MICROWAVE SEMICONDUCTOR DEVICE TECHNOLOGIES 4.

ENERGY BANDS AND CHARGE CARRIER

1.

ENERGY BANDS IN SOLIDS:

We know that in the simple atomic description of matter, electrons are restricted to sets of discrete energy levels (electrons orbit around the nucleus only at predetermined energy values) and large gaps exist in which no energy states are available. In an analogous manner, electrons in solids are also restricted to certain energies. The basic difference between the case of an electron in an isolated atom and that of an electron in a solid is that in the solid the electron has a range or band of available energies. The discrete energy levels of the isolated atom spread into bands of energies in the solid. (a) Bonding forces in Solids The interaction of electrons in neighbouring atoms of a solid serves the very important function of holding the crystal together. There are three major types of interatomic bonding:

(i)

Ionic bonding: The archetypal example is the NaCl crystal where during bonding the exchange of electrons between the Na and Cl atoms leads to net charge transfer giving rise to Na+ and Cl- ions. An important observation in the NaCl structure is that all electrons are tightly bound to atoms and once the electrons exchange is completed between the Na and Cl atoms, there are no loosely bound electrons to participate in current flow and hence NaCl is a good insulator.

(ii)

Metallic bonding: In a metal atom, the outer electron shell is only partially filled usually by no more than 3 electrons. These electrons are loosely bound .The forces holding the metal lattice together arise from the interaction between the positive ion cores and the surrounding free electrons. This type of bonding leads to metals having high electrical conductivity because of the "sea" of free electrons available.

(iii)

Covalent bonding : This type of bonding is exhibited by a great deal of elemental ( Si and Ge) and compound semiconductors ( GaAs, InP etc..). In these crystals each atom shares its valence electrons with its four neighbour (Figure 1.). The bonding force arises from a quantum mechanical
4-1

interaction between the shared electrons. This is known as covalent bonding, each electron pair constitutes a covalent bond. As in the case of ionic crystals, no free electrons are Available to the lattice and hence semiconductors should behave as insulators. However for semiconductors we will see that there are mechanisms for inducing conduction in the lattice.

(b) Energy bands As isolated atoms are brought together to form a solid, various interactions occur between neighbouring atoms. The forces of attraction and repulsion between atoms will find a balance at the proper interatomic spacing for the crystal. In the process important changes occur in the electron energy level configuration and these changes give rise to the properties of solids. In a solid, as the atoms are brought together their discrete energy levels begin to split into new levels that belong to the solid rather than each individual atom. This splitting eventually leads to the picture in figure 2, which shows the formation of energy bands of diamond (C) as the isolated carbon atoms are brought together.

4-2

Figure 2.

Energy band formation

(c) Metal, Semiconductors and Insulators The implications of the preceding section concerning bonding in solids naturally leads to the classification of solids according their final band structure. Figure 3 shows the energy band diagrams for an insulator, a semiconductor and a metal. For the insulator the band structure consists of a filled valence band separated from an empty conduction band by a large energy gap. There can be no charge transport within the valence band since there are no empty states for the electrons to go into and there is no charge transport in the conduction band since there are no electrons there. Insulators are therefore characterised by high resistivity. Semiconductor materials at O K are basically insulators. The difference lies in the size of the energy gap : for common semiconductors it is about 1 to 2 eV compared with over 5 eV for a typical insulator. The relatively small band gaps of semiconductors allow for excitation of electrons from the valence band into the conduction band by reasonable amount of thermal or optical energy. In metals the bands either overlap or are only partially filled. Thus electrons and empty energy states are intermixed within the bands so that electrons can move freely under the influence of an electric field. Metals have high electrical conductivity.

4-3

Figure 3.

Typical band structure at O K.

2.

E-k DIAGRAMS : (a) Energy-Momentum relationships

When quantitative calculations are made of the band structures, a single electron is assumed to travel through a perfectly periodic crystal. The wavefunction that describes the motion of electron through the crystal is assumed to be in the form of a plane wave moving in the x-direction with propagation constant k called a wave vector.

