Beruflich Dokumente
Kultur Dokumente
1 Molecular orbitals and energies To understand the energy band formation in solid, let us first consider the gy , formation of energy levels when two hydrogen atoms are brought together to form a hydrogen molecule (H2). The H2 molecule are bonded by the b th covalent b di l t bonding. The electronic configuration of H atoms is H: 1s1 The wave function of the 1s state: 1s 1s d describes a quantum state with n = 1 l = 0 & ml = 0 ib t t t ith 1, The energy of the 1s state: E1s
A
When two hydrogen atoms, A & B say, approach each other, the 1s wave function 1s of each individual H f h i di id l atom interact producing two new wave functions, a & b, with energies Ea & Eb respectively. See Fig.3.1.
E1s
1 1s
Ea
a 1 1s
E bonding energy
E1s
Eb b
H atom (A) H2 molecule Fig. 3.1 H atom (B)
The wave functions a and b are obtained by a linear combination of the individual atomic orbitals/wave functions (LCAO)
molecular l l orbital
a = 1s ( A ) + 1s ( B ) b = 1s ( A ) 1s ( B )
H3 Semiconductor Physics & Devices
(3.1)
The formation of the two new wave functions a & b is consistent p p , g y with the Pauli exclusion principle, i.e. in a given electrons system, there are no two electrons may have the same quantum state. In our case here, this means we cannot have two identical 1s wave functions in the H2 molecule. The wave functions a & b are called the molecular orbital orbital. Specifically, a with energy Ea is called the anti-bonding orbital antib with energy Eb is called the bonding orbital Taking the spin into consideration, each wave function will have two distinct states: one spin up & one spin down. In other words, each wave function can accommodate two electrons, one electron with spin up & one electron with spin down.
Before the interaction, each individual i di id l H atom h t has one electron in the 1s state/wave function. function After the interaction in interaction, the ground state, the two electrons will occupy the lowest energy wave function b, one electron with spin up & the other electron with spin d ith i down.
E1s
1s
Ea
a 1s
E bonding energy g gy
E1s
Eb b
H atom (A) H2 molecule Fig. 3.1 H atom (B)
The lowering of the original energy before the interaction (E1 ) & the 1s ground state energy after the interaction (Eb) is the bonding energy of the H2 molecule (E). The formation of the bonding and anti-bonding energy, Eb & Ea respectively, from the original energy E1s is referred to as energy splitting
A
The energy splitting or the splitting, separation of Ea & Eb , varies with the interatomic distance r between the two H atoms as shown in Fig. 3.2 i Fi 3 2 It is important to note here that:
E
1 H atom 1s
Ea Eb r0
E1s
Fig. 3.2
number of quantum states before interaction = number of quantum states after interaction number of electrons b f b f l t before i t interaction = number of electrons after ti b f l t ft interaction
3.2 Energy band: conduction and valence bands Having d H i understood the f d h formation of the molecular orbitals & energies, we i f h l l bi l i can now look into the energy band formation in semiconductor crystals. Consider a silicon crystal consisting of N silicon atoms. The electronic atoms configuration of an isolated silicon atom is
14Si:
Table 3.1
The number of states & electrons of N isolated Si atoms is given in Table 3.1
States 1s 2s 2p 3s 3 3p
No. of states 2N 2N 6N 2N 2 6N
No. of electrons 2N 2N 6N 2N 2 2N
The N silicon atoms are now brought together to form silicon crystal. How are these silicon atoms bonded? They are bonded by the covalent bonding. Each Si atom share its four valence electrons (2 electrons from each 3s & 3p states) with the four nearest neighbors to form stable configuration or to complete their subshells. As a result each Si atom is surrounded by 8 electrons. This is shown in Fig 3 3 Fig. 3.3
Si Si Si Si
Si
Si
Si
Si
Si
Si
Si
Note that the 2-D representation in Fig.3.3b is used to conveniently show that each silicon atom shares its 4 valence electrons with 4 nearest neighbors of silicon atoms forming covalent bonds. Since the figure is 2-D, it could not show the angle of the bonds. The angle of the bonds can be seen in the 3-D representation of Fig.3.3a. The formation of the energy band in Si crystal can be explained as follow. As the inter-atomic distance decreases the outer shells 3s & 3p states interact. See Fig.3.4. The 2N states of 3s interact. The energy levels split forming 3s band. There are 2N states & 2N electrons in the 3s band. The 6N states of 3p interact. The energy levels split forming 3p band. There are 6N states & 2N electrons in this band. The degree of the energy splitting increases with decreasing interp g atomic spacing.
