Artificial Transmutation of Elements by a Fast Proton Beam

The telegraph has brought news that, in Rutherford’s laboratory, two of his young collaborators, Cockcroft and Walton, have carried out the artificial transformation of certain light elements under the action of rapidly moving protons. This new major success was the culmination of attempts made in various laboratories and aimed at passing from the natural source of fast particles—radioactive preparations, used by Rutherford, as well as by Pettersson and Kirsch in their first works—to an artificial source controllable by the experimenter. Such a source is a stream of fast protons, carried in a discharge tube and accelerated by the voltage applied to the tube. It is easy to understand the enormous advantage of such a method. Quite apart from the possibility of controlling the source, which makes it possible to vary the conditions of the experiment over wide limits, such a source offers an enormous advantage with respect to intensity. Indeed, the number of $\alpha$-particles emitted per second in all directions by one gram of radium in equilibrium with the products of its decay is $3.7 \times 10^{10}$. Consequently, even the strongest sources of $\alpha$-particles used in actual experiments do not in any case give more than $10^9$ $\alpha$-particles. It is easy to see that, with a current of 1 microampere, the number of protons in the discharge tube carrying this current will be $3 \times 10^{12}$, i.e. in any case 1000 times greater than the number of $\alpha$-particles from the strongest of the radioactive preparations that have been used. If to this

one should also add the extremely important circumstance that the aforementioned number of protons is not scattered in all directions, but is carried within a certain small solid angle; this gives a further, and moreover very considerable, increase in the relative power of the source in comparison with radioactive preparations. Proceeding from these considerations, a fairly large number of investigators in almost all countries of the world tried to develop the appropriate experimental technique for the artificial transmutation of elements. In doing so, all the investigators started from the seemingly obvious premise that, in order to carry out artificial transmutation, it was necessary to accelerate the bombarding particles to enormous velocities, reckoned in any case in millions of volts. A considerable number of schemes were developed for conveniently obtaining ultra-high voltages; indeed, an attempt was even made to use for this purpose the high potentials borrowed from thunderstorm electricity; in other cases ingenious schemes were devised that made it possible to create the required proton velocities of the order of a million volts by means of very simple devices with primary voltage sources giving only a few thousand volts. A survey of these methods will be given in one of the forthcoming issues of Uspekhi.

However, positive results with the artificial transmutation of elements by means of protons accelerated in this way were first obtained in the same Cavendish Laboratory where, 13 years earlier, the very phenomenon of artificial transmutation of elements had been discovered. At the present moment (the end of May 1932) the details of the experiments have not yet been described. Their results, however, have been reported in two short communications by Cockcroft and Walton, printed in Nature, and also in the report of a discussion held at the Royal Society on April 28, 1932. Since these communications are very brief, we shall quote them in full.

The first communication, published on February 13 and dated February 2, is entitled: “The Artificial Production of Fast Protons”:

“In the Cavendish Laboratory a high-voltage laboratory has been organized for studying the properties of positive ions of high velocities. The potential from a high-voltage transformer is rectified and increased fourfold by means of a special arrangement of rectifiers and capacitors, giving a constant working potential of 800 kilovolts. In this way currents of the order of 1 milliampere are obtained at a potential whose constancy is maintained within 1—2%.

Protons from a discharge in hydrogen are directed along the axis of two glass cylinders 35 cm in diameter and 90 cm long and are accelerated by the constant potential of the rectifier. After this they pass into the experimental chamber at atmospheric pressure through a mica window with a retarding power equivalent to 1 millimeter of air. Under these conditions the luminescence of the air can easily be observed.

The ranges of protons in air and hydrogen were measured by means of a fluorescent screen as detector. The range in air, under normal conditions, of a proton having a velocity of \(10^9\) cm/sec was found to be 8.2 mm, whereas the corresponding range for hydrogen was 3.2 cm. The observed ranges on the whole support Blackett’s conclusion concerning the relative ranges of protons and \(\alpha\)-particles, although the absolute value of the ranges for both gases is less than the theoretical values. The ranges and the stopping power were measured more carefully by the ionization method.

