Cathodoluminescence (CL) is a method found in a wide range of geological applications. These applications include igneous petrology, sedimentary petrology, crystal growth, metamorphic petrology as well as gemstone identification [e.g., 1-11]. This study’s focus concerns the use of CL as applied to minerals – specifically color-zoned sapphires.
CL is known to be frequently used on zircons to obtain zircon growth crystallization . Thanks to recent CL advancement and color zonation in some sapphires, this has inspired the evaluation of using CL on sapphires.
Cathodoluminescence is one of the more widely known methods for studying minerals. CL emission is created after the interaction between the electron beam and a luminescent material, which produces electromagnetic radiation in the spectral range, from ultraviolet to infrared. To be specific, cathodoluminescence occurs when the crystal absorbs energy from the electron beam and reaches an excited state, after which this excitation energy is (partially) emitted as light as the crystal returns to the unexcited ground state [e.g., 3, 10, 13, 14].
CL can be applied to a wide range of minerals and is used both as a microscopic and a spectroscopic method. As stated, CL is applicable for a wide variety of uses, including imaging cement and detrital material in sedimentary rocks (e.g., limestone/sandstone), as well as the visualization of crystallization histories and subsequent zonation patterns in minerals, such as zircon [6, 8, 15, 16, 17].
Though using CL for natural sapphires is not a common occurrence, it has been used in the study of synthetic sapphires and their growth processes [e.g., 1]. Natural sapphire, otherwise known as the gem variety of the mineral corundum (Al2 O3) is a highly prized precious gemstone. The reasons for this are multiple: it is the second-hardest naturally occurring mineral (Moh’s hardness scale = 9, after diamond), relatively dense (~4 g/cm3), and very chemically inert. Geological research focusses on the natural processes behind the sapphire formation, given that sapphires require slow crystal growth in a formation environment that is both rich in oxygen (O) and aluminum (Al) while simultaneously depleted in silicon (Si).
Of interest is the geological processes which can result in these environments. In natural sapphires, color zonation is a common feature which may hold the key to understanding potential changes in growth environments. To examine zonation and trace element defects, as a way of understanding the conditions of sapphire formation, this study combines both laser ablation inductively coupled Plasma mass spectrometry (LA-ICP-MS) with hyperspectral CL of natural sapphires.
Figure 1. Petrographic images of the samples (10x objectives) in A) plane-polarized light, and B) reflected light conditions. The numbering for the sapphires (in circles) as well as the different ROIs (not circled) are shown in the image.
Sapphires and rubies are members of the trigonal system and possess a crystal structure made up of alternating layers of O2- anions and Al3+ cations . Trace amounts of other elements substitute for Al3+ (Al2 O3 usually 97–99 wt%); frequent substituting elements are transition metals such as Fe2+/3+, Cr3+, Ti3+/4+, V3+, Mn2+/3+, in addition to Mg2+ and Ga3+ . Sapphires and rubies only differ in their color and trace element content: rubies are red due to Cr3+ substitution, and sapphires can be any other color, though most frequently they are blue due to intervalence charge transfer between Fe2+ and Ti4+ .
Most frequently, CL has been applied to synthetic sapphires and rubies rather than natural crystals. Significantly, Chapoulie et al.,  observed that the CL emission spectrum is directly linked to crystal lattice defects (in the local structure) in industrial sapphires. CL studies have proven that that Fe2+ acts as a quenching center, Ti3+ is a near-infrared emitting center (broad peak between 728 and 794 nm), and Ti4+ is a blue-emitting center (broad peak at 470 nm; ).
Spectral CL studies of synthetic rubies observed narrow, well-defined spectral peaks of the well-known Cr3+ substitution at ~694 nm [9, 10]. Both the Cr3+ and Ti3+ centers have the potential to be incredibly efficient light emitters, and, given this quality, they are employed in Ti: sapphire and ruby laser systems respectively [21, 22].
F centers are created when O2- anions are taken from the crystal lattice of corundum, wherein removal of O2- creates an F center with typical spectral peaks at 415 nm, and removal of O- produces an F+- center with a peak at 325 nm .
CL imaging is commonly used in the visualizations of growth zones of zircon (ZrSiO4) [24-27], a mineral frequently found in sapphires as inclusion, or found as megacrysts co-occurring in alluvial gemfields with sapphires. Though both sapphires and rubies have distinct zonation patterns, very few studies use CL to analyze crystal zonation quantitively.
