Coulomb Scattering: Dragica Vasileska Professor Arizona State University

Download as pdf or txt
Download as pdf or txt
You are on page 1of 39

Coulomb Scattering

Dragica Vasileska
Professor
Arizona State University
What contributes to Γkk′ ?

Scattering Mechanisms

Defect Scattering Carrier-Carrier Scattering Lattice Scattering

Crystal
Impurity Alloy Intravalley Intervalley
Defects

Neutral Ionized Acoustic Optical Acoustic Optical

Deformation Piezo- Nonpolar Polar


Elastic Scattering potential electric

processes
 Where is Coulomb Scattering Important:
◦ Heavily doped bulk systems
◦ Semiconductor devices
 Nanoscale MOSFETs
 MESFETs
 PN Diodes
 BJTs
 Contact regions of FD SOI devices, Dual-gate
transistors, FinFETS
 HEMTs -> remote Coulomb scattering is important.
Since remote Coulomb scattering is much weaker, we
have very high mobilities in HEMT structures that are
used in RF applications

Coulomb Scattering
Location of the Channel in Different FET Types

MESFET HEMT MOSFET


S G D S G D S G D

n+ n+ n+ n AlGaAs n+ n+ n+

i GaAs p-type Si
S.I. GaAs
S.I. GaAs
Channel Channel Channel
bulk-like 2DEG 2DEG
Inversion layer
EC EC
EC

EF EF EF
EV EV
EV
Ionized impurity scattering is due to:

(Ionized donors/acceptors, substitutional impurities, charged


surface states, etc.)

• The potential due to a single ionized impurity with net charge


Ze is: 2
0 Ze
Vi (r ) = − mks units
4πεr
• In the one electron picture, the actual potential seen by
electrons is screened by the other electrons in the system.
What is Screening?

λD - Debye screening length


-

r 1 screening cloud 1 − r / λD
+ → e
r r
-

Ways of treating screening:


• Thomas-Fermi Method
static potentials + slowly varying in space
• Mean-Field Approximation (Random Phase Approximation)
time-dependent and not slowly varying in space
 Screening is the damping of electric fields caused by the
presence of mobile charge carriers.
 It is an important part of the behavior of charge-carrying fluids,
such as ionized gases (classical plasmas) and conduction
electrons in metals.
 In a fluid composed of electrically charged constituent particles,
each pair of particles interact through the Coulomb force,

 In reality, these long-range effects are suppressed by the flow


of the fluid particles in response to electric fields. This flow
reduces the effective interaction between particles to a short-
range "screened" Coulomb interaction.
 For example, consider a fluid composed of electrons. Each
electron possesses an electric field which repels other electrons.
As a result, it is surrounded by a region in which the density of
electrons is lower than usual. This region can be treated as a
positively-charged "screening hole". Viewed from a large
distance, this screening hole has the effect of an overlaid
positive charge which cancels the electric field produced by the
electron. Only at short distances, inside the hole region, can the
electron's field be detected.
 The first theoretical treatment of screening, due to Debye and Hückel (1923), dealt with a
stationary point charge embedded in a fluid. This is known as electrostatic screening.
 Consider a fluid of electrons in a background of heavy, positively-charged ions. For
simplicity, we ignore the motion and spatial distribution of the ions, approximating them as
a uniform background charge. In condensed matter physics, this model is referred to as
jellium.
 Let ρ denote the number density of electrons, and φ the electric potential. At first, the
electrons are evenly distributed so that there is zero net charge at every point. Therefore,
φ is initially a constant as well.
 We now introduce a fixed point charge Q at the origin. The associated charge density is
Qδ(r), where δ(r) is the Dirac delta function. After the system has returned to equilibrium,
let the change in the electron density and electric potential be ∆ρ(r) and ∆φ(r) respectively.
The charge density and electric potential are related by the first of Maxwell's equations,
which gives

 To proceed, we must find a second independent equation relating ∆ρ and ∆φ. There are
two possible approximations, under which the two quantities are proportional: the Debye-
Hückel approximation, valid at high temperatures, and the Fermi-Thomas approximation,
valid at low temperatures.
 In the Debye-Hückel approximation, we maintain the system in
thermodynamic equilibrium, at a temperature T high enough
that the fluid particles obey Maxwell-Boltzmann statistics. At
each point in space, the density of electrons with energy j has
the form

Debye-Hückel approximation
 In the Fermi-Thomas approximation, we maintain the system at
a constant chemical potential and at low temperatures. (The
former condition corresponds, in a real experiment, to keeping
the fluid in electrical contact at a fixed potential difference with
ground.) The chemical potential µ is, by definition, the energy
of adding an extra electron to the fluid. This energy may be
decomposed into a kinetic energy T and the potential energy -
eφ. Since the chemical potential is kept constant,

∆µ = ∆T − e∆φ = 0.

 If the temperature is extremely low, the behavior of the


electrons comes close to the quantum mechanical model of a
free electron gas. We thus approximate T by the kinetic energy
of an additional electron in the free electron gas, which is
simply the Fermi energy EF. The Fermi energy is related to the
density of electrons (including spin degeneracy) by

Fermi-Thomas approximation
 Our results from the Debye-Hückel or Fermi-Thomas
approximation may now be inserted into the first Maxwell
equation. The result is

 which is called a screened Coulomb potential. It is a Coulomb


potential multiplied by an exponential damping term, with the
strength of the damping factor given by the magnitude of k0, the
Debye or Fermi-Thomas wave vector. Note that this potential has
the same form as the Yukawa potential.

