Dr. M. M.

Abdel Aziz

Chapter 6

Photonic Devices

Excess carriers
Most semiconductor devices operate by the creation of charge carriers in excess of
the thermal equilibrium values. These excess carriers can be created by:
1. optical excitation
2. Thermal excitation
3. electron bombardment,
4. injection across a forward-biased p-n junction.

Whenever the excess carriers arise, they can dominate the conduction processes in
the semiconductor material. In this chapter we shall investigate the creation of excess
carriers by optical absorption in photonic devices and the resulting properties of
photoluminescence and photoconductivity.

The Origin of Radiation

Fig.1

Fig. 1 shows the band diagram of a certain material with bandgap energy Eg = Ec –
Ev . Photons with energies greater than the band gap energy are absorbed while photons
with energies less than the band gap are transmitted. Since the valence band contains many
electrons and the conduction band has many empty states into which the electrons may be
excited, the probability of photon absorption is high.
As Fig.1 indicates, an electron excited to the conduction band by optical absorption
may initially have more energy than is common for conduction band electrons (almost all
electrons are near Ec unless the sample is very heavily doped). Thus the excited electron
loses energy to the lattice in scattering events until its velocity reaches the thermal

44
 Dr. M. M. Abdel Aziz

equilibrium velocity of other conduction band electrons. The electron and hole created by
this absorption process are excess carriers; since they are out of balance with their
environment, they must eventually recombine. While the excess carriers exist in their
respective bands, however, they are free to contribute to the conductivity of the material.

A photon with energy less than Eg is unable to excite an electron from the valence
band to the conduction band. Thus in a pure semiconductor, there is negligible absorption of
photons with hυ < Eg (υ = f ).
This explains why some materials are transparent in certain wavelength ranges. We
are able to “see through” certain insulators, such as a good NaCl crystal, because a large
energy gap containing no electron states exists in the material. If the bandgap is about 2
eV wide, only long wavelengths ( infrared ) and the red part of the visible spectrum are
transmitted; on the other hand, a band gap of about 3 eV allows infrared and the entire
visible spectrum to be transmitted.

Fluorescence

Direct recombination described in Fig.1 is a fast process; the mean lifetime of the
EHP is usually on the order of 10-8 s or less. Thus the emission of photons stops within
approximately 10-8 s after the excitation is turned off. Such fast luminescent processes are
often referred to as fluorescence.

Phosphorescence

In some materials, however, emission continues for periods up to seconds or minutes

after the excitation is removed. These slow processes are called phosphorescence, and the
materials are called phosphors. An example of a slow process is shown in Fig. 2.

Fig.2

This material contains a defect level (perhaps due to an impurity) in the band gap
which has a strong tendency to temporarily capture (trap) electrons from the conduction
band. The events depicted in the figure are as follows:
(a) Incoming photon with hυ1 > Eg is absorbed, creating an EHP;

45
 Dr. M. M. Abdel Aziz

(b) The excited electron gives up energy to the lattice by scattering until it nears the
bottom of the conduction band.
(c) The electron is trapped by the impurity level Et and remains trapped until it can
be thermally reexcited to the conduction band
(d); (e) Finally direct recombination occurs as the electron falls to an empty state in
the valence band, giving off a photon (hυ2) of approximately the band gap energy.

The delay time between excitation and recombination can be relatively long if the
probability of thermal reexcitation from the trap (d) is small. Even longer delay times result
if the electron is retrapped several times before recombination.

Light- Emitting Materials

Fig.3

Figure 3 indicates the band gap energies of some of the common semiconductors,
relative to the visible, infrared, and ultraviolet portions of the spectrum. We observe that
GaAs, Si, Ge, and InSb lie outside the visible region, in the infrared. Other semiconductors,
such as GaP and CdS, have band gaps wide enough to pass photons in the visible range. It is
important to note here that a semiconductor absorbs photons with energies equal to the band
gap, or larger. Thus Si absorbs not only band gap light (1 µm) but also shorter wavelengths,
including those in the visible part of the spectrum.
The relation between photon energy and wavelength is E = hc/λ.

If E is given in electron volts and λ in micrometers, this becomes E = 1.24/ λ.

