Penetrating Radiation of Light Elements and Experimental Evidence for the Existence of Neutrons

E. V. Shpolsky, Moscow

In the very recent period our knowledge of the nature and structure of the atomic nucleus has been enriched by two very substantial discoveries: first, an isotope of hydrogen with mass 2 was discovered, i.e., the possibility of a close combination of two protons and one electron was proved[^1]; second, a whole series of facts was found, very convincingly testifying in favor of the real existence of particles that are a close combination of one proton and one electron and therefore have mass 1 and zero charge—the so-called neutrons.

Both discoveries were made in recent months, but both had been predicted with complete clarity already 12 years ago. It was precisely in his well-known Bakerian Lecture[^2], discussing the artificial disintegration of atoms that had just then been discovered, that Rutherford wrote: “It seems very probable that one electron can bind two H-nuclei, and possibly even one H-nucleus. If the first supposition is correct, then it indicates the possibility of the existence of an atom with mass about 2 and with one charge. Such a substance must be regarded as an isotope of hydrogen. The second supposition contains the idea of the possible existence of an atom with mass 1 and nuclear charge equal to zero. Such a formation appears entirely possible. From the modern point of view, a neutral atom of hydrogen should be regarded as a nucleus with a unit nuclear charge, with which is associated an electron situated at a certain distance from it, and the spectrum

hydrogen is attributed to the motion of this latter electron. However, under such conditions it seems probable that one electron may combine more closely with the H nucleus, forming something like a neutral doublet. Such an atom would possess quite fantastic properties. Its external field should practically be equal to zero, except in regions very close to the nucleus; as a consequence, it should have the ability to pass freely through matter. The existence of such an atom would probably be difficult to detect with a spectroscope, and it could not be held in a closed vessel. On the other hand, it should readily enter into the structure of an atom and either combine with its nucleus, or be destroyed by the intense field of the latter, giving rise to a charged H atom or an electron, or to both.”

The recently found experimental evidence for the existence of these remarkable particles is connected with the investigation of the nature and properties of the penetrating radiation of light elements—chiefly beryllium, discovered by Bothe and Becker in 1930. The two questions are so closely intertwined that it is impossible to set them out separately; moreover, the discovery of Bothe and Becker has not yet been covered in the pages of Uspekhi.

When light elements are bombarded with the α-rays of polonium, Bothe and Becker discovered that certain elements begin to emit an artificially excited penetrating radiation of the γ-ray type. The apparatus with which the authors made their observations is shown in Fig. 1. Here \(Z\) is an ordinary Geiger counter 5 cm in diameter, \(P\) is a polonium preparation deposited on a silver disk. The preparation had a strength of from 7 to 3 millicuries and was turned with its active side upward, i.e. turned away from the counter. The substance under investigation was placed directly on the preparation; in order that two substances could conveniently be studied one after the other, these substances made up two \(120^\circ\) sectors of the disk \(S\), which could easily be turned by means of the cord \(Sch\).

The third sector was left free so that it would be possible to investigate the null effect caused by the weak γ-radiation of polonium, unexpectedly discovered in these experiments.⁴ To protect against the surrounding radioactive radiation, the entire apparatus was surrounded by a lead armor 5 cm thick.

Testing a number of substances with the aid of this apparatus, Bothe and Becker found that some of them produced an increase in the number of deflections of the electrometer far exceeding the limits of error. From Table 1 it is clear that such substances are, beyond any doubt, lithium, beryllium, and boron. A considerably smaller, but apparently also qualitatively unquestionable effect is exhibited by fluorine (CaF₂), magnesium, and aluminum. Thus all these substances, under the influence of the α-rays of polonium, themselves begin to emit penetrating radiation, with the emission of beryllium being distinguished by particular intensity.

Fig. 1.

It further turned out that the new radiation is distinguished by a very strong penetrating power. Namely, according to the measurements of Bothe and Becker, 1 cm of lead weakens the intensity of the beryllium rays by only 30%, and of the boron rays by 53%. Although these measurements were of an approximate character and were quite imprecise, they definitely indicate that the new radiation belongs to the penetrating type, as a result of which Bothe and Becker quite naturally recognized it as artificially induced γ-rays.

The discovery of Bothe and Becker was soon confirmed by Irène Curie,⁵ Joliot, and somewhat later by Webster¹⁵ in the Cavendish Laboratory. It was shown that the elements Li, Be, B, F, Na, Mg, Al, when bombarded by the α-particles of polonium, begin to emit

TABLE I

strongly penetrating radiation, not subject to the action of electric and magnetic fields. A more exact measurement of the absorption coefficient of beryllium rays, carried out by Curie, showed that, in order to reduce their intensity by half, a lead screen must have a thickness of 47 mm. For comparison one may note that the most penetrating \(\gamma\)-rays of ThC'', after being first filtered through 20 mm of Pb, are weakened by half by a screen of 16 mm of lead.

