Read sections: 5.1 to 5.6; 5.8; 5.10; 5.11; 5.12; 5.

13

Principles of Electronic
Materials and Devices
Fourth Edition, S.O. Kasap (© McGraw-Hill Education, 2018)

Insert Photo Credit Here

 Copyright Information and Permission Part I
This Power Point presentation is a copyrighted supplemental material to the
textbook Principles of Electronic Materials and Devices, Fourth Edition, S. O.
Kasap, McGraw-Hill Education (USA), ISBN-13: 978-0078028182, ISBN-10:
0078028183 © 2018 McGraw-Hill Education. Permission is given to instructors to
use these Power Point slides in their lectures provided that the above book has
been adopted as a primary required textbook for the course. Slides may be used in
research seminars at research meetings, symposia and conferences provided that
the author, book title, and copyright information are clearly displayed under each
figure. It is unlawful to use the slides for teaching if the textbook is not a required
primary book for the course. The slides cannot be distributed in any form
whatsoever, especially on the internet, without the written permission of McGraw-
Hill Education.

Please report typos and errors directly to the author: e-mail

2
©McGraw-Hill Education. Insert Photo Credit Here
 Copyright Information and Permission: Part II
This Power Point presentation is a copyrighted supplemental material to the textbook Principles of
Electronic Materials and Devices, Fourth Edition, S. O. Kasap, McGraw-Hill Education (USA),
ISBN-13: 978-0078028182, ISBN-10: 0078028183 © 2018 McGraw-Hill Education. Slides may be
used in research seminars at research meetings, symposia and conferences provided that the
author, book title, and copyright information are clearly displayed under each figure. It is unlawful
to use the slides for teaching if the textbook is not a required primary book for the course. The
slides cannot be distributed in any form whatsoever, especially on the internet, without the written
permission of McGraw-Hill Education. It is unlawful to post these slides, or part of a slide or slides,
on the internet.

Copyright © 2018, 2006, 2002, 2000, 1997 by McGraw-Hill Education, 2 Penn Plaza, New York, NY
10121, USA. All rights reserved. This publication is protected by Copyright and permission should
be obtained from the publisher prior to any prohibited reproduction, storage in a retrieval system,
or transmission in any form or by any means, electronic, mechanical, photocopying, recording, or
likewise. For information regarding permission(s), write to: Rights and Permissions Department,
McGraw-Hill Education, 2 Penn Plaza, New York, NY 10121, USA. (Permissions)

3
©McGraw-Hill Education. Insert Photo Credit Here
 Important Note
You may use color illustrations from this Power Point in your research-
related seminars or research-related presentations at scientific or
technical meetings, symposia or conferences provided that you fully cite
the following reference under each figure

From: S.O. Kasap, Principles of Electronic Materials and Devices, Fourth

Edition, © 2018 McGraw-Hill Education

THIRD PARTY PHOTOS AND ARTWORK

All third party images such as photos and artwork have been used with
full permission and, in most case, copyright fees have been paid. If you
wish to extract and use these images in other Power Points or any other
work, you need to contact the original copyright holder.

Report errors and corrections directly to the author at e-mail

4
©McGraw-Hill Education. Insert Photo Credit Here
 Learning Outcomes
What You Will Learn in This Chapter & Why It Matters
This chapter investigates the fundamental principles of semiconductors, which are materials central to contemporary
technology. Upon completion and a solid understanding of the principles presented, you will possess a
comprehensive understanding of the fundamental operation of semiconductors and the methods by which they can
be manipulated to fabricate various devices, including transistors and solar cells. By mastering these concepts, you
will acquire a deeper appreciation for the basics of underlying science and engineering that drive our world and the
tools necessary to expand the boundaries of what is feasible in electronics and beyond.

Key Takeaways:
Explore the behavior of electrons and holes in semiconductors – One will gain insights into their movement,
interactions, and contributions to electrical conductivity.
Understand the significance of doping – Adding minute quantities of impurities (dopants) to silicon can
fundamentally alter its electrical properties.
Examine the Mass Action Law and its limitations – The mass-action principle n⋅p = ni² (where n is the electron
concentration, p represents hole concentration, and nᵢ signifies intrinsic carrier concentration) is applicable in
numerous cases; however, you will learn some of the conditions under which it may fail.
Investigate charge carrier movement mechanisms – Whether through drift resulting from electric fields or
diffusion arising from concentration gradients, these transport mechanisms are pivotal to the functioning of
semiconductor-based devices.
Consider the effects of temperature and doping levels – A comprehensive understanding of how these factors
influence the behavior of semiconductors is essential for developing more efficient technologies and materials.
Analyze junction behavior – Analyzing the distinctions between Schottky and Ohmic contacts will elucidate their
significance in semiconductor applications in electronics technology.
Building the modern world and powering innovation – Transistors, LEDs, solar cells, and microchips all rely on
these principles. Moreover, semiconductor engineering advances AI chips, quantum computing, and ultra -efficient
technology processors.

5
©McGraw-Hill Education. Insert Photo Credit Here
 Semiconductors 1

200 mm and 300 mm Si wafers.

Courtesy of SunEdison Semiconductor
(See Ch 1, p84, 4th Edition. )

6
©McGraw-Hill Education. Insert Photo Credit Here
 Semiconductors 2

GaAs ingots and wafers.

GaAs is used in high speed

electronic devices, and
optoelectronics.

Courtesy of Sumitomo Electric Industries

(See Ch 1, p2, 4th Edition.)

7
©McGraw-Hill Education. Insert Photo Credit Here
 Energy Band for Silicon

(a) A simplified two-dimensional illustration of a Si atom with four hybrid orbitals hyb. Each orbital has one electron.
(b) A simplified two-dimensional view of a region of the Si crystal showing covalent bonds.
(c) The energy band diagram at absolute zero of temperature.
Note: the energy gap can be seen beautifully from the solution of Schrodinger equation where a periodic potential
with the periodicity of the lattice is used. If one can change the periodicity spacing, the bandgap can change
(e.g. stressing the crystal!) 8
©McGraw-Hill Education. Insert Photo Credit Here
 Two-dimensional pictorial view of the Si crystal

A two-dimensional pictorial view of the Si crystal showing covalent bonds as two lines
where each line is a valence electron.
9
©McGraw-Hill Education. Insert Photo Credit Here
 Excitation across the bandgap by photoexcitation

(a) A photon with an energy greater than Eg can excite an electron from the VB to the CB.
(b) When a photon breaks a Si-Si bond, a free electron and a hole in the Si-Si bond is
created. 10
©McGraw-Hill Education. Insert Photo Credit Here
 Thermal generation of electron and holes pairs

Thermal vibrations of atoms can break bonds and thereby create electron-hole pairs.

