SUMMER 2013

VOLUME XLIX

THE QUARTERLY JOURNAL OF THE GEMOLOGICAL INSTITUTE OF AMERICA

Contrast Patterns for Optimizing Brilliance in Colored Gemstone Faceting

Fluorescence Produced by Common Optical Defects in Diamond
Quick Reference Chart for Optical Defects in Diamond
Nephrite Jade from the Italian Alps
 Summer 2013
VOLUME 49, No. 2
EDITORIAL
63 Integrating G&G with GIA’s Gemology Education
Duncan Pay

FEATURE ARTICLES
pg. 65
64 Optimizing Face-Up Appearance in Colored Gemstone Faceting
Al Gilbertson
Understanding the human visual system of interpreting contrast patterns can
reveal facet-planning techniques to increase perceived brightness and optimize a
gem’s appearance.

82 Fluorescence Produced by Optical Defects in Diamond:

Measurement, Characterization, and Challenges
Yun Luo and Christopher M. Breeding
Studies of 3D fluorescence spectra from diamonds with different optical centers
reveal that variations in UV lamp output can greatly affect observed
fluorescence color and intensity.
pg. 105
98 Nephrite Jade from Val Malenco, Italy: Review and Update
Ilaria Adamo and Rosangela Bocchio
Analysis of 21 rough and gem-quality samples offers an update on the composi-
tion and properties of this pale green gem from the Italian Alps.

NOTES AND NEW TECHNIQUES

107 Optical Defects in Diamond: A Quick Reference Chart
James E. Shigley and Christopher M. Breeding
A simple chart on optical defects in diamond, including those that produce
coloration and fluorescence colors.
pg. 117

REGULAR FEATURES
112 Thank You, Donors
113 Lab Notes
Freshwater cultured pearls with damaged nacre • Calcium fluoride coating found on 13 “Fancy” pink
diamonds • Diamond with fish-like inclusion • Large irradiated green-yellow diamond • Strong color
zoning representing complex diamond growth environment • Very large single-crystal rough diamond
• Unusual curved color zoning in emerald • Rare faceted neptunite

120 Gem News International

Ancient tourmaline and beryl from Afghanistan • Orangy pink coated “soft coral” • Strands of tiny akoya
keshi pearls • Green kyanite • Banded mimetite • Turquoise-rock crystal composite • Imitation Larimar
Editorial Staff
Editor-in-Chief Editorial Assistants
Duncan Pay Brooke Goedert
[Link]/gems-gemology
dpay@[Link] Nathan Renfro
Managing Editor Editors, Lab Notes Subscriptions
Justin Hunter Thomas M. Moses Copies of the current issue may be purchased for
[Link]@[Link] Shane F. McClure $29.95 plus shipping. Subscriptions are $79.99 for
one year (4 issues) in the U.S. and $99.99 else-
Editor Contributing Editors where. Canadian subscribers should add GST.
Stuart D. Overlin Discounts are available for group subscriptions,
James E. Shigley
GIA alumni, and current GIA students. For insti-
soverlin@[Link] Andy Lucas tutional rates, contact Customer Service.
Technical Editors Editor-in-Chief Emeritus To purchase print subscriptions, visit [Link]
Tao Z. Hsu Alice S. Keller or contact Customer Service.
[Link]@[Link]
Customer Service Database Coverage
Jennifer Stone-Sundberg Gems & Gemology is abstracted in Thomson
Martha Erickson
(760) 603-4502 Reuters products (Current Contents: Physical,
Associate Editor
Chemical & Earth Sciences and Science Citation
Claire Harlin gandg@[Link]
Index—Expanded, including the Web of Knowl-
charlin@[Link] edge) and other databases. For a complete list of
sources abstracting G&G, go to [Link]/gems-
gemology.
Production Staff
Manuscript Submissions
Director, Content Development Photographer and Photo Editor Gems & Gemology welcomes the submission of
Jan Iverson Robert Weldon articles on all aspects of the field. Please see the
Guidelines for Authors at [Link]/gandg or con-
Creative Director Multimedia Specialists tact the Managing Editor. Letters on articles pub-
Faizah Bhatti Joseph Kaus lished in Gems & Gemology are also welcome.
Juan Zanahuria
Image Specialist Copyright and Reprint Permission
Kevin Schumacher Production Supervisor Abstracting is permitted with credit to the source.
Richard Canedo Libraries are permitted to photocopy beyond the
Senior Illustrator limits of U.S. copyright law for private use of pa-
Peter Johnston Video Producer trons. Instructors are permitted to photocopy iso-
Pedro Padua lated articles for noncommercial classroom use
without fee. Copying of the photographs by any
means other than traditional photocopying tech-
Editorial Review Board
niques (Xerox, etc.) is prohibited without the ex-
Ahmadjan Abduriyim A.J.A. (Bram) Janse George R. Rossman press permission of the photographer (where
Tokyo, Japan Perth, Australia Pasadena, California listed) or author of the article in which the photo
appears (where no photographer is listed). For
Shigeru Akamatsu E. Alan Jobbins Kenneth Scarratt other copying, reprint, or republication permis-
Tokyo, Japan Caterham, UK Bangkok, Thailand sion, please contact the Managing Editor.
Edward W. Boehm Mary L. Johnson James E. Shigley Gems & Gemology is published quarterly by the
Chattanooga, Tennessee San Diego, California Carlsbad, California Gemological Institute of America, a nonprofit ed-
ucational organization for the gem and jewelry in-
James E. Butler Anthony R. Kampf Christopher P. Smith dustry.
Washington, DC Los Angeles, California New York, New York
Postmaster: Return undeliverable copies of Gems
Alan T. Collins Robert E. Kane Wuyi Wang & Gemology to GIA, The Robert Mouawad
London, UK Helena, Montana New York, New York Campus, 5345 Armada Drive, Carlsbad, CA
92008.
John L. Emmett Lore Kiefert Christopher M. Welbourn Our Canadian goods and service registration num-
Brush Prairie, Washington Lucerne, Switzerland Reading, UK ber is 126142892RT.
Emmanuel Fritsch Thomas M. Moses Any opinions expressed in signed articles are un-
Nantes, France New York, New York derstood to be opinions of the authors and not of
the publisher.
Jaroslav Hyršl Mark Newton
Prague, Czech Republic Coventry, UK

About the Cover

The lead article in this issue examines the visual contrast patterns that optimize brightness in faceted gemstones.
Brightness is superbly demonstrated by the gemstones shown on the cover, part of a collection of more than 100 tril-
liant-cut stones on display at GIA Carlsbad. Clockwise from the top: 61.98 ct kunzite, 48.31 ct sphalerite, 9.67 ct
grossular garnet, 7.71 ct zircon, 12.27 ct sphalerite, and 29.59 ct spessartine. Center: 25.15 ct fluorite. Collection cour-
tesy of Roz and Gene Meieran; photo by Robert Weldon.
Printing is by Allen Press, Lawrence, Kansas.
GIA World Headquarters The Robert Mouawad Campus 5345 Armada Drive Carlsbad, CA 92008 USA
© 2013 Gemological Institute of America All rights reserved. ISSN 0016-626X
 INTEGRATING G&G with
GIA’s Gemology Education
One of the things I’ve always wanted to do is create synergy between
GIA’s gemology courses and Gems & Gemology. Many of the student
assignments incorporate research from the journal, and adding G&G
articles as references would provide support and validation for the infor-
mation in our coursework.
As we added video media to our eLearning courses, we featured
interviews with many of GIA’s research scientists, some of whom were
also G&G authors. Providing their articles to our students was a natural
next step.
In the past, I had discussed this idea with Alice Keller when she was editor-in-chief of G&G.
Although everyone was willing, we had no practical way to achieve it at the time. Now that every
issue is available free of charge on the redesigned GIA website at [Link], we can realize this
ambition.
Beginning with our Diamonds & Diamond Grading eLearning course, we’ll provide links at the
end of every assignment to specially chosen G&G feature articles.
Besides illuminating some of the source material for our coursework, links to supporting articles
will provide a truly meaningful pathway into the resources on our new website. Students with
inquiring minds will find many
avenues of interest. Besides illuminating some of the source material
For example, I believe a good number for our coursework, links to supporting G&G arti-
of students would choose to explore cles will provide a truly meaningful pathway into
the important G&G articles on the resources on our new website.
colored diamond, diamond treatment,
or diamond synthesis. Others might also download articles focused on mining, production, and
the diamond supply chain, or read about the many famous gems examined or graded by GIA’s
laboratory.
This tie-in exposes the tremendous amount of important work that G&G authors—both
internal and external to GIA—have done to enrich our collective knowledge on the many aspects
of diamond, and makes it available to a wider, and younger, audience.
We’ll also provide a short student study guide that explains the structure of a typical feature
article and explains how to navigate one.
In time, we’ll add similar links to relevant G&G articles in our other gemology courses.
Let me take this opportunity to thank all our current G&G print subscribers and online users. I
hope to welcome many more of you in the near future.

Duncan Pay | Editor-in-Chief | dpay@[Link]

EDITORIAL GEMS & GEMOLOGY SUMMER 2013 63

FEATURE ARTICLES

OPTIMIZING FACE-UP APPEARANCE IN

COLORED GEMSTONE FACETING
Al Gilbertson

The human visual system interprets visual cues to perceive different intensities of brightness. Patterns of
light and dark contrast create impressions of brightness in faceted gemstones. These patterns generally
remain coherent in the viewer’s eye when the gemstone is tilted or rotated. Using computer-generated
color-coded contrast maps in facet planning can improve apparent brightness, optimizing a gemstone’s
appearance.

T
here are many challenges in choosing a facet will lead to a faceted gemstone with maximum appeal
arrangement that optimizes a colored stone’s given the constraints of the material at hand. This
face-up appearance. What contributes to an at- paper focuses on gems with colors that are not easily
tractive appearance? In the field of cognitive science, weakened by pronounced brightness or darkness, sim-
experts have identified rules that govern our percep- ilar to the medium range of tone and strong saturation
tion of visual cues such as line, color, form, brightness, found in peridot (figure 1).
contrast, and motion. The subconscious processing of
these cues is the work of each person’s “visual intelli- UNDERSTANDING BRIGHTNESS
gence system,” wherein every aspect of our visual ex- AND CONTRAST
perience is framed. In gemstones, brightness depends Understanding what creates the impression of bright-
on how we process visual cues, so some patterns ap- ness can help us map and plan areas in a gemstone
pear brighter than others even when the measurable to enhance observed brightness.
light return is identical. Studying these rules can help Cognitive scientists understand much of the com-
formulate design strategies to make gemstones plexity of the stimuli processed by our visual system.
brighter and more interesting. Tools exist for exploring For example, even though variations in lighting re-
these aspects of appearance. Using these, cutters can sult in varying wavelengths of light reflected from a
choose elements that will optimize the design. Effec- colored surface, our visual intelligence system actu-
tive design requires an exploration of the best angle ally works as a subconscious processor, assigning a
combinations and placement of facets for a given ma- constant hue, saturation, and tone (which scientists
terial. Nevertheless, choices about appearance are refer to as “lightness”). This collaboration between
personal. And while preferences vary, this study as- the visual system and the brain is constantly extract-
sumes the goal of improving scintillation while main- ing information from a flood of sensations to con-
taining as much brightness as possible. Most colored struct a visual world. In the words of leading
stone cutters understand their own preferences regard- cognitive scientist Donald Hoffman (1998), “You are
ing visual contrast, depth of color, spread (shallow or a creative genius. Your creative genius is so accom-
deep), brightness, and scintillation. By understanding plished that it appears, to you and others, as effort-
certain aspects of optimization, they can modify de- less. Yet, it far outstrips the most valiant efforts of
signs to suit their own preferences. Ultimately, this today’s fastest supercomputers. To invoke it, you
need only open your eyes.”
See end of article for About the Author and Acknowledgments.
These scientists work primarily with two-di-
GEMS & GEMOLOGY, Vol. 49, No. 2, pp. 64–81, mensional printed (opaque) images or flat images on
[Link] a computer monitor, and they consider brightness a
© 2013 Gemological Institute of America subjective attribute or property of an object being

64 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
 Figure 1. The concepts
dealing with bright-
ness, scintillation, and
overall appearance are
best evaluated in a gem
whose color is not eas-
ily weakened by mak-
ing it bright or overly
dark, such as peridot.
Photo by Robert Wel-
don, © GIA.

observed. Gemstones are three-dimensional and The following is a general overview of the visual
transparent, and the light reflected from within cues relevant to faceted, transparent gems.
them has a greater impact on their face-up appear-
ance than the light reflected from the surface. Our Invisible Surfaces. Termed by some vision scientists
visual intelligence processes visual cues so we see as “the invisible surface that glows,” the illustration
a gem that seems to generate light from within it- created by G. Kanizsa (1955) shows two different tri-
self, in a pattern that changes as the gem moves. angles that are more luminous (i.e., brighter) than the
This pattern is generated by the interaction of light background, and whose borders are easily observed
with the faceting design. even though there are no lines defining them (figure
Jose Sasián, a professor of optical sciences at the 2). Using a photometer, one cannot detect any edges
University of Arizona, explains that while looking at of the triangle. The borders and luminosity are purely
a gem, the observer appears to see more facets than ac- the construction of the observer’s visual intelligence
tually exist. Sasián (2007) calls these “virtual” facets. system. The brain uses a variety of cues to construct
These perceived facets are a result of different levels these images, which we will refer to as “invisible sur-
of contrast, caused by light striking all the facets and face maps.” In figure 3, the invisible surface map on
then splitting, creating the appearance of many more the left has a more luminous square area than the
facets. map on the right. How the brain perceives the lines
How we interpret visual cues is critical to under- that resemble small “check marks” causes this duller
standing what creates a gemstone’s beauty, and by effect.
studying these cues we can identify the elements to In the three “plus-symbol” outlines in figure 4, the
include in gem design. first contains a square luminous area in the center. In
the second image, the sides of the square bow inward
Figure 2. Observers see a bright, luminous triangle in
the center of both figures, even though there are no
defining lines. Cognitive scientists attribute this “in- Figure 3. The visual contrast cues in this pair of invis-
visible surface map” to how our brain interprets visual ible surface maps are different, causing our brain to
cues provided by the contrast of light and dark. From interpret the left square as brighter than the right
Kanizsa (1955). square. From Albert (1955).

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 65
 an edge that is not really there—the invisible surface.
Different types of pattern elements (such as blunt
lines, dots, or wedges) can produce similar appear-
ances, depending on how they are arranged. Elements
of varying brightness can also be embedded or
stacked within each other (figures 5H–5J), creating
Figure 4. These “plus-symbol” outlines have slightly
multiple bright areas. Shifting the center of these em-
different visual cues. The first contains a square
bright middle area. In the second, the square appears
bedded areas (figure 5I) does not disturb the impres-
to have concave sides due to the four dots. With the sions of brightness; the center has moved along with
third, the bowing has been removed by shifting the the areas of contrast. Although these areas show
four sets of lines. From Petry (1987). strong contrast, they are still coherent and bright,
much like when a faceted gemstone is rocked. They
are also present in non-round shapes such as the mar-
due to the addition of visual cues (the four black dots). quise or oval, as shown in figure 5J.
In the third image, the bowing disappears because the The ability to see these illusory bright areas in an
four sets of lines have been shortened to square up invisible surface map is enhanced by movement.
with the black dots. (Note: The brain can interpret vi- When the image in figure 6 was “flickered” rapidly
sual cues in more than one way. For example, some on a computer monitor, the illusory center became
see a circular rather than a square luminous area in brighter. When the flickering image was also rotated,
the leftmost image.) the apparent brightness increased significantly. The
Varying combinations of bright and dark can flashing of light and dark patterns caused by rotating
cause the observer to perceive different intensities of and flickering the image equates to the scintillation
brightness. This does not address personal prefer- seen when we move a gemstone back and forth.
ences for different combinations. For example, the Stronger contrast (black or gray areas against white
invisible surface maps in figures 5A–5G demonstrate backgrounds) also equates to stronger scintillation.
various visual cues that create a bright circular area Vision scientists have found that in certain cases an
in the center. In each of these examples, the different image stays organized or coherent in a viewer’s mind
cues and their arrangements result in varying degrees when the direction of observation is changed, as
of perceived brightness, leading to the impression of through movement.

Figure 5. These invisible surface maps have slightly different contrast cues, resulting in the impression of bright or
luminous circles (A–G). Contrast cues can also be stacked or embedded within each other (H–J) to give the impres-
sion of multiple areas of brightness.

66 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
Brightness Contrast. Appraiser Michael Cowing,
FGA, has published several articles on evaluating di-
amond cut quality, coining the term “brilliance con-
trast” to describe how contrast causes the brightness
observed in diamonds (2009). Each of the circles in
figure 7 is 50% dark and 50% light by surface area.
A light meter confirms that each one is equally
bright when printed on paper (even the gray ones).
The checkered pattern of the bottom left circle
makes it perhaps the most visually interesting. If a
gem were able to return 100% of the light (all white),
and no dark areas were visible, it would measure

In Brief
• A gem’s brightness and attractiveness depends not only
on how much light is returned but also on the contrast
pattern or contrast distribution created by virtual facets.
• An idealized contrast plot shows the types of contrast
distribution that contribute to effective gem designs to Figure 7. Each of these circles is 50% dark and 50%
be used by faceters. light by surface area. Yet the circles on the right are
• Following rules for optimization, such as creating a dull due to weak contrast. A gemstone’s visual interest
design with the right contrast distribution, improves a depends on good contrast. If it returned all light and
gem’s appearance and increases its fire. had no contrast, the gemstone would be uninteresting.

brighter than a stone with dark areas of contrast— Figure 8 shows a common optical illusion known
but its appearance would be far less appealing. For in- as “White’s Illusion.” Although the rectangles in B
stance, the right column in figure 7 seems duller than seem darker than those in A, they are actually of
the left due to weaker contrast. These images show equal luminance. In fact, they have the same gray
that while good light return is an important aspect color. The gray areas embedded in dark stripes (A) ap-
of “brilliance,” contrast is a critical factor in face-up pear brighter than the gray areas embedded in white
brightness. Obviously, there comes a point when too stripes (B). This illustrates the concept of “lateral in-
much darkness or a poor distribution of darkness is hibition,” in which the perceived brightness of an
less pleasing. area depends upon its surroundings (White, 1979).

Figure 6. Observers see a markedly increased bright- Figure 8. The rectangles on the right appear darker
ness in the center circle when this invisible surface than those on the left, but they are actually the same
map is flickered on a computer monitor. When the shade of gray. This demonstrates that the perceived
image is rotated and flickered simultaneously, the brightness of an area depends on the surroundings
brightness further increases. From Petry (1987). and the contrast. From White (1979).

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 67
 In figure 9, the horizontal stripe and all six boxes
have the same brightness and color, but the squares’
A
apparent brightness depends on the contrast of the sur-
rounding area. In this illusion demonstrating “simul-
taneous lightness contrast” (an example of lateral
inhibition), apparent lightness is different from quan-
titative lightness, the real measurement of the total
B
lightness reflected from an object. The phenomenon
also occurs when a colored area is placed on a dark
background, making it appear clearer, brighter, or
more luminous. In figure 10, for example, the navette
shapes at the bottom are identical, but they appear to
C
be different colors when placed on alternating light
and dark backgrounds (C). The difference is accentu-
ated with the addition of an interwoven background
(A and B).

