On the Work of Ionization and Dissociation of Hydrogen.

Tea Krüger. Ionisations- und Dissociationsarbeit d. Wasserstoffs. Ann. d. Phys. 64, p. 288 (1921).

Among the many consequences that followed from the application of Bohr’s model of the hydrogen atom, a prominent place is occupied by the explanation of the work of ionization of the atom and of the work of dissociation of the hydrogen molecule into atoms. Until recently, however, there were no exhaustive experimental studies on this question, despite the fact that a whole series of authors had already carried out investigations that always gave, however, only very approximate conclusions. In none of the preceding investigations, in particular, were (a) the work of ionization and the work, absorbed by resonance radiation, separated; (b) it was not taken into account that free electrons, in their impacts, encounter not atoms but molecules of hydrogen.

The author of the paper under review succeeded in investigating in detail the dissociation and ionization of hydrogen and in following all stages of these processes. The source of free electrons was a heated tungsten filament \(P\), surrounded by two coaxial platinum grids \(D_1\) and \(D_2\) and, finally, by a solid platinum cylinder \(Z\).

Between \(P\), \(D_1\), \(D_2\), and \(Z\) arbitrary potential differences could be established; the cylinder \(Z\) was led to earth through a sensitive \((10^{-8}a — 5 \cdot 10^{-11}a)\) galvanometer.

The distances from \(P\) to \(D_1\) and from \(D_2\) to \(Z\) were not greater than the free path of an electron (at the pressures present in the vessel, measured with a Mac Leod manometer). The distance from \(D_1\) to \(D_2\), however, considerably exceeded this value. The electric fields between \(P\) and \(D_1\), \(D_1\) and \(D_2\), \(D_2\) and \(Z\), were chosen so that electrons, accelerated on the path \(P D_1\), could not reach the cylinder \(Z\) (the field \(D_1 D_2\) and \(D_2 Z\) was directed ...).

was applied oppositely to \(P D_1\), while the positive ions, formed as a result of collisions in the region \(D_1D_2\), were directed toward the cylinder \(Z\), producing a current in the galvanometer. The author succeeded in eliminating the errors introduced by the initial velocity of the electrons flying out from the heated filament, by the potential drop in the latter, and by the contact potentials of the grids.

Plotting on a diagram the corrected values of the potentials that cause ionization (along the abscissa axis) and the current strength in the galvanometer, proportional to the number of positive ions formed (along the ordinate axis), it was possible to detect rather sharp breaks in the curves—breaks corresponding to the moments of a sharp increase in ionization.

These moments, as it turned out, correspond to potentials of \(17.1 \pm 0.25\) volts and \(30.4 \pm 0.5\) volts. But, besides ionization by collision, resonance radiation may also arise in the apparatus, on which work is evidently also expended. In order to trace the latter phenomenon, the author used the methods of Bergen Davis and Goucher. The potential difference between \(D_1\) and \(D_2\) was set equal to \(+36\) volts, while between \(D_2\) and \(Z\) the potential difference was \(-10\) volts. When resonance radiation arises, the latter acts on the grid \(D_2\), and this grid begins to emit electrons, which are carried by the field toward the cylinder and impart to it a negative charge with respect to the earth. Thus a current appears in the galvanometer, directed in the direction opposite to the ionization current. The curve \(i=f(v)\) bends downward from the abscissa axis, and from the moment of bending one may judge the moment at which resonance radiation arises.

The general course of the curves obtained by the author showed the following:

a) at \(11.5 \pm 0.7\) volts, weak ionization and weak ultraviolet radiation set in;

b) at \(13.6 \pm 0.7\) volts—strong radiation;

c) at \(17.1 \pm 0.25\) volts—the first stage of strong ionization;

d) at \(30.4 \pm 0.5\) volts—the second stage of strong ionization.

In order to interpret the results obtained from the point of view of Bohr’s theory, the author considers the changes that may occur in Bohr’s hydrogen molecule:

1) If the existence of molecule-ions is possible, then, at the corresponding ionizing potential, one electron is removed from the molecule, and it is transformed into a positive moleculion.

2) If the molecules absorb energy equal to the sum of the energy of resonance radiation \(R\) and the energy of dissociation \(D\), then one neutral stationary atom and one radiating atom arise.

3) If the energy absorbed by the molecule proves equal to the dissociation energy \(D\) plus the ionization energy of one atom \(J\), then one neutral atom and one atomion arise.

4) If the absorbed energy is equal to the sum \(J+R+D\), then one atomion and one radiating atom arise.

5) If the absorbed energy is equal to \(D+2R\), then two radiating atoms arise (the first stage of strong ionization).

6) If the absorbed energy is equal to \(D+2J\), then two atomions arise (the second stage of strong ionization).

If cases 4 and 5, which require unusually sensitive methods for their detection, are set aside, then all the remaining cases very well encompass the author’s experimental results.

Indeed, comparing c with 5 and d with 6, one may conclude that

\[ J+D=17.1 \text{ volt} \]

\[ 2J+D=30.4 \text{ volt} \]

Hence (taking into account the weight of the observations and finding the mean) the author finally obtains

\[ J = 13.3 \pm 0.25 \]

which is in excellent agreement with the value \(13.5\ V\), obtained by Bohr theoretically.

\[ D = 3.53 \pm 0.3\ \text{volt} \]

which, when converted into thermal units, gives

\[ D = 81300 \pm 5700\ \text{gr-cal.} \]

a number differing from the theoretical one (60000) by 25%.

The value \(D\) is also computed by the author in a somewhat different way. Namely, the strong radiation observed at \(13.6\ V\) corresponds, evidently, to case 2, a; consequently

\[ 13.6 = D + R. \]

But the resonance potential \(R\) can be determined from the study of the lines of the absorption spectrum.

The calculations give for \(R\) the value \(R = 10.1\), and therefore:

\[ D = 13.6 - 10.1 = 3.5 \]

which agrees well with the figure found above.

It remains to mention also case 1, when a molecule turns into a molecule. This case, in the author’s opinion, occurs precisely at \(11.5\) volts, when weak ionization is observed, varying within wide limits with a change in pressure and ceasing completely at pressures greater than \(0.08\) mm of mercury.

Such a dependence on pressure is not observed for the regions of strong dissociation (at \(17.6\ V\) and \(30\ V\)), which the author explains by the large volume of the ion in comparison with the atomic ion.

In conclusion, the work gives a whole series of observations by other investigators, indirectly confirming the author’s conclusions.

Vas. Shuleikin.