NEUTRON DIFFRACTION IN GASES

The scattering of X-rays in gases and liquids is substantially affected by intratomic, intramolecular, and intermolecular interference (see, for example, ¹). Intramolecular interference plays the predominant role in scattering in gases; external (intermolecular) interference plays the predominant role in scattering in liquids and in gases under significant pressure. Fig. 1 shows an interference curve for X-ray scattering in gases. The quantity \(\rho\) characterizes the density of the gas.

At high densities the curve corresponds to X-ray scattering in a liquid. The curve shown is constructed without taking the atomic factor into account, i.e., the intratomic interference; it should therefore also be valid for neutrons whose scattering by nuclei is spherically symmetric.

Fig. 1.

In the apparatus shown in Fig. 2, neutron diffraction in gases was studied³. Monochromatic neutrons with an energy of \(0.07\) eV, obtained by reflection from the (200) plane of an NaCl crystal, were passed through oxygen and carbon dioxide. The gases under investigation were contained in a steel cylindrical vessel at room temperature and a pressure of 60 atmospheres. The intensity of the neutrons scattered by the gases was measured with a counter filled with BF\(_3\).

Fig. 2.

Since the scattering of neutrons by gases is very small, special attention was paid to reducing the background of fast neutrons and of neutrons scattered by the walls of the vessel. For this purpose the steel of the vessel was chosen so that it did not give coherent scattering of neutrons with an energy of \(0.07\) eV in the angular range from \(0\) to \(30^\circ\). In the region of large angles (\(>30^\circ\)) the counter was protected from neutrons scattered by the walls by special screens. Despite all precautions, the background nevertheless amounted to half of all counted neutrons. In addition, double scattering was possible (theoretically up to 10%), for which

a correction was introduced, but no attempts were made to reduce it.

Figures 3 and 4 show the experimental results for both gases, as well as theoretical curves derived from semiclassical

Fig. 3.

Fig. 4.

representations: the neutron flux is represented in the form of a wave, and the molecules are replaced by a rigid system of scattering centers. In calculating the theoretical curves, corrections for the Doppler effect were made.

Paramagnetism of oxygen was not taken into account. The vertical scale was chosen so that the experimental points coincided with the theoretical curves. The positions of the maxima and minima are due to the geometrical shape of the molecule.

The good agreement between experiment and theory, which is seen when comparing Figs. 3 and 4 with Fig. 1, indicates the applicability of the semiclassical theory to this question. A divergence of the experimental data from the theory in the case of CO₂ was to be expected: at 60 atmospheres the density of this gas is such that intermolecular interference becomes significant. For oxygen at 60 atmospheres this phenomenon does not occur, since this is above its critical pressure.

R. Ozerov

References

1. V. I. Danilov, Scattering of X-rays in Liquids. Problems of Modern Physics, issue XXV, ONTI, 1935.
2. N. Z. Alcock, D. G. Hurst, Phys. Rev. 75, 1609 (1949).