The decimal ratio \(^{4}\mathrm{He}:{}^{16}\mathrm{O}\) differs from the previously accepted value by \(3/10000\). The fifth column of Table 1 gives the masses of the elements obtained with this correction taken into account. As the table shows, Bethe’s data agree fairly well with the data of Rutherford and others.

Confirmation of the correctness of the introduced corrections is provided by the fact that the energies of nuclear reactions, calculated from the new, corrected masses, are in better agreement with the experimental data than are the energies calculated from mass-spectrographic data. This can be seen from Table 2.

Recently Aston\(^{4}\) measured the masses of several elements by means of a new method. The preliminary data from these measurements are given in the sixth column of Table 1. Their accuracy is no better than 1 in 10,000. As is clear from the table, there is a large difference between the new and old Aston data, and precisely in the direction that follows from the data of Oliphant, Kempton, and Rutherford.

In conclusion, let us note that the introduction of corrections to the masses of the elements resolves the question of the instability of the beryllium nucleus. According to mass-spectrographic data, beryllium is unstable, since its mass \((9.0155)\) is greater than the mass of two \(\alpha\)-particles and a neutron \((8.0043 + 1.0080 = 9.0123)\). However, experimental investigations do not confirm the instability of beryllium. According to the corrected data, beryllium is a stable element—its mass \((9.0135)\) is less than the mass of two \(\alpha\)-particles and a neutron \((8.0068 + 1.0083 = 9.0151)\).

L. Groshev

LITERATURE

1. Oliphant; Kempton, Rutherford, Proc. Roy. Soc., 149, 406, 1935.
2. Oliphant, Kempton, Rutherford, Proc. Roy. Soc., 150, 241, 1935.
3. Bethe, Phys. Rev. 47, 633, 1935.
4. Aston, Nature, 135, 541, 1935.

\(\gamma\)-RADIATION PRODUCED BY THE ACTION OF NEUTRONS ON MATTER

In studying the scattering of neutrons by various substances, Li\(^{1}\) established that, when paraffin or liquid hydrogen is irradiated with neutrons, \(\gamma\)-radiation with an energy of several million volts arises in them. These observations were then confirmed by Fleischmann\(^{2}\). In his last work Li investigated this effect in greater detail; it was established that the production of \(\gamma\)-rays under the action of neutrons on matter is also observed for a number of other elements. Although the accuracy of the measurements carried out is low, which is explained by the smallness of the effect itself, nevertheless the work mentioned deserves special attention in view of the importance of the question under consideration.

In essence, Li’s experiments consist of the following. Above the article, a thick-walled ionization chamber with high pressure contains a source of neutrons (usually polonium + beryllium, 10–15 millicuries). Under these conditions, in the chamber there is observed a certain current caused by \(\gamma\)-rays and neutrons coming directly from the source. Then a layer of the substance under investigation (scatterer) is placed over the source, and the current in the chamber is measured again. The current in the second case proves to be larger. The additional ionization obtained in the chamber when the scatterer is placed over the source is usually very small and amounts to 2–3% of the ionization observed in the absence of the scatterer. This addit—

ionization is caused partly by neutrons scattered through large angles, and partly by $\gamma$-rays arising when neutrons act on the substance of the scatterer. To determine the ionization caused separately by neutrons and by $\gamma$-rays, measurements are made with two ionization chambers, one of which is filled with argon at a pressure of 90 atm, the other with hydrogen at 60 atm. The first chamber is more sensitive to $\gamma$-rays, while the second is more sensitive to neutrons. Making certain assumptions about the effectiveness of the action of $\gamma$-rays and neutrons in either chamber, from the results of measurements in both chambers one can determine what part of the additional ionization belongs to neutrons and what part to $\gamma$-rays.

Control experiments carried out with a radiothorium preparation giving only $\gamma$-rays (energy 2.65 MeV) showed the absence of additional ionization upon insertion of the scatterer. Hence it may be concluded that the part of the additional ionization produced by $\gamma$-rays is due to the action on the substance of the scatterer of neutrons, and not of $\gamma$-rays. This is further confirmed by the fact that for a boron + polonium source, emitting $\gamma$-rays (3 MeV) and neutrons, additional ionization is observed; at the same time, if it is calculated per 1 neutron from boron, it amounts to approximately $2/3$ of the ionization calculated per 1 neutron from beryllium, which can be explained by the lower velocities of neutrons from boron.

The results obtained for the case of hydrogen* show that the additional ionization produced in the chamber in this case is created exclusively by $\gamma$-rays, as should be expected, since in single collisions of neutrons with protons the former cannot be deflected from their initial direction by angles greater than 90°.

In the following table Li gives the results obtained for various elements. The values given are expressed in arbitrary units for the additional ionization caused in the chamber only by $\gamma$-rays. They are calculated per 1 atom of scatterer.

When the dependence of the effect on the atomic number is represented graphically, a smooth curve increasing more slowly than $Z$ is obtained. At the same time, it is striking that in the case of hydrogen the effect is much larger than would follow from the course of this curve.

* In these experiments paraffin served as the scatterer. The action of the carbon in the paraffin was taken into account by means of additional experiments with a graphite scatterer.

