Photodiode Construction
Silicon photodiodes are
constructed from single crystal silicon wafers similar to those used in the
manufacture of integrated circuits. The major difference is that photodiodes
require higher purity silicon. The purity of silicon is directly related to its
resistivity, with higher resistivity indicating higher purity silicon. Centro
Vision products utilize silicon whose resistivities range from 10 Ohm-cm to
10,000 Ohm-cm.
A cross section of a
typical silicon photodiode is shown in the figure. N type silicon is the
starting material. A thin "p" layer is formed on the front surface of
the device by thermal diffusion or ion implantation of the appropriate doping
material (usually boron). The interface between the "p" layer and the
"n" silicon is known as a pn junction. Small metal contacts are
applied to the front surface of the device and the entire back is coated with a
contact metal. The back contact is the cathode, the front contact is the anode.
The active area is coated with either silicon nitride, silicon monoxide or
silicon dioxide for protection and to serve as an anti-reflection coating. The
thickness of this coating is optimized for particular irradiation wavelengths.
As an example, a Centro Vision Series 5-T photodiode has a coating which
enhances its response to the blue part of the
spectrum.
The characteristics of
pn junctions are well known. However, photodiode junctions are unusual because
the top "p" layer is very thin. The thickness of this layer is
determined by the wavelength of radiation to be detected. Near the pn junction
the silicon becomes depleted of electrical charges. This is known as the
"depletion region". The depth of the depletion region can be varied
by applying a reverse bias voltage across the junction. When the depletion
region reaches the back of the diode the photodiode is said to be "fully
depleted". The depletion region is important to photodiode performance
since most of the sensitivity to radiation originates there.
The capacitance of the
pn junction depends on the thickness of this variable depletion region.
Increasing the bias voltage increases the depth of this region and lowers
capacitance until the fully depleted condition is achieved. Junction
capacitance is also a function of the resistivity of silicon used and active
area size. The relationship between junction capacitance, bias voltage and area
is shown in the graph below.
When light is absorbed
in the active area an electron-hole pair is formed. The electrons and holes are
separated electrons passing to the "n" region and holes to the
"p" region. This results in a current generated by light (usually
abbreviated Isc). The migration of electrons and holes to their respective
region is called "The Photovoltaic Effect".
Silicon photodiodes are
most useful as current generators although a voltage is also generated by
illumination. Most of the data supplied in this manual refers to the short
circuit current characteristics of the photodiodes. The short circuit current
is a linear function of the irradiance over a very wide range of at least seven
orders of magnitude. The Isc is only slightly affected by temperature, varying
less than 0.2% per degree C for visible wavelengths. A recently published
independent laboratory study has shown Centro Vision photodiodes to have Isc
stability better than +/-0.25% per year.
Approximate Photdiode Short Circuit Currents for Various Light
Sources
Part Number
|
Sunlight at Noon, mA
|
Room Light On Table, microA
|
Super Red LED at 10 mA, 1 CM Away
, microA
|
Laser Pointer @ 1 meter, mA
|
OSD1-5T
|
0.47
|
0.45
|
0.32
|
0.71
|
OSD5-5T
|
1.80
|
2.10
|
1.70
|
1.00
|
OSD15-5T
|
4.50
|
5.60
|
2.60
|
1.00
|
OSD35-5T
|
11.00
|
14.00
|
3.80
|
1.10
|
OSD60-5T
|
28.00
|
39.00
|
7.20
|
1.10
|
It must be noted that
when a reverse bias is applied some current will flow without illumination. The
"dark current" is specified for every device. In cases where a very
low bias voltage is used, shunt resistance is specified. This is determined by
measuring dark current with +/-0.010 volts applied bias.
A photodiode has two
terminals, a cathode and an anode. It has a low forward resistance (anode
positive) and high reverse resistance (anode negative). Normal biased operation
of most photodiodes described in this catalog calls for negative biasing the
active area of the device which is the anode or positive biasing the backside
of the device, which is the cathode.
In the photovoltaic and
zero bias modes, the generated current or voltage is in the diode forward
direction. Hence the generated polarity is opposite to that required for the
biased mode.
The measure of
sensitivity is the ratio of radiant energy (in watts) incident on the
photodiode to the photocurrent output in amperes. It is expressed as the
absolute responsivity in amps per watt. Please note that radiant energy is
usually expressed as watts/cm^2 and that photodiode current as amps/cm^2. The
cm^2 term cancels and we are left with amps/watt (A/W). A typical responsivity
curve that shows A/W as a function of wavelength is given below.
