On the basis of an experimental study in a large temperature range, it is shown that “helium densities” of adsorbents measured at room temperature could be erroneous due to a non-negligible effect of helium adsorption. It is proposed that the density obtained with helium at high temperature, for instance, at the regeneration temperature of the adsorbent, be considered as the adsorbent density. Using the corrected densities of 3A, 4A, 5A, and 13X zeolites and of activated and graphitized carbons and of silica gel, we experimentally determined the adsorption of helium on the above mentioned adsorbents at room temperature and in a large pressure range up to 500 MPa. The shape of the adsorption isotherm reveals no saturation at high pressure. These experimental data are in agreement with Monte Carlo simulations of adsorption of a Lennard-Jones gas by a rigid plane as well as by a microporous rigid solid interface. We also examined implications of the new helium density of activated carbon for our previous measurements of adsorption at high pressure. The result is the disappearance of the inexplicable negative part of the isotherms and even a renewed increase in the curves at high pressure. Moreover, a comparison with Monte Carlo simulations of argon adsorption on microporous graphite is in good agreement with the shape of the adsorption curve at high pressure. Finally, the role of the microporous structure of adsorbents and of the gas−adsorbent interaction in adsorption at high pressure is discussed.

No abstract

Density data for helium and neon up to \0 kbar at 25 'C are reported. The earlier experimental results for the Clausius-Mossotti function have been extended to higher densities for the whole noble gas series (i.e .• up to reduced density pO"",!). These results. combined with the machine calculation of the dipole-induced-dipole contribution with respect to density for a Lennard·Jones fluid. allow the determination of the mean individual atomic polarizability of noble gases as a function of density. In an appropriate system of units. the values of the mean polarizability fall on a single universal curve for all the reduced temperatures investigated. III. RESULTSFor the 25 °C isotherm, a knowledge of the molar density along with the dielectric constant value for a given pressure allows us to deduce the Clausius-Mos-

The static dielectric constant of noble gases (excepting xenon) and nitrogen has been measured near room temperature in a large range of pressures (up to 12 kbar for He, Ne, Ar, and N 2 ). The Clausius-Mossotti function (C.M.) has been deduced, when density data are known, with a precision of 10-3 in the moderate pressure range and 3 X 10-3 above 1.0 kbar. Values of the first, second, and in some cases third dielectric virial coefficient have been determined. We have found, in agreement with previous results obtained at low pressure, that the light gases and the heaviest ones show a different behavior of their C.M. function in terms of the density. For He and Ne, the second dielectric virial coefficient is negative; for Ar, Kr, and N2 it is positive, but the C.M. function goes through a maximum and then decreases. The effect of density on the C.M. function of Ar and N2 is approximately a drop of 3.5% at 12 kbar. These results have been compared with the predictions of molecular theories of dielectric properties of nonpolar fluids in the case of a hard spheres model. We have found that the Kirkwood-Yvon theory based upon the DID approximation does not reproduce completely the results of experiments. By contrast, we have been able to show that the semiquantal theory of Heller and Gelbart which takes into account the influence of the electronic distortion is more in agreement for a wide range of densities. This is the case for Ne and Ar when the excess of the pair polarizability tensor of a diatom is computed within the electron-gas approximation, and for He only if it is deduced from the Hartree-Fock scheme.