11
©McGraw-Hill Education. Insert Photo Credit Here
 Holes are free in the valence band

A pictorial illustration of a hole in the valence band wandering around the crystal due to
the tunneling of electrons from neighboring bonds.
12
©McGraw-Hill Education. Insert Photo Credit Here
 Bands bend in an applied field

Holes are the absence of electrons; electrons in

valence band will move the same way as the
conduction band ones under the influence of E, this
gives rise to holes moving uphill (floating)

When an electric field is applied, electrons in the CB and holes in the VB can drift and
contribute to the conductivity.
(a) A simplified illustration of drift in Ex.
(b) Applied field bends the energy bands since the electrostatic PE of the electron is –eV(x)
and V(x) decreases in the direction of Ex, whereas PE increases. 13
©McGraw-Hill Education. Insert Photo Credit Here
 Electron and Hole Drift Velocities
vde = μeEx and vdh = μhEx
vde = drift velocity of the electrons, μe = electron drift mobility, Ex = applied electric field,
vdh = drift velocity of the holes, μh = hole drift mobility

Newton’ s 2nd law F = me* a and v = at where a is the acceleration, m e* is the

electron effective mass. We define 𝛕 as the mean free time (time before second
collision). Using this, we have
v = a𝛕 , then v = (F/me*)𝛕. Which results in μ proportional to 1/ m e*

Conductivity of a Semiconductor
σ = enμe + epμh
σ = conductivity, e = electronic charge, n = electron concentration in the CB, μe = electron drift
mobility, p = hole concentration in the VB, μh = hole drift mobility.
The question is how do we determine n and p?
14
©McGraw-Hill Education. Insert Photo Credit Here
 Electron and hole concentrations from density of
states and statistics

(a) Energy band diagram.

(b) Density of states (number of states per unit energy per unit volume).
(c) Fermi-Dirac probability function (probability of occupancy of a state).
(d) The product of g(E) and f (E) is the energy density of electrons in the CB (number of electrons per unit energy
per unit volume). The area under nE(E) versus E is the electron concentration. Note gCB is not the same for both
bands. One has to use a different g for the valence band ( gVB )
3/ 2

g ( E ) = (8 2 )
 me 
 2
1/ 2
Recall: From Free Electron model, we have: E 1/ 2
h  15
©McGraw-Hill Education. Insert Photo Credit Here
 Electron Concentration in CB
Key assumption: Ec- EF >> kT. That is, the number of electrons in conduction band is way less
than the number of empty states there. Thus we can use Boltzmann stats.

Under normal conditions, in pure silicon

 (Ec − E F )  at room temperature, ni = 1010/cm3

n = Nc exp −  BUT the number of silicon atoms

 kT is 1022 atoms/cm3

Roughly 1012 silicon atoms contribute

1 electron per cm3 to the conduction band!

n = electron concentration in the CB, Nc = effective density of states at the CB edge, Ec =

conduction band edge, EF = Fermi energy, k = Boltzmann constant, T = temperature

Effective Density of States at CB Edge

 2m*e kT  3 / 2
Nc = 2 
 h 2


Nc = effective density of states at the CB edge, me* = effective mass of the electron in the
CB, k = Boltzmann constant, T = temperature, h = Planck’s constant
16
©McGraw-Hill Education. Insert Photo Credit Here
 Hole Concentration in VB
 (EF − Ev ) 
p = Nv exp− 
In pure silicon at room


temperature, p = 1010/cm3
kT
Remember: if f(E) is the probability that an electron is occupying a state in the conduction band,
[1 – f(E)] is the probability of a hole occupying a state in the valence band!

p = hole concentration in the VB, Nv = effective density of states at the VB edge, EF

= Fermi energy, Ev = valence band edge, k = Boltzmann constant, T = temperature

Effective Density of States at VB Edge (confusing name!)

 2m*h kT  3 / 2
Nv = 2 

 h 2

Nv = effective density of states at the VB edge, mh* = effective mass of a hole in the
VB, k = Boltzmann constant, T = temperature, h = Planck’s constant
17
©McGraw-Hill Education. Insert Photo Credit Here
 Mass Action Law
Intrinsic Semiconductors

 Eg 
np = n = N c N v exp  −
2
i  (ni = pi for intrinsic semiconductor)

 kT 
where ni = intrinsic concentration

The np product is a constant, ni2, that depends on the material properties Nc,
Nv, Eg, and the temperature. If somehow n is increased (e.g. by doping), p must
decrease to keep np constant.

Mass action law applies in

thermal equilibrium and in
the dark (no illumination)
18
©McGraw-Hill Education. Insert Photo Credit Here
 Fermi Energy in Intrinsic Semiconductors
Using n= p, one can derive the equation for the Fermi energy:
1 1  Nc 
EFi = Ev + Eg − kT ln  
2 2  Nv 
EFi = Fermi energy in the intrinsic semiconductor,
Ev = Valence band edge,
Eg = Ec − Ev = Bandgap energy,
k = Boltzmann constant, T = temperature,
Nc = Effective density of states at the CB edge,
Nv = Effective density of states at the VB edge

1 3  m* 
EFi = Ev + Eg − kT ln  e* 
2 4  mh 
me* = electron effective mass (CB), mh* = hole effective mass (VB)

If the effective masses are equal, then the Fermi level for an intrinsic material is in the center
of the bandgap!

Normally, me* and mh* are not equal! In most cases, EFi is slightly below mid bandgap.

19
©McGraw-Hill Education. Insert Photo Credit Here
 Note that in semiconductors the Fermi energy is also where the work function is
measured with respect to the vacuum level

There are NO electrons or holes at the EFi

©McGraw-Hill Education. Insert Photo Credit Here

 Average Electron Energy in CB
3
E CB = Ec + kT
2

𝐸ത𝐶𝐵 = average energy of electrons in the CB, Ec = conduction band edge, k = Boltzmann
constant, T = temperature.
(3/2)kT is also the average kinetic energy per atom in a monatomic gas (kinetic molecular
theory) in which the gas atoms move around freely and randomly inside a container.
The electron in the CB behaves as if it were “free” with a mean kinetic energy that is
(3/2)kT and an effective mass me*.
In metals, a considerable number of electrons that can easily excited into the conduction band (no bandgap). The
number of electrons is close to the density of states, and thus, electrons do interact with each other and are not
really entirely free! We had to use quantum mechanics to find the electron energy, which turns out to be very close
to (3EF )/5:
3  5 2  kT  2 
Eav (T ) = EFO 1 +   
5  12  EFO  


