THE PROBLEM OF THE TRANSFORMATION OF HYDROGEN INTO HELIUM

E. Rabinovich.

The recently published results of experiments by the Berlin chemists Paneth and Peters (1) compel physicists’ attention to return to the problem of the possibility of transforming hydrogen into helium. Theoretically the matter stands as follows: according to present-day conceptions (not directly proven, but having a high degree of probability in their favor), the helium nucleus consists of 4 hydrogen nuclei and 2 electrons. These six particles form a system whose arrangement is unknown to us, but with respect to which we know from the experiments of Rutherford, Chadwick, and Bieler (2) that outwardly it acts as an elastic ellipsoid with semiaxes equal to \(8 \cdot 10^{-13}\) and \(4 \cdot 10^{-13}\) cm. Until the particle approaching the helium nucleus touches the surface of this ellipsoid, it is acted upon by the ordinary Coulomb force, inversely proportional to the square of the distance from the center of the ellipsoid, as though the charge of the helium nucleus, equal to \(2e\), were concentrated at this center. When the extraneous particle (the hydrogen nucleus in Rutherford’s experiments) reaches the surface of the ellipsoid, the repulsive force begins to increase much more rapidly than according to Coulomb’s law; there occurs, as it were, an elastic reflection of the particle from the surface of the ellipsoid. The calculations of Chadwick and Bieler showed that it is impossible to devise such a spatial arrangement of four protons and two electrons, interacting according to Coulomb’s law, for which the external action of the system would be in agreement with these experimental results. Rutherford is therefore inclined to think that the constituent parts of the helium nucleus are bound to one another by other forces, and not by electrostatic attraction according to Coulomb’s law. Attempts were made by Pettersson (3) to explain the sudden increase of repulsion in the immediate vicinity of the nucleus by the polarization of the latter under the action of the approaching particle, and thus to avoid the necessity of abandoning Coulomb’s law inside the nucleus; but this attempt, too, has not yet led to any concrete conceptions about the arrangement of the constituent parts

in the helium nucleus. Thus we know nothing definite about the structure of the helium nucleus, and the problem of this structure still appears theoretically inaccessible.

On the other hand, we have at our disposal an indirect indication of the energy of formation of the helium nucleus, in the form of the well-known hypothesis of the “mass defect,” first proposed by Rutherford (4) and then developed by Harkins and Wilson (5). According to this hypothesis, the “defect” of \(\Delta M = 0.032\, g\), observed in the transformation of four gram-atoms of hydrogen (\(4.032\, g\)) into 1 gram-atom of helium (\(4.000\, g\)), must correspond to a “heat of reaction” equal, according to Einstein’s formula, to \(\Delta E = \Delta M \cdot c^2 = 0.032 \cdot 9 \cdot 10^{20}\) ergs \(= 6.9 \cdot 10^8\) large calories. Let us represent 2 “gram-atoms” of electrons and 4 gram-atoms of protons as two monatomic gases and suppose that the chemical constant of these gases can be calculated by the formulas by which this calculation is carried out for ordinary monatomic gases. Then we can, together with Tolman (6), by means of Nernst’s theorem, calculate the equilibrium of the reaction \(4\mathrm{H}^{+} + 2\mathrm{El} \rightleftarrows \mathrm{He}^{++}\). As is easy to suppose a priori, the calculation leads to the result that even at a temperature up to \(10^6\) degrees and a pressure of \(10^{-100}\) atmospheres, i.e. under conditions most favorable for dissociation, the equilibrium must still lie entirely on the side of association. In other words, hydrogen can in general exist in nature in finite quantity only by virtue of one of the following three reasons: either the hypothesis of the formation of He from four H with the indicated energy is false; or this reaction never, and under no conditions concretely realized in nature, proceeds with finite velocity; or else the quantity of hydrogen disappearing owing to transformation into helium is compensated by the new formation of hydrogen through the disintegration of heavier elements or of helium itself. The latter process cannot, of course, occur spontaneously, but only with the absorption of an enormous amount of energy; the sole possibility for it consists, therefore, in the presence of some cosmic radiation of extraordinarily short wavelength.

