A New Source of Garnets with Directional Color Shift

Abstract

This study investigates color-change garnets from Tanzania, with one sample displaying a secondary color-change phenomenon potentially linked to the Usambara Effect (UE). While the sample's thickness was too thin to directly observe the effect, internal reflections extended the path length, causing the UE to manifest at the stone's periphery. The garnets showed not only a color change but also an unusual hue shift at the edges, which is atypical for garnets. The aim of this research is to explore the cause of the secondary color change and determine whether these garnets differ from other color-change garnets. Gemological and spectroscopic data are presented alongside a comparative analysis to enhance understanding of these optical properties.

Introduction

Most color-change garnets found in the gem market exhibit compositions that are intermediate or solid-solution blends between pyrope and spessartine end-members, often with a trace presence of chromium or vanadium-plus-chromium, which accounts for their distinctive color-change properties. Previous studies have outlined two primary types of color-change garnets: those composed of pyrope with exceptionally high Cr3+ content (Carstens, 1973) and pyrope-spessartine combinations containing both Cr3+ and V3+ (Karl Schmetzer et al., 2001, 2002; Karl Schmetzer & Heinz-Jürgen Bernhardt, 1999; V. D. Manson & Stockton, 1981; V. Manson & Stockton, 1984; Michael S krzemnicki et al., 2001; Pamela Cevallos & Ziyin Sun, 2018; Qiu & Guo, 2021; Stoclzton, 1982; Ziyin Sun et al., 2015; Ziyin Sun & Jonathan Muyal, 2017; Zwaan, 2020). Within trade circles, most color-change garnets are commonly referred as pyrope-spessartine (Stockton & Manson, 1985) . Recent discoveries have also unveiled color-change garnets composed of grossular and pyrope-spessartine with a notably high concentration of grossular garnet (Pamela Cevallos & Ziyin Sun, 2018; Promwongnan & Buathong, 2018; Ziyin Sun & Jonathan Muyal, 2017). In all these variations of color-change garnets, chromium (Cr) and vanadium (V) serve as trace elements responsible for the distinctive color-change phenomenon. The concentration of these elements, alongside other chromophoric trace elements such as iron (Fe) and manganese (Mn), significantly influences both the inherent body color and the observed color-change characteristics of these gemstones.

Color-change garnets display a range of hues, from blue, green, and yellow in daylight to orange, red, and purple under incandescent light. They show two distinct colors when switching between these lighting conditions, and in some cases, they change color with varying path lengths under constant lighting. This color change is called Dichromatism or the Usambara Effect in trade terms, a special case of the Alexandrite effect. The Usambara effect occurs when one of the dual transmission windows is depleted based on the thickness of the sample.

In 1997, Halvorsen and Jensen defined the Usambara effect, explaining the color shift in chrome-vanadium tourmaline from Tanzania’s Usambara region as a result of increased light path length (Halvorsen & Jensen, 1997). Longer path lengths absorb shorter wavelengths (like green), shifting the color towards red. This differs from the Alexandrite effect, which is driven by light source wavelength changes. Earlier, V. Manson & Stockton (1984) observed similar shifts in Madagascar garnets but couldn’t explain the cause. Halvorsen and Jensen linked the change to greater absorption in thicker material, formally defining the Usambara effect.

In 2006, Halvorsen expanded on this with absorption-modified dispersion (AMD), observed in faceted Usambara tourmalines, where reflected light shows red and green flashes, linked to the stone’s transmission spectrum (Halvorsen, 2006). The study also revealed interactions between color-change phenomena, such as the influence of the Usambara effect on the Alexandrite effect, and interactions with thermochromism and pleochroism. These effects follow the Beer-Lambert law, where absorption increases with path length, causing a perceptual shift from green to red.

Grant M. Pearson & Donald B. Hoover (2013) also applied the Beer-Lambert law to explain dichromatism, where color changes with path length due to differential absorption in two spectral bands. This was observed in materials like dravite tourmaline, rhodolite garnet, and some dyes.

In February 2023, Gemological Science International (GSI) received a new batch of garnets sourced primarily from Tanzania, supplied by Ales Patrick Krivanek of Ravenstein Gem Co. These garnets displayed unusual optical properties. At first glance, they appeared to exhibit pleochroism, showing different colors and intensities depending on the viewing angle. This was unexpected, as garnets are isotropic minerals with a cubic crystal structure, and pleochroism is typically found in anisotropic minerals. Additionally, the garnets displayed a subtle color change depending on the light source, shifting hues under fluorescent and incandescent lighting. To investigate these phenomena, four rough garnet samples were obtained for further analysis.

