Sample preparation

It's common knowledge that poor sample preparation is, next to poor sampling, the largest source of bias in an exploration or resource evaluation project. Sample preparation methods should therefore be selected as carefully as the actual analytical methods.

The objective of a precise sample preparation scheme is to produce a representative and meaningful test sample (regularly about 100 - 150 g) from a large bulk sample. The grain size of the prepared sample must be so fine that the element of interest (or host mineral) can be properly liberated from the bulk matrix and distributed in the pulp to produce a homogeneous distribution to ensure sufficient representativity for the following analytical methods. This is particularly important for low-concentration ores (e.g. Au and PGE's) where the number of mineral particles producing ore concentration is always low.

Different minerals behave differently during pulverisation – most (brittle) minerals will easily break down to small particles while some (e.g. native gold) will just change their shape if proper sample preparation methods are not used.

It is commonly accepted that poor sample preparation is, next to poor sampling, the largest source of bias in an exploration or resource evaluation project. Sample preparation methods should therefore be selected as carefully as the actual analytical methods.

For routine sample preparation we recommend our automated sample preparation system in Rovaniemi laboratory.

Conventional, manual sample preparation (crushing, splitting, pulverizing) is available in our laboratories in Rovaniemi, Kuopio and Sodankylä. Petrological sample preparation is carried out in Espoo laboratory.

1.1.1 Drying

All samples are always dried no matter what the earlier sample preparation history is (Method 10). Exceptionally wet and large samples (RC-, chip-samples etc.) require longer drying in elevated temperature (Method 14).

1.1.2 Crushing

The standard scheme consists of direct one–stage fine crushing using a special type jaw crushers (nominal particle size > 70 % < 2 mm), precision riffle splitting (Method 31) and pulverising the split subsample of 100 –150g. This is a suitable method if crushed reject is needed for future work. The use of this method is meaningful to maximum size of 2000 g samples, because if more than 3 - 4 splittings are required the representativity of the split subsample can not be assured. If larger than 2000 g samples are prepared using this procedure an extra pre - crushing (Method 30) is invoiced.

For samples containing visible gold and/or for unusually big or heterogeneous samples, (max. 3,5 kg) we recommend standard crushing to 5-10mm (Method 30) and followed by pulverising the entire test sample (Methods 50) using Essa LM5 mills.

If crushed reject is required for future work the crushed material can be split to two (e.g. 1-2 kg) splits (riffle splitting Method 35) – the other for storage and the other for pulverising (Method 50).

1.1.3 Pulverising

Pulverising will always cause unavoidable contamination of wear metals at trace level from the grinding surfaces. This contamination may vary depending on material of the bowl, hardness of the sample material, pulverising time etc. The pulverising method must be selected to best serve the requirements of the client.

Some examples of bowl materials used at LABTIUM and expected contamination:

• carbon steel (< 0.2 % Fe, no base metals)
• hardened steel (< 0.2 % Fe, low Mn, Ni, Cu, Cr, Co)
• chrome steel (up to 200 ppm Cr, < 0.2 % Fe, traces Mn, Cu, Co)
• tungsten carbide (W, Co)
• agate (Si)

To minimise cross-contamination, cleaning of pulverising bowls between samples (pulverising with barren quartzite) is included in the price in all Labtium pulverising methods. The pulverisers and jaw crushers are cleaned with compressed air and brushes between every sample

The standard scheme consists of direct one–stage fine crushing using a special type jaw crushers (Method 31 nominal particle size > 70 % < 2 mm), precision riffle splitting (Method 35) and pulverising the split subsample of 100 –150g (Method 40).

For samples containing visible gold and/or for unusually big or heterogeneous samples, (max. 3,5 kg) we recommend standard crushing (Method 30) and followed by pulverising the entire test sample (Methods 50, hardened steel bowl) using Essa LM5 mills, avoiding any sample splitting which may deteriorate representativity of large samples. The pulverising takes place in large bowl and provides a large homogenised test sample for representative subsampling