A
The band formation is consistent with the Pauli exclusion principle principle: no two electrons in a given system may have the same quantum states. states As the inter-atomic spacing decreases further, the 3s & 3p bands grow p g , p g & merge into a single band composed of mixture of energy levels. This 3s-3p band contains 8N states & 4N electrons. Further decreasing the inter-atomic distance, approaching the equilibrium inter-atomic spacing, the 3s-3p band splits into two bands q p g, p p separated by an energy gap or a bandgap Eg. The upper band, called the conduction band contains 4N states, & band, the lower band, called the valence band, also contains 4N states. band In a perfect crystal, the are no allowed energy levels (quantum states) in the bandgap for electrons to occupy. The bandgap is also called a forbidden band f bidd b d forbidden band. f d
A
At 0 K, the electrons will occupy the lowest energy states & therefore all 4N available electrons occupy all 4N states i th valence b d I ll il bl l t ll t t in the l band. In other words, at 0 K the valence band is fully occupied & the conduction band is empty. In this state, the semiconductor is an insulator, no py , , electrical conduction can take place. for the electrical conduction to take place, electrons must be able to move into empty states. It follows that for the electrical conduction, the electrons must be excited to the conduction band. band
Electron energy E e
Eg
r=
inter-atomic distance, r
The energy of the conduction & valence bands versus the physical conduction distance x of th semiconductor sample di t f the i d t l band can be represented by an energy band g g diagram as shown in Fig. 3.5 the energy refers to the electron gy energy the physical distance x is normally not indicated
valence band
Ec
bandgap energy Eg
Ev
Fig. 3.5
Ec is the energy of the bottom of the conduction band, very often called the conduction band edge Ev is the energy of the top of the valence band, very often called the valence band edge
A
If an electric field is applied to the semiconductor sample the energy band is not flat anymore as in Fig. 3.5 the energy bands will be inclined at a gradient proportional to the electric field according to equation (3.2). The energy band diagram is shown in Fig. 3.6
Ec
Eg
Ev
1 dE = q dx
A
(3.2)
3.3 Electrons and holes At T = 0 K, the valence band is fully occupied by electrons & the conduction band is empty The semiconductor behaves like an insulator. The electrons in the valence band can be excited to the conduction by increasing the temperature illuminating crystal with light of a suitable wavelength
Let us, first of all, consider the excitation of electrons from the valence band to the conduction by increasing the temperature. As the temperature temperature of the semiconductor is raised above 0 K, i.e. T > 0 K, some electrons in the valence band gain enough thermal energy to be excited across the b d i d h bandgap to the conduction b d This process h d i band. hi corresponds to the electrons breaking the Si-Si covalent bonds leaving behind broken bonds This is illustrated in Fig 3 7 bonds. Fig. 3.7
conduction band
Si Si Si Si
Ec Eg
Si Si Si Si
valence band
Ev
Si Si Si Si
(b) Electron, negatively charged ( 1.610-19 C) ec o , ega ve y c a ged ( .6 0 hole (broken bond), positively charged (+1.610-19 C)
A
conduction band
Si
Si
Si
Si
Ec Eg Ev
Si Si Si Si Si Si Si Si
T=0K
valence band conduction band
Si
Si
Si
Si
Ec Eg
+ (a)
Si Si Si Si
T>0K
valence band
Ev
Si
Si
Si
Si
(b)
Electron, negatively charged ( 1.610-19 C) , g y g ( ) hole (broken bond), positively charged (+1.610-19 C)
A
Some of the states in the conduction band is now occupied by electrons & some of the states in the valence band is now empty. An empty state in the valences band is referred to as a hole, it corresponds to the broken bond in the Si-Si bonding The hole is bonding. positively charged with a value of 1.610-19 C. These electrons & h l can f l move th Th l t holes freely f freely throughout th crystal. h t the t l The term free is obviously not in the real sense since the movement y of the electrons and holes in the crystal is under the influence of periodic potential formed by the atoms/ions in the crystal (compare to the real free electron in vacuum). The electrons & holes can only move freely for a short period of time, after which they will collide with vibrating atoms, impurity atoms & defects.
