Statistics in Semiconductors

1
Statistics of carriers in semiconductors

1. Known facts:
- in solids the allowed energy states are grouped in bands. The
last one is theconduction band. The last but one is the valence band.
Between them lies the forbidden gap with its width E
g
.
- the valence band is almost completely filled with electrons, the
conduction band is almost completely empty (see the computations
below for better estimations). The electrons movement in the valence
band could be expressed by introducing a virtual particle called hole,
which is the absence of an electron.

List of band gaps (from Wikipedia)
Material Sym Band gap (eV) @
2

bol 300K
Silicon Si 1.11
Germanium Ge 0.67
Silicon carbide SiC 2.86
Aluminum
phosphide
AlP 2.45
Aluminium
arsenide
AlAs 2.16
Aluminium
antimonide
AlSb 1.6
Aluminium
nitride
AlN 6.3
Diamond C 5.5
Gallium(III)
phosphide
GaP 2.26
Gallium(III)
arsenide
GaA
s
1.43
Gallium(III)
nitride
GaN 3.4
Gallium(II)
sulfide
GaS 2.5 (@ 295 K)
Gallium
antimonide
GaS
b
0.7
Indium(III)
nitride
InN 0.7
Indium(III)
phosphide
InP 1.35
Indium(III)
arsenide
InAs 0.36
Zinc oxide ZnO 3.37
3

Zinc sulfide ZnS 3.6
Zinc selenide ZnSe 2.7
Zinc telluride
ZnT
e
2.25
Cadmium sulfide CdS 2.42
Cadmium
selenide
CdS
e
1.73
Cadmium
telluride
CdT
e
1.49
Lead(II) sulfide PbS 0.37
Lead(II) selenide PbSe 0.27
Lead(II) telluride PbTe 0.29
Copper(II) oxide
Cu
2
O
2.17

- the real band structure is complex

Band structure
Al
x
Ga
1-x
As for x<0.41-
0.45. Important minima
of the conduction band
and maxima of the
valence band

Carriers have an effective mass which is not equal to the free electron mass. It is
defined by
4

(1)

where ) (k c is the dispersion equation relating the energy with the wave vector.
From relation (1) and the figures above it is obvious that if ) (k c is a convex
function, as e.g. for the holes in the valence band, the effective mass is negative. This
means that a hole moves in the opposite direction than an electron would move in an
external electric field.
There are heavy and light electrons in the conduction band, heavy and light holes
in the valence band. They participate to the conduction process.

2. Density of carriers

We want to compute the carrier density, i.e. the number of charge carriers in the
unit of volume which participate to the conduction process. Task: intersect the Fermi
surface with the dispersion surface ) (k c c = .
Simplified isotropic model:

5

The density of electrons is obtained multiplying the density of levels by the
Fermi function (index n for negative):

1 exp
1
) (
+
(

=
T k
E E
E f
B
c
n
(2)
One gets:
( )
dE
T k
E E
E E
h
m
E n
B
F
c n
1 exp
2
4 ) ( d
2 / 1
2 / 3
2
+
(
|
|
.
|
\
|
=
-
t
c
E E >
(3)
A hole is the absence of an electron, the Fermi function is replaced by (index p
for positive):

(

= =
T k
E E
E f E f
B
F
n p
exp
1
) ( 1 ) (
V
E E < (4)
The result is:

( )
dE
T k
E E
E E
h
m
E p
B
F
V
p
1 exp
2
4 ) ( d
2 / 1
2 / 3
2
+
(
|
|
.
|
\
|
=
-
t
V
E E < (5)

The total concentrations of carriers in the two bands are:

) (
2
2 ) (
2 / 1
2 / 3
2
o
t
F
h
T k m
E dn n
B n
E
C
|
|
.
|
\
|
= =
-
}
(6)

where
T k
E E
B
C F

= o

and
6

) (
2
2 ) (
2 / 1
2 / 3
2
0
|
t
F
h
T k m
E dp p
B p
E
V
|
|
.
|
\
|
= =
-
}
(7)

where
T k
E E
B
F V

= | .

