1

Outline
1. Thermoelectric Effect
Seeback Effect
Peltier Effect
Thompson Effect
2. Basics of Thermodynamics
3. Derivation of Thermo-Electric
Coefficients Using the Relaxation Time
Approximation
4. Thermo-Magnetic Effects
5. Thermal Effects in Nano Devices
2
3
1. Thermoelectric Effect
The thermoelectric effect is the direct conversion of
temperature differences to electric voltage and vice
versa.
Simply put, a thermoelectric device creates a voltage
when there is a different temperature on each side, and
when a voltage is applied to it, it creates a temperature
difference.
This effect can be used to generate electricity, to
measure temperature, to cool objects, or to heat them.
Because the direction of heating and cooling is
determined by the sign of the applied voltage,
thermoelectric devices make very convenient
temperature controllers.
4
Traditionally, the term thermoelectric effect or
thermoelectricity encompasses three separately
identified effects, the Seebeck effect, the Peltier
effect, and the Thomson effect.
The Peltier and Seebeck effects are the most basic of
the three and are essentially the inverses of one
another, for which reason the thermoelectric effect
may also be called the PeltierSeebeck effect.
Joule heating, the heat that is generated whenever
a voltage difference is applied across a resistive
material, is somewhat related, though it is not
generally termed a thermoelectric effect (and it is
usually regarded as being a loss mechanism or non-
ideality in thermoelectric devices).
The PeltierSeebeck and Thomson effects are
reversible, whereas Joule heating is not.
5
Seebeck Effect
In 1821, Thomas Seebeck found that an electric
current would flow continuously in a closed circuit
made up of two dissimilar metals, if the junctions of
the metals were maintained at two different
temperatures.

S= dV / dT;
S is the Seebeck Coefficient with units of Volts per Kelvin
S is positive when the direction of electric current is same
as the direction of thermal current
6
Seebeck Effect
The Seebeck effect is the conversion of temperature
differences directly into electricity.

The effect is that a voltage, the thermoelectric EMF,
is created in the presence of a temperature difference
between two different metals or semiconductors.

This causes a continuous current to flow in the
conductors if they form a complete loop. The voltage
created is of the order of several microvolts per
degree difference.
7
the voltage developed can be derived from:


S
A
and S
B
are the Seebeck coefficients (also called
thermoelectric power or thermopower) of the metals A
and B as a function of temperature, and
T1 and T2 are the temperatures of the two junctions.

The Seebeck coefficients are non-linear as a function
of temperature, and depend on the conductors'
absolute temperature, material, and molecular
structure.

If the Seebeck coefficients are effectively constant for
the measured temperature range, the above formula
can be approximated as:
8
Thermopower
The thermopower, or thermoelectric power, or
Seebeck coefficient of a material is a measure of the
magnitude of an induced thermoelectric voltage in
response to a temperature difference across that
material.
The thermopower has units of (V / K), though in
practice it is more common to use microvolts per
kelvin. Values in the hundreds of V/K, negative or
positive, are typical of good thermoelectric materials.
The term thermopower is a misnomer since it
measures the voltage or electric field induced in
response to a temperature difference, not the electric
power.
9
Peltier Effect
In 1834, a French watchmaker and part time
physicist, Jean Peltier found that an electrical
current would produce a temperature gradient at the
junction of two dissimilar metals.
<0 ; Negative Peltier coefficient
High energy electrons move from
right to left.
Thermal current and electric current
flow in opposite directions.
(electronic)
10
Peltier Cooling
>0 ; Positive Peltier coefficient

High energy holes move from left to
right.

