Hanna Instruments HI 904 - Manual (Page 119)

9-7
METHODS OPTIMIZATION
9.2 The Sample
9.2.1 Proper Sampling Procedure
Proper sampling is essential for accurately determining the water content of bulk materials,
particularly with non-homogeneous samples. Many standard methods detail instructions to
ensure proper sampling. As a general rule, the following guidelines should be followed:
1. The sample must be representative. The water content of the sample taken is the
same as the average water content of the bulk material.
2. Avoid exposing samples to the contaminating effects of atmospheric moisture.
Take samples as quickly as possible and protect the sample during transport and/or
storage.
3. Take samples from the interior of bulk materials. Surfaces of hygroscopic materials
may contain higher levels of moisture relative to the rest of the material. Surfaces
of materials which release water may contain less water relative to the rest of the
material.
4. Taking large samples of bulk materials will result in a more representative sample.
9.2.2 Determining the Optimal Sample Size
The HI 904 titrates optimally in the range of 0.5 to 2.0 mg water per sample. Ideally, the
sample size would be scaled to always be in this range, but it becomes impractical to add the
large sample sizes that would be required for concentrations of 500 ppm and lower. Attempting
to add more than 10 g of sample increases the chance of erroneus results due to the sample
significantly changing the composition of the reagent. In addition, the titration vessel will
quickly fill after a couple of titrations if the sample size is 10 g. For samples below 500 ppm,
the sample size should be a balance between titration accuracy (bigger sample size) and
economy (less reagent waste). The following table shows the recommended sample size
based on the moisture content:
9.2.3 Solid Samples
Solid samples should never be analyzed directly in the titration vessel. Solids greatly increase
the risk of clogging the diaphragm of the generator, which could cause permanent damage.
In addition, opening the titration vessel in order to add a solid sample would introduce a
significant amount of moisture causing false high readings and increasing the drift between
titrations. These samples should be analyzed by either external extraction or external
dissolution.
t n e t n o C r e t a W m p p 1 m p p 0 2 m p p 0 0 1 m p p 0 5 2 m p p 0 0 0 1 % 1 % 5
e z i S e l p m a S g 0 1 g 5 g 3 g 2 g 1 g 1 . 0 g 2 0 . 0