The samples were examined at the GSI Mumbai laboratory. Samples 1, 3, and 4 showed a change in color saturation when illuminated with different light sources, appearing warmer under incandescent light (Fig.1). Sample 2, however, exhibited more complex behavior. Under fluorescent light, it appeared pale grayish-green, while under incandescent light, it shifted to a brownish-red color. Moreover, a purplish-red hue appeared at the periphery of the stone when viewed from the opposite end of the light source, most pronounced under incandescent light (Fig. 2). Sample 3 also exhibited a similar dual color shift.

Of the four samples, Samples 2 and 3 exhibited a distinctive feature: under both light sources, two different colors were visible. One color appeared when light reflected off the interior of the stone, while another appeared when light passed through it. Due to the irregular shape of the rough samples, it was difficult to determine whether these color changes were caused by internal reflection, the stone’s thickness, or other optical factors. To explore this further, the rough samples were cut to create two parallel flat faces, allowing for more controlled analysis of the optical effects.

The primary goal of the study was to determine whether the secondary color phenomena observed in Samples 2 and 3 can be attributed to the Usambara effect (UE), or if other factors, such as internal reflection. Additionally, the study aims to assess whether these garnets exhibit unique characteristics compared to other known color-change garnets. The following sections present the gemological and spectroscopic data for these Tanzanian garnets, along with a comparative analysis of other documented color-change garnets.

Methods

All garnet samples were tested using a combination of non-destructive basic gemological techniques and advanced spectroscopic methods.

For detailed microscopic observations, a Zeiss Stemi 508 trinocular microscope equipped with a camera setup was used. The luminescence properties of the stones were examined using a Fable 4-watt UV lamp emitting both long-wave (365 nm) and short-wave (253.7 nm) UV radiation. The examination was conducted in a darkroom with the lamp positioned 10 cm away from the sample.

Raman spectroscopy was performed with an Enspectra confocal Raman microscope equipped with a 532 nm laser excitation source. The spectra were recorded in the range of 250–1250 cm⁻¹ with a spectral resolution of 5 cm⁻¹. Data acquisition included a 3 second exposure time with ten accumulations, and the laser power on the sample was set at 50 mW. The instrument was calibrated using the 521 cm⁻¹ Raman peak of silicon.

Fourier-transform infrared (FTIR) spectra were obtained using a Nicolet iS50 spectrophotometer with a spectral range of 400–2000 cm⁻¹, a resolution of 4 cm⁻¹, and 32 scans.

Ultraviolet-visible-near infrared (UV-Vis-NIR) spectroscopy was performed with an Agilent Cary 5000 spectrophotometer over the range of 400–700 nm. All spectra were recorded in an unpolarized state

Chemical analysis was conducted using a Horiba XGT-7000 energy-dispersive X-ray fluorescence (EDXRF) spectrometer with a 1.2 mm aperture, X-ray tube voltage of 50 kV, and an acquisition time of 100 seconds under partial vacuum conditions. The spectra were collected using the mapping method, optimizing analyzer settings for the best results. EDXRF analysis revealed the presence of Si, Al, and Mn, along with trace amounts of Fe, Ca, Cr, and Ti. For end-member composition calculations, the elemental data was converted to oxides (Deer et al., 2013).

Results and Discussion

The gemological properties of the four garnet samples are summarized in Table 1. The refractive index (RI) of the samples is consistently measured at 1.77, with specific gravity (SG) ranging from 3.82 to 3.86, confirming the samples as pyrope-spessartine garnets. Under both long and short-wave ultraviolet (UV) radiation, the garnets remained inert, indicating no fluorescence. Polariscope examination revealed slight to moderate anomalous extinction, indicating internal strain in the garnet structure. Needle-like inclusions were observed in most of the samples, regardless of orientation, and some stones also exhibited fingerprints and stress cracks (Fig.3).

Raman spectra of four samples, showed dominants peaks in the range of 358, 555, 913 cm−1 (Fig.4). The spectra of spessartine and pyrope samples from RRUFF database shows peaks at 175, 221, 350, 552, 905 cm−1 and 222, 364, 563, 916 cm-1. The close analysis of this data confirms that all these samples are intermediate members of the pyrope-spessartine solid solution series (Fu et al., 2022). The FTIR spectra of all these specimen showed bands at 985, 897, 866, 638, 576, 525, 483 and 454 cm-1 (Fig.4) consistent with solid solution of pyrope-spessartine garnets (Qiu & Guo, 2021).