There exist very few CL studies of natural sapphires and rubies which demonstrate zonation [5, 25, 28, 29]. Oscillatory zoning can be observed in hand specimens, in some Scottish sapphires. It has been suggested by Upton et al.  that CL in these natural sapphires is quenched by both Fe and Ti. In Burmese rubies, CL imaging has identified skeletal growth features , whereas subtle zonation features parallel to externa prismatic faces have been identified in Moguk rubies . CL studies before this have largely employed panchromatic CL imaging, instead of wavelength-sensitive hyperspectral CL.
As we have stated, the results obtained from studies of synthetic and industrial sapphires and rubies suggest that Fe is likely associated with quenching effects. Given that Fe is a minor element that comprises 0.1 - 1 wt%, the potential exists for this to (strongly) lessen the observed CL signal. However, the CL signal can be boosted by enlarging the beam current and primary electron energy (which is controlled by the acceleration voltage). This analysis pairs spectral CL with LA-ICP-MS to quantitatively analyze patterns of natural sapphire zonation. Insight into crystallization history, and thus formation processes and conditions, can be obtained by understanding zonation in sapphire.
Samples and Methods
Sample Description and Preparation
Alluvial gemfields in Eastern Australia provided the six sapphires for study. As Figure 1 displays, these samples include both uniformly colored and color-zoned crystals. The sapphires which are zoned include Sample 2 (from Anakie), which has zonation from brown to colorless, and Sample 3 (from Weldborough), which displays zonation from blue to colorless. As controls, three single-colored sapphires were selected.
As seen, Sample 1 is a light pink sapphire from the St. Arnaud district. Sample 5, from Glen Innes, is a light blue sapphire and, from the Toombullup district. Sample 6 is a light blue sapphire. Lastly, for comparative purposes, Sample 4 is from Inverell and is a dark-colored opaque sapphire.
After cleaning, the sapphires were mounted in a 15 mm-diameter resin mount and then ground until cross-sections of crystal faces were exposed. Next, the resin mounts were polished to obtain a 0.25 µm diamond solution finish.
A SPARC CL system (Delmic B.V.) was used to obtain the CL measurements, which was mounted on a Thermo Fisher Scientific/FEI FEG scanning electron microscope (SEM). For each region of interest (ROI), a map of 40 x 40 pixels was collected in a 20 x 20 µm2 area. Using an aluminum paraboloid mirror, CL was collected. Using a motorized micropositioning stage, the paraboloid was aligned with respect to the electron beam.
At every ROI, the optical alignment was checked to confirm that the CL intensities could be compared between ROIs. Hyperspectral measurements were performed using a Czerny-Turner spectrograph with a 300 lines/mm grating blazed for a wavelength of 800 nm and a deep-depletion CCD array (1340 x 100 pixels). The experiment was undertaken using a combination of a dwell time of 80 ms per pixel, 10 kV acceleration voltage, and a 400 pA beam current. For an improved signal-to-noise ratio, the camera was fully binned in the vertical direction and four times in the horizontal direction.
Figure 2. An example of the CL spectrum fitted with two Gaussian line shapes (indicated by the dashed lines) using a least-squares optimization routine to determine peak amplitudes, peak positions, and peak widths (full width at half maximum, FWHM). Sample 6 was used in this example.
Based on measurements and calculations for a single-crystal aluminum substrate , the spectra were corrected for the relative system response of the CL system. Because it is only a relative correction, the absolute scaling of the y-axes in the spectra becomes rather arbitrary due to this correction. Nonetheless, the measurements can still undergo a direct comparison from one to the other (from ROI to ROI and from sapphire to sapphire), and the relative peak amplitudes should be representative for the actual CL emission as well.
By fitting the blue and red peaks with Gaussian line shapes using a least-squares optimization routine (see Fig. 2 for an example), the peak amplitudes, peak positions, and peak widths (full width at half maximum, FWHM) were obtained. Three Gaussians were used for fitting in spectra where the peak associated with Cr3+ was present. The spectra were corrected with the appropriate Jacobian conversion factor so that the spectra could be fitted in the energy domain .
