FROM THE CURRENT LITERATURE

A “CHERENKOV COUNTER” FOR COSMIC-RAY PARTICLES

A new type of luminescence, arising when charged particles pass through a substance with velocities exceeding the phase velocity of light \(\left(\dfrac{c}{n}\right.\), where \(n\) is the refractive index\(\left.\right)\), was discovered by the Soviet physicist P. A. Cherenkov in 1934–1940[^1][^2].

The theory of this phenomenon is given in detail in the works of I. E. Tamm and I. M. Frank[^3][^4]. Owing to its directionality and other properties, Cherenkov radiation is of great interest to the experimentalist as a means of detecting and determining the direction and velocity of fast charged particles. Attempts made in 1947 to use the Cherenkov effect for counting fast electrons (Dicke, 1947[^5][^6]), and also particles of cosmic radiation[^5][^7], did not give sufficiently reliable results.

In the present note successful experiments are described on the registration of charged particles of cosmic radiation (\(\mu\)-mesons) by means of the Cherenkov scintillations produced by them in distilled water[^8]. The scintillations were recorded with the help of a photomultiplier. The success of the experiments is explained by the low noise level of the photomultiplier and the small time constant of the amplification circuit used. In Fig. 1 is shown the arrangement of the Geiger counters and the “Cherenkov counter,” which in this case operated in coincidence. Geiger counters with an effective area of \(120\ \mathrm{cm}^2\) were placed above a cylindrical glass cuvette filled with distilled water. To reduce the loss of light, the walls of the cuvette were silvered, and the upper end was covered with black paper. At the lower end of the cuvette there was a light-collecting cone coated with magnesium oxide, which illuminated the photocathode \(K\) of an 11-stage photomultiplier of type 5311. The pulses at the output of the multiplier were amplified and, after passing through a discriminator, entered one of the channels of a coincidence circuit with a resolving power \(\tau = 1.8\ \mu\mathrm{sec}\).

Fig. 1.

The Geiger counters, through a cathode follower, were connected to the second input of the coincidence circuit. The amplifier had a time constant equal to \(3.2 \cdot 10^{-8}\ \mathrm{sec}\), which ensured a considerable reduction of the photomultiplier noise level relative to the short pulse from Cherenkov radiation, whose duration in the present case did not exceed \(10^{-9}\ \mathrm{sec}\). The choice of water as the medium in which the luminescence arose was made in order to avoid the registration of scintillations

of “non-Cherenkov” origin. The cylindrical shape of the cuvette was convenient for control experiments, when the cuvette and the multiplier were rotated upward.

With the arrangement of the counters shown in Fig. 1, a series of countings of the number of coincidences between the pulses of the Geiger counters and the photomultiplier was carried out. The number of random coincidences was calculated and subtracted from the results obtained.

Decisive evidence that the recorded scintillations are due to Cherenkov radiation may be considered the strong directional effect found by the author. As is known, in water the glow must be contained within a cone bounded by an angle of \(41^\circ\) with the direction of the particle path. The radiation angle \(\theta\) is related to the refractive index of the medium \(n\) and to the ratio of the particle velocity to the velocity of light in vacuum \(\beta\) by the relation \(\cos \theta = \dfrac{1}{\beta n}\); for water \(n = 1.33\), and at \(\beta \to 1\), \(\theta = 41^\circ\).

Series of countings of the number of coincidences were carried out for two positions of the counter: with the photocathode under the cuvette (position \(B\)) and with the photocathode at the point \(K'\) above the cuvette (position \(A\)). Series of control countings were also carried out with an empty cuvette in positions \(B\) and \(A\), in order to estimate the number of coincidences recorded when a pulse from a fast particle arises in the photomultiplier itself. The table gives the mean results of such measurements for three different discriminator bias potentials, cutting off part of the pulses that arise.