The Fermi function is given by:

x
x
x
F d
1 ] exp[
2
) (
2 / 1
2 / 1
}
+
=
o t
o (8)

For 1 , << | o ,
o
o e F ~ ) (
2 / 1
, and we find for nondegenerate semiconductors:

(

=
(

|
|
.
|
\
|
=
-
T k
E E
N
T k
E E
h
T k m
n
B
C F
C
B
C F B n
exp exp
2
2
2 / 3
2
t
(9)

(

=
(

|
|
.
|
\
|
=
-
T k
E E
N
T k
E E
h
T k m
p
B
F V
V
B
F V
B p
exp exp
2
2
2 / 3
2
t

(10)
N
C
and N
V
are the effective densities of carriers.
In all the preceding relations by m
p
we understand the modulus of m
p
.
Relations (9) and (10) are true for all non-degenerate semiconductors, for which
the Fermi level is in the forbidden gap, at least at 3-5 k
B
T

from the allowed bands.

3. The Fermi level

3.1. Intrinsic semiconductors

Neutrality condition: n=p

7

Result:

-
-
+
+
=
n
p
B
C V
F
m
m
T k
E E
E ln
4
3
2
(11)
The intrinsic concentration:

(
=
T k
E
N N n
B
g
V C i
exp (12)
It depends on T through the exponential factor but also through N
C
and N
V
(see
(9) and (10)).
Useful relation:
2
i
n np = correct for every non-degenerate semiconductor.

3.2. Extrinsic (doped) semiconductors

Donors: group V elements: P, As, Sb. Levels just below E
C
.
Acceptors: group III elements: Al, B, Ga, In. Levels just above E
V
.

Denote the densities of impurities by N
d
and N
a
. The ionized impurities have
densities
+
d
N and
a
N . The neutrality condition is:

+
+ = +
d a
N p N n (13)

8

The densities of ionized impurities, the only one participating to the conduction,
are given by:

1 exp
1
+
(

=
+
T k
E E
g
N
N
B
d F
d
d
and
1 exp +
(
T k
E E
g
N
N
B
F a
a
a
(14)

where g is the statistical weight of the level (we assume g=2).

Type- n semiconductors
All quantities have an index n. Usually
i d
n N >> . The neutrality condition for
non-degenerate semiconductors is
+ +
~ + =
d n d n
N p N n .
+
d
N is the concentration of
ionized donors. These ionized donors have lost their electrons, so their concentration
is given by (14) with g=2. One finds:

1 exp
2
1
exp
+
(

=
(

T k
E E
N
T k
E E
N
B
d F
d
B
C F
C

Denoting
(
=
kT
E
x
F
exp
0 exp exp 2
2
=
(
+
(
d
C
C
C d
C
N x
kT
E
N x
kT
E E
N , or

0 exp
2 2
exp
2
=
(
+
kT
E E
N
N
x
kT
E
x
d C
C
d
d

The result for the unknown x is (we keep only the positive root):

|
|
.
|
\
|
(

+ +
(
=
kT
E E
N
N kT
E
x
d C
C
d
d
exp
8
1 1
4
exp
or
9

|
|
.
|
\
|
(

+ + + =
kT
E E
N
N
kT E E
d C
C
d
d F
exp
8
1 1
4
1
ln (15)
d C
E E is the small difference between the bottom of the conduction band and
the donor levels ( ~0.01 eV).
Limit cases.
At T=0 K the Fermi level is at half the distance between E
C
and E
D
. When T
grows E
F
approaches the conduction band and later drops as in the figure below.

a) Very low temperatures (a few K). 1 exp 8 >>
(

T k
E E
N
N
B
d C
C
d
. We neglect the
unity in (15) and find

|
|
.
|
\
|
+
+
~
C
d B d C
F
N
N T k E E
E
2
ln
2 2
(16)