To measure local trace element concentrations in the sapphires, LA-ICP-MS was used. To directly correlate trace element and spectral data, spot locations for CL and LA-ICP-MS analyses were the same. At the Queensland University of Technology, following a revised methodology from Wong et al., trace element analyses were performed on an LA-ICP-MS. . A new Wave Excimer laser ablation system determined the concentrations of elements (wavelength = 193 nm) using He gas flow (0.4 L/min), coupled with an Agilent 8800 triple quadrupole ICP–MS via a junction with 1.0 L/min Ar flow in Teflon tubing. The instrument was run out in “single-quad” mode, and the reaction cell was idle.
By using Al2 O3 as an internal standard and assuming that Al2 O3 = 99 wt% based on the average Al2 O3 -corrected composition derived from electron probe microanalysis (EPMA) , elemental concentrations were determined. The ablation spot’s diameter was 85 μm, ablating at 6 Hz over 30 seconds laser-on time, and with a 30-second laser-off period that allowed the signal to return to background levels.
In Figure 1b, the laser ablation spots are visible in the reflection image. The masses measured include 9Be, 11B, 24Mg, 44Ca, 45Sc, 47Ti, 51V, 52Cr, 57Fe, 60Ni, 63Cu, 66Zn, 71Ga, 118Sn, and 208Pb. Using NIST 612 as the external standard, concentrations were calculated from raw counts. Using Iolite , a software package based on the calculation style of Longerich et al., data was processed  for the determination of individual concentrations and associated detection limits.
This study used the same reference materials and internal standards as in the work of Wong et al. . Two of the six samples were zoned (Samples 2 and 3), three samples were transparent and single-colored (Samples 1, 5, and 6), and just one dark-colored and opaque (Sample 4).
One spot was selected in the center of the crystal for the single-color samples (Samples 1, 4, 5, and 6). For zoned crystals, multiple spots were analyzed. For Sample 2, a fragmented hexagonal sample with good crystal shape, five spots were selected. At the rim of this sample, four spots were analyzed, and at the core, one was analyzed. Sample 3, which lacked good crystal shape and was rounded, only necessitated the selection of four spots, which spanned different shades of blue.
Trace Element Compositions
The trace element concentrations which were detectable were of B, Mg, Ti, V, Fe, Ga, and Sn, whereas Be, Ca, Sc, Cr, Ni, Cu, and Zn had low concentrations or were below detection limits (bdl).
The highest concentrations were noted of Fe (3160–9580 ppm) and Ti (31.7–3920 ppm). Intermediate concentrations were observed of Ga (53.5–550.3 ppm), V (4.7–92 ppm), Mg (2.48–88.8 ppm), B (44.3–78.1 ppm), and Sn (0.157–39.2 ppm). Be (bdl–14.2), Ca (bdl–29), Sc (bdl–0.59), Cr (bdl–2.6; with the exception of Sample 1 (pink sapphire; 436 ppm Cr), Ni (bdl– 4.11), Cu (bdl–1.18), Zn (bdl–5.39), and Pb (bdl–0.5) were found in low concentrations.
The detection limits are as per the following: Be (0.26), B (9.8), Mg (126), Ca (22.9), Sc (0.10), Ti (0.82), V (0.05), Cr (0.75), Fe (9.2), Ni (0.28), Cu (0.47), Zn (0.65), Ga (0.07), Sn (0.084), and Pb (bdl– 0.05).
The single-colored sapphires have varying spectral emission features, as seen in Fig. 3.
Sample 1 is a light pink sapphire that, as shown in Figure 3, has a prominent narrow spike at ~694 nm. This is related to Cr3+ substitution in the crystal lattice, and the accompanying LA-ICP-MS analysis spot indeed has a Cr concentration of 436 ppm. Sample 4, an opaque dark sapphire, is dominated by a broad peak at ~775 nm (Fig. 3). In general, this peak relates to Ti3+ substitution and complicated processes of the energy transition in the crystal field at higher energy levels, related to fitting Ti3+ into the crystal lattice and resulting in distortion .
Samples 5 and 6 are light blue transparent sapphires, which have two peaks in the spectrum with different CL intensities. Both have one peak in the blue (446 nm and 475 nm) and one in the red (both 775 nm; Fig 3), but with different CL intensity between blue and red emissions. In Sample 5, we see low CL intensity of the blue peak (~446 nm), which is in contrast to a much higher CL intensity in the red peak (~775 nm).