46
 Dr. M. M. Abdel Aziz

The Light Emitting Diode (LED)

There are many ways by which electrical energy can be used to generate photon
emission in a solid. In LEDs an electric current causes the injection of minority carriers into
regions of the crystal where they can recombine with majority carriers, resulting in the
emission of recombination radiation.
This important effect (injection electroluminescence) will be explained in terms of p-n
junction theory with the aid of the Fig.4
For a direct bandgap semiconductor material where the empty electron states in the
conduction band ( CB) of the p-type material and the normally hole empty states in the
valence band (VB) of the n-type material are populated by injected carriers which
recombine across the bandgap.
This recombination may be radiative or non-radiative. In non-radiative
recombination , the energy released is dissipated in the form of lattice vibrations and thus
heat. However in band – to – band radiative recombination, the energy is released with the
creation of a photon of energy :

Fig.4

ℎ𝑐
𝐸𝑔 = ℎ𝑓 =
𝜆
The optical wavelength is

1.24
𝜆=
𝐸𝑔
(λ in µm ,Eg in eV )

This spontaneous emission is known as electroluminance . The light is emitted at the

site of carrier recombination which is primarily close to the junction , although

47
 Dr. M. M. Abdel Aziz

recombination may take place through hole diode structure as carrier diffuse away from the
junction, Fig.5

p- type
p-n junction
n- type

Fig.5

Other radiative recombination processes:

Energy levels may be introduced into the bandgap by impurities or lattice defects within the
material structure which greatly increase the electron – hole pair recombination ( reduce the
carrier life time ).
Major radiative recombination processes are shown as ;
a- Band-to-band transition
b- Conduction band to acceptor impurity, and donor impurity to valance band
transition
c- Donor impurity to acceptor impurity transition
d- Isoelectronic impurity to valance band.

Fig.6

48
 Dr. M. M. Abdel Aziz

The Double Heterojunction LED

The principle of operation is illustrated in the following figure. A p-type GaAs layer
is sandwitched between p- and n- type AlGaAs layers. At forward bias, electrons from n-
type are injected into the p-type GaAs layer and become minority carriers which are diffused
away from the junction and recombine with majority carriers (holes ) as they do so. The
injected electrons are inhibited from diffusing into the p-type AlGaAs layer because of the
potential barrier presented by the p – p heterojunction. Hence radiation only occurs in the
GaAs layer, providing both good internal quantum efficiency and high radiance emission .
Also, light is emitted from the device without absorption.
The double heterojunction LED is considered the most incoherent source for optical
communications. The structure is given below.

Fig.7

49
 Dr. M. M. Abdel Aziz

Optical Detectors
Introduction:
Two types of optical detectors are commonly used: thermal devices, and photon
devices. In thermal detectors, the absorption of light raises the temperature of the device and
this in turn results in changes in some temperature dependent parameter ( e.g electrical
conductivity ). The output of the thermal detectors is usually proportional to the amount of
energy absorbed per unit time by the detector provided the absorption efficiency is the same
for all wavelengths.

In photon detectors, the absorption process results directly in some specific quantum
event ( such as photoelectric emission ) which is then counted by the detection system.
Thus the output of the photodetectors is governed by the rate of absorption of light quanta
and not directly on their energy. Also, a certain minimum photon energy is required to
initiate the process.

Since 𝐸 = ℎ𝑓 = ℎ𝑐/𝜆 photon detectors have a cutoff wavelength, that is maximum

𝜆, beyond which they will not operate. A problem encountered with photon detectors
operating in the IR, is that the photon energy > the thermal energy of atoms in the detector (
= KT ). Thermal excitation will occur rather than light absorption, and this will generate
noise. An obvious way to reduce this noise is to lower detector temperature [ most photon
detectors operating above 𝜆 = 3µm are cooled in liquid nitrogen at 77K ]

Photodetectors

When absorption of light energy produces a corresponding electrical output, the

device is called a 'photodetector'. We shall limit our discussion here for the semiconductor
junction detectors.

Two important properties of photodetectors are : the gain, and the response time.
The gain is defined as the number of charge carriers passing through the contact electrodes
per second for each photon absorbed per second. The response time is the ability of the
detector to follow a step change of light input. These two properties are summarized below
for some photodetectors considered here

Device Gain Response time (s)

Photoconductor 105 10-1

p –n photodiode 1 10-10

p-i-n photodiode 1 10-12

Avalanche photodiode 102 10-10

Bipolar Transistor 102 10-7

50
 Dr. M. M. Abdel Aziz

The Junction Detectors

1. The p-n junction photodiode

Fig.8

If the diode is made transparent, then radiation can penetrate to the junction region
and electron – hole pair ( EHP ) will be generated both near the junction and away from it.
Electrons released near the junction (within a diffusion length ) will be swept into the n-
region, and holes to the p-region by the junction field. EHP formed away from the junction
may recombine before charge separation.

Charge separation may be detected in three distinct ways:

a- Photovoltaic Operation Mode : If the device is left open circuit ( or very high
resistance between the external contacts), an externally measurable potential will
appear between the p – and n – regions

b- Photoamperic Mode: A very low external resistance is connected between the

external contacts and a measured photo generated current flows.

c- Photoconductive Operation Mode: This is the most usual way to operate the
device. A reverse bias is applied across the junction and the resulting current flows
through a load resistor is measured.

