Using FTIR Spectroscopy to Analyze Gemstones

Due to the increasing popularity and high prices of quality gemstones, there is a large market for less expensive stones that resemble the real thing (Figure 1). Among the most valuable naturally occurring gemstones are rubies, sapphires, and emeralds. It is essential for industry experts to be able to distinguish a treated or synthetic stone from a real one. This article discusses how an FTIR spectroscopy technique can be used to determine the genuine article from treated or synthetic stones. 

Figure 1 - A ruby (a), sapphire (b), and assorted gemstones (c). Natural and synthetic stones cannot be easily differentiated visually.

Example 1: Rubies and sapphires

Rubies and sapphires are gem varieties of the same chemical material (corundum or aluminum oxide). FTIR spectroscopy can be used to confirm that a stone is corundum and not a simulant; however, since the infrared spectra of ruby and sapphire are virtually identical, it is not possible to easily differentiate them. In fact, synthetic sapphire is colorless and is often used as a window material in near infrared and visible optics because of its unique spectral properties.

Today, red stones are usually classified as rubies, while the other remaining colors are classified as sapphires, even though the name “sapphire” was originally reserved for blue stones. The color differences are created by the presence of trace amounts of metal ions that cannot be detected by infrared spectroscopy. The red color of rubies comes from chromium and iron, the blue color of sapphires from titanium and iron. Pink, green, and yellow stones contain different states of chromium or iron. As with many natural materials, corundum is not particularly rare, and colored forms of low-quality stones are readily available. In contrast, high-quality, natural gemstones are rare and very valuable, particularly when the stone is a nontraditional color.1

Recently, a beryllium treatment process was reported that can greatly enhance the colors of poor-quality stones. This high-temperature treatment diffuses beryllium atoms into the corundum crystal and significantly improves the appearance of the stone. Most natural rubies and sapphires have a small peak in the infrared spectrum near 3310 cm–1 that corresponds to the O–H stretching mode of water. While FTIR spectroscopy cannot detect the presence of the beryllium atoms, the high temperature of the treatment usually eliminates the water that is trapped in most natural stones and, consequently, this peak loss can be measured.2

Using FTIR spectroscopy, it is possible to verify the presence of trapped water in a ruby or sapphire. Although the amount of OH found in natural stones can vary greatly, FTIR provides a rapid, nondestructive technique to identify stones that might require further investigation.

Using FTIR spectroscopy to distinguish treated rubies and sapphires

For the purposes of this article, a Thermo Scientific Nicolet™ 6700 FTIR spectrometer with a 4× beam condenser accessory (Thermo Fisher Scientific, Madison, WI) was used to demonstrate the benefit of using the FTIR technique for the function described above. The spectrometer was equipped with a high-sensitivity mercury cadmium telluride (MCT) detector and extended-range KBr beamsplitter. The 32 scans were acquired at 4 cm–1 resolution for a measurement time of 20 sec. 

Figure 2 - Infrared spectrum from a blue sapphire.

The spectrum of a sapphire in Figure 2 shows the strong absorption from the Al–O bonds below 1500 cm–1 and a lack of other major features in the spectrum. The unique spectrum of corundum makes it very easy to differentiate ruby and sapphire from most of the simulants or other similarly colored gemstones. 

Figure 3 - Comparison of the OH spectral region for four stones.

The peak used in this example is barely visible in the 3300 cm–1 region of the spectrum in Figure 2. Figure 3 shows the scale-expanded spectra from several stones in the region from 3450 cm–1 to 3100 cm–1. On this scale, the peak of interest is clearly visible in three of the four spectra.

The lack of an OH peak in the spectrum of the small pink sapphire at the bottom of Figure 3 suggests that this might be a treated stone. While a visual confirmation is straightforward, this is an excellent application for an automated analysis technique. A classical least squares (CLS) analysis method has been used to verify if the peak is present and to determine the magnitude. A CLS method minimizes the difference between a sample spectrum and a set of reference spectra containing the components to be analyzed.

Furthermore, an important feature of the CLS technique is the calculation of a standard error term. In most cases, a concentration value that is several times larger than the reported standard error gives a very high confidence that the peak is present. In this case, the peak is 13 times larger than the calculated error, clearly confirming its presence (see inset, Figure 3).