<<

SPECTROSCOPIC STUDIES ON DEMANTOID . IS IT POSSIBLE TO ASSIGN ORIGINS BASED ON SCIENTIFIC EVIDENCE? Clemens Schwarzinger, Bettina Schwarzinger, Manuela List Institute for Chemical Technology of Organic Materials, Johannes Kepler University Linz, Austria

Introduction: Demantoid garnets belong to the family of Ugrandites, more exactly they are

Andradites with the general formula Ca3Fe2[SiO4]3. As their luster and are comparable to that of (thus the name demantoid, meaning diamond- like), combined with a pleasing green to yellowish green color they are sought after Fig. 3. Two demantoid parcels of different color and unknown origin were and highly valued by connoiseurs. Gem material can be found in bought and could be identified as originating from Namibia. Namibia, Madagascar, Italy, Canada, Iran, Pakistan but most famed are the stones from Russia. Demantoid is probably the only gemstones which has an increased value when a certain type of inclusion is present - the so called horse tails, a bunch of byssolith Results and Discussion: fibers typically emerging from a chromite crystal. This feature can be present at A very indicative parameter is the content. It is the chromophor that several localities and if so limits the identification of origin to Russia, Canada or gives garnets the typical green color which is coveted most. However, Italy, but it is not always present. not all stones have chromium: Namibia and Madagascar typically has none, while It is the purpose of this research to determine if a scientific approach based on Canada, Italy, Iran, Pakistan and Russia can have concentrations in the %-range. spectroscopy is capable of determining the origin of demntoid gemstones when no Fig. 2. shows the Cr-concentrations of Russian and Namibian stones; whilst in the typical inclusions are present. first half concentrations of 0 ppm to 2% can be found, only one Namibian sample shows a significant amount of Cr (~100 ppm). Very high Cr values have also been The task: found in Iran, Pakistan and some Canadian samples. The Russian Ural region and the area of Tubussis in Namibia are the most prolific Another indicative element is , which is very low (<10ppm) in Namibian mining areas today so it is utmost interest to be able to distinguish stones from stones, shows highest concentrations in Madagascar and Canada (>1500 ppm); these localities. Russian stones have between 0 and 1200 ppm. We have tested specimens from all possible locations and set up an analytical Several statistical methods have been applied to best separate the dataset database using mainly Laser Ablation - icp Mass Spectrometry (LA-icp-MS) to including principal component analysis, which gives best results in separating all determine various trace elements and set up a map. Then two parcels demantois possible localities. However, if the task is to separate Russian from Namibian of different color and from unknown locality were purchased and analyzed material we have found out that using the chromium content and the ratio of similarily to assign their origin. manganese over titanium gives very good results. When both axes are plotted in a logarythmic scale a clear separation line can be drawn with very little overlap. In fact only one Russian sample (all three values) can be found in the „Namibian zone” and only two Namibian single values are in the „Russian zone” as can be seen in Figure 4. When these settings are applied to the two parcels of unknown gemstones it becomes evident that they must have originated from Namibia. In addition ot LA-icp MS all samples ware analyzed by FTIR and Raman Fig. 1. Demanoid reference samples from Tubussis, Namibia showing different spectroscopy. These techniques also show features that seem to be typical for color shades (photos (c) www.csgemstones.com). certain locations, however, the space available here is limited and those data could not be incorporated so far.

100000 I G 1

Experimental: R 17 / 1 R 17/ 1 R 17 /1 C 3 Around 50 samples from various localities have been analyzed using LA-icp MS. I G 1 C 3 R G 4 R G 4 10000 C 3 For each sample 3 different spots were measured using a 100 µm spot placed on R G 4 P CS I G 1 P CS P C S the girdle of the gemstones or a flat spot on crystals and gem rough. R G 9 R 17 /3 R G 9 R 17 /3 R 17 /3 The trace elements Mg, Ti, V, Cr and Mn were measured and quantified, the main R 33 R G 9 R 17/ 2 R 32 /1 C 2 constituents Si, Ca and Fe were monitored but not used quantitatively as their 1000 R 17/ 2 R 17/ 2 R 32/ 1 C 2

] C 1 R 33 R 33 C 2 concentrations are too high to give reliable results. Calibration was done with NIST C 1 R 32/ 1

m C 1

610, 612 and 614 glass standards. p N 3 /1 _4 R G3 p

[ R G 3 VM G 2 N 3/ 1_4 100 R G 3 r V M G2 N 3/ 1_ 4

C V M G 2 R 32 /2 M CS 2 N 3/ 1_7 25000 R 3 2/ 4

2 R 3 2/ 4 N R 3 R 3 1 M C S

5 N 3/ 1_9 N 3 /1 _9 Urals, Russia Namibia N R 4 R 32/ 2 M i G B 10 R 3 2/ 2 R 32 /4 M i G B R 18 0 R 3 1 R 17 9 M i G B I VY 11 IV Y3 20000 M C S 2 IV Y 12 I VY 4 I VY 11 N 3/ 1_5 M C S 2 I VY 5 R 161 /3 N R 4 I VY 12 I V Y4 R 179 I V Y2 N 3/ 1_7 N R 3 N 3/ 1_9 I VY 6 IV Y 4 I VY 10 I V Y6 N 3/ 1_1 I VY 1 I V Y1 2 N 3 /1 _6 1 R 16 1/ 3 I V Y1 0 I VY 2 IV Y 1 I VY 3 N 86 N 8 6 N 3/ 1_1 R 180 N 121 N 3/ 1_5 N R3 R 16 1/ 3 N 123 N 3/ 1_3 N 1 23 N 100 I VY 10 N 10 0 I VY 1 N 8 6 N 100

] 15000 N 3/ 1_8 IV Y 11 R 179 I VY 5 IV Y5 N 12 1 IV Y2 I V Y6 m R 18 0 R 31 N 3/ 1_1 0 N 3/ 1_2

p N 3 /1 _3 N 3/ 1_ 10

p 0.1 [

r 0.1 1 10 100 1000

C 10000 2

5 Mn/Ti

5000 Fig. 4. Graphical separation based on trace elements. Green dots... Russia; yellow dots... Namibia; brown dots ... unknown samples.

0 Acknowledgment: We would like to thank Prof. Leopold Rössler from the Austrian Gemological Fig. 2. Chromium concentration of the reference samples from the Urals, Society as well as Gerhard Brandstetter for providing numerous samples. Russia (left) and Namibia (right).