Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
WO 94/20837 ~ ~ ~ PCT/GB94/00415
1
DI STI NGUI SHI NG NATURAL FROM SYNTHETI C DI AMOND
Background of the Invention
This invention relates to a method and apparatus for
distinguishing natural diamond from synthetic diamond by
observing the luminescence, for example by investigating
the arrangement of growth sectors in the diamond.
Synthetic diamonds differ from natural diamonds in that
synthetic diamonds show mixed habit growth, having
different growth sectors through the body of the
crystal. These different growth sectors all incorporate
impurities as they grow, but do so at different rates
and in different ways, and this gives them different
spectroscopic properties. In an unpolished stone, the
presence of different growth sectors in synthetic
diamonds will produce characteristic growth faces on the
surface of the diamond quite distinct from those formed
on the surface of a natural diamond, which will tend to
exhibit single habit octahedral growth. A skilled
person can identify an unpolished synthetic diamond from
a natural di amond j us t by 1 ooki ng at i t, but al l thi s
surface information is removed if the stone is polished.
WO 94/20837 ~ PCT/GB94/00415
2
A technique known as cathodoluminescence has been
developed and discussed in Woods & Lang (J Crystal
Growth vol 28 (1975) page 215), Burns et al (J Crystal
Growth vol 104 (1990) page 257), Shigley et al (Gems and
Gemology, vol 23 (1987) page 187), and Marshall
("Cathodoluminescence of Geological Materials", 1988,
published by Unwin Hyman, pages 19 to 36). This
technique involves subjecting a diamond to an electron
beam in an evacuated cathode ray chamber. With a
typical energy of 18 keV, electrons will penetrate and
cause excitation in a surface region to a depth of about
3~.m. Luminescence known as cathodoluminescence will
be generated in this region. An image of the surface
being excited can be then formed, this image showing the
cathodoluminescence pattern.
Ponahlo (J Gemology, vol 21 (1988) page 182) describes
the use of the cathodoluminescence technique for
distinguishing natural from synthetic emeralds and
rubi a s .
The cathodoluminescence technique has significant
disadvantages in the form of the apparatus required.
The gemstone has to be placed in a vacuum chamber, which
is expensive and increases the time required to make a
measurement, and the electron beam generates X-rays
which have to be screened. In addition, the cathode ray
apparatus itself is expensive.
CA 02157469 2004-10-08
3
It is desirable to provide a method of distinguishing
natural diamond from synthetic diamond without using
complex and expensive apparatus or vacuum chambers.
Shigley et al (supra) discloses a method of short wave
ultraviolet illumination of synthetic diamonds to study
growth sectors, using ultraviolet radiation of a
wavelength of 254nm. This causes excitation into
impurity energy levels) specific to the particular type
of diamond being studied. The technique described in
this disclosure will only work with those diamonds that
have a strong extrinsic absorption at 254nm.
It is desirable to provide an observation technique that
will work with a single wavelength or single band of
wavelengths for all diamonds.
A paper by Welsh et al (Journal Of Luminescence, Volume
4 (1971) page 369) describes the thermoluminescence
technique in natural and synthetic semi-conducting
diamonds, and relates it to phosphorescence phenomena.
Thermoluminescence is the thermal generation of
luminescence following excitation at a low temperature,
for example 77R. The sample is excited using an
electron beam or photo-excitation from an arc lamp.
Upon raising the temperature of the sample,
thermoluminescence peaks at different temperatures are
observed. Phosphorescence may also be observed.
..
~~.~'~~~
Phosphorescence is the decay of stimulated luminescence,
following the removal of the excitation source, over a
period of milliseconds to tens of seconds and sometimes
over a period of minutes.
Thermoluminescence apparatus is complicated and
expensive and may not be suitable for differentiating
synthetic diamonds and natural diamonds.
