On the Possible Discovery of an Element with Atomic Number Greater than 92*

E. Fermi, Rome

Until recently it was usually assumed that an atom obtained by artificial disintegration of elements must be a stable isotope. However, I. Curie and F. Joliot found data refuting this assumption; they showed that in certain cases the atom produced may prove to be radioactive, i.e., to have a measurable half-life and to pass into a stable form only after the emission of a positron.

The number of elements that can be activated by impact of an α-particle (Joliot), a proton (Cockcroft, Gilbert, Walton), or a deuton (Crane, Lauritsen, Henderson, Livingston, Lawrence) is inevitably limited by the fact that, owing to the presence of Coulomb repulsive forces, only the lightest of the elements can be disintegrated.

This limitation is invalid in the case when neutrons are used as the bombarding particles. The high effectiveness of these particles in regard to the disintegrations they produce fully compensates for the weakness of the neutron source in comparison with sources of α-particles or protons. Indeed, as the corresponding experiments show**, a large number of elements (47 out of the 68 investigated so far) of any atomic weight can be activated with the aid of even such a simple neutron source as a glass tube filled with beryllium powder and radon (about 800 millicuries). The yield of such a source is approximately 1 million neutrons per second.

All elements activated in this way have shown in their radiation only negative electrons. Theoretically this is quite understandable, since capture of the bombarding neutron by the nucleus creates within it an excess in the number of neutrons, and a stable state can be reached only through the transformation

* Nature, 1934, translated by A. A. Ilyina.
** E. Fermi, Ricerca Scientifica 1, 5, 283, 6, 33; Nature 113, 757, 1934. E. Amaldi, O. D’Agostino, E. Fermi, F. Rasetti, E. Segrè, Ricerca Scientifica 8, 452, 1934.

neutron into a proton, which is inevitably connected with the emission of a β-particle.

In a few cases it proved possible to carry out the chemical separation of the β-active element, following the usual technique of adding small quantities of related elements to the irradiated substance. These elements are then separated chemically and tested separately for β-activity by means of a Geiger–Müller counter. The activity always accompanies a definite element, with which, in this way, the active element can be identified. In three cases (aluminium, chlorine, cobalt) the active element formed by bombardment of an element of atomic number \(Z\) has atomic number \(Z - 2\). In four cases (phosphorus, sulphur, iron, zinc) the atomic number of the active product proved to be \(Z - 1\), and, finally, in two cases (bromine, iodine) the active element proved to be an isotope of the bombarded element.

These data apparently indicate that three processes are possible: a) capture of a neutron with the instantaneous emission of an \(\alpha\)-particle, b) capture of a neutron with the emission of a proton, c) capture of a neutron with the emission of a \(\gamma\)-quantum, freeing the nucleus from its excess energy.

From the theoretical point of view, the probability of processes “a” and “b” depends mainly on the energy emitted by the \(\alpha\)- or H-particles and to an still greater degree on the atomic weight of the element. The probability of process “c,” given the present state of nuclear theory, can be estimated only very roughly.

It seems expedient to devote special attention to the study of the heavy radioactive elements thorium and uranium, since the greatest instability of these heavy nuclei may initiate a whole chain of subsequent transformations. Proceeding from this, the author, in collaboration with F. Rasetti and O. D’Agostino, undertook a study of the heavy elements.

Experiment showed that both elements, preliminarily purified of the usual active impurities, can be strongly activated by neutron impacts. The initial induced activity in our experiments amounted to 1,000 counts per minute in a Geiger–Müller counter made of aluminium foil \(0.1\) mm thick. The decay curves of this activity indicate the complexity of the process taking place. A rough analysis of the activity curves of thorium reveals in this element at least two periods.

Uranium was studied best of all. Here periods of approximately 10, 40 sec., 13 min. and, finally, two periods from 40 min. to one day were firmly established. Owing to the considerable uncertainty of the decay curves, caused by statistical fluctuations, it is very difficult to establish whether these periods correspond to successive or alternative decay processes.

Attempts were made to identify chemically the β-active element with a period of 13 min. The essence of these investigations consists in

On the Possible Discovery of an Element with Atomic Number 83

upon adding to the irradiated substance (uranyl nitrate in a concentrated solution, purified from decay products) such an amount of an ordinary β-active element as would give about one hundred counts per minute in the counter. If it were possible to prove that the induced activity, which can be recognized by its characteristic period, can be chemically separated from the added activity, then it would be quite permissible to consider that these activities do not originate from isotopes.

Below is a description of a reaction that makes it possible to identify the 13-minute product from among the heaviest elements. The irradiated uranium solution is diluted with 50% nitric acid; then a small amount of a manganese salt is added to it, which, by the addition of NaCl, is precipitated from the boiling solution in the form of dioxide ($\mathrm{MnO_2}$). The precipitate of manganese dioxide obtained in this way carries with it a significant percentage of the activity.

This reaction proves first of all that the 13-minute activity is not an isotope of uranium. In order to exclude the possibility that this activity originates from element 90 (thorium) or 91 (palladium), we repeated this reaction at least 10 times with the addition of a certain amount of uranium $X_1 + X_2$, giving about 2,000 counts per minute. In addition, some cerium and lanthanum were added.

Under these conditions the manganese reaction gave only the 13-minute activity. No traces of the 2,000 counts of uranium $X_1$ (period 24 days) were found in the precipitate. The same was true for uranium $X_2$ (period 75 sec.), although all operations were carried out in less than 2 min. after the precipitation of manganese dioxide, so that the several hundred counts given by uranium $X_2$ would have been easily distinguishable.

Similar negative results were obtained in controls for radium (88) and actinium (89). For this control mesothorium 1 and 2 were used, and barium and lanthanum were added. The results, as in the preceding cases, were definitely negative. The final precipitation of uranium $X_1$ and mesothorium 1, which gave no β-rays penetrating enough to be detected by our counter, should have been revealed by the subsequent formation, respectively, of uranium $X_2$ and mesothorium 2.

Finally, we added to the irradiated uranium solution a certain amount of inactive lead or bismuth and proved that the conditions of the manganese dioxide reaction can be chosen in such a way that precipitation of manganese dioxide with 13-minute activity is obtained without the slightest traces of lead or bismuth in the precipitate.

It seems to us that, by means of all these operations, we have excluded any possibility that the 13-minute active substance under investigation is an isotope of uranium (92), protactinium (91), thorium (90), actinium (89), radium (88), bismuth (83), or lead (82). Its behavior also excludes eka-caesium (87) and emanation (86).

These results, which rule out the identity of the 13-minute active substance with a large number of heavy elements, suggest the possibility that the atomic number of the element found may be greater than 92.

If this is indeed an element with atomic number 93, then it should be a chemical homologue of manganese and rhenium. This supposition is also supported to some extent by the observation that the 13-minute activity was also detected in the precipitate of rhenium sulfide, which is insoluble in hydrochloric acid. However, owing to the fact that rare elements are also readily precipitated by this method, these data cannot be regarded as entirely reliable.

The possibility of atomic numbers 94 and 95 is difficult to distinguish from the preceding one, since the chemical properties of such elements would probably be quite similar to those of element 93.

Valuable information about these intricate processes may be obtained by investigating the possible emission of heavy particles. This has not yet been done, since, for their observation, the active product must be in the form of a very thin layer.

From all that has been set forth above it follows that at present it is still too early to formulate definite hypotheses about the intricate decay processes of artificially produced heavy atoms.