PART I SEMICONDUCTOR PHYSICS

CHAPTER 1
ENERGY BANDS AND CARRIER
CONCENTRATION IN THERMAL EQUILIBRIUM
 The physics of semiconductor devices is naturally
dependent on the physics of semiconductor materials
themselves. We will cover important material properties
of semiconductors, especially silicon and gallium
arsenide

 In this chapter we consider :

 Element and compound semiconductors and their basic properties
 The diamond structure and crystal planes
 Valence bonds and energy bands related to carrier concentration in
semiconductors
 Intrinsic carrier concentration
 Donors and acceptors, and Fermi level
 SEMICONDUCTOR MATERIALS
 There are three classes of solid state materials (at 300K)
 Figure 1 shows the typical range of conductivities for insulators,
semiconductors, and conductors

 Insulator: σ (conductivity) ~10-18 - 10-8 Siemens⁄cm ( S⁄cm )

 Semiconductor : σ ~ 10-8 - 103 ( S⁄cm )
 Conductor : σ ~ 104 - 106 ( S⁄cm )
 Semiconductor’s conductivity is sensitive to small amount of impurities
(typically, 1 μg to 1g impurity atoms in 1 kg of semiconductor materials)
 ELEMENT AND COMPOUND
SEMICONDUCTORS

 The element semiconductors, those composed of single species of

atoms, can be found in Column IV. The most important element
semiconductor is silicon (Si) ― over 97% of all semiconductor devices
are made in Si, because Si devices exhibit good properties at room
temperature, high-quality SiO2 can be grown thermally, and silicon in
the form of silica and silicates comprises 25% of the Earth’s crust (a
low cost material).
 ELEMENT AND COMPOUND
SEMICONDUCTORS(Cont.)
 The compound semiconductors are combinations of two or more elements
from the periodic table, such as binary (two elements), ternary (3 elements)
and quaternary (4 elements) compound semiconductors.
SILICON AND COMPOUND SEMICONDUCTORS
 Silicon : > 97% of all semiconductor devices sold (1.12eV)
 III-V Compounds (~2%)
 Binary : GaAs (1.42), InP (1.35), GaN (3.44)
 Ternary : AlxGa1-xAs (1.42 - 2.51)
 Quaternary : GaxIn1-xAsyP1-y (0.36 - 2.2)
 II-VI Compound (~0.1%) CdSe (1.7) ZnS (3.66)
see p.305, Fig. 20
p.553, Appendix F World Production of
Single-Crystal Si
and III-V Compounds
 CRYSTAL STRUCTURE
 Semiconductor materials are single crystals with atoms arranged in a
3D periodic fashion. The periodic arrangement is called a lattice.
 Unit cell
 Unit cell represents the lattice, by repeating the unit cell throughout the
crystal, one can generate the entire lattice. Figure 2 shows a
generalized primitive unit cell.
 A unit cell is characterized by 3 vectors, 𝒂𝒂, 𝒃𝒃, 𝒄𝒄
𝑅𝑅 = m𝒂𝒂 + n𝒃𝒃 + p𝒄𝒄
where m, n, and p are integers
 CUBIC-CRYSTAL UNIT CELLS
 Figure 3 shows three cubic-crystal unit cells. (a) Simple cubic. (b) Body-
centered cubic. (c) Face-centered cubic.
𝔃𝔃 𝔃𝔃 𝔃𝔃

 Single cubic (sc) has 6 equidistant nearest-neighbor atoms

(polonium ― Po)
 Body-centered cubic (bcc) has 8 equidistant nearest-neighbor atoms
(Na, W, etc.)
 Face-centered cubic (fcc) has 12 equidistant nearest-neighbor atoms
(Al, Cu, Au, Pt etc)
 EXAMPLE
 If we pack hard spheres in a bcc lattice so that the atom in the center just
touches the atoms at the corners of the cube, find the fraction of the bcc
unit cell volume filled with hard spheres.

 SOLUTION
Each corner sphere in a bcc unit cell is shared with eight neighboring cells; thus,
each unit cell contains one-eighth of a sphere at each of the eight corners for a
total of one sphere. In addition, each unit cell contains one central sphere. We
have the following :

Spheres (atoms) per unit cell = (1/8) × 8 (corner) + 1 (center) = 2;

Nearest-neighbor distance (along the diagonal AE in Fig. 3b) = 𝑎𝑎 3⁄2;
Radius of each sphere =𝑎𝑎 3⁄4;
Volume of each sphere = 4𝜋𝜋⁄3 × (𝑎𝑎 3⁄4)3 =𝜋𝜋𝑎𝑎3 3⁄16; and
Maximum fraction of unit cell filled = Number of spheres × volume of each
sphere/total volume of each unit cell = 2(𝜋𝜋𝑎𝑎3 3⁄16)⁄𝑎𝑎3 = 𝜋𝜋 3⁄8 ≈ 0.68.
Therefore, about 68% of the bcc unit cell volume is filled with hard spheres,
and about 32% of the volume is empty.
EXAMPLE(Cont.)
 THE DIAMOND STRUCTURE
 Figure 4 shows (a) Diamond lattice. (b) Zincblende lattice.

 Element semiconductors (Ge, Si) have the diamond lattice structure (as two
interpenetration fcc sublattices with one sublattice displaced from the other
𝟏𝟏𝟏𝟏 𝟑𝟑𝟑𝟑
by distance along the diagonal of the cube (i.e., a displacement of 𝒂𝒂𝒂𝒂 ).
𝟒𝟒𝟒𝟒 𝟒𝟒𝟒𝟒
A unit cell consists of a tetrahedron (the spheres each connected by 4
darkened bars)

 Most III-V compound semiconductors (e.g., GaAs) have a zincblende lattice,

which is identical to a diamond lattice except that one fcc sublattice has
column III atoms (Ga) and the other has column V atoms (As).
 CRYSTAL PLANES AND
MILLER INDICES
 To define a crystal plane we use Miller indices :
 Find the intercepts
 Take the reciprocals, reduce to the smallest three integers
 Enclose them in parentheses (hkl) as the Miller indices

 Example :

• a, 3a, 2a
1 1 1 1 1
• a , 3a , 2a , → 1, 3 , 2 → 6, 2, 3

• (632) ― crystal plane

MILLER INDICES OF IMPORTANT PLANE IN
A CUBIC CRYSTAL
 Figure 6 shows the Miller indices of some important planes in a cubic crystal.

 � 𝒌𝒌𝒌𝒌) : negative x-axis. e.g., 𝟏𝟏

(𝒉𝒉 �𝟏𝟏𝟏𝟏 plane
 𝒉𝒉𝒉𝒉𝒉𝒉 : equivalent symmetry plane 𝟏𝟏𝟏𝟏𝟏𝟏  𝟏𝟏𝟏𝟏𝟏𝟏 , (𝟎𝟎𝟎𝟎𝟎𝟎)
�𝟎𝟎𝟎𝟎)
𝟎𝟎𝟎𝟎𝟎𝟎 , (𝟏𝟏
�𝟎𝟎 , (𝟎𝟎𝟎𝟎𝟏𝟏
𝟎𝟎𝟏𝟏 �)
 [𝒉𝒉𝒉𝒉𝒉𝒉] : direction [𝟏𝟏𝟏𝟏𝟏𝟏] for x-axis which is perpendicular to (𝟏𝟏𝟏𝟏𝟏𝟏) plane
 < 𝒉𝒉𝒉𝒉𝒉𝒉 > : equivalent direction < 𝟏𝟏𝟏𝟏𝟏𝟏 >  𝟏𝟏𝟏𝟏𝟏𝟏 , [𝟎𝟎𝟎𝟎𝟎𝟎]
�𝟎𝟎𝟎𝟎]
𝟎𝟎𝟎𝟎𝟎𝟎 , [𝟏𝟏
�𝟎𝟎 , [𝟎𝟎𝟎𝟎𝟏𝟏
𝟎𝟎𝟏𝟏 �]
 VALENCE BONDS
 Figure 7 (a) A tetrahedron bond. (b) Schematic two-dimensional
representation of a tetrahedron bond (c) dwelling time of electrons along
vertical A location A
Time

(c)

 Diamond lattice has tetrahedron bonds.

