Radioactivity and the Structure of the Atom

L. Meitner. Radioaktivität und Atomkonstitution.
Zeitschr. für Physik, 5, p. 124, 1921. Die Naturwissenschaften 9, p. 423, 1921.

The author’s main ideas amount to the following: as is known, the atomic weight of all elements can be expressed by the general formula \(4n + p\), where \(n\) is an integer and \(p = 0, 1, 2, 3\). If \(p = 0\), i.e., an element of the type \(4n\), then its nucleus is built only of helium nuclei; if \(p = 1, 2,\) or \(3\), then, in addition to \(n\) helium nuclei, the atomic nucleus of the given element also contains, respectively, 1, 2, or 3 hydrogen nuclei. Let the atomic number of the element be \(Z\). If this \(Z\) is even, then the element will be built of \(\frac{Z}{2}\) helium nuclei with free charge, \(n - \frac{Z}{2}\) helium nuclei with charge compensated by \(2\left(n - \frac{Z}{2}\right)\) electrons and, generally speaking, \(p\) hydrogen nuclei whose charge is compensated by \(p\) electrons. In the case of odd \(Z\), there is also one extra electron or an \(H\) nucleus with free charge. From radioactive data it follows, with great plausibility, that the former is much more probable than the latter. The following notation is then introduced: a helium nucleus with free charges is denoted by the symbol \(\alpha\), a helium nucleus with compen-

with a screened charge, an \(a'\) electron; the electron compensating the charge \(a'\) is denoted by \(\beta\). Thus the uranium nucleus has the following composition \((Z=92)\):

\[ 46a + 13(a' + 2\beta) + 2H + 2e, \]

where \(e\) is the electron neutralizing the charge \(H\).

If such assumptions are made, then the sequence of transformations in radioactive series becomes to a considerable degree comprehensible. Indeed, it appears that equilibrium is disturbed in that part of the nucleus which is built of free helium nuclei. Then, with great probability, a series of \(\alpha\)-transformations should follow one after another: \(\alpha-\alpha-\alpha-\cdots\). When equilibrium is disturbed in the part consisting of the formations \(a' + 2\beta\), two cases are possible: 1) first an \(a'\)-particle is emitted; in this process two \(\beta\)-particles are liberated, and the transformation must proceed according to the scheme \(a' - \beta - \beta\); 2) first one of the \(\beta\)-particles is emitted; then the second \(\beta\)-particle and the \(a'\)-particle remain free. Therefore the subsequent transformation may consist in the emission of either the one or the other particle; the radioactive series may undergo a branching, which then closes again according to the scheme

\[ \beta - \begin{matrix} & a' & \beta \\ \beta & \diamond & a' \end{matrix} \]

It is possible, however, also that equilibrium is disturbed both in the electrically neutral and in the charged part of the nucleus. In this case a branching must occur which does not close again, but leads to the formation of a new branch of transformations. The branching schemes here will be

\[ \begin{matrix} & a \\ a' & \lt \end{matrix} \quad\text{or}\quad \begin{matrix} & a \\ \beta & \lt \end{matrix} \quad \text{(cf. the formation of the actinium series).} \]

All the proposed schemes of transformations are excellently confirmed when the three radioactive series are considered. For example, let us write out the uranium–radium–actinium family:

\[ \mathrm{U}\,\mathrm{UX}_1\,\mathrm{UX}_2\,\mathrm{UII}\,\mathrm{Io}\,\mathrm{Ra}\,\mathrm{Em}\,\mathrm{RaA}\,\mathrm{RaB}\,\mathrm{RaC} \begin{matrix} & \mathrm{RaC'} & a' \\ \alpha & \lt & \beta \\ & \mathrm{RaC''} & \end{matrix} \mathrm{RaD}\,\mathrm{RaE}\,\mathrm{RaF}\,\mathrm{RaG}\,\mathrm{Pb} \]

\[ \downarrow \]

\[ \mathrm{UY}\,\mathrm{Pa}\,\mathrm{Ac}\,\mathrm{RaAc}\,\mathrm{AcX}\,\mathrm{Em}\,\mathrm{AcA}\,\mathrm{AcB}\,\mathrm{AcC} \begin{matrix} & \mathrm{AcC'} & a' \\ \alpha & \lt & \beta \\ & \mathrm{AcC''} & \end{matrix} \mathrm{Pb}. \]

Here only the branch \(UY\) and the subsequent transformations up to actinium proceed according to the unusual scheme

\[ \begin{matrix} & \beta \\ & \lt \\ \alpha' - \beta - \alpha - \beta \end{matrix} \]

However, the place of the \(UY\) branching has not yet been established with complete certainty. If \(UY\) arose not from \(\mathrm{U}_2\), but from \(\mathrm{U}_1\), then the transformation would proceed according to the scheme

\[ \begin{matrix} \alpha' \\ \lt \\ \alpha - \beta - \alpha' - \beta - \alpha - \alpha - \cdots \end{matrix} \]

in full agreement with the ideas of L. Meitner.

The thorium series gives no deviations from the proposed schemes.

E. Shpolsky.