Screened Coulomb interactions


Brooks-Herring Approach
• For the scattering rate due to impurities, we need for Fermi’s
rule the matrix element between initial and final Bloch states
− ik′⋅r
n , k Vi (r ) n, k = V ∫ drun′,k ′e Vi (r )un,k e ik⋅r
′ ′ −1 *

Since the u’s have periodicity of lattice, expand in reciprical


space
− ik′⋅r
−1
= ∑V ∫ dre Vi (r )e ik⋅r e −iG⋅rU nn′kk ′ (G )
G
= ∑V −1 ∫ dre −ik′⋅rVi (r )e ik⋅r e −iG⋅r ∫ dr′un* ′,k ′ (r′)un,k (r′)e iG⋅r′
G Ω

• For impurity scattering, the matrix element has a 1/q type


dependence which usually means G≠0 terms are small

= V −1 ∫ dre −ik′⋅rVi (r )e ik⋅r ∫ dr′un* ′,k ′ (r′)un,k (r′) = Vi (q )Ikk ′


nn ′

• The usual argument is that since the u’s are normalized within
a unit cell (i.e. equal to 1), the Bloch overlap integral I, is
approximately 1 for n′=n [interband(valley)]. Therefore, for
impurity scattering, the matrix element for scattering is
approximately
2 2 Z 2e 4
k′ Vi (r ) k = Vi (q ) ≅ 2 2 ; V = volume
( )
2 2
V q + λ εsc
where the scattered wavevector is: q = k − k′
• This is the scattering rate for a single impurity. If we assume
that there are Ni impurities in the whole crystal, and that
scattering is completely uncorrelated between impurities:

kk ′ N i Z 2e 4 ni Z 2e 4
Vi ≅ 2 2 =
( )
V q + λ εsc V q 2 + λ2 εsc
2 2 2
( )
where ni is the impurity density (per unit volume).
From M. Lundstrom, private communication.
• The total scattering rate from k to k′ is given from Fermi’s
golden rule as:
2πni Z 2e 4
Γki k ′ = δ(Ek′ − Ek )
( 2
)
2 2
V q + λ εsc
If θ is the angle between k and k′, then:
q = k − k′ = k 2 + k ′2 − 2kk ′ cos θ = 2k 2 (1 − cos θ )
• Comments on the behavior of this scattering mechanism:
- Increases linearly with impurity concentration
- Decreases with increasing energy (k2), favors lower T
- Favors small angle scattering
- Ionized Impurity-Dominates at low temperature, or room
temperature in impure samples (highly doped regions)
• Integration over all k′ gives the total scattering rate Γk :
n Z 2 4
e m *  4k 2

Γki = i 2 3 3  2 ; qD = 1 / λ
(
8πεsc  k qD 4k 2 + qD2 )
From M. Lundstrom, private communication.
From M. Lundstrom, private communication.
From M. Lundstrom, private communication.
From M. Lundstrom, private communication.
Electron and hole mobility in bulk silicon at T=300K.

Doping Dependence of the Mobility


From M. Lundstrom, private communication.
From M. Lundstrom, private communication.
Neutral Impurity Scattering:

• This scattering mechanism is due to unionized donors, neutral


defects; short range, point-like potential.
• May be modeled as bound hydrogenic potential.
• Usually not strong unless very high concentrations
(>1x1019/cm3).
• It is a dipole type of interaction and is weaker than direct
Coulomb interaction
http://engineering.dartmouth.edu/microeng/otherweb/henning/papers/mobility.pdf
Mobility modeling can be separated in
three parts:
◦ Low-field mobility characterization for
bulk or inversion layers
◦ High-field mobility characterization to
account for velocity saturation effect
◦ Smooth interpolation between the low-field
and high-field regions

Mobility Modeling
Phonon scattering:
- Simple power-law dependence of the temperature
- Sah et al. model:
acoustic + optical and intervalley phonons
combined via Mathiessen’s rule

Ionized impurity scattering:


- Conwell-Weiskopf model
- Brooks-Herring model

Low-Field Models for Bulk Materials


Combined phonon and ionized
impurity scattering:
- Dorkel and Leturg model:
temperature-dependent phonon
scattering +
ionized impurity scattering + carrier-
carrier interactions

- Caughey and Thomas model:


temperature independent phonon
scattering + ionized impurity scattering
- Sharfetter-Gummel model:
phonon scattering + ionized impurity
scattering (parameterized expression –
does not use the Mathiessen’s rule)

- Arora model:
similar to Caughey and Thomas, but with
temperature dependent phonon
scattering
1400

1200

1000
-s
/V-
Mobility cm2/V

800

600

400

200

0
16 17 18 19
10 10 10 10

Doping density [cm-3]


Neutral impurity scattering:
- Li and Thorber model:
mobility component due to neutral
impurity scattering is combined with the
mobility due to lattice, ionized impurity
and carrier-carrier scattering via the
Mathiessen’s rule
The field-dependent mobility model provides smooth transition
between low-field and high-field behavior
µ0
µ (E) = β = 1 for electrons
 µ E β 1 / β
β = 2 for holes
1 +  0  
  vsat  

vsat is modeled as a temperature-dependent quantity:

2.4 × 10 7
vsat (T ) = cm/s
 TL 
1 + 0.8 exp 
 600 

Field-Dependent Mobility

You might also like