TABLE OF CONTENTS Duncan (1962, p. 330) has stated that "Randomness in the selection of original increments is indeed the number one problem in the sampling of bulk material." Fortunately, Gy (1998, p. 28) has developed a principle for randomness to apply to bulk sampling and has found ways to implement it.
Every part of the lot has an equal chance of being in the sample and
the integrity of the sample is preserved during and after sampling.
This principle is the guide for the random sampling of amorphous material. In the first part, we have to make sure that we
theoretically define the sample we will take so that it follows this principle and
physically obtain the sample we have correctly defined. [7]
Failure to carry out these two steps properly contributes most often to bias in sampling. If part of the lot is inaccessible or the sampling tool cannot take the sample we have correctly defined, then getting a random sample is not possible. In the second part of the principle, we must ensure that we
preserve the integrity of the sample while it is being taken and also between the time it is taken and the time it is analyzed. [8]
Oxidation, abrasion, and evaporation are examples of improper sample handling because the sample integrity is not preserved. Getting a representative sample is useless if the value of the property...
