Photon counting applications play an important role in advancements in many different areas such as astronomy, particle sizing, disease diagnosis, medical analysis and imaging methods as well as the most recent area of quantum cryptography. The most spectacular application is surely biomedical research. The use of confocal and STED microscopy led to major advancements in the fluorescence analysis of the structure and function of biological molecules.
Primarily two types of technology are used to detect a small amount of photons:
Photomultiplier – PMT
A photomultiplier typically consists of a photocathode and a downstream secondary electron multiplier in an evacuated (10-6...10-5 Pa) glass tube.
Working principle. The photons encounter a photocathode which, depending on the material used, emits an electron in a spectral range starting from 115 nm. The electrons that are set free are accelerated in an electrical field and encounter other electrodes – called dynodes. To ensure that the electrons remain free, the entire assembly is located in a vacuum tube; a high voltage supply of 1-2 kV is required. The electron that encounters the dynode then causes several electrons, so-called secondary electrons, to be emitted. These secondary electrons are accelerated toward the next dynode.
This process is repeated by several dynodes connected in series, producing an avalanche of electrons. This large number of electrons is finally absorbed by the anode, and an electrical pulse is generated that is detected by an electronic counter.
Single Photon Avalanche Diode – SPAD
Avalanche photodiodes, APDs, are highly sensitive and fast photodiodes. They differ from “normal” PIN photodiodes in that during the detection of photons a charge carrier avalanche occurs internally. For this to happen, it is necessary that a bias voltage be applied to the APD to expand the depletion layer.
Specially manufactured Si APDs can also be used as photon counters in “Geiger mode” above the breakdown voltage (VR > VBR). Here, a single photoelectron can initiate an avalanche pulse of approximately 108 charge carriers. These types of APDs are also referred to as single photon avalanche diodes (SPADs).
Working principle. Normally, an APD that is operated above the breakdown voltage will conduct a very large current. By using an appropriate circuit, the diode has to be prevented from staying conductive, which would be the result of the large current; the easiest way to accomplish this is by using a series resistor. The voltage drop at the series resistor leads to a reduction in the bias voltage across the APD, which then returns to its ready state as a result. This is referred to as passive quenching. This process is repeated automatically, and the current pulses can be counted. In active quenching, the bias voltage is electronically actively reduced within a few nanoseconds of a breakthrough current being detected. Increasing the bias voltage again to exceed the breakdown voltage leads to the reactivation of the SPAD. Signal processing by the electronics produces dead times of approximately 50 ns allowing count rates of up to 10 MHz to be achieved.
Single Photon Counting Modules
For the user to get the most out of these types of SPADs in terms of performance, complete single photon counting modules are available. In addition to the cooled SPAD, these modules contain complete electronics, including a stabilized high voltage supply and temperature control, in a compact housing. At the output end of these modules, a count pulse can be measured.
Advantages of Each Individual Technology
SPADs have significantly higher quantum efficiency (QE) and a large measurement range from 300 nm to the NIR range.
Quantum efficiency refers to the relationship of electrons produced to the incident photons in percentage; it depends on the wavelength. If all other detector properties are the same, the detector with the greatest quantum efficiency is the best choice.
When counting single photons, the noise produced by the detector plays an important role. Because one generally counts single photons, the noise of such detectors is no longer given in fW/sqrt [Hz], but rather in c/s (counts/second), which is referred to as the dark count rate.
It is important to know that both the quantum efficiency and the dark count rate depend on the operating voltage applied to the detector (see table). The trick is to produce a special diode that makes it possible to achieve maximum quantum efficiency at the smallest possible dark count rate.
Overvol- tage [V] | QE @ 405 nm | QE @ 670 nm | QE @ 810 nm | Dark count rate |
---|---|---|---|---|
2.0 | 30% | 55% | 32% | 15.4 c/s |
4.1 | 36% | 69% | 43% | 31.4 c/s |
6.3 | 40% | 79% | 51% | 57.4 c/s |
8.0 | 43% | 85% | 55% | 91.4 c/s |
10.7 | 45% | 90% | 60% | 138.2 c/s |
The “VLoK APD” developed by LASER COMPONENTS meets these requirements. The VLoK APD makes that possible today which was unimaginable in the past: dark count rates < 10 c/s at a simultaneous quantum efficiency of > 80 % in the red spectral range. Diodes produced now even exhibit efficiencies > 90 % @ 670 nm.
Alternatives at Shorter Wavelengths
For single photon counting at shorter wavelengths there used to be no way around using PMTs, particularly because the quantum efficiency of SPADs lacked significantly. Depending on the photocathode material used photomultipliers detect single photons up to 115 nm. The quantum efficiency in this range is approximately 10 - 20 %, in the blue spectral range approximately 30 %. With the COUNTblue LASER COMPONENTS has recently started offering a UV-enhanced single photon counting module that exhibits efficiencies of > 50 % at 405 nm.