This method for the measurement of sodium and potassium in silicate rocks uses a borate fusion technique.
The sample is fused with a suitable fusion agent such as boric acid (H3BO3) and lithium carbonate (Li2CO3) and the ground product is dissolved in dilute citric acid. The presence of other metals such as aluminium, calcium, iron, magnesium and silicon should not interfere with the measurement of sodium or potassium using this method.
The prefusion mixture is prepared by mixing and grinding together 200g H3BO3, 60g of Li2CO3, 30g of SrCO3 and 10g of Cobalt (II,III) oxide.
This method utilises the breakdown of Cobalt (II,III) Oxide at 950 degrees centigrade to form Cobalt (II) Oxide, thus it is imperative that the oven is able to hit this temperature for the analysis.
The sample is prepared by mixing 200mg of the ground rock powder with 4g of fusion agent. The fusion is carried out at 960ºC in a graphite crucible over a 3-hour period.
The sample is then removed from the oven and allowed to cool to handling temperatures. The pellet is then removed from the base of the crucible and ground using a mortar and pestle into a fine powder.
A 500mg sample of the powdered fusion product is dissolved in 25ml of 10% HNO3. After filtering the precipitated SiO2 using a medium grade filter paper, the solution is diluted to
50ml using distilled water. In some cases, a second filtration step may be required prior to dilution.
A 500mg sample of the powdered fusion product is added to 25mls of 2.5% citric acid and heated to 90ºC.
The fusion product should completely dissolve, apart from residual carbon-based powder from the crucible, which is filtered off using a medium grade filter paper.
The filtrate is transferred to a 50ml volumetric flask and made up to volume. The solutions are then analysed using standards prepared from a similar solution (10% HNO3 + H2O or 2.5% citric acid + H2O).
If more in depth analysis is required for the analysis, the standards can be matched to the sample matrix by forming a blank solution. This entails that the pellet formed in the crucible must be formed without the addition of the sample, then carried through the process as the sample had been, then utilising the “blank” option when calibrating the BWB-Tech flame photometer.
1 Gavindaraju, K. ‘Rapid Flame Photometric Determination of Sodium and Potassium in Silicate Rocks’, Appl.Spectroscopy, 20 (1966), p.302-304.