sum of
atomic orbitals
Variation Method for H2+
 Assume that the molecular orbital Ψ will be a linear
combination of hydrogen 1s atomic orbitals

 Let ΨMO → molecular orbital for H2+

A → atomic 1s orbital (Ψ100) for one H nucleus (A)
B → atomic 1s orbital (Ψ100) for the other H nucleus (B)

Trial wavefunction: ΨMO = φA + φB

 Insert arbitrary parameters (cA and cB)
ΨMO = cAφA + cB φ B
Variation Method
Step 2: Determine the expectation
value
for the energy
Variation Method
 Coulomb Integral (HAA and HBB)
 E of an e- in a 1s hydrogenic orbital plus the attractive E of
the nucleus for this e-

 Resonance Integral (HAB)

 e- is not restricted to the 1s orbital of either atom A or B
but can exchange places bet the 2 orbitals
 accounts for the calculated binding energy in H2+

 Overlap Integral (s)

 degree of overlap of the two 1s orbitals which are centered
on nuclei A and B
Variation Method
 Let
Variation Method
 Step 3: Minimize the energy

 Use Matrix Method to calculate for E

Energy of H2+
 Relationship of cA and cB

Calculate the normalization constant, cA, for

Ψ+ and Ψ-
Bonding and anti-bonding MO
Bonding and anti-bonding MO
 Electron probability density along the
line of the 2 nuclei are given by Ψ* Ψ or Ψ2
 Hydrogen Molecule-Ion, H2+

Fig. 3. Schematic diagram of orbital regions for (a) two

hydrogen atoms (b) bonding orbital in H2+. (c) antibonding
orbital in H2+
Bonding and anti-bonding MO
( Potential Energy Curve)
 Bonding Orbital – an orbital which, if
occupied, helps to bind two atoms together
 Bonding and anti-bonding MO
(Potential Energy Curve)
 Anti-bonding Orbital – an orbital that, if
occupied, contributes to a reduction in the cohesion
between two atoms and helps to raise the energy
of the molecule relative to the separated atoms.
Hydrogen Molecule, H2
Simplest stable molecule
 2 protons, 2 electrons
Complete Hamiltonian for H2
Hydrogen Molecule, H2
 The Hamiltonian operator for H2 in atomic units

+ 1/r12 ( express in atomic units)

 The approximation of neglecting the nuclear

motion ( Born – Oppenheimer Approximation )
Huckel Molecular Orbital Theory
 Huckel invented a method of building up molecular orbitals
for hydrocarbons by assuming that overlap could be
ignored

Assumptions:
 Based on the approximation that the πelectrons move in a
fixed, electrostatic potential due to the σframework of the
molecule. The σelectrons are described in terms of
localized hybrid bond orbitals and πelectrons are
described by molecular wavefunctions that extend over
each of the atoms (delocalized)

 The π – electron wavefunction is expressed as a linear

combination of px atomic orbitals.
Ethylene
 Ψπ = c1Ψp1 + c2 Ψp1
 Ĥ Ψπ = E Ψπ

 E = ∫Ψπ Ĥ Ψπ dτ/ ∫Ψπ Ψπ dτ

 Energy

E =α + β

 Ψπ = 1/√2 Ψp1 + 1/√2 Ψp2

E =α - β

 Ψπ = 1/√2 Ψp1 - 1/√2 Ψp2

Butadiene
 Ψπ = c1Ψp1 + c2 Ψp2 + c3Ψp3 + c4 Ψp4

 Energy Levels of butadiene

____ E4 α – 1.6180 β
____ E3 α – 0.6180 β
____ 2e- E2 α + 0.6180 β
____ 2e- E1 α + 1.6180 β

Normalized wavefunction for Butadiene

Ψ1 = 0.3717 x1 + 0.6015 (x2+x3) + 0.3717 x4

Ψ2 = 0.6015 x1 + 0.3717 (x2 - x3) - 0.6015 x4
Ψ3 = 0.6015 x1 - 0.3717 (x2 - x3) + 0.6015 x4
Ψ4 = 0.3717 x1 - 0.6015 (x2 + x3) - 0.3717 x4
Π – Molecular Orbitals of 1,3 -butadiene
Energy Diagram
Delocalization energy for Butadiene
 This is the energy by which butadiene is
stabilized relative to two isolated
double bonds
Charge Distribution and Bond Orders
 Coefficients in Huckel MO can be used
to calculate charge distribution and bond
orders
 Electron density (q ) at atom r
r
 q = Σn C 2
r i ir
 The sum of electron densities contributed
by each electron in each MO
 C = coefficients of atom r in the jth MO,
ir
which is occupied by ni electrons
Charge Distribution and Bond Orders
 Bond order = a rough physical
significance associated with the binding
power of a bond.