21
©McGraw-Hill Education. Insert Photo Credit Here
 Table 5.1
Table 5.1 Selected typical properties of Ge, Si, InP, and GaAs at 300 K
Semi
condu Eg χ Nv ni μe (cm2 μh (cm2
ctor (eV) (eV) Nc (cm−3) (cm−3) (cm−3) V−1 s−1) V−1 s−1) me*/me mh*/me εr
Ge 0.66 4.13 1.04×1019 6.0×1018 2.3×1013 3900 1900 0.12a 0.23a 16
0.56b 0.40b
Si 1.10 4.01 2.8×1019 1.2×1019 1.0×1010 1400 450 0.26a 0.38a 11.9
1.08b 0.60b
InP 1.34 4.50 5.2×1017 1.1×1019 1.3×107 4600 190 0.079a, b 0.46a 12.6
0.58b
GaAs 1.42 4.07 4.4×1017 7.7×1018 2.1×106 8800 400 0.067a, b 0.40a 13.0
0.50b
NOTE: Ge and Si are indirect whereas InP and GaAs are direct bandgap semiconductors. Effective mass
related to conductivity (labeled a) is different than that for density of states (labeled b). In numerous
textbooks, ni is taken as 1.45×1010 cm−3 and is therefore the most widely used value of ni for Si, though the
correct value is actually 1.0×1010 cm−3. (Green, M.A., Journal of Applied Physics, 67, 2944, 1990.) (Data
combined from various sources.)

22
©McGraw-Hill Education. Insert Photo Credit Here
 n-Type Silicon 1
Typically, 10,000,000 of Si to 1 atom of As; range is 10 14 to 1018 As/cm3 of silicon (light to heavy doping range). This
corresponds 1-As to 108-Si (light doping) and 1 to 10 4 (heavy doping)!
Group 15 in the IUPAC (International Union of
Pure and Applied Chemistry) system

For moderate doping, n = 10 16/cm3 which yields

p = 10 4/cm3 (why?)

Arsenic-doped Si crystal.
The four valence electrons of As allow it to bond just like Si, but the fifth electron is left
orbiting the As site. The energy required to release the free fifth electron into the CB is
very small.
23
©McGraw-Hill Education. Insert Photo Credit Here
 n-Type Silicon 2

Note: p.n = ni2 still holds

here too!

Energy band diagram for an n-type Si doped with 1 ppm As. There are donor
energy levels just below Ec around As+ sites.
This explanation is not entirely correct but pragmatic: Nd >> ni. This means that n = Nd and “p”
decrease to accommodate this increase in electrons. This happens through recombination of some
conduction electrons with holes in the valence band, thus keeping np = ni2 valid in thermal equilibrium.
In this case, p is less than that in the intrinsic case!

However, what happens in realty is that when we dope an intrinsic semiconductor, let us say with
n-type dopant, the overall increase in electron concentration causes an upward shift in the Fermi level.
This new reality reduces the number of holes, and thus the product “np” is always constant!
24
©McGraw-Hill Education. Insert Photo Credit Here
 n-Type Conductivity
 ni 2

 = eN d e + e   h  eN d  e

 Nd  Dominated by dopant concentration!

σ = electrical conductivity

e = electronic charge

Nd = donor atom concentration in the crystal

μe = electron drift mobility, ni = intrinsic concentration,

μh = hole drift mobility

25
©McGraw-Hill Education. Insert Photo Credit Here
 Occupation Probability at a Donor
1
f d (Ed ) =  (Ed − EF ) 
1
1 + exp  
2  kT 

Due to the fact that ONLY one electron can belong to the donor site with either spin up or down!
Think of the 2 as the degeneracy of that donor state!

fd (Ed ) = probability of finding an electron in a state with

energy Ed at a donor (remember, not all donors give their
electron at low temperatures!)
Ed = energy level of donor
EF = Fermi energy
k = Boltzmann constant, T = temperature

26
©McGraw-Hill Education. Insert Photo Credit Here
 p-Type Silicon 1

Boron-doped Si crystal.
B has only three valence electrons. When it substitutes for a Si atom, one of its bonds has an
electron missing and therefore a hole, as shown in (a). The hole orbits around the B- site by the
tunneling of electrons from neighboring bonds, as shown in (b). Eventually, thermally vibrating
Si atoms provide enough energy to free the hole from the B- site into the VB, as shown.
27
©McGraw-Hill Education. Insert Photo Credit Here
 p-Type Silicon 2
1
- Energy band diagram for a p-type Si doped with 1 ppm B. f d (Ed ) =  (E(E −– EE ) ) 
1
There are acceptor energy levels Ea just above Ev around B-  kT
-
1 + exp  d a FF 

2
sites. These acceptor levels accept electrons from the VB and 4
therefore create holes in the VB.
For holes, the factor used is not 1/2 but ¼
due to heavy holes(HH) and light holes (LH).

HH
LH
Split-off band

Figure 1. Energy band diagrams and major carrier transition processes

in InP (Left) and silicon crystals (Right). Reprinted with permission from
Ref. [5]. Copyright 2010 Nature Publishing Group. https ://www.mdpi.com/2304-6732/10/2/161

Note: isp.n
- Degeneracy = ni2 still
the number holds
of different here
quantum too!
states having the same energy level! In our case, silicon is the host; for the donor state (B),
the hole in the valence band can be in 4 degenerate states: 2 light-hole states and two heavy-hole states! Remember: B donates holes to the
valence band!
- One way to rationalize this degeneracy is that when an electron leaves the valence band, a hole is created with a spin opposite to that of the
electron (we two electrons with opposite spins per energy state and two bands-HH &LH). Remember, we defined holes as the absence o
electrons!
28
©McGraw-Hill Education. Insert Photo Credit Here
 Table 5.2

Quasi-Fermi Level
Under nonequilibrium conditions, the charge behavior cannot be described by the Fermi Level
but instead by what is termed as Quasi-Fermi levels!

https://www.spie.org/news/4681-solar-cells-without-p-n-junctions

A semiconductor under illumination (e.g. , Si-solar cell) will have electron and hole generation. These concentrations
of the two different charges represent Fermi levels close to the conduction band (for electrons) and the valence
band (for holes), as shown in the figure. These levels are called Quasi-Fermi levels. 29
©McGraw-Hill Education. Insert Photo Credit Here
 Compensation Doping
Controlling the properties of a semiconductor
More donors than acceptors Nd − Na >> ni
p-type to n-type
2 2
n n
n = Nd − Na p= = i i
n Nd − Na

More acceptors than donors Na − Nd >> ni

n-type to p-type
2 2
n n
p = Na − Nd n= = i i
p Na − Nd
Recombination between electrons (from donors) and holes (from acceptors)
allows keeping np = ni2 30
©McGraw-Hill Education. Insert Photo Credit Here
 Energy bands in an applied field

Energy band diagram of an n-type

semiconductor connected to a voltage
supply of V volts.
The whole energy diagram tilts
because the electron now also has an
electrostatic potential energy and are
trying to minimize it by moving
towards the + terminal.

