In order to obtain correct data it is required to remove all physically adsorbed material from the adsorbent surface to ensure a reproducible initial state of the adsorbent surface, especially one in which pores are obstructed by foreign species. This can be accomplished by vacuum pumping or purging with an inert gas at elevated temperatures. Vacuum is attractive, because it prepares the surface under the same conditions that are required to start a static volumetric adsorption experiment (i.e., to start a such an adsorption experiment the sample cell with adsorbent has to be evacuated). In addition, it also allows outgassing at lower temperatures than one would need if flow outgassing (purging) under atmospheric conditions were applied. A drawback of the vacuum method is certainly the problem of powder elutriation, which does not occur as readily in the flow method. An additional advantage of flow is that its setup is very easy (i.e., no expensive vacuum system is required).

The sample should be outgassed at the highest temperature that will not cause a structural change to the sample. In general, too low an outgassing temperature will cause lengthy preparation, and may result in lower than expected surface areas and pore volumes. In general, outgassing organics must be performed with care since most have quite low softening or glass transition points (e.g., magnesium stearate). In contrast, most carbon samples for instance can be outgassed quite safely at 573 K. Physisorbed water in nonporous or mesoporous materials will be lost at relatively low temperatures « 473 K) under the influence of vacuum, but if adsorbed in narrow micropores, as they are present in some zeolites, high temperatures (up to 573 K) and long outgassing periods (often no less than 8 hours) are required. A special heating program is often needed, one which allows for a slow removal of most of the preadsorbed water at temperatures below 373 K accompanied by a stepwise increase in temperature until the final outgassing temperature is reached. This is done to avoid potential structural damage of the sample due to surface tension effects and so-called "steaming", i.e., hydrothermal alteration. In particular zeolites are sensitive to steaming, where the possibility of vaporization and re-condensation inside the pores can lead to structural changes.

In those instances where samples cannot be heated, the method of repetitive cycling investigated by Lopez-Gonzales et af can be utilized. They found that by repetitive adsorption and desorption the surface can be adequately cleaned to allow reproducible measurements. Usually three to six cycles are sufficient to produce a decontaminated surface. Presumably the process of desorption, as the sample temperature is raised, results in momentum exchange between the highly dense adsorbate leaving the surface and the contaminants. As the impurities are removed from the surface they will be carried out of the sample cell by the flowing gas. Thus the technique of repetitive cycling is an efficient means for removal of contaminants from the surface of a solid.

If the vacuum method is performed outgassing of the sample to a residual pressure of about 1 Pa to 0.01 Pa is considered to be satisfactory for most nonporous and mesoporous materials. This can be readily achieved by a combination of a rotary and diffusion pump. The usual precautions should be taken to insure no contamination by the pump oil. However, as already mentioned, microporous materials such as zeolites require an outgassing at much lower pressures, i.e., below 0.01 Pa. Nitrogen adsorption occurs here at relative pressures P/Po even below 10^-7 for pores of diameter below ca. 6A. Hence, the sample should also be outgassed at these very low pressures. This can be achieved by using a turbomolecular pump which, if coupled with a diaphragm roughing pump, allows the sample to be outgassed in a completely oil free system.