 Prs = ΣPrsj = Σ nj CjrCjs

Charge Distribution
 qr = Σni Cir2

 q1 = 2 C112 + 2 C21 2
+ 0 C312 + 0 C412
= 2 (0.3717)2 + 2 (0.6015)2 = 1.00
 q2 = 2 C122 + 2 C222 + 0 C322 + 0C422
= 2 (0.6015)2 + 2 (0.3717)2 = 1.00
 q3 = 2 C132 + 2 C232 + 0 C332 + 0C432
= 2 (0.6015)2 + 2(-0.3017)2 = 1.00
 q4= 2 C142 + 2 C242 + 0 C342 + 0C442
= 2 ( 0.3717)2 + (-0.6015)2 = 1.00
Bond Order
 Prs = ΣPrsπ = Σ nj CjrCjs

 P12= n1C11C12+n2C21C22+n3C31C32 +n4C41C42

= 2 C11C12 + 2 C21C22
= 2 (0.3717) (0.6015) + 2 (0.6015) (0.3717)
= 0.8942

P23 = n1C12C13 + n2C22C23 + n3C32C33 + n4C42C43

= 2 (0.6015) (0.6015) + 2 ( 0.3717) (-0.3717)
= 0.4473
Valence bond theory
 Describes molecule as a result of overlap
of the atomic orbital of their constituent
atoms
 This is quite clearly exemplified in
Hybridization theory.
SP hybrid Orbitals are equivalent and are
directed 180 from each other
o

 E.g BeCl2
 The two Be- H bonds in BeH2 are
equivalent and H – Be –H bond angle is
180o.
 To obtain directionality of the two Be –H
bonds, take a linear combination of the
berrylium 2s orbital and one of the
berrylium 2p orbitals.
SP hybrid orbitals
 ζ1 = a12SBe + b1 2PzBe
 ζ2 = a22SBe + b2 2PzBe

 We can appproximate the 2s and 2p orbitals

of the Be atom by the Slater orbitals.
 Ψ2s = (1/4π)1/2 R (r)

 Ψ2pz= (3/4π)1/2 R (r) cosθ

SP hybrid Orbital
 ζ1= R(r)/(4π)1/2 (a1 + (3)1/2b1cos θ)
 ζ2= R(r)/(4π)1/2 (a2 + (3)1/2b2cos θ)

 ζ1 and ζ2 are orthogonal

∫ (0 to ∞)∫ (0 to π) 2πr2 sinθdrdθ ζ1(r, θ) ζ2(r, θ) = 0

Assume that the radial function is normalized

∫ (0 to ∞ ) R2(r) r2 dr =1

∫ (0 to π) sinθdθ (a1 + (3)1/2b1cos θ) (a2 + (3)1/2b2cos θ)

= a1a2 + b1b2 = 0
 Because the two Be-H bonds in BeH2 are equivalent, we shall
require that the hybrid orbitals have the same shape.
SP Hybrid Orbital
 Thus a1 = a2 and b1 = -b2
 a12 – b12 = 0 and a = ±b
 a1 = a2; a12 + a12 = 1
a1 = 1/ (2)1/2

ζ1 = 1/(2)1/2[ 2s + 2pz]
ζ2 = 1/(2)1/2[ 2s - 2pz]

Or ζ1 = 1/(2)1/2 (2s) + b1 2pz

Because ζ1 is normalized : [(1/(2)1/2]2 + b12 =1
b1= 1/(2)1/2
ζ1 = 1/(2)1/2[ 2s + 2pz]
SP hybrid Orbital
 since ζ1 and ζ2 are orthogonal

 ζ2 =a22SBe + b2 2PzBe
 ζ2 = 1/(2)1/2[ 2s - 2pz]

 ζ1 is lying along z –axis

 ζ1= R(r)/(4π)1/2 (a1 + (3)1/2b1cos θ)

 dζ/dθ = - (3)1/2b1 sinθ = 0

sin nπ = 0
θ = 1800
SP2 Hybrid Orbitals
 Are equivalent, lie in a plane, and are directed 120o from
each other.
 e.g BH3