31
©McGraw-Hill Education. Insert Photo Credit Here
 Ionization, Extrinsic and Intrinsic Regimes
What are the charge concentrations at other than room temperature?

Nd >> ni Almost independent of T: n=Nd

n-Type semiconductor

(the ½ is from donor statistics) Dominant: thermal generation of

charges across the bandgap! p = n

Ts is the saturation temperature. At Ts, all donors are ionized (each donated
the outermost electron)
(a) Below Ts, the electron concentration is controlled by the ionization of the donors.
(b) Between Ts and Ti, the electron concentration is equal to the concentration of donors
since they would all have ionized.
(a) At high temperatures, thermally generated electrons from the VB exceed the number
of electrons from ionized donors and the semiconductor behaves as if intrinsic.

32
©McGraw-Hill Education. Insert Photo Credit Here
 Electron concentration in an n-type semiconductor

Si-Si bonds are

broken! This
contribution
dominates n in
conduction
band.

ΔE = Ec – Ed ;
where Ed is doner
energy levels

The temperature dependence of the electron concentration in an n-type semiconductor.

33
©McGraw-Hill Education. Insert Photo Credit Here
 Intrinsic concentration

The temperature dependence

of the intrinsic concentration

34
©McGraw-Hill Education. Insert Photo Credit Here
 Charge Drift Mobility (in one dimension)

vd = μE
F = -eE (For an electron, the direction of the force is opposite to E)

Recall: Newton’ s 2nd law F = me* a and v = at where a is the acceleration, m e*

is the electron effective mass. We define 𝛕 as the mean free time between
scattering events (time before second collision, i.e. between collisions). Using this,
we have
v = a𝛕 , then v = (F/me*)𝛕.
Charge drift mobility is

μ = (e 𝛕)/me*

35
©McGraw-Hill Education. Insert Photo Credit Here
 (also called phonon scattering)
Lattice-Scattering-Limited Mobility (𝛕 ∝ T—3/2)
High Temp (can be room temp)
−3 / 2
L  T Example: This mobility dominates in silicon at room temp.

μL = lattice vibration scattering limited mobility, T = temperature.

Vibrations in crystals, such as semiconductors, are quantized into phonons—similar to how light is quantized into
photons, and for electrons in solids, the energy is quantized into discrete but closely spaced levels!

Ionized Impurity Scattering Limited Mobility (𝛕 ∝ T3/2)

Low Temp
3/ 2
T
I  Example: This mobility dominates in heavily doped silicon at low temperatures

NI
μI = ionized impurity scattering limited mobility, NI = concentration of the ionized impurities (all
ionized impurities including donors and acceptors) 36
©McGraw-Hill Education. Insert Photo Credit Here
 Effective or Overall Mobility

1 1 1
= +
e I L
μe = effective drift mobility
μI = ionized impurity scattering limited mobility (decreases with
increased ionized dopant concentration)
μL = lattice vibration scattering limited mobility

Of course, this is the best situation we hope for, but it does not provide a complete picture,
especially regarding devices, and even more so with nano-sized devices! Some of the other
factors that impact mobility are, but not limited to, Surface Scattering, Alloy Disorder
Scattering due to random distribution of dopant atoms, Carrier-Carrier scattering, Neutral
Impurity and Defect Scattering, etc..
37
©McGraw-Hill Education. Insert Photo Credit Here
 Drift Mobility vs Temperature

A more complicated approach is needed

to explain the behavior of mobility as
seen by experiments on some materials.

Note: axis is for T not 1/T

Log-log plot of drift mobility versus temperature for n-type Ge and n-type Si samples.
Various donor concentrations for Si are shown. Nd are in cm-3. The upper right inset is
the simple theory for lattice limited mobility, whereas the lower left inset is the simple
theory for impurity scattering limited mobility.
38
©McGraw-Hill Education. Insert Photo Credit Here
 Drift Mobility vs Dopant Concentration

The variation of the drift mobility with dopant concentration in Si for electrons and holes at 300 K.

39
©McGraw-Hill Education. Insert Photo Credit Here
 Conductivity vs Temperature

Recall, 𝛔 = e.n.𝝁

n = Nd

Remember the exp (…)

dependence on n and thus
𝜇.

Note: 1/T

Schematic illustration of the temperature dependence of electrical conductivity

for a doped (n-type) semiconductor.
40
©McGraw-Hill Education. Insert Photo Credit Here
 Doped Semiconductors and np = ni2
We assumed that Boltzmann statistics hold true in the conduction band since there are so many
accessible unoccupied states that no two electrons can ever get close to sharing one! That is,
this law holds only if the conduction band edge is a few kT above the Fermi level.

Energy band diagrams for

(a) Intrinsic,
(b) n-type, and
(c) p-type semiconductors.
In all cases, np = ni2

The book uses EFp and EFn to indicate the Fermi level in a p-type and an n-type, respectively. Other
books use the same notation for quasi-Fermi levels only. Please do not confuse the two concepts. Here,
we have equilibrium, so no quasi-Fermi Levels are present!

n << Nc
Non-degenerate
p << Nv semiconductor

Keep in mind, in thermal equilibrium, a single Fermi level describes the distribution of both electrons and holes in the semiconductor,
regardless of whether it is doped or intrinsic. Whereas two separate quasi-Fermi levels are used to describe electrons and holes!

©McGraw-Hill Education. Insert Photo Credit Here

 Degenerate Semiconductors
(heavy doping. n > Nc , p > Nv)
Dopants are at 10 19 /cm3 and more, Must use Fermi-Dirac stats!

Valence e- of
dopant here

Degenerate n-type semiconductor. Large number of donors form a band that overlaps the CB
Degenerate p-type semiconductor. Boltzmann statistics cannot apply here, we have to use Fermi-Dirac function!

NOTE: Mass action law is no longer valid here np ≠ ni2

n = Nd or p = Na is not necessarily valid due to dopant-dopant interaction

Resistivity is proportional to T, just like in metals!

Applications: Ohmic contacts; Laser diodes, Microelectronics 42
©McGraw-Hill Education. Insert Photo Credit Here
 Direct recombination in GaAs

Direct recombination in GaAs

kcb = kvb so that momentum
conservation is satisfied
43
©McGraw-Hill Education. Insert Photo Credit Here
 Direct and Indirect Recombination
Recombination is when an electron in the conduction band falls back into the valence band and recombines
with a hole. In direct recombination, the electron and hole recombine without changing their momentum (k). Indirect
recombination requires the electron to change its momentum to recombine with a hole because the conduction band
minimum and the valence band maximum occur at different k values.

To conserve momentum, a phonon

Intervenes (or a defect, etc..) to allow
the electron to recombine!