Summary of Vision Science. Combining the ideas of Figure 10. The navette shapes throughout this image
invisible surfaces and brightness contrast, we can de- have the same tone but appear darker or lighter de-
rive that darker contrast, placed in the right locations, pending on the adjacent areas. From Sarcone and
can give faceted gems a brighter appearance. The areas Waeber (undated).
of contrast need to be evenly distributed and not
grouped together; this is particularly true for areas
under the table. Our eye tends to gravitate toward the shape results from certain visual cues relating to dif-
center to gather an overall impression of brightness. ferences of light and dark. The effect is further en-
The diagrams in figure 11 illustrate how the place- hanced by movement, and invisible surfaces can be
ment of dark areas affects brightness. The patterns are stacked or embedded within each other. Strong con-
identical but contain varying levels of contrast. The trast makes a gemstone appear brighter and more ap-
image of highest contrast (bottom right, for most ob- pealing, and the position of the contrast is also
servers) also has the brightest-looking center. This is
a common goal in facet design, as dark centers, or
“nail heads,” are undesirable in the jewelry trade. Figure 11. Varying the levels of contrast affects the ap-
From cognitive science we know that the appar- pearance of brightness. Each circle has the same pat-
ent brightness of an invisible surface of definable tern, but the areas of contrast vary in strength. The
circle with the highest contrast (bottom right) has the
brightest-looking center.

Figure 9. The stripe and the small squares are all the
same shade of gray, again illustrating that perceived
brightness relies upon its surroundings and the con-
trast, which changes as the environment changes.
From Purves (2011).

68 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
important. Scintillation is more dynamic when there
is strong contrast between adjacent virtual facets.
If we understand what causes the cues for bright-
ness in any shape, we can use this information to de-
sign effective facet patterns by using strong contrast
and ending the lines of contrast to produce a bright
center. A gem’s brightness and attractiveness depends
not only on how much light is returned—angles are
still important—but also on the contrast pattern cre-
ated by virtual facets. Once an effective pattern is es-
tablished, the major elements of that pattern must
remain in effect when the gem is tilted to maintain
the appearance of brightness. This means that the
pavilion angles cannot be cut near an angle that allows
the gem to “window,” letting light leak through. If the
slightest tilt interrupts the pattern due to a window, Figure 12. A gemstone’s liveliness is the result of scin-
the pattern is not effective because the contrast pat- tillation: the strong contrast between neighboring
tern has been eliminated. The impact of these effects facets when the gem, the light source, or the observer
may vary according to the level of brightness, as dic- moves. An observer’s head (shown from above) ob-
tated by refractive index and color saturation. structs overhead light, causing a sharp change in the
light streaming into the gem. Note that at a viewing
TOOLS FOR ASSESSING CONTRAST distance of one foot (approximately 30 cm), the angle
If contrast is an important part of faceting various gem between the two reflected rays that return to both
eyes is about 12°. From Harding (1975, figure 8).
materials, there needs to be a reliable method to de-
termine elements of contrast. Most critical is the con-
trast caused by the reflection of the observer’s head
and torso. Harding (1975) first identified this critical patterns in diamond appearance, as shown in an un-
element: the effect of an observer’s head blocking rays dated FireScope brochure, circa 1987.
of overhead illumination. He calculated the rays re- It should be noted that an observer both blocks
turned from the gem to the viewer’s eyes, as well as light from the rear and reflects light from the front
the illumination blocked by the viewer’s head. or side. In most environments, however, far less light
Regarding head obstruction, Harding writes, “For
ultimate liveliness it should be possible to see reflec-
tions in the table from both sides of the pavilion with Figure 13. The FireScope by JDM uses the black reflec-
tion of the viewing lens as a dark contrast against a
both eyes at once. At a viewing distance of one foot,
red reflecting field. Studying the patterns of different
as shown in [figure 12], the angle between reflections
diamonds, JDM concluded that the most attractive
to both eyes (two different rays) is about 12°. To see diamonds have a certain contrast pattern. From an
reflections from both sides with both eyes, therefore, undated FireScope brochure, circa 1987.
the minimum external table reflection angle must be
at least 6°.” While not directly discussing the need
for contrast, Harding alluded to its importance. Of
course, liveliness stems from scintillation, the result
of strong contrast between neighboring facets when
the gem, the light source, or the observer moves.
This concept was also part of Kazumi Okuda’s
groundbreaking work in about 1980. His experiments
with reflectors—“hearts and arrows” viewers have
evolved from these devices—led to the FireScope™
introduced by JDM in Japan. JDM used the black re-
flection of the lens as a dark contrast against a red re-
flecting field (figure 13). Studying the black reflection,
they determined the importance of certain reflection

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 69
 A B C
Figure 14. These photos show
the same gemstone under dif-
ferent viewing conditions: (A)
under totally diffused white
light; (B) under the same en-
vironment, with the observer
wearing a red mask and gar-
ment; and (C) with the ob-
server standing closer. From
these, it is apparent that the
observer sees much of their
own reflection, as dark con-
trast in the gemstone.

is reflected into the gem from the observer’s head and the author’s own environment. GemCad is a popular
shoulders than from around and above the observer. computer-aided design (CAD) program used to create
The observer’s reflection therefore serves as the accurate 3D models of a faceted gemstone. DiamCalc
major source of contrast in the lighting environment. accepts GemCad and other designs and allows
To better understand this, consider what a faceted changes in table size and girdle thickness, as well as
gem might look like in a totally diffuse white-lit en- facet angles and azimuths, for a variety of standard
vironment (figure 14). If the gem reflected nothing shapes. DiamCalc can also create motion pictures of
but white light, it would have no contrast (A). If you a gem being rocked or otherwise moved from side to
were to cover your face and shoulders with a fluores- side in these lighting environments. The software is
cent red mask and view the gem in the same envi- widely used in the diamond cutting industry to plan
ronment, you would see red contrast (B). If you rough and design new cuts. It can accept files from
moved closer to the gem, there would be even more GemCad (.asc format), Sarin (.srn), Helium (.mmd),
red reflected throughout it (C). and Autodesk (.dxf or .stl), as well as certain .txt files.
In 1997, this author started experimenting with It can generate the following file formats: Autodesk
color coding of light (and contrast) entering a stone DXF (.dxf), GemCad (.asc), and binary and ASCII STL
from various angles (“Reflector technologies,” un- (both in .stl format). For colored stone cutters, it al-
dated). By constructing a hemisphere or dome with lows variations in refractive index for modeling var-
concentric rings of color (figure 15) and viewing the ious gem materials.
stone through an aperture at the top of the dome, one DiamCalc does not model double refraction, so the
can see that the gem gathers its light from the col- effects of a closed or dark c-axis cannot be predicted.
ored rings. DiamCalc, a 3D modeling program intro- Many stones are dichroic, which means one sees dif-
duced in Russia in March 1999, uses ray tracing to ferent colors in different directions through the gem.
alter the proportions and angles of a diamond image Some gems with this property will show nearly iden-
viewed on a computer monitor. By 2000, color-coded tical colors in all directions, while some will show only
lighting environments were added, allowing the differences in color density (darker or lighter tone). For
modification of angle of arc as well as color. Diam- example, tourmaline often has extreme dichroism, and
Calc can model virtual polished gemstones in a vari- this characteristic is referred to by many gem cutters
ety of realistic lighting environments, including as a “closed c-axis” or “dark c-axis.” In other words,
GemCad’s ISO and COS, Fire Scope™, Ideal-Scope, light does not pass through one axis (direction) of the
ASET (Angular Spectrum Evaluation Tool), and AG, stone, or only passes in a limited amount.

Figure 15. Gemstones

gather light from the sur-
rounding viewing environ-
ment. When placed under
a hemisphere lined with
concentric rings of various
colors (left and center), a
gemstone (right) returns
the colors of the dome.

70 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
Comparison of Two Important Models. While Diam- which can vary depending on RI, it is advisable to de-
Calc supports a variety of lighting environments, termine personal preference before analyzing pat-
this paper deals with two of particular value in de- terns with AG and ASET lighting.
sign optimization: AG and ASET (figure 16). These DiamCalc also allows the background to be color-
are variations of the color coding of light entering a coded. A black background is the default setting for
diamond from specific surrounding arcs, patented in both the AG and ASET environments, but white is
2000 (Gilbertson, 2003) and 2004 (Caudill et al. also available. Areas where light passes through the
2008), respectively. While either can be used as a pri- gem from the background (“windows”), are indicated
mary environment for planning or as a secondary as white in the contrast plot when using the white
double check, the AG environment offers certain ad- background.
vantages as the primary. Comparing the three-color
ASET and the five-color AG patterns for the same PRACTICAL APPLICATION
stone, observe how their angles of light orientation Past work can be useful for further understanding the
differ. Key information about both the observer AG lighting environment. For example, both GIA
(black, 83°–90°) and the lower angles of light entry and the American Gem Society have determined
likely producing a strong contrast (dark violet, that the most appealing round brilliant-cut diamonds
40°–55°) are not represented well in the ASET image. have a balance of contrast and brightness (GIA, 2006;
Three color-coded zones do not provide enough in- Sasián, 2007). The contrast maps of a typical well-cut
formation, as certain patterns with strong contrast diamond are shown in figures 16 and 17 to illustrate
(black) can produce some of the visual cues leading the relative balance of the various color-coded light
to different preferences. entry angle ranges. Much like the invisible surface
If a design is planned with AG and then checked map examples in figure 5, these color-coded contrast
with ASET, however, nuances and slight weaknesses maps can explain the pattern seen in the cut gem
in design can be captured and modified later. These with AG lighting (figure 17).
examples represent diamond’s RI and the sets of an- With the AG environment, note that there is very
gles considered optimum for diamond appearance. little black (83º–90º). The black areas representing the
Since individuals may prefer different patterns, retro-reflection of the observer should be minimized

ASET (+ white) AG (+ white)

Green = 0° (horizon) to 45° Light Blue = 0° (horizon) to 40°
Red = 45° to 75° Dark Violet = 40° to 55°
Blue = 75° to 90° Green = 55° to 70°
Red = 70° to 83°
Black = 83° to 90° Figure 16. Two of the
light color-coding mod-
els supported by Diam-
calc—AG and
ASET—can be used for
mapping contrast in
gemstone designs. Dif-
ferent aspects of con-
trast in the same
gemstone can be seen
using the three-color
ASET environment and
the five-color AG envi-
ronment.

None

Blue Red Green Light Blue Violet Green Red Black

75°-90° 45°-75° 0° (horizon) to 45° 0° (horizon) to 40° 40°- 55° 55°- 70° 70°- 83° 83°- 90°

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 71
or removed in planning colored stones. The lowest an- EXAMPLES
gles of light entry represented by violet (40º–55º) will Figure 18 shows examples of various gems purchased
appear darker, along with any areas of windowing, from the trade for GIA’s Dr. Eduard J. Gübelin Col-
which produces a very nice balance of varying con- lection. The individual gems were scanned using a
trasts within the gem. These examples are comparable non-contact measuring device to derive the facet an-
to the dark areas in the invisible surface maps. gles and arrangement. A photo of each gem, showing
The contrast maps display a substantial amount the representative pattern viewed face-up, is accom-
of green, which is represented by light coming from panied by an AG contrast map derived from Diam-
around the observer’s head (55º–70º). The red (70º– Calc. The observer’s face and torso are about 18–20
83º) is arranged in a narrow, wagon-wheel pattern inches from the stone, and the photo captures a good
from the girdle edge to the near-center. Red areas can representation of what the observer would typically
appear dark or bright, depending on the observer’s see in the gem. The following is a brief discussion on
proximity to the gem or how much it is tilted. Both the appearance of each gem.
of these colors indicate the primary areas of bright- Figure 18A: This native-cut golden sapphire (RI
ness in a gem, comparable to the white area in the near 1.77) is particularly bright because the center of
invisible surface maps. An effective design requires the gem reflects light gathered from around the ob-
these areas of brightness to be prominent and well server and returns that light to the observer. Yet the
distributed, with the dark areas creating a balanced, outer areas window slightly, and in lighter colored
contrasting pattern in the gem. gems this is distracting. This view can be explained
Contrast maps are similar to a much more com- by looking at the color-coded contrast map (AG with
plex stacked invisible surface set (such as figure 5H– white contrast map). Red and green indicate higher-
J). As the invisible surface maps are moved and contrast areas that provide some brightness in those
rotated, the apparent brightness increases dramatically parts of the gem. Note that these extend out to the
and remains organized in a viewer’s mind even when girdle, which is very important. The outer areas con-
the direction of observation is changed. This means tain blue, resulting from lower-angle lighting. The
that movement (rotating and flickering) of the invisi- outer white areas are the windows where the ob-
ble surface maps is comparable to the scintillation server sees through the gem.
seen when a gemstone moves back and forth. With Figure 18B: When gently rocking this apatite (RI
the proper types of contrast in adjacent virtual facets near 1.64), the observer sees much of their own dark
of the red areas, the gem will be more interesting and reflection in the middle. The apparently bright outer
attractive, and in many cases appear brighter. Note edges in the photograph of the stone are actually dull;
that while brightness may be the goal, too many areas the observer sees through the gem. This produces a
of brightness can lighten the saturation of the color in stone that is not very attractive or visually dynamic.
large portions of the gem, lowering its market value. The black area of the color-coded contrast map (AG

Figure 17. For this sample gemstone, there is little black (retro-reflection of the observer) in the AG environment.
Red reflection (70°–83° arc) is arranged in a narrow, wagon-wheel pattern from the girdle edge to the near-center.
Red will be bright when the gemstone is held at some distance, but dark if the gem is close to the observer. The
evenly distributed green shows where light is returned from around the observer’s head (55º–70º arc). Violet (40º–
55º arc) comes from a low angle on the horizon and is generally dark. The lowest angles (0º–40º arc) would be seen
as light blue. Together, these work to compose a uniformity of varying contrasts within the gem.

72 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
 A B C D
Figure 18. Four gem-
stones from GIA’s Dr.
Eduard J. Gübelin Col-
lection were scanned
using a non-contact op-
tical measuring device
to obtain a 3D image of
each. Photos showing
the representative pat-
terns when viewed face-
up are equated to the
AG contrast maps de-
rived from DiamCalc.

with white contrast map) is a reflection of the ob- cutting schemes. Six different GemCad designs specif-
server. Red and green indicate higher contrast areas ically for quartz (or other gems with an RI near 1.55)
of brightness, and some of the outer areas contain were chosen from the DataVue file of GemCad de-
dull blue-green, resulting from lower-angle lighting. signs. The name of the design, the creator, and the pub-
Again, the outer white areas are windows. lication where it first appeared are listed. Each shows
The next two examples show better cut optimization: a table where the pavilion depth and crown height
Figure 18C: Despite this amber’s soft appearance change by 1% of girdle width (and all pavilion and
(RI near 1.54) caused by facets that are slightly crown angles change accordingly). The resulting angles
rounded and less crisp, its brightness—even when can be obtained from the GemCad file, but for simplic-
rocked—comes from effective use of high-angle entry ity this article uses crown and pavilion percentage dif-
light to provide dynamic contrast and minimal ob- ferences. The table size is constant, as altering it would
server reflection (black). In the red areas of the map, result in many more combinations. The original de-
the observer’s reflection may come into view when sign’s proportions are outlined and marked “original.”
the gem is slightly tilted. The prominent green areas For reference, the graphic design is also shown, accom-
represent a large amount of bright light entering the panied by a contrast plot for an idealized round bril-
gem. The even distribution of red and green from liant. The idealized contrast plot demonstrates the
culet to girdle adds to the attractiveness. Minimal types of color distribution that contribute to effective
areas of blue (lower-angle lighting) and white (win- designs. It serves as a visual reference for the types of
dowing) keep this gem bright. dominant colors and balance that might be desired.
Figure 18D: This grossular garnet (RI near 1.74) is Design 13022 (figure 19): This design performs well
bright and does not window easily, even when with a variety of angle differences. It is a very forgiving
rocked. Its brightness comes from the effective use design that can be used with a variety of crown heights
of high-angle light entry to create a good mix of dy- to better utilize the rough. But note that all of these
namic contrast, with little reflection of the observer. contrast plots have a predominance of green in the
Red areas indicate that when the gem is slightly middle. Where does the contrast belong? Green repre-
tilted, the observer’s reflection may come into view. sents brightness, but brightness alone is not terribly
The abundance of green areas helps by directing interesting. For more visual interest, a cutter could try
bright light into the gem. For maximum visual ap- a 46% pavilion/21% crown or a 47% pavilion/12%
peal, the red and green areas should be evenly distrib- crown modification to add contrast. As mentioned be-
uted and extend from culet to girdle. In this map, fore, trying several patterns will help identify which
there is little blue (lower-angle lighting) or white specific elements are the brightest and most attractive.
(windowing). Design 13061 (figure 20): This example demon-
strates the need for caution in trusting a cutting plan,
APPLICATION TO GEMCAD-BASED OR regardless of the source. The original gem is fairly
3-D WIREFRAME DESIGNS dull, perhaps due to a typographical error in the plan,
The following discussion provides examples of how to a design not meant for quartz, or incorrect entering
read and interpret the AG contrast maps to optimize of the angles. While a large amount of red around the

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 73
 Norman Steele, Seattle Facet Design, Oct. 1988, p. 3 13022
London Shield

PAVILION DEPTH % DATAVUE #

RI 1.55
44% 45% 46% 47% 48% 49% 50%

11%
CROWN HEIGHT %

12%

13%

14%

15%

16%

17%
ORIGINAL

18%

19%

20%

21%

Figure 19. In this design, the six center pavilion main facets tend to gather light from the same direction, without
breaking it up much (seen here as large green center areas). For more visual interest, a cutter could try a shallower
crown and pavilion, which would give it more contrast into the center.

girdle edge does not always produce the best design, abundant contrast, and the resulting gemstone
the 41% pavilion/11% crown combination may pro- would be quite dynamic. Whereas the contrast
vide the most contrast, with some blue at the girdle areas with combinations of red and red/black can
edge to break up the red. slightly darken the stone, green/red is generally a
Design 13096 (figure 21): This design offers better combination. Designs should minimize blue

74 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
 Ben Dawson, Facets, June 1989 13061
Barion Old Mine Triangle

PAVILION DEPTH % DATAVUE #

RI 1.55
36% 37% 38% 39% 40% 41% 42%

10%
CROWN HEIGHT %

11%

12%
ORIGINAL

13%

14%

15%

16%

17%

18%

19%

20%

Figure 20. This design is fairly dull, and a much steeper pavilion and shallow crown would dramatically improve
its appearance.

points in the center. For this design, a 45% pavil- most dynamic example. It is another forgiving design
ion/11% crown or 41% pavilion/17% crown would and also has high contrast all the way into the middle
be more effective. of the stone, with greens and reds throughout. While
Design 13138 (figure 22): Of the triangular designs the original proportion set has a little too much blue
chosen for these illustrations, this is probably the in that quadrant of the plots, it would still produce a

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 75
 Jim Summers, New Mexico Facetor, Aug. 1994 13096
GemFaire 94

PAVILION DEPTH %
DATAVUE #

RI 1.55
40% 41% 42% 43% 44% 45% 46%

9%
CROWN HEIGHT %

10%

11%

12%

13%

14%

15%

16%
ORIGINAL

17%

18%

19%

Figure 21. This design presents high contrast. The cutter merely needs to choose steeper pavilion angles (≥44%
pavilion depth) and slightly shallower crown angles to avoid windowing.

very good appearance. Slightly increasing the pavilion slightly tilted, sometimes darkening the stone
depth—for instance, 47% pavilion/14% crown— slightly. Therefore, this design should not be used
would improve the design. with dark material, and only with slightly light or
Design 13141 (figure 23): Note the prominence of medium colors. The original angle combinations
red among these choices. The red areas start to pick offer a good balance between contrasts of different
up reflections of the observer (black areas) when light-entry angles. If the material is slightly dark, a

76 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
 Robert Strickland, TFG Newsletter, Oct.-Dec. 1996, Vol. 17, No. 4, p. 18 13138
Trilogy

PAVILION DEPTH % DATAVUE #

RI 1.55
43% 44% 45% 46% 47% 48% 49%

11%
CROWN HEIGHT %

12%

13%

14%
ORIGINAL

15%

16%

17%

18%

19%

20%

21%

Figure 22. This is a forgiving design with high contrast into the center. With pavilion depths from 43% to 49% and
a crown ranging from 12% to 17%, all working effectively, this design can be adapted to shallow or thick rough.