Thermal current and electric current
flow in same direction.

q=*j, where q is thermal current density and j is electrical current
density.
= S*T (Volts) S ~ 2.5 k
B
/e for typical TE materials
T is the Absolute Temperature
11
12
When a current I is made to flow through the circuit,
heat is evolved at the upper junction (at T2), and
absorbed at the lower junction (at T1). The Peltier heat
absorbed by the lower junction per unit time, is equal to




Where is the Peltier coefficient
AB
of the entire
thermocouple, and
A
and
B
are the coefficients of
each material. P-type silicon typically has a positive
Peltier coefficient (though not above ~550 K), and n-
type silicon is typically negative, as the names suggest.
13
The Peltier coefficients represent how much heat current is
carried per unit charge through a given material.

Since charge current must be continuous across a junction,
the associated heat flow will develop a discontinuity if
A

and
B
are different. This causes a non-zero divergence at
the junction and so heat must accumulate or deplete there,
depending on the sign of the current.

Another way to understand how this effect could cool a
junction is to note that when electrons flow from a region of
high density to a region of low density, they expand (as with
an ideal gas) and cool.
14
The conductors are attempting to return to the electron
equilibrium that existed before the current was applied by
absorbing energy at one connector and releasing it at the
other. The individual couples can be connected in series to
enhance the effect.

An interesting consequence of this effect is that the direction
of heat transfer is controlled by the polarity of the current;
reversing the polarity will change the direction of transfer
and thus the sign of the heat absorbed/evolved.

A Peltier cooler/heater or thermoelectric heat pump is a
solid-state active heat pump which transfers heat from one
side of the device to the other. Peltier cooling is also called
thermo-electric cooling (TEC).
15
Applications of Peltier Effect
Water/Beer Cooler
Cooled
Car Seat
Electronic Cooling
Cryogenic IR Night Vision
Laser/OE Cooling
TE
Si bench
16
Thomson Effect
Discovered by William Thomson (Lord Kelvin)
When an electric current flows through a
conductor, the ends of which are maintained at
different temperatures, heat is evolved at a rate
approximately proportional to the product of the
current and the temperature gradient.

dx
dT
I
dx
dQ
t =
t
is the Thomson coefficient in Volts/Kelvin
Seebeck coeff. S is temperature dependent
dT
dS
T = t
Relation given by Kelvin:
17
The Thomson effect was predicted and
subsequently experimentally observed by William
Thomson (Lord Kelvin) in 1851. It describes the
heating or cooling of a current-carrying conductor
with a temperature gradient.

Any current-carrying conductor (except for a
superconductor), with a temperature difference
between two points, will either absorb or emit heat,
depending on the material.

If a current density J is passed through a
homogeneous conductor, heat production per unit
volume;
18
is the resistivity of the material
dT/dx is the temperature gradient along the wire
is the Thomson coefficient.
The first term J is simply the Joule heating, which is not
reversible.
The second term is the Thomson heat, which changes sign
when J changes direction.
19
In metals such as zinc and copper, which have a hotter end
at a higher potential and a cooler end at a lower potential,
when current moves from the hotter end to the colder end,
it is moving from a high to a low potential, so there is an
evolution of heat. This is called the positive Thomson
effect.

In metals such as cobalt, nickel, and iron, which have a
cooler end at a higher potential and a hotter end at a lower
potential, when current moves from the hotter end to the
colder end, it is moving from a low to a high potential, there
is an absorption of heat. This is called the negative
Thomson effect.
20
The Thomson coefficient is unique among the three
main thermoelectric coefficient because it is the only
thermoelectric coefficient directly measurable for
individual materials.

The Peltier and Seebeck coefficients can only be
determined for pairs of materials. Thus, there is no
direct experimental method to determine an absolute
Seebeck coefficient (i.e. thermopower) or absolute
Peltier coefficient for an individual material.

However, there are two equations, the Thomson
relations, also known as the Kelvin relations (see
below), relating the three thermoelectric coefficients.
21
The Seebeck effect is actually a combination of the Peltier
and Thomson effects. In fact, in 1854 Thomson found two
relationships, now called the Thomson or Kelvin
relationships, between the corresponding coefficients. The
absolute temperature T, the Peltier coefficient and
Seebeck coefficient S are related by the first Thomson
relation

which predicted the Thomson effect before it was actually
formalized. These are related to the Thomson coefficient
by the second Thomson relation
22
The figure of merit for thermoelectric devices is defined as

where is the electrical conductivity, is the thermal
conductivity, and S is the Seebeck coefficient or
thermopower (conventionally in V/K).

