Sample Preparation – A Few Important Considerations

User Rating:  / 1
Published Date Written by Frances O'Neill

Sampling errors arise both in the taking of the original sample and in attempting to take a small representative sub-sample from that sample. The accuracy of reported results, be it a solution, fusion bead or powder briquette, is dependent on the integrity of the sample preparation. This necessitates that all procedures in the preparation of the sample must be performed in a reproducible, ordered and systematic manner.

The goal of sample preparation is to obtain an analytical sized portion that is representative of the bulk material submitted so that once prepared and analysed, results can be confidently reported as a true composition reflection of the original material.

In some instances sample preparation can be quite elaborate and include drying, crushing, sample division/splitting and pulverisation/milling, perhaps even multiple steps of the aforementioned depending on the mass of bulk sample received!

Particle size of the sample also plays a vital role in the preparation process. It is important to crush bulk material to as small a particle size as possible before dividing and continuing along the preparation process. Error, for a given mass and component concentration, is minimised in a finer particle size sub-sample, thus shrinkage in particle size must precede any mass or volume reduction.

Contamination always poses a threat during sample preparation, especially when analysing for minor or trace element concentrations. Successful preparation and analysis depends on recognising potential contamination sources and controlling them. The crusher jaws and the milling/pulverisation pots are in direct contact with the sample and are possible spoilage sources. In order to minimise interference define, prior to analysis, the analytes and concentrations of importance  then choose jaws and pots made from appropriate materials.

Bias is another consideration. Different techniques and types of equipment are available to assist in splitting a sample into a representative sub-sample. Options include random sampling, cone-and-quartering, riffle dividers and the more preferred rotary sample dividers. Essentially, with a rotary sample divider, bulk material is distributed evenly between a number of receiving containers (cups) rotating on a carousel. The theory of operation is that by rotating the containers each cup receives an equal split of sample. Furthermore each portion should contain an even distribution of particle fineness and be a representative sub-sample, avoiding sample bias.

During the pulverisation step, it is important that the correct amount of sample is milled in a given vessel. It is not desirable, for example, to grind a small sample in a large vessel, as self-attrition of the lining (eg tungsten carbide or zirconia) of the vessel is more pronounced, greatly increasing the contamination. Furthermore, loading too little sample or using too large a milling vessel can result in pronounced conglomeration (caking) of the sample during the milling cycle. Inserting too much sample in too small a vessel will result in ineffective grinding, even to the point of failure. When selecting the optimal pulverisation time and milling speed, care should be taken to avoid unintentionally changing the characteristics of the original sample. Excessive grinding could result in oxidation of the sample (eg samples containing ferrous iron), or even loss of volatiles due to excessive heating of the sample (such as sulphur in pyrite concentrations).

It is only when the analysed volume of the prepared sub-sample and the original bulk sample are of the same composition that the laboratory analyst is able to provide an analysis that is truly representative of the original material.

Copyright © 2018 UIS Analytical Services. All Rights Reserved.
Joomla! is Free Software released under the GNU General Public License.
Joomla 1.6 Templates designed by Joomla Hosting Reviews