44% pavilion/12% crown will produce a little less the example in the 46% column will also produce win-
darkness in the stone. dowing. At 50%, the substantial amount of blue (low-
Design 13146 (figure 24): Note that this gem has a angle light entry) does not allow much brightness. The
narrow range of feasible pavilion depth. At 45%, there original proportions will work very well, but higher
is strong windowing, which also occurs in the white proportions of green in a contrast map—for instance,
area in the 46% column. While not shown here, tilting 48% pavilion/17% crown—are generally preferable. To

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 77
 Richard C. Walker, Facets, Sept. 1988, p. 3 13141
Brazil Cushion Triangle 1

PAVILION DEPTH % DATAVUE #

RI 1.55
44% 45% 46% 47% 48% 49% 50%

9%
CROWN HEIGHT %

10%

11%

12%
ORIGINAL

13%

14%

15%

16%

17%

18%

19%

Figure 23. Another forgiving design with high contrast, this should be used with lighter materials. The red areas
start to pick up reflections of the observer when the stone is tilted, darkening its appearance.

achieve good contrast, the adjoining facets need to have modifying the design (adding or removing facets, or
offset angles. This design only does so in a very narrow changing their placement). By using lighting schemes
range of proportions. such as AG and ASET in design planning, facet
Computer modeling can assist in effective design arrangement can be optimized to produce the most
by simulating the best angle combinations and/or visually interesting results.

78 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
 Ernie Hawes, New Mexico Facetor, Nov.-Dec. 1999 13146
Third Millennium

PAVILION DEPTH % DATAVUE #

RI 1.55
45% 46% 47% 48% 49% 50% 51%

9%
CROWN HEIGHT %

10%

11%

12%

13%

14%
ORIGINAL

15%

16%

17%

18%

19%

Figure 24. This design is effective in a limited range of proportions and should be used with lighter materials.

GENERAL CUTTING CONSIDERATIONS 2. Step cuts are more difficult to optimize. With
1. Designs for equal-sided gems (e.g., square, round, the addition of four or more rows on the pavil-
equilateral triangular) can easily be made more ion, the difficulty increases. In optimizing step
dynamic. Elongated shapes are inherently more cuts, very careful attention must be paid to step
difficult to optimize to the same degree. width, as well as the design execution.

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 79
Figure 25. Each image shows a 2º rotation clockwise from the previous image. A spot light source contributes a
small amount of fire (seen here as red, blue, green, or yellow virtual facets). As the gem is rotated, some virtual
facets go from light to dark (examples are circled). As a facet nears the border between light and dark, the observer
will see fire. Some of these flares of color last through larger degrees of rotation than others (larger red circles).
While not illustrated, when a facet only slightly darkens or lightens, fire is faint. Strong changes in contrast create
stronger fire. More fire is exhibited by materials of higher dispersion values.

3. Some designs are more effective for certain RI source directed at the gem), the observer
ranges than others. For example, the standard will see little fire. If the lighting environ-
round brilliant is far more effective for gems ment provides strong contrast, using bright
with a high RI (diamond, zircon, etc.) than for spot-lighting against a dark background, fire
lower-RI materials. is strongly enhanced.
4. The gem’s physical size also dictates design b. Strong contrast: Areas of strong contrast in
considerations. For example, when cutting a 10 the gem will create fire as those areas alter-
mm gem, a simple single-cut design with eight nate between dark and bright when the gem
pavilion and eight crown facets (plus the table is rotated. Figure 25 shows a gem being ro-
facet) is not as effective as a 57-facet round bril- tated in 2º increments with the light source
liant design. A 181-facet design, for instance, in a fixed position. As the gem is rotated, the
does not succeed unless the gem is of substan- movement of facets causes them to fluctu-
tial size, because tiny facets are less distinct ate between dark and light. At the threshold
and look “fuzzy.” between light and dark, the observer will see
5. Adding small facets to designs that contain fire. If the gem is moved slowly, with facets
large facets is rarely practical. For example, only slightly changing in relation to the
splitting stars into three small facets while light source and the reflection to the eye, the
leaving the others at a normal size will have lit- change in color of the facet will be gradual.
tle effect and may actually diminish the design. If movement is sudden, the observer may
Often that center facet provides less contrast only see where the facet is dark and then
than if there were only two facets. light, without seeing the fire. When a facet
6. Every now and then, faceters concern them- only slightly darkens or lightens, fire is
selves with dispersion, which jewelers refer faint.
to as “fire.” It is a measurement that indi- Following these rules for optimization, such as
cates how much light is spread by different creating a design with strong contrast, will also
materials. Generally, lower-RI gems do not improve the appearance of a gem and increase its
contain much visible fire. Adding color satu- fire.
ration to the material hides what little fire Before implementing personal preferences, the
might be there. On the other hand, highly dis- faceter should be aware that the colored gem market
persive gem materials (with dispersion values is very focused on color: hue, saturation and tone.
above 0.039) provide some unique opportuni- Color quality is critical, but often suffers in lightly
ties to display fire. This effect depends on two colored gems when the design is modified toward
considerations: brightness. Instead, optimization may be needed to
enhance the color first and foremost, with scintilla-
a. Lighting environment: Where the lighting is tion a secondary priority. Understanding the ele-
primarily diffused (e.g., reflected from a ments of brightness can help the faceter achieve it
white ceiling with no direct spot-type without compromising color.

80 OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013
 ABOUT THE AUTHOR suggestions and the organization of this paper. The author greatly
Mr. Gilbertson is the project manager of cut research for GIA in appreciates comments and suggestions about content from Dale
Carlsbad, California. Carriere, Lisa Elser, Nathan Renfro, Michael Cowing, and Wayne
ACKNOWLEDGMENTS Emery.
The author acknowledges Dr. Ilene Reinitz for her many helpful

REFERENCES
Albert M.K. (1995) Genericity and the Perception of Visual Con- stimulated fields]. Rivista di Psicologia, Vol. 49, No. 1, pp. 7–
tours and Surfaces. Ph.D. dissertation, University of California, 30. Translated by W. Gerbino, in Petry and Meyer, 1987, pp. 40–
Irvine, p. 25. 49).
Caudill J., Sasian J., Yantzer P. (2008) Methods, apparatus, and sys- Petry S.J., Meyer G.E. (1987) The Perception of Illusory Contours.
tems for evaluating gemstones. US Patent number: 7336347, Springer-Verlag, New York, NY.
filed Dec 20, 2004, issued Feb. 26, 2008. Purves D. (2011) An empirical explanation: Simultaneous brightness
Cowing M. (2009) Describing diamond beauty—assessing the op- contrast. Center for Cognitive Neuroscience, Duke University,
tical performance of a diamond. ACA Gem Laboratory, [Link]
[Link] [date ac- [Link] [date accessed: Aug. 30, 2012].
cessed: Aug. 30, 2012]. Reflector technologies (undated) Good Old Gold, [Link].
Gilbertson A. (2003) Device for judging symmetry, brightness, and com/technologies/reflectortechnologies/history [date accessed:
efficacy of light return in precious stones. U.S. Patent 6665058, June 5, 2013].
filed May 24, 2000, issued Dec. 16, 2003. Sarcone G., Waeber M.-J. (undated) Colors of the mind,
Harding B. (1975) Faceting limits. G&G, Vol. 15, No. 3, pp. 78–88. [Link]
Hoffman D. (1998) Visual Intelligence: How We Create What We [accessed on Aug. 30, 2012].
See. W.W. Norton & Co, New York, NY. Sasián J., Quick J., Sheffield J., Caudill J., Yantzer P. (2007) Evaluation
GIA (2006) Diamond Grading Lab Manual. Gemological Institute of brilliance, fire, and scintillation in round brilliant gemstones.
of America, Carlsbad, California, p. 21. Optical Engineering, Vol. 46, No. 9, p. 60936014–6093625.
Kanizsa G. (1955) Margini quasi-percettivi in campi con stimo- White M. (1979) A new effect on perceived lightness. Perception,
lazione omogenea [Quasiperceptual margins in homogenously Vol. 8, pp. 413–416.

For online access to all issues of GEMS & GEMOLOGY, visit:

[Link]/gems-gemology

OPTIMIZING FACE-UP APPEARANCE IN GEMSTONE FACETING GEMS & GEMOLOGY SUMMER 2013 81
FEATURE ARTICLES

FLUORESCENCE PRODUCED BY OPTICAL

DEFECTS IN DIAMOND: MEASUREMENT,
CHARACTERIZATION, AND CHALLENGES
Yun Luo and Christopher M. Breeding

Three-dimensional fluorescence spectra were collected from both natural-color and treated diamonds
with common color centers (including N3, H3, H4, 480 nm, and N-V) to characterize the fluorescence
produced by each defect. Unlike individual spectra, 3D presentations of multiple spectra allow quick
and simultaneous determination of the fluorescence-producing defect(s), the excitation energy that yields
maximum fluorescence intensity, the variation of fluorescence with excitation, and the peak position
and band shape of individual and overall fluorescence emissions. The combination of 3D fluorescence
spectra from common defects and emission spectra from several standard ultraviolet light sources re-
vealed noticeable inconsistencies in the fluorescence observed. Our data indicate that variations in UV
lamp output can significantly affect the fluorescence color observed in gem diamonds.

F
luorescence is the emission of electromagnetic fluorescence from one color center may be absorbed
radiation (e.g., visible light, ultraviolet light, X- by other defects, dramatically reducing the overall
rays, and gamma rays) from a substance, stimu- fluorescence. Phosphorescence commonly refers to
lated by the absorption of incident electromagnetic visible light emitted by a diamond after the stimu-
radiation. The emission persists only as long as the lating UV light source has been discontinued. While
stimulating radiation is continued. The terms fluo- related to fluorescence, it is a relatively uncommon
rescence and photoluminescence often appear inter- visual property of gem diamonds and not a focus of
changeably, but gemologists commonly use the this study.
former to describe the visible light emission from ul- Fluorescence in diamonds has been studied for
traviolet (UV) excitation. In most cases the emitted nearly a century (e.g., Becquerel, 1868; Mani, 1944;
light has a longer wavelength, and therefore lower en- Shipley, 1947; Wild and Biegel, 1947; Cotty, 1956;
ergy, than the absorbed radiation. Diamonds contain- Dyer and Matthews, 1958; Collins, 1974, 1982; Fritsch
ing no defects or impurities generally do not absorb and Waychunas, 1994; Eaton-Magaña et al., 2007; Hol-
visible and UV light (>230 nm) and thus produce no loway, 2009; Shigley and Breeding, 2013). While much
color or fluorescence. But when impurities or defects is known and published about diamond defects and
(often referred to as color centers) are present, they fluorescence, most gemologists do not have ready ac-
may absorb visible and UV light to produce color, flu- cess to this information. This article seeks to clearly
orescence, or both. In diamonds with multiple color outline the defects that cause fluorescence in dia-
centers, the fluorescence from one defect may even mond, and explain how the gemological tools used to
excite the emission from another center more effi- evaluate this property may impact observations. For
ciently than the external UV light. Conversely, the decades, investigations of diamond fluorescence have
mainly focused on visual observation (Moses et al.,
1997; figure 1), while more recent studies have dealt
See end of article for About the Authors and Acknowledgments.
with individual spectra and the physics of individual
GEMS & GEMOLOGY, Vol. 49, No. 2, pp. 82–97,
[Link] defects. The application of 3D fluorescence spec-
© 2013 Gemological Institute of America troscopy has been limited. Eaton-Magaña et al. (2007)

82 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
Figure 1. For decades, UV lamps have been used in the jewelry industry and by gemological laboratories to assess
diamond fluorescence. Left: A gemologist examines fluorescence reactions from a suite of colorless faceted dia-
monds in a long-wave UV fluorescence box. Right: A group of rough stones illustrates the range of fluorescence col-
ors seen in diamond under LWUV excitation. Note that GIA evaluates fluorescence of individual stones, not in
bulk as shown here. Photos by Eric Welch.

presented a few spectra for diamond in this format, the materials used in the UV filters, and aging of
and Hoover and Theisen (1993) investigated other col- lamps and filters can also affect the radiation output.
ored gemstones with this approach. Secondly, multiple defect centers may exist within
Visual evaluation of fluorescence is standard prac- the same diamond and produce different emission
tice in gemological laboratories, and fluorescence peaks simultaneously. To the human eye, multiple
color and intensity are often documented on labora- emissions combine to produce a single, albeit mixed,
tory reports. When selecting diamonds, jewelers also fluorescence color (e.g., blue + yellow = apparent
rely on fluorescence to ensure a good color match whitish fluorescence). Visual evaluation of fluores-
under a variety of lighting circumstances. The most cence does not allow for identification of multiple
common excitation source is a handheld UV lamp, defects or irregular emission peaks. Finally, the dis-
which often has dual modes for long-wave (LW) and tance between the radiation source and the diamond
short-wave (SW) UV radiation. The typical color cen- as well as the stone’s orientation (table-down versus
ters in diamond fluoresce more intensely under face-up) may produce noticeably different fluores-
LWUV excitation. Fluorescence assessment relies cence by changing the amount of excitation energy
mainly on visual observation at room temperature— that interacts with the diamond.
fluorescence can vary in intensity and color at differ- In this study, a high-resolution luminescence
ent temperatures—and sometimes requires a set of spectrometer was used to study fluorescence emis-
reference diamonds for comparison. Observation of sions from common color centers in gem diamonds
fluorescence can be affected by three main factors: (1) by obtaining three-dimensional scans of excitation,
the nature of the emission from the UV light source; emission, and fluorescence intensity. From the 3D
(2) the nature of the defect(s) responsible for the fluo- scans, we can quickly and simultaneously determine
rescence; and (3) methodology, including the viewing the nature of observed fluorescence and the energy
geometry and the distance from the radiation source. at which the fluorescence maximizes or changes by
Within current industry practices, complications exist systematically varying the excitation energy wave-
for each of these factors. length. Furthermore, emission characteristics such
First, the excitation produced by different light as peak position and band shape that are characteris-
sources may vary and not consist of “pure” LWUV tic of particular defect centers can be evaluated.
or SWUV light, which generally have wavelengths of With detailed, direct knowledge of fluorescence-
365 nm and 254 nm, respectively. Differences in ex- producing diamond defects in hand, we surveyed sev-
citation wavelengths between lamps may produce eral common UV light sources to better understand
inconsistent fluorescence reactions. Warm-up time, how much variability exists in UV excitation and to

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 83
 Figure 2. UV light sources from
this study include (A) a standard
gemological UV lamp, (B) the Min-
eralight hobbyist lamp, (C) a com-
mercial UV fluorescence loupe, (D)
a fiber-optic LED light source, and
(E) the Thermo ABII luminescence
spectrometer. Photos by Eric Welch
(A) and Christopher M. Breeding
(B–E).

further investigate the effect of UV source on observed Emission Spectra from Industry UV Light Sources.
fluorescence colors. While a few previous articles have To evaluate the excitation energy produced, we col-
made similar investigations (e.g., Eaton-Magaña et al., lected emission spectra from various UV light
2007; Pearson, 2011), our survey includes newer LED- sources over the 200–900 nm range using an Ocean
based sources that have only recently seen widespread Optics USB 2000 CCD spectrometer with a UV-
use in the industry. The outcome of this study could transmitting fiber-optic cable. Output spectra from
be beneficial to both the diamond industry and gemo- five UV sources—including a GIA UV lamp (model
logical laboratories in that accurate, consistent, and
reproducible fluorescence measurements are vital for
industry confidence. In addition, better methods for
measurement of fluorescence from different defect In Brief
centers may lead to more effective methods of detec- • 3D fluorescence spectra allow quick, simultaneous char-
tion and separation of natural, synthetic, and treated acterization of the excitation energy that maximizes fluo-
diamonds. rescence, variation in fluorescence with excitation, as
well as position and band shape of individual and overall
fluorescence emissions.
MATERIALS AND METHODS • Common mercury-based UV lamps and LED-based UV
Samples. Three-dimensional fluorescence spectra loupes do not produce pure 365 nm LWUV emissions.
were collected from 25 faceted diamonds, including • UV lamp output variation can significantly affect fluores-
19 natural-color samples and six treated by high-pres- cence color and intensity observed in diamond.
sure, high-temperature (HPHT) and irradiation (table
• A standardized UV excitation source is necessary for ac-
1). GIA issues reports on thousands of colored dia- curate and reproducible determination of fluorescence.
monds each year, providing a large selection of sam-
ples. The diamonds in this study, all of which carried
GIA reports indicating natural- or treated-color ori-
gin, were chosen as representative of their respective 745000), a custom-made GIA fluorescence measure-
defect group. While multiple fluorescence-producing ment device with UV LEDs (not shown), a handheld
color centers occur in most diamonds, each of our UVP Mineralight hobbyist lamp, a UV LED loupe,
samples was dominated by one of the following: and a commercial 365 nm LED light source (figure 2,
A–D)—were measured and output peak positions
1. N3 center were recorded.
2. H3 or H4 center
3. 480 nm band 3D Luminescence Spectroscopy. A Thermo Aminco
4. N-V center Bowman II (ABII) luminescence spectrometer (figure

84 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
 Fluorescence Spectroscopy N3 defect ZPL = 415 nm
1.4
Excitation Artifact
90000

UV-VIS Absorbance (a.u.)

N3-related fluorescence

(-0.003 – 9.999)
FLUORESCENCE 1.2

PL Intensity (counts)
5 Excitation artifact 1 80000

.8 N3-related absorption N3-related emission

0
50 UV-VIS 325 nm PL 70000
0 .6

EX 40
CI 0
TA 700 .4 60000
TIO 30 600
N 0 500
(nm 22 400
ION (nm) .2
) EMISS
0 300
50000

360 380 400 420 440 460 500

WAVELENGTH (nm)

FL 365 nm excitation 10 440 nm emission PLE

FLUORESCENCE (arb. units)

FLUORESCENCE (arb. units)

8
N3 defect (ZPL = 415 nm)

N3 defect (ZPL = 415 nm) 8

Excitation Artifact

Excitation Artifact
6
6

4
4

2
2

0 0

350 400 450 500 550 600 650 700 750 250 300 350 400 450 500

EMISSION WAVELENGTH (nm) EXCITATION WAVELENGTH (nm)

Figure 3. Top left: 3D fluorescence spectra are composed of many individual fluorescence spectra collected at dif-
ferent excitation wavelengths. This example from a colorless diamond containing N3 defects (sample 2) shows
how a single spectrum (top left, blue line) at a given excitation (365 nm) can be extracted and examined in detail
(bottom left). PLE excitation spectra can also be extracted from the 3D data (top left, red line and bottom right).
Top right: Absorption (UV-Vis) and PL emission spectra confirm the presence of a defect such as N3 and relate to
the data taken from a 3D spectrum. Fluorescence (bottom left) and PL (top right, blue line) emissions from the de-
fect are similar and longer in wavelength than the ZPL, whereas PLE (bottom right) and UV-Vis absorption spectra
(top right, red line) correlate and show defect-produced bands at wavelengths shorter than the ZPL.

2E) was used to investigate the samples’ fluores- presented, the 2× and 3× artifacts have been re-
cence. The analyses were conducted at room tem- moved. The 1× artifact was retained to avoid poten-
perature to simulate typical observation conditions. tial data loss.
Three-dimensional fluorescence spectra were Each 3D scan took approximately four hours to
recorded in the 300–750 nm emission range (2 nm complete and consisted of 57 individual spectra. The
resolution, 4 nm bandpass) using an excitation of advantage of a 3D plot is that all aspects of excitation
220 to 500 nm (5 nm intervals, 4 nm bandpass). De- and emission can be derived. Any of the 57 fluores-
tector voltage was adjusted so that the highest peak cence emission spectra for a given excitation can be
in the spectrum comprised 80%–90% of the instru- separated, and data for excitations not specifically an-
ment’s maximum fluorescence intensity range. The alyzed can be interpolated. In addition, photolumi-
ABII spectrometer is not configured with excitation nescence excitation (PLE) spectra for a given
filters, so most of the 3D spectra collected contained emission wavelength can be obtained to see the ef-
first-, second-, and sometimes third-order artifacts fect of excitation light frequency on fluorescence.
produced by the spectrometer grating at 1×, 2×, and Figure 3 illustrates these features in the anatomy of
3× the excitation wavelength. In most of the spectra a 3D fluorescence spectrum.

