20
VI. SOURCES OF ERROR
Non-ideal gas behavior
Equation (8) was derived using the equation of state for an ideal gas; therefore dry helium is
recommended for use in the MULTIPYCNOMETER. However, dry nitrogen can also be used at
room temperature often with no adverse effect. The use of gases which do not behave in a near ideal
fashion at room temperature should be avoided.
Diffusion/Absorption
When analyzing vegetable matter, materials containing cellulose or low density polymers (including
foams) it is preferred that nitrogen (or sulfur hexafluoride, SF
6
) be used instead of helium because
helium can diffuse into the solid matter and across cell walls.
Impure gases
If air or other gases which contain adsorbable impurities are used, the pressure readings will be
affected due to adsorption on the sample surface. The extent of the resulting error depends upon the
amount and nature of the impurities as well as the solid's surface area. Always use high purity gases
and clean, metal gas lines – never plastic gas tubing.
Insufficiently prepared samples
Many samples contain impurities, usually adsorbed moisture, on their surface and within pores that
should be removed prior to analysis. The presence of these impurities can affect the results in several
ways:
1. The actual weight of the sample is less that the weight measured.
2. Contaminants fill pores causing a larger sample volume to be determined.
3. Volatile impurities will cause erroneous readings.
Successive volume determinations yielding results trending in one direction are usually an indication
that contaminants are being removed after each cycle. Measurements should be repeated until two or
three successive determinations are obtained to within 0.2%.
Another indication of the presence of volatile contaminants is a gradual pressure increase when the
sample is included in the flow path (selector valve in the CELL position, V
R
+ V
C
) after purging
with dry gas. This occurs as the contaminants leave the surface and establish their own partial
pressure.
The size of the gas molecule
An additional source of error in high surface area powders can be the annulus volume created
between the powder surface and the center of mass of the gas phase molecules at the interface.
Assuming that the closest approach of the center of mass of the gas molecules to the powder surface
is 0.5Å (5 x 10
-11
meter) and that the powder surface is in the order of 1000 square meters per gram,
there will exist an annulus volume of 5 x 10
-8
cubic meters (5 x 10
-2
cm
3
) per gram of powder. Thus,
with samples of about 1 gram of high specific surface area, volume errors of 0.05 cm
3
can occur.
Corrections for this error can be made with knowledge of the effective diameter (i.e., Van der Waals
diameter) of the gas molecules and the powder's specific surface area.
To Next Page
Loading...