PIN / PN photodiode structure
Although an ordinary p-n junction can be used as the basis of a photodiode, the p-i-n junction provides a far more satisfactory photodiode structure. In the photodiode fabrication process a thick intrinsic layer is inserted between the p-type and n-type layers. The middle layer may be either completely instrinsic, or very lightly doped to make it and n- layer. In some instances it may be grown as an epitaxial layer onto the substrate, or alternatively it may be contained within the substrate itself.
PIN photodiode structure
One of the main requirements of the photodiode is to ensure that the maximum amount of light reaches the intrinsic layer. One of the most efficient ways of achieving this is to place the electrical contacts at the side of the device as shown. This enables the maximum amount of light to reach the active area. It is found that as the substrate is heavily doped, there is very little loss of light due to the fact that this is not the active area.
As light is mostly absorbed within a certain distance, the thickness of the intrinsic layer is normally made to match this. Any increase in thickness over this will tend to reduce the speed of operation - a vital factor in many applications, and it will not improve the efficiency greatly.
It is also possible to have the light enter the photo diode from the side of the junction. By operating the photo diode in this fashion the intrinsic layer can be made much less to increase the speed of operation, although the efficiency is reduced.
In some instances a heterojunction may be used. This form of structure has the additional flexibility that light can be received from the substrate and this has a larger energy gap which makes it transparent to light.
PIN photodiode heterojunction structure
The heterojunction format for a PIN photodiode uses less standard technology often using materials such as the InGaAs and InP depicted in the diagram. Being a less standard process, it is more expensive to implement and as a result tends to be used for more specialist products.
PN / PIN photodiode materials
The materials used within a photodiode determine many of its critical properties. The wavelength of light to which it responds and the level of noise are both critical parameters that are dependent upon the material used in the photodiode.
The wavelength sensitivity of the different materials occurs because only photons with sufficient energy to excite an electron across the bandgap of the material will produce significant energy to develop the current from the photodiode.
| MATERIAL | WAVELENGTH SENSITIVITY (NM) |
|---|---|
| Germanium | 800 - 1700 |
| Indium gallium arsenide | 800 - 2600 |
| Lead sulphide | ~1000 - 3500 |
| Silicon | 190 - 1100 |
While the wavelength sensitivity of the material is very important, another parameter that can have a major impact on the performance of the photodiode is the level of noise that is produced. Because of their greater bandgap, silicon photodiodes generate less noise than germanium photodiodes. However it is also necessary to consider the wavelengths for which the photodiode is required and germanium photodiodes must be used for wavelengths longer than approximately 1000 nm.
Avalanche photodiode structure
The avalanche photodiode structure is relatively similar to that of the more commonly used PN photodiode structure or the structure of the PIN photodiode. However as the avalanche photodiode is operated under a high level of reverse bias a guard ring is placed around the perimeter of the diode junction. This prevents surface breakdown mechanisms.
Avalanche PIN photodiode structure
Avalanche photodiode materials
Like the standard PN or PIN photodiodes, the materials used have a major effect on determining the characteristics of the avalanche diode.
| MATERIAL | PROPERTIES |
|---|---|
| Germanium | Can be used for wavelengths in the region 800 - 1700 nm. Has a high level of multiplication noise. |
| Silicon | Can be used for wavelengths in the region between 190 - 1100 nm. Diodes exhibit a comparatively low level of multiplication noise when compared to those using other materials, and in particular germanium. |
| Indium gallium arsenide | Can be used for wavelengths to 1600 nm and has a lower level of multiplication noise than germanium. |
For optimum noise performance the large difference in the ionisation coefficients for electrons and holes is needed. Silicon provides a good noise performance with a ratio between the different coefficients of 50. Germanium and many group III-V compounds only have ratios of less than 2. While the noise performance of these materials is much inferior, they need to be used for longer wavelengths that require the smaller energy gap offered.
Photodiode Operation & Theory
Although there are several different types of photodiode, they all utilise the same basic principles, although some are enhanced by other effects.
different types of photodiode work in slightly different ways, the basis of operation of all photodiodes remains the same.
Light energy can be considered in terms of photons or packets of light. When a photon of sufficient energy enters the depletion region of a semiconductor diode, it may strike an atom with sufficient energy to release the electron from the atomic structure. This creates a free electron and a hole (i.e. an atom with a space for an electron). The electron is negatively charged, while the hole is positively charged.
The electrons and holes may remain free, or other electrons may combine with holes to form complete atoms again in the crystal lattice. However it is possible that the electrons and holes may remain free and be pulled away from the depletion region by an external field. In this way the current through the diode will change and a photocurrent is produced.
PIN / PN photodiode Operation
The photodiode is operated under a moderate reverse bias. This keeps the depletion layer free of any carriers and normally no current will flow. However when a light photon enters the intrinsic region it can strike an atom in the crystal lattice and dislodge an electron. In this way a hole-electron pair is generated. The hole and electron will then migrate in opposite directions under the action of the electric field across the intrinsic region and a small current can be seen to flow. It is found that the size of the current is proportional to the amount of light entering the intrinsic region. The more light, the greater the numbers of hole electron pairs that are generated and the greater the current flowing.
Operating diodes under reverse bias increases the sensitivity as it widens the depletion layer where the photo action occurs. In this way increasing the reverse bias has the effect of increasing the active area of the photodiode and strengthens what may be termed as the photocurrent.
It is also possible to operate photodiodes under zero bias conditions in what is termed as a photovoltaic mode. In zero bias, light falling on the diode causes a current across the device, leading to forward bias which in turn induces "dark current" in the opposite direction to the photocurrent. This is called the photovoltaic effect, and is the basis for solar cells. It is therefore possible to construct a solar cell using a large number of individual photodiodes. Also when photodiodes are used in a solar cell, the diodes are made larger so that there is a larger active area, and they are able to handle higher currents. For those used for data applications, speed is normally very important and the diode junctions are smaller to reduce the effects of capacitance.
When not exposed to light the photo diode follows a normal V-I characteristic expected of a diode. In the reverse direction virtually no current flows, but in the forward direction it steadily increases, especially after the knee or turn on voltage is reached. This is modified in the presence of light. When used as a photo-diode it can be seen that the greatest effect is seen in the reverse direction. Here the largest changes are noticed, and the normal forward current does not mask the effects due to the light.
Avalanche photodiode diode operation
Light enters the un-doped region of the avalanche photodiode and causes the generation of hole-electron pairs. Under the action of the electric field the electrons migrate towards the avalanche region. Here the electric field causes their velocity to increase to the extent that collisions with the crystal lattice create further hole electron pairs. In turn these electrons may collide with the crystal lattice to create even more hole electron pairs. In this way a single electron created by light in the un-doped region may result in many more being created.
The avalanche photodiode has a number of differences when compared to the ordinary p-i-n diode. The avalanche process means that a single electron produced by light in the un-doped region is multiplied several times by the avalanche process. As a result the avalanche photo diode is far more sensitive. However it is found that it is not nearly as linear, and additionally the avalanche process means that the resultant signal is far noisier than one from a p-i-n diode. The structure of the avalanche diode is also more complicated. An n-type guard ring is required around the p-n junction to minimise the electric field around the edge of the junction. It is also found that the current gain is dependent not only on the bias applied, but also thermal fluctuations. As a result it is necessary to ensure the devices are placed on an adequate heat sink.
0 comments:
Post a Comment