 Three hybrid orbitals on the Boron atom

 ζ1 = a12S + b1 2Pz + c1 2px
 ζ2 = a22S + b2 2Pz + c2 2px
 ζ3 = a32S + b32Pz + c3 2px

(Slater orbital)

 Ψ2px = (3/4π)1/2 R (r) sin θcos φ

 Ψ2s = (1/4π)1/2 R (r)
 Ψ2pz = (3/4π)1/2 R (r) cosθ
SP2Hybrid Orbital
 One hybrid orbital will lie along the z – axis (2PX)
ζ1 = a1 2S + b1 2Pz

 Take a1 = a2 = a3

 Because there is one 2s orbital that is distributed

among three hybrid orbitals

 a12 + a22 + a32 = 1

 a1 = 1/(3)1/2 = a2 = a3
 ζ1 = 1/(3)1/2 2S + b1 2Pz
SP2 Hybrid Orbital
 Because ζ1 is normalized
 1/3 + b12 = 1
 b1 = (2/3)1/2
 ζ1 = 1/(3)1/2 2S + (2/3)1/2 2Pz

 The second hybrid orbital

 ζ2 = 1/(3)1/2 2S + b2 2Pz + c2 2px
 Since ζ1 and ζ2 are orthogonal
 1/3 + b2(2/3)1/2 = 0
b2 = -1/(6)1/2
 SP2 Hybrid Orbital
 ζ2 = 1/(3)1/2 2S - 1/(6)1/2 2Pz + c2 2px

 Because ζ2 is normalized

 1/3 + 1/6 + c22 = 1

c2 =1/(2)1/2

 ζ2 = 1/(3)1/2 2S - 1/(6)1/2 2Pz + 1/(2)1/2 2px

SP2 Hybrid Orbital
 Directional properties of ζ2 and ζ1
 ζ1 = directed along the positive z –axis along the line
θ =0
 2px orbital lies in the x –z plane φ = 0

 ζ2 = R(r)/(4π)1/2 [ ( 1/(3)1/2 –(3/6)1/2 cosθ+

(3/2)1/2 sinθ)]
 dζ2/ dθ = (3/6)1/2 sin θ+ (3/2)1/2 cos θ = 0
 Tan θ =-(3)1/2
 Θ= 120o

 Thus, the orbital ζ2 and ζ1 lie in a plane and ζ2 is

directed 120o from ζ1.
Problem Set 3
1.) Use the variational principle to estimate the ground –
state energy of a harmonic oscillator using the trial
function.
φ(x) = 1/ (1 + βx2)
and the Ĥ = - ħ2/2 μ (d2/dx2 ) + (k/2)x2
The following integrals are useful:
∫(-∞ to ∞) dx / (1+ βx2)3 = 3π/8β1/2
∫(-∞ to ∞) x2dx / (1+ βx2)4 = π/16β1/2
∫(-∞ to ∞) x2dx / (1+ βx2)3 = π/2β3/2
∫(-∞ to ∞) dx / (1+ βx2)2 = π/2β1/2

2.) Calculate the a.)normalized wavefunctions b.) Energy of

the wavefunctions c.) HOMO and LUMO energy diagram
d.) charge distribution and e.) bond order of 1,3,5,7 -
octatetraene
QUANTUM MECHANICAL
METHODS
 Ab Initio
 Semi-Empirical
Density Functional Theory
Basis Set

mathematical representation of the molecular orbital within a molecule ( the
set of atomic functions used to construct LCAO- MOs)
 restricts each electron to a particular region of space
 assigns a group of basis functions to each atom to approximate its orbitals

 Basis function = linear combination of Gaussian type atomic functions or

primitives.
Note : Gaussian-type orbitals is a modified Slater type orbitals.

e.g. ψ1sSTO( r, ξ) = S100 ( r, ξ) = (ξ3/π)1/2 e-ξr

ψ1sGF ( r, α) = G100 ( r, α) = (2α/π)3/4 e-αr2

Advantage of Gaussian type orbital over the Slater type orbital

 all the multicenter integrals are easier to evaluate.

