Intrinsic Photovoltaic Detectors

Peterson and Casselman [8.71] have been extending Petersen s earlier calculations of Auger recombination in n-type Hg, jjCd,Te to p-type material. The purpose is to determine how important Auger recombination in fact is in p-type Hg, jjCd Te. Their calculations assume Au r transitions involving the light-hole valence band as well as the heavy-hole valence band. The result will help determine the maximum Z)J really possible in Hg, Cd Te photovoltaic detectors. 

Development of Hg, jjCd Te photovoltaic detector technology continued. See, for example, the publications of Fiorito and co-workers [8.72a] and of Becla and Pawlikowski [8.72b], and the review by Dornhaus and Nimtz [8.73]. 

---

Recent Advances in Optical and Infrared Detector Technology 307 8.2.1 Intrinsic Photovoltaic Detectors... 

The second photon effect of general utility is the photovoltaic effect. Unlike the photoconductive effect, it requires an internal potential barrier with a built-in electric field to separate a photoexcited hole-electron pair. Although it is possible to have an extrinsic photovoltaic effect, see Ryvkin [2.32], almost all practical photovoltaic detectors employ the intrinsic photoeffect. Usually this occurs at a simple p — n junction. However, other structures employed include those of an avalanche, p—i — n, Schottky barrier and heterojunction photodiode. There is also a photovoltaic effect occuring in the bulk. Each will be discussed, with emphasis on the p—n junction photoeffect. 

In contrast to photoconductivity, the photovoltaic effect depends largely upon the minority carrier lifetime. This is because the presence of both the photoexcited electron and photoexdted hole is required for the intrinsic effect to be observed. Because the minority carrier lifetime is usually shorter than the majority carrier one, the photovoltaic signal terminates when the minority carrier recombines. The time dependent photosignal is given by (2.1S) and (2.16), but the appropriate lifetime is the minority carrier one. For this reason, photovoltaic detectors are usually faster than photoconductive ones made from the same material. 

The important photovoltaic detectors use intrinsic photoexcitation and so are made of semiconductors with energy gaps satisfying condition 1 for the wavelength to be detected. Using the /j(K) term of (4.10) and letting G = 1 for these photovoltaic detectors, condition 2 becomes... 

We saw in Subsection 4.2.2 that satisfaction of condition 1 was almost sufficient for a semiconductor to be a satisfactory photovoltaic detector material conditions 2 through 4 placed demands mainly on device design and technology rather than on fundamental properties of the semiconductor material. The situation is more restrictive for intrinsic photoconductive detectors, because condition 3 places specific demands on fundamental material properties which eliminate some classes of semiconductors as satisfactory high-performance detector materials. Let us consider condition 3 in a qualitative way next to determine which materials may be satisfactory. We shall treat conditions 2 and 3 in terms of a quantitative example later. 

Thus the zinc blende structure semiconductors can be useful for intrinsic photoconductive detectors. Compounds such as InSb have been used as intrinsic photoconductors [4.20], as well as for photovoltaic detectors, but greater versatility of wavelength response is possible with the Hg j tCd Te alloy system. The Hgi j.Cd,Te alloys have received considerable development effort in recent years and are the most prominent intrinsic photoconductor materials they will be analyzed in this subsection. The development of Hg, Cd Te has concentrated almost entirely on n-type material since it provides high photoconductive gain however, p-type Hg, Cd,(Te crystals may be useful for intrinsic photoconductive detectors also [4.21]. 

Condition 4 provides the highest possible BLIP detectivity by requiring that the quantum efficiency approach its maximum value of unity. This condition is easily met by relatively thin photovoltaic and intrinsic photoconductive detectors. However, it is a major problem for extrinsic Si photoconductors, because limited maximum values of dopant concentrations and absorption cross sections give rather low absorption coefficients, requiring undesirably thick detectors for high quantum efficiencies. 

The detector characteristic may very well be included in the filter design. For example, an indium arsenide photovoltaic detector, operating at 195 K, has a very sharp cut-off at 3.6 m. In combination with a thin germanium window, a well-defined 1.9-3.6 m response function is obtained. However, with a limited number of substances available for the design of filters based on intrinsic absorption and reflection phenomena other methods must be found to constmct filters where the transmission limits can be set by the scientific objectives and not so much by the absorption properties of available substances such methods are based on the interference principle, to be discussed in Section 5.6, but first we deal with prism spectrometers, gas filters, and pressure modulation. 

In photovoltaic detectors, also called photodiodes, electrons and holes generated by photons are separated by an electric field formed at a potential barrier within the device. This field is formed at a specifically introduced interface in the material, for instance a p-n junction. Photodiodes used in astronomy are almost always made of intrinsic material. When an electron-hole pair is created at the junction by a photon the electron drifts to the -region and the hole to the p-region. The separation of charge causes a voltage to be generated across the detector terminals, which can be sensed directly or, alternatively, a current can be measured when the circuit is completed. 

The basic theory of photovoltaic and photoconductive detectors shall be presented in Section 4.1 in a unified form convenient for intercomparison of the two effects and of the various detector materials. Then Sections 4.2,4.3, and 4.4 shall cover photovoltaic, intrinsic photoconductive, and extrinsic photoconductive detectors, respectively, each of these sections including first a subsection in which the general theory of Section 4.1 is specialized to that class of detector, and then a subsection in which specific materials suitable for that class of detector are evaluated in terms of the theory. Finally in Section 4.5 we will draw some conclusions about the status and prospects of photovoltaic and photoconductive infrared detectors. Symbols used in this chapter which are not defined in the text are defined in Table 4.1. 

This work is dedicated to an analysis of the possible improvements of the performance of photonic detectors with a goal to use them at room temperature and to approach them as much as possible to the previously described idealized IR detector element. The presentation is limited to photoconductive and photovoltaic intrinsic detectors, although for the most part it can be generalized to any type of... 

---