The Invention
The invention provides a method and apparatus for
distinguishing natural from synthetic diamond as set out
in Claim 1, 9, 16 or 17, preferred and/or optional
features of the method and apparatus being set out in
the Claims 2 to 8 and 10 to 15.
Using the invention, it is possible to deter:,~::~.ne for
most diamonds whether the diamond is natural or
synthetic, employing apparatus which can be portable and
can be of a cost and size which make it usable in say
small-scale gemological laboratories, jewellery
wholesale manufacturers and large jeweller retail
establishments. Although ultraviolet of a wavelength
less than 225 nm is dangerous to the eyes and skin if it
escapes, complete shielding can be provided. Although
normal optics cannot be used for delivering the
ultraviolet to the sample, as they would attenuate and
,AMENDED SHEET
WO 94/20837 ~ ~ ~ ~ /~ ~ PCT/GB94/00415
absorb the ultraviolet of a wavelength less than 225 nm,
it is possible to provide special optics which do not
' grossly attenuate the ultraviolet.
The diamond is observed under ultraviolet excitation of
such a wavelength that the surface structure can be
distinguished, possibly with the assistance of the
different colours of luminescence produced by different
diamonds. A permanent image or surface topograph of the
diamond can be formed, preferably magnified; however the
method is preferably operated by observing the diamond
by eye through a microscope. Once the eye has
accomodated itself to a low light level, a good image is
s een.
Ultraviolet radiation of wavelengths shorter than 225 nm
interacts strongly with all types of diamond to cause
luminescence, and thus the method and apparatus of the
invention are suitable for use with any form of
diamond. The luminescence bands observed for various
types.of diamond (natural or synthetic) fall within a
wide range of wavelengths, generally in the visible part
of the spectrum. For instance, one can identify
patterns indicating the disposition of growth sectors in
the diamond and hence the types of growth sectors
~ present. Sectors of different growth type will luminesce
with different colours, and may thus be distinguished.
Growth sectors of the same type will have the same
WO 94/20837 PCT/GB94/00415
luminescence colour but may show different intensities.
In this way, natural diamond can be distinguished from
synthetic diamond as synthetic diamonds will in general
show more than one type of growth sector. The method
can also enable one to detect a surface deposition of
synthetic diamond on a natural diamond. The invention
is particularly useful for examining worked stones, i.e.
fully polished or part-polished stones.
If it is difficult to distinguish natural diamond from
synthetic diamond by looking at the growth sectors,
particularly if there is only a single growth sector
which gives little topographical information, the colour
can usually enable the observer to identify whether the
diamond is natural or synthetic.
General luminescence microscopy is the technique of
illuminating a diamond with.ultraviolet light and
observing through a microscope the characteristics of
the stimulated luminescence from the bulk of the sample
diamond, the whole of the diamond being excited.
General luminescence microscopy is not suitable for
studying the geometrical forms of the luminescence; even
if a limited plane is focused upon, the rest of the
luminescence severely reduces the contrast. If a
composite image for the whole diamond were generated,
the image would be very hard to interpret. However, the
,", WO 94/20837
PCT/GB94100415
7
success of such a technique would be determined by the
type of diamond material and strength of the bands
emitted, and would vary from diamond to diamond.
In the invention, the exciting radiation does not
penetrate very deeply into the crystal and substantially
only the surface region is penetrated and irradiated.
In this way, the surface region of the diamond may be
observed. Luminescence emitted by regions of the
diamond which are deeper than the surface region must be
insufficient to render indistinct the luminescence
produced by the surface regions; in other words that
irradiating radiation which induces luminescence must be
substantially absorbed in the surface region.
The degree of attenuation of the ultraviolet radiation
within the diamond and hence the irradiated depth is
dependent on the coefficient of absorption of the
diamond at the irradiating wavelength; the attenuation
follows an inverse exponential behaviour with depth.