 Each atoms has 4 electrons in the outer orbit, each atom shares these
electrons (valence electrons) with its four neighbors.
 Covalent bonding: each pair of atoms forms a covalent bond.
 Each electron spends most its time between the two nuclei. Fig. 7c shows
the dwelling time of electrons along the vertical A location.
 The force of attraction for the electrons by both nuclei holds the two atoms
together.
 INTRINSIC SILICON
 Figure 8 shows the basic bond representation of intrinsic silicon. (a) A
broken bond at position A, resulting in a conduction electron and a hole.
(b) A broken bond at position B.

 At low temperatures, no bond is broken.

 At higher temperature, thermal vibrations will break the bonds.
 When a bond is broken, a free electron can participate in current
conduction.
 An electron deficiency is left in the bond.
 The deficiency is call a hole (it carries a positive charge) which can also
participate in current conduction.
 ENERGY BANDS

 Energy Levels
 For an isolated atom, the electron has discrete energy levels, e.g.,

hydrogen atom:
−13.6
𝐸𝐸𝑛𝑛 = eV
𝑛𝑛2

𝐸𝐸1 = −13.6 eV
𝐸𝐸2 = −3.4 eV
𝐸𝐸3 = −1.5 eV

 For N atoms to be brought together to form a solid, each discrete

level will split to form separate but closely spaced levels a
continuous band of energy.
 ENERGY BANDS(Cont.)

 Figure 11 shows the formation of energy bands as a diamond lattice

crystal is formed by bringing isolated silicon atoms together.

-3.14 eV

-13.6 eV
 ENERGY BANDGAP VERSUS TEMPERATURE
 The upper band is called the conduction band, and the lower band is called the
valence band.
 There is an energy bandgap Eg between these two bands. Eg is the energy
required to break a bond in a semiconductor to free an electron to the
conduction band and leave a hole in the valence band.
 The band gap is temperature dependent :
𝟒𝟒. 𝟗𝟗 × 𝟏𝟏𝟏𝟏−𝟒𝟒 𝑻𝑻𝟐𝟐
𝑬𝑬𝒈𝒈 (𝑻𝑻) = 𝟏𝟏. 𝟏𝟏𝟏𝟏 −
(𝑻𝑻 + 𝟔𝟔𝟔𝟔𝟔𝟔) Si
𝟓𝟓. 𝟒𝟒 × 𝟏𝟏𝟏𝟏−𝟒𝟒 𝑻𝑻𝟐𝟐
𝑬𝑬𝒈𝒈 𝑻𝑻 = 𝟏𝟏. 𝟓𝟓𝟓𝟓 − GaAs
(𝑻𝑻 + 𝟐𝟐𝟐𝟐𝟐𝟐)
at 𝑻𝑻 = 𝟎𝟎𝐊𝐊,
𝑬𝑬𝒈𝒈 (𝐒𝐒𝐒𝐒) = 𝟏𝟏. 𝟏𝟏𝟏𝟏 eV
𝑬𝑬𝒈𝒈 (𝐆𝐆𝐆𝐆𝐆𝐆𝐆𝐆) = 𝟏𝟏. 𝟓𝟓𝟓𝟓 eV
at 𝑻𝑻 = 𝟑𝟑𝟑𝟑𝟑𝟑𝟑𝟑,
𝑬𝑬𝒈𝒈 (𝐒𝐒𝐒𝐒) = 𝟏𝟏. 𝟏𝟏𝟏𝟏 eV
0.9
𝑬𝑬𝒈𝒈 (𝐆𝐆𝐆𝐆𝐆𝐆𝐆𝐆) = 𝟏𝟏. 𝟒𝟒𝟒𝟒 eV
𝒅𝒅𝑬𝑬𝒈𝒈
( < 𝟎𝟎 for Si, GaAs)
𝒅𝒅𝒅𝒅
 ENERGY-MOMENTUM DIAGRAM
𝒑𝒑𝟐𝟐
 For free electrons 𝑬𝑬 =
𝟐𝟐𝒎𝒎𝟎𝟎
p = particle momentum, m0 = free electron mass)
 Figure 12 shows the parabolic energy (E) vs.
momentum (p) curve for a free electron.

 For an electron in a semiconductor, we can replace the free electron mass by

an effective mass 𝒎𝒎𝒏𝒏
𝒑𝒑𝟐𝟐 mn = electron effective mass
𝑬𝑬 = mp = hole effective mass
𝟐𝟐𝒎𝒎𝒏𝒏
 If we have the energy-momentum relationship, we can obtain 𝒎𝒎𝒏𝒏:
𝟐𝟐 𝟐𝟐 −𝟏𝟏
𝒅𝒅𝑬𝑬 𝒑𝒑 𝒅𝒅 𝒅𝒅𝒅𝒅 𝒅𝒅 𝑬𝑬 𝟏𝟏 𝒅𝒅 𝑬𝑬
= → = 𝟐𝟐
= → 𝒎𝒎𝒏𝒏 =
𝒅𝒅𝒅𝒅 𝒎𝒎𝒏𝒏 𝒅𝒅𝒅𝒅 𝒅𝒅𝒅𝒅 𝒅𝒅𝒅𝒅 𝒎𝒎𝒏𝒏 𝒅𝒅𝒑𝒑𝟐𝟐
Therefore, the narrower the parabola, corresponding to a larger second
derivative, the smaller the effective mass.
 ENERGY BAND STRUCTURE OF Si AND GaAs

 Figure 14 shows the energy band structures of (a) Si and (b) GaAs.
Circles (o) indicate holes in the valence bands and dots (•) indicate
electrons in the conduction bands.
 ENERGY BAND STRUCTURE OF Si AND
GaAs(Cont.)
 Near the minimum of the conduction band or the maximum of the
valence band, the E-p curves are essentially parabolic, we can thus
obtain mn and mp.

 Silicon is called a indirect semiconductor, because the maximum in

the valence band occur at p=0, but the minimum in the conduction
band occur at pǂ0.

 GaAs is called a direct semiconductor, because the maximum and

minimum both occur at p=0. Thus an electron making a transition from
the valence band to the conduction band can do so without a change
in momentum.
 CONDUCTION IN METAL, SEMICONDUCTOR,
AND INSULATOR
 Figure 15 shows the schematic energy band representations of (a) a
conductor with two possibilities (either the partially filled conduction band
shown at the upper portion or the overlapping bands shown at the lower
portion), (b) a semiconductor, and (c) an insulator.
 CONDUCTION IN METAL, SEMICONDUCTOR,
AND INSULATOR(Cont.)