In operation in the last mode, the excess carriers near the junction provide a
photocurrent IP which adds directly to the current generated thermally ( Id ). In a reverse
biased photodiode with zero illumination Id is known as the dark current. After
illumination , the reverse current rises from Id to Id + Ip. Ip is related to the power of the
incident radiation Po. If there are rp photons incident per second on the diode, then

𝑃𝑜 = 𝑟𝑝 . ℎ𝑓 = 𝑟𝑝 . ℎ𝑐/𝜆

51
 Dr. M. M. Abdel Aziz

And if there are re electron-hole pairs are created per second by the radiation, then
𝑟𝑒
𝐼𝑃 = 𝑟𝑒 . 𝑒 , 𝜂=
𝑟𝑝
, and
𝐼𝑃 𝑟𝑒 𝑒 𝜆 𝑒𝜆𝜂
= =
𝑃𝑂 𝑟𝑝 ℎ 𝑐 ℎ𝑐

Where η is the quantum efficiency. It is a measure of the diode efficiency.

𝐼𝑃
is called the responsivity and is proportional to the power of the incident
𝑃𝑂
radiation at a particular wavelength.

1- The p - n Diode

The I – V characteristics of a silicon p – n photo diode is shown. The reverse

current IR = Id + IP is almost independent on VR , since the field across the junction
is sufficiently strong to extract all the current carriers created at the junction. The
reverse current is proportional to illumination ( optically generated EHP or gop) and
can be measured directly. A typical sensitivity is 0.7 A mW-1 cm-2

Fig.9

2- The Phototransistor

Greater sensitivity is obtained if the junction is the collector – base junction of a

phototransistor. This can be a normal transistor in a transparent encapsulation and connected
in common-emitter configuration with the base open circuit.

In the absence of radiation, the collector current is:

𝐼𝐶 = (ℎ𝐹𝐸 + 1)𝐼𝐶𝐵0 ≅ ℎ𝐹𝐸 𝐼𝐶𝐵0

ICB0 is the leakage collector current . When radiation falls on the collector base junction, the
collector current becomes:

52
 Dr. M. M. Abdel Aziz

𝐼𝐶 = ℎ𝐹𝐸 (𝐼𝐶𝐵0 + 𝐼𝑃 ) ≅ ℎ𝐹𝐸 𝐼𝑃

Thus owing to the current gain of the transistor , the photocurrent is greater than the diode
current. The optical gain of the transistor is given by :

𝐼𝐶 ℎ𝑓 𝐼𝐶
𝐺0 = η ℎ𝐹𝐸 = η = .
𝐼𝑃 𝑒 𝑃0

Fig.10

Solar Cells
Since power can be delivered to an external circuit by an illuminated junction, it is
possible to convert solar energy into electrical energy. Consider the fourth quadrant of the I
– V characteristics of a p-n diode, Fig. 11. It appears doubtful that much power can be
delivered by an individual device if extra holes move to the p-region and extra electrons
move to the n-region, i.e increasing the junction area. The voltage is restricted to values less
than the contact potential, which in turn is generally less than the band gap voltage Eg/q.
For Si the voltage Voc is less than about 1 V. The current generated depends on the
illuminated area, but typically Iop is in the range 10-100 mA for a junction with an area of
about 1 cm2.
A photo diode may be represented by the circuit shown in Fig 12, where the effect of
illumination is represented by a current sources Ip

Fig.11

53
 Dr. M. M. Abdel Aziz

In general, the diode current and voltage are related as:

𝑒𝑉
𝐼 = 𝐼0 (𝑒 𝐾𝑇 − 1)
And
𝐾𝑇 𝐼
𝑉= 𝑙𝑛 (1 + )
𝑒 𝐼0
When the diode is open circuit I = Ip ( due to illumination )
𝐾𝑇 𝐼𝑝
𝑉𝑜𝑐 = 𝑙𝑛 (1 + )
𝑒 𝐼0
With RL connected across the diode, a load line can be drawn as shown so that
I = Ip - IL , where IL = V / RL, and
𝐾𝑇 𝐼𝑝 − 𝐼𝐿
𝑉= 𝑙𝑛 (1 + ) < 𝑉𝑜𝑐
𝑒 𝐼0
For RL = 0, diode behaves as a current source with IL = Ip and V = 0

Ip RL

Fig.12

To utilize a maximum amount of available optical energy, it is necessary to design a solar

cell with a large area junction located near the surface of the device (Fig. 13)
(a) enlarged view of the planar junction; (b) top view, showing metal contact “fingers.”

Fig.13

a)The array of junctions in solar cell “paddles” attached to the main body of the satellite
b)A 72MW solar cell plant in Italy
54