Analytical Instrumentation

Infrared Spectroscopy
Transmission infrared data were obtained using a Thermo-Nicolet 6700 Fourier-transform infrared spectrometer (FTIR) equipped with a KBr beamsplitter and MCT-B detector. Spectra were collected in the mid-infrared range (400–6000 cm-1; 1 to 4 cm-1 resolution) with a number of scans to improve signal-noise ratios (between 128 and 512). A 6x beam condenser was used to focus the incident beam through the pavilion and girdle region (a direction parallel to the table facet), or in other sample orientations, to obtain the maximum signal. Taking into account the measured minimum path length, the y-axis of each spectrum is presented in absorption coefficient (cm-1). The sample chamber was purged with dry air. The spectra were recorded at room temperature. The samples were cleaned in alcohol to remove any surface contamination. In many instances, absorption at certain wavenumber ranges of the infrared spectrum exceeded the measurement limits of the detector—these portions of the spectra possessed no useful information, and were replaced by horizontal straight lines in the spectra diagrams.

Visible Spectroscopy
Absorption spectra in the ultraviolet to near-infrared range (250–1000 nm) were collected using a Hitachi U4001 spectrometer. Instrument operating conditions included a slit width of 2.0 nm, data collection interval of 1.0 nm, and scan speed of 120 nm/min. Spectra were collected at room temperature using an unpolarized light beam unless otherwise noted. The incident light beam was focused on the culet area in a direction perpendicular to the table facet, which was positioned against the instrument detector for maximum light collection. The spectra were presented over the 350–750 nm range. In instances where the absorbance exceeded the absorption range of the instrument in the region below ~400 nm, the spectra were truncated using a horizontal line on the spectrum plot. Polarization artifacts related to the instrument have also been removed from the spectra.

The absorption spectra were plotted with wavelength (nm) along the horizontal axis and absorption coefficient (cm-1) along the vertical axis. Although it is very difficult to determine the exact path length of light through the faceted samples, the consistent orientation of the samples during analysis allowed a minimum path length to be estimated (equivalent to the depth measurement of the stone) and the absorption coefficient values to be calculated.

The testing of gemstones requires that analytical work be conducted with minimal or no damage to the sample. Gemstones comprise a wide range of sample size, shape and facet arrangement, color, and degree of transparency. In rare instances, there is a relationship between the shape and facet arrangement of the gemstone and the optical directions of the original crystal. Frequently, however, there is no such relationship because:

a) faceting is undertaken to maximize the size of the cut stone relative to that of the original crystal to retain as much weight as possible, and/or to reveal color or some other desirable appearance (such as asterism or chatoyancy), so there may be little effort made to orient the cut stone with respect to the optical directions of the material, and

b) light travels through a faceted gemstone in a complicated path that is affected by several factors, such as indices of refraction, optical orientation of the material, dispersion, angle of light incidence, surface reflections, wavelength composition of the incident light, sample size, and facet size and arrangement. In cases where the optical axes can be reliably determined, we have illustrated their positions for reference.

For these reasons, it is often difficult to position a faceted gemstone in an analytical instrument to record an optically oriented spectrum. Therefore, unless noted otherwise, the spectra presented in this database were recorded in a random optical orientation for the sample. Spectra recorded in this way correspond more closely to the situation faced by gemologists who must examine gemstones with a desk-model spectroscope, and who are unable for the reasons mentioned above to obtain optically oriented spectra.

The visible spectra are plotted with an expanded vertical scale so that absorption lines and bands could more easily be seen. Because of this procedure, weaker absorption features that were plotted more prominently in the visible spectra could not always be seen visually with the desk-model spectroscope.

Raman Spectroscopy
Raman spectra were collected using a Renishaw InVia Raman microscope with sample excitation produced by an Argon-ion laser operating at a wavelength of 514.5 nm. Three scans were collected and summed over the 100–2000 cm-1 Raman shift to provide good signal-noise ratios. Analysis was done at room temperature using a focused beam on the table facet of a randomly oriented sample, unless otherwise noted. In a few rare instances, certain gemstones fluoresced too strongly under 514.5 nm laser excitation to collect a spectrum. In these cases, a 633 nm helium-neon laser was used.

The Raman spectra were plotted with Raman shift in inverse centimeters (cm-1) along the horizontal axis, and counts along the vertical axis, and the spectra have been baseline-corrected to better show the peaks.

Photoluminescence Spectroscopy
Photoluminescence spectra were collected using a Renishaw InVia Raman microscope with sample excitation produced by an Argon-ion laser operating at a wavelength of 514.5 nm. A single scan was collected over the 517–1000 nm range. Analysis was done at room temperature using a focused beam on the table facet of a randomly oriented sample, unless otherwise noted.

EDXRF Chemical Analysis
Qualitative energy-dispersive X-ray fluorescence (EDXRF) data were collected using a Thermo ARL QuantíX EDXRF analyzer. This method can detect elements above sodium (Na) in the periodic table. Operating conditions were varied depending on the gemstone and the elements of interest for analysis, but the following were representative: voltage 6–35 kV, current 0.02–1.98 mA, 100 second live time, vacuum atmosphere, and any of several filters including no filter, aluminum, and palladium (thin). The data presented for most stones was collected with an aluminum filter at 15 kV, and X-ray fluorescence peaks for the significant elements have been labeled. Samples were oriented to minimize the occurrence of diffraction peaks in the spectra. Peaks in the spectra that are due to particular elements are labeled with the element symbol (such as Fe for iron). Artifact peaks which may be prominent in the XRF spectra of some samples are not labeled. Depending on the excitation conditions chosen, the X-ray fluorescence of different elements in a material will vary in intensity. For this reason, the height and width of fluorescence peaks in the spectrum are not directly related to the concentration of elements in the sample. Furthermore, lighter-atomic-weight elements (below sodium) do not produce strong X-ray fluorescence, and they cannot be detected by this technique with our equipment.