Chemical analysis via EDXRF (Table 1) confirmed that the garnets are intermediate members of the pyrope-spessartine solid solution series. Pyrope content ranged from 37.25 to 41.66 mol%, while spessartine content ranged from 45.30 to 49.95 mol%. Additionally, the samples contained 5.20 to 11.48 mol% of grossular, and small amounts of iron (1.50–1.77 wt% Fe). The iron was considered in its Fe²⁺ (FeO) form, consistent with the almandine component, though estimation of Fe³⁺ was not reliable due to the low concentration of iron. Small amounts of vanadium (0.13–0.29 wt% V₂O₃) and chromium (0.06–0.17 wt% Cr₂O₃) were also detected.

The absorption spectra of the four garnet samples (Fig.5) showed two primary absorption bands. The first is a strong band in the blue-to-violet range with maxima at 410, 422, and 431 nm, while the second is a broad absorption band with a maximum at 573 nm. Between these two bands, weaker absorption features were observed at 459, 486, 506, and 526 nm. These bands are typical for vanadium (V³⁺ at 573 nm), manganese (Mn²⁺ at 410, 422, and 486 nm), and iron (Fe³⁺ at 431 nm and Fe²⁺ at 459, 506, and 526 nm)(Karl Schmetzer et al., 2002; Karl Schmetzer & Heinz-Jürgen Bernhardt, 1999; Qiu & Guo, 2021; Ziyin Sun et al., 2015). The absorption bands of Cr³⁺ (around 570 nm) overlap with the vanadium absorption at 573 nm, making it difficult to separate these two absorptions. This overlap suggests that chromium and vanadium are both contributing to the observed color change.

The color of the garnets is primarily influenced by the presence of manganese (Mn²⁺), which imparts the characteristic brownish-red hue, while vanadium (V³⁺) and chromium (Cr³⁺) modify the color by shifting the base hues. Previous studies on color-change garnets have reported refractive indices ranging from 1.74–1.773 and specific gravity ranges of 3.816–3.98, consistent with the properties of the current samples(Halvorsen & Jensen, 1997; Karl Schmetzer et al., 2002; V. D. Manson & Stockton, 1981; V. Manson & Stockton, 1984; Michael S krzemnicki et al., 2001; Pamela Cevallos & Ziyin Sun, 2018; Qiu & Guo, 2021; Stockton & Manson, 1985; Stoclzton, 1982; Zwaan, 2020). These samples fall within the established ranges for pyrope-spessartine garnets.

The color change effect in the garnets is most noticeable in Sample 2, which exhibited a marked shift in color between fluorescent white light and incandescent light. In contrast, Samples 1 and 4 showed minimal color variation, appearing warmer under incandescent light due to their lower concentrations of vanadium and chromium (Table 1 and Fig. 4). Although Samples 2 and 3 have similar chemical compositions and UV-Vis spectra, Sample 2 demonstrated a more pronounced color change. This discrepancy could be attributed to differences in sample shape, size, or internal features.

Samples 2 and 3 exhibited not just a single-color change, but two distinct color phenomena. These stones demonstrated color variation under incandescent versus fluorescent lighting, as well as a secondary color difference depending on whether the light passed through the stone or was internally reflected. This variation is not solely attributed to changes in illumination but rather to the differing amounts of light perceived by the viewer depending on the path it takes through the stone. When light is transmitted through the stone, it reveals the garnet's body color; when light is internally reflected, it produces a different color appearance.

The irregular shapes of the rough samples made it difficult to clearly observe this behavior, but the increased optical path length due to multiple internal reflections contributed to the observed color at the periphery of the stones. The phenomenon of varying color with light reflection versus transmission is like the Usambara Effect (UE), which occurs when changes in optical path length affect the observed color, especially when the viewer and the light source are on the same side of the gem.

In summary, while the color change in these garnets is influenced by the concentrations of vanadium and chromium, the additional color phenomena observed in Samples 2 and 3 may be a result of internal reflection and varying optical path lengths. This suggests a complex interaction between the garnet's internal features, its geometry, and the lighting conditions. Further investigation into these factors is needed to fully understand the underlying causes of these dual color effects.

Conclusion

The gemological investigation of garnets sourced from Tanzania, specifically Sample 2, revealed a fascinating and rare phenomenon: the Usambara effect. Sample 2 exhibited a striking color shift, showing pale grayish green under fluorescent light and transitioning to a deep brownish red under incandescent illumination. This color change, along with the associated dichroism, underscores the aesthetic appeal and challenges the conventional understanding of isotropic materials like garnet.

However, the limited size of the samples, particularly Samples 2 and 3, restricted further investigation into the full extent of the Usambara effect. Despite these constraints, gemological analysis and spectroscopy confirmed that the color change and dichroic effects are driven by the interplay between light transmission through the stone and internal reflections, supporting the presence of the Usambara effect.