This is more commonly expressed as the dimensionless
figure of merit ZT by multiplying it with the average
temperature ((T
2
+ T
1
) / 2). The value of the figure of merit
is usually proportional to the efficiency of the device,
subject to certain provisions, particularly the requirement
that the two materials of the couple have similar Z values.
23
ZT is therefore a very convenient figure for
comparing the potential efficiency of devices using
different materials.

Values of ZT=1 are considered good, and values of
at least the 34 range are considered be essential for
thermoelectrics to compete with mechanical
generation and refrigeration in efficiency.

To date, the best reported ZT values have been in
the 23 range. Much research in thermoelectric
materials has focused on increasing the Seebeck
coefficient and reducing the thermal conductivity,
especially by manipulating the nanostructure of the
materials.

24
Progress in ZT
Fundamental limitations:

k and o, S and o are coupled.
25
Bismuth telluride is the best bulk TE material with
ZT=1
Trends in TE devices:
Superlattices and nanowires: Increase in S, reduction in k
Nonequilibrium effects: decoupling of electron and phonon
transport
Bulk nanomaterial synthesis
Trends in TE systems
Microrefrigeration based on thin film technologies
Automobile refrigeration
TE combined with fluidics for better heat exchangers
To match a refrigerator, an effective ZT= 4 is needed
To efficiently recover waste heat from car, ZT = 2 is
needed
26
27
28
2. Basics of Thermodynamics
Statistical thermodynamics is based on the fundamental
assumption that all possible configurations of a given
system, which satisfy the given boundary conditions
such as temperature, volume and number of particles,
are equally likely to occur.

The overall system will therefore be in the statistically
most probable configuration.

The entropy of a system is defined as the logarithm of
the number of possible configurations multiplied with
Boltzmann constant. While such definition does not
immediately provide insight into the meaning of entropy,
it does provide a straightforward analysis since the
number of configurations can be calculated for any given
system.
29
Laws of Thermodynamics
0
th
Law of Thermodynamics

It is observed that a higher temperature object which is in contact with a lower
temperature object will transfer heat to the lower temperature object. The
objects will approach the same temperature, and in the absence of loss to other
objects, they will then maintain a constant temperature. They are then said to
be in thermal equilibrium.

The "zeroth law" states that if two systems are at the same time in thermal
equilibrium with a third system, they are in thermal equilibrium with each other.
30
Laws of Thermodynamics
1
st
Law of Thermodynamics

The first law of thermodynamics is the application of the conservation of energy
principle to heat and thermodynamic processes:










In the context of physics, the common scenario is one of adding heat to a volume of
gas and using the expansion of that gas to do work, as in the pushing down of a
piston in an internal combustion engine.
31
Internal Energy
Internal energy is defined as the energy associated with the random,
disordered motion of molecules.

it refers to the invisible microscopic energy on the atomic and molecular
scale.








U is the most common symbol used for internal energy.
Related energy quantities which are particularly useful in chemical
thermodynamics are enthalpy, Helmholtz free energy, and Gibbs free
energy.
32
Enthalpy
Four quantities called thermodynamics potentials " are useful in
the chemical thermodynamics of reactions and non-cyclic
processes. They are internal energy, the enthalpy, the
Helmholtz free energy, and the Gibbs free energy.

Enthalpy is defined by H = U + PV

where P and V are the pressure and volume, and U is internal
energy. Enthalpy is then a precisely measurable state variable
state, since it is defined in terms of three other precisely
definable state variables.