 Metal no bandgap. Electrons at the uppermost levels can

move freely for current conduction, thus high
conductivity.

 Semiconductor: smaller bandgap (1.12eV for Si). At room

temperature, the thermal energy can cause some
bond-breaking for electron transition from the
valence band to the conduction band, giving to a
moderate conductivity.

 Insulator: large bandgap (9eV for SiO2). The bonds between

neighboring atoms are difficult to break, no free
electrons for current conduction, thus very low
conductivity.
 INTRINSIC CARRIER CONDUCTION

 Intrinsic semiconductor: small amount of impurity compared to

thermally generated electrons and holes.

 Electron density in 𝒅𝒅𝒅𝒅 ≡ 𝒏𝒏 𝑬𝑬 = 𝑵𝑵(𝑬𝑬) (Density of states or the density

of allowed energy states/cm𝟑𝟑 ) X 𝑭𝑭(𝑬𝑬) (probability of occupying that
energy range).
∴ 𝒏𝒏 𝑬𝑬 = 𝑵𝑵 𝑬𝑬 𝑭𝑭(𝑬𝑬)

 Total electron density in conduction band = number of electrons per

cm3:
𝑬𝑬𝒕𝒕𝒕𝒕𝒕𝒕𝑬𝑬𝒕𝒕𝒕𝒕𝒕𝒕 𝑬𝑬𝑬𝑬
𝒕𝒕𝒕𝒕𝒕𝒕
𝒕𝒕𝒕𝒕𝒕𝒕
𝒏𝒏 =𝒏𝒏�= � 𝒏𝒏 𝒏𝒏
𝑬𝑬 𝑬𝑬𝒅𝒅𝒅𝒅
𝒅𝒅𝒅𝒅==�� 𝑵𝑵𝑵𝑵 𝑬𝑬𝑬𝑬 𝑭𝑭 𝑬𝑬 𝒅𝒅𝒅𝒅
𝑭𝑭 𝑬𝑬 𝒅𝒅(𝑬𝑬)
𝑬𝑬𝒄𝒄 𝑬𝑬𝒄𝒄 𝑬𝑬𝑬𝑬
𝒐𝒐 𝒄𝒄
 DERIVATION OF THE DENSITY OF STATES

semiconductor
 Consider a semiconductor bar with length L
𝑳𝑳
𝑳𝑳
=
= 𝒏𝒏𝒏𝒏𝒙𝒙𝒙𝒙 (integer) λ
𝝀𝝀
𝝀𝝀
𝒉𝒉
𝒉𝒉
𝛌𝛌𝛌𝛌 =
= de broglie hypothesis, px is the momentum in x direction
𝒑𝒑
𝒑𝒑𝒙𝒙𝒙𝒙
∴ 𝑳𝑳𝒑𝒑𝒙𝒙 = 𝒉𝒉𝒏𝒏𝒙𝒙
𝑳𝑳𝒅𝒅𝒑𝒑𝒙𝒙 = 𝒉𝒉 (incremental momentum dpx for a unity increase of nx)
∴ 𝑳𝑳𝟑𝟑 𝒅𝒅𝒑𝒑𝒙𝒙 𝒅𝒅𝒑𝒑𝒚𝒚 𝒅𝒅𝒑𝒑𝒛𝒛 = 𝒉𝒉𝟑𝟑 (3 dimensional cubic of side L)
Let L=1 ∴ 𝒅𝒅𝒑𝒑𝒙𝒙 𝒅𝒅𝒑𝒑𝒚𝒚 𝒅𝒅𝒑𝒑𝒛𝒛 = 𝒉𝒉𝟑𝟑

 The volume 𝒅𝒅𝒅𝒅𝒙𝒙𝒅𝒅𝒅𝒅𝒚𝒚𝒅𝒅𝒅𝒅𝒛𝒛 in the momentum space for a unit cube

(L=1) is h3. Each incremental change in n corresponds to a unique set
of integers (𝒏𝒏𝒙𝒙, 𝒏𝒏𝒚𝒚, 𝒏𝒏𝒛𝒛)  corresponds to an allowed energy state.
Therefore, the volume in momentum space for an energy state is h3.
 MOMENTUM SPACE
 The volume between two concentrate spheres (𝑝𝑝 𝑝𝑝 + 𝑑𝑑𝑑𝑑) is 4𝜋𝜋𝜌𝜌2 𝑑𝑑𝑑𝑑
 Number of energy states =𝑁𝑁(𝐸𝐸)𝑑𝑑𝑑𝑑
2 3 3
𝑝𝑝𝑧𝑧
= 2 4𝜋𝜋𝑝𝑝 𝑑𝑑𝑑𝑑 /ℎ = 8𝜋𝜋𝑝𝑝 𝑝𝑝𝑝𝑝𝑝𝑝 /ℎ
𝑑𝑑𝑑𝑑
electron spin up & down
θ
𝑝𝑝2 p
𝒑𝒑𝒑𝒑𝒑𝒑 𝑝𝑝𝑦𝑦
∴ ∵𝒅𝒅𝒅𝒅
𝐸𝐸 =
= 2𝑚𝑚 , 𝑝𝑝 = 2𝑚𝑚𝑛𝑛 𝐸𝐸
𝒎𝒎𝒏𝒏𝑛𝑛 φ
∴∴𝑵𝑵𝑑𝑑𝑑𝑑 𝒅𝒅𝒅𝒅𝑝𝑝𝑝𝑝𝑝𝑝
𝑬𝑬 = ∴
𝑚𝑚𝑛𝑛 𝑝𝑝𝑥𝑥 The momentum space in
3 spherical coordinates.
8𝜋𝜋 2𝑚𝑚𝑛𝑛 𝐸𝐸 𝑚𝑚𝑛𝑛 𝑑𝑑𝑑𝑑 2𝑚𝑚𝑛𝑛 2
∴ 𝑁𝑁 𝐸𝐸 𝑑𝑑𝑑𝑑 = = 4𝜋𝜋 𝐸𝐸𝑑𝑑𝑑𝑑
ℎ3 ℎ2
3
N(E) E
2𝑚𝑚𝑛𝑛 2
𝑁𝑁 𝐸𝐸 = 4𝜋𝜋 𝐸𝐸
ℎ2
~ E
Density of states, i.e., the density of E N(E)
allowed energy states per unit volume
 FERMI DISTRIBUTION FUNCTION
1 𝟏𝟏
 𝑭𝑭(𝑬𝑬) is the Fermi distribution function given by 𝑭𝑭𝑭𝑭 𝑬𝑬
𝑬𝑬 =
=
𝟏𝟏1+exp[(𝑬𝑬−𝑬𝑬 − 𝑬𝑬𝑭𝑭 )/𝒌𝒌𝒌𝒌]
+ 𝐞𝐞𝐞𝐞𝐞𝐞[(𝑬𝑬𝑭𝑭)/𝒌𝒌𝒌𝒌]
where k is the Boltzmann constant, 𝑬𝑬𝑭𝑭 is the Fermi level.
 Figure 16 shows the Fermi distribution function 𝐹𝐹(𝐸𝐸) versus (𝐸𝐸 – 𝐸𝐸𝐹𝐹 ) for
various temperatures.

 At 𝑬𝑬 = 𝑬𝑬𝑭𝑭 , the probability of occupation by an electron in exactly 0.5.