(x) = U( k , x) e
k x

jk x

Where the function U(kx,x) modulates the wave function according to the periodicity of the lattice. In such a calculation, allowed values of energy can be plotted vs. the propagation constant k. The propagation constant k is directly proportional to the momentum of the electrons.

4-4

(b) Direct and Indirect Semiconductors The band structure of GaAs has a minimum in the conduction band and a maximum in the valence and for the same value of k. On the other hand Si has its valence band maximum at a different value of k than its conduction band minimum. (See figure 4). Thus an electron making a smallest energy transition from the conduction band to the valence band in GaAs can do so without a change in k which is not the case for Si.

Figure 4.

Direct and indirect gap semiconductors.

Thus there are two classes of semiconductors energy bands: direct and indirect. In the case of the indirect transition a change in k, requiring a change of momentum, is involved.

4-5

In the case of the direct band gap semiconductor the minimum in energy at k=0 in the conduction band is called the Gamma band Two other important conduction bands situated at other values of k are the X and L -bands The valence bands are all situated at k=0.

3.

ELECTRONS AND HOLES: (a) Effective mass

The electrons in a crystal are not completely free but instead interact with the periodic potential of the lattice. As a result their motion is not the same as that of electrons in free space. To account for most of the influences of the lattice, the electrons are treated as being "almost free" with an effective mass that is different from the usual mass of the electron in free space. The calculation of effective mass must take into account the shape of the E-k relationship. The effective mass of an electron in a band with a given (E, k) relationship is:

h m= E d dk
* 2 2

Thus the curvature of the band determines the electron effective mass. For example in GaAs m* is much smaller in the direct minimum than in the X or L minimum. An interesting feature in figure 4 is that the curvature ( m*) is positive at the conduction band but negative at the valence band maxima. Thus the electrons near the top of the valence band have negative effective mass. These negative effective mass electrons move in the same direction as a positive particles and are thus termed holes. (b) Intrinsic materials A perfect semiconductor crystal with no impurities is called an intrinsic semiconductor. In such materials there are no charge carriers at O K. At higher temperatures electron-hole pairs (EHP) are generated as valence band electrons are exited thermally across the band gap to the conduction band. These EHP are the only charge carriers in intrinsic materials.. The generation of EHP can be visualised by the breaking of the covalent bonds in the crystal. We have n = p = ni
-6-

c) Extrinsic materials It is possible to create carriers in semiconductors by purposely introducing impurities into the crystal. This process called doping is the most common technique for varying the conductivity of semiconductors. By doping a semiconductor can be altered so that conduction is predominantly by electrons or holes giving rise to two types of semiconductors: n-type (mostly electron conduction) : p-type (mostly hole conduction) : impurities have excess electrons to donate. impurities have deficiency of electrons.

When impurities are introduced into a semiconductor crystal additional energy levels are created within the band gap. For an n-type semiconductor the energy level is nearer the conduction band (Figure 5). The process is explained by the fact that the "donor " level is filled with electrons at O K but requires very little energy to donate its electron to the conduction band (hence by 5O K all the electrons are exited into the conduction band)

Figure 5.

Donation of an electron from a donor level

-7-

Similarly impurities that introduce holes in the valence band are called acceptors and the process is depicted in figure 6.

Figure 6.

Acceptance of valence band electrons by an acceptor level and the resulting creation of holes.

For the most common semiconductors, donor and acceptor levels are around 2 to 20 meV respectively. The physical process of donor and acceptor formation is illustrated in figure 7 for Al (p-type) and Sb (n-type) doping in Silicon.

Figure 7.

Formation of acceptors and donors in a silicon crystal.

(d) The Fermi level At thermal equilibrium the Fermi level must be constant throughout the semiconductor materials. The Fermi-level is determined by the condition of electrical neutrality ie n + N-A = p + N+D where N-A and N+D are the densities of ionised acceptors and donors respectively.