When using radiation of a wavelength longer than 225 nm,
for certain types of diamonds the absorption is not as
strong as at 225 nm and the depth of diamond illuminated
is much greater. The actual depth of diamond
illuminated at a wavelength longer than 225 nm depends
on the magnitude of the extrinsic defect (nitrogen)
induced absorption at the exciting wavelength, which
will vary considerably between diamonds. Thus, the
W~ 94/20837 PCT/GB94/00415
8
induced luminescence may be generated all through the
body of the crystal (which will be the case if the
absorption is sufficiently weak) and the contrast of the
images produced at the surface may be swamped.
At wavelengths less than about 225 nm, the absorption
coefficient of all types of diamond is very high and the
absorption coefficient of diamond increases rapidly with
decreasing wavelength. In the absence of absorption
from impurities, the absorption coefficient at about 225
nm is about 50 cm 1 and at 223 nm it is greater than
400 cm 1. The limit of absorption spectroscopy is
achieved at about 208 nm with an absorption coefficient
of about 5000 cm 1 - below about 208 nm, the
absorption coefficient is too large to be measured. The
"attenuation depth" at a given wavelength is the depth
at which the radiation has been attenuated to about 13~
of its incident intensity. This is given by 2/A, where
A is the absorption coefficient at a given wavelength.
From 225 nm to 208 nm the attenuation depth has
decreased from 400 ~.m to 4 ~.m, being 50~.m at 223
nm. In general, in all diamonds, using a suitable
filter allowing only radiation with wavelengths below
225 nm to be incident on the diamond, most of the
radiation would be substantially absorbed within a depth
of about 50~m of the surface of the diamond - more
specifically, the effective penetration depth for a
uniform intensity distribution from 225 nm to 190 nm is
CA 02157469 2004-10-08
D
9
substantially less than 50 Vim. Thus the surface
region can be considered to have a depth of about 50
Vim, it being preferred that the irradiating
ultraviolet radiation of a wavelength down to 190 nm, be
attenuated to about 13% of the incident intensity (i.e.
integrated over wavelength) within this depth; this
depth is satisfactory provided that the intensity of any
illumination having wavelengths above 223 nm or 225 nm
is insignificant at a depth of not many times, e.g.
three or four times, greater than 50 Vim. Ideally,
most of the irradiating radiation should be of a
wavelength just slightly less than 225 nm.
To avoid excessive luminescence from deeper regions, the
irradiating source should not contain any high intensity
visible radiation (radiation above 380 nm) that could be
transmitted to the body of the diamond to cause
luminescence. Moreover, the absence of visible
excitation means that there would be no contamination of
the luminescence produced in the surface regions from
scattered visible excitation light. Ideally, the
radiation should be confined to ultraviolet radiation
below 225 nm. Nonetheless, in practical embodiments
some radiation of over 225 nm is permitted and as a
general statement, the intensity of the irradiating
radiation above 225 nm and' up to 380 nm (the limit of
visible radiation) is preferably not more than about 50~,
or not more than about 250, or not more than about 150, or
not more than about 10%, or not more than about 5~ of
WO 94/20837 PCT/GB94/00415
the irradiating ultraviolet radiation below 225 nm. The
oxygen in air cuts off radiation below about 180 nm and
ultraviolet optics usually cut off radiation below about
180 or 190 nm. Thus, in considering the radiation above
225 nm and up to 380 nm as a percentage of the
ultraviolet radiation below 225 nm, radiation below say
190 nm can be ignored.
Natural diamond can be distinguished from certain types
of synthetic diamond merely by irradiating as set forth
above, terminating the irradiation, and observing the
phosphorescence. In general, merely observing the
phosphorescence will not identify a number of different
types of diamond, and this technique is preferably used
as a step additional to that of observing the
luminescence, when it can provide additional
di s c ri mi nati on.
The phosphorescence is found to originate only from
certain growth sectors in synthetic diamonds and lasts
long enough to be observed, for up to several minutes.
