FROM CURRENT LITERATURE

NEW PHOTOCELLS

Following the step forward that consisted in the development of antimony-cesium photocells, quite recently another major step was taken—this time in the field of photocells with blocking layers. At the end of last year, in the semiconductor group of the Leningrad Physico-Technical Institute, headed by Yu. P. Maslakovets, research associate B. T. Kolomiiets developed selenium-thallium photocells with blocking layers, distinguished by quite unusual properties.

The features that bring these photocells to the fore are their enormous integral sensitivity, exceeding the sensitivity of good selenium photocells with blocking layers by 1–2 orders of magnitude and capable of reaching \(8\,000\)—\(10\,000\ \mu\mathrm{A}/\mathrm{Lm}\), as well as the nature of the distribution of sensitivity over the spectrum (Fig. 1), which makes these photocells applicable in the wavelength region that until now was inaccessible to photocells with blocking layers and is of considerable technical interest. In the rest of their properties these photocells¹ do not differ substantially from other photocells with blocking layers.

Fig. 1. Spectral characteristic of the new photocell

Fig. 2. Frequency characteristics of the new photocell at different loads

Of interest is the fact that the experience of using these photocells in sound cinema gave positive results. Here the new photocells provide certain substantial advantages over ordinary photocells with an external photoeffect and electronic amplifiers. Thus, for example, they do not produce their own noise. The frequency characteristic of such a photocell at small load resistances (Fig. 2, curve 1) falls off sufficiently weakly for the

so that it can easily be corrected with an amplifier, and sound reproduction is quite satisfactory. In the case of a large ohmic load, the frequency characteristic deteriorates considerably (curve 2, Fig. 2), because this photocell, like all photocells with blocking layers, has a large intrinsic capacitance, and as the external resistance increases the time constant assumes large values. Therefore, for an amplifier intended to work with such photocells over a considerable range of frequencies, it is necessary to have a transformer input.

No less remarkable than the photocells themselves is the history of their creation, which is a brilliant example of the correct interaction of theory and practice. Before we proceed further, let us point out that in these photocells there occurs not the ordinary photoeffect of the blocking layer, in which electrons pass from the semiconductor into the metal (which may give a face or a back effect, depending on the thickness of the semiconductor layer, the properties of the electrodes and of the interface), but a new effect, never previously observed before these experiments, called by the authors the “positive photoeffect of the blocking layer.” In this effect, the transfer of electrons takes place from the metal (the transparent upper electrode) into the semiconductor.

Fig. 3. Spectral characteristic of the sulfur-thallium photocell under the normal photoeffect of the blocking layer

Fig. 4. Polarity of a photocell with a blocking layer under the normal face (a), back (b) and positive face (c) effect; the arrows indicate the incident light.

In addition to the positive effect, the very same semiconductor can also give the normal photoeffect of the blocking layer. In this case the integral sensitivity of the photocells is low (~10 μA/lm), and the spectral distribution of the sensitivity has an entirely different character (Fig. 3).

In the process of developing new photocells, owing to the peculiarities of the adopted technology, as a rule, photocells of this latter type were obtained. But one day a photocell of exceptionally high sensitivity, ≈ 100 μA/lm, was obtained. It was the first photocell with a positive photoeffect. Investigation showed that its polarity was the reverse of that characteristic of the normal face effect (Fig. 4). Naturally, the idea arose that this was due to predominance of the back effect over the face effect, although this was not connected with its high sensitivity—naturally because nothing was known about the existence of the “positive photoeffect.” But experiment disproved this assumption: no back effect was detected. Thus the positive photoeffect of the blocking layer was discovered.

Thus it turned out that the same materials in one and the same combination can behave in entirely different ways. There was something here to be surprised at and something to think about. It is more than likely that, had the theory of semiconductors not been brought to bear, the new photocells would only have multiplied the number of “miracles of technology” (in the mystical sense), i.e., more precisely, the number of misunderstood and unexplained things. Fortunately this did not happen. The theory was used, and it provided what a correct theory can and should provide: it explained the new effect and indicated how one must proceed in order to obtain precisely photocells with a positive effect.

Fig. 5. Energy diagram of a normal (a) and positive (b) barrier-layer photocell. M—metal, B—blocking layer, S—semiconductor; complex dashed line — impurity energy level; arrow at E—direction of increase of the electron’s potential energy; dashed arrows—direction of motion of the electron (a) and the hole (b).

As Yu. P. Maslakovets showed, the whole matter here lay in the nature of the conductivity of the semiconductor, which determines the energy structure of the photocell. Under the conditions of the existing technology the conductivity obtained was hole conductivity. The energy properties of the system for this case are shown in Fig. 5a. An electron knocked out of the filled levels and transferred into the conduction band encounters no obstacles to passing into the metal—the normal effect is obtained. In the case when the conductivity is electronic, the scheme of Fig. 5b operates. Electrons that have entered the conduction band do not enter the metal because of the presence of a potential barrier, but the electrons of the metal have every chance to occupy the places vacated in the band of filled levels—a positive photoeffect is obtained.

As soon as all this had been clarified, it became possible to construct a technology for photocells that would ensure the production of conductivity of the required type. After this, as is known, the new photocells became an entirely undoubted and very valuable reality.

N. Khlebnikov, Moscow

LITERATURE

1. B. T. Kolomiets, Izvestiya of the Academy of Sciences of the USSR, Physical Series, Nos. 5–6, 1938.