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 85
Additional Absorption and PL Spectroscopy. Ultravi- 365 nm (the typical LWUV emission) was more in-
olet/visible/near-infrared (UV-Vis-NIR), Fourier- tense than the 252 nm SWUV peak (figure 4B), likely
transform infrared (FTIR), and photoluminescence due to filter aging. The existence of this intense 365
(PL) spectra were collected to complement the fluo- nm peak in the excitation may have a pronounced
rescence data, providing additional detail about the effect on what is observed as SWUV fluorescence. Be-
defects under more readily measurable conditions cause most gemologists use UV lamps for many
(again, see figure 3). UV-Vis-NIR spectra were col- years and never replace the filters, these tests are rep-
lected with an Ocean Optics integrating sphere and resentative of industry practices.
HR 4000 CCD spectrometer over the 380–1000 nm
range (~1 nm resolution) at ~77K temperature. FTIR UVP Mineralight Lamp. The UVP Mineralight is an-
spectra (not shown) were collected to identify dia- other dual LW-SW lamp widely used in the industry
mond type using a Thermo Nexus 6700 FTIR spec- and by hobbyists (figure 2B). Though similar in ap-
trometer (6000–400 cm–1, 1 cm–1 resolution, 128 pearance and function to the GIA UV lamps, this
scans, room temperature). Breeding and Shigley lamp showed different results. LWUV excitation pro-
(2009) describes diamond type and its use in gemol- duced an intense broad band at 355 nm, along with
ogy. Photoluminescence spectra were collected using the primary LW peak at 365 nm (FWHM = 37 nm)
a Renishaw inVia Raman microscope (325, 488, 514, and weak emissions at 313, 405, and 437 nm. Short-
633, and 830 nm laser excitations, various scan wave excitation showed a similar array of peaks
ranges, and ~77K temperature). using the GIA lamp, with major emissions at 254
(FWHM = 4 nm), 313, and 366 nm, as well as minor
RESULTS peaks at 297, 303, 335, and 405 nm (figures 4C and
UV Light Sources. GIA UV Lamp. Gemologists use 4D).
conventional UV lamps (figure 2A) to observe a Note: Both the GIA and Mineralight lamps are
stone’s fluorescence color in response to long- or able to produce LWUV and SWUV because they con-
short- wave radiation. The lamps in this study were tain low-pressure mercury lamps that emit initially
sold by GIA Instruments until 2009 and are still sharp mercury lines, but the phosphors inside the
prevalent in the trade. It is well known that tradi- lamp are creating a conversion to broadband wave-
tional UV lamps do not provide single characteristic lengths at a longer wavelength. The lamps emit a
365 nm (LWUV) and 254 nm (SWUV) emission lines range of light that spans both LWUV and SWUV.
(Williams, 2007; Pearson, 2011) due to the variety of They are equipped with a daylight filter that removes
materials used in UV filters and the aging of lamps most of the visible light so that mainly the desired
and filters. Rather than a single 365 nm emission UV band passes through the filter (R. Geurts, per-
peak, LWUV lamps often emit 404 and 435 nm lines sonal communication, 2012).
and a broad band that extends from the UV to the vis-
Digital GIA LED Fluorescence Device. GIA uses a
ible region of the spectrum. Similarly, SWUV lamps
custom-made fluorescence meter with LWUV LEDs
always have distinct peaks at 254, 315, and 365 nm
to measure the fluorescence intensity documented
instead of a single 254 nm emission. All of these
on its diamond grading reports. Designed to improve
emission lines, as well as several weaker ones, are
fluorescence measurement consistency, this instru-
produced by the mercury lamp inside the units. The
ment uses an LED excitation source with a single
filter materials used to remove mercury radiation are
emission peak at 367 nm and an FWHM of 14 nm
not 100% efficient and leak undesired emissions. As
(figure 4E).
the filters age, the leaked emissions become stronger
and more prominent. UV LED Loupe. A generic commercial UV loupe
Examination of two handheld UV lamps manu- sometimes used in the trade (figure 2C) was also ex-
factured by GIA and used at its laboratory revealed a amined. The LED excitation source from this loupe
major peak at 368 nm (FWHM = 17 nm) with LWUV gave a single peak at 403 nm with an FWHM of 17
excitation, accompanied by a small peak at 404 nm nm (figure 4F).
(figure 4A). In addition to the 252 nm (FWHM = 3
nm) SWUV emission, which shifted 2 nm from typ- Commercial LED Light Source. For comparison with
ical 254 nm SWUV, there were major peaks at 312, the other LED sources, we also tested a commer-
365, 404, and 435 nm, as well as minor peaks at 296, cially available “365 nm” LED light source manufac-
302, and 334 nm. Interestingly, the peak intensity at tured by Ocean Optics (figure 2D). This excitation

86 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
 SPECTRA FROM UV LIGHT SOURCES
A 4000 B
368 nm UV lamp - LW 365 nm UV lamp - SW
3000 252 nm
3000

2000
2000 312 nm

404 nm
1000 1000
404 nm 435 nm

0 0

200 300 400 500 600 700 800 200 300 400 500 600 700 800

4000 C D 366 nm
365 nm UV mineral lamp - LW 4000 UV mineral lamp - SW
355 nm
3000 3000
313 nm
254 nm
2000 2000

1000 1000
313 nm 405 nm
405 nm 437 nm
0 0

200 300 400 500 600 700 800 200 300 400 500 600 700 800

3500 F
E 367 nm 403 nm
4000 GIA digital UV loupe
INTENSITY (COUNTS)

fluorescence meter 3000

3000 2500

2000
2000
1500

1000
1000
500

0 0

200 300 400 500 600 700 800 200 300 400 500 600 700 800

G 364 nm H 366 nm
LED - 365 nm ABII - 365 nm
3000
3000 (bandpass = 4 nm)
2500

2000
2000
1500

1000 1000

500

0 0

200 300 400 500 600 700 800 200 300 400 500 600 700 800

4000
I
367 nm ABII - 365 nm
(bandpass = 16 nm)
3000

2000

1000

200 300 400 500 600 700 800

WAVELENGTH (nm)

Figure 4. The UV light sources from this study showed clear differences in emission characteristics.

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 87
TABLE 1. Diamonds examined in this study.

Sample Natural color Colora Weight Shape LWUV SWUV Prominent

or treated (ct) reactionb reactionb color center

1 Natural Colorless (D) 1.10 Round Medium green None

2 Natural Colorless (D) 0.37 Round Strong blue Medium green
3 Natural Colorless (E) 0.31 Round Medium blue None
4 Natural Near-colorless (I) 1.12 Round Strong blue Weak green N3
5 Natural Near-colorless (I) 1.51 Round Strong blue Weak green
6 Natural Faint yellow (L) 1.01 Cushion Strong blue Medium green
7 Natural Very Light pink 1.02 Heart Strong blue Medium blue
8 Natural Fancy Light greenish yellow 0.14 Round Strong yellow Medium yellow
9 Natural Fancy yellow-green 0.69 Round Strong blue + Medium blue +
yellow (uneven) yellow (uneven)
10 Natural Fancy Intense greenish yellow 1.01 Heart Medium to strong Weak yellow H3
yellow + blue
11 HPHT-treated Fancy Vivid green-yellow 2.34 Round Strong green + Medium green
weak blue
12 Natural Fancy Intense yellowish green 3.10 Rectangle Very strong green Strong green
13 Natural Fancy Light yellow-green 0.58 Round Strong green Strong green H4
14 Natural Fancy Light yellowish green 1.04 Round Strong green Medium green
15 Natural Fancy Deep orange-yellow 0.40 Round Strong yellow Medium yellow
16 Natural Fancy Deep brownish 1.62 Pear Strong orange Medium to strong
orangy yellow orange
17 Natural Fancy Vivid yellow-orange 0.72 Pear Strong orange Strong orange 480 nm band
18 Natural Fancy Intense orange-yellow 1.02 Marquise Strong orange Strong yellow
19 HPHT-treated Fancy Intense orangy yellow 0.19 Round Strong yellow Medium yellow
20 HPHT-treated Fancy Intense yellow 0.39 Round Strong orange Medium orange
(partial)
21 Natural Fancy brown-pink 1.51 Old Mine Weak red Weak to medium red
22 Natural Very Light pinkish brown 2.01 Pear Very weak orange Very weak orange
23 HPHT-treated Fancy red 1.49 Square Weak red Medium red
and Irradiated N-V
24 HPHT-treated Fancy Deep pink 0.90 Oval — —
and Irradiated
25 Irradiated Fancy red 0.17 Round Strong red Very strong red

a
Assigned using standard GIA color grading practices.
b
LWUV and SWUV reactions were observed using excitation from a standard handheld UV lamp.

source showed a single peak at 364 nm with an emission showed a narrow peak at 366 nm with an
FWHM of 10 nm (figure 4G). FWHM of 5 nm. With a 16 nm bandpass parameter
(which most closely resembled the output of a GIA
ABII Excitation Light Source. The ABII spectrometer LW-UV lamp), a peak was observed at 367 nm with
(figure 2E) was selected to provide well-constrained, an FWHM of 15 nm. To precisely characterize the
high-resolution fluorescence spectra. To understand fluorescence from a given excitation wavelength, we
how its excitation compared to that of other sources, chose the smallest available bandpass setting (4 nm),
different excitation wavelengths of the ABII CW and thus the narrowest excitation emission, for the
Xenon source were examined (the instrument can 3D-fluorescence spectra collection.
produce excitation of 200–950 nm). As shown in fig-
ures 4H and 4I, when the instrument was set to 365 N3 Center. The N3 center is the most common
nm excitation with 4 nm bandpass, the resultant color-producing defect in diamond, consisting of a va-

88 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
 Figure 5. Diamonds
with N3 defect centers
include samples 1–3,
which are colorless (1
and 3 have a brownish
hue); samples 4–6,
which are near color-
less to faint yellow (not
shown); and sample 7,
a pink diamond with
both N3 and H3 defect
centers. Samples with
the H3 center (8–11)
and H4 center (12–14)
are yellow to light
green. Diamonds with
the 480 nm band (15–
20) are yellow to or-
ange, while those with
N-V centers (21–25) are
pale to intense pink or
red. Photos by Don
Mengason and Robison
McMurtry.

cancy surrounded by three nitrogen atoms on a {111} 1982), possibly explaining the weaker fluorescence
plane. The zero phonon line (ZPL), the identifying observed (table 1). PL spectra confirmed the pres-
emission for N3, occurs at 415 nm. This line, along ence of N3 in sample 2 (figure 3). The 503.2 peak
with a broadband emission with dominant peaks at in the PL spectra of samples 1, 3, and 7 revealed an
429, 439, and 450 nm, results in typical blue fluores- H3 center as well. The H3 center in sample 1 was
cence (Collins, 1982). the source of the visible greenish component of the
fluorescence.
Absorption and PL Spectroscopy. UV-Vis-NIR spec-
tra for samples 1–7 all showed the N3-related ZPL Fluorescence Spectroscopy. Among the seven dia-
at 415 nm. Sample 7 also showed a 550 nm band, monds examined in this category, six have color in
which causes the stone’s pink bodycolor. FTIR the D-to-Z range (colorless to pale yellow) and one has
analysis identified them as type Ia diamonds with a light pink color (figure 5). All seven showed rela-
both A and B nitrogen aggregates in various concen- tively intense fluorescence emission, with the ZPL at
trations. Samples 1 and 3 showed more A aggregates, 415 nm and three peaks at 429, 439, and 450 nm su-
which are known to quench fluorescence (Collins, perimposed on a ~400–550 nm broadband, all charac-

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 89
 N3 CENTER
Sample #2 A Sample #2 B
BP = 4 nm BP = 16 nm
FLUORESCENCE

FLUORESCENCE
N3-related fluorescence
(-0.01 – 9.999)

(-0.01 – 9.999)
N3-related fluorescence
5 5

0 0
50 50
0 0

EX EX
CI CI
TA 700 TA 700
TIO 600 TIO 600
N 500 N 500
(nm 22 400 (nm) (nm 22 400
ION (n
m)
SION
0 0
300 300
)
EMIS ) EMISS

C D
Sample #7 N3
FLUORESCENCE

N3-related fluorescence
(-0.01 – 9.999)

ABSORBANCE (offset)
5
H3-related fluorescence

H3
0
50
0
EX
C
IT
AT
I O
N 22 700 Sample 7
0 600
500
(n 300 400
Sample 2
m
) EMISSION (nm)

400 450 500 550 600 650 700

WAVELENGTH (nm)

Figure 6. These 3D fluorescence spectra were collected from diamonds with N3 as the major defect center. The top
two spectra are from the same sample using different excitation energy peak widths. Sample 7 (bottom left) con-
tains both N3 and H3 centers. UV-Vis-NIR spectra (bottom right) confirm the particular fluorescence defects. Exci-
tation artifacts are marked in blue.

teristic of the N3 center (Eaton-Magaña et al., 2007). H3 center, which was observed from photolumines-
The 3D fluorescence spectra indicated that this N3- cence spectra in these three samples but not in the
related emission reaches its maximum at 395 nm ex- other four from this category. As stated earlier, most
citation energy and is not excited above 430 nm diamonds contain multiple color centers.
(figure 6). Figures 6A and 6B show fluorescence spec-
tra collected on sample 2 but with different bandpass H3 and H4 Center. The H3 center consists of a va-
settings (4 and 16 nm). From these two spectra, it is cancy trapped at an A aggregate of nitrogen, which
clear that the bandpass does not affect the position or gives an uncharged defect consisting of two nitrogen
maximum intensity of N3 fluorescence. Similar re- atoms separated by a vacancy—for instance, (N-V-
sults occurred for other defects. Although the band- N)0. Optical absorption at this center has a ZPL at
pass of 16 nm most closely resembles the output of a 503 nm and creates a yellow bodycolor in diamond,
handheld LWUV light source, the 4 nm bandpass cho- while also commonly producing green luminescence
sen for all the 3D fluorescence spectra in this study with visible-light stimulation. The H4 center con-
adequately represented the corresponding excitation. sists of four nitrogen atoms separated by two vacan-
Aside from the N3-related emission, which is not cies. It is created when a vacancy migrates through
active with excitation energy above 435 nm, sample the diamond lattice and combines with a B aggregate
7 showed a 520 nm emission band with a maximum of nitrogen. This center has a ZPL at 496 nm and typ-
at 470 nm excitation. This feature is caused by the ically produces yellow color in diamond. Both H3

90 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
 H3 AND H4 CENTER

Sample #8 A Sample #10 B

H3-related fluorescence
FLUORESCENCE H3-related
H 3-relat
e ed fluorescence

FLUORESCENCE
N3-related fluorescenc
fluorescence
ce
(-0.01 – 9.999)

(-0.01 – 9.999)
5 5

0 0
50 50
0 0

EX EX
CI CI
TA 700 TA 700
TIO 600 TIO 600
N 500 N 500
(nm m) (nm
22 400 22 400 m)
0 300 ION (n ) 0 300
ION (n
) EMISS EMISS

Sample #12 C D
FLUORESCENCE

H4-related fluorescence H3
(-0.01 – 9.999)

PL INTENSITY (offset)
5 H4

0
50
0 Sample 10

EX Sample 8
CI 700
TA
TIO 600
N 500
(nm 22 400 )
)
0 300
IO N (nm
EMISS Sample 12

494 496 498 500 502 504 506

WAVELENGTH (nm)

Figure 7. These 3D fluorescence spectra were collected on diamonds with H3 (samples 8 and 10) and H4 (sample
12) as the major defect centers. Both H3 samples also contained weak N3 emissions. In the bottom right corner,
PL spectra show the presence of H3 and H4 in the respective samples. Excitation artifacts are marked in blue.

and H4 centers can occur naturally or during treat- 8, 9, 10, and 11, as well as ZPL at 415 nm from the
ment (Collins, 1982). N3 center. The H3 defect requires the presence of ag-
Seven diamonds were examined in this category, gregated nitrogen, making the occurrence of N3 in
including six samples determined by GIA to be nat- such stones very likely. Samples 12, 13, and 14
urally colored and dominated by H3 and H4 (three showed H4-related ZPL at 496 nm in the UV-Vis-NIR
each), as well as one HPHT-treated diamond domi- spectra. FTIR analysis identified sample 8 as type IaA
nated by H3 (samples 8–14; figure 5). Sample 8 is a diamond, and samples 9, 10, and 11 as type Ia with
Fancy Light greenish yellow stone that fluoresced both A and B aggregates in various concentrations.
strong yellow under LWUV and medium yellow Samples 12, 13, and 14 are type IaB diamonds. Strong
under SWUV. Samples 9, 10, and 11 all have some 503 nm peaks for samples 8, 9, 10, and 11 and 496
combination of yellow and green bodycolors and ex- nm peaks for samples 12, 13, and 14 in the PL spectra
hibited an uneven fluorescence with blue and yellow further confirmed the presence of dominant H3 and
zones. Samples 12, 13, and 14 have yellowish green H4 centers, respectively (figure 7D).
bodycolors and showed very strong green fluores-
cence under both long- and short-wave UV. Fluorescence Spectroscopy. The 3D fluorescence
spectra of samples 8–11 show an emission maximum
Absorption and PL Spectroscopy. UV-Vis-NIR spec- at 520 nm with a wide band extending from ~480 to
tra showed the H3-related ZPL at 503 nm for samples 650 nm, a feature that manifested itself as a green or

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 91
yellowish green fluorescence (figures 7A and 7B). 10 470 nm 520 nm

This emission feature is caused by H3 and has a max-

imum at 470 nm excitation. Unlike N3, the 503 nm 8

ZPL of H3 is not seen at room temperature. The N3-

related fluorescence feature was also observed with 6

various intensities due to different defect concentra-

tions (figure 7B). The spectrum for 365 nm excita- 4

tion, which most closely approximates the LWUV

FLUORESCENCE INTENSITY
2
lamp, showed both H3 and N3 emissions at rela-
tively low intensity. Fluorescence spectra from the 0
H3

HPHT-treated sample (11) were not noticeably differ-

ent from those of other H3-dominated natural-col- 300 400 500 600 700
ored samples.
The 3D fluorescence spectra of samples 12–14
10
(see sample 12 in figure 7C) showed an emission peak
centered at 512 nm with a band from ~470 to 630 nm
8
that resulted in green color from H4 fluorescence.
This emission feature increases at 395 nm excitation 6
and maximizes at 470 nm. Similar to H3, the ZPL for
H4 (496 nm) is not visible at room temperature but 4

was clearly seen in liquid nitrogen–cooled PL spectra

for these samples. 2

Both H3- and H4-related fluorescence emissions H4

showed a broad fluorescence band at ~470 to 650 nm. 0

But the H3 fluorescence peak is shifted to the higher- 300 400 500 600 700
wavelength side and has a wider tail toward this di- WAVELENGTH (nm)
rection, resulting in a yellowish fluorescence rather
than the green fluorescence of H4 (figure 8). Figure 8. Fluorescence spectra collected at 470 nm ex-
citation from H3 (top, sample 8) and H4 centers (bot-
480 nm Band. Although the structure of the 480 tom, sample 12) using the ABII luminescence
nm band is not understood, it is known to be asso- spectrometer show distinct differences in emission
ciated with yellow fluorescence in diamond. Six di- peak position and distribution (represented by dashed
lines). Excitation artifacts are marked in blue.
amonds with 480 nm bands in the UV-visible
absorption spectra were examined (samples 15–20;
figures 5 and 9). Four of the samples were naturally
colored, and two were HPHT-treated. Samples 15– wide band from 600 nm to more than 850 nm, typical
19 had a combination of orange and yellow in body- of diamond with 480 nm bands.
color and fluoresced strong yellow to orange and
medium yellow to orange under long- and short- Fluorescence Spectroscopy. 3D fluorescence spectra
wave UV lamps, respectively. Sample 16 had a from 480 nm-band diamonds are complex. All of the
brownish color component as well. Sample 20 had samples showed two major features: (1) a wide band
yellow bodycolor and showed strong orange and centered at ~653 nm (maximum shifts between 630
medium orange fluorescence under long- and short- and 655 nm) that maximized with 485 nm excitation
wave UV, respectively. energy, and (2) a ~539 nm-band (maximum shifts be-
tween 505 and 541 nm) that increased with 420, 345,
Absorption and PL Spectroscopy. UV-Vis-NIR spectra and 285 nm excitation energies (figure 9, A–C). The
showed the typical 480 nm absorption band for all six combination of both bands ranged from ~500 to 700
samples (figure 9D). FTIR analysis identified all of nm and produced yellow to orange fluorescence
them as type IaA diamonds with traces of isolated ni- when excited by visible light (near-LWUV). The band
trogen. Sample 18 contained much higher concentra- in the red part of the spectrum (630–655 nm) could
tions of aggregated nitrogen impurities than the other be excited independently of the ~539 nm band using
samples. PL spectra at 514 nm excitation showed a longer-wavelength light. The fluorescence from this