If it is assumed that the γ-radiation under consideration is isotropic and does not differ very greatly in its energy for different elements, then the numbers given above will be proportional to the effective cross section (calculated per nucleus) for the process of formation of a γ-quantum under the action of neutrons on matter. In the case of lead it was shown experimentally (measurements in the intervals 0–90° and 90–180°) that the γ-radiation arising in it is distributed approximately isotropically. The energy of the γ-radiation was determined from absorption curves in lead for the cases Pb, Fe, H. However, in view of the small number of points taken for constructing the absorption curves, and the great inaccuracy of the measurements, the values obtained for the energies should be regarded only as approximate. These measurements showed that in the case of hydrogen the γ-rays have an energy equal to 3–4 MeV, and in the case of iron and lead—approximately 1.5 MeV.

In explaining the origin of γ-rays under the action of neutrons, let us dwell first on complex nuclei (all the elements investigated, with the exception of H). As is known, when neutrons pass through matter they interact practically only with the nuclei of atoms; in doing so there occurs either capture of the neutron by the nucleus, or its scattering (elastic or inelastic). In some cases, as a result of capture, a radioactive element is formed, decaying sometimes with emission of γ-radiation^4. One might try to identify the radiation under investigation with this radiation. Control experiments, however, show that under the conditions of the experiment artificial radioactivity practically plays no role.

From the work of Fermi and his collaborators^4 it is also known that in a number of cases γ-radiation arises in the process of formation of the radioactive atoms themselves. However, this γ-radiation too cannot be identified with the radiation under investigation, because the probability of formation of radioactive atoms upon capture of a neutron by a nucleus, and consequently also the intensity of the corresponding γ-radiation, changes from element to element by sharp jumps, whereas the γ-radiation under investigation increases smoothly with increasing atomic number of the elements.

In order to determine whether the appearance of γ-rays is connected with absorption of neutrons or not, Li determined the coefficients of absorption and scattering of neutrons and calculated from them the effective cross sections for these processes. Comparison of these effective cross sections with the effective cross section for the process of appearance of γ-rays from neutrons shows that the appearance of this radiation is not connected with absorption of neutrons. Therefore Li ultimately comes to the conclusion that, for the complex elements considered, the effect under investigation is explained by inelastic scattering of neutrons, as a result of which the nucleus is excited and then, passing into the normal state, emits a γ-quantum.

Li’s effect for hydrogen is of exceptionally great interest, since in this case there is obtained a sharp contradiction with other experimental data. It is quite clear that the explanation given for the case of complex nuclei is inapplicable here, unless one assumes that in the collision of a neutron with a nucleus it is the neutron that is excited, and not the nucleus. In his first note on the formation of γ-rays when paraffin or liquid hydrogen is irradiated with neutrons, Li explained this effect in the following way. When a neutron collides with a proton, a deuteron is formed; in this process its binding energy together with half the kinetic energy of the deuteron (this follows readily from the conservation laws) is emitted in the form of a γ-quantum. The deuteron thus formed must fly in a direction close to the direction of the incident neutron. It was expected that, in this circumstance, those tracks which rather often arise in a Wilson chamber filled with hydrogen, if it is irradiated with neutrons, would be deuteron tracks. However, his experiments showed that these tracks do not have a preferential direction along the path of the incident neutrons. Measurement of the curvature of these tracks in a magnetic field showed that they are caused by protons. We note that the absence of corresponding deuteron tracks does not yet prove conclusively

against Li’s explanation, since it is unknown what energy neutrons must have in order to form a deuteron with a proton. It may be that in this formation the deuteron receives so little kinetic energy that it cannot be detected in a Wilson chamber. However, if the origin of the $\gamma$ radiation is explained by the formation of deuterons, then Li’s data are in sharp contradiction with other experimental data. The point is that Chadwick and Goldhaber observed an effect inverse to that discovered by Li, namely, the splitting of deuterons by $\gamma$ rays. The effective cross section for such a process was found to be equal to $6 \cdot 10^{-28}\ \text{cm}^2$. On the other hand, general thermodynamic and statistical considerations show$^{6,7}$ that the formation of a deuteron from a neutron and a proton with the emission of a $\gamma$ quantum is a considerably less probable process than its splitting by a $\gamma$ quantum. In reality, however, there is an opposite relationship between the data of Li and of Chadwick and Goldhaber. The cause of this contradiction, indicated above, cannot at present be regarded as possible. Therefore the question of the mechanism of the formation of $\gamma$ rays when hydrogen is irradiated with neutrons remains as yet unresolved.

L. Groshev

LITERATURE

1. Lea, Nature, 133, 24, 1934.
2. Fleischmann, Naturwiss. 22, 839, 1934.
3. Lea, Proc. Roy. Soc., 150, 627, 1936.
4. Fermi, Amaldi, D’Agostino, Pontecorvo, Rasetti, Segrè, Proc. Roy. Soc. 149, 522, 1935; see also Advances in Physical Sciences No. 7, 1935.
5. Auger, C. R., 198, 365, 1934.
6. Chadwick and Goldhaber, Nature, 134, 237, 1934; Advances in Physical Sciences, 14, 953, 1934.
7. Bethe and Peierls, Proc. Roy. Soc., 148, 146, 1925.