The electron and hole concentrations are given by : n = Nc(E-Ec)fcdE and p = Nv(Ev -E)(1-fv)dE

where fc and fv are given by : fc = 1/{exp((E-EFn)/kT) + 1} fv = 1/{exp((E-EFp)/kT) + 1} and Nc(E - Ec)dE and Nv(Ev - E)dE are the density of available electron ( or hole) states in the conduction ( or valence ) band within the element of energy dE. fc and fv are the Fermi-Dirac distribution functions After some lengthy algebra, the expression for the electron concentration comes out to be :

( EF EC ) n = Nce kT
Where Nc is now: 3 2 m * kT Nc = 2( )2 2 h where m* is the electron effective mass in the conduction band , k is the Boltzmann constant, T the temperature and h is Plancks constant.

-9-

3 ( EF EC ) 2 m * kT And hence n = 2( ) 2 e kT 2 h When the exponential approximation to the Fermi-Dirac function is not valid, then n = NcF1/2(EF-Ec)/kT

Where F1/2 is the Fermi-Dirac integral of order 1/2.

Similarly for holes: p = NvF1/2(Ev-EF)/kT Where Nv = 2(2kT/h2)3/2 [ (m*ph)3/2 + (m*pl)3/2 ] If Boltzmann statistics are valid and n/Nc or p/Nv is less 0.1, then all donors are ionised and the electrical neutrality condition is given by n = ND or if NA is not negligible n = ND - NA ie n E F = E C + kT q [ln Nc ] .................... (A) The effect of degeneracy: The Joyce-Dixon approximation When n/Nc or p/Nv are > 0.1 then the more exact Fermi-Dirac integral must be used in the calculation. However Joyce and Dixon have extended the Boltzmann approximation for the Fermi level given by equation (A) above to the degenerate case as follows:
n 1 n] E F = E C + kT q [ln Nc + 8 Nc

- 10 -

PROBLEMS 1. 2. 3. Show that the element Beryllium (Be) of the group II of the periodic table acts as an acceptor when it is substituted for a Gallium atom in the GaAs lattice. Show that the element Sulfur (S) of the group VI of the periodic table acts as a donor acceptor when it is substituted for an Arsenic atom in the GaAs lattice. How does the element Silicon (Group IV) behave in GaAs when it replaces the Ga atom? How does it behave when it replaces an arsenic atom? (we call such a dopant an emphoteric dopant) A GaAs sample has a free electron concentration of 1 x 1017 cm-3. Calculate the position of the Fermi level using the Boltzmann approximation. Repeat the calculations using the Joyce-Dixon approximation Nc = 4.45 x 1017 cm-3 for GaAs at room temperature (Answer: -0.0388 eV and 0.036 eV) A GaAs sample has a free electron concentration of 1 x 1018 cm-3. Calculate the position of the Fermi level using the Boltzmann approximation. Repeat the calculations using the Joyce-Dixon approximation Nc = 4.45 x 1017 cm-3 for GaAs at room temperature (Answer: 0.0584 eV and 0.0379 eV) In a GaAs sample at 300K, the Fermi level coincides with the valence band edge. Calculate the hole density using: (a) The Boltzmann approximation (b) The Joyce Dixon approximation Nv = 7.72 x1018 cm-3 Hint: use trial and error to get to the solution (Answer: 5.87x1018 cm-3) 7. Calculate the position of the Fermi level for a GaAs crystal doped a concentration of 5x1016 cm-3 both at room temperature and 77K using both Boltzmann and Joyce-Dixon approximations. The electron effective mass for electron in the conduction band is m*=0.067m0 mo= 9.109 x 10-31 kg Comment on the results (Answer: at room temperature -0.056 and 0.0551 eV, at 77K 0.0032 eV and 0.005 eV Difference at RT is 1meV, which is quite negligible but is 8 meV at 77K which is not negligible!)
- 11 -

4.

5.

6.