92 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
 480 nm BAND
A Emission at ~539 nm B
Sample #15 Emission at ~539 nm
Sample #18
FLUORESCENCE
Emission at ~653 nm

FLUORESCENCE
Emission at ~653 nm
(-0.01 – 9.999)

(-0.01 – 9.999)
5
5

0
50 0
0 50
0
EX
CI 700 EX
TA CIT 700
TIO 600 AT
N IO 600
500
(nm m) N( 500
nm)
22 400 N (n nm ON (
) 0
300 ISSIO )
22 400 SI
EMIS
0
EM 300

Emission at ~539 nm C D
Sample #19
FLUORESCENCE

Emission at ~653 nm
480 nm band

ABSORBANCE (offset)
(-0.01 – 9.999)

0
50
0
Sample 19
EX
CI 700
TA
TIO 600
N 500
(nm)
Sample 18
(nm 22 400 ION
) 0
300 EMISS
Sample 15

400 450 500 550 600 650 700

WAVELENGTH (nm)

Figure 9. These 3D fluorescence spectra collected on diamonds with 480 nm visible absorption bands as the major
defect center vary considerably. In the bottom right corner, UV-Vis-NIR spectra show differences in the shape and
intensity of the 480 nm bands. Excitation artifacts are marked in blue.

feature is likely the same as the PL broadband seen pinkish brown natural-color diamonds that showed
after excitation with a 514 nm laser. In general, the weak orange to red fluorescence under both long- and
band’s intensity increased with the size of the 480 short-wave UV lamps. Samples 23–25 were red and
nm absorption band. No clear correlations explain deep pink diamonds that had undergone combina-
the differences in the ~539 nm band. The two HPHT- tion treatments (HPHT + irradiation + annealing;
treated samples showed no obvious systematic dif- samples 23–24) or standard irradiation + annealing
ferences from the naturally colored samples. treatment (sample 25). The treated diamonds exhib-
ited red fluorescence under both LWUV and SWUV
N-V Center. The N-V center consists of a single ni- lights.
trogen atom adjacent to a vacancy in the diamond
lattice. The center can exist in an uncharged state, Absorption and PL Spectroscopy. The UV-Vis-NIR
(N-V)0 with ZPL at 575 nm, or in a negative charge spectra of samples 21 and 22 featured a broad 550 nm
state, (N-V)– with ZPL at 637 nm. N-V defects are band (thought to be caused by plastic deformation)
commonly associated with orange and red fluores- that contributed to their pink color, as well as the (N-
cence (Collins, 1982). V)– center at 637 nm. FTIR indicated type IIa stones
We examined five diamonds with N-V centers with no measureable nitrogen or hydrogen impuri-
(samples 21–25; figures 5 and 10): two naturally col- ties. PL spectra showed evidence for significant con-
ored, two multi-treated, and one irradiation-treated. centrations of (N-V)0, H3, and (N-V)– defects (figure
Samples 21 and 22 were brownish pink and very light 10D). For samples 23 and 24, UV-Vis spectra indicated

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 93
 N-V CENTER

Sample #22 A Sample #24 B

H3-related fluorescence
(H3)-related fluorescence
FLUORESCENCE

FLUORESCENCE
(N-V)˚-related fluorescence (N-V)¯-related
(N-V)¯
¯-related fluorescence

(-0.01 – 9.999)
(-0.01 – 9.999) (N-V)˚-related fluorescence
5 N3-related fluorescence 5
N3-related
fluorescence

0 0
50 50
0 0

EX 700
EX 700
CI CI
TA 600
TA 600
TIO TIO
500
N
(nm
500
(nm) N
(nm (nm)
22 400 SION 22 400 SION
EMIS EMIS
) 0 ) 0 300
300

Sample #25 C (N-V)˚ (N-V)¯ D

FLUORESCENCE

(N-V)¯-related
(N-V)¯
)¯-related fluorescence

PL INTENSITY (offset)
(-0.01 – 9.999)

(N-V)˚-related fluorescence
5

Sample 25
0
50
0

EX 700
CIT
AT 600
IO 500
N nm) Sample 22
(nm 22 400 SION (
EMIS
) 0 300
Sample 24

570 580 590 600 610 620 630 640 650

WAVELENGTH (nm)

Figure 10. These 3D fluorescence spectra collected on diamonds with N-V centers show combinations of several
defect centers. In the bottom right corner, PL spectra from corresponding samples confirm the presence of N-V cen-
ters. Excitation artifacts are marked in blue.

an abundance of (N-V)– centers; FTIR identified them peak that maximized at 470 nm excitation). Despite
as type IaAB with traces of isolated nitrogen. PL spec- its presence in PL spectra, fluorescence from (N-V)–
tra showed both (N-V)0 and (N-V)– centers as well as was not observed in these samples. Sample 21 also
H3 defects. For sample 25, UV-Vis spectra indicated displayed fluorescence of unknown origin at ~340–
the presence of a strong (N-V)– center and FTIR 360 nm. The treated samples (23–25) contained both
demonstrated a low concentration of type Ib isolated the (N-V)0 and (N-V)– centers as overlapping bands,
nitrogen defects. PL spectra showed both (N-V)0 and with the latter fluorescence band dominating at 660
(N-V)– centers, together with H3 centers. nm and extending beyond 750 nm (figures 10B and
10C). Sample 25 (figure 10C) featured only N-V de-
Fluorescence Spectroscopy. The 3D fluorescence fects in the fluorescence spectrum, consistent with
spectra for most of the N-V center samples are com- the occurrence of mainly isolated nitrogen impurities
plicated by multiple fluorescence-producing defects, (type Ib). Samples 23 and 24 contain aggregated ni-
sometimes as many as three. Both naturally colored trogen impurities and, as expected, showed fluores-
samples (21 and 22; figure 10A) showed the fluores- cence from N3 and H3 color centers.
cence emission feature of the (N-V)0 center, with ZPL
at 575 nm and an emission band extending from 575 DISCUSSION
to more than 700 nm with peaks at 586, 600, and 618 UV Light Sources. Emissions from the widely used
nm. We also saw clear fluorescence from two defects: mercury-based UV lamps (GIA and UVP Minera-
N3 (triplet peaks at 410–450 nm) and H3 (a 520 nm light lamps) do not contain pure 365 nm LWUV and

94 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
254 nm SWUV peaks, but instead show multiple ex- peaks resulting from mercury lamps and leaking fil-
citation energies of widely varying bandwidths. Sim- ters. Our 3D fluorescence spectra demonstrate that
ilarly, the popular LED UV loupe has a visible small changes in excitation wavelength, even from a
emission of 405 nm, rather than the standard for “pure” LWUV emission, can significantly affect the
LWUV (365 nm). The emissions from GIA’s LED flu- intensity and possibly the color of the fluorescence.
orescence meter and commercial LED light sources For example, the intensity of N3 luminescence (mea-
show a well-constrained LWUV emission in the sured at 439 nm) when excited by 400 nm excitation
364–367 nm range. The variability in excitation is approximately double that of the same emission
wavelengths and bandwidths among commonly measured at 360 nm excitation (figure 6). The mix-
used lamps and LED UV sources demonstrates that ture of the emission peaks from the UV lamp will af-
the colors and intensities of observed fluorescence fect the apparent color and intensity of the
in a single sample can vary depending on the light fluorescence, limiting the effectiveness and consis-
source. UV lamp emissions tend to fluctuate accord- tency of visual observation.
ing to the age of the bulbs and filters as well as To illustrate the potential variations introduced
warm-up times, making it very difficult to achieve by “extra” excitation energy, figure 11A shows the
consistent results from one source to another. LED fluorescence emission peaks from sample 15 (ex-
sources are far more constrained in emission band- tracted from the 3D fluorescence spectra of this
width and purity, but they are manufactured in such sample containing the 480 nm band) at 365, 405,
a wide range of wavelengths that consistency among and 435 nm excitation energies. These are the three
different products remains a problem. Nevertheless, emission peaks measured from the GIA LWUV
the well-controlled emissions suggest that LEDs can lamp (365 nm being the strongest). The spectra
provide better accuracy and consistency in interpret- were processed to produce equivalent colors by
ing the fluorescence response to a LWUV source, as conversion to equivalent CIE L*a*b* color values
long as an LED near 365 nm is used. using GRAMS AI software (including artifact re-
moval, 2° viewing angle, and CIE D65 illumination;
Fluorescence Colors from Different Defect Centers. all spectral intensities were scaled up 30× to facili-
3D fluorescence spectra provide an excellent oppor- tate color reproduction). The fluorescence peak
tunity to examine fluorescence produced by individ- generated by 365 nm excitation is centered at 520
ual defects as well as the contributions of multiple nm, producing a green color with a slightly yellow
defects. By directly depicting the relationship be- tint, whereas the fluorescence peaks caused by 405
tween UV excitation and fluorescence emission, and 435 nm excitation energies are shifted up to
these spectra illustrate the emission resulting from ~545 nm with a tail extending beyond 750 nm,
the various optical centers. Fluorescence spectra adding an orange-red component to the yellowish
from diamonds dominated by individual color cen- green fluorescence. The strong yellow fluorescence
ters are relatively simple and consistent, as discussed of this diamond under the conventional gemologi-
for each defect in the Results section. cal UV lamp is a mixture of the three emission col-
In most diamonds, multiple defect centers occur to- ors from these LW-UV excitation peaks. Under a
gether. Even under pure 365 nm LWUV excitation, the “pure” 365 nm LED source, this diamond shows a
fluorescence color will be a mixture of the fluorescence yellowish green fluorescence. While it would be
from different defect centers. The fluorescence may ideal to illustrate this effect with photos, we could
also have uneven zoning due to defect distributions. not achieve a long enough exposure to photograph
Visual observation alone is not an adequate basis for the fluorescence.
interpreting a diamond’s fluorescence color with re- These multiple emission peaks from the hand-
spect to the corresponding defect centers. The energy held UV lamp can also affect the intensity of the flu-
of the excitation source can have a profound effect on orescence color observed. Figure 11B shows the
the resulting fluorescence color and luminescence fluorescence emission peaks from sample 7 (ex-
spectra, and it must be carefully considered. tracted from the 3D fluorescence spectra of its N3
center) at the same 365, 405, and 435 nm excitation
Potential Effects of UV Light Sources on Fluores- energies used in the example above. Compared to
cence Color and Intensity. Analysis clearly shows 365 nm excitation, the fluorescence color with 405
that LWUV sources emit not only varying intensities nm excitation is almost twice as intense. The 435
and bandwidths of light at 365 nm, but also “extra” nm excitation gives a relatively small emission peak

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 95
 A. Sample 15 B. Sample 7

10 365 nm 10 365 nm
excitation

8 8

6 6

4 2nd order 4
excitation

2 2

0 0

10 405 nm 10 405 nm
FLUORESCENCE INTENSITY

8 8

6 6

4 4

2 2

0 0

10 435 nm 10 435 nm

8 8

6 6

4 4

2 2

0 0

300 400 500 600 700 300 400 500 600 700

WAVELENGTH (nm)

Figure 11. A comparison of diamond fluorescence color (samples 7 and 15) from a conventional UV lamp and a
single 365 nm excitation shows the effect of multiple emissions on the color and intensity of diamond fluores-
cence. Colors were reproduced from spectra by conversion to equivalent CIE L*a*b* values with CIE D55 illumi-
nation using Thermo GRAMS AI software. Excitation artifacts are marked in blue.

centered at ~520 nm due to the presence of the H3 Fluorescence as a Tool for Identifying Treatment. Al-
center, resulting in a weak green color. “Extra” emis- though we tested a limited number of samples, some
sion from the conventional UV lamp at 405 nm differences in the fluorescence behavior of natural
causes the diamond’s blue fluorescence to be more and treated diamonds were observed. In diamonds
intense than that from a “pure” 365 nm LED source. with H3 or 480 nm band defects, 3D fluorescence
Such discrepancies contribute to problems with con- spectra did not show significant differences between
sistency and reproducibility of fluorescence observa- the natural-color and treated samples. Yet treated di-
tions within the industry. amonds tended to show much more fluorescence

96 FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013
from the (N-V)– center, partially overlapping with the characteristics, a UV source with consistent output
band from the (N-V)0 center, whereas naturally col- energy is necessary. Traditional gemological UV
ored stones showed a dominant (N-V)0 defect (again, lamps and light sources used in the diamond indus-
see figure 10). Because the fluorescence from the (N- try produce widely variable emissions, making it
V)– center was much more intense in the treated di- difficult to achieve reproducible fluorescence obser-
amonds and shifted to longer wavelengths of >750 vations. LED light sources can provide narrow, well
nm, the fluorescence color from the treated dia- constrained UV emissions to produce more reliable
monds with both N-V centers was a more intense red diamond fluorescence reactions.
color than in the untreated samples with only the (N- This study characterizes the fluorescence gener-
V)0 center, which produced an orange color. While ated by the most common defect centers in dia-
this trend for N-V centers after HPHT treatment is mond. Changes in fluorescence relative to
well documented (e.g., Fisher and Spits, 2000) and excitation energy were carefully examined to deter-
also proved true in this study, gemologists should not mine where fluorescence maxima occur for each
rely solely on the observation of orange or red fluo- defect, as well as the peak positions and band
rescence to ascertain diamond treatment. shapes that can impact fluorescence colors and in-
tensities. Our findings demonstrate the diamond
CONCLUDING REMARKS industry’s need for a standardized UV excitation
Testing of diamonds with various primary defect source, to provide consistent and reproducible de-
centers has shown that to define their fluorescence terminations of fluorescence.

ABOUT THE AUTHORS ACKNOWLEDGMENTS

Dr. Luo ([Link]@[Link]) is supervisor of global metrology and The authors thank GIA’s Ronnie Geurts, Wuyi Wang, Tom Moses,
Dr. Breeding is a research scientist at GIA in Carlsbad. Jim Shigley, and Sally Eaton-Magaña, and GIA consultants Martin
Cooper and Chris Welbourn, for many helpful discussions. Assis-
tance from David Nelson, John Innis, and Dylan Hand in collection
of spectroscopic data at GIA is also greatly appreciated.

REFERENCES
Becquerel M.E. (1868) Sources de Lumière: Ses Causes et Ses Effets Holloway G. (2009) Blue fluorescence in diamond. The Australian
[Sources of Light: Its Causes and Effects]. Librairie de Firmin Gemmologist, Vol. 23, No. 9, pp. 408–414.
Didot Frères, Paris, 431 pp. Hoover D.B., Theisen A.F. (1993) Fluorescence excitation-emission
Breeding C.M., Shigley J.E. (2009) The “type” classification system spectra of chromium-containing gems: An explanation for the
of diamonds and its importance in gemology. G&G, Vol. 45, effectiveness of the crossed filter method. The Australian Gem-
No. 2, pp. 96–111, [Link] mologist, Vol. 18, No. 6, pp. 182–187.
Collins A.T. (1982) Colour centers in diamond. The Journal of Mani A. (1944) The fluorescence and absorption spectra of dia-
Gemmology, Vol. 18, No. 1, pp. 37–75. mond in the visible region. Proceedings of the Indian Academy
——— (1974) Visible luminescence from diamond. Industrial Di- of Sciences A, Vol. 19, pp. 231–252.
amond Review, Vol. 34, pp. 131–137. Moses T.M., Reinitz I.M., Johnson M.L., King J.M., Shigley J.E.
Cotty W.F. (1956) Identification of diamonds by their fluorescence. (1997) A contribution to understanding the effect of blue fluo-
The Journal of Gemmology, Vol. 5, No. 7, pp. 339–341. rescence on the appearance of diamonds. G&G, Vol. 33, No. 4,
Dyer H.B., Matthews I.G. (1958) The fluorescence of diamond. Pro- pp. 244–259, [Link]
ceedings of the Royal Society of London: Series A, Mathemat- Pearson G. (2011) Review of ultraviolet sources for gem fluores-
ical and Physical Sciences, Vol. 243, No. 1234, pp. 320–335, cence and testing. The Journal of Gemmology, Vol. 32, No. 5-
[Link] 8, pp. 211–222.
Eaton-Magaña S., Post J.E., Heaney P.J., Walters R.A., Breeding Shigley J.E., Breeding C.M. (2013) Optical defects in diamond: A
C.M., Butler J.E. (2007) Fluorescence spectra of colored dia- quick reference chart. G&G, Vol. 49, No. 2, pp. 107–111,
monds using a rapid, mobile spectrometer. G&G, Vol. 43, No. [Link] 107.
4, pp. 332–351, [Link] Shipley R.M. (1947) Notes on the abundance and color of fluores-
Fisher D., Spits R.A. (2000) Spectroscopic evidence of GE POL cent diamonds. G&G, Vol. 5, No. 9, pp. 395–397.
HPHT-treated natural type IIa diamonds. G&G, Vol. 36, No. 1, Wild G.O., Biegel H. (1947) The fluorescence of diamonds. The
pp. 42–49, [Link] Gemmologist, Vol. 16, No. 193, pp. 228–229.
Fritsch E., Waychunas G.A. (1994) Gemstones. In M. Robbins, Ed., Williams B. (2007) Technology update—ultraviolet light. Gem
Fluorescence: Gems and Minerals Under Ultraviolet Light, Market News, Vol. 26, No. 1, pp. 8–11.
Geoscience Press Inc., Phoenix, AZ.

FLUORESCENCE PRODUCED BY OPTICAL DEFECTS IN DIAMOND GEMS & GEMOLOGY SUMMER 2013 97
FEATURE ARTICLES

NEPHRITE JADE FROM VAL MALENCO,

ITALY: REVIEW AND UPDATE
Ilaria Adamo and Rosangela Bocchio

Alpe Mastabia, in the Val Malenco district of northern Italy, has been a source of nephrite jade since the
early 2000s. Twenty-one samples from this locality were investigated by classical gemological methods;
X-ray powder diffraction, combined with quantitative phase analysis; scanning electron microscopy in
combination with energy-dispersive spectrometry; electron microprobe analysis; mass spectrometry;
and mid-infrared spectroscopy. From a mineralogical standpoint, this jade consists mainly of tremolite
amphibole, with variable amounts of other constituents, especially calcite (up to approximately 30
wt.%), but also pyroxene, apatite, and sulfide minerals. Its pale green color is related to the low iron
content of the tremolite amphibole, whereas the other minerals are responsible for different colors (cal-
cite for white, molybdenite and galena for gray). On the basis of minor and trace-element composition,
we can classify this jade as dolomite-related nephrite (para-nephrite). Although new material could be
recovered from this area, future production will probably be limited by access difficulties.