The wavelength of the
radiation to be detected is an important parameter. As can be seen from the
graph, silicon becomes transparent to radiation of longer than 1100 nm
wavelength. It is not therefore suitable for use at wavelengths appreciably
longer than this. Ultraviolet light is, conversely, absorbed in the first 100
nm thickness of the silicon. Even the most careful surface preparation leaves
some surface damage which reduces the collection efficiency for this
wavelength. Surface coatings further affect the spectral response of the
device. It is normal to apply anti-reflection coatings which enhance the
response (by up to 25%) at the required wavelength. These coatings may reduce
the efficiency at other wavelengths which they reflect. The package window further
modifies the spectral response. The standard glass window absorbs wavelengths
shorter than 300 nm. For UV detection, a fused silica or UV transmitting glass
window is necessary. Various filter windows are also available to tailor the
spectral response to suit the application. One specific filter which is of
great interest, modifies the normal silicon response to approximate the
spectral response of the human eye.
The output of photodiode
when reverse-biased is extremely linear with respect to the illuminance applied
to the photodiode junction, as shown in the graph.
Effect of Reverse Bias on Photodiode Linearity
A photodiode's
capability to convert light energy to electrical energy, expressed as a
percentage, is its Quantum Efficiency, (Q.E.). The sensitivity of a photodiode
may also be exppressed in practical units of amps of photodiode current per
watt of incident illumination. The QE is related to the photdiode's
responsivity by the following equation:
Operating under ideal
conditions of reflectance, crystal structure and internal resistance, a high
quality silicon photodiode of optimum design would be capable of approaching a
Q.E. of 80%. The following reference table identifies, at a Q.E. of 100%, the
responsivity of an ideal photodiode over the 200-1100 nm wavelength range. It
should be noted that a Q.E. of 100% is not attainable.
Wavelength, nm
|
Responsivity at 100% Q.E. A/W
|
200
|
0.161
|
300
|
0.242
|
400
|
0.323
|
500
|
0.403
|
600
|
0.484
|
700
|
0.565
|
800
|
0.645
|
900
|
0.726
|
1000
|
0.806
|
1100
|
0.887
|
Increasing the operating
temperature of a photodiode device results in two distinct changes in operating
characteristics. The first change is a shift in the Quantum Efficiency (Q.E.)
due to changes in the radiation absorbtion of the device. Q.E. values shift
lower in the UV region and higher in the IR region. See figure below:
The second change is
caused by exponsntial increases in the thermally excited electron-hole pairs
resulting in increasing dark current. This leakage doubles for each 8 to 10 deg
C temperature increase, as shown below:
In many design
applications, the designer needs to know the minimum detectable light (power)
of the photodiode. The minimum incident power required on a photdiode to
generate a photocurrent eual to the toatal photodiode noise current is defined
as the noise equivalent power, or NEP.
The NEP is dependent on
the bandwidth of the measuring system; to remove this dependence the figure is
divided by the square root of the bandwidth. This gives the NEP the units of
watts/HzE-0.5. Since the photodiode light power to current conversion depends
on the radiation wavelength, the NEP power is quoted at a particular
wavelength. The NEP is non-linear over the wavelength range, as is
responsivity.
The noise generated by a
silicon photodiode, operating under reverse bias, is a combination of shot
noise, due to dark leakage current, and Johnson noise due to the shunt
resistance of the device and the ambient temperature. The Shot Noise current
produced by the reverse leakage current of a device is given by the formula:
The Johnson noise
contribution is provided by the shunt resistance of the device, series
resistance and the load resistance. The Johnson noise is given by:
The total noise current
is the root mean square sum of the individual noise current contributions.
As an example: If a
photodiode has a dark leakage current of 2 nA and a shunt resistance of 5E8
Ohms, and a responsivity of 0.5 A/W, and letting the bandwidth of the system be
1 Hz,
Shot noise is the
dominant component of the noise current of a reverse-biased photodiode. This is
particularly true at higher voltages. If devices are operated in a photovoltaic
mode with zero bias, the Johnson noise dominates, as dark current approaches
zero. When operating in the zero bias mode the noise current is reduced such
that the NEP, and hence the minimum detectable signal, is reduced in spite of
some loss of absolute sensitivity.