The table gives the true numbers of coincidences per minute (random ones have been subtracted). The letters \(\Phi\), \(K\), and \(\mathcal{Ж}\) denote, respectively, the photomultiplier, the cuvette, and the water in it. The ratio of the numbers of coincidences \(\dfrac{\mathcal{Ж}_B}{\mathcal{Ж}_A}\) considerably exceeds unity and increases with the discriminator bias potential. In the control experiments the cuvette, instead of water, was filled with a 0.5% solution of terphenyl in xylene (see \(^{6,9}\)). One might have expected that the “ordinary” scintillations arising in this case would give a ratio \(\dfrac{\mathcal{Ж}_B}{\mathcal{Ж}_A}\) less than unity, owing to the greater density of the particle flux at the photocathode in position \(A\). The experiment, however, gave a ratio equal to \(1.26 \pm 0.06\) at a bias of 2 volts. Its magnitude, exceeding unity, may be explained by the fact that in xylene a considerable fraction of the scintillations are Cherenkov ones.

The ratio \(\dfrac{\mathcal{Ж}_B}{\mathcal{Ж}_A}\) for water, which does not tend to infinity, is due to the fact that even in the inverted position (\(A\)), as a result of reflection from the walls, some of the scintillations are capable of producing a recorded pulse. The increase of the ratio \(\dfrac{\mathcal{Ж}_B}{\mathcal{Ж}_A}\) with the discriminator bias is consistent with this.

According to current data on the intensity of cosmic radiation at sea level (see, for example, \(^{10}\)), the number of coincidences per minute for a counter of 100% efficiency and an area of \(57\ \text{cm}^2\), placed at point \(S\) (see Fig. 1), should be 7 per minute, which may be compared with 3.4 for the photomultiplier in position \(B\). In the case of filling the cuvette with a solution of terphenyl in xylene, the number of pulses per minute was 3.5.

Control experiments, in which a 10 cm layer of lead was placed above the Geiger counters, gave a number of coincidences in position \(B\) equal to \(2.67 \pm 0.11\) per minute, compared with \(3.35 \pm 0.04\) without lead. Hence it follows—

Table

gives the authors grounds for the conclusion that 80% of the observed effect is caused by μ-mesons.

Fig. 2 gives the dependence of the number of pulses per minute recorded by the “Cherenkov counter” on the depth of water in the cuvette. From the fact that the asymptotic value is not reached even at a water depth of 20 cm, it is clear that the efficiency of the counter is less than unity. Fig. 3 shows the dependence of the number of pulses per minute on the angle between the axis of symmetry of the “Cherenkov counter” and the vertical. The course of the curve agrees sufficiently well with what is expected if one takes into account that, because of the radiation angle of 41° and internal reflection, a large number of pulses should be expected with the counter lying horizontally, and even at angles \(180^\circ > \varphi > 90^\circ\). The curve of the cosine distribution \((\cos^2 \varphi)\), agreement with which was not to be expected, is given for comparison.

Fig. 2.

Fig. 3.

It is quite possible that, as photomultipliers are improved and, in particular, scintillation schemes with short time constants are strengthened, “Cherenkov counters” will become a widespread and valuable addition to the range of instruments for recording individual fast particles. Their especially important properties should be considered to be their directionality of action and the dependence of the limiting radiation angle on the velocity of the particles, owing to which direct determination of velocity is in principle possible. Like other scintillation counters, the “Cherenkov counter” is characterized by an extremely high speed of response. At the same time, it is possible to obtain very large working volumes. Its sensitivity to nuclear “stars” and γ-quanta (in particular, to the laboratory background) is low, which also constitutes an advantage.

V. Vavilov

CITED LITERATURE

1. P. A. Cherenkov, DAN 2, 451 (1934).
2. P. A. Cherenkov, Transactions of the Phys. Inst. of the Acad. Sci. USSR 2, No. 4 (1944).
3. I. E. Tamm and I. M. Frank, DAN 14, 107 (1937).
4. I. M. Frank, UFN 30, 149 (1946).
5. Dicke, Phys. Rev. 71, 737 (1947).
6. V. M. Kharitonov, UFN 39, 402 (1949).
7. Weisz and Anderson, Phys. Rev. 72, 431 (1947).
8. Jelley, Proc. Phys. Soc. A64, 82 (1951).
9. Reynolds et al., Phys. Rev. 78, 488 (1950).
10. B. Rosen, UFN 38, 222 (1949).