Photons carry
negligible momentum

Good for LED applications

(a) The electron energy (E) versus electron’s crystal momentum (ħk) in a direct bandgap semiconductor. Each circle represents a
possible state, an electron wavefunction (y), a solution of Schrodinger’s equation in a crystal, with a wave vector k. These solutions fall
either into the CB or the VB; there are no solutions within the bandgap. The sketches are highly exaggerated because the circles are so
close that they form a continuous energy versus momentum behavior. (b) Energy versus crystal momentum for an indirect bandgap
semiconductor such as Si. 44
©McGraw-Hill Education. Insert Photo Credit Here
 Recombination and Trapping
Localized state below
conduction band

The “recombination center-RC” Photons also, but not often

makes up for the momentum needed

to conserve the overall momentum of
the process and allows for
recombination to happen!

RC can be an Impurity, crystal

defect, or dopant. This recombination
process is less efficient than Direct
recombination! RC are not suitable for
LEDs and Solar cells (non-radiative
recombination).
Trapping center is a temporary detention
center of conduction electrons. Can be
an impurity or crystal defect. Such centers
delay recombination! Not good for
Transistors (altering carrier concentration)!

Holes also have RC and trapping centers!

Recombination and trapping. (a) Recombination in Si via a recombination center which has a localized energy level at Er in the bandgap,
usually near the middle. (b) Trapping and detrapping of electrons by trapping centers. A trapping center has a localized energy level in
the band gap.
45
©McGraw-Hill Education. Insert Photo Credit Here
 Weak photoinjection into an n-type semiconductor

Photoexcitation:

Subscript “o” stands for thermal equilb. condition

During photoexcitation,

In general, the mass action law is not obeyed under

illumination! BUT, there are instances where we can
still use it!

Holes are minority carriers

(in valence band) Majority carrier concentration
at thermal equilb. in the dark
via thermal ionization! Note:
at such condition mass action
Low-level photoinjection into an n-type semiconductor in which Δnn < nno law is valid.

In other words, if the excess carrier concentrations generated by the photoexcitation are small compared to the majority
carrier concentration, the mass action law still holds here. 46
©McGraw-Hill Education. Insert Photo Credit Here
 Weak photogeneration in an n-type semiconductor

Assume weak illumination leads to 10% change in nno for an n-type

semiconductor doped with 5x10 16 /cm3 donor atoms:

Photogeneration

In this case, the minority charge carriers is given by:

Dark Photogeneration
np = ni2 np ≠ ni2
Low-level injection in an n-type semiconductor does not significantly affect nn
but drastically affects the minority carrier concentration pn
47
©McGraw-Hill Education. Insert Photo Credit Here
 Recombination of Excess Carriers

where Gph is rate of

photogeneration

pn is instantaneous minority
carrier concentration!

Note that Δpn = Δnn (1 electron and 1 hole

generated upon photon absorption)

- 𝝉h is minority carrier lifetime or mean recombination time. It depends on semiconductor material, crystal defects.
temperature, impurities, number density of recombination centers, etc. 𝝉h is independent of majority carrier
concentration!

- Equilib. charge concentration of holes and electron is Temp dependent only!

- Electrons recombine with holes in valence band. You can think of 𝝉h as the hole’s waiting period until an electron
recombines with it.

- Recombination centers are needed for the recombination to occur! The concentration of such centers affect 𝝉h as
well.
Illumination of an n-type semiconductor results in excess electron and hole concentrations
(departure from equilib.). After the illumination, the recombination process restores
equilibrium; the excess electrons and holes simply recombine.
48
©McGraw-Hill Education. Insert Photo Credit Here
 Photoconductivity

τh = minority carrier lifetime (mean recombination time).

1/ τh is the average probability per unit time for hole to recombine with an electron
49
©McGraw-Hill Education. Insert Photo Credit Here
 Photoconductivity Assumptions

SUMMARY OF ASSUMPTIONS
- Uniform illumination
- Material is direct bandgap semiconductor
- All incident light is transmitted into the semiconductor (No surface reflection –
antireflection coating used)
- All light transmitted into the semiconductor is absorbed
- Absorption is uniform
- Steady state operation
- -- τh, μe, μh do not depend on the carrier concentration (weak
photogeneration)
- Contacts do not block the flow of carriers

50
©McGraw-Hill Education. Insert Photo Credit Here
 Transient Photoconductivity: Uniform Photoexcitation
What is the shape of minority carries concentration for a step illumination?

dpn p
Integrate equation: = Gph − n
dt h
Illumination is switched on at time t = 0 and then off at t = toff (use these boundary conditions to solve the diff.
equation). At t=0, ∆pn = 0; At t=toff, Gph = 0.
The excess minority carrier concentration ∆pn(t) rises exponentially to its steady-state value with a time constant τh.
From toff, the excess minority carrier concentration decays exponentially to its equilibrium value with a time
constant τh. 51
©McGraw-Hill Education. Insert Photo Credit Here
 Excess Minority Carrier Concentration
dpn pn
= Gph −
dt h
Δpn = excess hole (minority carrier) concentration in n-type
t = time
Gph = rate of photogeneration (maybe depend on t)
τh = minority carrier lifetime (mean recombination time)
I is the light intensity
𝜂 is the quantum efficiency

52
©McGraw-Hill Education. Insert Photo Credit Here
 Photoconductive Detectors

PbS (lead sulfide) photoconductive detectors for the detection of IR

radiation up to 2.9 mm. They are typically used in such applications
as radiation thermometers, flame monitors, water content and food
ingredient analyzers, spectrophotometers etc. (P9217 series)
(Courtesy of Hamamatsu, Japan)
53
©McGraw-Hill Education. Insert Photo Credit Here
 DIFFUSION

https://scentuparomamarketing.com/2020/08/21/what-is-scent-diffusion/

54
©McGraw-Hill Education. Insert Photo Credit Here
 Definition of Particle Flux Density

N
=
At
Γ = particle flux density, N = number of particles crossing A in a
time interval t, A = area, t = time interval

Definition of Current Density

J = QΓ
J = electric current density, Q = charge of the particle, Γ = particle
flux
55
©McGraw-Hill Education. Insert Photo Credit Here
 Electron Diffusion
- Decreasing electron concentration in x direction (similar to
having light shining on, briefly, at end of semiconductor
then turned off)
- No electric field, thus, motion is random

Scattering happens
at each ℓ (mean free path).