On the other hand, Grain 6 has much less contrast of CL intensity between its blue peak (475 nm) and red peak (775 nm). It is important to note that in terms of absolute CL intensity, Samples 1 and 4 are significantly brighter (max amplitude of ~3000 versus ~ 50).
In Samples 2 and 3 (color-zoned crystals), we saw a variation in CL intensity influenced by the location of the spot analysis (Figs. 4 and 5). Across the hexagonal face of the sapphire, Sample 2 is color-zoned (Fig. 4). Generally, spectra from the five spots show a similar shape, which presents as a broad peak with a narrow spike in some spectra (~694 nm), again corresponding to the presence of Cr3+.
In the ‘colorless’ section of the crystal, analysis of spot 1 has the most prominent spike and the highest CL intensity. Sample 3, zoned from blue to colorless, shows two broad peaks: namely, a low CL intensity broad peak around 470 nm, and a second broad peak with high CL intensity around 742 nm (Fig. 5).
Figure 3. CL spectra for Sample 1 (top left), Sample 4 (bottom left), Sample 5 (top right), and Sample 6 (bottom right)
Figure 4. System response corrected CL spectra for different positions on Sample 2. The corresponding petrographic images are shown as inset.
Figure 5. CL spectra for Sample 3. The corresponding petrographic images are shown as inset.
Based on the combined spectral and trace element compositions of natural sapphires, the chief ions involved in CL of sapphire are Fe2+, Ti3+, Ti4+, and Cr3+, ions which are also found in synthetic, industrial, and natural sapphires and rubies. Variation in CL intensity was shown in CL spectra that correlate with trace element compositions determined by LA-ICP-MS. As well as variations in CL emission, Fe and Ti concentrations are responsible for the color zonation in sapphires.
Differences in Fe and Ti concentrations are responsible for the differences in the peak amplitude for the blue emission in the two single-colored transparent light blue sapphires. Sample 5’s higher Ti and Fe concentrations (Table 1), may have meant that the 446 nm blue peak might have shifted due to the contribution of more F centers . Thanks to the blue peak’s lower CL intensity (Ti4+), this naturally results in a proportion of Ti3+ in the total Ti concentration, which causes the bimodal CL intensity distribution as seen in Sample 5. On the other hand, Sample 6 has both lower Fe and Ti concentrations in general, and a more even distribution between Ti3+ and Ti4+ .
Figure 6. Peak amplitude of CL peaks as extracted from the numerical fits against Fe (A and B) and Ti (C and D) concentrations. Top figures A and C use blue peaks, and bottom figures B and D use red peaks. Note that the number of counts (cts) is larger than in the spectra due to the Jacobian conversion factor which converts the data from units of counts per nm to counts per eV.
Figure 7. Ti and Fe concentrations of the six samples.
It seems clear that there is a defined positive trend between Ti concentrations and CL emission, in addition to a negative correlation between Fe and CL emission (Fig. 6). Despite this, however, no clear correlation exists between concentrations of Fe and Ti (Fig. 7).
To determine if the rate of change between Fe and Ti affected the CL spectra instead, Fe/Ti ratios were calculated. There is a negative correlation between Fe/Ti and Ti concentration (Fig. 8), which means two conclusions can be drawn in regards to the presence of Fe, Ti3+, and Ti4+ in the corundum crystal lattice (Fig. 8). The first of these is that the distribution of Ti in sapphires is in the dominant form of Ti3+ when Fe/Ti is low, and subsequently, as Fe/Ti increases, the ratio of Ti4+ relative to Ti3+ increases.
Visually, this is demonstrated as the spectral content shifts from a more dominant peak to two peaks with a similar amplitude as ratios between Fe and Ti change from low Fe/Ti to high Fe/Ti (Fig. 8). Secondly, the CL intensity decreases as Fe/Ti increases, due to the presence of Fe and the associated quenching effects (this is depicted in Fig 8B.; examples of Sample 4 on the left and Sample 2 on the right are shown, corresponding to the same samples in Fig. 8A).
The presence of Cr3+ in their crystal lattice changes the spectral shape of both the pink sapphire (Sample 1) and zoned brown to colorless sapphire (Sample 2). It is important to note that though for Sample 2, the average concentration is as low as 2 ppm, the Cr3+ contribution can still be seen in the spectrum. This proves that when detecting bright and well-defined chromophores such as Cr3+ in sapphire, CL is extremely sensitive.
