ADAMAS GEMOLOGICAL LABORATORY publishes gemologically oriented software for the jewelry appraiser, jeweler, or student in gemology or mineralogy. The laboratory also provides technical consulting and jewelry appraisal services to the trade and the public and expert witness services to the legal profession.

SAS2000: Developed, Produced And Supported In The USA   Questions, Comments, Product Support

On June 28th John Caruso and I, while examining various diamonds in the SAS2000 collection discovered what may turn out to be a unique diagnostic characteristic on near colorless synthetic diamond. We examined four (at the present time) Russian production near colorless synthetic diamonds, including early Chatham synthetic diamond to current production from unknown Russian sources. Each diamond, when examined with SAS2000 Raman Photoluminescence spectroscopy exhibited a 694nm characteristic, ranging from very very weak to very strong, which may turn out to be (and to our knowledge is) the FIRST spectral diagnostic characteristic found in NEAR COLORLESS SYNTHETIC DIAMOND.  We will ignore the saturated 1332wn Diamond Raman Line which is shown at approximately 573nm in this illustration.

Note: The apparent strength of the 694nm defect is FAR LESS than the strength of the characteristic diamond Raman signal at ~1332cm-1.

Note: The  Swiss Gemological Institute (SSEF) has reported in a private communication that they have seen the same feature at 693.4nm using a 514nm laser on samples purchased in Tucson last year.  Reviewing the data we have taken with the SAS2000 (data quantized at ~ 0.35nm) indicate a peak in the 693.6 to 694nm range.

NOTE:  This feature has now been seen in natural hydrogen rich diamonds!
 
 

 Tom Chatham of  Chatham Created Emeralds kindly supplied twenty four large near colorless rough synthetic diamond crystals produced a few years ago in Kiev for additional studies. These samples varied averaged over one half carat in size.  Each of these were checked at room temperature for the existence of the 694nm defect. Twenty one (21) of the twenty four (24) samples showed the 694nm feature at room temperature, with strengths varying from barely visible to very strong.

We also obtained nine smaller faceted near colorless synthetic diamonds supplied by  The Morion Company which are of recent Ukraine production. These stones averaged about 0.13ct in size each and were very very heavily included. Surprisingly, only two (2) of the nine (9) samples showed any 694nm features, however there may be a correlation between level of inclusions/contamination and the lack of fluorescence.

 Konstantin Iakoubovskii  of the Semiconductor Physics Laboratory, Katholieke Universiteit Leuven, Belgium had previously conducted studies on a diamond grown by Dr. Hisao Kanda . He indicated that "The sample is grown by him {Dr. Kanda} at NIRIM with a Co catalyst , but the growth chamber consists of NaCl+ZrO2 ...".  He also indicated that the sample was NOT colorless as there was no getter.

Mr. Iakoubovskii provided the following commentary on the 694 feature:

"I have seen the 694-nm peak in one sample of synthetic diamond grown by Hisao Kanda with pure Co metal catalyst. The sample consists of the main large (several millimeters) crystal (curve 1 in figure {below}) and some dozens of spontaneous crystallites smaller than 1 mm. The center was seen only in crystallites (curve 2). The strongest peak has the position at 1.7871 eV at room temperature and it is the sharpest PL feature at room temperature I have ever seen in diamond, only 1.5 meV half width! This doublet has a relaxation time of 7 ms and it is not seen in crystals grown with Ni-Fe catalyst by H. Kanda or B. Feigelson (Russia). Relying on the long relaxation time and doping results I suggest (but do not  insist) that it is Co related. "

We thank Mr. Iakoubovskii and Dr. Kanda for the permission to publish their findings here....

SAS2000 data shown below provides similar, but lower resolution evidence of the doublet shown above.

GIA is quoted in an article on this topic in Professional Jeweler "GIA, however, has not detected the feature. "The feature described is not present in the colorless synthetic samples we have observed and would need further investigation to determine its significance with regards to an identification criterion," a GIA spokesperson says."

Unfortunately, GIA's comment conveys the public image that GIA has conducted similar Raman - Photoluminescence spectra on all the near colorless synthetic diamonds they have seen, at least that's the way I read it.  I find this very puzzling, as I believe that GIA must have seen most, if not all, of Tom Chatham's synthetic diamond produced in Russia.  I have taken a fairly representative sample set (33) of that production and have seen the feature on 27 out of 33 samples.  Only taking statistics based on the 14 stones out of 46 total stones tested which had VERY VERY easily discernable features (moderate and above), the probability of NOT detecting the 694nm feature if you had tested 10 stones, would be 2.6%.  GIA's experience appears to be a statistical anomaly and I wonder how they could have missed it, given that they have seen far more samples than I have in total.  Maybe they only tested a very limited sample, or maybe they need an SAS2000!!!!!!!