N
ephrite jade is an almost monominer- discovery of the Alpe Mastabia nephrite as well as
alic rock, composed primarily of tremo- the geologic environment bear striking similarities
lite [Ca2Mg5Si8O22(OH)2] to actinolite to those reported by Dietrich and De Quervain (1968)
[Ca2(Mg,Fe)5Si8O22(OH)2] amphiboles (Leake et al., for the better-known nephrite deposit at Scortaseo
1997). Although tremolite-actinolite is considered (Val Poschiavo, Switzerland), situated less than 20
the predominant phase in nephrite jade, its specific km away.
weight percentage range is still debatable.
Major sources include the Kunlun Mountains in
Qinghai Province and the Xinjiang Uygur Au- Figure 1. Nephrite jade from Val Malenco is used in gems
and ornamental carvings, such as these bird statuettes
tonomous Region of China; the East Sayan Moun-
(approximately 15 cm high). Courtesy of Pietro Nana.
tains of Siberia; Chuncheon in South Korea; South
Westland in the South Island of New Zealand; and
Cowell, Australia (Harlow and Sorensen, 2005; Liu
et al., 2011a,b; Zhang et al., 2011).
A new deposit of gem-quality nephrite jade (figure
1) was discovered at the beginning of the 2000s at
Alpe Mastabia, located in the Val Malenco district in
the Sondrio province of northern Italy (figure 2). Mr.
Pietro Nana first noticed an attractive green stone in
the discarded waste materials of an abandoned talc
mine (figure 3) located at an altitude of 2,077 meters
(Nichol and Geiss, 2005). The events leading to the

See end of article for About the Authors and Acknowledgments.

GEMS & GEMOLOGY, Vol. 49, No. 2, pp. 98–106,
[Link]
© 2013 Gemological Institute of America

98 NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013
 Austria
using X-ray powder diffraction combined with quan-
Switzerland titative phase analysis based on the Rietveld method;
Alpe Mastabia, scanning electron microscopy in combination with
Val Malenco
energy-dispersive spectrometry (SEM-EDS); electron
microprobe analysis in wavelength dispersion mode
France (EMPA-WDS); mass spectrometry; and mid-infrared
Austroalpine nappes
Italy
ly (IR) spectroscopy.

Malenco Unit
GEOLOGIC SETTING
Val Malenco is an extremely interesting geological and
mineralogical district (Adamo et al., 2009) situated in
the Rhetic Alps near the Italian-Swiss border between
46˚ 15'N
the Southern Alps and the so-called “root zone” of the
Alpe Mastabia N Alpine nappes. The main geological unit is an ultra-
mafic body (the “Malenco unit”) that is one of the
0 5 km largest ophiolitic masses of the Alps (figure 2). It is ex-
posed in an area of about 130 km2 and consists of ser-
pentinized peridotite with minor relicts of lherzolite
Figure 2. Alpe Mastabia in Val Malenco, northern and harzburgite (Trommsdorff et al., 1993). The Alpe
Italy, has been producing gem-quality nephrite jade Mastabia talc mine is situated in a narrow tectonic
since the beginning of the 2000s. The inset shows
zone (Lanzada-Scermendone) at the southern margin
the area’s main geologic features. The Alpe
Mastabia talc/nephrite deposit is enclosed in
of the Malenco unit (figure 2). About 300 meters away
schists, gneiss, and marbles. is serpentinized rock, and the rocks around the talc
mine are orthogneisses and schists (see box A) of the
pre-Mesozoic crystalline basement, situated among
This study aims to provide a review and update Triassic white calcitic to dolomitic marbles. The ori-
of the nephrite jade from Val Malenco, by investigat- gin of the talc deposit, as well as of tremolite and
ing a suite of rough and cut gem-quality samples nephrite, is ascribed to metasomatic processes within

Figure 3. Left: A view of

the Alpe Mastabia talc
mine with the nephrite
deposit. Right: Pietro
Nana stands near the
entrance. Courtesy of
Pietro Nana and Vin-
cenzo De Michele.

NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013 99
 BOX A: GEOLOGICAL GLOSSARY

Nappe: A large body or sheet of rock that has shifted clinopyroxene and orthopyroxene. Harzburgite contains
far from its original position by thrust faulting during olivine (40–90%), orthopyroxene, and small amounts of
continental collisions. Also known as an allochton. clinopyroxene (<10%).

Ultramafic (or ultrabasic): Describing igneous and Orthogneiss: Rock derived from the metamorphism of
meta-igneous rocks with very low silica content (less igneous rocks. Orthogneiss is distinguished from para-
than 45% SiO2), generally greater than 18% MgO, high gneiss, which derives from sedimentary rocks.
FeO, and low potassium. Ultramafic rocks are usually
composed of greater than 90% mafic minerals, which Schist: Metamorphic rock having a foliated or plated
are dark and have high magnesium and iron contents. structure called schistosity, in which the flaky miner-
als (micas) are visible to the eye.
Peridotite: Coarse-grained ultramafic rock consisting
mainly of olivine (at least 40% by volume) and pyroxene Metasomatic: Referring to a process by which the
(ortho-/clino-). Peridotites are distinguished according to chemical composition of a rock or portion of rock is al-
their different mineralogical composition. Dunite con- tered in a pervasive manner. Metasomatism involves
tains more than 90% olivine. Wherlite is mostly com- the introduction or removal of chemical components
posed of olivine (40–90%), clinopyroxene, and a minor as a result of the rock’s interaction with aqueous solu-
amount of orthopyroxene (<10%). Lherzolite contains tions. During metasomatism, the rock remains in a
olivine (40–90%) and approximately equal amounts of solid state.

the dolomitic marbles during the Alpine metamor- ples underwent standard gemological testing to de-
phism (Montrasio, 1984; Nichol and Giess, 2005). termine their refractive index (RI), specific gravity
(SG), and ultraviolet (UV) fluorescence.
HISTORY AND PRODUCTION X-ray powder diffraction measurements were car-
During the years of operation at the Alpe Mastabia ried out on six rough specimens to determine the
talc mine, from 1964 to 1994, nephrite boulders asso- jade’s mineralogical composition. Data were col-
ciated with the talc were ignored or discarded as lected by means of a Panalytical X’Pert-PROMPD
waste (Andreis, 1970; De Michele et al., 2002; Nichol X’Celerator X-ray powder diffractometer, using
and Giess, 2005). The mine was abandoned, and en- CuKα radiation (λ=1.518 Å) at a beam voltage of 40
trances to its horizontal tunnels and galleries that
once led to the ore body have mostly caved in or been
barricaded (again, see figure 3), making the deposit dif- Figure 4. These cut nephrite jades from Val Malenco
ficult to access. In 1995, while examining the waste (3.33–25.34 ct) were investigated in this study. Photo
material outside the mine, Mr. Nana noticed some by Monica Odoli.
attractive nephrite boulders. Recognizing their gemo-
logical value, he consigned rough samples to lapidary
workshops in China and Idar-Oberstein, Germany.
The production and marketing of nephrite jade from
Alpe Mastabia started at the beginning of the new
millennium, with about 25 tons produced since the
discovery. Production still flourishes due to the high
quality of the finished jewelry pieces and other orna-
mental objects.

MATERIALS AND METHODS

We examined 21 samples from the Val Malenco de-
posit, consisting of 13 cut (3.33–25.34 ct; figure 4) and
eight rough (0.35–2.50 ct) specimens. All 13 cut sam-

100 NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013
kV and a current of 40 mA. X-ray powder diffraction ical analyses were performed using the electron mi-
patterns were collected over the 9–80° range of the croscope’s EDS system (ISIS 300 Oxford).
scattering angle 2θ, with steps of 0.02° 2θ and a count We also performed quantitative chemical analy-
time of 25 seconds per step. The phase identification ses in situ of the fibrous crystals constituting the four
was based on data from the PDF-2 database (Interna- rough nephrite samples previously analyzed with X-
tional Center for Diffraction Data, Newton Square, ray powder diffraction. We used a JEOL JXA-8200
Pennsylvania). Quantitative phase analysis (box B) electron microprobe in wavelength-dispersive mode
was performed with the Rietveld method using the (EMPA-WDS) under the following conditions: 15 kV
accelerating voltage, 15 nA beam current, and a
count time of 60 seconds on peaks and 30 seconds on
the background. The following standards were used:
In Brief natural grossular (for Si and Ca), anorthite (Al), fay-
• Nephrite jade is an almost monomineralic rock, com- alite (Fe), olivine (Mg), rhodonite (Mn), omphacite
posed mainly of tremolite to actinolite amphiboles. (Na), ilmenite (Ti), K-feldspar (K), and pure V and Cr
• The nephrite deposit at Alpe Mastabia, located in for those elements. The raw data were corrected for
northern Italy’s Val Malenco district, has been produc- matrix effects using a conventional ΦρZ routine in
ing nephrite jade since the beginning of the 2000s. the JEOL software package.
• The nephrite from Val Malenco is associated with The trace-element composition of the same fi-
dolomitic marbles (para-nephrite). brous crystals in four rough samples was determined
• This jade’s pale green color is due to the low concen- by laser ablation–inductively coupled plasma–mass
tration of iron, although other minerals (i.e., calcite, spectroscopy (LA-ICP-MS). The instrument con-
molybdenite, and galena) can also influence the color. sisted of a Quantel Brilliant 266 nm Nd:YAG laser
coupled to a Perkin Elmer DRCe quadrupole ICP-
MS. The spot size was 40 µm, using NIST SRM 610
glass as an external standard and Ca as an internal
GSAS software package (Larson and Von Dreele, standard, as analyzed by microprobe. Precision and
1994) to treat the experimental 2θ-profile. accuracy estimated on the basaltic glass standard
The microstructural features of six rough samples BCR2 were better than 10%.
were investigated using a Cambridge STEREOSCAN Additional information was derived from hydro-
360 scanning electron microscope (SEM), with an ac- gen isotope composition, obtained through multiple
celeration current of 15 kV. Semi-quantitative chem- analyses of a few milligrams of selected tremolite

BOX B: QUANTITATIVE PHASE ANALYSIS

Quantitative phase analysis (QPA) is used to deter-

mine the concentration of various phases present in a
mixture after the identity of every phase has been es-
where Y1obs is the observed and Y1calc is the calculated in-
tablished (qualitative phase analysis). Powder diffrac-
wi is the weight assigned to the i th data point.
tensity of a point i of the powder diffraction pattern, and
tion is a direct method to identify and quantify phases
on the basis of their unique crystal structures
The entire calculated powder diffraction pattern is
(Pecharsky and Zavalij, 2003; Dinnebier and Billinge,
based on simultaneously refined models of the crystal
2008).
structures, diffraction optics effects, instrumental factors,
Among existing QPA methods, the Rietveld (1969)
and other specimen features. Rietveld refinement of multi-
technique appears to be one of the fastest and most reli-
phase samples can generate a relatively accurate QPA, be-
able. The foundation of the Rietveld method is that the
cause the Rietveld scale factors determined during the re-
difference between the measured and calculated whole
finements for every phase in the mixture are proportional
powder diffraction profile should be close to zero by
to the weight of the corresponding phases (Pecharsky and
means of the following minimized function (Pecharsky
Zavalij, 2003; Dinnebier and Billinge, 2008).
and Zavalij, 2003):

NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013 101
 TABLE 1. Gemological properties of the 13 cut nephrite jade samples from Val Malenco, Italy, investigated in
this study.
A B C D E F G H I L M N O
Whitish Greenish White- Greenish Greenish Whitish Greenish White White Greenish Yellowish Yellowish Green
Color
green white green white white green white white green green
Diaphaneity Opaque Opaque Opaque Opaque Opaque Opaque Opaque Opaque Opaque Opaque Opaque Opaque Opaque
Refractive
1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60
indexa
Specific
2.95 2.89 2.89 2.89 2.96 2.90 2.86 2.77 2.74 2.85 2.96 2.96 2.96
gravity
Mohs
6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5 6.5
hardness
UV
Inert Inert Inert Inert Inert Inert Inert Inert Inert Inert Inert Inert Inert
fluorescence

a
Measured by the distant vision method

fibers by mass spectrometry following standard pro- X-ray Powder Diffraction Data. Quantitative phase
cedures (Vennemann and O’Neil, 1993). analysis based on XRPD data showed that five spec-
Mid-infrared absorption spectroscopy (4000–600 imens consisted mainly of tremolite amphibole, with
cm–1) was carried out on four rough specimens using various amounts of other minerals, especially calcite.
a Nicolet Nexus Fourier-transform infrared (FTIR) This mineral was generally less than 5 wt.%, though
spectrometer, operating in transmission mode with exceptional values of about 25–30 wt.% were found
KBr pellets, at a resolution of 4 cm–1 and 200 scans per in the two most whitish samples (table 2; figure 5).
sample.
Microstructure. The nephrite jade from Val Malenco
RESULTS showed a micro- to crypto-crystalline texture that
consisted of a dense intergrowth of fine (about 10–20
Gemological Properties. The gemological properties
of the 13 cut jades from Val Malenco are listed in μm long) randomly oriented bundles of tremolite
table [Link] color of both the polished and rough sam-
ples ranged from a common white and white-green Figure [Link] on Rietveld refinement of the X-ray
to a rarer yellowish green and green (figure 4). Two powder diffraction pattern (9–80° 2θ-angle range) of
rough samples had a blackish gray color (similar to the white nephrite sample 4C, the main constituent
the bracelet’s color in figure 11). The RI was constant, phases are about 70 wt.% tremolite and 30 wt.% cal-
while SG ranged from 2.74 to 2.96, with the lower cite (peak positions marked in black and red, respec-
values in the whitest samples. The SG values of the tively). The lower pattern represents the residual
two blackish gray rough samples were 2.80 and 2.86. between the calculated and experimental curves.
All the stones were opaque and inert to long- and
short-wave UV radiation. Irregular striped or spotted QUANTITATIVE PHASE ANALYSIS
color zoning was observed in nearly every sample.
1.5
INTENSITY (a.u.)

1.0

TABLE 2. Mineralogical composition (wt.%) of the

six rough nephrite jade samples from Val Malenco. 0.5

Sample 4S 6A 6B 13 12 4C 0.0

Color Green Green White- White- White White

green green 0.5
Tremolite 100 99 97 95 76 70 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0
Calcite 0 1 3 5 24 30 2θ (degrees)

102 NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013
Figure 6. Nephrite jade from Val Malenco is characterized by a finely fibrous texture, as shown in these SEM photos.

fibers (figure 6). Other minerals were identified by samples, whereas the other minerals were rarer. In
SEM-EDS as calcite, diopside, apatite, and opaque particular, molybdenite and galena were more concen-
minerals (molybdenum, iron, lead, and zinc sulfides; trated in the blackish gray nephrite samples. These
see figure 7). Calcite was widespread in almost all minerals were distributed unevenly in nephrite.

Chemical Composition. The chemical composition

Figure 7. SEM photos show the mineralogical compo- of the fibrous crystals in the jade is reported for four
sition of the blackish gray (top) and white specimens samples in table 3.
(bottom). Tremolite is labeled T, calcite C, galena G, Chemical analyses of the fibrous mineral showed
and molybdenite M.
a composition close to that of tremolite amphibole,
Ca2Mg5Si8O22(OH)2, according to the classification of
Leake et al. (1997), with a low concentration of most
trace elements. All elements of the first transition se-
ries (Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, and Zn), as well
as the alkaline earth metals (Sr, Ba), and alkaline met-
als (K, Rb, Cs) were always less than 0.1 wt.%, with
the lone exception of Na, ranging from 0.14 to 0.20
wt.% as Na2O. Among the chromophore ions, iron
was the most abundant, with contents ranging from
527 to 670 ppm (as Fe), followed by vanadium at
about 200 ppm (as V); chromium was much lower,
with concentration ≤4 ppm (as Cr; table 3).
In terms of stable isotopic composition, the

Figure 8. The IR spectrum of the white nephrite sam-

ple 4C shows absorption bands of tremolite (labeled
T) and calcite (C).

IR SPECTRUM
1.0 1000 949
0.9 T T
949
0.8 1067
ABSORBANCE

1108 T
T
0.7 T
876
1423 C
0.6 C
0.5
758
0.4 1799 C
3674 C 686
0.3 T 2513 T
C
0.3 712
C
0.1

3000 2000 1000

WAVENUMBERS (cm-1)

NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013 103
 CALCITE CONTENT TABLE 3. Average chemical composition of tremolite
amphibole in the nephrite jade from Val Malenco.
30

Sample 4S 6A 6B 4C
CALCITE (wt. %)

Electron microprobe analysesa

20 Oxides (wt.%)
SiO 2 59.41 59.87 59.60 58.98
TiO 2 0.01 0.02 0.02 0.02
10
Al O 2 3 0.26 0.04 0.29 0.48
FeO 0.03 bdlb 0.01 0.08
MnO 0.01 0.01 0.01 0.02
0
MgO 23.14 23.01 23.12 22.89
Green Green White- White- White White
green green CaO 13.22 13.16 13.12 13.42
SAMPLES Na O 2 0.14 0.15 0.18 0.20
KO 0.04 0.04 0.05 0.04
2
________ ________ ________ ________
Figure 9. Calcite contents obtained by X-ray powder Total 96.26 96.31 96.39 96.12
diffraction analyses (see table 2) are shown for six
Fe/(Fe+Mg) 0.001 0.000 0.000 0.002
nephrite jade samples.
Ions per 23 oxygens
Si 8.127 8.176 8.137 8.093
nephrite samples had a hydrogen isotopic mean Ti 0.001 0.002 0.002 0.002
value (δD) of –113± 4.8‰. Al 0.041 0.007 0.046 0.077
Fe 0.003 bdl 0.001 0.009
Mid-Infrared Spectroscopy. The samples’ mid-IR ab- Mn 0.001 0.001 0.001 0.002
sorption spectra, collected in transmission mode Mg 4.719 4.684 4.705 4.681
with KBr pellets (figure 8), were characterized by Ca 1.937 1.926 1.919 1.972
Na 0.038 0.039 0.047 0.053
bands at 3674, 1108, 1067, 1000, 949, 922, 758, and
K 0.007 0.007 0.008 0.008
688 cm–1, typical of tremolite (Hawthorne and Della
Ventura, 2007). The whitest samples (12 and 4C) also LA-ICP-MS analysesc
Element (ppm)
contained the absorption features of calcite (2513, Sc 1.24 1.42 1.38 1.16
1799, 1423, 876, and 712 cm–1; Farmer, 1974). Ti 21.0 12.2 18.7 19.2
V 215 271 293 220
DISCUSSION Cr 2.10 2.47 4.20 2.20
The various samples from Val Malenco exhibited RI Mn 81.2 80.7 84.5 87.8
Fe 584 527 572 670
values typical for nephrite jade (O’Donoghue, 2006).
Co 0.19 0.19 0.29 0.19
Their SG values, which showed variation with the
Ni 0.46 0.57 0.81 0.66
color of the stones, were in agreement with those com- Zn 52.7 42.0 51.5 38.9
monly reported for this jade (SG=2.9–3.1; O’Donoghue, Sr 6.88 7.21 6.46 5.34
2006), except for the lower values in the two white Y 0.72 0.68 0.93 0.23
samples (2.77 and 2.74, respectively; see H and I, table Zr 0.14 0.18 0.39 0.66
1). These SG values are compatible with a significant a
Ten points per sample were analyzed. FeO is the total iron. Cr is below
amount of calcite, whose SG (2.70 ± 0.01) is lower than detection limit in all samples.
that of tremolite. The Mohs hardness of 6.5 and the b
Abbreviation: bdl = below detection limit (0.01 wt.%).
c
finely felted fibrous texture are typical of nephrite from Five points per sample were analyzed.

localities worldwide (O’Donoghue, 2006).

