(also photoconductive effect), the increase in the electrical conductivity of a semiconductor when the semiconductor is exposed to electromagnetic radiation. Photoconductivity was first observed in selenium by W. Smith (Great Britain) in 1873.
Photoconductivity is usually caused by an increase in the concentration of charge carriers upon exposure to light; this effect is called primary photoconductivity. The photoconductive effect is the result of several processes whereby photons cause electrons to be ejected from the valence band and injected into the conduction band (Figure 1). The number of conduction electrons and holes increases simultaneously, and the effect is called intrinsic photoconductivity. When electrons from a filled band are injected into vacant impurity levels, the number of holes increases; this effect is referred to as p-type extrinsic photoconductivity. If electrons are ejected from impurity levels and injected into the conduction band, the effect is known as n-type extrinsic photoconductivity. The combined excitation of intrinsic and extrinsic photoconductivity is also possible. Such combined excitation is called exciton-induced photoconductivity and occurs when the excitation of intrinsic photoconductivity leads, as a result of the ensuing processes of carrier trapping, to the occupation of impurity centers and, consequently, to the occurrence of extrinsic photoconductivity. Primary photoconductivity can result only from
Figure 1
excitation by sufficiently short-wavelength radiation, in which the photon energy exceeds either the width of the forbidden band (in the case of intrinsic and exciton-induced photoconductivity) or the distance between the valence or conduction band and an impurity center (in the case of extrinsic photoconductivity).
All nonmetallic solids exhibit photoconductivity to some extent. The photoconductivity of semiconductors, such as Ge, Si, Se, CdS, CdSe, InSb, GaAs, and PbS, is the best studied and most widely used in technology. The magnitude of the primary photoconductive effect is proportional to the quantum efficiency –n, which is the ratio of the number of carriers produced and the total number of absorbed photons, and to the lifetime of excess photocarriers, that is, of excess charge carriers generated by the light. When visible light is used for illumination, η, is usually less than unity because of competing processes that result in the absorption of light but are not associated with the production of photocarriers, that is, associated with the excitation or production of excitons, impurity atoms, or lattice vibrations. When a substance is exposed to ultraviolet or harder radiation, η > 1, since the photon energy is high enough not only to eject an electron from a filled band but also to provide the electron with sufficient kinetic energy for impact ionization.
Figure 2. A typical relative spectral response characteristic for intrinsic photoconductivity. The sharp decrease at longer wavelengths is caused by an absorption edge, that is, by the cutoff of intrinsic absorption that occurs when the photon energy becomes less than the width of the forbidden band. The smooth decrease at shorter wavelengths is caused by surface absorption of light.
The free-carrier lifetime, that is, the mean time a carrier is free, is determined by recombination processes. During direct recombination, a photoelectron migrates directly from the conduction band to the valence band. In the case of recombination at impurities called recombination centers, an electron is first trapped by such a center and then enters the valence band. Depending on the material’s structure, purity, and temperature, the free-carrier lifetime may range from a few fractions of a second to 10–8sec.
The dependence of the photoconductive effect on radiation frequency is determined by the absorption spectrum of a given semiconductor. As the absorption coefficient increases, photoconductivity first reaches a maximum and then declines (Figure 2). The decrease in photoconductivity is explained by the fact that when the absorption coefficient is large, all the light is absorbed in the surface layer of the semiconductor, where the free-carrier recombination rate is very high. Recombination in the surface layer is called surface recombination.
Other types of photoconductivity are possible which are not associated with a change in the free-carrier concentration. For example, when long-wavelength electromagnetic radiation, which does not cause interband migration and does not ionize impurity centers, is absorbed by free carriers, the energy of the carriers is increased. Such an increase leads to a change in carrier mobility and, consequently, to an increase in electrical conductivity. Such secondary photoconductivity decreases at high frequencies and is not frequency dependent at low frequencies. The change in mobility upon exposure to radiation may be caused not only by an increase in carrier energy but also by the effect of the radiation on electron scattering in the crystal lattice.
The study of photoconductivity is one of the most effective ways to investigate the properties of solids. The photoconductive effect is used to produce photoconductive cells and radiation detectors with a low time constant that are sensitive over a very broad wavelength range, namely, from γ-rays to microwaves.
REFERENCES
Ryvkin, S. M. Fotoelektricheskie iavleniia v poluprovodnikakh. Moscow, 1963.
Stil’bans, L. S. Fizika poluprovodnikov. Moscow, 1967.
See also reference under SEMICONDUCTOR.E. M. EPSHTEIN