MAGI Application note:

Detecting HPHT treatment of natural type IIa colorless diamonds

Sep. 3. 2013 Mikko Åström, Alberto Scarani, Marco Torelli

HPHT (High Pressure High Temperature) treatment can be used to enhance the color grade of brownish or grayish type IIa diamonds to colorless or near colorless. Most of the larger HPHT-treated diamonds has been sold branded as GE-POL, Bellataire, Pegasus or Monarch and can be readily identified by laser inscription located on the girdle of the stone. Unfortunately it is possible to remove these laser markings by repolishing the girdle, and small stones may have entered the market without any inscription at all.

Photoluminescence spectroscopy is one of the very few methods available for determining HPHT treatment of colorless type IIa diamonds. However, this method can not be used alone without other instrumentation, because as a preliminary requirement, the sample under study has to be determined as natural type IIa diamond.

Important note: This application note applies only to colorless or near colorless diamonds.

Partially repolished "HPHT-processed"-inscription on the girdle of type IIa natural HPHT treated diamond.

Preface

About 2% of natural diamonds belong to type IIa, which by description do not have enough nitrogen impurities to be detected by FTIR (Fourier-Transform Infra Red) spectroscopy. This means about 98% of colorless to yellow (cape series) diamonds can be determined as natural and likely not HPHT treated by single or combination of following methods:

1.) Detecting the N3 absorption peak at 415 nm by optical spectroscope or spectrometer.
2.) Detecting IaA and/or IaB- type nitrogen defect peaks by FTIR spectroscopy.
3.) Observing lack of SWUV transparency by a screening device or UV-VIS-NIR spectrometer

If above tests indicate the sample is of type IIa it's origin may be one of the followings:

a) Natural untreated type IIa diamond
b) Natural HPHT color enhanced type IIa diamond
c) Synthetic type IIa diamond manufactured by HPHT process
d) Synthetic type IIa diamond manufactured by CVD process

There is no single tool capable of distinguishing all the possible origins of colorless IIa diamonds. A combination of multiple techniques is mandatory in order to achieve for reliable results. These include, but are not limited to: microscopic study of anomalous double refraction, deep UV fluorescence examination with DiamondView or similar devices, fluorescence microscope & spectrometer and photoluminescence studies at room temperature and in liquid nitrogen immersion with a sensitive Raman spectrometer such GemmoRaman-532.

Microscopic examination of ADR:

Anomalous double refraction (ADR) in diamonds is caused by internal stress of the crystal. It is observed at the microscope between crossed polarizer filters.

• Natural untreated IIa	Tatami pattern
• Natural HPHT-treated IIa:	Tatami pattern
• Synthetic IIa, HPHT method:	No ADR
• Synthetic IIa, CVD method:	ADR visible, but does not look like typical tatami pattern.

Tatami pattern of natural HPHT-treated type II diamond between crossed polarizers.

			SWUV and LWUV examination & magnetism Synthetic type IIa diamonds exhibit stronger fluorescence in SWUV than LWUV excitation. For natural stones the reaction is opposite. HPHT synthetic diamonds have been grown in metal flux compounds. In many cases flux relics remains trapped inside the crystal, actually less than 10% of these diamonds are VVS clarity grade. Since these flux compounds are ferromagnetic elements (Ni, Co), even a small inclusion can result the diamond to be attracted by a magnet. A fast and easy screening can be performed by using a cheap neodymium magnet. We had an opportunity to test a very large parcel of 8-30 points synthetic HPHT diamonds, they were more or less all in the VS-SI clarity range, more than half of the stones sticked to the magnet only by hovering it on them.
Synthetic (HPHT) type IIa colorless diamonds picked up by a neodymium magnet.			CVD synthetic diamonds grow in a very different environment compared to HPHT, no metal flux is added during the process so eventual inclusions are not ferromagnetic.

DiamondView or similar fluorescence microscope

Virtually all diamonds fluoresce in Deep-UV excitation. When wavelength of UV radiation is less than 225 nm it does not enter into diamond but reveals it's growth structure on the surface.

• Natural untreated IIa:	Typically blue fluorescence with dislocations network.
• Natural HPHT treated IIa:	Typically blue fluorescence with dislocations network.
• Synthetic IIa, HPHT method:	Whitish blue fluorescence, cubo-octahedral growth zones. Long lasting blue phosphorecence.
• Synthetic IIa, CVD method:	Blue, green or orange fluorescence with curved striations, best seen from pavilion. Observable phosphorecence.

Curved striations in type IIa synthetic CVD diamond

Photoluminescence studies with GemmoRaman-532 or GemmoRaman-532SG

Before proceeding

The reason for all above tests in the preface section of this application note is that the following photoluminescence tests can only be applied to natural type IIa diamond. IaB natural diamonds are indeed extremely rare and, as the IIa type, can be successfully submitted to HPHT color enhancing treatment. Unfortunately, due their rarity there is not enough material available to perform statistically valid PL studies.