The fine fibers of tremolite amphibole occurred to-
gether with other unevenly distributed constituents. iron as the most abundant chromophore ion. Yet its
Calcite was the most abundant, although its con- content shows comparable values in the various sam-
tent varied from 0 to about 30 wt.%; pyroxene, ap- ple colors from white to green (again, see figure 9). This
atite, and sulfide minerals were rarer. The variable suggests that the occurrence of other mineral phases
amount of calcite corresponded with a white hue (e.g., calcite) could influence the jade’s color variation.
(figure 9), while molybdenite and galena were re- Nephrite’s minor and trace-element chemical
sponsible for the gray and black hues. composition plays an important role in tracing its ge-
Chemical analysis of the tremolite fibers revealed ologic origin. It is used to distinguish nephrite asso-
a low abundance of minor and trace elements, with ciated with dolomitic marbles (dolomite-related

104 NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013
 CR VS. CO AND NI
Serpentinite-Related Serpentinite-Related
Nephrite Nephrite

50

40 1000
Figure 10. Chromium
versus cobalt and
nickel plots of the Ital-
Co 30 Ni Dolomite-Related ian nephrite jade (green
Nephrite
triangles) are compared
20 500 with nephrite data
nephrite from from Siqin et al. (2012).
Val Malenco, Italy Dolomite-Related nephrite from
Nephrite Val Malenco, Italy
10

0 0
0 10 20 30 40 50 0 10 20 30 40 50

Cr Cr

nephrite or para-nephrite) from that associated with and the enclosing schists (Nichol and Giess, 2005).
serpentinite rock (serpentinite-related nephrite or The very low δD-depleted value determined on this
ortho-nephrite); see, for instance, Nichol (2000) and nephrite also suggests a possible contamination of the
Siqin et al. (2012). In particular, various authors (e.g., hydrothermal fluid with water of meteoric origin (Yui
Siqin et al.) have shown that serpentine- and and Kwon, 2002; Harlow and Sorensen, 2005).
dolomite-related nephrite minerals can be separated
on the basis of their Fe/(Fe+Mg) ratio (0.064–0.118 and CONCLUSIONS
0.001–0.074, respectively) and their concentrations of Nephrite jade from Val Malenco, northern Italy, en-
Cr, Co, and Ni, which are higher in nephrite associ- tered the market at the beginning of the 2000s. Since
ated with serpentinite. Based on the very low the discovery of nephrite there in 1995, the deposit
Fe/(Fe+Mg) ratio (always <0.002), as well as the low has produced about 25 tons of gem-quality rough,
contents of Co (0.19–0.29 ppm), Cr (2.10–4.20 ppm),
and Ni (0.46–0.81 ppm) found in the nephrite jade
from Val Malenco (table 3 and figure 10), we can clas- Figure 11. These necklaces (top) and bracelet (bot-
sify this material as dolomite-related nephrite or para- tom) consisting of approximately 10 mm diameter
spheres are fashioned from nephrite jade from Val
nephrite.
Malenco. Photos by Pietro Nana.
The possible formation process for this type of
nephrite deposit is commonly ascribed to a metaso-
matic reaction involving dolomite replacement by sili-
cic fluids (Harlow and Sorensen, 2005). This is
consistent with the reaction proposed by Nichol and
Giess (2005) for nephrite from Val Malenco, which
produces excess calcite as isolated grains or aggregates:
5CaMg(CO3)2 + 8SiO2 + H2O = Ca2Mg5Si8O22(OH)2 + 3CaCO3
(dolomite) (silica) (tremolite) (calcite)

Yet the occurrence of diopside grains as rare relicts

within tremolite also suggests a second stage of for-
mation involving the intermediate formation of diop-
side from dolomite, subsequently replaced by
tremolite. In any case, the formation of the Alpe
Mastabia deposit in Val Malenco implies an intense
fluid-rock reaction by hydrothermal solutions, perco-
lating at the contact between the dolomitic marbles

NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013 105
carved into ornamental objects or fashioned into jew- color: Calcite is responsible for the rather common
elry (figure 11). This study, performed on both cut and white hue, while molybdenite and galena cause the
rough specimens, shows that nephrite jade from Val black and gray colors. The nephrite’s compact mi-
Malenco is composed of tremolite amphibole. It con- crostructure and soft color make it an attractive gem
tains a low concentration of iron as a chromophore, material. Although new finds could be limited by ac-
in agreement with its pale green color. Variable cess difficulties, the geologic features of the deposit
amounts of other constituents also influence the suggest significant potential for further production.

ABOUT THE AUTHORS Dr. Alberto Zanetti (CNR-IGG, Pavia) for assistance with LA-ICP-
Dr. Adamo ([Link]@[Link]) is junior researcher, and Dr. Boc- MS measurements; Dr. Valeria Diella and Mr. Agostino Rizzi
chio is professor of mineralogy, in the Earth Sciences department (CNR-IDPA, Milan) for SEM-EDS analyses; and Dr. Monica Dapi-
of the Università degli Studi di Milano, Italy. aggi (Università degli Studi di Milano) for collaboration during the
X-ray powder diffraction data collection and analysis. The au-
ACKNOWLEDGMENTS thors are also indebted to Dr. Luigi Dallai (CNR-IGG, Stable Iso-
The authors thank Mr. Pietro Nana (Sondrio, Italy) for providing topes Laboratory, Pisa) for hydrogen isotope measurement. Dr.
samples and useful information about the material investigated in Loredana Prosperi (Istituto Gemmologico Italiano, Milan) and
this study. We also thank Mr. Andrea Risplendente (Università Mrs. Nerina Saggiomo (Milan) are acknowledged for collabora-
degli Studi di Milano) for assistance with microprobe analyses; tion during this research.

REFERENCES
Adamo I., Bocchio R., Diella V., Pavese A., Vignola P., Prosperi L., [Link]
Palanza V. (2009) Demantoid from Val Malenco, Italy: Review Liu Y., Deng J., Shi G., Yui T-F., Zhang G., Abuduwayiti M.,
and update. G&G, Vol. 45, No. 4, pp. 280–287, [Link] Yang L., Sun X. (2011b) Geochemistry and petrology of
10.5741/GEMS.45.4.280. nephrite from Alamas, Xinjiang, China. Journal of Asian
Andreis C. (1970) Il talco di Mastabia in Val Malenco (Prov. di Son- Earth Sciences, Vol. 42, No. 3, pp. 440–451, [Link]
drio). Master’s thesis, Università degli Studi di Milano (Italy). 10.1016/[Link].2011.05.012.
De Michele V., Liborio G., Nana P., Piacenza B. (2002) Nefrite della Montrasio A. (1984) The “Lanzada-Scermendone Zone”: An ophi-
Val Malenco (Provincia Sondrio, Lombardia, Italia). IVM Mag- olitic unit of continental affinity in the southern Rhaetic Alps
azine, Bollettino dell’Istituto di Mineralogia “F. Grazioli,” Vol. (Prov. Sondrio-Italy). Schweizerische Mineralogische und Pet-
1, pp. 13–15. rographische Mitteilungen, Vol. 64, No. 1/2, pp.111–129.
Dinnebier R.E., Billinge S.J.L., Eds. (2008) Powder Diffraction: The- Nichol D. (2000) Two contrasting nephrite jade types. Journal of
ory and Practice. Royal Society of Chemistry, Cambridge, UK. Gemmology, Vol. 27, No. 4, pp. 193–200.
Dietrich V., De Quervain F. (1968) Die Nephrit-Talklagerstätte Nichol D., Giess H. (2005) Nephrite jade from Mastabia in Val Ma-
Scortaseo (Puschlav, Kanton Graubünden) [The Scortaseo lenco, Italy. Journal of Gemmology, Vol. 29, No. 5/6, pp. 305–
nephrite-talc deposit (Poschiavo, Grisons Canton)]. Beiträge zur 311.
Geologie der Schweiz, Geotecnische Serie, Vol. 46. O’Donoghue M., Ed. (2006) Gems, 6th ed. Butterworth-Heine-
Farmer V.C. (1974) The Infrared Spectra of Minerals. Mineralogical mann, Oxford, UK.
Society, London. Pecharsky V. K., Zavalij P.Y. (2003) Fundamental of Powder Dif-
Harlow G.E., Sorensen S.S. (2005) Jade (nephrite and fraction and Structural Characterization of Materials. Kluver
jadeitite) and serpentinite: metasomatic connections. Academic Publishers, USA.
International Geology Review, Vol. 47, No. 2, pp. 113–146, Rietveld H.M. (1969) A profile refinement method for nuclear and mag-
[Link] netic structures. Journal of Applied Crystallography, Vol. 2, No. 2,
Hawthorne F.C., Della Ventura G. (2007) Short-range order in am- pp. 65–71, [Link]
phiboles. In Hawthorne F.C., Oberti R., Della Ventura G., Mot- Siqin B., Qjan R., Zhuo S., Gan F., Dong M., Hua Y. (2012) Glow dis-
tana A., Eds., Amphiboles: Crystal Chemistry, Occurrence and charge mass spectrometry studies on nephrite minerals formed
Health Issues, Reviews in Mineralogy and Geochemistry, Vol. by different metallogenic mechanisms and geological environ-
67, pp. 173–222. ments. International Journal of Mass Spectrometry, Vol. 309, No.
Larson A.C., Von Dreele R.B. (1994) General structure analysis sys- 1, pp. 206–211, [Link]
tem (GSAS). Los Alamos National Laboratory Report, LAUR Trommsdorff V., Piccardo G.B., Montrasio A. (1993) From magma-
86-748. tism through metamorphism to sea floor emplacement of sub-
Leake B.E., Woolley A.R., Arps C.E.S., Birch W.D., Gilbert M.C., continental Adria lithosphere during pre-Alpine rifting
Grice J.D., Hawthorne F.C., Kato A., Kisch H.J., Krivovichev (Malenco, Italy). Schweizerische Mineralogische und Petro-
V.G., Linthouut K., Laird J., Mandarino J.A., Maresch W.V., graphische Mitteilungen, Vol. 73, No. 2, pp. 191–203.
Nickel E.H., Rock N.M.S., Schumacher J.C., Smith D.C., Vennemann T.W., O’Neil J.R. (1993) A simple and inexpensive
Stephenson N.C.N., Ungaretti L., Whittaker E.J.W., Youzhi G. method of hydrogen isotope and water analyses of minerals and
(1997) Nomenclature of amphiboles: Report of the Subcommit- rocks based on zinc reagent. Chemical Geology, Vol. 103, No. 1-
tee on Amphiboles of the International Mineralogical Associ- 4, pp. 227–234, [Link]
ation, Commission on New Minerals and Mineral Names. Yui T.-F., Kwon S.-T. (2002) Origin of a dolomite-related jade de-
American Mineralogist, Vol. 82, No. 9/10, pp. 1019–1037, posit at Chuncheon, Korea. Economic Geology, Vol. 97, No. 3,
[Link] pp. 593–601, [Link]
Liu Y., Deng J., Shi G., Sun X, Yang L. (2011a) Geochemistry and Zhang B., Lu T., Chen H., Ke J. (2011) Research and identification
petrogenesis of placer nephrite from Hetian, Xinjiang, North- of natural and treated nephrite in China. G&G, Vol. 47, No. 2,
west China. Ore Geology Review, Vol. 41, No. 1, pp. 122–132, pp. 122.

106 NEPHRITE JADE FROM VAL MALENCO, ITALY GEMS & GEMOLOGY SUMMER 2013
NOTES & NEW TECHNIQUES

OPTICAL DEFECTS IN DIAMOND:

A QUICK REFERENCE CHART
James E. Shigley and Christopher M. Breeding

diamond to absorb particular energies of incident

Gem diamonds owe much of their value to their light or radiation. Defects can occur randomly or in
color, or lack thereof. Defects in the atomic struc- particular locations within the lattice. Diamonds can
ture of diamond are responsible for this color and contain more than one type of defect, and in natural
are important for the identification of color treat- diamonds, defects can be altered over geologic time
ments. This article and its tables are intended as in the earth or by exposure to heat or radiation during
a quick reference for gemologists as they read color treatment.
about various common diamond defects in the So-called optical defects (or optical centers) cause
gemological literature. absorption in the visible or near-visible portions of the
electromagnetic spectrum, often producing coloration
(e.g., figure 1). Luminescence reactions result when
defects absorb higher-energy incident radiation and
then reemit lower-energy radiation as visible light.
D espite the commercial value of natural-color di-
amonds, distinguishing them from treated dia-
monds remains a significant identification challenge.
Optical defects occur in very low concentrations in
all diamonds, and their presence can be detected using
While some diagnostic visual features exist (inclu- spectroscopic techniques. A theoretically “pure and
sions, color or growth zoning, and absorption bands perfect” diamond containing no such defects would
seen with a spectroscope), the separation of natural appear colorless.
from synthetic or treated diamonds is not always
possible using standard gemological methods. In
Figure 1. The red color of the graining in this Fancy red
such cases, advanced spectroscopic analysis at a pro- diamond from Brazil is caused by absorption related
fessional gem-testing laboratory is required. Imaging to the 550 nm band. This band, the most common
of luminescence distribution patterns is also a help- cause of pink to red color in natural, untreated dia-
ful tool for recognizing synthetic diamonds (Mar- monds, is thought to be the result of a defect created
tineau et al., 2004; Shigley et al., 2004). by plastic deformation. Photomicrograph by Jian Xin
In a laboratory setting, the identification of dia- (Jae) Liao; magnified 50×.
monds is based mainly on the detection of tiny im-
perfections in the atomic lattice. These “defect
centers” may include foreign impurity atoms (typi-
cally nitrogen, and occasionally boron or hydrogen);
carbon atom vacancies in the lattice (either single or
clusters of neighboring vacancies); carbon atoms po-
sitioned in between normal lattice locations (intersti-
tials); and dislocations where planes of carbon atoms
are offset from one another due to plastic deforma-
tion. Not all of these lattice imperfections create spec-
troscopic features, but several do so by allowing the

See end of article for About the Authors.

GEMS & GEMOLOGY, Vol. 49, No. 2, pp. 107–111,
[Link]
© 2013 Gemological Institute of America

NOTES & NEW TECHNIQUES GEMS & GEMOLOGY SUMMER 2013 107
TABLE 1. Important optical defects in diamond and their effect on color and luminescence.
LWUV DiamondView
Optical defect and spectroscopic
Color fluorescence luminescence
means of detection (~365 nm lamp) (<220 nm source)

ND1: A defect with an absorption line at 393.6 nm (3.150 eV).

No effect Thought to consist of a vacancy in the negative charge state (V–). No effect No effect
Produced by natural or artificial irradiation. UV
N3: An impurity and intrinsic defect with an absorption line at
415.2 nm (2.985 eV) and associated bands. Thought to consist of
three substitutional nitrogen atoms surrounding a vacancy (3N+V).
Often occurs with the associated N2 (477.2 nm) and several other
related bands (465, 452, 435, and 423 nm) in “Cape” diamond
spectra. All are naturally occurring. UV, PL, EPR
480 band: A naturally occurring optical defect of uncertain struc-
ture (sometimes attributed to substitutional oxygen) in type I
diamonds with a broad absorption band centered at 480 nm
(2.580 eV). UV

H4: An impurity and intrinsic defect with an absorption line at

496.2 nm (2.498 eV). Thought to consist of four substitutional
nitrogen atoms surrounding two vacancies (4N+2V). Occurs
naturally or can be produced by irradiation followed by annealing.
UV, PL

H3: An impurity and intrinsic defect with an absorption line at

503.2 nm (2.463 eV) and associated bands. Thought to consist of
two substitutional nitrogen atoms separated by a vacancy in a
neutral charge state (N-V-N)0. Occurs naturally or can be produced
by irradiation followed by annealing or by high-pressure, high-
temperature annealing. UV, PL

May contribute 3H: A defect with an absorption line at 503.4 nm (2.462 eV).
Thought to be related to interstitial carbon (I). Produced by natural No effect No effect
to a green color
or artificial irradiation. UV, PL
550 band: An optical center of uncertain structure with a broad
absorption band centered at 550 nm (2.250 eV). Thought to result
from plastic deformation of the lattice structure. Occurs naturally.
UV No effect No effect

NV0: An impurity and intrinsic defect with an absorption line at

575 nm (2.156 eV) and associated bands. Thought to consist of a
single substitutional nitrogen atom associated with a vacancy in a
neutral charge state (NV0). Occurs naturally or can be produced by
irradiation followed by annealing. PL, EPR, UV

595 band: An optical defect of uncertain structure with an absorption

May contribute band at 594.4 nm (2.086 eV). Thought to be related to nitrogen.
No effect No effect
to other colors Occurs naturally or can be produced by irradiation followed by
annealing. UV
NV – : An impurity and intrinsic defect with an absorption line at
637 nm (1.945 eV) and associated bands. Thought to consist of a
single substitutional nitrogen atom associated with a vacancy in a
negative charge state (NV –). Occurs naturally or can be produced
by irradiation followed by annealing or by high-pressure, high-
temperature annealing. PL, EPR, UV

GR1: A defect with a pair of absorption lines at 740.9 nm (1.673 eV)

and at 744.4 nm (1.665 eV) and associated bands. Thought to
consist of a vacancy in a neutral charge state (V 0). Produced by
natural or artificial irradiation. UV, PL No effect No effect

108 NOTES & NEW TECHNIQUES GEMS & GEMOLOGY SUMMER 2013
 Color LWUV DiamondView
Optical defect and spectroscopic fluorescence luminescence
means of detection (~365 nm lamp) (<220 nm source)

H2: An impurity and intrinsic defect with an absorption line at

986.3 nm (1.256 eV, 10125 cm –1) and associated bands. Thought
May contribute to consist of two substitutional nitrogen atoms separated by a
No effect No effect
to a green color vacancy in a negative charge state (N-V-N) –. Occurs naturally or
can be produced by irradiation followed by annealing or by high-
pressure, high-temperature annealing. IR, PL
H1c: An impurity and intrinsic defect of uncertain structure with
an infrared absorption line at 1934 nm (0.6408 eV, 5171 cm –1).
No effect No effect No effect
Thought to be associated with nitrogen B centers. Occurs naturally
or can be produced by irradiation followed by annealing. IR
H1b: An impurity and intrinsic defect of uncertain structure with
an infrared absorption line at 2024 nm (0.612 eV, 4941 cm –1).
No effect No effect No effect
Thought to be associated with nitrogen A centers. Occurs naturally
or can be produced by irradiation followed by annealing. IR
Hydrogen: Defect(s) of uncertain structure with many related
infrared absorption lines, most notably at 3107 cm –1 (0.385 eV).
Occurs naturally. This defect can also produce yellow and violet No effect No effect
colors. IR, UV

Boron: A defect with a primary infrared absorption line at 2803 Phosphorescence Phosphorescence
cm –1 (0.348 eV) and associated lines, and a band extending into
the visible region. Thought to consist of single substitutional boron
atoms. Occurs naturally. This defect produces red phosphores-
cence. IR

A center: A defect with an infrared absorption band at 1282 cm –1

(0.159 eV). Thought to consist of two adjacent substitutional
nitrogen atoms (N-N). Occurs naturally. IR Quenches Quenches
luminescence luminescence

B center: A defect with an infrared absorption band at 1175 cm –1

(0.146 eV). Thought to consist of four adjacent substitutional nitrogen
atoms surrounding a vacancy (4N+V). Occurs naturally. IR No effect No effect

C center: A defect with an infrared absorption band at 1130 cm –1

(0.140 eV). Thought to consist of a single substitutional nitrogen
atom (N). Occurs naturally or can be produced by high-pressure, No effect No effect
high-temperature annealing of diamonds containing A or B
centers. IR, UV, EPR

H1a: An impurity and intrinsic defect of uncertain structure with

an infrared absorption line at 1450 cm –1 (0.180 eV). Thought to be
No effect No effect No effect
associated with interstitial nitrogen. Occurs naturally or can be
produced by irradiation followed by annealing. IR
Platelet: An impurity and intrinsic defect of uncertain structure
with an infrared absorption line at about 1360 cm –1 (0.169 eV).
No effect No effect No effect
Thought to be associated with groups of interstitial carbon atoms.
Occurs naturally. IR
Vacancy cluster: A defect of uncertain structure with increasing
absorption toward the blue end of the spectrum. Thought to consist
of groups of vacancies. Occurs naturally. UV No effect No effect

UV = Ultraviolet-visible absorption spectroscopy PL = Photoluminescence spectroscopy

IR = Infrared absorption spectroscopy EPR = Electron paramagnetic resonance spectroscopy

NOTES & NEW TECHNIQUES GEMS & GEMOLOGY SUMMER 2013 109
TABLE 2. Spectroscopic techniques for characterizing lattice defects in diamond.