Phosphorescence in natural diamonds is a rare phenomenon
and is almost exclusively found in diamonds of type IIb
character, which contain an excess concentration of
boron over nitrogen impurities. These diamonds are
generally blue in colour and have semi-conducting
properties. On the other hand, phosphorescence is a
common phenomenon in synthetic diamonds having a low
~ - ~'~4~~ .
m
nitrogen concentration, including those which are
colourless or near colourless and those doped with boron
to produce a blue colour. Stones are either identified
as natural (no phosphorescence) or are referred for
further testing (phosphorescence). Although the
technique cannot distinguish between natural and
synthetic blue-coloured diamonds, natural blue diamonds
are very rare so that only a small number of natural
stones will be referred for further testing.
Observing the phosphorescence is useful in two ways;
i) In the absence of any growth sectorial information,
if emissions persist for a perceptible time after
removing the exciting ultraviolet radiation, the
diamond should be suspected as being of synthetic
origin;
ii) If the growth sectors are poorly defined due to the
luminescence emissions being of a similar colour,
the contrast between the sectors may be enhanced
when the ultraviolet radiation is removed; those
growth sectors which do not phosphoresce will appear
black.
AMEPlDEJ SHEET
~ _ _
12
It has been found that the observation of the
luminescence caused by the ultraviolet enables one to
detect a layer of synthetic diamond deposited upon the
stone. This method is suitable for detecting synthetic
diamond deposited on natural diamond, and is suitable
for worked and especially fully polished stones.
ANI~~~~~~ Sh~cT
CA 02157469 2004-10-08
13
In all cases, the intensity of the irradiation must be
sufficient to obtain observable luminescence.
The invention will be further described by way of
example and with reference to the accompanying drawings,
in which:
Brief Description of the Drawings
Figure 1 shows the components of an ultraviolet
photoluminescence topography apparatus according to the
invention, and shows a possible modification;
Figure 2 is an isometric projection of the apparatus of
Figure 1;
Figures 3 and 4 are diagrammatic representations of the
major features of examples of ultraviolet
photoluminescence topographs of synthetic diamonds,
obtained using the method and apparatus of the invention;
CA 02157469 2004-10-08
14
Figures 5 and 6 show diagrammatic representations of
the major features of examples of ultraviolet
photoluminescence topographs of natural diamonds,
obtained using the method and apparatus of the
invention; and
Figure 7 is a diagrammatic representation, corresponding
to Figures 3, 4, 5 and 6, but showing the major
features of a natural diamond having a surface
deposition of synthetic diamond.
Preferred Embodiments
Figures 1 and 2
Figure 1 shows the layout of apparatus for ultraviolet
photoluminescence topography according to the
invention. A polished diamond 1 is mounted in mounting
means 2. The diamond 1 and mounting means 2 are shown
to a larger scale than the rest of the apparatus. Any
suitable mounting means 2 can be used, such as
spring-loaded tweezers, vacuum tweezers or a conical
depression. The mounting means 2 shown, purely as an
example, are spring-loaded tweezers with rubber pads for
gripping the diamond 1 and a protruding handle 3 for
manipulation of the diamond 1, with a rubber bellows 4
CA 02157469 2004-10-08
for ultraviolet shielding. The girdle of the diamond 1
can be gripped (as shown) for some views, and the table
and culet point gripped for other views.
The diamond 1 is illuminated with short wavelength
(225nm or less) ultraviolet radiation produced by a
light source 5. A water-cooled Hamamatsu 150 W
deuterium light source may be used, but the preferred
source is a W rich xenon flash lamp, which contains all
wavelengths from 190 nm to beyond 1000 nm. The
preferred xenon lamp illumination system consists of the
xenon lamp unit 5, which comprises a xenon lamp, a
capacitor and a trigger transformer, and a power supply
unit 6. The capacitor is connected across the lamp and
is charged to a defined voltage by the power supply
unit. In response to a signal from the signal
generator, the power supply unit provides a trigger
pulse to the trigger transformer, which ie connected to
the lamp, and initiates the flash. A suitable
illumination system consists of a PS-450AC power supply,
used with an integrated trigger transformer and lamp
base FyD 506 Lite Pac, a CP-1229M1 ~F, 600 V capacitor
and an FX-504U xenon flash lamp, all supplied by EG&G
Electro-Optics of Salem, MA, USA. The lamp has a
maximum mean operating power of 20 W and is operated at
a frequency of 50 to 200 Hz, an Increase in frequency
causing an increase in effective intensity. The signal
generator can be any suitable TTL square wave generator,
one suitable generator being a Levell function generator
type TG 301, supplied by Digitron Instruments Ltd, of
Hertford, Great Britain.
