Natural colored diamonds are so varied and rare they have defied efforts to conduct extensive studies of their properties.
Thanks to the Aurora Butterfly of Peace, a collection of 240 diamonds of various colors (figure 1), a team of researchers had a unique opportunity to conduct such a study when the Butterfly was on exhibit at the Smithsonian Institution. Specifically, these researchers examined fluorescence trends of natural fancy-colored diamonds using a new portable fluorescence spectrometer.
The colored diamonds in the Aurora Butterfly of Peace (166.94 carats total weight), assembled by Alan Bronstein and Harry Rodman of Aurora Gems Inc., New York, demonstrate nearly the full spectrum of color and cut styles available in natural colored diamonds. The collection is shown here in standard daylight (left) and long-wave UV radiation (right). Photos by Robert Weldon.
Their findings, published in a Winter 2007 Gems & Gemology article titled “Fluorescence Spectra of Colored Diamonds Using a Rapid, Mobile Spectrometer,” could help gemologists more easily identify treated-color diamonds.
Researchers Sally Eaton-Magaña, Jeffrey Post, Peter Heaney, Roy Walters, Christopher Breeding and James Butler examined 48 diamonds in the Butterfly collection, 2 fancy-color diamonds housed by the Smithsonian (the DeYoung Red and the DeYoung Pink) and 22 samples from GIA. They also looked at 10 irradiated colored diamonds.
The team selected diamonds in a range of bodycolors that included pink, yellow, yellow-green, orange, brown and blue-gray, as well as rare specimens such as fancy white, purple and chameleon. For the most part, they limited their picks to those that showed fluorescence to long-wave ultraviolet (UV) radiation.
After testing, most of the diamonds were grouped into three categories, according to shared characteristics in their fluorescence spectra (figure 2). The spectra of category 1 diamonds generally appeared as bluish fluorescence. This category encompassed most of the pinks, the fancy white and several yellow natural-color diamonds. By contrast, the irradiated diamonds in this category had green to greenish blue bodycolors.
Most of the natural-color diamonds tested could be grouped into three categories based on their fluorescence spectra. Three diamonds with particularly rare colors were inert, and two had a different type of fluorescence spectrum.
The majority of diamonds in category 2, which included the yellow-green and brown stones, had green or yellowish green fluorescence. In most cases, these natural, untreated diamonds had weak to moderate fluorescence spectra, whereas two greenish yellow irradiated samples showed strong to very strong spectra.
The UV fluorescence of the stones in category 3, which contained most of the orange and gray-green chameleon diamonds, generally appeared as yellow or yellow-orange. None of the irradiated diamonds fell in this category, but a broader study of treated diamonds would be needed to confirm this as a trend.
With a few exceptions, the fluorescence of natural diamonds in the three main groups corresponded to their bodycolor. However, the irradiated samples in the first two categories showed either significantly different bodycolors or different spectral intensities. These types of indicators may prove useful in separating natural from treated diamonds.
While not conclusive, these spectra may also provide clues to the origins of the color in these diamonds, which would help guide future studies. For example, the origins of yellow (nitrogen), blue (boron) and green (radiation) bodycolor are well known, but researchers are still trying to pin down what creates the color in pinks, browns and others.
These dramatically different fluorescence colors were produced by minor differences in the diamonds’ fluorescence spectra. The similarity of the spectra suggests that this range of fluorescence colors is caused by the same atomic-level defect. Photo by Shane Elen.
The study also demonstrated that fluorescence spectra are more accurate than fluorescence seen with a UV lamp, which can produce different results for different observers, especially given the nonstandard lighting environments in which fluorescence is often viewed. The fluorescence spectra measured by a spectrometer can be quite similar, even if the fluorescence colors appear different to the eye (figure 3). In addition, fluorescence spectra provide far more information than visible fluorescence alone, such as a semiquantitative indication of intensity and the presence of subordinate peaks.
One objective of the research was to evaluate a relatively low-cost, compact fluorescence spectrometer produced by Ocean Optics. This instrument, the size of a deck of cards, enabled the researchers to study diamonds that could not be removed to a laboratory. Its mobility and rapid collection time will make it easier for future research on important gemstones held in museums and private collections, and could prove useful in a commercial setting.
Fluorescence spectroscopy can help characterize large numbers of important diamonds, such as these 0.51–2.03 ct stones from the October 2007 Argyle collection. Courtesy of Argyle Diamond; photo by Robert Weldon.
From this research, the authors also found that no two diamonds showed identical fluorescence spectra; they all demonstrated slight differences in peak intensity or relative intensity between peaks. These subtleties are part of what makes each stone unique (figure 4). Everything about a diamond—from clarity features to the atomic-scale defects that cause fluorescence—tells that diamond’s story. Its fluorescence spectrum is one more chapter in that story, one more tool by which individual diamonds may be characterized.