 The Fermi distribution is symmetrical around 𝑬𝑬𝑭𝑭
𝑬𝑬−𝑬𝑬
𝑬𝑬−𝑬𝑬 𝑬𝑬−𝑬𝑬𝑭𝑭
𝑭𝑭 𝑭𝑭 −(𝑬𝑬−𝑬𝑬
−(𝑬𝑬−𝑬𝑬 ) 𝑭𝑭𝑬𝑬−𝑬𝑬
𝑭𝑭− ) 𝑭𝑭
� � �𝒌𝒌𝒌𝒌 �
𝐅𝐅𝐅𝐅𝐅𝐅 (𝑬𝑬
(𝑬𝑬−−𝑬𝑬𝑬𝑬𝑭𝑭𝑭𝑭)) >> 𝟑𝟑𝟑𝟑𝟑𝟑,
𝟑𝟑𝟑𝟑𝟑𝟑,𝒆𝒆𝒆𝒆 𝒆𝒆 𝒌𝒌𝒌𝒌 >>
>>𝟏𝟏,𝟏𝟏,
>>
𝒌𝒌𝒌𝒌𝒌𝒌𝒌𝒌 𝑭𝑭(𝑬𝑬)
𝟏𝟏, →→𝒆𝒆 𝒆𝒆→ 𝒆𝒆
𝑭𝑭(𝑬𝑬)
𝑭𝑭(𝑬𝑬) 𝒌𝒌𝒌𝒌
𝒌𝒌𝒌𝒌
𝑬𝑬−𝑬𝑬
𝑬𝑬−𝑬𝑬 𝑭𝑭 � 𝑭𝑭 −(𝑬𝑬𝑭𝑭 −𝑬𝑬) 𝑬𝑬𝑭𝑭 −𝑬𝑬
− �𝒌𝒌𝒌𝒌
𝐅𝐅𝐅𝐅𝐅𝐅 𝑬𝑬𝑬𝑬−−𝑬𝑬𝑬𝑬𝑭𝑭𝑭𝑭 <<−−𝟑𝟑𝟑𝟑𝟑𝟑,
𝟑𝟑𝟑𝟑𝟑𝟑,𝒆𝒆 𝒆𝒆 𝒌𝒌𝒌𝒌𝒌𝒌𝒌𝒌 << 𝟏𝟏, 𝑭𝑭(𝑬𝑬)𝑭𝑭(𝑬𝑬)
→ 𝟏𝟏→− 𝒆𝒆𝟏𝟏 − 𝒆𝒆 𝒌𝒌𝒌𝒌
 INTRINSIC SEMICONDUCTOR

 Figure 17 shows the intrinsic semiconductor. (a) Schematic band

diagram. (b) Density of states. (c) Fermi distribution function. (d)
Carrier concentration
𝐸𝐸

 The electron concentration can be obtained graphically from the

product of N(E) in Fig.17b and F(E) in Fig.17c. The result is shown in
Fig.17d for n(E) versus E curve (upper curve).
 The upper shaded area in Fig.17d is the electron density.
 The lower shaded area in Fig.17d is the hole density which is equal to
the electron density for an intrinsic semiconductor.
 ELECTRON DENSITY IN
CONDUCTION BAND
 Derivation of the electron density in the conduction band:
3
𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡
2𝑚𝑚𝑛𝑛 2 1 1
𝑛𝑛 = � 𝑁𝑁 𝐸𝐸 𝐹𝐹 𝐸𝐸 𝑑𝑑𝑑𝑑 = � 4𝜋𝜋 𝐸𝐸 2 𝑑𝑑𝑑𝑑
ℎ2 𝐸𝐸 − 𝐸𝐸𝐹𝐹
𝐸𝐸𝑐𝑐 𝐸𝐸𝑐𝑐 1 + exp
𝑘𝑘𝑘𝑘
3
∞
2𝑚𝑚𝑛𝑛 2 1
≅ � 4𝜋𝜋 𝐸𝐸 2 exp [−(𝐸𝐸 − 𝐸𝐸𝐹𝐹 )/𝑘𝑘𝑘𝑘)]𝑑𝑑𝑑𝑑
0 ℎ2
Let 𝑥𝑥 = 𝐸𝐸/ 𝑘𝑘𝑘𝑘 3
∞
2𝑚𝑚𝑛𝑛 2 1 𝐸𝐸𝐹𝐹
∴ 𝑛𝑛 = 4𝜋𝜋 � 𝑥𝑥𝑥𝑥𝑥𝑥 2 𝑒𝑒 −𝑥𝑥 𝑒𝑒 𝑘𝑘𝑘𝑘 𝑑𝑑(𝑥𝑥𝑥𝑥𝑥𝑥)
ℎ2 0
3
∞
2𝑚𝑚𝑛𝑛 2 3 𝐸𝐸𝐹𝐹 1
= 4𝜋𝜋 𝑘𝑘𝑘𝑘 2 𝑒𝑒 𝑘𝑘𝑘𝑘 � 𝑥𝑥 2 𝑒𝑒 −𝑥𝑥 𝑑𝑑𝑑𝑑
ℎ2 0

Γ( 23 )
 ELECTRON DENSITY IN CONDUCTION
BAND(Cont.)
 Gamma Function
∞ 𝒏𝒏−𝟏𝟏 −𝒙𝒙
𝚪𝚪(𝒏𝒏)
𝛤𝛤(𝑛𝑛) ≡�
≡ ∫𝟎𝟎 𝑥𝑥𝒙𝒙𝑛𝑛−1 𝒆𝒆𝑒𝑒 −𝑥𝑥𝒅𝒅𝒅𝒅
𝑑𝑑𝑑𝑑
𝟏𝟏 ∞0 𝟏𝟏
Γ ≡ ∫𝟎𝟎 ∞𝒙𝒙− �1𝟐𝟐 𝒆𝒆−𝒙𝒙 𝒅𝒅𝒅𝒅 = 𝝅𝝅
1
𝟐𝟐
𝛤𝛤 𝟑𝟑 ≡ �∞ 𝑥𝑥𝟏𝟏 −2 𝑒𝑒 −𝑥𝑥 𝑑𝑑𝑑𝑑 = 𝟑𝟑 𝜋𝜋 𝟑𝟑 𝟏𝟏
Γ 2 ≡ ∫𝟎𝟎0 𝒙𝒙 �𝟐𝟐 𝒆𝒆−𝒙𝒙 𝒅𝒅𝒅𝒅 = − 𝟏𝟏 Γ − 𝟏𝟏 = 𝝅𝝅
𝟐𝟐 𝟐𝟐 𝟐𝟐 𝟐𝟐
∞ 1
3 −𝑥𝑥
3 3 1
𝛤𝛤 ≡ � 𝑥𝑥 𝑒𝑒 𝑑𝑑𝑑𝑑 =
2 −1 𝛤𝛤 −1 = 𝜋𝜋
2 0 2 2 2
 Electron density
3
2𝑚𝑚𝑛𝑛 2 3 𝐸𝐸𝐹𝐹 1
𝑛𝑛 = 4𝜋𝜋 𝑘𝑘𝑘𝑘 2 𝑒𝑒 𝑘𝑘𝑘𝑘 𝜋𝜋
ℎ2 2
3
2𝜋𝜋𝑚𝑚𝑛𝑛 𝑘𝑘𝑘𝑘 2
𝑛𝑛 = 2 exp 𝐸𝐸𝐹𝐹 ⁄𝑘𝑘𝑘𝑘
ℎ2
 EFFECTIVE DENSITY OF STATES
 Refer bottom of conduction band as 𝑬𝑬𝑪𝑪.
3 3�
2𝜋𝜋𝑚𝑚 2
2𝜋𝜋𝑚𝑚 𝑘𝑘𝑘𝑘 2 𝑛𝑛𝑘𝑘𝑘𝑘 𝐸𝐸 − 𝐸𝐸 𝐸𝐸𝑐𝑐 − 𝐸𝐸𝐹𝐹 𝐸𝐸 − 𝐸𝐸 𝐸𝐸𝑐𝑐 − 𝐸𝐸𝐹𝐹
𝑛𝑛 = 2 𝑛𝑛 =22 − 𝑒𝑒𝑒𝑒𝑒𝑒(− =𝑘𝑘𝑘𝑘 )=−
𝑁𝑁𝑐𝑐 𝑒𝑒𝑒𝑒𝑒𝑒 (− )
𝑛𝑛 𝑐𝑐 𝐹𝐹 𝑐𝑐 𝐹𝐹
exp
ℎ 2 𝑁𝑁𝑐𝑐 exp 𝑘𝑘𝑘𝑘
ℎ 𝑘𝑘𝑘𝑘
3
𝑘𝑘𝑘𝑘
2𝜋𝜋𝑚𝑚𝑛𝑛 𝑘𝑘𝑘𝑘 2
Where 𝑁𝑁𝑐𝑐 ≡ 2 for one minimum
ℎ2