This is the measure of
the photodiode response speed to a stepped light input signal. It is the time
required for the photodiode to increase its output from 10% to 90% of final
output level (see response Time, below)
Maximum Reverse Voltage (Vr)
Applying excessive
reverse voltage to photodiodes may cause breakdown and severe degradation of
device performance. Any reverse voltage applied must be kept lower than the
maximum rated vale, (Vr max).
Response Time
In many applications the
most important parameter is dynamic performance. Photodiode response time is
the root mean square sum of the charge collection time and the RC time constant
arising from series plus load resistances and the junction and stray
capacitances. Charge collection time is voltage dependent and is made up of a
fast and a slow component. The fast component is the transit time of the charge
carriers (electrons and holes) through the depletion region, producing carriers
that are collected by diffusion. The transit time of these carriers will be
relatively slow. The figure below illustrates the transient response of a
photodiode to a square pulse of radiation.
When a photodiode is
operated in the unbiased mode, the slow diffusion component dominates, giving
risetimes on the order of 0.5 microseconds.
For a fast response
time, silicon resistivity and operating voltage must be chosen to produce a
depletion layer within which the majority of the carriers are generated. In
this case the transit time will be dependent on both the electron and hole
drift velocities. The depletion depth necessary for full absorbtion increases
rapidly with operationg wavelength. Response times increase correspondingly.
This makes it difficult to achieve risetimes faster than 15-20 ns at 1064 nm,
whereas risetimes of less than 2 ns are obtainable below 900 nm.
The Centro Vision -3T
and -4X series take advantage of the increase in drift velocity resulting from
a very high electric field. In this structure silicon thickness is reduced to
just contain the required depletion depth, and a heavily doped back layer is
used to supply the necessary charge to support the depletion region at higher
voltage. In this way the operating field, and hence the carrier drift drift
velocities, may be increased without a significant increase in depletion depth.
Further increase in speed may be obtained at the expense of overall sensitivity
by using silicon which is not thick enough to allow full absorbtion of incident
radiation.
The equivalent circuit
of a photodiode is shown in the figure.
Fundamentally a
photodiode is a current generator. The junction capacitance of the photodiode depends
on the depletion layer depth and hence bias voltage. The value of the shunt
resistance is usually high (megohms). The series resistance is low. The effect
of the load resistor value on the current/voltage characteristics is shown in
the following figure:
Photovoltaic Operation - Rl>>Rd, load line (a)
The generated
photocurrent flows through Rd causing a voltage across the diode. This voltage
opposes the band gap potnetial of the photodiode junction, forward biasing it.
The value of Rd drops exponsntially as the illumination increases. Thus the
photo-generated voltage is a logarithmic function of incident light intensity.
The major disadvantage of this circuit is that the signal depends on Td, which
typically has a wide spread of values over different production batches. The
basic circuit is shown below:
Zero Bias Operation - Rl<<Rd, load line (b)
The generated
photocurrent flows through Rl which is fixed. The resultant voltage is
therefore linearly dependent on the incident radiation level. One way to
achieve sufficiently low load resistance, and an amplified output voltage, is
by feeding the photocurrent to an operational amplifier virtual ground as shown
below. The circuit has alinear response and has low noise due to the almost
complete elimination of leakage current.
Photoconductive Operation - load line (c)
In the photoconductive
mode, the generated photocurrent produces a voltage across a load resistor in
parallel with the shunt resistance. Since, in the reverse biased mode Rd is
substantially constant, large values of Rl may be used still giving a linear response
between output voltage and applied radiation intensity. This form of circuit is
required for high speed of response. The main disadvantage of this mode of
operation is the increased leakage current due to the bias voltage, giving
higher noise than the other circuit modes already described. Practical
photoconductive mode circuits are shown below. (Note that in both circuits the
photodiode is reverse-biased.)
Hybrid Amplifiers
It is now possible to
produce a miniature hybrid photodiode and transimpedance amplifier in a package
little different from the basic photodiode. This reduces lead lengths and stray
capacitances at the small signal, high impedance amplifier inputs. Noise pickup
and amplifier generated noise are therefore both kept to the absolute minimum
using this technique. Hence, for low noise, high frequency and user convenience
a hybrid circuit is the optimum device. Centro Vision has several standard
photodiode/op-amp hybrids. Please see our OSI Series.
Junction Capacitance
The junction capacitance
of a photodiode depends on its area and the bias voltage, as shown below:
No comments:
Post a Comment