Lecture 6: Many Particle Applications

Topics to be covered:

• Identical particles
• Many electron atoms
• Electrons in crystalline solids
• Fermi energy
• Degeneracy pressure
• Band structure
• (Slides 30-32 not examinable)

1
 Identical Particles

In a multi-electron atom how many electrons can share the same

state?
The answer is given by the Pauli exclusion principle:

‘No two electrons in the same atom can be in the same quantum
state’

Pauli’s principle depends on the fact that all electrons are identical.

To discuss multi-electron atoms and condensed matter we need to

know how quantum theory deals with identical particles.

2
 A collision between two identical particles

• Outcomes are classically distinguishable (can label and

track particles)
• Quantum mechanics cannot track particles so collision
outcome is indeterminate

1’
1’
2 2

1
2’ 1
2’
3
 A system of two identical spinless particles
The amplitude to find one at x and the other at x  is   x , x  
  x , x  is also the amplitude to find the particles at x , x
Does it follow that   x, x      x , x  ?
No guarentee as  is not observable
We can expect that:
  x , x      x , x     x , x    exp  i    x , x 
2 2

 exchanging arguments increments the phase by 

a second exchange 
  x , x    exp  2i    x, x  
This  exp  2i   1  exp  i   1
 so   n 
4
 A system of two identical spinless particles
As exp  i n   cos   n   i sin   n    1
n

and
  x, x    exp  i    x , x  with   n

It follows that:
The symmetry of  under the exchange of identical particles
obeys the rule:

n even    x , x      x , x    is symmetric
n odd    x , x     x , x    is anti-symmetric

How do we decide which exchange symmetry holds? 5

 Wave function of two identical spin-s particles
Including spin, particle exchange 
  ms , x, ms , x   exp  i    ms , x, ms , x  
Swapping a second time leads to
  ms , x, ms , x    exp  2i    ms , x, ms , x      n
In this case it transpies that:
   ms , x, ms , x  n even (for bosons) 
  ms , x, ms , x     
  s
 m  , x , ms , x  n odd (for fermions) 
bosons: particles with integer spin  s  0,1, 2,3,....
 1 3 
fermions: particles with half-integer spin  s  , ,.... 
 2 2 
6
 Fermions and Bosons (elementary & composite)
Fermions Bosons
Electron (1/2) Pi meson (0)
Proton (1/2) Alpha particle (0)
Neutron (1/2) He4 (0)
He3 (1/2) Photon (1)

Exchange symmetry and spin are linked by the spin statistics

theorem, a result from relativistic quantum field theory.
Exchange symmetry extends to any pair of particles in a system
of N identical particles.
7
 The Pauli Exclusion Principle

For same state fermions, at the same location,

  ms , x, ms , x     ms , x, ms , x     ms , x, ms , x   0

Pauli exclusion principle:

Two independent fermions in the same system
cannot be in the same state

Bosons can happily inhabit the same state.

8
 The Helium atom
The CM system Hamiltonian (ignoring spin-orbit coupling)

Hˆ  (1, 2)  E (1, 2)

2
Hˆ  

2
 1
2
  2  V
2

e2  Z Z   e2 
V     
4 0  r1 r2   4 0 r12 2 

By ignoring the Coulomb interaction, we can solve for

each electron independently of the other. 9
The Helium atom – lowest three energy levels

E1S
0.8 eV
E1A

20 eV

E0S

E0S a singlet state (antisymmetric) at -78.9 eV

E1A a triplet state (symmetric)
10
 Possible state wave functions for the He atom
 1 
  1 1 2 2 1 2 2 1
u ( r )  u ( r )  u ( r )  u ( r )  (1)   (2) 
 2 
 1 1 
  1 1 2 2 1 2 2 1 
u ( r )  u ( r )  u ( r )  u ( r )  (1)   (2)   (2)   (1) 
 2 2 
 1 
  u1 (r1 )  u2 (r2 )  u1 (r2 )  u2 (r1 )   (1)   (2) 
 2 
A triplet with spin S  1
1 1
 u1 (r1 )  u2 (r2 )  u1 (r2 )  u2 (r1 )    (1)   (2)   (2)   (1) 
2 2
A singlet with spin S  0
If u1  u2 only the singlet is allowed by the Pauli principle
   spin up,   spin down 
11
 Example: 1-D box
Consider two non-interacting spin ½ particles of
mass m confined to a 1-D box, 0<x<L so that En =
n2 E1 .
What are the four lowest energy levels?