 Silicon has 6 minimum along <100>, and “GaAs” has only one
minimum at p=0 z
3
𝟑𝟑
2𝜋𝜋𝑚𝑚𝑛𝑛𝑛𝑛𝑘𝑘𝑘𝑘 𝟐𝟐2
2𝜋𝜋𝑚𝑚 𝑘𝑘𝑘𝑘 y
𝑵𝑵𝑐𝑐𝒄𝒄 ≡≡2𝟐𝟐
𝑁𝑁 2 ∙∙ 𝑴𝑴
𝑀𝑀𝑐𝑐𝒄𝒄
ℎℎ2
-x x
Mc = 6 for Si
Mc = 1 for GaAs -y
-z

 𝑵𝑵𝒄𝒄 is the effective density of states in the conduction band. At room

temperature, 𝑁𝑁𝑐𝑐 = 2.86 × 1019 cm−3 for Si
𝑁𝑁𝑐𝑐 = 4.7 × 1017 cm−3 for GaAs
 INTRINSIC FERMI LEVEL

 For holes in the valence band

𝐸𝐸𝐹𝐹 − 𝐸𝐸𝑣𝑣
∴ 𝑝𝑝 = 𝑁𝑁𝑣𝑣 exp −
𝑘𝑘𝑘𝑘
3
2𝜋𝜋𝑚𝑚𝑝𝑝 𝑘𝑘𝑘𝑘 2
𝑁𝑁𝑣𝑣 ≡ 2 = effective density of states in the valence band
ℎ2

At 300K

𝑁𝑁𝑣𝑣 = 2.66 × 1019 cm−3 for Si

𝑁𝑁𝑣𝑣 = 7 × 1018 cm−3 for GaAs
 INTRINSIC FERMI LEVEL(Cont.)

 For intrinsic semiconductor 𝑛𝑛 = 𝑝𝑝 = 𝑛𝑛𝑖𝑖 (intrinsic carrier concentration)

∵ 𝑛𝑛 = 𝑝𝑝
𝐸𝐸𝑬𝑬 −𝐸𝐸
−𝑬𝑬 𝐸𝐸 −𝐸𝐸 𝑬𝑬 −𝑬𝑬 (𝐸𝐸 +𝐸𝐸𝑣𝑣 � 𝑬𝑬 2𝐸𝐸
+𝑬𝑬 2𝐸𝐸𝐹𝐹 𝑁𝑁 𝑬𝑬
−−𝐸𝐸𝑐𝑐𝑐𝑐𝒄𝒄−𝐸𝐸𝐹𝐹
𝐹𝐹𝑭𝑭 − −𝐹𝐹𝐸𝐸− 𝑣𝑣 𝑭𝑭𝑣𝑣 𝒗𝒗 − 𝑐𝑐 𝐸𝐸𝑐𝑐 +𝐸𝐸𝑣𝑣 𝒄𝒄 𝐹𝐹𝒗𝒗
𝐹𝐹 −𝐸𝐸 − 𝑭𝑭
𝒒𝒒𝑣𝑣 𝒌𝒌𝒌𝒌
𝑁𝑁 𝑵𝑵𝒗𝒗
∴∴ 𝑵𝑵𝑐𝑐𝒄𝒄𝑐𝑐𝑒𝑒𝒆𝒆 𝑘𝑘𝑘𝑘𝒌𝒌𝒌𝒌 ==𝑁𝑁𝑣𝑣𝑵𝑵
∴ 𝑁𝑁 𝑁𝑁 𝑒𝑒
− 𝑘𝑘𝑘𝑘 = 𝑁𝑁
𝑣𝑣 𝑒𝑒𝒗𝒗 𝒆𝒆 𝑘𝑘𝑘𝑘 ⟹ 𝑒𝑒
𝑒𝑒 𝑘𝑘𝑘𝑘 ⟹𝒌𝒌𝒌𝒌 𝑒𝑒 ⟹ 𝒆𝒆𝑘𝑘𝑘𝑘 ∙𝒌𝒌𝒌𝒌𝑒𝑒 𝑘𝑘𝑘𝑘∙ 𝒆𝒆𝑁𝑁= 𝑣𝑣
−𝑘𝑘𝑘𝑘 ∙ 𝑒𝑒 𝑘𝑘𝑘𝑘 = =
𝑐𝑐 𝑁𝑁𝑐𝑐 𝑵𝑵𝒄𝒄
𝐸𝐸𝐸𝐸𝑐𝑐𝑐𝑐+
𝑬𝑬𝒄𝒄𝐸𝐸
+𝐸𝐸 +𝑬𝑬 𝑘𝑘𝑘𝑘 𝑁𝑁 𝐸𝐸𝑐𝑐 +
𝑣𝑣𝑣𝑣 𝒗𝒗 𝑘𝑘𝑘𝑘 𝒌𝒌𝒌𝒌 𝑁𝑁𝑣𝑣𝑣𝑣 𝑵𝑵𝒗𝒗𝐸𝐸𝑐𝑐 +𝐸𝐸 𝑚𝑚𝑝𝑝𝑚𝑚𝑝𝑝 𝒎𝒎𝒑𝒑
𝐸𝐸 𝑘𝑘𝑘𝑘 𝒗𝒗𝑘𝑘𝑘𝑘 𝒌𝒌𝒌𝒌
𝑣𝑣 𝑬𝑬𝒄𝒄𝑣𝑣+𝑬𝑬
∴ 𝑬𝑬𝐹𝐹𝑭𝑭𝐹𝐹≡≡ 𝐸𝐸𝑬𝑬
∴ 𝐸𝐸
∴ 𝐸𝐸 = 𝐸𝐸𝑖𝑖𝑖𝑖 𝒊𝒊== 2
=
2 𝟐𝟐
+ + ℓ𝑛𝑛 ln𝑁𝑁 = 2= + ℓ𝑛𝑛+ 𝑚𝑚 ln
+ 𝑙𝑙𝑙𝑙
22 𝟐𝟐 𝑁𝑁𝑐𝑐𝑐𝑐 𝑵𝑵𝒄𝒄
= + 𝑙𝑙𝑙𝑙
2 𝟐𝟐2 2 𝟐𝟐𝑛𝑛𝑚𝑚𝑛𝑛 𝒎𝒎𝒏𝒏
𝒎𝒎𝒑𝒑𝑚𝑚𝑝𝑝
If 𝑚𝑚𝑝𝑝 = 0.5 𝑚𝑚0, 𝑚𝑚𝑛𝑛 = 0.063 𝑚𝑚0 ∴∴ln𝑙𝑙𝑙𝑙𝒎𝒎 = =ln𝑙𝑙𝑙𝑙𝟖𝟖8== 𝟐𝟐2
𝑚𝑚𝑛𝑛
𝒏𝒏
3𝑘𝑘𝑘𝑘
𝟑𝟑𝟑𝟑𝟑𝟑 𝐸𝐸𝑔𝑔 𝑬𝑬𝒈𝒈
∴∴ ln(𝟐𝟐) ≈ 𝟎𝟎. 𝟎𝟎𝟎𝟎 ≪
𝑙𝑙𝑙𝑙(2) ≈ 0.04 ≪
𝟒𝟒4 2 𝟐𝟐
The intrinsic Fermi level Ei lies very close to the middle of the bandgap.
Ec
~Eg /2
EF

