NEW DATA ON THE ABUNDANCE OF ATOMIC NUCLEI*

The author of the paper under review has obtained a new detailed table of the relative abundance of atomic nuclei in nature, on the basis of the most reliable data on the composition of the Earth, the Sun, stars, and meteorites.

A similar table was compiled in 1937 by Goldschmidt. Until the present time it has been considered the most reliable, but—

* H. Brown, Rev. of Mod. Phys. 21, 625 (1949).

the author's work introduces significant corrections into Goldschmidt’s data, although on the average there remains a sharp drop in the curve of relative abundance of elements up to nuclei with mass number about 100, subsequently passing on the average into a smooth decrease of this curve.

On the path toward obtaining the relative abundance of atoms from data on stellar spectra there are two principal difficulties:

1. The difficulty of determining the abundance of atoms from the intensities of spectral lines.

2. The absence of certainty that the abundance of atomic nuclei is the same both at the surface and in the inner regions of a star. For example, in the depths of the Sun, where the temperature is of the order of 20 million degrees, such elements as lithium and boron cannot exist, since they have very short (Li \(\sim 1\) min.) lifetimes in thermonuclear reactions with fast protons, which constitute the main mass of the Sun.

Astrophysical data, although they require refinement, indicate the presence of lithium and boron in the solar atmosphere, and consequently must reflect the density gradient for these elements as one goes deeper into the interior of the Sun.

The changes that the author introduced into Goldschmidt’s tables were made on the basis of the following data:

1. The latest and most accurately determined stellar data were used.

2. The abundance of elements in stars was brought into agreement with their abundance in meteorites.

3. Modern information on the isotopic composition of the elements on Earth was used.

4. The metallic and silicate components in meteorites were estimated.

5. A statistical estimate was used for the deviations of the abundance of elements in meteorites.

6. The composition of metallic meteorites, consisting mainly of iron and nickel, and the composition of stony meteorites were compared with the composition of the Earth’s crust and with the density of the Earth’s inner regions obtained from seismic data.

Observations of discontinuities in the propagation of seismic waves give the boundary between the Earth’s crust and the inner part of the Earth—the core, consisting of iron and nickel. On this basis one can estimate the ratio:

\[ \frac{\text{weight of the core}}{\text{weight of the crust}} \simeq 0.5 . \]

The Earth’s crust consists to 10% of the metallic phase; therefore on Earth the ratio of the metallic phase to the silicate phase, of which the crust mainly consists, is equal to 0.67.

An analogous ratio was obtained for the composition of meteorites; the ratio of the number of iron atoms to the number of silicon atoms proved to be \(N_{\mathrm{Fe}}/N_{\mathrm{Si}} = 1.7\). According to Unsöld’s data, the same ratio for the composition of the Sun gives \(N_{\mathrm{Fe}}/N_{\mathrm{Si}} = 2.7\). These numbers may be considered to be in good agreement, taking into account the difficulty of obtaining stellar data.

In compiling the table, the greater strength of iron meteorites as compared with stony ones during interaction with the Earth’s atmosphere at the time of fall was taken into account. The oxidation of meteorites and the difficulties of detecting stony ones were taken into account.

of meteorites. The ratio found between the metallic and silicate phases makes it possible to reduce the inaccuracies arising from this.

In the author’s tables, wherever possible, data obtained from the composition of meteorites were used. Solar and stellar data were used for those elements for which meteoritic data are unreliable: H, He, C, N, O, Ne, A, and also, in doubtful cases, when meteoritic data are insufficiently accurate: F, S, Cl, B. In the work, very detailed tables of the relative abundance of atomic nuclei were obtained. In Fig. 1 a graph is given, constructed on the basis of the tables, for nuclei of odd atomic number.

Fig. 1.

Of exceptional interest is the relation obtained in the work between the abundances of elements in meteorites and in the solar atmosphere. Figure 2 gives a graph illustrating this relation.

Fig. 2.

The normalization used in obtaining the graph in Fig. 2 was carried out by superposing the calcium abundance points from the solar and meteoritic data. The length of the lines in Fig. 2 gives an estimate of the magnitude of the possible errors. The straight line passing through the points makes an angle of \(45^\circ\) with the axes, whence it follows that the known abundances of the elements in the solar atmosphere and in the composition of meteorites coincide.

K. Tolstov