It is somewhat parallel to the first law of thermodynamics for a
constant pressure system

Q = U + PV since in this case Q=H
33
An increase in the enthalpy H = U + PV might be associated
with an increase in internal energy which could be measured by
calorimetry, or with work done by the system, or a combination
of the two.

The internal energy U might be thought of as the energy
required to create a system in the absence of changes in
temperature or volume.

But if the process changes the volume, as in a chemical reaction
which produces a gaseous product, then work must be done to
produce the change in volume.

For a constant pressure process the work you must do to produce
a volume change V is PV.

Then the term PV can be interpreted as the work you must do to
"create room" for the system if you presume it started at zero
volume.
34
Laws of Thermodynamics
2
nd
Law of Thermodynamics

The second law of thermodynamics is a general principle which places
constraints upon the direction of heat transfer and the attainable efficiencies
of heat engines. In so doing, it goes beyond the limitations imposed by the
first law of thermodynamics. It's implications may be visualized in terms of
the waterfall analogy.
35
Second Law of Thermodynamics: It is impossible to extract an amount of
heat Q
H
from a hot reservoir and use it all to do work W . Some amount of
heat Q
C
must be exhausted to a cold reservoir. This precludes a perfect
heat engine. This is sometimes called the "first form" of the second law,
and is referred to as the Kelvin-Planck statement of the second law.
36
Entropy
Second Law of Thermodynamics: In any cyclic process the entropy
will either increase or remain the same.

Entropy: a state variable whose change is defined for a reversible
process at T where Q is the heat absorbed.

Since entropy gives information about the evolution of an isolated system
with time, it is said to give us the direction of " time arrow " . If snapshots
of a system at two different times shows one state which is more
disordered, then it could be implied that this state came later in time. For
an isolated system, the natural course of events takes the system to a
more disordered (higher entropy) state.
37
Entropy a state variable whose change is defined for a
reversible process at T where Q is the heat
absorbed.
a measure of the amount of energy which is
unavailable to do work.
a measure of the disorder of a system.
a measure of the multiplicity of a system.
38
Laws of Thermodynamics
3
rd
Law of Thermodynamics

The third law of thermodynamics states that if all the thermal motion of
molecules (kinetic energy) could be removed, a state called absolute
zero would occur. Absolute zero results in a temperature of 0 Kelvins or
-273.15 Celsius.

Absolute Zero = 0 Kelvins = -273.15 Celsius

The Universe will attain absolute zero when all energy and matter is
randomly distributed across space. The current temperature of empty
space in the Universe is about 2.7 Kelvins.
39
Thermodynamic Identity

The thermodynamic identity states that a change in energy can be caused
by adding heat, work or particles. Mathematically this is expressed by:

dU=dQ+dW+dN

where U is the total energy, Q is the heat and W is the work.

is the energy added to a system when adding one particle without adding
either heat or work. This energy is also called the electro-chemical potential.

N is the number of particles.
40
3. Derivations
In the discussion of RTA we introduced the concept
of electrical conductivity assuming no spatial
gradients in the system. In general, the term that
involves spatial derivative of the distribution function
in the BTE leads to:
Diffusion effects, when the electron density is position
dependent
A variety of thermoelectric effects when the temperature
and/or the Fermi energy are spatially varying functions
41
General case variation in both T and E
F













Substituting into BTE gives
0
0
1
,
1
A
C k F
B
s s k C k F s
r s r r r
k k
f f f
E E E
f
e k T
f f E E E E f
f T
E T E
u
u
u u
u u
= +
+
= =
+
c c c + c
| |
V = V = V = V
|
c c c c
\ .
C k F s s A
r
k k m
E E E f f f
Tv e v
T E E
c
t
+ c c
| |
V =
|
c c
\ .
42
Solving for the asymmetric component of the DF gives:















1
cos ( )
n s
A m z k n
k
F f T
f e v E F
eT z z e E
t u c
( c c c
| |
= + +
| (
c c c
\ .