If the Cr3+ spike is disregarded for the colorless to brown zoned sapphire, the results are consistent with the interpretation outlined above. In this analysis, any light absorption in the sapphire is disregarded, as it is difficult to quantify in CL. Nonetheless, it is anticipated that this effect is insignificant in size, given that light only propagates a few 100 nm at most before coupling out.
It is also possible that there could be a contribution from the F-center in the blue CL band. However, there is a suggestion that this contribution is minor, due to the observed peak wavelengths (~ 450 - 490 nm) and the fact that this peak can be fitted well with a single Gaussian (as displayed in Figure 2).
Figure 8. Fe/Ti plotted against A) Ti, B) CL spectra peaks (Sample 4, left; Sample 2, right), C) CL peak amplitudes of blue peak against Fe/Ti ratios, and D) CL peak amplitudes of red peak against Fe/Ti ratios.
The interest taken in the oxidation state of Fe and Ti in sapphires is largely because these two elements are important chromophores (that is to say, they are color-causing elements). The trace element compositions and CL spectra observed varying proportions between Ti3+ and Ti4+, which may be important indicators of the relative reducing or oxidizing conditions that the sapphires may have experienced. Ti4+ is generally more stable and common between the different valence states of Ti, in contrast to Ti3+, which is more reactive as a reducing agent and, as a result, less stable.
It is suggested by the negative power trendline between Fe/Ti and Ti that the conditions of sapphire formation change at a steady pace from reducing (dominantly Ti3+) to oxidizing (relatively more Ti4+) conditions.
It is known that both Ti3+ and Ti4+ exist in sapphires [36, 37]. The interpretation of Ti peaks  determines the finding of a negative correlation. A more dominant Ti3+ peak is found when there are low Fe/Ti ratios, and a bimodal peak of Ti3+ and Ti4+ is attained with high Fe/Ti (Fig. 8).
It is commonly accepted that the intervalence charge transfer (IVCT) between Fe2+ and Ti4+ is the cause of the blue color in sapphires . Though it may be suggested that IVCT results in this change between Ti3+ and Ti4+, the fact remains that newer XANES research indicates that it is the presence of Fe3+ and Ti4+ in mixed acceptor states (through an energy band model) that is the true reason behind the color in blue sapphires . It seems that more work remains in terms of the oxidation states of Fe and Ti, and the understanding of their lattice positioning may offer insight towards understanding the conditions required for sapphire formation.
More rigorous high-resolution zonation mapping could be performed in the future, in terms of CL spectroscopy, which would go beyond point ROI acquisitions. Systematic hyperspectral imaging could be performed over mm to cm distances if, during acquisition, the SEM stage is scanned in an automated fashion.
To assess their chemical zonation, natural alluvial sapphires from eastern Australia underwent analysis by both CL and LA-ICP-MS, as well as test spectral CL as a method by which to identify sapphire zonation. Variation in zoned sapphires was shown by spectral data, which is correlated with variation in Fe and Ti concentrations measured by LA-ICP-MS.
The CL spectral emissions in sapphires are controlled by the relationship between Fe/Ti and Ti concentrations. This is due to defects in the crystal lattice, which are related to the substitution of different ions.
Two main observations related to the presence of Fe, Ti3+ and Ti4+ in the corundum crystal lattice were made by the negative trendline between Fe/Ti and Ti. These are as follows: firstly, the distribution of Ti is predominantly Ti3+ when Fe/Ti is low, then, as Fe/Ti increases, it evens out the ratio between Ti3+ and Ti4+; and secondly, as Fe/Ti increases, thanks to the presence of Fe and its quenching effects, the CL intensity decreases.
This is the first study of its kind to detail the correlation between Fe and Ti, as recorded by spectral and trace element data, which therefore may reveal variably reducing or oxidizing conditions linked to sapphire formation.