The beam of light from the light source 5 is focused by
mm focal length ultraviolet grade quartz lenses 7 and
8, and a cut-on interference filter 9 is provided. The
filter 9 may be for instance an Omega 200 nm ultraviolet
CA 02157469 2004-10-08
16
mM
transmission filter, but is preferably a G25 206F filter
supplied by the Corion Corp. of Holliston, MA., USA,
which has a peak transmission at 206 nm; having regard
to the spectral output of the lamp 5, the intensity of
the ultraviolet radiation transmitted by the Coriori
filter 9 above 225 nm (up to 380 nm) is about 4% of the
ultraviolet radiation transmitted between 190 nm and 225
nm, i.e. the radiation transmitted is preponderant?y or
substantially of a wavelength of less than 225 nm.
A steerable ultraviolet grade mirror 10 is provided to
direct the light onto the surface of the diamond 1. An
image of the photo-luminescence patterns produced on the
surface of the diamond being studied is produced and
studied by eye using magnifying means in the form of a
microscope I1 in the preferred embodimeat, but a camera
or CCD image recorder 12 (shown schematically) may be
provided for later study or processiag of the results.
The glass of the microscope optics cuts out any hard
ultraviolet radiation, say that having a wave length of
less than 300 to 330 nm, and very little ultraviolet
radiation Would in any case pass up through the
microscope 11. Any suitable microscrope ii may be used,
preferably having a microscope ocular with a zoom
attachment, and the microscope 11 may be a standard
gemologicai microscope with a facility or camera 12 to
record the image on fast film. Colour film can be used.
CA 02157469 2004-10-08
17
Figure 2 shows just by way of example the apparatus
contained in a casing 13 to provide ultraviolet
shielding, the mounting means 2 being in a sliding
drawer 14 with suitable rubber seals for shielding.
Controls 15, 16, 17 ara shown schematically for power
supply on/off, power supply frequency and mirror
steering. The apparatus may be constructed so that it
is portable and occupies a fairly small space, to
provide a handy and portable device for distinguishing
natural diamonds and synthetic diamonds. For example,
the signal generator and power supply 6 for the light
source 5 may occupy a space of only 215 mm x 310 mm x
110 mm and the length, breadth and height of the casing
12 can be about 500 mm x 250 mm x 250 mm. However, it
is believed that these casing dimensions can be reduced
to say 300 mm x 200 mm x 200 mm.
A skilled operator observing a polished diamond 1 using
the apparatus of Figure 1 would look for geometric
patterns indicating mixed habit growth, which indicate
synthetic diamond.
As an example, a 1 ct. (0.2 g) circular brilliant-cut
diamond 1 can be examined using the EG&G xenon lamp 5
using a 200 Hz frequency. For examination by eye, the
signal generator is left on long enough to complete the
examination. For making a photographic togograph, the
signal generator can be left on several minutes,
WO 94/20837 PCT/GB94/00415
18
depending on the photographic film used and the
aperture. For recording using a CCD image recorder 12,
a few seconds may be sufficient.