( )
2
1
1
1

n
z m z m m n m
m
n
zz z zz n
m
n
z zz z
m
n
m
F ne T
J E F
m eT z z e
E
F T
F
eT z z e
F T
J Q
z z e
E
Q F Seebeck Coefficient
eT
t c t t t
t
o c o
t
c
t
t
( c c
| |
= + +
| (
c c
\ .

(
| |
c c
| |
= + + ( |
|
|
c c
\ .
(
\ .

c c
| |
= +
|
c c
\ .
| |
= |
|
\ .
43
When the momentum relaxation time obeys power-law dependence.
then





Seeback coefficient is:
- negative for n-type sample
- positive for p-type sample
5/ 2 , ln
F C C B
F C B
B
E E N k
Q s E E k T
e k T n
| |

| |
= + =
|
|
\ .
\ .
44
Thermal Power Flow

A proper definition for the heat current is suggested by the
thermodynamic relation:

dU=dQ+E
F
dN

where: dU -> increase in internal energy
dQ -> increase in heat
E
F
dN -> increase in internal energy when a
small number of carriers is added
at a constant temperature

J
Q
= J
U
- E
F
J
n
Heat flux
Flux of internal energy
Particle current density
45
Thus:
| |
( )
0
2
2
2
2
2
2
2
1
1
1
Relation
=-
Q C k F A
k
m m
F
Qz n m k n
m
m m
n z m
m m
m
n
m
Qz z
m
m
m
J E E E vf
V
ne E
E n T
J F E F
m z e mT z
E E
n T
F J E
e mT z
E
F TQ Kelvin
e
T
J J
z
E
n
E
mT
t t
t
t
t t
t
t t
t
t
t
t k
t
k t
t
= +
(
c c
| |
= +
(
|
c c
\ .
(

(
(
c
(
= + +
(
c
(
(


(
= =
(
(

c
=

(
(
(

Peltier coefficient
Thermal conductivity
46
When the momentum relaxation time obeys power-law dependence.
then




2
( 5/ 2)
B
k
T s
e
k
| |
= +
|
\ .
47
4. Thermo-Magnetic Effects
Nernst Effect
Ettingshausen Effect
Righi-Leduc Effect
48
Nernst Effect
49

Mobile energy carriers (for example conduction-band
electrons in a semiconductor) will move along
temperature gradients due to statistics and the
relationship between temperature and kinetic energy.

If there is a magnetic field transversal to the
temperature gradient and the carriers are electrically
charged, they experience a force perpendicular to
their direction of motion (also the direction of the
temperature gradient) and to the magnetic field.
Thus, a perpendicular electric field is induced.
50
51
Ettingshausen effect
52
53
Righi-Leduc Effect
54
5. Thermal Effects in Nano
Devices
Innovation Innovation- -Enabled Enabled
Technology Pipeline Technology Pipeline
Future options subject to change
Tri-Gate Tri-Gate
S
G
D
III-V
S
S
G
D
III-V
S
Carbon
Nanotube FET
Carbon
Nanotube FET
50 nm 50 nm
35 nm 35 nm
SiGe S/D
Strained
Silicon
SiGe S/D
2
nd
generation
Strained
Silicon
20 nm 20 nm
9
0
n
m
6
5
n
m
4
5
n
m
3
2
n
m
2
0
0
3
2
0
0
5
2
0
0
7
2
0
0
9
2
0
1
1
+
M
a
n
u
fa
c
tu
rin
g
D
e
v
e
lo
p
m
e
n
t
R
e
s
e
a
rc
h
10 nm 10 nm
5 nm 5 nm
5 nm
Nanowire
Metal Gate
High-k
Si Substrate
Low
resistance
layer
Metal Gate
High-k
Si Substrate
Low
resistance
layer
High-k/
metal gate
55
Average Power Density