 Chapoulie, R., Capdupuy, C., Schvoerer, M., and Bechtel, F., Cathodoluminescence and crystal growth of sapphire, Physica status solidi (a), 171, 613-621 (1999)
 Schaltegger, U., Fanning, C. M., Günther, D., Maurin, J. C., Schulmann, K., and Gebauer, D., Growth, annealing and recrystallization of zircon and preservation of monazite in high-grade metamorphism: conventional and in-situ U-Pb isotope, cathodoluminescence and microchemical evidence, Contributions to Mineralogy and Petrology, 134, 186-201 (1999)
 Pagel, M., Barbin, V., Blanc, P., and Ohnenstetter, D., Cathodoluminescence in Geosciences, Berlin/Heidelberg, Berlin/Heidelberg: Springer Berlin Heidelberg (2000)
 Ponahlo, J., Cathodoluminescence as a tool in gemstone identification, Cathodoluminescence in Geosciences, Springer, 479-500 (2000)
 Garnier, V., Ohnenstetter, D., Giuliani, G., Blanc, P., and Schwarz, D., Trace-element contents and cathodoluminescence of “trapiche” rubies from Mong Hsu, Myanmar (Burma): geological significance, Mineralogy and Petrology, 76, 179-193 (2002)
 Richter, D. K., Götte, T., Götze, J., and Neuser, R. D., Progress in application of cathodoluminescence (CL) in white paper 13 sedimentary petrology, Mineralogy and Petrology, 79, 127- 166 (2003)
 Schertl, H.-P., Neuser, R. D., Sobolev, N. V., and Shatsky, V. S., UHP-metamorphic rocks from Dora Maira/Western Alps and Kokchetav/Kazakhstan; new insights using cathodoluminescence petrography, European Journal of Mineralogy, 16, 49-57 (2004)
 Boggs, S., and Krinsley, D. H., Application of cathodoluminescence imaging to the study of sedimentary rocks / Sam Boggs, Jr. and David Krinsley, Cambridge, UK ; New York, Cambridge, UK ; New York : Cambridge University Press (2006)
 Guguschev, C., Götze, J., and Gobbels, M., 2010, Cathodoluminescence microscopy and spectroscopy of synthetic ruby crystals grown by the optical floating zone technique.(Author abstract)(Report), American Mineralogist, 95, 449 (2010)
 Götze, J., Application of Cathodoluminescence Microscopy and Spectroscopy in Geosciences, Microscopy and Microanalysis, 18, 1270-1284 (2012)
 Siebel, W., Shang, C. K., Thern, E., Danišík, M., and Rohrmüller, J., Zircon response to high-grade metamorphism as revealed by U–Pb and cathodoluminescence studies, International Journal of Earth Sciences, 101, 2105- 2123 (2012)
 Götze, J., and Kempe, U., Physical Principles of Cathodoluminescence (CL) and its Applications in Geosciences, in, Cathodoluminescence and its Application in the Planetary Sciences, 1-22 (2009)
 Yacobi, B. G., and Holt, D. B., Cathodoluminescence Microscopy of Inorganic Solids, Boston, MA, Boston, MA : Springer US, Imprint: Springer (1990)
 Coenen, T., and Haegel, N. M., Cathodoluminescence for the 21st century: Learning more from light, Applied Physics Reviews, 4 (2017)
 Houseknecht, D. W., Use of cathodoluminescence petrographyfor understanding compaction, quartz cementation,and porosity in sandstones: Luminescence Microscopy and Spectroscopy: Quantitative and Qualitative Applications, SEPM Short Course, 25, 59-66 (1991)
 Milliken, K. L., Reed, R. M., and Laubach, S. E., Quantifying compaction and cementation in deformation bands in porous sandstones, AAPG Memoir, 85, 237-249 (2005)
 Coenen, T., Cathodoluminescence Imaging on Sedimentary Rocks: Quartz Sandstones: whitepaper (2016)
 Dobrovinskaya, E. R., Lytvynov, L. A., and Pishchik, V., Properties of Sapphire, Sapphire Materials, Manufacturing, Applications, Springer, 55-176 (2009)
 Hughes, R. W., Ruby & Sapphire, RWH Publishing (1997)
 Pisutha-Arnond, V., Hager, T., Atichat, W., and Wathanakul, P., The role of Be, Mg, Fe, and Ti in causing colour in corundum, Journal of Gemmology, 30, 131-143 (2006)
 Eggleston, J. M., Deshazer, L. G., and Kangas, K. W., Characteristics and kinetics of laser-pumped Ti:sapphire oscillators, Quantum Electronics, IEEE Journal of, 24, 1009- 1015 (1988)