This study emphasizes the importance of sample size in exploring the nuances of this phenomenon. Larger samples would allow for more pronounced and observable effects, facilitating a deeper understanding of the Usambara effect. Specifically, larger samples would enable: More accurate observations of color shifts in relation to the thickness of the material, and a clearer understanding of how light transmission and internal reflection contribute to the color change.

Sample 2’s ability to display multiple hues, depending on the light source and the material's thickness, enriches its gemological identity and will undoubtedly captivate gem enthusiasts and collectors. This investigation serves as a foundation for future research, highlighting the need for larger specimens to fully uncover the complexities of the Usambara effect.

We would like to dedicate this article to Mr. Ales Patrick Krivanek of Ravenstein Gem Co., whose contributions helped make this study possible.

References

Carstens, H. (1973). The Red-Green Change in Chromium-Bearing Garnets. In Contr. Mineral. and Petrol (Vol. 41). Springer-Verlag.

Deer, W. A. ., Howie, R. A. ., & Zussman, J. . (2013). An introduction to the rock-forming minerals. Mineralogical Society of Great Britain and Ireland. https://doi.org/10.1180/DHZ

Fu, M., Dai, J., & Zhao, L. (2022). A Study on the Raman Spectral Characteristics of Garnet from the Jiama Copper Polymetallic Deposit in Tibet. Minerals, 12(12). https://doi.org/10.3390/min12121578

Grant M. Pearson, & Donald B. Hoover. (2013). Dichromatism, the cause of the Usambara and the alexandrite colour-change effects. The Australian Gemmologist, 62–70. www.gem.org.au

Halvorsen, A. (2006). The Usambara effect and its interaction with other colour change phenomena. Journal of Gemology, 1(2), 1–21.

Halvorsen, A., & Jensen, B. B. (1997). A new colour-change effect. Journal of Gemology, 25(5), 325–330.

Karl Schmetzer, & Heinz-Jürgen Bernhardt. (1999). Garnets from Madagascar with a Color Change of Blue-Green to Purple. Gems & Gemology, 196–201.

Karl Schmetzer, Thomas Hainschwang, Heinz-Jürgen Bernhardt, & Lore Kiefert. (2002). New Chromium and Vanadium Bearing Garnets from Tranoroa, Madagascar. Gems & Gemology, 38(2), 148–155.

Karl Schmetzer, Thomas Hainschwang, Lore Kiefert, & Heinz-Jürgen Bernhardt. (2001). Pink to Pinkish Orange Malaya Garnets from Bekily, Madagascar. Gems & Gemology, 37(4), 296–308.

Manson, V. D., & Stockton, C. M. (1981). Gem Garnets in the Red-To-Violet Color Range. Gems & Gemology, Winter, 191–204.

Manson, V., & Stockton, C. M. (1984). Pyrope-Spessartine Garnets with Unusual Color. Gems & Gemology, 200–207.

Michael S krzemnicki, Henry A Hanni, & Eric Reusser. (2001). Colour-change garnets from Madagascar comparison of colorimetric with chemical data. Journal of Gemology, 27, 395–408.

Pamela Cevallos, & Ziyin Sun. (2018). Unusual Orange Pyrope Spessartine-Grossular Garnet_Lab Note. Gems & Gemology, 54(1), 59–60.

Promwongnan, S., & Buathong, A. (2018). A Rare Nearly-Pure End-Member Grossular Garnet with Color-Change Effect.

Qiu, Y., & Guo, Y. (2021). Explaining colour change in pyrope-spessartine garnets. Minerals, 11(8). https://doi.org/10.3390/min11080865

Stockton, C. M., & Manson, D. V. (1985). A Proposed New Classification for Gem-Quality Garnets. Gems & Gemology, 205–207.

Stoclzton, C. M. (1982). Two Notable Color-Change Garnets. Gems & Gemology, 100–101.

Ziyin Sun, Aaron C. Palke, & Nathan Renfro. (2015). Vanadium and Chromium-Bearing Pink Pyrope Garnet Characterization and Quantitative Colorimetric Analysis. Gems and Gemology, 51(4), 418–427. https://doi.org/10.5741/GEMS.51.4.348

Ziyin Sun, & Jonathan Muyal. (2017). Pyrope-Spessartine Color Change Garnet with a High Grossular Component. In GEMS & GEMOLOGY (Vol. 53, Issue 1).

Zwaan, J. C. (Hanco). (2020). Colour-Change Garnet from Tanga, Tanzania. Journal of Gemmology, 37(2), 133–134. https://doi.org/10.15506/jog.2020.37.2.133