(a) Arbitrary electron concentration n(x, t) profile in a semiconductor. There is a net

diffusion (flux) of electrons from higher to lower concentrations.
(b) Expanded view of two adjacent sections at xo. There are more electrons crossing xo
coming from the left (xo−ℓ) than coming from the right (xo+ℓ)
56
©McGraw-Hill Education. Insert Photo Credit Here
 Electron Diffusion Flux Density Γe

- ℓ is mean free path (distance covered until charge experiences one lattice
scattering event)
- τ is mean free time for charge between scattering occurrences
57
©McGraw-Hill Education. Insert Photo Credit Here
 Electron Diffusion Flux Density Γe
Fick’s First Law
But is very small

dn
e = − De
dx
58
©McGraw-Hill Education. Insert Photo Credit Here
 Fick’s First Law 1
dn
e = − De
dx
Γe = electron flux density, De = diffusion coefficient of electrons and measures the ease by
which electrons diffuse in the material (in a sense, similar to the meaning of drift mobility but
E is the driving force here!), dn/dx = electron concentration gradient (driving force for
diffusion of charges AND is negative as it represents decreasing concentration)

Remember: for DRIFT, the driving force is the gradient (decreasing) of the potential (the
electric field E= -dV/dx)

Electron Diffusion Current Density

dn
J D,e = −ee = eDe
dx
JD, e = electric current density due to electron diffusion, Γe = electron flux, e =
electronic charge, De = diffusion coefficient of electrons, dn/dx = electron
concentration gradient
59
©McGraw-Hill Education. Insert Photo Credit Here
 Hole Diffusion

Arbitrary hole concentration p (x, t) profile in a semiconductor.

There is a net diffusion (flux) of holes from higher to lower concentrations. There are
more holes crossing x0 coming from the left (x0−ℓ) than coming from the right (x0+ℓ).
60
©McGraw-Hill Education. Insert Photo Credit Here
 Fick’s First Law 2
dp
h = − Dh
dx
Γh = hole flux density, Dh = diffusion coefficient of holes, dp/dx = hole
concentration gradient AND is negative! Thus making the flux positive.

Hole Diffusion Current Density

dp
J D ,h = +eh = −eDh
dx
JD, h = electric current density due to hole diffusion is POSITIVE, which means
the current density is in the direction of decreased concentration, Γh = hole flux
density, e = electronic charge, Dh = diffusion coefficient of holes, dp/dx = hole
concentration gradient
61
©McGraw-Hill Education. Insert Photo Credit Here
 Drift and Diffusion in a Semiconductor

When there is an electric field and also a concentration gradient, charge carriers
move both by diffusion and drift.
62
©McGraw-Hill Education. Insert Photo Credit Here
 Total Electron Current Due to Drift and Diffusion

dn
J e = en eE x + eDe
dx
Je = electron current density due to drift and diffusion

n = electron concentration

μe = electron drift mobility (measures the ease by which charges move under E in a
medium)

Ex = electric field in the x direction

De = diffusion coefficient of electrons

dn/dx = electron concentration gradient 63

©McGraw-Hill Education. Insert Photo Credit Here
 Total Hole Current Due to Drift and Diffusion
dp
J h = ep hE x − eDh
dx
Jh = hole current density due to drift and diffusion, p = hole concentration, μh = hole
drift mobility, Ex = electric field in the x direction, Dh = diffusion coefficient of holes,
dp/dx = hole concentration gradient

Einstein Relation
De kT Dh kT
= = Makes sense: increasing the temperature (within reasonable

e
limits-for you to think about) should lead to increased ease in
e h e charge mobility and thus expedites its diffusion. Note: This is
not referring to drift mobility, which depends on the dominant
scattering effects (due to impurities and/or lattice)!

De = diffusion coefficient of electrons, μe = electron drift, Dh = diffusion coefficient

of the holes, μh = hole drift mobility
64
©McGraw-Hill Education. Insert Photo Credit Here
 Comments

dn De kT
e = − De where =
dx e e

ℓ𝟐 , this is a bit simplified due to our original assumption that all

But we saw that De =
2τ electrons within a ℓ cross the plane in one direction! Accurate
statistical mechanics approach proves that in general D( for
electrons and holes) should be given by:

ℓ𝟐
D=
τ
65
©McGraw-Hill Education. Insert Photo Credit Here
 Nonuniform Doping Profile 1

Nonuniform doping profile results in electron diffusion toward the less concentrated regions.
This exposes positively charged donors and sets up a built-in field Ex. In the steady state, the
diffusion of electrons toward the right is balanced by their drift toward the left.
66
©McGraw-Hill Education. Insert Photo Credit Here
 Nonuniform Doping Profile 2
The net current = 0 (drift balances diffusion)

dn
J e = en eE x + eDe =0
dx
dV
Ex = −
dx
dV dn
− ene + eDe =0
dx dx
kT
De = e
kT dn e
 dV =
e n
V2 kT n2 dn kT  n2 
Potential difference

 V1
dV =
e 
n1 n
 V2 − V1 = ln 
e  n1 
67
©McGraw-Hill Education. Insert Photo Credit Here
 Built-In Potential and Concentration
kT  n2 
V2 − V1 = ln  
e  n1 
V2 = potential at point 2, V1 = potential at point 1, k = Boltzmann constant, T =
temperature, e = electronic charge, n2 = electron concentration at point 2, n1 =
electron concentration at point 1

Assume that dn kT
n(x) ≈ Nd(x) = Ndoexp(−x/b)  en eE x + eDe =0  Ex =
dx be
kT
De = e
e 68
©McGraw-Hill Education. Insert Photo Credit Here
 Photogeneration and Thermalization

Optical absorption generates electron-hole pairs.

Energetic electrons must lose their excess energy (thermalization process) to lattice
vibrations until their average energy is (3/2)kT in the CB.
69
©McGraw-Hill Education. Insert Photo Credit Here
 Band-to-Band Optical Absorption
Assume photon energy is > Eg

Absorption of photons within a small

elemental volume of width δx
I is energy/(area x time), i.e. energy per
unit area per unit time☺

Io = hfΓph
Not all electrons in valence band are near band edge (i.e., not all incident photons are absorbed!).This is why optical absorption depends on how much material is present
(thickness in this case)

δI = - αI δx, where α = α(λ) is the optical absorption coefficient of the material.

Over a distance of 1/α, the intensity, I, decreases by 63%. 1/α is called the penetration depth.
70
©McGraw-Hill Education. Insert Photo Credit Here
 Definition of Optical Absorption Coefficient
I
 =−
Ix
α = absorption coefficient, I = light intensity, δI = change in the light intensity
in a small elemental volume of thickness δx at x

Beer-Lambert Law

I(x) = Ioexp(−αx)
I(x) = light intensity at x, Io = initial light intensity, α = absorption coefficient,
x = distance from the surface (location) where I = Io. Note: Light propagates
along x.

71
©McGraw-Hill Education. Insert Photo Credit Here
 Absorption coefficient α depends on the photon energy hf
Direct transition due to
absorption of higher
energy photons.