Ev
 INTRINSIC CARRIER CONCENTRATION

 Figure 18 shows the intrinsic carrier densities in Si and GaAs as a

function of the reciprocal of temperature.

𝑛𝑛𝑛𝑛 = 𝑛𝑛𝑖𝑖2
𝒏𝒏𝟐𝟐
2 = 𝑵𝑵 𝑵𝑵 exp − 𝐸𝐸𝑬𝑬𝑐𝑐𝒄𝒄 −−𝐸𝐸𝑬𝑬𝒗𝒗
𝑣𝑣 = 𝑵𝑵 𝑵𝑵 exp(− 𝐸𝐸𝑬𝑬𝑔𝑔𝒈𝒈 )
𝑛𝑛𝒊𝒊𝑖𝑖 = 𝑁𝑁𝑐𝑐𝒄𝒄 𝑁𝑁𝑣𝑣𝒗𝒗 exp − 𝒌𝒌𝒌𝒌 = 𝑁𝑁𝑐𝑐 𝒄𝒄𝑁𝑁𝑣𝑣 𝒗𝒗exp − 𝒌𝒌𝒌𝒌
𝑘𝑘𝑘𝑘
𝑬𝑬𝑬𝑬 𝑘𝑘𝑘𝑘
∴ 𝒏𝒏𝒊𝒊 = 𝑵𝑵𝒄𝒄 𝑵𝑵𝒗𝒗 exp − 𝐸𝐸𝑔𝑔
∴ 𝑛𝑛𝑖𝑖 = 𝑁𝑁𝑐𝑐 𝑁𝑁𝑣𝑣 exp − 𝟐𝟐𝟐𝟐𝟐𝟐
𝟑𝟑� 2𝑘𝑘𝑘𝑘
𝟐𝟐𝜋𝜋𝒎𝒎𝒏𝒏 𝒌𝒌𝒌𝒌 𝟐𝟐 𝑬𝑬𝒈𝒈
= 4 𝟐𝟐
3 exp(− )3
𝒉𝒉
2𝜋𝜋𝑚𝑚𝑛𝑛 𝑘𝑘𝑘𝑘 2 𝟐𝟐𝟐𝟐𝟐𝟐
2𝜋𝜋𝑚𝑚𝑝𝑝 𝑘𝑘𝑘𝑘 2 𝐸𝐸𝑔𝑔
= 4 𝑀𝑀𝑐𝑐 exp −
ℎ2 ℎ2 2𝑘𝑘𝑘𝑘

𝑛𝑛𝑖𝑖 = 9.65 × 109 cm−3 for Si at 300K

𝑛𝑛𝑖𝑖 = 2.25 × 106 cm−3 for GaAs at 300K
 DONORS AND ACCEPTORS FOR
EXTRINSIC SEMICONDUCTOR
 Figure 19 shows the schematic bond pictures for (a) n-type Si with
donor (arsenic) and (b) p-type Si with acceptor (boron).
 DONORS AND ACCEPTORS FOR
EXTRINSIC SEMICONDUCTOR(Cont.)

 Donor can donate an electron to the conduction band.

 Acceptor can accept an electron to create a hole in the valence band.
 To estimate the ionization energy, we use the hydrogen model, Eq.2:

2
𝝐𝝐 𝟐𝟐𝜖𝜖0 𝒎𝒎
𝑚𝑚𝑛𝑛
𝐸𝐸
=𝐷𝐷 =𝟎𝟎 EH 𝐸𝐸𝐻𝐻
𝒏𝒏
𝜖𝜖𝑠𝑠 𝒎𝒎𝟎𝟎 𝑚𝑚
𝝐𝝐𝒔𝒔 0
= 0.025 eV Si
= 0.007 eV GaAs

𝐸𝐸𝐴𝐴 = 0.05 eV Si
= 0.03 eV GaAs
 IONIZATION ENERGIES
 Figure 20 shows the measured ionization energies (in eV) for various
impurities in Si and GaAs. The levels below the gap center are measured
from the top of the valence band and are acceptor levels unless labeled by
D for donor level. The levels above the gap center are measured from the
bottom of the conduction band and are donor levels unless indicated by A
for acceptor level.

 The most important donor-type impurities in Si are P and As.

 The most important acceptor-type impurity in Si is B.
 A single impurity may have multiple level; for example oxygen in Si has
two donor levels and two acceptor levels; Au in Si has one donor and one
acceptor level.
 NONDEGENERATE SEMICONDUCTORS

 A nondegenerate semiconductor has electron or hole concentration

much lower than 𝑵𝑵𝒄𝒄 or 𝑵𝑵𝒗𝒗 . Therefore, the Fermi level is at least 3kT
below 𝑬𝑬𝒄𝒄 or 3kT above 𝑬𝑬𝒗𝒗 .
 For shallow donors in n-type semiconductor (donor level 𝑬𝑬𝑫𝑫), there is
enough thermal energy to supply the energy 𝑬𝑬𝑫𝑫 to ionize all donor
impurities at 𝟑𝟑𝟑𝟑𝟑𝟑𝐊𝐊. 𝒏𝒏 = ND = donor concentration.
𝐸𝐸𝑐𝑐 − 𝐸𝐸𝐹𝐹 𝑭𝑭
𝑬𝑬𝒄𝒄 −𝑬𝑬
 From Eq.13 𝑛𝑛 = 𝒏𝒏 𝑁𝑁
=𝑐𝑐 exp
𝑵𝑵𝒄𝒄 exp(−
− ), we have
𝑘𝑘𝑘𝑘𝒌𝒌𝒌𝒌
𝑬𝑬 −
𝑵𝑵𝑫𝑫 = 𝑵𝑵𝒄𝒄 exp(− 𝒄𝒄𝐸𝐸𝑐𝑐)− 𝐸𝐸𝐹𝐹
𝑁𝑁𝐷𝐷 = 𝑁𝑁𝑐𝑐 exp −𝒌𝒌𝒌𝒌
𝑘𝑘𝑘𝑘
𝑁𝑁𝑐𝑐
𝐸𝐸𝑐𝑐 − 𝐸𝐸𝐹𝐹 = 𝑘𝑘𝑘𝑘 ℓ𝑛𝑛
𝑁𝑁𝐷𝐷
 Similarly we have for shallow acceptor with concentration NA:
𝑁𝑁𝑣𝑣
𝐸𝐸𝐹𝐹 − 𝐸𝐸𝑉𝑉 = 𝑘𝑘𝑘𝑘 ℓ𝑛𝑛
𝑁𝑁𝐴𝐴
 COMPLETE IONIZATION

 Figure 21 shows schematic energy band representation of

extrinsic semiconductors with (a) donor ions and (b) acceptor
ions.