Let us suppose that the hypothesis of the structure of \(\mathrm{He}^{++}\) from \(4\mathrm{He}^{+} + 2\mathrm{El}\) is correct. What can prevent their combining into the helium nucleus everywhere that electrons and protons are simultaneously present? All six particles participating in the reaction, according to our present conceptions, are elementary particles, not possessing any internal structure. Consequently, there can be no delays in the reaction analogous to those in ordinary chemical reactions. For example, \(\mathrm{O}_2\) and \(\mathrm{H}_2\) do not combine at ordinary temperature because the first stage of the process must be dissociation, or excitation, requiring energy which the mutual collisions of the molecules cannot supply. Similar delays due to

of the endothermicity of the first stages of the reaction in the formation of helium from hydrogen we cannot at present conceive. But one difficulty, accounting for the infinitely slow rate of the reaction, can readily be found: it lies in the infinitesimally small probability of the sixfold collision necessary for the formation of a helium nucleus. This probability is so negligible that, even if the density of hydrogen were increased to ten thousand atmospheres, collisions of this kind would still be unable to produce any appreciable rate of reaction.

Thus, from the theoretical point of view, there are grounds for considering the transformation of hydrogen into helium an inevitable, but infinitely slow, process, which cannot be appreciably accelerated by any means accessible to the experimenter; for the application of concentrated external energy can play no role in an exothermic process that has no preliminary endothermic stages, while the range of pressures accessible to us is far too insignificant to increase appreciably the probability of the reaction. The only possibility consists in assuming a stepwise reaction, the gradual construction of the helium nucleus with the formation of intermediate products, for example, an isotope of hydrogen with atomic weight 2 from two protons and one electron, or “neutron” from 1 proton and 1 electron, etc.

Such is the theoretical side of the question. As for the experimental side, assertions of a successful transformation of hydrogen into helium have long been encountered in the literature. Let us recall only the experiments, much discussed in their time, of Ramsay himself and of his collaborators Collie, Patterson, and Masson (7) concerning the formation of helium (and neon) in discharge tubes filled with hydrogen. Verification of these experiments by Strutt (8), Merton (9), Egerton (10), Piutti (11), and others gave negative results, while the fact of the presence of neon, the quantities of which usually even exceeded the quantities of helium found, makes it very probable that the results of Collie, Patterson, and Masson are to be explained by the penetration into the apparatus of traces of air; for air, as is well known, contains neon and helium in the ratio 3:1. Paneth and Peters report that they too obtained negative results when attempting to repeat the experiments on the formation of helium in a discharge tube.

All the more unexpected and paradoxical is the positive result obtained by these investigators when palladium acts on hydrogen. At first they detected helium in hydrogen passed in vacuo through a heated palladium capillary; later they found that still more considerable quantities of helium can be detected in hydrogen after it has simply stood over an active palladium preparation—palladium sponge, palladium asbestos, and the like.

THE PROBLEM OF THE TRANSFORMATION OF HYDROGEN INTO HELIUM

The experiments were carried out in such a way that hydrogen, which had stood for some time over palladium, was mixed with oxygen and burned by means of the very same palladium apparatus. Then all the gases (except helium and neon) were absorbed by means of activated charcoal at the temperature of liquid air, and the residue was subjected to spectral analysis. The absence of neon lines in the spectrum was considered proof of the “purity” of the work; in that case the helium could not have come from air that had accidentally entered the apparatus.

As a result of these experiments a parallelism was found between the duration of the hydrogen’s standing over palladium and the amount of helium; however, it turned out that different palladium preparations possess very different intensities of action, and that even the most active preparations gradually lose their activity; sometimes the latter can be restored by ordinary methods—by heating the preparation (in hydrogen, oxygen, or vacuum). In the most successful experiment, with the preparation standing alternately in hydrogen and oxygen for several hours, after standing in hydrogen there was each time found about \(10^{-8}—10^{-7}\ \mathrm{cm}^3\) of helium, whereas after standing in oxygen the amount of helium was 10–100 times smaller.