High energy photons will excite electrons from

Sharp rise due
VB to CB in an indirect bandgap semiconductor
to direct band!
without phonon (lattice vibration) emission!

Remember: 1) right near the band edges, the density of states is relatively low for each charge species!
2) transitions between the two bands must obey quantum mechanical rules and conservation of momentum!

The absorption coefficient α depends on the photon energy hf and hence on the
wavelength. Density of states increases from band edges and usually exhibits peaks and
troughs. Generally α increases with the photon energy greater than Eg because more
energetic photons can excite electrons from populated regions of the VB to numerous
available states deep in the CB.
72
©McGraw-Hill Education. Insert Photo Credit Here
 The Schottky Junction 1
  eV  
J = J o exp   − 1
  kT  
John Bardeen, Walter Schottky, and Walter Brattain.
Walter H. Schottky (1886–1976) obtained his PhD
from the University of Berlin in 1912. He made
many distinct contributions to physical electronics.
He invented the screen grid vacuum tube in 1915,
and the tetrode vacuum tube in 1919 while at
Siemens. The Schottky junction theory was
formulated in 1938. He also made distinct
contributions to thermal and shot noise in devices.
His book Thermodynamik was published in 1929
and included an explanation of the Schottky defect
(Chapter 1).
© Brattain Collection/AIP/Science Source.

73
©McGraw-Hill Education. Insert Photo Credit Here
 Metal-Semiconductor Junction (interface, contact…)

When metal surface is in an intimate contact with semiconductor

surface, two cases can arise:

1. A rectifying contact (that is current is allowed to pass in one

direction). This is known as Schottky Junction (φmetal > φsemi)
2. Ohmic contact (current can pass in either direction across
the junction). (φmetal < φsemi)

74
©McGraw-Hill Education. Insert Photo Credit Here
 Schottky Junction Diodes 1

High frequency applications,

e.g. communications
Fast photodetectors
UV detectors
Photovoltaics
Power electronics e.g. rectifier A SiC based Schottky photodiode used for
UV detection (Courtesy of sglux, Germany)

75
©McGraw-Hill Education. Insert Photo Credit Here
 Schottky Junction 1 Note: The highly concentrated charge
After region near the interface leads to a
Before nonuniform electric field, which in
turn causes the nonlinear band
bending. You can prove this using
Poisson’s equation! We will see linear
behavior for the band bending in
p-n junctions, to be discussed in
chapter 6!

Remember: Metals do not allow

electric fields inside them☺

eVo is the built-in

potential

ΦB is the Schottky
barrier height
Note: Eo is the maximum electric field in depletion
region!

Formation of a Schottky junction between a metal - No significant depletion region in the metal due to
the very large charge carrier density compared to
the semiconductor.
and an n-type semiconductor when Φm > Φn.

At equilibrium after contact, the Fermi levels MUST align!

Energy bands always bend in the direction of the electric field!
Recall: Electrons have zero KE at vacuum level and they are free from the solid surface of the material!
Also, Fermi energy in doped semiconductors depend on doping levels! 76
©McGraw-Hill Education. Insert Photo Credit Here
 Schottky Junction in Equlibrium 1
Schottky Junction: Metal to n-Type Semiconductor

Built-in potential Vo Reminder: Φm is constant, while Φn is dopant

concentration dependent!
eVo = Φm − Φn
where The built-in potential can be measured by measuring the
capacitance! You cannot use a voltmeter to measure it
Φm – work function of the metal from the circuit!
Measuring capacitance at various bias voltages can be
Φn – work function of the semiconductor used to find Nd and/or Vo.
Note: Capacitance is voltage-dependent (depletion region
width changes with voltage!) and is essential for the
operation of the device!

Barrier height (or Schottky Barrier) ΦB from the metal to the semiconductor (this is the barrier that prevents
movement of more electrons to the metal, and vice versa! In other words, the energy needed by electrons in the
metal to hop into the semiconductor’s conduction band!)

ΦB = Φm − χ = eVo + (Ec − EFn)

where
Φm – work function of the metal
χ – electron affinity of the semiconductors
eVo – built-in potential
Ec – Efn – distance of Fermi level from CB in the semiconductor 77
©McGraw-Hill Education. Insert Photo Credit Here
 Schottky Junction Characteristics: Schottky Diode 1

Forward biased Schottky junction. Electrons in the CB of the Reverse biased Schottky junction. Electrons in the
semiconductor can readily overcome the small PE barrier to metal can not easily overcome the PE barrier FB to
enter the metal via thermal activation (there is still some enter the semiconductor.
barrier for electrons to cross into the metal by impact of V
only!).
Most of the applied voltage drops across the depletion region (since I
t is way more resistive than the rest of the structure!)

Upon forward bias, energy

of respective carries goes
up!

No changes in ΦB occur!

No Bias I vs V characteristics of an ideal Schottky

junction exhibits rectifying properties
(negative current axis is in microamps) 78
©McGraw-Hill Education. Insert Photo Credit Here
 Forward Biased Schottky Junction 1
  eV  
J = J o exp   − 1
  kT  
For ideal case-no recombination!

Quasi-Fermi level

I vs. V Characteristics of a Schottky Forward biased Schottky junction.

junction exhibits rectifying properties. Electrons in the CB of the semiconductor
Notice the scale change in current and can readily overcome the small PE barrier
voltage axes. to enter the metal.
79
©McGraw-Hill Education. Insert Photo Credit Here
 Reverse Biased Schottky
Junction 1
PE barrier against
electron diffusion
from the CB in the = e(Vo + Vr)
semiconductor to
the metal

PE barrier for electron emission

from the metal into CB = ΦB
No change

Negligibly small

  
J1 = BeT 2 exp  − B 
For Metal-n-Si junction, Be is near
110 A cm−2 K−2, and for a metal-p-Si
The net reverse current = – J1  kT  junction (holes), Be is near 30 A cm−2 K−2.

Important: This equation is similar to Richardson–Dushman equation for thermionic emission of charge over a potential barrier Φ 80
©McGraw-Hill Education. Insert Photo Credit Here
 Reverse Biased Schottky Junction 2

Reverse biased Schottky junction. I vs. V Characteristics of a Schottky

Electrons in the metal can not easily junction exhibits rectifying
overcome the PE barrier FB to enter properties (negative current axis is in
the semiconductor. microamps)

81
©McGraw-Hill Education. Insert Photo Credit Here
 The Schottky Junction 2
J = current density
  eV  
J = J o exp   − 1
Jo = constant that depends on the metal and the semiconductor

  kT   e.g. ΦB, and also on the surface properties

V = voltage, e = electronic charge

k = Boltzmann constant, T = temperature

- Wide depletion zone

- Light doping of semiconductor
- Electrons are empowered to cross the Schottky barrier via thermal energy (NOT the bias voltage!)
- The current depends on the DIFFUSION of the majority carriers (electrons in n-type semiconductor)!