( EC − EF ) at𝑬𝑬E𝑫𝑫D
EF
EF
at𝑬𝑬E𝑨𝑨A ( EF − EV )
 COMPLETE IONIZATION(Cont.)

 For complete ionization n = ND (n-type)

𝑵𝑵
𝑁𝑁𝑐𝑐𝑪𝑪
Since 𝑬𝑬𝐸𝐸𝑪𝑪𝑐𝑐 −
− 𝑬𝑬
𝐸𝐸𝑭𝑭𝐹𝐹 = 𝒌𝒌
𝑘𝑘 𝑻𝑻 ln
𝑇𝑇 ℓ𝑛𝑛
𝑵𝑵𝐷𝐷𝑫𝑫
𝑁𝑁
𝑁𝑁𝑐𝑐
as 𝑁𝑁𝐷𝐷 ↑ ℓ𝑛𝑛 ln ↓
𝑁𝑁𝐷𝐷

∴ (𝑬𝑬𝒄𝒄 − 𝑬𝑬𝑭𝑭 ) ↓ or 𝑬𝑬𝑭𝑭 is closer to the bottom of the conduction band.

𝑁𝑁𝑐𝑐
 Similarly, as 𝑁𝑁𝐴𝐴 ↑ ℓ𝑛𝑛 ↓
𝑁𝑁𝐴𝐴

(𝑬𝑬𝑭𝑭 − 𝑬𝑬𝒗𝒗 ) ↓ or 𝑬𝑬𝑭𝑭 is closer to the top of the valence band.

 N-TYPE SEMICONDUCTOR
 Figure 22 shows n-type semiconductor. (a)Schematic band diagram. (b)
Density of states. (c) Fermi distribution function. (d)Carrier concentration.
Note that 𝒏𝒏𝒏𝒏 = 𝒏𝒏𝒊𝒊𝟐𝟐

 For an n-type nondegenerate semiconductor , because of the ionized

donors, the electron concentration ( upper shaded area in the conduction
band, Fig.22 d) is much larger than the hole concentration (lower shaded
area) in the valence band.
 Similarly for a p-type nondegenerate semiconductor, the hole concentration
in the valence band will be much larger then the electron concentration in
the conduction band.
 MASS ACTION LAW
 For intrinsic semiconductors, we have from Equations 13 and 14 with
𝑬𝑬𝑭𝑭 → 𝑬𝑬𝒊𝒊 𝐸𝐸𝑐𝑐 − 𝐸𝐸𝑖𝑖 𝐸𝐸𝑖𝑖 − 𝐸𝐸𝑣𝑣
𝑛𝑛𝑖𝑖 = 𝑁𝑁𝑐𝑐 exp − or 𝑛𝑛𝑖𝑖 = 𝑁𝑁𝑣𝑣 exp −
𝒏𝒏𝒊𝒊 = 𝑵𝑵𝒄𝒄 𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘
 For extrinsic semiconductors：
𝐸𝐸𝑬𝑬𝑐𝑐𝒄𝒄−
−𝑬𝑬𝐸𝐸𝑭𝑭𝐹𝐹 𝐸𝐸𝑬𝑬𝑐𝑐𝒄𝒄 −𝑬𝑬
− 𝐸𝐸
𝒊𝒊 𝑖𝑖 𝑬𝑬𝑭𝑭𝐸𝐸𝐹𝐹 𝒊𝒊−
−𝑬𝑬 𝐸𝐸𝑖𝑖 𝐸𝐸𝐹𝐹𝒊𝒊 − 𝐸𝐸𝑖𝑖
𝑬𝑬𝑭𝑭 −𝑬𝑬
𝒏𝒏𝑛𝑛 = 𝑵𝑵 exp[−
= 𝑁𝑁𝒄𝒄𝑐𝑐 exp − ] == 𝑵𝑵
𝑁𝑁𝒄𝒄
𝑐𝑐
exp
exp −
− exp(
exp ) = 𝒏𝒏
=𝒊𝒊 exp(
𝑛𝑛𝑖𝑖 exp )
𝒌𝒌𝒌𝒌
𝑘𝑘𝑘𝑘 𝒌𝒌𝒌𝒌
𝑘𝑘𝑘𝑘 𝒌𝒌𝒌𝒌 𝑘𝑘𝑘𝑘 𝒌𝒌𝒌𝒌 𝑘𝑘𝑘𝑘
ni
𝐸𝐸𝐹𝐹 − 𝐸𝐸𝑉𝑉 𝐸𝐸𝑖𝑖 − 𝐸𝐸𝑣𝑣 𝐸𝐸𝑖𝑖 − 𝐸𝐸𝐹𝐹 𝐸𝐸𝑖𝑖 − 𝐸𝐸𝐹𝐹
𝑝𝑝 = 𝑁𝑁𝑣𝑣 exp − = 𝑁𝑁𝑣𝑣 exp − exp = 𝑛𝑛𝑖𝑖 exp
𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘
n
𝐸𝐸𝑐𝑐 − 𝐸𝐸𝑉𝑉 i
𝐸𝐸𝑔𝑔
∴ 𝑛𝑛𝑝𝑝 = 𝑁𝑁𝑐𝑐 𝑁𝑁𝑣𝑣 exp − = 𝑁𝑁𝑐𝑐 𝑁𝑁𝑣𝑣 exp − = 𝑛𝑛𝑖𝑖 2
𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘
∴ The product of the two types of carriers will remain constant as a
given temperature → mass action law

 For example, in Si at 300K, 𝒏𝒏𝒊𝒊 𝟐𝟐 = (𝟗𝟗. 𝟔𝟔𝟔𝟔 × 𝟏𝟏𝟎𝟎𝟗𝟗 )2 = 𝟗𝟗. 𝟑𝟑𝟑𝟑 × 𝟏𝟏𝟎𝟎𝟏𝟏𝟏𝟏 /𝐜𝐜𝐜𝐜6
If 𝑛𝑛 = 9.31 × 1017 /cm3 then 𝑝𝑝 = 102 /cm3
If 𝑛𝑛 = 9.31 × 1014 /cm3 then 𝑝𝑝 = 105 /cm3
 EXAMPLE 4
 A silicon ingot is doped with 1016 arsenic atoms/cm3. Find the carrier
concentrations and the Fermi level at room temperature (300K)
 SOLUTION
At 300K, we can assume complete ionization of impurity atoms.
We have 𝑛𝑛 ≈ 𝑁𝑁𝐷𝐷 = 1016 cm-3.
𝑛𝑛𝑛𝑛𝑖𝑖2𝑖𝑖 2 9.65 × 10
9.65×10 9 92 2
From Eq.17, 𝑝𝑝 ≈
𝑝𝑝 ≈ 𝑁𝑁 == 16 ==9.3
9.3 × 1033cm
× 10 cm-3-3
𝑁𝑁𝐷𝐷𝐷𝐷 10
10 16

