Nick Davies, Sema Firat, David Fisher, Thoufieq Shaik
De Beers Science & Technology, De Beers Group, Belmont Road, Maidenhead, SL6 6JW, UK
Abstract
A 6.87 ct synthetic diamond has been analysed; it had been fashioned into an approximately octahedral shape similar to that of a natural rough diamond. Detailed visual examination of the stone revealed features inconsistent with a natural diamond and more detailed analysis confirmed that it was a synthetic diamond grown by chemical vapour deposition. This is the first report of such a stone and highlights that diligence is required in using traditional observational techniques when assessing rough diamonds, especially if more complex shaping methods were to be used in the future to further mask the identity of such synthetic diamonds.
Introduction
Natural rough diamonds typically have shapes between an octahedron and a rounded dodecahedron (or modifications thereof) with surface growth steps and so-called trigons due to surface etching (for a recent review see Harris et al. (2022)). These differ significantly from as-grown synthetic diamonds. Those grown by high pressure high temperature (HPHT) methods show characteristic cubo-octahedral morphologies (Koivula and Fryer, 1984; Sunagawa, 1984; Welbourn et al., 1996) whilst chemical vapour deposition (CVD) growth results in tabular or blocky crystals (Wang et al., 2003) quite unlike natural diamonds. Rough natural diamonds can therefore be confirmed to be neither synthetic diamonds nor simulants through careful observation of the crystal morphology and surface features. Simulants have been fashioned into shapes similar to those of rough diamonds in an attempt to mask their identity and potentially pass them off as natural diamonds under situations where stones are not being carefully assessed. Examples of cubic zirconia (Crowningshield and Moses, 1996), topaz (Sehgal and Befi, 2015), phenakite (Beaton et al., 2007), sapphire (Ahmed and Singbamroong, 2009) and Moissanite (Robb and Arden, 2022) have been reported. To date there have been no reports of synthetic diamonds being fashioned into the shape of a rough natural diamond, presumably due to the difficulties associated with carrying out such modifications. However, the precipitous drop in the value of synthetic diamonds does incentivise attempts to produce one that could be mistaken for its much more valuable natural counterpart.
Sample
The 6.87 ct colorless crystal shown above in figure 1 was passed to us for assessment. Those referring the stone were confident that it was diamond as confirmed by a Raman measurement, but there were concerns over the general appearance and “feel” of the stone in the hand. Their experience in handling of diamonds indicated that the stone’s “heft” (i.e. its assessed density) was consistent with that of diamond, suggesting that this was not a simulant.
Visual Observations and Microscopy
The stone has an approximately octahedral shape with flat faces and some evidence of rounding at the junctions between the faces. Closer examination of the shape of the stone showed that it was not a genuine octahedron. Whilst uneven growth can lead to “stretching” of the octahedral shape in many natural diamonds, the opposite faces on such stones should remain parallel. In this case, only two of the faces were found to be parallel and the angles between faces were inconsistent with a genuine octahedron (Figure 2).
The faces themselves showed very little by way of features. Observation under a microscope using differential interference contrast showed them to be extremely flat with the only significant features being series of parallel polishing lines (Figure 3). Microscopy also allowed the rounding of the junctions to be examined in detail (Figure 4). These regions were shown to be extremely rough. Our own experiments indicate that these features are consistent with those seen at facet junctions on polished diamonds that have been subjected to high enough temperatures to induce a degree of graphitisation. The graphitisation will not have been confined to these junctions and would have extended to cover the flatter faces of the stone. This suggests that the stone’s faces have been repolished to remove any graphite on them that was produced during the heat treatment stage.
Crystallography
X-ray Laue diffraction was used to confirm the orientation of the faces. In a natural diamond octahedron, all eight faces would be parallel to {111} planes. None of the faces of this stone were parallel to {111} planes. Two of the opposite faces were parallel to {100} planes, whilst the other faces showed no specific orientation. The presence of two parallel {100} faces is consistent with the preparation of a stone from a block of CVD-grown synthetic diamond. The favoured direction of growth in CVD synthesis is in a (100) direction on a substrate whose planar face is in a {100} plane. An octahedral block maximising the height of the as-grown material could be prepared from this with two {100} faces aligned with the substrate interface and the final growth surface.