How would the charge (electrons) cross the barrier in case of heavily
doped semiconductor-metal junction in forward bias?

Within reasonable limits, the doping level of the semiconductor has no bearing on the resistance of the junction!

82
©McGraw-Hill Education. Insert Photo Credit Here
 The Schottky Junction 3

The three inventors of the transistor: William

Shockley (seated left), Walter Brattain
(middle), and John Bardeen (right). The three
inventors shared the Nobel prize in 1956.
What is the diagram on the chalkboard?
©Photo12/The Image Works.

83
©McGraw-Hill Education. Insert Photo Credit Here
 The principle of
the Schottky
junction solar
cell

Carriers DRIFT in the

depletion region creating
a photocurrent!

The principle of the Schottky junction solar cell. The built-in field and built-in voltage are
reduced under illumination
It is also possible for the light excitation to excite electrons from the metal even if the Schottky barrier is less than
Eg of the semiconductor! These electrons also roll down the potential hill and contribute to the current! 84
©McGraw-Hill Education. Insert Photo Credit Here
 Schottky photodetectors
Increasing E causes increase in drift velocity of charges
causing faster response time of the device! FAST photodetection applications.

Reverse-biased Schottky photodiodes are frequently

used as fast photodetectors
Schottky junction is also used as “Gate” in field effect transistor!

©McGraw-Hill Education. Insert Photo Credit Here

 Ohmic Contacts
“A junction that does not limit the current flow”

Charge accumulation
region

(Heavily doped > 1018/cm 3)

Pretty much, only the resistance of the semiconductor outside the contact region is responsible for limiting the
current flow (This does not mean that I-V relation is necessarily linear in the contact region).
J = σE
When a metal with a smaller workfunction than an n-type semiconductor are put into contact, The resulting junction is an
ohmic contact.
- Narrow depletion zone (due to heavy doping), which leads to tunneling into the metal!
- Electrons tunnel through the barrier
- Heavily doped and degenerate semiconductors are needed to make
the depletion region so narrow that it facilitates tunnelling of electrons.
- Upon bias, the current is dominantly a drift one modulated by the tunneling probability!
- Used for source and drain in FET (field effect transistor) 86
©McGraw-Hill Education. Insert Photo Credit Here
 Ohmic vs Schotkky Junction

Ohmic Schotkky
Ohmic-non rectifying Rectifying (current
((same current out of
and into the junction)
in one direction)

Normally used for

Source and Drain in Normally used for
transistors Gate in transistors

Heavily doped

Lightly doped
Very narrow
depletion region
Wide depletion
region

Electrons tunnel Electrons are

through the junction thermally emitted
(temp. independent across junction
process)

Relatively high
Very low capacitance capacitance

Note: It is possible to convert an Ohmic contact to Schotkky junction by heavily doping (degenerate) the
semiconductor!
87
©McGraw-Hill Education. Insert Photo Credit Here
 Peltier Effect 1
(Heat release or absorption at the junction due to flow of charge!)

- When pushing charges uphill, heat is needed to do so (heat absorption-cooling environment around it)
- When charges roll down hill, they release heat (heating environment around it).

By Convention: CURRENT MOVES OPPOSITE TO CHARGES

Generate heat (temperature differential)

from applying power! As opposed to
Seebeck effect seen in
thermocouples, for example!, where
voltage is generated from heat!

(a) Current from an n-type semiconductor to the metal results in heat absorption at the junction.
(b) Current from the metal to an n-type semiconductor results in heat release at the junction.
88
©McGraw-Hill Education. Insert Photo Credit Here
 Peltier Effect 2

https://www.researchgate.net/figure/Figure-S3-Band-structures-of-metal-
semiconductor-heterojunctions-There-are-two-kinds-of_fig1_295424859

When a dc current is passed through a semiconductor to which metal contacts have

been made, one junction absorbs heat and cools (the cold junction) and the other
releases heat and warms (the hot junction).

89
©McGraw-Hill Education. Insert Photo Credit Here
 Thermoelectric Cooler 1

Cross section of a typical thermoelectric cooler.

90
©McGraw-Hill Education. Insert Photo Credit Here
 Thermoelectric Cooler 2

Typical structure of a commercial thermoelectric cooler.

Some Semiconductors used: Bi2Te3, Bi2Se3 91

©McGraw-Hill Education. Insert Photo Credit Here
 Thermoelectric Cooler 3

A two-stage thermoelectric cooler (3 cm × 3 cm × 1.09 cm), capable of

generating a temperature difference of 76 °C at a current of 5.70 A. Courtesy
of Laird, USA.

92
©McGraw-Hill Education. Insert Photo Credit Here
 Some Notes on Effective Mass
𝑑 2
𝐸 -1
m* ∝
𝑑𝑘2

Example of a special case (massless particle behavior):

Massless
Particle * * *
(undefined mass)
E∝ k E∝ k2 E∝ k2
Undefined!
https://www.researchgate.net/figur e/We-describe-the ore tically-the-distor sion-of-the-Dir ac-cone-
in-aI3-by-a-curvature-term_fig3_320796727

• m* can be negative!
• If electrons have negative m* at the top of the valence band, they will move opposite to the
applied force (see next slide)!
• m* can be zero (see graphene example above. It should be undefined at the vortex, but some
use “0” instead)
• For a given material, the effective mass can change values depending on crystal
planes and directions! Remember E vs. k is not a simple nice-looking curve like we
been dealing with.
- In the special case for the cone energy band structure (see above figure. Ex: Graphene!), the effective
mass is termed “massless” at the cone vertex (called Dirac point)! Defining m* in
such systems is more complex than what we have been doing so far! The undefined mass of electrons
in graphene accounts for their unusually high speed, which results in highly conductive material
property! 93
©McGraw-Hill Education. Insert Photo Credit Here
 A hole is a missing electron in the valence band 2
E (electric field)
N
J N = – e v gi = 0
Classical electrons movement

(energy)
i =1
Top VB electrons movement
(energy)

Hole movement

𝑑 2 𝐸 -1
m* ∝ ( 𝑑𝑘2 )

By convention, it is better to deal with holes having positive effective mass (regardless of the formula
above!) than electrons at the top of the valence band with an apparent negative effective mass. This
choice makes charge transport equations more intuitive to us (positive quasi-particle-not real particle-
with positive mass!). Remember: we invented the hole concept to represent (N-1) electrons’ behavior
in the valence band! The missing electron is in the conduction band now☺ Under the same electric
field, the conduction band electron moves as we know classically (they have positive mass now!) 94
©McGraw-Hill Education. Insert Photo Credit Here