With nine more samples of Tom Chatham's Russian synthetic diamond production added to our original 24, only  six (6) of thirty three (33) samples were without a measurable feature at 694nm.  Fifteen (15) of the thirty three (33) samples had 694nm features which were very easy to recognize as anomalous

Data taken on thirty three (33) out of forty six (46), or 71.7% of the sample set of synthetic near colorless diamonds produced in Russia or the former Russian republics have shown a systematic, apparently previously undocumented, photoluminescence feature at approximately 693.7nm.  Thirty percent (30%) of the total sample set had 694nm characteristics that literally could not be missed if tested with signitures in the moderate to very strong category. There appears to be a qualitative correlation between the existence of short wave fluorescence or phosphorescent sensitivity and the magnitude (or existence) of the 693.7nm photoluminescence feature found in these synthetic near colorless diamonds.  As to this qualitative correlation, it has been suggested that the relatively high level of impurities visibly evident in most of these samples (Fe ?, Ni ?, Ti ?, Al ? ) may serve to quench any short wave fluorescence or phosphorescence.  This feature may not exist in near colorless synthetic diamonds produced by methods atypical to the source(s) of the diamonds tested.
 
 

SAS2000 Liquid Nitrogen Immersion Spectroscopy (LNIS) techniques were applied to 3 GEPOL diamonds, an E, H, and K color.  The 575nm feature was visible using the SAS2000 differential analyses.  The 637nm N-V site was visible in the H and K color diamonds (also at room temperature) however, at present we were unable to detect any 637nm defect in the GEPOL E color diamond.  It has been suggested that it might not exist.
 
 

On May 22th, John Caruso and I took the first photoluminescence / Raman spectra with the SAS2000 Spectrophotometer Analysis System.  The SAS2000 external probe was linked to the SAS2000 external test chamber and the GE/POL diamond excited with a 532nm Green laser with about 50 milliwatts power level.  SAS2000 software was modified to process and display the Raman/photoluminescence signals.

The following is a Raman / Photoluminescence spectra of the new General Electric GEPOL High Pressure, High Temperature [HTHP] treated colorless type IIa diamond.  This diamond is of E color and has a GIAGTL report indicating that it had been subjected to a GE process enchancing its appearance. The diamond is laser inscribed GEPOL. The Raman data shown below has been corrected for dark current only and then normalized to the peak intensity seen.  SAS2000 differential processing is being added to enhance the Raman signal processing and remove unwanted low frequency background caused by the long pass filters currently being used in the SAS2000 Raman development.

The raw spectra shown above was optimally adjusted for theoretical background level caused by flourescence of the long pass filtering initially used with the adjusted data shown below. The low frequency background adjustment is shown below in yellow. 1st differential analysis (shown in green) indicated the presence of a weak line at 575nm, consistant with DeBeers data published in Gems & Gemology Spring 2000.  We did not see any evidence in this E color diamond of a 637N-V site, but according to published data in Gems & Gemology, this could be expected in type IIa diamonds. The 2467 wavenumber feature corresponding to about 613nm, was also seen in type Ia diamonds.
 
 

We also ran SAS2000 Raman on the GEPOL "E" color diamond at liquid nitrogen temperature, using SAS2000 LNIS immersion techniques, which will be modified to accomodate the longer integration times required in Raman data sampling.

Because of the interest in detecting GEPOL diamonds, and the recent article in the Spring 2000 Gems and Gemology indicating the presence of a 637nm N-V site in some off color GEPOL diamonds, we decided to test an irradiated purple-pink diamond which had a strong 637nm signature typical of these treated pinks.  The data shown below indicates the strong  photo-
luminescence/fluorescence bands excited by the 532nm laser on this diamond.  No 1332 raman line was seen on this diamond as it appears that the relative signal strengths of the Raman and any photoluminescence are not balanced as in the case on the near colorless GEPOL diamond.  We are obtaining several different higher wave length lasers for incorporation into the SAS2000 Raman suite.  More discovery to follow...

The adjusted raman data for this diamond shows a 637N-V site as well as strong photoluminescence bands at about 3690 and 4152 wavenumbers.  SAS2000 Raman will have a quantization level of about 5 wavenumbers, due to the approximate 0.35nm per pixel quantization of the spectrometers. It appears that this should be entirely sufficient for gemstone identification and treatment detection in the Raman domain.
 
 

We also experimented with a NovaDiamond HPHT treated diamond  with the SAS2000 Raman/Photoluminescence Spectra showing up the 1332 wavenumber diamond Raman line with differential processing, but almost hidden in the spectra due to the extreme photoluminescence of this treated diamond. The 1778 wavenumber feature corresponds to approximately 587nm,  the 3121 wavenumber feature corresponds to approximately the 637N-V center which is not visible as a defect in the visible spectra even at liquid nitrogen temperatures in this stone. The 3645 wavenumber feature is at approximately 659nm and the 4131 wavenumber corresponds to approximately 681nm.

Raman / Photoluminescence spectra may prove to reveal more information on trace elements for colored stone identification. The vis-near IR spectra (blue) of this yellow sapphire revealed a beneign spectra with only the fluorescence band at 694 due to Chromium, the photoluminescence spectra (shown in red) show features which I believe are related to Iron content.