X-ray fluorescence testing

Assay Office Birmingham has had to adapt to changes in the market throughout its history, and particularly in the last two decades. A prime example is the way that items are ‘sampled’ and tested, or ‘assayed’, to check the fineness of all components of the article prior to hallmarking. Technical director Dippal Manchanda explains how this process has changed.

Until the latter part of the 20th century the majority of items submitted for hallmarking were presented unfinished, and the removal of a minute sample of metal for assaying did not create too much of a problem as the customer had yet to ‘finish’ the item. However, the steady increase in imported items and the decline in the volume UK jewellery manufacturing trade in the late 1990s resulted in an influx of finished, highly polished goods from overseas for hallmarking. This presented the UK assay offices with an unprecedented challenge. Suddenly hallmarking parcels consisted predominately of complete, finished products, destined to be shipped straight to the retailer or importer, as opposed to semi-finished items awaiting polishing and finishing by the UK manufacturer.

Up to this point all sampling and assaying was carried out by physically scraping off tiny amounts of metal from the article and then testing it using one of the traditional referee methods: cupellation for gold, potentiometric titration for silver, and ICP/AAS based technique for platinum. X-ray fluorescence (XRF) testing was already being used within the jewellery industry to determine plating thickness, and this was identified as a possible alternative method, more suited to the new demands of the market. Recognising the need to develop more appropriate techniques, the assay offices sought expertise from the XRF manufacturers and worked closely with them to adapt the software driving the XRF technology, and therefore to create equipment to meet their needs. The criteria were that the results must be accurate, repeatable, consistent and userfriendly for their staff. In order to deliver this the XRF unit had to be able to accurately identify 100 per cent of elements of interest.

This was successfully delivered, and since the beginning of the 21st century XRF has become the standard recognised for assaying for hallmarking. As well as being conducted without damage to the product, assaying by XRF means that an item can quickly be tested in several places to check that solders and fastenings are compliant with the Hallmarking Act. XRF also has the benefit of being more environmentally friendly than the traditional cupellation method, which requires the use of lead and energy-hungry melting furnaces.

To explain the XRF process in simple terms, the machine emits a concentrated beam of X-rays onto a focused spot. The size of this target is adjusted by the operator, who has different sizes to select from, always aiming to choose the largest possible ‘spot size’ where the item has a flat surface to receive the beams evenly.

The X-rays penetrate the surface of the material and ‘excite’ the atoms in the alloy. The intensity of the energy that these atoms emit is then analysed and collated automatically to provide a reading as to the percentage of each element in the item. If used and calibrated correctly by an experienced operator, XRF can produce extremely accurate results. Accuracy is maintained by continual calibration against expensive reference standards and subsequent adjustment of the reading. The machines used by samplers at Assay Office Birmingham  are programmed to deliver confirmation that the article tested complies with a given test standard – for example nine carat yellow gold. The same technology with a stronger electric current and a longer exposure to the beams is used by the Laboratory to identify every element of interest in complicated and unusual alloys. In this case a highly trained technician will use his or her expertise to interpret the results from a graph and challenge every irregularity to ensure nothing has escaped scrutiny.

However sophisticated the XRF equipment, it will never deliver a true result for a heavily plated item – rhodium plating on white gold being a classic example that continually raises queries. In these cases some scraping of the plating may be required and operators need to be well aware of the various scenarios which could result in a grossly inaccurate result. At the Assay Office, the sampler may occasionally have to resort to the traditional skills of touch acid testing and the referee methods of cupellation and titration to ensure he or she obtains a realistic result.

In recent years smaller, hand held XRF machines have been launched onto the UK market and many jewellers are starting to use these to assess second hand jewellery. These will provide a useful guide to the approximate composition of an article but should always be used with caution, as they are difficult to calibrate, resulting in a wide ‘drift’ of results.

Particular care should be taken when assessing items with curved surfaces which do not make equal contact with the ‘window’ of the machine. Such equipment is not yet sufficiently accurate to deserve a place in the strict process necessary for the UK’s negative tolerance hallmarking regime. However, the development of XRF and the software associated with it continues to move on and it certainly has not realised its full potential for the jewellery industry just yet.