'Neon' green diopside

The challenge was then to know if the presented material was tremolite or diopside. Mean refractive index is significantly higher for diopside (1.695) than for tremolite (1.618), as well as the mean specific gravity which is 3.30 for diopside against only 3.01 for tremolite. For both materials, visible spectra are quite similar and it is difficult - likely impossible - to make the difference with a common hand-held spectroscope whenever they are observables.

The properties of the 2.12 ct gemstone are summarized in table 1, those of the 0.79 ct one are identical.

Shape	oval facetted
Size	10.5 x 6.7 x 4.2 mm
Color	strongly yellowish green
Lustre	vitreous
Weight	2.12
SG	3.24
RI	1.669 - 1.698
DR	0.029 B+
Pleochroism	strongly yellowish green / yellow-green  (no third color, color down to C-axis is slightly more saturated strongly yellowish green)
Polariscope / Conoscope	anisotrope, bi-axial
SWUV	weak chalky light greenish yellow
LWUV	inert
Magnetic susceptibility	none (N52)

Table 1 : Observational and measured properties

Infrared reflectance spectroscopy:

The infrared reflectance spectrum was acquired from the table of the 2.12 ct stone which is large enough to get a good signal without any difficulty. Except if the stone was cut in a particular orientation of the crystal, the spectrum is to be considered as unoriented.

The spectrum shown in figure 2 is typical and diagnostic of the diopside mineral specie.

UV-VIS-NIR spectroscopy:

Getting a reliable Uv-Vis-Nir spectrum from a faceted gemstone requires to find a light path in the stone. The retained solution is to use the girdle as entry and exit windows. Since optical axis have not been located in the faceted stone, the acquired spectrum is to be considered as unoriented.

The spectrum shows two main absorption regions located around 450 nm and 660 nm followed by a third one around 1060 nm as shown in figure 3. The first two ones have a FWHM of about 115 to 125 nm although the NIR one is much wider and flatter.

Photoluminescence spectroscopy:

A 20 mW 405 nm laser was used to obtained the photoluminescence spectra shown in figure 4. Even, with such power, the photoluminescence phenomenon is rather weak but clearly shows two main emission peaks located around 585 nm and 775 nm. The intensity of the peaks of the spectrum identified by 'exc 405nm (ray || length) - 2.12 ct diopside' being rather weak compared to the other one, it is resized for easy reading. The peaks around 580 nm are narrow although the 770 nm peaks are much wider. In the 690 nm region there is a weak shoulder on the left side of the 770 nm emission peaks.

After Gorobets & Rogojine ^[3], the 580 nm emission peak in diopside photoluminescence is caused by the Mn²⁺ activator in M2 site replacing Ca²⁺ (Ca_VIII) and another emission peak located around 730 nm should be caused by the Fe³⁺ activator replacing Si⁴⁺. The 690 nm emission feature could possibly be assigned to Cr³⁺ and/or V²⁺. Usually, Cr³⁺ shows typical emission peaks with its particular shape at 684 nm and ~710 nm that cannot be recognized here. Another possible explanation to the later emission can be attributed the Mn²⁺ in M1 site replacing Mg (Mg_VI).

Even if the observed Mn²⁺ emission peak is in accordance with Gorobets & Rogojine ^[3] data, the 759 and 775 nm emission peaks are slightly shifted to lower energies. The question rose here is to know whether the later peaks are only caused by Fe³⁺ ?

As the luminescence emission always occurs at lower energies than the energies that are required for excitation, using an excitation source at a given wavelength makes any absorption at a wavelength higher than that of the excitation source and lower to that of the emission peak a possible candidate as a cause for the emission.

As most of the absorptions of this diopside are located in the 400-500 nm and 600-700 nm ranges, using a 405 nm excitation source can bring to light that any of those energies absorptions are possible candidates for the emissions located above 700 nm.

A solution to reduce the set of the excitations causing the ~750 nm emission is to use an excitation source working at lower energies than one provided by the 405 nm laser. In the present case, the use of a 653-655 nm laser diode as a source is an interesting solution for studying the emission peaks in the 660-800 nm range. The figure 5 shows the acquired photoluminescence obtained with a 655 nm excitation, the emission around 755 - 775 nm is still observed. From that experiment, it is now possible to state that the Fe³⁺ emission is caused by some absorption of energies of wavelength included in the restricted range 655 - 750 nm. 

The 690 nm emission feature is also still noticeable with the 655 nm excitation.

As the 655 nm excitation produces the ~750 nm emission, a 678 nm laser diode is used to check whether that emission is produced. The acquired spectrum is available figure 6 showing again the ~750 nm emission, the maximum being around 765 nm. This result constraint the range of absorptions causing the Fe³⁺ emission to 680 - 750 nm and more reasonably to 680 - 720 nm.

Conclusion:

.

---

^[1] Yellowish Green Diopside and Tremolite from Merelani, Tanzania - Eric A. Fritz, Brendan M. Laurs, Robert T. Downs, and Gelu Costin

^[2] Diopside vanadifère et/ou chromifère vert pomme de Tanzanie - www.geminterest.com

^[3] Luminescent Spectra of Minerals, Boris S. Gorobets and Alexandre A. Rogojine, Moscow, 2002, ISBN: 5901837053

^[4] Luminescence Spectroscopy of Minerals and Materials, M. Gaft, R. Reisfeld, G. Panczer, Springer Editor, ISBN: 103540219188

^[5] Le Diopside, P. Tahiri-Jouti, 2008, DUG Nantes

^[6] Le vanadium trivalent (V3+) en tant que colorant dans les gemmes, E. Thoreux, 2011, DUG Nantes

^[7] Mineralogical Applications of Crystal Field Theory, R. G. Burns, 2005, ISBN: 0521017858 | 978-0521017855

^[8] Pyroxene Visible Spectra, CIT 7524 & RDS 62047-70 diopsides, California Institute of Technology