ABSTRACTS

NUCLEAR PHOTOELECTRIC EFFECT

As was already reported earlier in this journal¹, Chadwick and Goldhaber succeeded in showing that, when heavy hydrogen is irradiated with ThC′ γ-rays, protons arise as a result of the splitting of deuterium by a γ-quantum. Soon after the appearance of Chadwick and Goldhaber’s paper, a work by Szilard and Chalmers² was published, in which the authors showed that the very same effect of destruction of a nucleus by a γ-quantum—the nuclear photoelectric effect—is also observed when beryllium is irradiated with the γ-rays of radium. Subsequently this result was confirmed by a number of works carried out in other laboratories³. In a new paper by Chadwick and Goldhaber⁴ the question of the destruction of nuclei by γ-rays is considered in greater detail. In this work a larger number of elements was investigated; however, positive results were obtained only for the case of heavy hydrogen and beryllium—elements having the smallest mass defects.

Heavy hydrogen

In the case of heavy hydrogen the nuclear reaction must proceed according to the following formula:

\[ {}^{2}_{1}\mathrm{D}+h\nu \to {}^{1}_{1}\mathrm{H}+{}^{1}_{0}\mathrm{n}, \tag{1} \]

moreover, because of the small momentum of the γ-quantum and the approximate equality of the masses of the neutron and proton, the energy liberated in the reaction \((h\nu-W,\) where \(W\) is the binding energy of the neutron and proton in the deuteron) is distributed approximately equally between the two particles.

The authors detected the protons arising in the process under consideration in the following way. An ionization chamber filled with heavy hydrogen, connected to a linear amplifier and an oscillograph, was irradiated with ThC′ γ-rays (an RdTh preparation, equivalent in γ-rays to 9 g Ra), and the number of deflections of the oscillograph caused by ionization of the chamber gas by heavy particles was observed. Control experiments showed that an increase in the number of deflections above the background occurs only for heavy hydrogen and is completely absent in the case of ordinary hydrogen or nitrogen. It follows from these experiments that the additional deflections of the oscillograph observed in the presence of γ-rays are caused by protons arising in the splitting of deuterons by γ-quanta.

From the magnitude of the oscillograph deflections the authors were able to calculate the energy of the protons. For it a value of 240 kV was found (error not more than 80 kV). Knowing the energy of the proton and neutron*, one can calculate, from the energy of the incident γ-quantum (2.6 MeV), the binding energy of the neutron and proton in the deuteron. For it one obtains a value of about 2.1 MeV \((2.6-0.5)\).

To determine the probability of the process under consideration, the authors carried out analogous experiments with a source in the form of an active layer of Th (B + C). From the number of quanta emitted by this source (energy 2.6 MeV) one can calculate the effective cross section for the destruction of the deuteron by a γ-quantum, if

* They are approximately equal.

the known number of deuterons in the ionization chamber and the number of observed deflections of the oscillograph. For the case of the \(\gamma\)-radiation of ThC\('\), the value obtained for the effective cross section was \(6.6\cdot 10^{-28}\ \text{cm}^2\), in rather good agreement with the value calculated from the Bethe and Peierls theory (\(8\cdot 10^{-28}\ \text{cm}^2\)).

The neutrons arising in reaction (1) can be detected by the artificial radioactivity they induce if, in order to increase the effect they produce, they are first slowed down by paraffin. However, for quantitative investigation another method is more convenient, based on the fact that slow neutrons, on entering lithium or boron nuclei, cause their disintegration, accompanied by the emission of heavy charged particles, which can be detected by means of an ionization chamber. In this method of detecting neutrons, the ionization chamber is coated on the inside with a layer of lithium or boron and is connected in the usual way to a linear amplifier and an oscillograph. The advantage of this method is that the reactions mentioned proceed, in the case of slow neutrons, with a very high probability, thanks to which the chamber under these conditions operates with a large efficiency factor. Placing such a chamber together with a source of \(\gamma\)-rays and heavy water in paraffin, the authors established the presence of a large number of neutrons arising from the splitting of deuterons by \(\gamma\)-quanta. For example, for a source of RdTh in \(8\ \text{mg}\) and \(15\ \text{cm}^3\) of heavy water they observed more than 2 thousand deflections per hour, with a background (in the absence of \(\gamma\)-rays) of 40 deflections per hour. By this method the authors compared the magnitude of the effect of deuteron splitting by the \(\gamma\)-rays of ThC\('\) and Ra; it turned out that, calculated for one and the same \(\gamma\)-intensity, the effect in the first case is 27 times greater than in the second. However, one cannot yet conclude from this that in the first case 27 times more neutrons are produced than in the second. The point is that the neutrons arising in the two cases have different velocities, and consequently have different ranges in paraffin and unequal probabilities for destroying lithium or boron nuclei, and therefore they are counted by the chamber in different ratios.

The same method was used by the authors to compare the number of neutrons emitted from heavy water in the direction of the incident \(\gamma\)-rays and at an angle of \(90^\circ\) to them. These experiments showed that in the second case the number of neutrons is approximately twice as large as in the first; thus the neutrons from heavy water are distributed asymmetrically with respect to angle.