The transmission filter 9 and the ultraviolet grade
mirror 10 can be replaced by a single dichroic fused
silica steerable reflection filter 21. The reflection
filter 21 can be inserted in the same position as the
mirror 10. The transmission characteristic of this
filter 21 shows it to have a peak front surface
reflection at about 200 nm. There is on average about
2~ reflection at wavelengths greater than 225 nm and the
intensity of radiation reflected in the range 225-380 nm
is about 20~ of the ultraviolet radiation reflected
between 190 nm and 225 nm. Accordingly, a wavelength
band in which substantially or preponderantly all
wavelengths are less than 225 nm can be projected on to
the diamond 1. The emitted photoluminescence is
transmitted by the filter 21 to the microscope 11. An
anti-reflection coating on the back surface of the
filter 21 for visible wavelengths of 400 to 700 nm
enhances the transmission.
CA 02157469 2004-10-08
19
Figures 3, 4, 5, 6, and 7
Figures 3 and 4 are diagrammatic representations of major
features of the table of an emerald-cut synthetic diamond
and the table of a circular brilliant-cut synthetic
diamond, respectively, only part of the stone being shown
in Figure 4. The photoluminescence of the synthetic stone
diagrammatically represented in Figure 3 was essentially
pale blue and showed coloured banding; that of the
synthetic stone represented in Figure 4 was prediminantly
yellow in colour, though with regions of blue, and showed
yellow and green banding. Regions 31, 32, 33, 34 are
identified in Figure 3 and Figure 4, showing different
growth sectors, the striated portions 35 in Figure 4 being
due to growth banding. Both of these views show
characteristic mixed-habit growth of synthetic diamonds.
Figures 5 and 6 are diagrammatic representations of major
features of the tables of two circular brilliant-cut
natural diamonds taken under identical conditions as those
represented in Figure 3 and 4. The natural stones of
Figures 5 and 6 were blue in photoluminescence, which is
typical for natural stones. Most natural diamonds are
CA 02157469 2004-10-08
type Ia and emit blue luminescence due in part to a band
commencing at 4I5 nm (N3 band) and extending to longer
wavelengths. This band is almost coincident with the
blue bands emitted by natural and synthetic type IIb
semi-conducting diamonds. Natural type IIa diamonds
(containing very low point defect concentrations) tend
to luminesce very weakly under ultraviolet excitation at
225nm or less.
The patterns on the images of the natural stones are
almost entirely formed by the cut of the crystal and
there is very little distinct banding. A distorted
square 36 is identified in Figure 5 but this is the
only evidence of growth banding. Figure 6 shows some
growth banding 37 near the centre of the table only. A
person skilled in the art studying these diamonds would
recognise the octahedral growth only. The bright area
38 in the corner of Figure 6 is due to light scattering
from an internal crack.
Figure 7 shows the pattern on the image of a circular
brilliant-cut natural diamond to tho table of which has
been applied a vapour deposition layer prior to
polishing; only part of the stone is shown. The crown
facets in the centre of the field of view lie
approximately in tho focal plane of the microscope.
Figure 7 shows growth bands 39 in the natural diamond
substrate and a distinctly ditferent luminescence 40
CA 02157469 2004-10-08
21
from the CVD layer on the table and adjacent parts of
the crown facets, the luminescence being different in
intensity and/or colour.
The illuminating ultraviolet radiation is subsequently
cut off so that fluorescence (which forms the bulk of
the ultraviolet photoluminescence) ceases; any
phosphorescence will be revealed and can be observed by
eye or using the camera or image recorder 12. In
general form, the topographs are similar to those represented
diagranTnatically in Figures 3 and 4, and the shape of the sectors is the
same; however with some diamonds, the presence of
individual sectors is accentuated because phosphorescing
sectors stand out more clearly against a
non-phosphorescing background. The irradiation can be
terminated by using a flash lamp for the source 5. or by
interposing a shutter 41 in the light path. Whatever
means are used for terminating the irradiation, it
should be such that the irradiation can be swiftly
terminated, e.g. by switching off the power supply 6 or
dropping the shutter 41.
The present invention has been described above purely by
way of example, and modifications may be made within the
invention.
