Precious metal analysis

Dippal Manchanda, technical director of Assay Office Birmingham, reviews the different techniques used for the analysis of precious metals.

Determining the precious metal content of jewellery is fundamental to the jewellery industry and, as prices soar, is increasingly important particularly for those valuing and buying and selling scrap gold or second hand jewellery over the counter.

While valuing and buying UK hallmarked jewellery is relatively simple and easy, jewellers in-store regularly use touch acid testing to determine the fineness of the un-hallmarked gold and silver items they are presented with, and to identify articles made from base metal. Such testing depends upon knowledgeable interpretation of the discolouration following a chemical reaction when the appropriate acid is applied to the metal.

Touch acids and reagents consist of nitric acids of different strengths, with or without the addition of copper, sodium chloride or hydrochloric acid. The nitric acid causes oxidisation, which manifests itself as a discolouration which can, for example, vary from shades of light brown to a dark brown/red to indicate different standards of gold. The preparation of the surface to be tested and the amount of acid used will impact on the accuracy of the result, but for an experienced operator the process is very reliable as a good indication of precious metal content. Touch acid testing is widely used by pawnbrokers and others buying second hand jewellery, and those carrying out jewellery repairs, to ensure they match the metal fineness. It is also the accepted practice for valuers assessing un-hallmarked pieces.

The trend towards silver means that there are millions of articles in the market that do not need to be hallmarked as they fall below the exemption weight of 7.78 grams. Such items need to comply with the Hallmarking Act in so far as if they are described as silver they must be at least 800 parts per thousand (ppt) silver, and if they are to be referred to as sterling they must be 925 ppt. Again, retailers can check these in-store using touch acids.

Touch acid is not appropriate for platinum group metals, and for these and for more accurate analysis laboratory equipment is required. All precious metals can now be measured fairly accurately by X-ray fluorescence. New machines are very much more accurate than in the past, and with an experienced operator good results can be achieved. To deliver reliable readings a large number of ‘standards’ are needed, against which to test the sample pieces, and machines need to be regularly recalibrated and their results compared with other machines or analytical techniques to prevent ‘drift’.

When buying or selling scrap jewellery that has been made into a bar, or when there is a need to determine precious metals in high value, complex concentrates, bullion, smelting products, recycled material, catalysts, electronic scrap and other precious metal containing materials, a variety of different methods are used. Such samples may require sophisticated new techniques, but for gold and silver the most accurate methods remain traditional cupellation and titration, which are still the referee methods.
Cupellation is essentially a refining process whereby all other elements are eradicated, to leave pure gold. Small samples of gold are weighed and wrapped in lead with a pre-determined quantity of silver to assist with the removal of the base metals when the samples are subjected to a temperature of 1,100ºC in the furnace. Wrapped samples are placed on porous blocks, known as cupels, and when in the furnace the cupels absorb the lead and any base metals as oxides. A pure gold and silver bead is left behind, which is then flattened out and rolled into a cornet. Boiling these cornets in nitric acid will dissolve the silver, leaving a residue of fine gold. When the weight of this is compared with the original sample the gold content by weight can be calculated.

Titration is used to analyse silver and is now carried out in an ‘automatic titrator’. The sample is dissolved in a specific amount of nitric acid and left to cool. Potassium chloride is then added automatically, prompting the precipitation of silver chloride. The equipment automatically calculates the silver content of the sample by measuring how much potassium chloride was required to complete the silver titration.

Platinum group metals require a completely different approach and the referee method utilises sophisticated equipment – either an Atomic Absorption Spectrometer (AAS) or an Inductively Coupled Plasma-Optical Emission Spectrometer (ICP). The ICP can identify and measure more than one metal at a time whereas the AAS can only deal with one element. The samples are dissolved in aqua regia (75 per cent hydrochloric acid, 25 per cent nitric acid), heated and then passed through the ICP.
Due to the industrial use of platinum and its very high value, it is not unusual to have to determine the platinum content of small samples of very low grade material. Analysis by ICP can give an accurate result to one part per million and is widely used to analyse complex materials