1980 1990 2000 2010 2020
0.0
0.2
0.4
0.6
0.8
1.0


Intel Processors
ITRS Estimates
A
v
e
r
a
g
e

P
o
w
e
r

D
e
n
s
i
t
y
,

W
/
m
m
2
Year
56
57
58
Joule heating
59
60
Small Structures:
61
Existing Modeling Packages:
THERMAL3D

Predicts heat flow and temperature rise for material systems and
any number of heat generating sources
Models validated using measured data
Three models for heat dependent thermal conductivity to choose
from for each of the materials in the system
User definable thermal conductivities and coefficients for each
material
Very fast simulation times allow many combinations to be tried
for system design optimization
Seamlessly integrated into Silvacos device simulation software
framework, ATLAS
Industry leading, easy to use, multi-dimensional visualization
tools for results analysis
Interactive, user friendly and flexible runtime environment for
quick result generation and analysis with numerous examples
62
In this example, Thermal3D was then used to investigate the optimal
spacing between the devices and the effect on operating temperature
profile. Here a cut line through the center of the devices is shown
together with surface temperature.
63
Existing Modeling Packages:
GIGA3D
Self-Consistent Lattice Temperature solver
Thermodynamically correct modelization
Coupled to Drift-Diffusion or Hydrodynamic equations
Able to model steady state, transient and small signal a.c.
biasing
Default parameters for thermal conductivity over a range of
materials
Default parameters for heat capacity over a range of materials
Lattice temperature dependence for a wide range of parameters
Joule and Peltier/Thomson heat generation terms
Flexible boundary condition specification
Choice of non-linear solvers for coupling to drift-diffusion
equations
Anisotropic thermal conductivity tensor
Flexible thermopower specification, including phonon drag

64
Possible uses of the solver include:
Analysis of ESD protection structures
Simulation of power devices such as rectifiers, thyristors,
mosfets, bipolar transistors
Modeling LED, SOI , HBT, HEMT devices
Thermo-Voltaic device simulation
Thermal runaway modeling
Device efficiency modeling

65
Lattice temperature
distribution for the SOI
transistor at a gate bias of 3
V and a drain bias of 4 V.
66
The ASU Approach for
Thermal Transport
( )
{ }
{ }
k+q k k+q k k k+q k k+q
,q , q , q ,q
q
k+q k k k+q
,q ,q
k
(k) r (13a)
( ) (13b)
e r k e a e a
p r e a
p p
e
v E f W W W W
t
g
v q g W W
t t

c | |
+ V + V = +

|
c
\ .
c c | | | |
+ V = +

| |
c c
\ . \ .
J. Lai and A. Majumdar, Concurent
thermal and electrical modeling of
submicrometer silicon devices, J.
Appl. Phys. , Vol. 79, 7353 (1996).
( )
2
* 3
, (14 )
2 2
3
. (14 )
2
LO e L d LO A B
LO LO
e LO e LO LO A
LO A e L A B
A A A LO
LO A e L
T T T nm v T T nk
C C a
t
T T T T T nk
C k T C b
t
t t t
t t


| | | | c
= +
| |
c
\ . \ .
| | | | c
= V V + +
| |
c
\ . \ .
67





















High Electric Field
Optical Phonon
Emission
Acoustic Phonon
Emission
Heat Conduction
in Semiconductor
t ~ 0.1ps t ~ 0.1ps

t ~ 10ps

t ~ 10ps

Hot Electron Transport
68


























Ensemble Monte
Carlo Device
Simulator
Phonon Energy
Balance Equations
Solver
T
A
T
LO
n
v
d

T
e

Find electron position in a grid:(i,j)
Find: T
L
(i,j)=T
A
(i,j) and T
LO
(i,j)
Select the scattering table with
coordinates: (T
L
(i,j)=T
LO
(i,j))
Generate a random number and
choose the scattering mechanism
for a given electron energy
0
25
50
75 75
0
5
10
0.5
1
1.5
x 10
25
E
l
e
c
t
r
o
n