The movement of the hole is actually the movement of the electron from the Si-Si bond to the b k bond. Thi i ill t t d i Fi f h Si- b d h broken b d This is illustrated in Fig. bond 3.8. It is easier & convenient to think in term of the displacement of holes rather than the actual displacement of electrons from Si-Si p bonds to the broken bonds.
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
B
Si Si Si Si Si Si Si Si Si Si Si Si
C
Si Si Si Si Si Si Si Si Si Si Si Si
The electrons & holes are created by the excitation of electrons from the valence band to the conduction band. This direct excitation of electrons from the valence band to the conduction band is referred to as direct band-to-band generation. Obviously, the number of hole in the valence band = the number of electrons in the conduction band. Since the electrons & holes are created in pair, they are also called electron-hole pairs electronpairs. There are empty states in the conduction & valence bands, therefore the electrons & h l are able to perform a di h l holes bl f directed motion under the d i d h action of an external electric field, conducting an electric current.
The electrons in the valence band can also be excited to the conduction band, creating electron-hole pairs, by illuminating the semiconductor with light having a photon energy Eph equal or greater than the bandgap energy Eg . This is illustrated in Fig. 3.9.
E ph E g E ph = h = hc hc
(3.3) conduction band
Eph
photon valence band b d + (a)
Si Si Si Si
Ec Eg Ev
(3.4)
Eg Eg
(3.5)
1240
Si
Si
Eph
Si Si Si Si Si
where in nm & Eg in eV
photon
Si
Fig. 3.9
A
(b)
Sample at temperature T
conduction band
Ec Eg Ev
Eph
photon valence band +
Ec Eg Ev
valence band
Si
Si
Si
Si
Si
Si
Si
Si
Eph
Si Si Si Si Si Si Si Si
photon
Si Si Si Si Si Si Si Si
3.3.1 Direct band-to-band generation and recombination We have l W h learnt that electrons can b excited f h l be i d from the valence b d to the h l band h conduction band, creating electron-hole pairs, by thermal excitation & illumination (excitation by light) light). What happen if a semiconductor is subjected to a certain temperature T, or illuminated with light (Eph Eg) or both? We have to ask our self a question: Do these generated continuously? continuously electron-hole pairs keep increasing
The answer is certainly NO, otherwise we will observe that all the th covalent b d of th semiconductor will b b k simply b l t bonds f the i d t ill be broken i l by putting a sample in a room. We can subject a semiconductor to room temperature by simply putting the sample in a room. p y pyp g p
A
So, what happens ? The reason is that as electron hole pairs are generated they can also electron-hole generated, recombine. In this process the electrons & holes are annihilated (i.e. pp ) p recombination. disappear). The process is called recombination This is illustrated in Fig. 3.10. The direct generation & recombination processes from band-to-band shown in Fig 3 10 Fig. 3.10 is called direct band-to-band band-togeneration & recombination
conduction band
Ec
recombination
Ev
+ Fig. 3.10
g p The generation rate G is the number of electron-hole pairs created per unit volume per second (units: #/cm3-s). The generation is the sum of the thermal generation rate & the generation rate by the external excitation (if any) t l it ti ) The recombination rate R is the number of electron-hole pairs recombining per unit volume per second (units: #/cm3-s). p p p It is expected that the recombination rate is proportional to the electron concentration in the conduction band & the hole concentration in the valence band. Obviously, in the steady state, the electron & hole concentrations are constant with time.
A
When a semiconductor is at temperature T & there is no external disturbance/excitation (e.g. external electric field, magnetic field or light illumination), which create electrons in the conduction band & holes in the valence band, then the electrons & holes are solely created by thermal energy An equilibrium will be established & this state is called energy. established, thermal equilibrium since there is only thermal energy involved. In thermal equilibrium: thermal generation rate Gth = recombination rate R (the (th generation rate i called th ti t is ll d thermal generation rate) l ti t ) there exist thermal equilibrium electron & hole concentrations, n0 & p0 respectively. i l g , g The thermal generation rate, the recombination rate & the generation rate by the external excitation play a very important role in semiconductors. In chapter 7, we will express the generation & recombination rates mathematically. bi ti t th ti ll
A
conduction band
n0
Ec
=
+ +
recombination rate, R
valence band
Ev
p0