Sample Handling
Sampling and uniform sample preparation and sample application before a drying is important to obtain repeatable moisture results.
Safety information
Some samples require particular care as they can
harm people or damage objects. These include combustible, explosive,
toxic or corrosive substances and/or samples which release these
substances when dried/heated.
- Never dry combustible or explosive substances. If in any doubt, use
small sample volumes (max. 1 gram) and low temperatures.
- Carry out a risk analysis (e.g. with regard to risk of explosion,
combustibility, toxicity and corrosiveness of sample and the vapors
released when heated).
- If necessary, work in a fume cupboard (adjustment should then be
undertaken there).
Warning: The surface temperature of the radiator is higher than
the measurement temperature and hence may ignite combustible vapours as
they pass by.
Note: Please note that the user bears all responsibility and
liability for damage resulting from the use of the aforementioned types
of samples.
|
The Halogen Moisture Analyzers offer
both precision and speed. You save time and costs and increase
efficiency.
|
Sample pans
Only use clean sample pans for moisture
determination.
- Do not use deformed sample pans.
Tip: Using single-use aluminum sample pans guarantees reliable
measurements free from the influence of residue remaining from previous
samples or cleaning agents. These aluminum sample pans are also
available in a reinforced version. These are suitable for samples which
contract when drying and may deform the pan.
Please dispose of the used pans properly.
|
Sampling
The way in which samples are taken has a major
impact on the reproducibility of
the measurement results:
- Representative of total volume

- Ensure homogeneity (mixed well), e.g. mix and stir the total volume
first etc.
- Sufficient sampling
- No addition or removal of moisture when taking samples (work as
quickly as possible)
- If measurements are not being taken straight away: Store in
airtight container without an air cushion (fully filled).
|
Sample preparation
The correct preparation of samples once
they have been taken is also key to repeatable and reliable
results.
- Ensure even granulation (particle size).
- If necessary increase the sample surface area by breaking up the
sample. This will ensure a better and faster release of moisture during
drying (faster diffusion of moisture to the surface).
The sample should not be heated at this stage as
this would cause moisture to be lost during preparations.
Mechanical crushing can be carried out e.g. using a mortar, grinder
(watercooled) or simply by cutting.
Tip: You can increase the surface area and therefore speed up
the drying of liquids by using a glass-fiber filter.
|
Sample application
An evenly spread sample results in a
homogeneous distribution of heat throughout the product being measured
and the moisture can diffuse evenly out of the sample. This generates
results which are easier to reproduce.
- Carefully mix your sample before adding to the sample pan.
- Always use the same volume of sample to achieve good repeatability.
Use the right sample volume. The pan should
be thinly and evenly covered with the sample over the total surface
area.
Spread the sample evenly over the pan (do
not build up piles).
Influence of sample weight on repeatability and length of
measurement:

A higher sample weight will mean that more water is evaporated and
moisture determination will take longer. In addition, too large a
sample volume can result in an uneven distribution of heat and thus to
less precise results. On the other hand, repeatability decreases
(higher standard deviation) as the sample weight decreases:
- Standard deviation for 2 g sample1: 0.05%
- Standard deviation for 10 g sample1: 0.01%

1Assuming an ideal sample where all of the moisture can
always be removed without causing decomposition (e.g. moist sand).
Deviations result from the substance-dependent uncertainty and the
repeatability guaranteed by the instrument (in this case: HR83). In
reality, differences in measurements occurring in one series of
measurements (not ideal samples) may be greater than the values shown
in the table. |