Photodiodes are structured from semiconductors p-n, but the basic radiation process involved is absorption. The light that falls on the junction leads to the formation of electron-hole pairs. In photovoltaic mode, i.e., no bias is applied, and the electron-hole pair migrates to the opposite side of the junction, resulting in a voltage (and current, if the device is connected in the circuit). However, most photodiodes operate in light guide mode, which applies a reverse bias to the junction. There are distinct advantages to working in this mode [1]. The reverse bias increases the width of the depletion region, resulting in a larger photosensitive region, allowing for more light collection. In addition, the bias creates a strong field in the junction that rapidly sweeps the carriers, making recombination less likely. This guarantees high quantum efficiency or efficient conversion of photons to charge carriers. There are also advantages in terms of response time. In a back-biased photodiode, the current generated by the bias and charge carriers is proportional to the incident light intensity over a wide dynamic range.
The key difference between a semiconductor photon source and a photon detector is that the former requires the use of direct bandgap semiconductors, while the latter can use indirect bandgap semiconductors. While the requirement for both conservation of energy and momentum greatly reduces the likelihood of photon radiation in indirect bandgap semiconductors, this is not the case with absorption. A two-step process is easier to implement, in which electrons are excited to a higher energy level in the conduction band, followed by a relaxation process in which the momentum of the electrons is transferred to the phonons. Since these two steps can occur sequentially, there is a much higher probability of a radiation process than if both steps had to occur at the same time. As a result, group IV element semiconductors such as silicon (Si) and germanium (Ge) are effective photonic detectors, similar to direct bandgap III-V systems such as Gaas or Ingaas. With SI being ubiquitous in electronic circuits and devices, it's no surprise that SI photodiodes are the most commonly used photodetectors in instrumentation (see Figure 1 for a typical device architecture). The spectral responsivity of the Si covers the UV, VIS and Nir bands. Photodiodes that utilize other semiconductor materials may cover other parts of the electromagnetic spectrum.
Photodiodes have some characteristics that distinguish them from thermal detectors. The photon-to-electron conversion is very rapid, so photodiodes have the potential to respond to rapidly changing radiation levels with a detection rate that can be significantly higher than that of a thermal detector. The detection mechanism is strongly wavelength-dependent, that is, there is a peak in responsivity, which decreases at short wavelengths due to the photon-to-watt conversion, and at long wavelengths due to the minimum photon energy required to generate electron-hole pairs. The dynamic range of a photodiode can be very large, with a single detector exceeding 10 (10). Due to their high detection rate and large dynamic range, photodiodes are often used to measure optical power over a wide range. For continuous or quasi-continuous light sources, this is intuitive; For pulsed light sources, the procedures outlined in this article can be used to estimate pulse energies. As long as the time response can accommodate pulse integration, photodiodes can also be used as energy sensors. This results in a decrease in the dynamic range, with the lower limit of the range being reduced due to the reduced detection sensitivity (due to the faster time response) and the upper limit of the range being due to the saturation of the detector's linear response. This is because the electron-hole pairs begin to recombine without flowing through the circuit.