Page 105 - Vogel's TEXTBOOK OF QUANTITATIVE CHEMICAL ANALYSIS

surface caused by rubbing, and (c) by its temperature being different from
that of the balance case. These errors may be largely eliminated by wiping
the vessel gently with a linen cloth, and allowing it to stand at least 30 minutes
in proximity to the balance before weighing. The electrification, which may
cause a comparatioely large error, particularly if both the atmosphere and the
cloth are dry, is slowly dissipated on standing; it may be removed by subjecting
the vessel to the discharge from an antistatic gun. Hygroscopic, efflorescent,
and volatile substances must be weighed in completely closed vessels.
Substances which have been heated in an air oven or ignited in a crucible
are generally allowed to cool in a desiccator containing a suitable drying
agent. The time of cooling in a desiccator cannot be exactly specified, since
it will depend upon the temperature and upon the size of the crucible as well
as upon the material of which it is composed. Platinum vessels require a
shorter time than those of porcelain, glass, or silica. It has been customary
to leave platinum crucibles in the desiccator for 20-25 minutes, and crucibles
of other materials for 30-35 minutes before being weighed. It is advisable to
cover crucibles and other open vessels.
2. When a substance is immersed in a fluid, its true weight is diminished by the
weight of the fluid which it displaces. If the object and the weights have the
same density, and consequently the same volume, no error will be introduced
on this account. If, however, as is usually the case, the density of the object
is different from that of the weights, the volumes of air displaced by each
will be different. If the substance has a lower density than the weights, as is
usual in analysis, the former will displace a greater volume of air than the
latter, and it will therefore weigh less in air than in a vacuum. Conversely,
if a denser material (e.g. one of the precious metals) is weighed, the weight
in a vacuum will be less than the apparent weight in air.
Consider the weighing of 1 litre of water, first in oacuo, and then in air.
It is assumed that the flask containing the water is tared by an exactly similar
flask, that the temperature of the air is 20 OC and the barometric pressure is
101 325 Pa (760 mm of mercury). The weight of 1 litre of water in oacuo under
these conditions is 998.23 g. If the water is weighed in air, it will be found
that 998.23 g are too heavy. We can readily calculate the difference. The
weight of 1 litre of air displaced by the water is 1.20g. Assuming the
weights to have a density of 8.0, they will displace 998.2318.0 = 124.8 mL, or
124.8 x 1.20/1000 = 0.15 g of air. The net difference in weight will therefore
be 1.20 - 0.15 = 1.05 g. Hence the weight in air of 1 litre of water under the
experimental conditions named is 998.23 - 1.05 = 997.18 g, a difference of
0.1 per cent from the weight in oacuo.
Now consider the case of a solid, such as potassium chloride, under the
above conditions. The density of potassium chloride is 1.99. If 2 g of the Salt
are weighed, the apparent loss in weight (= weight of air displaced) is
2 x 0.0012/ 1.99 = 0.0012 g. The apparent loss in weight for the weights is
2 x 0.0012/8.0 = 0.000 30 g. Hence 2 g of potassium chloride will weigh
0.0012 - 0.000 30 = 0.000 90 g less in air than in oacuo, a difference of
0.05 per cent.
It must be pointed out that for most analytical purposes where it is desired
to express the results in the form of a percentage, the ratio of the weights in
air, so far as solids are concerned, will give a result which is practically the
same as that which would be given by the weights in oacuo. Hence no buoyancy