The distinctive feature of Curie’s experiments was that she used an exceptionally strong polonium preparation (up to 100 millicuries) and, as a result, was able to pass from a Geiger counter to an ionization chamber connected to a sensitive Hoffmann electrometer. This, at first glance, not very substantial change in method, not prompted by any theoretical considerations whatever, nevertheless subsequently played a decisive role. Using their apparatus, Curie and Joliot showed that when beryllium rays pass through matter, entirely unexpected phenomena are observed.

of the phenomenon. Namely, it turned out that the ionization current in the chamber increases greatly when its aperture is closed with a substance containing hydrogen (paraffin, cellophane, water), whereas screens made of other substances (Al, Cu, Ag) produce no specific effect. At the same time, the rays that excite the additional ionization in the case of hydrogen-containing screens proved, in contrast to the beryllium rays themselves, to be very weakly penetrating: a sheet of aluminum \(0.2\) mm thick already absorbs them completely. On this basis, and also on the basis of a number of other indications, Curie and Joliot concluded that the additional ionization is caused by a stream of H-particles knocked out of the substance by the beryllium rays. This conclusion was very soon confirmed directly by observing the paths of the ionizing particles in a Wilson chamber, first by Curie and Joliot themselves,\(^8\) and then by Rasetti,\(^9\) who succeeded in photographing these tracks. In Fig. 2 we present one of Rasetti’s photographs. The tracks of \(\alpha\)-particles, visible in the photograph below, belong to a weak polonium preparation and serve to check the proper operation of the Wilson chamber. The polonium preparation, beryllium, and paraffin are located at the center of the chamber; the sharp long streak marked by the arrow undoubtedly belongs to an H-particle.

What is the origin of these fast ionizing particles? Curie and Joliot initially\(^7\) considered them to be recoil nuclei arising from the Compton scattering of quanta of the beryllium rays on H-nuclei. In that case, in order to impart to protons a velocity sufficient for them to have the ranges observed experimentally, the magnitude of the quantum of the beryllium rays would have to be \(50 \times 10^{6}\) volts. It is quite incomprehensible what processes, in the interaction of an \(\alpha\)-particle with a beryllium nucleus, could liberate such enormous quanta of radiation. The most “hard” quanta could be obtained upon capture of an \(\alpha\)-particle by a Be\(^9\) nucleus with the formation of a C\(^13\) nucleus. The mass defect of the latter is known sufficiently accurately, and from it one can calculate that the magnitude of the quantum of the beryllium rays is not

See article by E. V. Shpolsky

Fig. 2.

Fig. 3.

can exceed \(14 \times 10^6\). The question of the origin of the secondary protons observed when beryllium rays pass through was investigated in detail by Chadwick[^10] in Rutherford’s laboratory. Using a small ionization chamber connected to an oscillograph, Chadwick showed that not only hydrogen but also other light elements, when traversed by beryllium rays, give rapidly moving particles with great ionizing power. These particles, by all indications, are recoil nuclei of the corresponding elements. The path of one such recoil nucleus is visible, for example, in Rasetti’s photograph shown in Fig. 2, where the short track to the left of the proton marked by an arrow evidently belongs to a rapidly moving C nucleus knocked out of paraffin.

If one assumes that these recoil nuclei acquired their velocity in the Compton scattering of certain radiation quanta, then calculating the magnitude of these quanta from the ranges of the recoil nuclei of different elements gives completely contradictory results. Thus, the quantum of one and the same beryllium radiation, from the range of hydrogen nuclei, is obtained as \(50 \times 10^6\) V; from the range of nitrogen nuclei, \(100 \times 10^6\) V; and from the range of argon nuclei, \(150 \times 10^6\) V. The contradiction disappears, and the observed ratios of ranges agree well with the calculated ones, if it is assumed that beryllium radiation is not electromagnetic in character but is a stream of rapidly moving corpuscles, with charge 0 (enormous penetrating power and no influence of electric and magnetic fields!) and mass 1.

It is quite natural to suppose that an \(\alpha\)-particle, being captured by a \(\mathrm{Be}^9\) nucleus, emits a neutron and is transformed into \(\mathrm{C}^{12}\). In this case, from precise determinations of the atomic masses \(\mathrm{Be}^9\), \(\mathrm{He}^4\), \(\mathrm{C}^{12}\) according to Aston’s curve, one can find the mass defect, and from it the velocity of the emitted neutrons. The resulting velocity is equal to \(3 \times 10^9\) cm/sec, i.e. is of the order of magnitude of the velocity of fast \(\alpha\)-particles.

