Absorption and fluorescence not only occurs in visible range of the electromagnetic spectrum, but also in the x-ray region, although it posses very short wavelength and human eyes cannot perceive any mechanism in this region. When a substance is exposed to x-rays, as any other form of energy, the electron of the atoms get excited to the higher energy level and on falling back to the ground state, emit radiations in the form of fluorescence. The emitted light is thus, recorded by the detectors and hence, the method known as x-ray fluorescence spectrometry. The emitted energy corresponds to electron transitions of the elements under study and the fluorescence features are expressed in terms of energy, keV.
The emitted energy of excited elements is displayed as peaks, which are characteristic for a particular element; each peak can be assigned to a specific element, although some elements can have multiple peaks depending on the energy levels. While the position of the peak is dependent on the type of element, its intensity is dependent on the concentration level of that element in a sample. Hence, by correlating the intensities of the peaks, the amount or quantity of a particular element in the sample is determined.
A typical Energy Dispersive X-Ray Fluorescence (EDXRF) spectrometer consists of an excitation source i.e. x-ray tube, where anode can be of Rh, Ag, Cu, Mo, etc, a detector, which consists of a multi-channel analyzer to detect and sort fluorescence features on the basis of their energy levels, and a mathematical program i.e. software to generate the spectra.
Schematic principle of an EDXRF spectrometer
X-ray fluorescence spectrometry is used to determine the bulk chemical composition of the samples' surface. Due to the sensitivity levels of the spectrometers used, elements with atomic number 11 i.e. sodium (Na) or higher can be detected, while those falling below atm. no. 11 (such as H, C, O, B, Be, Li) cannot be detected due to their very low levels of emitted energies.
The main advantage of using EDXRF is that it provides a quick chemical composition of a sample, although care has to be taken to place the sample properly. In case, if a sample is coated with some impurities or an artificial coloured layer, true composition of the sample will not be obtained. On routine, largest polished surface i.e. table of the sample is analysed, as it provides the least scattering of the energies, and maximum energy is transmitted to the detector. However, depending on the type of the sample and information required, samples may be analysed from the pavilion or even girdle side. This becomes more important when a sample is coated from only the pavilion side, as in case of tanzanite or mystic quartz / topaz; similarly in case of glass filled diamonds or rubies, it is important to analyse the fracture, where it is breaking on the surface.
At the Gem Testing Laboratory, EDXRF system is used to:
Identify gemstone species / variety, especially in case of look-alikes such as separating natural turquoise from dyed magnesite, black spinel from black garnet, diopside from enstatite, orthoclase (moonstone) feldspar from plagioclase (sunstone / labradorite) feldspar, or albite from anorthite, types of garnets, black diamond from synthetic moissanite, and many more.
EDXRF analysis helps to identify members of minerals with an isomorphous series such as feldspar. The two main groups, K-feldspar and plagioclase (Na-Ca) feldspar can easily be differentiated on the basis of key element, potassium (K), and hence, differentiates between a moonstone and labradorite or aventurine orthoclase (red trace) and aventurine sunstone (blue trace).
Yellow green epidote (blue trace) can easily be separated from yellow green chrysoberyl (red trace) on the basis of different chemical composition.
Turquoise (blue trace) can conclusively be differentiated from magnesite (red trace) on the basis of their chemical composition.
Differentiate natural from synthetic or man-made counterparts, especially in case of rubies and sapphires, emeralds, HPHT grown diamonds, corals or turquoises from their man-made (imitation) counterparts, etc. Since, chemical impurities present in natural substances cannot be replicated exactly in a laboratory, or to increase and/or control the growth speed of synthetic materials, additional impurities are added, which can be detected by EDXRF analysis.
HPHT synthesis of diamonds require iron and nickel as catalysts for growth to take place, which can be reflected in a grown diamond, as in case of this yellow HPHT synthetic diamond (blue trace), while a natural diamond lacks these impurities (red trace); the weak peaks visible here are instrument artifacts.
Flux -grown synthetic ruby (red trace) may show the presence of lead, which is present in the flux used to grow crystals, while natural ruby (blue trace) does not have lead content, unless it is glass filled.
Natural coral (blue trace) and ceramic-imitation coral (red trace) is differentiated on the basis of presence or absence of calcium and titanium; the latter one is present in ceramic imitation.
Identify treatments, especially glass fillings in rubies, sapphires or diamonds, coatings in spinel, tanzanite, topaz or any other gem, diffusion in corundum or topaz, etc.; however, care has to be taken to place the stone properly in the measuring cup so that the filled fracture breaks on the surface or coated surface is exposed to the x-rays, especially in case of topaz or tanzanite.
Exact nature of the type of glass filling i.e. whether lead (yellow trace) or bismuth (red trace) in a ruby (blue trace) can be determined conclusively.
Coating in tanzanite can conclusively be identified by presence of cobalt (red trace), the impurity not present in a natural colour or a heated tanzanite (blue trace).
In addition to the above mentioned applications, chemical analysis using EDXRF also assists in origin determination of rubies, sapphires, emeralds or tourmaline, although the service is not provided at the Gem Testing Laboratory. The variation of chemical elements however, depends on the type of geologic setting and the parent rock and not specific to only a single geographical locality. For example, rubies from Burma, Kenya, Tanzania and Vietnam share a similar geological setting, and hence, the chemical fingerprinting also overlaps each other.