May these results be regarded as definitive proof of the successful transformation of hydrogen into helium? The large number of possible sources of error was taken into account by Paneth and Peters and eliminated by means of special experiments. These sources are of three kinds: 1) the presence of traces of helium in all materials (cf., for example, the investigations of Strutt (12)); 2) selective diffusion of helium from the air through glass and quartz (13); and 3) possible absorption of neon in charcoal. Of these three sources the first seems to us the most dangerous. It is true that Paneth and Peters showed by means of special experiments that palladium preparations completely give off the gas absorbed by them in an atmosphere of pure helium even upon brief heating, and that it is therefore impossible to suppose that these preparations are capable of absorbing helium from the air in appreciable quantity and retaining it as firmly as would be necessary to explain the results of the experiments. However, one may imagine that helium absorbed by palladium is in fact not completely released upon heating. Perhaps helium, like hydrogen, has the ability to penetrate into the crystal lattice of palladium—though in an incomparably smaller quantity than hydrogen; when hydrogen is absorbed, the helium “stuck” in the lattice is displaced by the hydrogen, which penetrates the whole lattice and expands it, and upon subsequent heating the helium volatilizes.

It will be possible to judge the probability of this and other explanations of a similar kind (a priori, of course, not very plausible) only after

E. RABINOVICH

the appearance of a detailed description of Paneth and Peters’ experiments, and the appearance of further investigations on this question. Until then, one must reckon with the fact that the results of Paneth and Peters are the fruit of three years of persistent and unusually painstaking work, carried out with the expenditure of substantial apparatus resources, and therefore deserve the most serious consideration and confidence.

If the fact of the transformation of hydrogen into helium by means of the catalytic action of palladium is confirmed, then, it seems to us, it will be necessary to recognize as most probable Tolman’s second hypothesis mentioned above. The conditions present in palladium-absorbed hydrogen cannot be in any particular, extraordinary degree more favorable for transmutation than those occurring in nature. According to the latest investigations, hydrogen in palladium, it seems, is not even in the form of free protons, but in the form of neutral H atoms; true, in an unusually high concentration.

Therefore, if the reaction really proceeds in palladium with the rate established by Paneth and Peters, it must also occur to an appreciable extent in nature; and, consequently, the presence of hydrogen in our world can be explained, according to Tolman’s calculations, only by its constant new formation from heavier atoms.

In connection with this conclusion, one may also recall that Eddington (14) and other scientists have repeatedly pointed out that the transformation of hydrogen into helium in nature could serve as a sufficient answer to the question of whence the expenditures of energy in the radiation of the sun and other stars are covered.

LITERATURE

1) F. Paneth und K. Peters. Berichte der Deut. Chem. Ges. 59, 2039 (1926), [Russian translation, see UFN, 6, issue 4—5 (1926)].

2) J. Chadwick and E. S. Bieler. Phil. Mag. 42, 923 (1921).

3) See, for example, Petterson und Kirsch in Handbuch der Physik von Geiger und Scheel, vol. XXII, pp. 171—178 (1926).

4) E. Rutherford. Phil. Mag. 27, 488 (1914).

5) W. D. Harkins und E. D. Wilson. J. Amer. chem. Soc. 37, 1367 (1915).

6) R. C. Tolman. J. Amer. chem. Soc. 44, 1902 (1922).

7) J. N. Collie and H. J. Patterson. J. chem. Soc. 103, 419 (1913); Proc. chem. Soc. 29, 22, 217, J. Masson. Proc. chem. Soc. 29, 233 (1913); J. N. Collie, H. J. Patterson and J. Masson. Proc. Roy. Soc. 91, 30 (1915).

8) R. J. Strutt. Proc. Roy. Soc. 89, 499 (1914).

9) T. H. Merton. Proc. Roy. Soc. 90, 549 (1914).

10) A. C. G. Egerton. Proc. Roy. Soc. 91, 180 (1915).

11) A. Piutti and E. Cardoso. Journ. d. chimie phys. 18, 81 (1920); A. Piutti. Z. f. Electrochemie 28, 452 (1922); A. Piutti and E. Boggio-Lera. Gazzetta chim. italiana 53, 473 (1923).

12) R. J. Strutt. Proc. Roy. Soc. 80, 572 (1907).

13) Jaquerod and Perrot. Comptes rendus 139, 789 (1904); 140, 1542. Archive des sciences phys. et nat. 18, 613 (1904); 20, 128 (1905); Jaquerod and Przemski, ibid., 34, 255 (1912); Williams and Ferguson (J. Amer. chem. Soc. 44, 2160 (1922); 46, 635 (1924)); Lo Surdo, Atti Accad. Lincei 30, I, 85 (1921).

14) A. S. Eddington. Nature; supplement to the issue of 9/XII 1923.