From (9)
d n
N n :

(
~

T
B
k
d
E
C
E
d C n
N N T n exp 2 / ) low very ( , (17)
10

The graph ( )
T
f n
1
ln = is virtually a straight line with slope |
.
|
\
|
kT
E E
D C
2
.
This is a method to find the position of the donor level below E
C
.
Question: what is the situation in an intrinsic semiconductor?
b) Low temperatures. 1 exp 8 <<
(

T k
E E
N
N
B
d C
C
d
, hence
D C
N N 8 >> . We
develop the square root and eventually find

|
|
.
|
\
|
+ ~
C
d
B C F
N
N
T k E E ln (18)
The argument of ln is negative, so the Fermi level comes close to the middle of
the gap. The e
-
concentration is (from (9)):
d n
N T n ~ ) low ( . This is the depletion
region (srcire in Romanian) when all the donors are ionized and the intrinsic
generation is low. All the impurities are ionized and the concentration of e
-
in the
conduction band equals the donor concentration. The majority carriers have an almost
constant concentration; the concentration of the minority carriers grows rapidly with
T: From
2
i
n np = one finds
n
g
V C
d
i
n
i
n
n
kT
E
N N
N
n
n
n
p <<
(
= = =
2
exp
2 2

For very high temperature, the intrinsic concentration becomes comparable with
the donor concentration and the intrinsic thermal generation supersedes the donor
ionization. The Fermi level moves down towards the center of the forbidden gap.
c) High temperatures. We go back to the neutrality condition and assume all the
donors are ionized
n
i
d n d n
n
n
N p N n
2
+ = + = . The positive solution is:

|
|
.
|
\
|
+ + =
2
2
4
1 1
2
d
i d
n
N
n N
n , and hence (19)

|
|
.
|
\
|
+ +
=
2
2
2
4
1 1
2
d
i
d
i
n
N
n
N
n
p (20)
11

The Fermi level is given by:

|
|
.
|
\
|
(
+ + + =
kT
E
N
N N
N
N
kT E E
g
d
V C
C
d
C F
exp
4
1 1
2
ln
2
(21)
For high T but with 1
4
2
2
<<
d
i
N
n
one find again the depletion region (show!).
Finally, if 1
4
2
2
>>
d
i
N
n
the donor impurities are overtaken by the intrinsic generation.
The majority carrier concentration as function of 1/T is given below:

Type- p semiconductors

Impurities from the III
rd
group e.g. Al, Ga. Extra levels in the gap close to E
V
as
in the drawing below.
a
N is the concentration of ionized acceptors, i.e. states filled
with electrons. They obey a Fermi-Dirac statistics (up to a factor 2 in front of the
exp):
1 exp 2 +
(
T k
E E
N
N
B
F A
a
a
The neutrality condition is
p a p
p N n = +

.
12

Neglecting n
p
one finds the dependence ) (T E
F
as in the figure above. The
analytic results are obtained analogous to the case of the n-type scond. E.g. (15)
becomes:
|
|
.
|
\
|
(

+ + =
kT
E E
N
N
kT E E
V A
V
A
A F
exp
8
1 1
4
1
(15)
Likewise, for high T,
A A
N N =
and
A i
N n >> :

|
|
.
|
\
|
(
+ + =
kT
E
N
N N
N
N
kT E E
g
A
V C
V
A
A F
exp
4
1 1
2
ln
2

|
|
|
.
|
\
|
+ + =
2
2
4
1 1
2
A
i A
p
N
n N
p
Remark. What are the temperatures for which one moves from one region to
another ? They depend strongly on the gap, on effective masses and on impurity
concentrations and as such may vary from 30 K to 3000 K !5

Statistics in Semiconductors

Uploaded by

Copyright:

Available Formats

Statistics in Semiconductors

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Statistics in Semiconductors

Uploaded by

Copyright:

Available Formats

1

Statistics of carriers in semiconductors

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.