 Kumari, S., Kushwaha, A., and Khare, A., Spatial distribution of electron temperature and ion density in laser induced ruby (Al2 O3 :Cr3+) plasma using Langmuir probe, J. Instrum., 7 (2012)
 Munisso, M. C., Zhu, W., Leto, A., and Pezzotti, G., Stress dependence of sapphire cathodoluminescence from optically active oxygen defects as a function of crystallographic orientation: The journal of physical chemistry, 111, 3526 (2007)
 Graham, I., Sutherland, L., Zaw, K., Nechaev, V., and Khanchuk, A., Advances in our understanding of the gem corundum deposits of the West Pacific continental margins intraplate basaltic fields, Ore Geology Reviews, 34, 200-215 (2008)
 Harlow, G., and Bender, W., A study of ruby (corundum) compositions from the Mogok Belt, Myanmar: Searching for chemical fingerprints, American Mineralogist, 98, 1120-1132 (2013) white paper 14
 Bindeman, I. N., Serebryakov, N. S., Schmitt, A. K., Vazquez, J. A., Guan, Y., Azimov, P. Y., Astafiev, B. Y., Palandri, J., and Dobrzhinetskaya, L., Field and microanalytical isotopic investigation of ultradepleted in 180 Paleoproterozoic “Slushball Earth” rocks from Karelia, Russia, Geosphere, 10, 308-339 (2014)
 Sutherland, F. L., Coenraads, R. R., Abduriyim, A., Meffre, S., Hoskin, P. W. O., Giuliani, G., Beattie, R., Wuhrer, R., and Sutherland, G. B., Corundum (sapphire) and zircon relationships, Lava Plains gem fields, NE Australia: Integrated mineralogy, geochemistry, age determination, genesis and geographical typing, Mineralogical Magazine, 79, 545-581 (2015)
 Sorokina, E. S., Hofmeister, W., Hager, T., Mertz-Kraus, R., Buhre, S., and Saul, J. M., Morphological and chemical evolution of corundum (ruby and sapphire): crystal ontogeny reconstructed by EMPA, LA-ICP-MS, and [Cr.sup.3+] Raman mapping.(Report)(Author abstract), American Mineralogist, 101, 2716 (2016)
 Upton, B. G. J., Hinton, R. W., Aspen, P., Finch, A. A., and Valley, J. W., Megacrysts and Associated Xenoliths: Evidence for Migration of Geochemically Enriched Melts in the Upper Mantle Beneath Scotland, Journal of Petrology, 40, 935-955 (1999)
 Brenny, B. J. M., Coenen, T., and Polman, A., Quantifying coherent and incoherent cathodoluminescence in semiconductors and metals, Journal of Applied Physics, 115, 244307 (2014)
 Meuret, S., Coenen, T., Zeijlemaker, H., Latzel, M., Christiansen, S., Conesa Boj, S., and Polman, A., Phys. Rev. B 96, 035308 (2017)
 Wong, J., Verdel, C., and Allen, C. M., Trace-element compositions of sapphire and ruby from the eastern Australian gemstone belt, Mineralogical Magazine, 81 (2017)
 Wong, J., Trace element and oxygen isotope geochemistry of east Australian sapphires: regional petrogenesis and tectonic implications, PhD thesis, University of Queensland (2018)
 Paton, C., Hellstrom, J., Paul, B., Woodhead, J., and Hergt, J., Iolite: Freeware for the visualisation and processing of mass spectrometric data, Journal of Analytical Atomic Spectrometry, 26, 2508-2518 (2011)
 Longerich, H. P., Jackson, S. E., and Gunther, D., Laser ablation inductively coupled plasma mass spectrometric transient signal data acquisition and analyte concentration calculation, Journal of analytical atomic spectrometry, 11, 899-904 (1996)
 Fritsch, E., and Rossman, G. R., An update on color in gems: Part 3: Colors caused by band gaps and physical phenomena, Gems & Gemology, 24, 81-102 (1988a)
 Fritsch, E., and Rossman, G. R., An update on color in gems: Part 3: Colors caused by band gaps and physical phenomena, Gems & Gemology, 24, 81-102 (1988b)
 Wongrawang, P., Monarumit, N., Thammajak, N., Wathanakul, P., and Wongkokua, W., Oxidation states of Fe and Ti in blue sapphire, Materials Research Express, 3, 026201 (2016)
This information has been sourced, reviewed and adapted from materials provided by Delmic B.V.
For more information on this source, please visit Delmic B.V.