John Pople developed a commercially available program called Gaussian 94 (the most widely used quantum
chemical computer program available today)
Examples of Gaussian Functions

 py
s

 dxy

is composed of x, y, and z

α is a constant determining the radial extent of the function

Types of the Basis Set
 1. Minimal basis sets
-contain the minimum number of basis
functions needed for each atom
 STO-3G [each STO approximated as a
linear combination of three gaussian
primitives]
Types of the Basis Set
 2. Split valence basis sets
-contain two (or more) sizes of basis function
for each valence orbital

 6-31G [each inner shell STO represented by a

sum of six gaussians and each valence shell STO
split into inner and outer parts (i.e., double-ζ)
described by three and one gaussian primitives,
respectively]
Types of the Basis Set
3. Polarized basis sets
- contain orbitals with angular momentum
beyond what is required for the ground
state of each atom

 6-31G* [the 6-31G basis set augmented

with six d-type gaussian primitives on
each heavy (Z>2) atom, to permit
polarization
Types of the Basis Set
4. Basis sets with diffuse functions
-contain large-size versions of s- and p-type
functions allowing orbitals to occupy a larger
region of space.

 6-31+G* [the 6-31G* basis set augmented with

a set of diffuse s- and p-type gaussian functions
on each heavy atom in order to permit
representation of diffuse electronic distribution,
as in anions]
Ab Initio Calculations
 (“from the beginning”)

 calculations based entirely on theory

 use mathematical transformation and

approximation techniques

 provide high quality quantitative predictions for a

broad range of systems

 require more expensive computing resources

Semi-Empirical Methods
 based on Hartree-Fock formalism but
use many approximations and empirical
parametrization

 much faster than ab initio BUT highly

dependent on parameters used
Semi-empirical methods in Hyperchem
 CNDO (Complete Neglect of Differential
Overlap) and INDO (Intermediate Neglect of
Differential Overlap)
 - parameters based primarily on theory to obtain
values as close as possible to Hartree-Fock
calculation

 PM3 (Parameterized Model number 3)

- primarily used for organic molecules, but is also
parameterized for many main group elements
Semi-empirical methods in Hyperchem
 MINDO/3 (Modified Intermediate
Neglect of
Differential Overlap), MNDO (Modified
Neglect of Differential Overlap), and AM1
(Austin Model 1)
 - use many parameters, such as heats of
formation and geometries of sample
molecules, to reproduce experimental
quantities
Semi-empirical methods in Hyperchem
 ZINDO/1 (Zerner's Intermediate Neglect
of Differential Overlap)
- for calculating ground state properties
such as bond lengths and bond angles
 ZINDO/S

- parameterized to reproduce electronic

transitions and other excited states
What is computational chemistry ?
 - Use of computer software as a tool in
understanding chemical reactions and
processes
Computational Chemistry Operations
 Calculation of Electronic Properties and
Processes
 Geometry Optimizations
 Frequency and Thermodynamics Calculations
 Potential Energy Surfaces
 Molecular Dynamics
 Monte Carlo Methods
 Quantitative Structure-Property Relationships
 Conformational Searching
What is HyperChem?
 - HyperChem is a versatile molecular
modeler and editor and a powerful
computational package

1. Building and displaying molecules

2. Geometry optimizations
3. Molecular reactivity investigations
4. Generation and viewing of orbitals and
electronic plots
Common methods of calculations
 Ab initio
 Calculations that start only from the atomic
numbers of the elements and mass and charge of
an electron

 Semi-empirical
 Calculations that make use of empirical data (like
IE) instead of evaluating all molecular integrals
numerically

 Density Functional Theory (DFT)

- Directly computes for the total electron density of
the system to calculate the energy of the system
Density Functional Theory
The Hohenberg-Kohn Theorem
 the ground state properties of a many-electron
system are uniquely determined by an electron
density (ρ) that depends on only 3 spatial
Coordinates

 reduces the many-body problem of N

electrons with 3N spatial coordinates to 3
spatial coordinates, through the use of
functionals of ρ
Density Functional Theory
 based upon a strategy of modeling electron
correlation via general functionals of ρ
approximate functionals
partition the electronic energy into several terms:
E = ET + EV + EJ + EXC
ET = KE term (due to motion of the electrons)
EV = PE term (Nu-e attraction and Nu-Nu
repulsion)
EJ = electron-electron repulsion term
EXC = exchange-correlation term
Density Functional Theory
 EXC = exchange-correlation energy determined
entirely by (is a functional of) ρ

EXC (ρ) = EX (ρ) + EC (ρ)

 EX (ρ) prevents two parallel-spin electrons

from being found at the same point in space

 EC (ρ) correlation between the spatial

position of electrons due to their Coulomb
repulsion (absent in Hartree-Fock SCF)
Density Functional Theory
  Hartree-Fock energy is always above
the exact solution of the non-relativistic
SE within the Born-Oppenheimer
approximation