Holographic Interferometry for Determination of Binary Diffusion Coefficients of Gas Mixtures of Helium with Krypton and Ammonia

Summary and outlook: This study presents experimental binary diffusion coefficient data measured with a Loschmidt cell combined with holographic interferometry for gas analysis. This measuring technique is supposed to be the most reliable absolute method for the determination of the concentration dependence of the binary diffusion coefficient of gases [3]. The Loschmidt cell is designed for a temperature range from 283.15 K to 353.15 K and a pressure range from vacuum to 10 bar. lt is equipped with two holographic interferometers, one for each half-ell. Starting the diffusion experiment with pure gases the upper half-cell provides data for the mole fraction range of the heavier compound 0 ::::: x1 ::::: 0.5 and the lower half-cell for 0.5 :::; x1 :::; 1. This way the concentration dependence can be determined over the complete mole fraction range in only one experimental run. Altogether our setup can investigate the concentration, temperature and pressure dependence of the binary diffusion coefficient in a timesaving way.

Measurements were conducted with the gas pairs krypton-helium and ammonia-helium. The system krypton-helium was investigated to validate the measuring range of our experimental setup. A noble gas pair was chosen because it shows weak real gas behaviour and the diffusion was supposed to be approximated by Fick's second law. There is also a variety of theoretical and experimental data for this system available in literature for comparison. Ammonia-helium was chosen as a system with a molecular compound due to its stronger real gas behaviour. The influence on the data evaluation approximated by ideal diffusion should be tested. This gas pair has also a technical relevance; it is used for the design of diffusion absorption refrigerators where especially the concentration and pressure dependence are of interest.

The system krypton-helium was investigated at 293.15, 313.15, 333.15 and 353.15 K and the pressures 1, 2, 5 and 10 bar. At 293.15 K we achieved increasing D12Pmix data with increasing mole fraction of krypton which accords with predictions from the kinetic theory of dilute gases. However, the upper half cell as weil as both half-cells at higher temperatures shows systematic deviations from the kinetic theory. Previous investigations suggested a correlation between the systematic deviation in the upper half-cell and the diffusive velocity. In this study the deviations in the upper half-cell increased with increasing temperature, i.e. with increasing velocity. However, a correlation for the lower half-cell could not be found. For the temperature dependence we produced increasing values of the binary diffusion coefficient with increasing temperature. Our data are in very good agreement with literature data and theoretical data based on the kinetic theory of dilute gases. Data for the pressure dependence of the binary diffusion coefficient is very rare. A comparison of our data with the only available literature data shows that both studies achieved slightly decreasing D12 ·pmix values with increasing pressure. We achieved uncertainties lower than 0.75% depending on the temperature and pressure.

For the system ammonia-helium sorption effects were discovered. According to the Kerl group [3, 32, 33] lubricants are the main reason for sorption. That is why the diffusion cell is constructed in a way that lubricants were necessary. However, we found sorption at the sealings and cell walls by investigating the pressure change inside the Loschmidt cell during diffusion of ammonia and helium. For compensation of this effect we tried saturating the cell with ammonia prior to the experiment. The saturation time had a strong influence on the concentration dependence of the diffusion coefficient data. As very long saturation times resulted in gas overflow between the two half-cells a compromise of two hours for saturation was applied. The sorption effect was not yet the uncertainties are smaller than 0. 75%. With our apparatus the determination of the concentration dependence of binary diffusion coefficients is still difficult, but we achieve very good results for the temperature and pressure dependence.

For future studies the data evaluation has to be extended to consider sorption effects as weil as further transport effects such as gravitational forces or convection. To learn more about these effects further measurements should be conducted for example on the noble gas system krypton-xenon. This gas pair is characterized by a very small binary diffusion coefficient and is therefore suitable for the characterization of the bending behaviour. Disagreements to the literature data should be less distinct if the correlation between the bending behaviour and the diffusive velocity holds.

Furthermore, systems with two molecular components should be investigated. The system carbon dioxide-nitrous oxide (C02-N20) is characterized by equal molar masses and approximately equal second pressure virial coefficients, i.e. the particle number in the two half-cells filled with pure gases are identical. This prevents convection due to difference in molar masses or particle numbers at the beginning of the diffusion measurement and allows the examination of the intluence of acceleration and delay of the particles in the upper and lower half-cell. The same weight force acts on the particles of the upper and lower half-cell, but the particles of the upper half-cell are accelerated whereas the particles from the lower half-cell get delayed. taken into account in the data evaluation.


Binary diffusion coefficients of the gas pair ammonia-helium were determined at 293.15, 313.15, 333.15 and 353.15 K and the pressures 1, 2, 5 and 8 bar. Due to the sorption effects it was not possible to reliably determine the concentration dependence. We found increasing as weil as decreasing values for D12 "Pmix in both half-cells. A correlation of the bending behaviour with the diffusive velocity was not observed for this system. However, our absolute values of D12 ·pmix agree very well with the literature data and theoretical values based on the kinetic theory of gases at zero density. For this system, too, our data for the temperature dependence are in good accordance with theoretical and experimental literature data. The pressure dependence of the diffusion coefficient for the system ammonia-helium was examined experimentally for the first time. Our data was compared to theoretical values calculated by the kinetic gas theory at zero density [17]. We obtained a stronger decrease for D12 ·pmix values with increasing pressure in comparison to the system krypton-helium. For the system ammonia-helium