Beryllium

Approximately the same experiments as in the case of heavy water were carried out with beryllium. Here, for the splitting of the nucleus by a \(\gamma\)-quantum, two reactions are energetically possible,

\[ {}^{9}_{4}\mathrm{Be}+h\nu \to {}^{8}_{4}\mathrm{Be}+{}^{1}_{0}\mathrm{n}, \tag{2} \]

\[ {}^{9}_{4}\mathrm{Be}+h\nu \to 2\,{}^{4}_{2}\mathrm{He}+{}^{1}_{0}\mathrm{n}. \tag{3} \]

The following facts indicate that the splitting process proceeds, in all probability, according to formula (2).

1. The mass difference calculated from (2) for \({}^{9}_{4}\mathrm{Be}\) and \({}^{8}_{4}\mathrm{Be}\) agrees with the data obtained by Rutherford, Kempton, and Oliphant for the same quantity from the reaction

\[ {}^{9}_{4}\mathrm{Be}+{}^{1}_{1}\mathrm{H}\to{}^{8}_{4}\mathrm{Be}+{}^{2}_{1}\mathrm{D}. \]

1. When an ionization chamber coated on the inside with a layer of beryllium is irradiated with \(\gamma\)-rays, no additional deflections of the oscillograph are observed. The absence of ionizing particles in this case can be explained by the fact that they have a very small range and cannot be detected in the chamber. This, evidently, can occur in the case of reaction (2).

2. Experiments on the absorption of neutrons arising from the irradiation of beryllium by \(\gamma\)-rays show that there are no slow neutrons here.

However, if the disintegration process proceeded with the formation of three particles (3), then such neutrons too would have to exist.

The neutrons produced in the disintegration of beryllium by a neutron can be detected with the aid of a chamber with lithium. A comparison of the magnitude of the disintegration effect for RdTh and Ra sources of the same γ-intensity showed that in the second case the effect is approximately twice as large as in the first. Here, however, too, one cannot compare from this figure the number of neutrons in the two cases, since the neutrons have different energies.

To establish the threshold of the nuclear photoelectric effect for Be, the authors proceeded as follows. An ionization chamber filled with helium was irradiated with neutrons arising in Be under the action of the γ-rays of ThC″ (energy 2.6 MeV). Then, from the magnitude of the deflections of the oscillograph, the maximum energy of the recoil atoms was determined, from which it was possible to calculate the maximum energy of the neutrons entering the chamber. Knowing it, one can, from the mass ratio of the neutron and \(^{8}\mathrm{Be}\), calculate the entire energy liberated in reaction (2). For it a value approximately equal to 1 MeV was obtained. Hence, for the photoeffect threshold for beryllium one obtains \(2.6 - 1 = 1.6\) MeV. Brasch, Lange and others\(^6\), working with X-rays, found that the disintegration of Be with the emission of a neutron already occurs for quanta with energies less than 2 MeV. On the other hand, Arzimowitsch and Palibin\(^7\) showed that for quanta with energies up to 1.3 MeV this effect is completely absent. As can be seen, the value given for the threshold falls within the limits established by the two latter studies.

The magnitude of the effective cross section for the destruction of the beryllium nucleus by a γ-quantum could be estimated only indirectly, by comparing the number of neutrons knocked out of beryllium and heavy water by the γ-rays of ThC″. For \(h\nu = 2.6\) MeV, \(\sigma_{\mathrm{Be}} = 10^{-28}\ \mathrm{cm}^{2}\).

Measurement of the number of neutrons knocked out of Be in the direction of the γ-rays and at an angle of 90° to them showed that in this case the distribution of neutrons over angles is approximately symmetric.

Attempts to split the nuclei of other elements by γ-quanta led in all cases to negative results.

In conclusion, we note that the study of the nuclear photoelectric effect may contribute to a considerable degree to the solution of the difficult question of the structure of the atomic nucleus.

L. Groshev, Moscow

LITERATURE

1. Uspekhi fizich. nauk, 14, no. 8, 953, 1934.
2. Szilard a. Chalmers, Nature, 134, 494, 1934.
3. Grosse a. Agruss, Phys. Rev., 47, 93, 1935; Ridenaur, Shinohara, Jost, Phys. Rev., 47, 318, 1935; Meitner, Naturwiss. 22, 759, 1934; Brasch, Lange, Szilard, a. others, Nature, 134, 880, 1934; Gentner, C.R. 199, 1211, 1934; 200, 311, 1935; Amaldi, D’Agostino, Fermi, Pontecorvo, Rasetti a. Segrè, Proc. Roy. Soc., 149, 522, 1935.
4. Chadwick a. Goldhaber. Proc. Roy. Soc., 151, 479, 1935.
5. Bethe a. Peierls, Proc. Roy. Soc., 148, 146, 1935.
6. Brasch, Lange, Szilard a. others, Nature, 134, 880, 1934.
7. Arzimowitsch u. Palibin, Phys. Z. Sowjet., 7, 245, 1935.