1, 1  E  2 E1 (1 state)
n 2 2  2 1, 2  E  5 E1 (4 states)
En 
2mL2 2, 2  E  8E1 (1 state)
1, 3  E  10 E1 (4 states)

12
 Central field models for multi-electron atoms
In a multi-electron atom we must account for the following :
• The Pauli exclusion principle
• Mutual electron repulsion.
• The partial screening of the nuclear charge by inner electrons.
These interactions cannot be treated exactly.

Central field models provide a useful approximate treatment.

They assume that each electron moves in some average
spherically symmetric effective potential, with limiting values:

lim Ze 2 lim e2
Veff  r    , Veff  r   
r 0 4 0 r r  4 0 r

13
 Central Field Model & Electron Screening

Outer electrons

Inner electrons

Nucleus

14
 Effective atomic number
Z eff e 2
Look for a Z eff to fit Veff  
4 0 r
Z eff  Z  no screening of an inner electron
Z eff  1  total screening of an outer electron
1

Thomas Fermi model (large Z model with aTF  Z 3
)
 r 
Z eff  r   Z exp   
 aTF 
Alkali model (atoms with one outer electron)
 a
Z eff  Z 0 1   , Z 0 , a : adjustable
 r
15
 Example: Zeff for the 3s electron in Sodium (Z = 11)

Sodium has the electron configuration (1s)2 (2s)2 (2p)6 (3s).

The 3s electron is the outermost electron.

The ionization energy for the 3s electron is 5.14 eV.

We use this to deduce Zeff for the 3s electron in Sodium.

Assuming the 3s electron is effectively screened by the inner

electrons we compare its ionization energy with the energy of
the n = 3 state of a hydrogen-like ion. So that,

Z eff2
E  5.14eV  13.6 2
 Z eff  1.84  11
3

16
Exercise: Discuss the properties (at small r and at large r) of
electron screening as modelled by the effective potential.

Z eff (r )e 2  r
Veff (r )  , Z eff (r )  ( Z  1) exp     1
(4 0 ) r  a

Final note: More complex, numerically self-consistent methods

are available to model multi-electron atoms: Hartree theory

17
 Electron Configurations
The specification of n and l for each atomic electron is termed the
electron configuration of that atom and defines a subshell.

The filling of subshells follows a minimum energy principle.

The Coulomb energy (of repulsion) between electrons is least

when the number of electrons with parallel spins has been
maximised. This forms the basis of Hund’s rule:

‘electrons usually fill different orbital’s with unpaired spins rather

than the same orbital with paired spins’

The configurations below illustrate this rule.

18
 19
Orbital filling and Hunds Rules
 Ionization Energies
The first ionization energy of an atom is the binding energy of
the most weakly bound electron.

The alkali metals (eg sodium) have one electron outside a filled
shell and this is easily removed because the inner electrons
provide effective screening of the nuclear charge.

Electrons in the same shell are unable to screen each other

effectively and so the ionization energy increases with Z until the
shell is full.

The highest ionization energies occur at the inert gases (filled

shells)

The lowest ionization energies occur at the alkali metals.

20
Ionization energies and the
elements

21
 Crystalline Solids: multi-electron systems

• In a crystalline solid such as sodium the outer electrons (one

from each atom) are not tied to their parent atoms.

• They can roam freely through the periodic potential of the

crystal lattice.

• These electrons can be treated as a free electron gas trapped

inside a 3-D potential box formed by the boundaries of the
solid.