D
e
n
s
i
t
y

(
m
-
3
)
Si/SiO2 interface
source
drain
channel
Si/BOX
interface
x (nm)
y (nm)
0
25
50
75 75
1
6
11
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
E
n
e
r
g
y

(
e
V
)
channel
source
drain
x (nm)
y (nm)
Exchange of Variables Between the Two Kernels
69
5 10 15 20 25 30
1.4
1.42
1.44
1.46
1.48
1.5
1.52
Number of iterations
C
u
r
r
e
n
t

(
m
A
/
u
m
)


k(Si)=143W/mK
k(Si)=13W/mK
Current Convergence
70
0 0.2 0.4 0.6 0.8 1 1.2
0
0.5
1
1.5
2
Vds (V)
I
d
s


(
m
A
/
u
m
)
isothermal
T=300K
isothermal
T=400K
thermal
simulations
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7
x 10
-8
-0.5
0
0.5
1
1.5
2
2.5
x 10
5
x (m)
V
e
l
o
c
i
t
y

(
m
/
s
)


Vds=1.2V
Vds=1.1V
Vds=1.0V
Vds=0.8V
Vds=0.6V
Vds=0.4V
channel source
drain
Modeling of 25 nm gate length FD SOI Device
Carriers are in the velocity
overshoot
71
25 nm FD SOI nMOSFET (Vgs=Vds=1.2V)


10 20 30 40 50 60 70
3
8
13 13
45 nm FD SOI nMOSFET (Vgs=Vds=1.2V)


20 40 60 80 100 120
3
12
21 21
60 nm FD SOI nMOSFET (Vgs=Vds=1.2V)


20 40 60 80 100 120 140 160 180
3
15
27 27
300
400
500
300
400
500
300
400
500
source contact drain contact
source region
drain region

80 nm FD SOI nMOSFET (Vgs=Vds=1.5V)


50 100 150 200
3
20
35 35
90 nm FD SOI nMOSFET (Vgs=Vds=1.5V)


50 100 150 200 250
3
21
39 39
100 nm FD SOI nMOSFET (Vgs=Vds=1.5V)


50 100 150 200 250 300
3
23
43 43
300
400
500
600
300
400
500
300
400
500



25 nm FD SOI nMOSFET (Vgs=Vds=1.2V)


10 20 30 40 50 60 70
3
8
13 13
45 nm FD SOI nMOSFET (Vgs=Vds=1.2V)


20 40 60 80 100 120
3
12
21 21
60 nm FD SOI nMOSFET (Vgs=Vds=1.2V)


20 40 60 80 100 120 140 160 180
3
15
27 27 300
400
500
300
400
500
300
400
500
600
source contact
drain contact

80 nm FD SOI nMOSFET (Vgs=Vds=1.5V)


50 100 150 200
3
20
35 35
90 nm FD SOI nMOSFET (Vgs=Vds=1.5V)


50 100 150 200 250
3
21
39 39
100 nm FD SOI nMOSFET (Vgs=Vds=1.5V)


50 100 150 200 250 300
3
23
43 43 300
400
500
600
400
600
300
400
500
600

T=300 K on gate T=400 K on gate
72
0 25 50 75 75
0
2
4
6
8
10
x (nm)
y

(
n
m
)


300
305
310
315
320
325
330
335
source drain
0 25 50 75 75
0
2
4
6
8
10
x (nm)
y

(
n
m
)


300
320
340
360
380
400
source drain
0 2.5 5 7.5 7.5
x 10
-8
-0.2
0.6
1.4
2.2
3
3.8
4.6
5.4
66
x 10
-8
x (m)
y

(
m
)


300
320
340
360
380
source drain
BOX
0 2.5 5 7.5 7.5
x 10
-8
-0.2
0.6
1.4
2.2
3
3.8
4.6
5.4
66
x 10
-8
x (m)
y

(
m
)


300
320
340
360
380
400
source drain
BOX
Possible future device designs