Diamond Type
Infrared absorption measurements indicated that this is a type IIa diamond, i.e. one that contains no detectable nitrogen-related absorption. Around 1 to 2 % of natural diamonds are type IIa but these are usually irregular in shape (Sunagawa, 2001; Wilks and Wilks, 1991). Natural octahedral diamonds with a well-formed crystal habit are almost always type Ia so the type and morphology combination observed for this stone is inconsistent with a natural diamond. All colourless synthetic diamonds are type IIa or very weak type IIb due to the presence of a small concentration of uncompensated boron.
DiamondView Imaging
The stone was definitively identified as a CVD synthetic diamond through DiamondView imaging of the flat faces (Figure 5). The background luminescence showed bands of orange emission of varying intensity. This emission is from nitrogen-vacancy centres incorporated during the CVD growth process. The bands are consistent with a stop-start growth process where periods of reasonably consistent growth are interrupted either by a significant change in growth parameters or possibly the stopping of the process to reposition the growing stone with respect to the plasma in the growth chamber. Within the coarse growth bands, it is also possible to see growth striations. These can be more clearly seen in Figure 6, where the green filter provided with the DiamondView has been inserted in front of the camera to remove the blue emission from the dislocations and enhance the orange emission from the striated regions. Such striations have been documented previously for CVD synthetic diamonds (Martineau et al., 2004) and are due to step flow growth where there is differential uptake of impurities at the terraces and risers of the steps.
Also evident in the image are “plumes” of blue luminescence that thread through the coarse bands roughly perpendicular to them (similar features are seen in Figure 3 of Wang et al. (2022) and Figure 11 of Eaton-Magaña et al. (2024)). These are associated with dislocations in the boundaries between small regions of the crystal that have grown with small misorientations with respect to each other. Such features of columnar growth have been widely reported for CVD synthetic diamonds (Martineau et al., 2004; Gaukroger et al., 2008). The dislocations run approximately parallel to the growth direction and form a network where they intersect the final growth surface. This can be seen in Figure 7 which is the view of the face adjacent to and parallel with the final growth surface. This region has very weak orange emission and the only significant features are the blue dislocation networks.
Screening Instruments
Diamond Verification Instruments (DVIs) are generally configured to analyse polished diamonds where any morphological indication of a diamond’s synthetic origin has been removed. Many of these techniques could be applied to stones such as this one due to its flat faces, although they would not always be suitable for routine screening of large volumes of natural rough diamonds. This stone was tested on a range of DVIs to confirm their ability to screen for CVD synthetic diamonds fashioned in this way. The results are shown in the table below.
Diamond Verification Instrument or Technique | Result | Successfully screened (Y/N) |
De Beers DiamondSure | REFER (type IIa) | Y |
De Beers DiamondView | Clear CVD growth features | Y |
De Beers DiamondPLus | REFER (CVD SYNTHETIC?) | Y |
De Beers SYNTHdetect | REFER | Y |
De Beers DiamondProof | REFER | Y |
GIA iD100 | Refer | Y |
Magilabs EXA | Refer | Y |
Gemlogis Belize | Type IIa HPHT CVD | Y |
lw/sw UV lamp | lw: orange, sw: orange | Y |
Conclusions and Identification
The stone presented in this work was a synthetic diamond grown by chemical vapour deposition. After growth, an approximately octahedral shape has been produced with two of the faces aligned with the growth and interface planes for ease. This shaping will almost certainly have been carried out using a laser saw. The resulting surfaces will have been covered in a layer of black non-diamond carbon. These surfaces appear to have then been cleaned by heat treatment in an oxygen-containing atmosphere, producing the observed rounding of the junctions between the faces. Polishing of the faces has then been carried out to produce the final stone.
The density of synthetic diamond is exactly the same as natural diamond so any assessment of the “heft” of the stone will not provide any indication that the stone is synthetic diamond. Visual examination beyond a quick assessment reveals a range of features that are inconsistent with a natural rough diamond, including deviation from an octahedral shape, smooth faces lacking growth steps and trigons and rounded junctions between faces. Academic research has been carried out since the 1960s into etching of diamonds in order to better understand natural etching process and replicate the surface features seen on natural rough diamonds (see for example Khokhryakov and Palyanov (2007)). Such processes could in theory be applied to stones such as this one to give a more natural appearance, but to date they have only been applied to relatively small research samples. Also, the fact that the current stone is not a genuine octahedron means that the etch features would differ from those produced on initially octahedral natural diamonds. For instance, trigons are only generated on {111} faces and no {111} faces were present on this stone. The polishing of faces parallel to {111} planes is also, in practice, very difficult (Watermeyer, 1991). Even producing a stone such as this one from a block of CVD material grown in a (100) direction requires careful processing and any resultant polished diamond would have a much lower yield than one produced directly from the as-grown block of material. The only motivation for such an action would appear to be a deliberate attempt at deceit.
