Rough diamonds

Diamond fingerprinting techniques should make it easier to enforce new trade controls on diamonds. In turn, this should help to curb illicit trading in the gemstones known to be behind the funding of several civil wars, writes Richard Stevenson

Diamonds are a warlord’s best friend. Several vicious civil wars of recent years - in Angola, Sierra Leone and the Democratic Republic of Congo (DRC) - have been funded with diamonds smuggled from mines in the war zone. A small packet of uncut diamonds will buy a lot of Kalashnikovs.

Non-governmental organisations (NGOs) such as the UK-based group Global Witness have campaigned against this trade in ’conflict diamonds’, and a UN report identified diamonds, along with cobalt and tantalum, as DRC mineral resources that were being systematically looted to pay for arms and to enrich politicians and soldiers. Terrorist groups, including al-Qa’eda, as well as international drug smuggling gangs are believed to be using uncut diamonds to move their assets around the world, while organised crime has always targeted diamonds for their intrinsic value and portability. Intelligence services, law enforcement agencies such as the Royal Canadian Mounted Police (RCMP), and customs officials around the world would like to be able to track and prevent this trade.

In December 2000, the UN passed a resolution to break the link between diamonds and war, and gave a mandate to a South African inspired initiative called the Kimberley process. Under this agreement, legitimate traders will only buy rough diamonds if they are accompanied by a certificate of origin. This initiative was supported by the diamond trade, which had been hurt by media and NGO allegations of involvement in murky trafficking. The Kimberley process has been adopted by the EU and the UK introduced the appropriate import and export controls on 10 February this year. The Foreign Office has set up a Government Diamond Office to oversee the UK’s commitments to the Kimberley Process. Draft legislation has also been introduced into Congress to bring the US into the process. Following the ending of the civil war in Angola, the Angolan government has reformed its diamond industry to be compliant with the Kimberley process and in February it restarted diamond exports. Having cleaned up its industry, Angola hopes to attract investment from De Beers, the world’s largest diamond trader.

But certificates aside, how can you tell where a packet of uncut diamonds really came from? Unlike coloured gemstones such as sapphire - where, as Harry Levy of the British Jewellers’ Association says, an experienced gemologist can instantly recognise, say, the distinctive cornflower blue of a Kashmiri stone - diamonds are largely colourless. The pale pinks, yellows and blues that sometimes occur are due to small amounts of nitrogen, hydrogen or even boron incorporated into the carbon lattice. These colours are not diagnostic and they can be manipulated using radiation.

Although some experts claim to be able to recognise stones from a particular source due to the range of colours, crystal forms and surface features, this subjective technique is not foolproof, especially when different parcels of stones are mixed. Mineral inclusions are not diagnostic either. Jeff Harris of Glasgow University, a consultant to both the UK Foreign Office and to the biggest diamond trader, De Beers, points out that ’diamond forms in a relatively restricted environment within the Earth’. This means that ’the same minerals occur in diamonds worldwide and, in consequence, can never guarantee source uniquely’. The trade needs a chemical fingerprinting technique that could discriminate between diamonds from different mines on the basis of their trace element chemistry.

Until now, most research on diamond chemistry has been to understand its geological origins, to improve synthetic diamond films for electronics, or to spot fake gemstones such as cubic zirconia. Recent conferences have brought together experts from a range of disciplines to assess the available analytical techniques, for example at the White House in January 2001, and at the May 2002 meeting of the American Geophysical Union. Understandably, the experts concentrated on techniques that mineralogists already understand and use.

Cathodoluminescence (CL) is one such technique. CL is light emitted in the wavelength range 160-2000nm when a non-metallic crystal is bombarded with electrons. This can be in an electron microscope or under an optical microscope with an electron gun attachment. CL is associated with lower energy transitions than X-ray emission, so it cannot be used as an elemental analysis technique. Instead, CL is governed by physical chemistry, optoelectronic properties, crystal structure, lattice strain etc. CL is popular with mineralogists because it can reveal zoning and overgrowths within a single crystal and thus reflect its crystallisation history. Diamonds from a single mine should share a crystallisation history and therefore show similar zonation.

Secondary ion mass spectrometry (SIMS) is another promising technique. A focused ion beam sputters secondary ions from the sample surface and these can be used to measure isotopic and trace element abundances. SIMS has been used, for example, to examine sulphide inclusions in diamond crystals. Proton-induced X-ray emission spectroscopy (PIXE) is an established technique for research on solid and fluid inclusions in minerals and has the advantage of being non-destructive (traditional research on fluid inclusions relied on crushing the diamond and analysing the gases by mass spectrometry).

Other groups are analysing diamond surfaces to see if they retain a geochemical signal from their surrounding minerals. George Rossman of the California Institute of Technology has taken this further by heating diamond samples to measure the relative abundance of hydrogen and deuterium boiled off. The D:H ratio in rainwater varies with latitude, and Rossman has reported the signature of subtropical water adsorbed on a diamond from Botswana. This could not be confused with rain from Cape Province, for example. Rossman believes that although this isotopic analysis technique is relatively sophisticated it has the potential to be made into a tabletop mass spectrometer similar to those already developed by the Pentagon. However, surface techniques are dependent on the degree to which the diamonds have been cleaned, and criminal gangs would soon learn to disguise their merchandise.

Many experts believe that the best hope lies with laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). A key paper on applying this technique to diamonds was published in 1995 by John Watling and his then colleagues at the Chemistry Centre (WA) in East Perth, Western Australia. In this process a laser blasts a microscopic pit in the sample and the resulting plasma of ions is passed into a mass spectrometer. Watling and his colleagues determined 43 trace elements in their samples. They found, for example, that tungsten was a good marker in their Russian sample, whereas lead, molybdenum and rubidium were discriminatory for their South African samples. The advantage of the technique is that it only needs to be semi-quantitative: it is the relative proportions of these minor trace elements that provide the fingerprint. This should make it relatively easy to standardise results from labs around the world.

Writing in 1995, Watling and his colleagues thought that their method would be useful for law enforcement agencies tracking stolen diamonds. The interest in conflict diamonds has provided additional impetus, and a number of groups are now evaluating LA-ICP-MS. The Belgian Diamond High Council (Hoge Raad voor Diamant, HRD), based in the diamond centre of Antwerp, has been providing technical input to the Kimberley process. Since 1999 it has been assessing LA-ICP-MS in a joint project with the Catholic University in Louvain, Belgium, but does not expect results before 2005. Other centres collaborating with HRD on LA-ICP-MS include the Australian National University in Canberra, the Central Forensic Laboratory of the Royal Canadian Mounted Police in Ottawa, and De Beers’ Geoscience Centre in Johannesburg, South Africa.

De Beers is refining the technique by using CL to determine which parts of a diamond to sample. The problems still to be overcome are the cost of the equipment - reported to be about ?2m - and the need to develop a database of analyses of diamonds of known provenance. At the end of the day, however, the problem is one of capacity. Even with only 2 per cent of world supply coming from war zones now that the wars in Angola and Sierra Leone are over, Harris points out that this still represents 2.5 million individual stones. ’Time is never on the good guys’ side’ he says. ’LA-ICP-MS is a fine technique’, but he estimates that it would take such a machine 39 years to analyse that many stones. There is an open market opportunity for new and faster analytical methods.

Source: Chemistry in Britain

Acknowledgements

Richard Stevenson