From Current Literature

Antimony–cesium photocells for the ultraviolet

The first indication of the presence of considerable sensitivity of antimony–cesium photocathodes in the ultraviolet region of the spectrum was contained already in a publication by Görlich[^1], who gave the spectral sensitivity distribution curve for these photocathodes, beginning at the red boundary (\(\sim 7500\) Å) and extending to the boundary of the visible region.

Subsequently, many authors[^2], who investigated the spectral sensitivity of Sb–Cs cathodes by means of monochromators having glass optics, ascribed to Sb–Cs cathodes a sensitivity-distribution curve quite unlike that indicated by Görlich. Judging from these data, their sensitivity in the near ultraviolet should not have been significant.

In 1939, while working with a double quartz monochromator (made by Kipp), the author of the present note found that the Sb–Cs photocathode does not give a sharp maximum in the region around 4500 Å followed by a strong decrease of sensitivity toward shorter wavelengths, but that, on the contrary, with an ordinary incandescent lamp, on an Sb–Cs photocell (both in glass bulbs) one can observe appreciable photocurrents down to \(\lambda = 3800\) Å, and in individual cases also have noticeable sensitivity at \(\lambda = 3600\) Å. Since the energy in the radiation of an incandescent lamp in this region is negligible, the absorption of glass is considerable, and the double decomposition excludes the influence of stray light, it remained to be assumed that the sensitivity of Sb–Cs cathodes in this wavelength region is still very high.

Fig. 1.

Further confirmation of this was found in the paper of Glover and Janes[^3], who measured the sensitivity of an Sb–Cs cathode (hidden by them under the pseudonym S-4) of a photocell which also had a glass bulb, down to \(\lambda = 3000\) Å. The course of their curve in the short-wavelength part is governed by the absorption of the glass, as the authors themselves indicated.

In 1941 the author undertook a more detailed investigation of the sensitivity of Sb–Cs photocathodes in the ultraviolet, showing that it extends at least to \(\lambda = 2537\) Å. The main difficulty in this work was to make, by the simplest possible method, a bulb transparent as far as possible into the ultraviolet region. For the indicated wavelength region this was solved in a very simple way, using ordinary glass bulbs with a flat, thin-walled window (Fig. 1). Such windows possess, at a small glass thickness (it was possible to obtain wall thicknesses down to 0.03 mm), a strength quite sufficient for carrying out all technological operations. This work was interrupted in connection with the war and was resumed only in 1943 in collaboration with A. E. Melamid[^4].

Fig. 2 shows the spectral characteristic of such an Sb–Cs photocell. Here the course of the curve in the short-wavelength part is also governed by the absorption of the glass. As follows from the form of the spectral characteristic,

the quantum yield for Sb–Cs cathodes, even in the ultraviolet, amounts to 2–3 tens of percent.

Fig. 2.

This circumstance, together with the simplicity of making such photocells (identical with the manufacture of ordinary Sb–Cs photocells) and their great constancy of sensitivity (Sb–Cs photocathodes exhibit almost no fatigue when operating in the ultraviolet), makes them, in our opinion, a very useful and convenient instrument for various kinds of investigations connected with the observation and measurement of ultraviolet radiation, thereby to the highest degree simply and successfully filling the gap that has existed here in a good photoelectric indicator.

N. Khlebnikov

Literature

1. P. Gorlich, Z. Physik, 101, 335, 1936.
2. See, for example, S. Yu. Luk’yanov, Journal of Technical Physics, 9, 1175, 1939; A. M. Gurevich, Journal of Technical Physics, 10, 943, 1940.
3. A. M. Glover and R. B. Janes, Electronics, Aug., 1940.
4. N. S. Khlebnikov and A. E. Melamid, Journal of Technical Physics, 1945 (in press).