The application of standard synthetic diamond screening techniques to this stone has shown that it can be routinely identified as a synthetic diamond and this has been confirmed by more detailed analysis, in particular DiamondView imaging to reveal growth structures. The material used in this case is now very familiar and typical of large synthetic diamonds being produced around the world. Care needs to be taken at any point in the diamond pipeline where there is a possibility of switching such a synthetic diamond with its natural counterpart and only a quick visual assessment is taking place. There is currently no evidence of any widespread switching in the rough diamond pipeline. Should this become more prevalent, an immediate screening solution for rough diamonds could be implemented through adaptation of existing screening instruments for polished diamonds.
References
Ahmed N., Singbamroong S. (2009) Colorless synthetic sapphire imitating rough natural diamond. Gems & Gemology, Vol. 45, No. 3, pp. 230-231.
Beaton D., Sheby J., Befi R. (2007) Phenakite as a rough diamond imitation. Gems & Gemology, Vol. 43, No. 3, p. 250.
Crowningshield G.R., Moses T. (1996) Lab Notes: Imitation crystals. Gems & Gemology, Vol. 32, No. 3, p. 205.
Eaton-Magaña S., Matthew F. Hardman M.F., Odake S. (2024) Laboratory-grown diamonds: An update on identification and products evaluated at GIA. Gems & Gemology, Vol. 60, No. 2, pp.
Gaukroger M.P., Martineau P.M., Crowder M.J., Friel I., Williams S.D., Twitchen D.J. (2008) X-ray topography studies of dislocations in single crystal CVD diamond, Diamond & Related Materials, Vol. 17, No. 3, pp. 262-269.
Harris J.W., Smit K.V., Fedortchouk, Y., Moore M. (2022) Morphology of monocrystalline diamond and its inclusions. Reviews in Mineralogy & Geochemistry, Vol. 88, pp. 119-166.
Khokhryakov A.F., Palyanov Y.N. (2007) The evolution of diamond morphology in the process of dissolution: experimental data. American Mineralogist, Vol. 92, pp. 909-917.
Koivula J.I., Fryer C.W. (1984) Identifying gem-quality synthetic diamonds: An update. Gems & Gemology, Vol. 20, No. 3, pp. 146-158.
Martineau P.M., Lawson S.C., Taylor A.J., Quinn S.J., Evans D.J.F., Crowder M.J. (2004) Identification of synthetic diamond grown using chemical vapor deposition (CVD). Gems & Gemology, Vol. 40, No. 1, pp. 2-25.
Robb C., Arden S. (2022) A synthetic Moissanite presented as natural diamond rough. Gems & Gemology, Vol. 58, No. 3, pp. 361-362.
Sehgal A., Befi R. (2015) Polished freeform topaz imitating diamond rough. Gems & Gemology, Vol. 51, No. 3, pp. 321-322.
Sunagawa I. (1984) Morphology of natural and synthetic diamond crystals. In I. Sunagawa, Ed., Materials Science of the Earth’s Interior, Terra Scientific Publishing Co., Tokyo, pp. 303-330.
Sunagawa I. (2001) A discussion on the origin of irregular shapes of type II diamonds. Journal of Gemmology, Vol. 27, No. 7, pp. 417-425.
Watermeyer B. (1991) Diamond Cutting (Fourth Edition), Basil Watermeyer, Johannesburg, p. 20.
Wang W., Moses T., Linares R.C., Shigley J.E., Hall M., Butler J.E. (2003) Gem-quality synthetic diamonds grown by a chemical vapor deposition (CVD) method. Gems & Gemology. Vol. 39, No. 4, pp. 268-283.
Wang W., Persaud S., Myagkaya E. (2022) New record size for CVD laboratory-grown diamond. Gems & Gemology, Vol. 58, No. 1, pp. 54-56.
Welbourn C.M., Cooper M., Spear P.M. (1996) De Beers natural versus synthetic diamond verification instruments. Gems & Gemology. Vol. 32, No. 3, pp. 156-169.
Wilks E.M., Wilks J. (1991) The morphology of diamond. In Properties and Applications of Diamond (Butterworth Heinemann Ltd., Oxford), pp. 108-144.