The correctness of this hypothesis of Chadwick was proved by Chadwick himself and by other collaborators of the Cavendish—

of the laboratory, with a long series of very interesting observations, of which we shall cite only the most striking.

1. It may be expected that, in passing through matter, a neutron collides not only with the nuclei of atoms, but sometimes also with electrons. If we apply to this collision the ordinary laws of mechanics, i.e. the law of conservation of momentum and the law of conservation of energy, then for the velocity of the electron set in motion we easily obtain:

\[ u = 2v \frac{m}{m+\mu}\cos\Theta, \]

where \(v\) is the velocity of the incident neutron, and \(m\) and \(\mu\) are the masses of the neutron and the electron. Since \(m \gg \mu\), in the case of a central impact—and only such cases can be at issue here, owing to the absence of an external field for the neutron—we obtain that the velocity of the electron is equal to twice the velocity of the neutron. Since the latter is of the order of magnitude of the velocity of \(\alpha\)-particles, this case of collision of a neutron with an electron is entirely analogous to the occurrence of the so-called \(\delta\)-rays, i.e. slow electrons of small range, formed when \(\alpha\)-particles pass through a gas.* Such short trajectories of electrons were discovered by Dee in Wilson photographs during the passage

* It is not difficult to convince oneself that these must indeed be slow electrons. According to what has been said in the text, the velocity of the electrons formed in collision with a neutron must be \(6.6 \times 10^9\) cm/sec. If the accelerating potential \(V\) is expressed in volts, then between \(v\) in cm/sec and \(V\) there is the relation:

\[ v^2 = 2 \cdot V \cdot 10^8 \cdot \frac{e}{\mu} = 2V \cdot 10^8 \cdot 1.76 \cdot 10^7 = 3.52 \cdot 10^{15}V . \]

Hence, for the velocity \(v = 3 \cdot 10^9\), the corresponding potential will be only

\[ V = \frac{36 \cdot 10^{18}}{3.52 \cdot 10^{15}} = 11\,000 \text{ volts}. \]

From the work of Williams and Nuttall (Phil. Mag. 2, 1109, 1926) it is known that electrons with such velocities must have a range in air of about \(3.4\) mm.

beryllium rays in moist gases of various kinds, and it is very difficult to devise for them any other explanation than the one indicated above.

If the beryllium rays had a quantum nature, then the magnitude of their quantum, as we have seen, would have to be no less than \(50 \times 10^6\) volts. The Compton electrons arising under the action of such enormous quanta would have very long tracks. In reality, such long-range electrons are indeed observed, but very rarely—considerably more rarely than short-range electrons. This fact nevertheless indicates that the beryllium rays have a nonuniform structure. We shall return to this question again at the end of the article.

1. The study of Wilson photographs during the passage of beryllium rays through various gases provides further interesting evidence for the existence of neutrons. Since the neutrons themselves have very small ionizing power, their paths are not visible in Wilson photographs; instead, the paths of those recoil atoms are visible which arise in a central collision of a neutron with the corresponding nucleus. In this process, before such a favorable collision the neutron may have undergone a number of other, less favorable collisions, which, however, changed the direction of flight of the neutrons. As a result, the direction of the recoil nuclei may be in no way connected with the position of the neutron source. This can be seen very well in Fig. 3 (Rasetti’s photograph), which is a Wilson photograph during the passage of beryllium rays through helium (the arrangement of the source is the same as in Fig. 2), where the arrow marks a helium nucleus set in motion under the influence of a neutron impact—a kind of artificially produced \(\alpha\)-particle.

Feather obtained analogous clear photographs in nitrogen. From a comparison of the ranges of the nitrogen recoil nuclei thus detected with the ranges of recoil protons obtained during the passage of neutrons through hydrogen, Chadwick was even able to calculate the mass and velocity of the neutrons.

In fact,

\[ u_{\mathrm H}=\frac{2m}{m+1}\,v, \]

\[ u_{\mathrm N}=\frac{2m}{m+14}\,v. \]

Hence, substituting for \(u_{\mathrm H}\) and \(u_{\mathrm N}\) the experimental values from the corresponding ranges, Chadwick obtained for \(m\) a value somewhat greater than unity, and for \(v\)—about \(3\times10^9\) cm/sec.