The Fermi level measured from the bottom of the conduction band
is given by Eq. 22：
19
𝑁𝑁𝑁𝑁𝐶𝐶𝑐𝑐 2.86 × 1019
2.86
𝐸𝐸𝐸𝐸𝑐𝑐 𝑐𝑐−−𝐸𝐸𝐸𝐸𝐹𝐹𝐹𝐹 =
= 𝑘𝑘𝑘𝑘
𝑘𝑘𝑘𝑘 𝑙𝑙𝑙𝑙
ℓ𝑛𝑛 =
= 0.0259
0.0259 𝑙𝑙𝑙𝑙
ℓ𝑛𝑛 = 0.205
= 0.205 eV.
eV.
𝑁𝑁𝑁𝑁𝐷𝐷𝐷𝐷 1016
16

The Fermi level measured from the intrinsic Fermi level is given by
Eq. 25：
𝑁𝑁𝑁𝑁𝐷𝐷𝐷𝐷 1016
𝐸𝐸𝐹𝐹𝐸𝐸𝐹𝐹−−𝐸𝐸𝐸𝐸𝑖𝑖 𝑖𝑖≈≈ 𝑘𝑘𝑘𝑘
𝑘𝑘𝑘𝑘𝑙𝑙𝑙𝑙
ℓ𝑛𝑛 == 0.0259
0.0259 𝑙𝑙𝑙𝑙
ℓ𝑛𝑛 = 0.358
= 0.358 eV.
eV.
𝑛𝑛𝑛𝑛𝑖𝑖 𝑖𝑖 9.65 × 109
9.65
These results are shown graphically in Fig. 23.
 EXAMPLE 4(Cont.)

 Figure 23 shows the band diagram showing Fermi level 𝐸𝐸𝐹𝐹 and
intrinsic Fermi level 𝐸𝐸𝑖𝑖
 CHARGE NEUTRALITY
 Charge neutrality, i.e., the total negative charges (electrons and ionized

acceptors) must equal the total positive charges (holes and ionized

donors).
− +
𝑛𝑛 + 𝑁𝑁𝐴𝐴 = 𝑝𝑝 + 𝑁𝑁𝐷𝐷
𝑛𝑛𝑛𝑛 = 𝑛𝑛𝑖𝑖 2

− 𝑛𝑛𝑛𝑛𝑖𝑖𝑖𝑖22 +
∴ 𝑛𝑛 + 𝑁𝑁𝐴𝐴 = + 𝑁𝑁𝐷𝐷
𝑛𝑛𝑛𝑛
+ −
𝑛𝑛2 − 𝑁𝑁𝐷𝐷 −𝑁𝑁𝐴𝐴 𝑛𝑛 − 𝑛𝑛𝑖𝑖 2 = 0
+ − +𝟐𝟐 − 2
𝑵𝑵𝑁𝑁
+𝐷𝐷 − − 𝑁𝑁𝐴𝐴 ++ 𝑁𝑁
− 𝐷𝐷 − 𝑁𝑁𝟐𝟐𝐴𝐴 + 4𝑛𝑛𝑖𝑖2
𝑫𝑫 −𝑵𝑵𝑨𝑨 + 𝑵𝑵𝑫𝑫 −𝑵𝑵𝑨𝑨 +4𝒏𝒏𝒊𝒊
Solving for 𝑛𝑛 𝒏𝒏𝑛𝑛𝒏𝒏𝑛𝑛0==
𝟐𝟐 2
𝑛𝑛𝑛𝑛0 is the majority carriers for 𝒏𝒏-type semiconductor (𝑛𝑛𝑛𝑛0 ≅ 𝑁𝑁𝐷𝐷 , if 𝑁𝑁𝐷𝐷+ ≫ 𝑁𝑁𝐴𝐴 )
𝒏𝒏𝟐𝟐𝒊𝒊𝑛𝑛𝑖𝑖2
𝒑𝒑𝑝𝑝𝒏𝒏𝟎𝟎𝑛𝑛0== ,𝑝𝑝𝑛𝑛0 is the minority carriers in 𝒏𝒏-type semiconductor
𝒏𝒏𝒏𝒏
𝑛𝑛 𝑛𝑛0
 FERMI LEVEL VERSUS TEMPERATURE
 Fermi level as a function of temperature can be calculated for a
given acceptor or donor concentration from Eqs. 13, 14, 28 and 31.

 Figure 24 shows the Fermi level for (a) Si and (b) GaAs as a
function of temperature and impurity concentration (also showing
is the dependence of the bandgap on temperature):

 As T↑ 𝑬𝑬𝒈𝒈 ↓ ( see Figure of Eg vs T )

T↑ 𝑬𝑬𝑭𝑭 → 𝑬𝑬𝒊𝒊
 ELECTRON DENSITY VERSUS TEMPERATURE
 Figure 25 shows electron density as a function of temperature for a Si sample
with a donor concentration of 𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏 cm-3 .

 For 𝑵𝑵𝑫𝑫 = 𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏 𝐜𝐜𝐜𝐜−𝟑𝟑

At low temperature ( < 150K ): freeze out region, 𝒏𝒏 < 𝑵𝑵𝑫𝑫
At medium temperature (150 ~ 550K ): extrinsic region, 𝒏𝒏 = 𝑵𝑵𝑫𝑫 > 𝒏𝒏𝒊𝒊
At high temperatures( > 550K ): intrinsic region, 𝒏𝒏𝒊𝒊 > 𝑵𝑵𝑫𝑫
 Intrinsic temperature: the temperature at which 𝒏𝒏𝒊𝒊 = 𝑵𝑵𝑫𝑫
for example for 𝑵𝑵𝑫𝑫 = 𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏 cm-3 , the intrinsic temperature 𝑻𝑻𝒊𝒊 = 𝟓𝟓𝟓𝟓𝟓𝟓𝐊𝐊
If 𝑵𝑵𝑫𝑫 = 𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏 𝐜𝐜𝐜𝐜−𝟑𝟑 , 𝑻𝑻𝒊𝒊 can be obtained from Fig.18.
𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏
𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏 𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏
𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏
For 𝒏𝒏𝒊𝒊 = 𝟏𝟏𝟏𝟏𝟏𝟏𝟏𝟏 𝐜𝐜𝐦𝐦−𝟑𝟑 , we have ==𝟏𝟏.𝟏𝟏.
5 𝟓𝟓∴ 𝑻𝑻∴𝒊𝒊 𝑻𝑻=𝒊𝒊 = = 𝟔𝟔𝟔𝟔𝟔𝟔𝐊𝐊
𝑻𝑻𝑻𝑻
𝒊𝒊 𝟏𝟏.𝟓𝟓 𝟏𝟏. 𝟓𝟓 = 𝟔𝟔𝟔𝟔𝟔𝟔𝟔𝟔
𝒊𝒊 47
 SUMMERY OF CHAPTER 1

 We have listed 32 semiconductor materials, the most

important is silicon.

 Silicon has a diamond lattice structure, its surface and

orientation are defined by Miller indices.

 We have introduced the effective-mass concept. By using

this concept, we can treat electrons and holes essentially as
classical charged particles.

 We have considered the carrier concentration in a

semiconductor which is dependent on the bandgap, impurity
concentration, and temperature.

48