Room temperature (297K)

Spectral acquisition at room temperature (RT, 297K) is the more convenient method of the two. In the case of colorless IIa diamonds it provides surprisingly good results. Therefore, it is suggested to perform PL tests first at room temperature and later in liquid nitrogen immersion (LNT, 77K).

Untreated stones

Natural untreated diamonds of type IIa exhibit a series of typical small photoluminescence peaks which tends to be annealed (removed) by HPHT treatment.

Figure 1. Photoluminescence peaks at 536 nm, 567 nm, 579-580 nm, 587 nm and 596 nm at room temperature are very strong indicators a type IIa natural diamond has not been HPHT processed.

Most untreated diamonds exhibit nitrogen-vacancy crystallographic defects in their uncharged state (N-V⁰) at 576 nm in room temperature. This defect can also be negatively charged (N-V^-) causing a PL peak at 637nm. Strong peak at 576 nm combined with absence of 637nm peak can be an additional non-diagnostic clue that the sample has not been HPHT processed.

HPHT-treated stones

HPHT treatment removes or severely attenuates most of the PL peaks present in untreated stones. Their spectrum may lack any PL features, leaving only first and second order diamond Raman peaks visible. In some cases Nitrogen-Vacancy defects may be visible as broad bumps at 576 nm (N-V⁰) and/or 637 nm (N-V^-).

Figure 2. HPHT treatment has annealed most of the photoluminescence peaks detected in untreated stones. While N-V peaks at 576 and 637nm anneal out in commercial high temperature HPHT-treatment, new N-V defects may formed if the diamond contains minor type Ib component. In that case N-V- peaks at 637 nm are typically stronger than N-V0 peaks at 576 nm, but this "rule" is not diagnostic and has proved to have many exceptions.

Crystallographic defects causing the above mentioned PL peaks for natural untreated diamonds are not fully understood. However, their tendency to anneal at high temperatures required for removing color from type IIa diamonds serves as very good tool for detecting the treatment.

Comparison of room temperature spectra

Looking at both treated and untreated samples in same figure reveals the differences in overall appearance of the spectra.

Figure 3. HPHT-treated and untreated natural type IIa diamonds plotted on same graph for revealing the obvious differences in overall shape of the spectra. All the lines has been vertically shifted for visual convenience.

Liquid nitrogen immersion, LNT (77K)

Cooling diamond sample near to liquid nitrogen temperature (LNT) yields stronger photoluminescence peaks, narrows down their spectral width and may reveal additional peaks which are not detected at room temperature. Cooling also shifts the wavelenght of PL peaks towards blue end of the spectrum.

Basically, all the diamonds which does not contain large inclusions or severe cracks can survive the rapid temperature changes involved in liquid nitrogen immersion. No special precautions are required to be taken before the immersion but, hoovering the stone just above the liquid nitrogen level for a while could slightly reduce the thermal shock.

Handling liquid nitrogen safely is out of scope of this application note. However, a very small amount of liquid is needed (about 5ml) when working with GemmoRaman-532. We supply separate instructions for safe handling of LN for the customers purchasing the LN setup.

Untreated stones

Figure 5. LN temperature reveals some additional PL peaks which are not visible at room temperature. Peaks at 558 nm, 566 nm, 569 nm, 575.8 nm and 579 nm are strong indicators type a IIa natural diamond was not submitted to HPHT treatment. Of course not all those peaks may be present in every untreated IIa and their strength can vary a lot. 558nm peak is destroyed even in relatively low temperatures, but 575.8nm and 579 nm peaks are known to survive even through higher temperature treatments. A solitary peak in the 560-580nm range is actually an indicator of treatment.

HPHT-treated stones

Similar to room temperature, LN immersion reveals most PL peaks are annealed at HPHT conditions. While commercial treatment temperatures generally destroys N-V defects, new ones may be formed depending on original concentration of single nitrogen defects and vacancy clusters in the stone. Therefore, both treated and untreated diamonds may contain N-V defects. Sometimes 637 nm peak can be very strong in HPHT treated stones, while in pristine IIa high color grade stones its intensity is almost negligible.

Figure 6. Typical spectra of HPHT-treated colorless type IIa diamond acquired in LNT shows only diamond Raman features and possibly N-V defects at 574.8nm and/or 637 nm. Strong 637 nm peak is a typical product of originally highly dislocated (darker brown) diamond. It is important to note, that some of the treated stones does not reveal any PL peaks, a very unlikely situation for untreated stones.

Comparison of LNT spectra

Looking at both treated and untreated samples in same figure reveals the differences in overall appearance shape of their LNT spectra.

Figure 7. HPHT-treated and untreated natural type IIa diamond LNT spectra plotted on same graph for revealing the obvious differences in overall appearance. All the lines has been vertically shifted for visual convenience.

References

Inga A. Dobrinets, Victor G. Vins, Alexander M. Zaitsev (2013) HPHT-treated Diamonds, ISBN 978-3-642-37489-0

We highly recommend this new publication for every gemologist interested about deep understanding of HPHT- treatment and it's detection. The book is in our opinion the most comprehensive review of dispersed publications written in past 15 years.