Chadwick obtained a more exact value for the neutron mass from consideration of the production of neutrons from boron. In this case the process apparently proceeds according to the equation:

\[ \mathrm B^{11}+\mathrm{He}^4\to \mathrm N^{14}+\mathrm n^{11}, \]

where \(\mathrm n^{11}\) denotes the neutron. The exact values of the masses of boron, helium, and nitrogen are known directly (for beryllium the mass has to be interpolated from Aston’s curve) from Aston’s measurements. If one also takes into account the kinetic energy of the \(\alpha\)-particle, then from considerations based on the energy balance one may conclude that the neutron must have a mass lying between 1.005 and 1.007. Since the mass of a free atom of hydrogen is 1.0078, on the basis of Einstein’s relation one can easily calculate that the binding energy of the electron with the proton in the neutron is about one million volts. This exceedingly interesting result shows that the neutron, like the \(\alpha\)-particle, is a very stable formation, apparently playing a large role in the structure of the nuclei of heavier elements.

The totality of the evidence listed, as well as of other evidence not given here, testifies with complete conclusiveness in favor of the reality of the existence of neutrons.

Observations on the artificial transformation of elements under the influence of neutrons are also very interesting. It is curious to note first of all that here we are dealing with a completely peculiar case of artificial trans-

penetrating radiation of light elements

transformation of elements, for if, in a collision with a nucleus, a neutron is captured by it and a proton is then ejected, the resulting nucleus will have the same mass as the original one, but an atomic number smaller by one; thus, for example, the nitrogen nucleus \(N^{14}\) must be transformed into the carbon isotope \(C^{14}\), etc. The peculiarity of the neutron is that its external field is practically equal to zero. Therefore a neutron can exert any effect only in a central collision, but in that case it has a high probability of penetrating into the nucleus and, consequently, of causing its disintegration. This explains the fact that the probability of artificial transformation under the action of neutrons is considerably higher than under the action of \(\alpha\)-particles. And indeed, of 180 Wilson “tracks” obtained by Feather in nitrogen, 30 were caused by artificial disintegration.

In conclusion we shall return once more to the question of the nature of the penetrating radiation of light elements. That this radiation, in large part, is a stream of neutrons, according to what has been said, can at present hardly be doubted. However, we have already mentioned in passing that the appearance, on some photographs, of long electron paths indicates that these rays are inhomogeneous and, along with corpuscular components, also consist of quantum radiation of the \(\gamma\)-ray type. Rasetti\(^9\) came to the very same conclusion on the basis of his observations of coincident deflections of two Geiger counters. Finally, Bothe and Becker\(^ {17}\) indicated that the counter generally records only \(\gamma\)-radiation, so that, if Curie and Joliot had not switched to the ionization chamber, neutrons would hardly have been discovered.

From the theoretical point of view, the simultaneous occurrence of corpuscular and quantum radiation when an \(\alpha\)-particle acts on a nucleus presents no special difficulties. A certain analogy here may be provided by the occurrence of X-rays, which begins with the ejection of an electron from one of the inner levels.

There is no doubt that the study of the properties of neutrons will enrich us with important information about the nature and structure of the atomic nucleus. And although the possibility of such a close combination of an electron with a proton appears enigmatic from the standpoint of present-day quantum mechanics—which has so brilliantly justified itself insofar as the electron shell of the atom is concerned—the very fact of the real existence of neutrons will undoubtedly serve theoretical physics as a powerful stimulus toward the development of a relativistic form of quantum mechanics and electrodynamics, to which intranuclear processes must be subject.

References

1. Cf. G. S. Landsberg, Advances in the Physical Sciences, XII.

2. E. Rutherford, Proc. Roy. Soc. A, 97, 374, 1920. Russian translation in the collection: E. Rutherford, The Structure of the Atom and the Artificial Disintegration of Elements. Prepared for publication by E. V. Shpolsky, Moscow—Leningrad, 1923.

3. W. Bothe und H. Becker, Z. Physik 66, 282, 1930; Naturwiss. 19, 753, 1931.

4. W. Bothe und H. Becker, Z. Physik 66, 307, 1930.

5. Irene Curie, C. r. 193, 1412, 1931.

6. See Handb. d. Physik von Geiger und Scheel Bd. XVI, S. 245.

7. I. Curie et F. Joliot, C. r. 194, 273, 1932.

8. I. Curie et F. Joliot, C. r. 194, 708, 1932.

9. F. Rasetti, Naturwiss. 20, 252, 1932 (Heft 14, 1 April 1932).

10. J. Chadwick, Nature, 129, 212, 1932.

11. E. Rutherford, Nature 129, 457, 1932 (of 26 March 1932).

12. Cf. Gamow’s article in the next issue of Advances in the Physical Sciences.

13. Cf., for example, E. Rutherford, J. Chadwick and C. Ellis, Radiations from Radioactive Substances, Cambridge 1930, p. 148.

14. Cf. especially ⁹.

15. Webster, Proc. Roy. Soc. A 136, 428, 1932.

16. C. D. Ellis, Natur, 1932, p. 674—676.

17. Bothe und Becker, Naturwiss.