• The crystal lattice structure causes the energies of the free

electrons to form into conduction bands of closely spaced levels
separated by forbidden gaps.
22
 The free electron gas
A solid like sodium has ~1023 valence electrons per cubic cm. Free
electrons in the lowest possible energy states will fill the available
states up to a maximum energy called the Fermi energy.
For free (non-interacting) electrons in a rectangular box:
  nx
2 2  2
n 2
y n 2 
 2 2
k
Enx ,ny ,nz   2  2  2  
z
, nx , ny , nz  1, 2,3,...
2me  ll l y l z  2me
lx , l y , lz  side lengths of box (V  0 inside, V   outside)
me  electron mass
k 2  k x2  k y2  k z2

We can fill each available energy state with two electrons (of
opposite spin).
23
 .
‘k-space’ and Fermi Energy
.
Each energy state can occupies an elementary volume in k-space,
out to a maximum kF that defines a Fermi surface (hence a Fermi energy)

3 3
k x k y k z   , V  volume of box
lxl y lz V 24
 Fermi Energy
Each energy state can occupies an elementary volume in k-space,
out to a maximum kF that defines a Fermi surface (hence a Fermi energy)

1  4 2  Nq   3  1
The Fermi surface:   k F      k F   3 
2 3

8 3  2 V 
Nq
 (free electron density) with q  free electrons per atom
V
The corresponding (maximum) Fermi energy is:
2
k 2 2 2 2
  3 Nq 
2 2

 
 3
EF   F
3 2 3
  
2me 2me 2me  V 
The total energy (the sum over all k-shells) is:
5
 2  3 2 Nq  3

2
Etot  V 3
10 me
2
25
 Degeneracy Pressure
The quantum mechanical Fermi energy exerts a degeneracy presure
5

dEtot  
2  2
 3 2
Nq  3

5
2
V dV   Etot
3
dV
3 10 2 me 3 V
This energy does work W on the surrounds  dW  PdV  .
So it exerts a pressure:
2

2 Etot  3   53
2 2 3

P  
3 V 5me
This pressure is the reason (in part) why a
cold solid object does not simply collapse.
(Ignoring Coulomb repulsion and thermal pressure)

26
 The Fermi Energy Example (Sodium)
Molecular weight = 23, Solid density = 0.97 [g cm 3 ]
Number of valence electrons per cm 3
N 0.97 [g cm 3 ] 1 3
 1
 6.02  10 23
[atoms mol ]  2.5  10 22
[atoms cm ]
V 23 [g mol ]
2
  3 Nq 
2 2 3
EF     3.1 eV
2me  V 
1
EF  me vF2  vF  1 106 m s 1
2
EF
TF   3.66 104 K  Na at room temperature is "cold"
kB
dP 5
Bulk modulus B  V  P  0.85 1010 Nm 2
dV 3
27
 The Fermi-Dirac Distribution Function
When dealing with the physics of. systems containing a very large
number of particles with a given total energy we have to apply
the principles of statistical mechanics to obtain probabilistic
information on how the particle energies are distributed. For
indistinguishable particles with half-integer spin (namely
fermions) the average number of particles in each energy state is
given by,
1
n E 
  E  EF  
exp   1
 k BT 
1, for E  EF
As T  0, n  E  
0, for E  EF
28
 The Fermi-Dirac Distribution Function

For most crystalline solids the vast number of

electrons are frozen in the lowest energy states.

29
 Band Structure of Crystalline Solids
The energy states of conduction electrons are dictated by the
periodic potential they experience. To formally solve the
Schrodinger equation with an accurate potential is challenging.

A simple 1-D model potential that contains the essential physics

is a 1-D comb of Dirac delta functions with amplitude α.

Realistic

Ideal

30
 The Bloch Ansatz
.
Solving for the simple periodic potential
. above is helped by the
Bloch anstaz that encodes the translational symmetry of the
potential.
The probability of finding the electron in each of the regions
between the barriers is invariant under translation.

  x  a  e   x    x  a    x
iKa 2 2

To deal with the ends of the 1-D crystal we impose the following
periodic condition:

  x  Na     x   eiNKa  1  NKa  2 n,
2 n
K ,  n  0, 1, 2,...
Na
31
Solution of the Schrodinger equation for a single cell
.

sin  z  2mE me a
f  z   cos  z    , z  ka, k  , 
z  2

Graph of f(z) for β=10 showing

allowed bands (shaded) separated by forbidden gaps
32
 .

The Electrical Properties of Solids

Using the Fermi-Dirac distribution we can show that: EF  Eg 2
.
Eg is the band gap between the top of the valence band and the
bottom of the conduction band.
At (absolute) temperature T the average number of electrons at
the bottom of the conduction band is:
1
n  Eg  
 Eg 
exp   1
 2 k BT 
If Eg  k BT then
 Eg 
n  Eg   exp   
 2 k B T 

33