The maximum energy of the protons obtained so far was 710 kilovolts at a velocity of \(1.16 \times 10^9\) cm/sec, and the corresponding range in air was 13.5 mm. We see no difficulties in operating our apparatus at potentials up to 800 kilovolts.”

The following communication, dated April 16, 1932, is entitled “The Disintegration of Lithium by Fast Protons.” The communication says:

“In the preceding communication we described a method of obtaining a steady stream of fast protons with energies up to 600 kilovolts by applying high potentials; we also described experiments on measuring the range of these protons

outside the tube. We used the very same method in order to investigate the effect occurring as a result of bombardment of a layer of lithium by a stream of these ions, the lithium being placed inside the tube at an angle of \(45^\circ\) to the beam. On the side of the tube there was a small window sealed with mica, with a stopping power equivalent to \(2\ \mathrm{cm}\) of air, and the emergence of radiation from the lithium was studied by the scintillation method outside the tube. The thickness of the mica window was considerably greater than that necessary to prevent scattered protons from passing out at the highest potentials used.

When the applied potential reached about 120 kilovolts, the appearance of a certain number of bright scintillations was suddenly observed; their number rapidly increased with the potential up to the highest potential employed, namely up to 400 kilovolts. Under these conditions several hundred scintillations per minute were observed when the proton current amounted to several microamperes. No scintillations were observed when the proton current was switched off, or when the lithium was screened from it by a metal shield. The range of the particles was measured by introducing mica screens into the path of the rays; it was found to be approximately \(8\ \mathrm{cm}\) in air and did not change noticeably when the potential was varied.

In order to clarify the nature of these particles, experiments were carried out with a Wilson–Shimizu chamber; tracks resembling the tracks of \(\alpha\)-particles were observed, and the ranges found from measurement of these tracks closely coincided with those found from the scintillations. It was estimated that at 250 kilovolts 1 particle is produced for approximately \(10^9\) protons.

The brightness of the scintillations and the thickness of the tracks observed in the Wilson chamber suggest that these particles are normal \(\alpha\)-particles. If this point of view proves correct, then it does not seem improbable that the isotope of lithium with mass 7 accidentally captures a proton and that the resulting nucleus with mass 8 disintegrates into two \(\alpha\)-particles, each of mass 4 and each with an energy of approximately eight million electron-volts. The release

energy, according to this view, is about 16 million electron-volts per act of disintegration, which agrees approximately with that calculated from the decrease in atomic mass occurring in this decomposition.

The experimental investigation of the phenomena obtained in other elements when bombarded by a beam of fast protons and other particles is at present “in progress.”

Some further information on the results of work in the Cavendish Laboratory in the same direction is contained in a brief report on the discussion on the topic “The Structure of Atomic Nuclei,” held on April 28 at the Royal Society. The experiments were continued with other elements, and it was found that beryllium, boron, carbon, possibly nitrogen, fluorine, and aluminum—all give particles with a characteristic range and with greater energy than the incident protons. Oxygen and copper gave no effect. The disintegration of boron and aluminum begins already at a proton energy of 150 kilovolts, whereas for the other elements more than 300 kilovolts is required in order to obtain a noticeable effect. It thus appears highly probable that a considerable majority of the light elements can be disintegrated by proton bombardment. There are as yet no definitive data on the nature of the products obtained in this way. Very probable assumptions may be made concerning them. Thus, for example, for fluorine and aluminum one may expect the following processes:

$$ \mathrm{F}^{19} + \mathrm{H} = \mathrm{He} + \mathrm{O}^{16} $$

$$ \mathrm{Al}^{27} + \mathrm{H} = \mathrm{He} + \mathrm{Mg}^{24}. $$

However, there are as yet no experimental data confirming these assumptions.

There is no doubt that the results set forth here, together with the discovery of neutrons, mark the beginning of a new era in research on the structure of the atomic nucleus. That is why we considered it necessary to acquaint the readers of Uspekhi with the still rather scant preliminary data that are now available.

E. Shpolsky.