Technique Type of Typical scan Commonly Test Advantages Disadvantages Comments

spectroscopy range detected features conditions
and defects

Ultraviolet/ Absorption 250–800 nm ND1, N3, N2, ~77 K Evaluation of color- Large samples Staple technique in
visible/near-infrared (transmission) H3, H4, 595 temperature causing defects; absorb too much gemological
(UV-Vis-NIR) nm, and GR1 relatively light, causing laboratories; can be
inexpensive detector saturation; performed with
difficulty quantifying scanning or
results from faceted dispersive detector
stones because of
the uncertain path
length of light travel
Mid- to near-infrared Absorption 400–11,000 A and B Room Determination of Large samples Staple technique in
(IR) (transmission) cm–1 aggregate temperature diamond type; absorb too much gemological
centers, C relatively in- light, causing laboratories
centers, various expensive; defect detector saturation
hydrogen-related concentrations can
defects, H1a, be quantified by
H1b, H1c, H2, normalization
and “amber
centers”
Raman Luminescence 100–2000 Diamond Room Identification of Expensive; yields Typically used for
cm–1 Raman Raman line temperature diamond; analysis little information diamond vs. non-
shift (1332 cm-1) of internal strain regarding treatment; diamond identifica-
difficulty analyzing tion (provides little
strongly fluorescent other information)
samples
Photoluminescence Luminescence 350–1000 nm N3, 490.7 nm, ~77 K Detection of Expensive; difficulty Lasers of various
(PL) H4, H3, NV0, temperature HPHT treatment; analyzing strongly wavelengths used
NV–, GR1, H- or lower characterization of fluorescent samples; for excitation (most
and Ni-related low-concentration variety of laser commonly 325,
defects defects; small excitations required 488, 514, 532, and
analysis area allows to activate various 633 nm); required
for detailed defects; requires for effective
investigation carefully controlled treatment detection
cryogenic test
temperatures
Cathodoluminescence Luminescence 400–700 nm A band, B band, Room Evaluation of Requires an electron Rarely used, limited
(CL) N3, H3, and H4 temperature internal structure; beam and a vacuum; data obtainable
detection of defects provides little
causing information about
luminescence color treatment
Electron- Resonance C centers, NV Room Detection of very Very expensive; Typically used for
spin/paramagnetic absorption in defects temperature low-concentration long sample run defect research
resonance (ESR, EPR) a changing defects; evaluation times; provides little
magnetic of specific defect information about
field structures and color treatment
charges

As a quick reference for gemologists, table 1 lists Some precautions should be kept in mind when
the most common lattice defects, including those using this table. Each lattice defect is known to pro-
that can create color and/or luminescence reactions duce a particular diamond color or luminescence. For
in diamond. The scientific name of the defect, as well instance, the GR1 center produces green or blue-
as the wavelength (or wavenumber) and electron volt green color. But the reverse is not necessarily true—
(eV) positions of the main and associated spectral for example, not all green diamonds owe their color
bands, are shown along with photos of representative to the GR1 center. In fact, there are several causes of
diamonds. The most common spectroscopic tech- green color. In some cases, a diamond contains more
nique used to detect the defect center is indicated by than one optical defect, and its color stems from a
a code shown in italics. combination of defects. In other cases, a diamond’s

110 NOTES & NEW TECHNIQUES GEMS & GEMOLOGY SUMMER 2013
color may arise from absorption caused by one opti- terization. For further reading on color in diamond,
cal center, while its luminescence may result from the Additional Reading list serves as a reference
another optical center. guide to access the much larger range of gemological
Table 2 provides a comparison of the most com- and technical literature on diamond identification.
mon spectroscopic techniques for diamond charac-

ADDITIONAL READING
Collins A.T. (2001) The colour of diamond and how it can be Ed., The Nature of Diamonds. Cambridge University Press in
changed. Journal of Gemmology, Vol. 27, No. 6, pp. 341– association with the American Museum of Natural History,
359. New York, pp. 23–47.
——— (2003) The detection of colour-enhanced and synthetic gem Hofer S.C. (1998) Collecting and Classifying Coloured Diamonds:
diamonds by optical spectroscopy. Diamond and Related Mate- An Illustrated Study of the Aurora Collection. Ashland Press,
rials, Vol. 12, No. 10/11, pp. 1976–1983, [Link] New York.
S0925-9635(03)00262-0. Martineau P.M., Lawson S.C., Taylor A.J., Quinn S.J., Evans D.J.F.,
Cunningham D. (2011) The Diamond Compendium. NAG Press, Crowder M.J. (2004) Identification of synthetic diamond grown
Robert Hale Ltd., London. using chemical vapor deposition (CVD). G&G, Vol. 40, No. 1,
Deljanin B., Simic D., Zaitsev A., Chapman J., Dobrinets I., Wide- pp. 2–25, [Link]
mann A., del Re N., Middleton T., Deljanin E., De Stefano A. (2008) Shigley J.E., Breeding C.M., Shen A.H. (2004) An updated chart
Characterization of pink diamonds of different origin: Natural (Ar- on the characteristics of HPHT-grown synthetic diamonds.
gyle, non-Argyle), irradiated and annealed, treated with multi- G&G, Vol. 40, No. 4, pp. 303–313, [Link]
process, coated and synthetic. Diamond and Related Materials, GEMS.40.4.303.
Vol. 17, No. 7/10, pp. 1169–1178, [Link] Tappert R., Tappert M.C. (2011) Diamonds in Nature: A Guide to
mond.2008.03.014. Rough Diamonds. Springer, New York, pp. 45–68.
Eaton-Magaña S., Post J.E., Heaney P.J., Walters R.A., Breeding Zaitsev A.M. (2000) Vibronic spectra of impurity-related optical
C.M., Butler J.E. (2007) Fluorescence spectra of colored dia- centers in diamond. Physical Review B, Vol. 61, No.19, pp.
monds using a rapid, mobile spectrometer. G&G, Vol. 43, No. 12909–12922, [Link]
4, pp. 332–351, [Link] ——— (2001) Lattice Defects in Diamonds: A Data Handbook.
Fritsch E. (1998) The nature of color in diamond. In G.E. Harlow, Springer-Verlag, Berlin.

ABOUT THE AUTHORS

Dr. Shigley (jshigley@[Link]) is a distinguished research fellow,
and Dr. Breeding is a research scientist, at GIA’s laboratory in
Carlsbad, California.

NOTES & NEW TECHNIQUES GEMS & GEMOLOGY SUMMER 2013 111
 GIA appreciates gifts to its permanent
collection, as well as gemstones, library materials,
and other non-cash assets to be used in education
and research activities. These contributions help
GIA further its public service mission while
offering donors philanthropic benefits. We
extend sincere thanks to all 2012 contributors.

CIRCLE OF HONOR*
$100,000 and higher
The Aaron Group PierLuigi Dalla Rovere Kazanjian Bros., Inc. Art Sexauer
Dr. Suman Agrawal The De Beers Group KCB Natural Pearls Shades of the Earth
(K.C. Bell) (Laura and Wayne Thompson)
Almaza Jewelers Debbie and Mark Ebert
(Ziad H. Noshie) William F. and Jeanne H. Larson Ambaji Shinde
Fabricjewelry
American Pearl Company Stephen Lentz S.H. Silver Company
Dallas R. Hales
(Stephen and Eileen Silver)
Amsterdam Sauer Sophie Leu
Dr. H. Tracy Hall
Dr. Geoffrey A. Smith
Robert and Marlene Anderson Honoring Betty H. Llewellyn
Dr. Gary R. and Barbara E.
D. Swarovski & Co.
Aurafin-Oroamerica Hansen Marshall and Janella Martin
Touraine Family Trust
Banks International Gemology Inc. James Y. Hung, M.D. Roz and Gene Meieran
(Daniel and Bo Banks) United States Pearl Co.
Inta Gems Inc.
Nancy B. & Co. (James and Venetia Peach)
The Bell Group-Rio Grande
J.O. Crystal Company Inc.
Dr. Kurt Nassau Robert H. Vanderkay
Allan Caplan (Judith Osmer)
John & Laura Ramsey Vicenza Fair
Chatham Created Gems Inc. Jewel America Inc.
(Thomas H. Chatham) (Zvi and Rachel Wertheimer) R. Ed Romack

2012 D0NORS*
Kaufman Enterprises (Mark Robert Sielecki
Kaufman)
$50,000 to $99,999 $2,500 to $4,999
Steven and Betty Lou Neely In Memory of Larry Crandall In Memory of Arthur Metzger
Terri Ottaway (Elise A. Skalwold and William A.
In Memory of Harry Nest
Bassett of Cornell University)
Hussain Rezayee
Kathryn Kimmel
$10,000 to $49,999
David & Gale Kempf
In Memory of Rudolf Loewenthal
$500 to $999
Cos Altobelli Renée Newman In Memory of Bobby R. Prade
In Memory of Dr. V. Premanda Nalini Pattni Kessler Mfg. & Imports Inc.
Jerry Bearman
William "Flint" Carter Aucoin Hart Jewelers Jeweler Profit Inc.
Dudley Blauwet
Cynthia Renée Inc. Dr. Jaroslav Hyršl Fabrice Danet
Mark Mauthner
Brendan M. Laurs
$1,000 to $2,499 Ginger Bartlett Hughes
Aurora Gems
Under $500
Jewelry Design by Ruth Fitzgerald Revere Academy of Jewelry Arts
$5,000 to $9,999
G. Scott Davies
Bill Barker Suzanne Harned Sea Hunt Pearls
House of Onyx
Alexandre Reza Marco Campos-Venuti David S. Bunevich
New Era Gems
(Steve Ulatowski) Rattanachart Mining Co., Ltd. Throwin' Stones/Intuitively John Catron
Inspired Designs
Joanna Seetoo Designs Oceanview Mines, LLC Cosmos Gems (Leonardo Silva
(Joanna Joy Seetoo) (Jeff Swanger) Mawingu Gems Souto)
Courtland Lee Gems TV R.T. Boyd Limited
In Memory of
Dr. Frederick H. Pough Stephen Lewis Solomon Rajneesh Bhandari [Link]
J.D., G.G. German Salazar (Mauro Pantò)
Albert & Louise Garcia

* All are cumulative donations

If you are interested in making a donation and receiving tax benefits information, please contact:
MCKENZIE SANTIMER
call: (760) 603-4150
fax: (760) 603-4056
email: [Link]@[Link]
 Editors
Thomas M. Moses | Shane F. McClure

Atypical Freshwater CULTURED plasma–mass spectrometry (ICP-MS) Finally, photographs of the X-ray
PEARLS with Damaged Nacre confirmed that both the outer and fluorescence reactions of all four
Recently the East Coast laboratory re- inner nacre layers originated from a pearls were taken in a dark environ-
ceived four baroque-shaped pearls freshwater environment. Some of ment (figure 3). Various intensities of
ranging in size from 9.07 × 13.50 × these pearls contained widely varying green and yellowish green fluores-
14.03 mm to 9.32 × 11.34 × 15.77 mm amounts of trace elements (i.e., Mn cence could be observed within each
and weighing between 10.63 and 13.60 and Mg) between the outer and inner nacreous layer, presumably due to
ct. On each of the four samples, the layers, but previous studies have varying concentrations of manganese
outermost surface nacre layers exhib- found that uneven distribution of between layers and among the pearls
ited large holes and gaps, revealing one trace elements can also occur in a themselves (H.A. Hanni et al., “X-ray
or more distinct nacreous layers un- pearl’s cross section (R. Lu et al., “Op- luminescence, a valuable test in pearl
derneath (figure 1). These underlying erational considerations of EDXRF, identification,” 2005 Journal of Gem-
nacreous layers displayed good luster LA-ICP-MS, and photoluminescence mology, 29, 5/6, pp. 325–329). More
and overlapping platelet surface struc- techniques in the analysis of pearls,” interestingly, reddish fluorescence
tures, and the large gaps between the Summer 2011 G&G, pp. 149–150). was also observed around the dam-
inner and outermost nacre layers indi- Two did not show any distinct in- aged areas of the nacre in samples 3
cated that smaller pearls had been used ternal demarcation when examined and 4. This was possibly caused by
as nuclei during the culturing process. with X-radiography in various direc- foreign material between the “nu-
Further gemological testing by en- tions (figure 2). This lack of a clear in- cleus-pearl” and the outer nacre lay-
ergy-dispersive X-ray fluorescence ternal boundary may be due to multiple ers, which may also contribute to the
(EDXRF) and inductively coupled culturing attempts using the same pearl lack of distinct demarcation observed
as a nucleus, which can cause over- in the cultured structure.
lapped internal structures and multiple While similar types of pearls with
Editors’ note: All items were written by staff distinct nacreous layers under the sur- intact outer nacre surfaces have been
members of GIA laboratories. face (figure 1, sample 3). In addition, reported on previously (E. Strack, “Chi-
structures typical of non-bead cultured nese freshwater cultured pearls beaded
GEMS & GEMOLOGY, Vol. 49, No. 2, pp. 113–118, pearls were observed in these micro X- with baroque freshwater cultured
[Link]
radiographs, suggesting that freshwater pearls,” Fall 2011 G&G, pp. 244–245),
cultured pearls were used as nuclei dur- the ambiguous internal structures, in-
© 2013 Gemological Institute of America
ing the culturing process. teresting X-ray fluorescence properties,

Figure 1. These four atypical freshwater cultured pearls with damaged nacre reveal one or more distinct nacreous
layers underneath. Left to right: samples 1–4.

LAB NOTES GEMS & GEMOLOGY SUMMER 2013 113

Figure 2. X-radiographs show partial demarcation between the outer and inner structures of cultured pearls 1 and
2, but fail to reveal any clear boundaries in samples 3 and 4.

and large gaps in the outer nacre layers ranged from 0.42 to 0.50 ct and had a that reminds us of a particular shape,
of these four samples offer a rare view Fancy pink to Fancy Intense pink an effect known as pareidolia (Winter
of their internal physical structure and color. Figure 4 also shows the patchy 2007 Lab Notes, pp. 363–364), is a geo-
appearance, obtained free of destruc- appearance characteristic of coated di- logic wonder.
tive testing. amonds. Microscopic examination The New York laboratory recently
Chunhui Zhou using reflected light revealed obvious received a natural yellow (graded as
coating on the pavilion facets, with Fancy yellow) “Cape” 0.51 ct rectangu-
patchy colorless areas where it had lar faceted diamond exhibiting such a
been removed (figure 5). feature. Examination of the stone re-
DIAMOND
The UV-visible spectra of all 13 di- vealed a series of table-reaching frac-
Calcium Fluoride Coating amonds exhibited a broad band cen- tures and clouds that seemed to form
Found On 13 “Fancy” Pinks tered at about 520 nm, not at 550 nm the outline and scale-like detail of a fish
Coating gemstones to alter or im- as expected. This band is responsible (figure 6). The inclusion is large in rela-
prove their color is the oldest known for the pink coloration, and the 520 tion to the stone, encompassing nearly
enhancement method. Early coatings, nm position identified the coating as 75% of the table.
particularly on diamond, were easily fluoride.
detected, often visible using a 10× This example demonstrates that
loupe. But advances made after World traditional treatments such as coating Figure 4. These face-up images of
War II resulted in more durable coat- are still prevalent in the trade, and 0.45 ct and 0.5 ct pink round bril-
ings and films that were subtler and great care must be taken to identify liants are representative of the 13
harder to detect. them. Because coatings are not per- coated pink diamonds submitted.
A common coating was calcium manent, all 13 diamonds were issued
fluoride (CaF2) doped with gold (Au) identification reports disclosing the
nanoparticles. CaF2 causes a broad ab- treatment.
sorption band at about 520 nm, simi-
lar to the band at about 550 nm that
largely accounts for the pink color of
natural diamonds.
Recently submitted to the East Fish-Like Inclusion
Coast laboratory for identification and It is always a pleasant surprise to find
grading was a group of 13 pink dia- a new and interesting inclusion during
monds (figure 4). The round brilliants routine observation. But an inclusion

Figure 3. X-ray fluorescence images of the four cultured pearls showing

various degrees of green or yellowish green fluorescence between inner
and outer layers. Atypical orangy red fluorescence is also visible between
the layers in samples 3 and 4.

114 LAB NOTES GEMS & GEMOLOGY SUMMER 2013

Figure 5. Photomicrographs of the diamonds from figure 4, taken with re-
flected light at 80× and 100× magnification, reveal the coating applied to
the pavilion facets.

Figure 7. This 24.16 ct oval-cut dia-

Nearly all fractures in diamond (figure 7). The color was distributed mond (20.47 × 14.00 × 10.72 mm)
result from the interaction among evenly throughout the stone, with no was color graded as Fancy Intense
multiple cleavage directions (The color zoning detected. It showed green-yellow, and identified as ar-
MicroWorld of Diamonds, Gemworld strong yellow and blue fluorescence to tificially irradiated and annealed.
International, 2000, p. 97). While often long-wave UV radiation, and moderate
detrimental to the value of a stone, yellow fluorescence to short-wave UV
this inclusion serves as a useful iden- radiation. This fluorescence reaction sorption or H1b/H1c center in the
tifier, and such findings help keep the and absence of chalkiness are seen in near-infrared region was recorded.
study of gems enjoyable. some natural diamonds with a similar Absorption spectrum of this diamond
Martha Altobelli bodycolor. Absorption spectrum in the in the ultraviolet/visible/near-infrared
mid-infrared region showed saturated (UV-Vis-NIR) region at liquid-nitrogen
absorptions in the one-phonon region, temperature (figure 8) showed a mod-
indicating a type Ia diamond with a erately strong absorption from the N3
A Large Irradiated Green-Yellow
high concentration of nitrogen. Also (415.2 nm) and H4 (496.0 nm) centers,
Artificially irradiated diamonds are
recorded were a moderate platelet as well as weak absorptions at 512.8,
common in the trade, and identifying
peak at 1368 cm–1 (approximately 2.1 594.4, and 741.2 nm (GR1). Absence of
them remains a challenge. The New
cm–1 intensity) and a weak absorption the H3 optical center was confirmed
York lab recently identified a large di-
at 1433 cm–1. No hydrogen-related ab- by photoluminescence spectroscopy at
amond as artificially irradiated based
on careful spectroscopic analysis and
Figure 8. After irradiation and annealing treatment, a significant amount
a review of the lab’s database.
of optical center H4 was introduced. This is mainly responsible for the
This 24.16 ct oval-cut diamond
color improvement from Fancy brownish greenish yellow to Fancy In-
(20.47 × 14.00 × 10.72 mm) was color
tense green-yellow. N3 absorptions in two spectra in this plot were nor-
graded as Fancy Intense green-yellow
malized to the same intensity.

Figure 6. This pavilion view of UV-VIS-NIR SPECTRA

the 4.76 × 4.30 × 2.75 mm Fancy
yellow diamond displays the .7 N3
“fish” inclusion scene. The fish
appears to be facing to the right. .6
ABSORBANCE

Magnified 40×.
.5

.4 H4

.3 512.8
741.2
594.4 After treatment (GR1)
.2

.1 Before treatment

450 500 550 600 650 700 750

WAVELENGTH (nm)

LAB NOTES GEMS & GEMOLOGY SUMMER 2013 115

liquid-nitrogen temperature with 488 The diamond’s unusual fluores-
nm laser excitation. A fluorescence cence under both long- and short-
image collected using the Diamond- wave UV illumination can be seen in
View showed a gradual change from a figures 10 and 11, respectively. Under
predominantly blue table region to a long-wave UV (handheld lamp) the
mostly green upper pavilion and girdle near-colorless section showed typical
area. The spectral features observed in blue fluorescence. The orangy brown
this large diamond are rarely encoun- section displayed a yellowish fluores-
tered in natural diamonds, raising the cence, which is common in type IaA
possibility of artificial irradiation yellow diamonds. Under the higher- Figure 10. Under long-wave UV il-
treatment. energy, shorter-wavelength UV exci- lumination at 10× magnification,
A search of the GIA database found tation of the DiamondView, the the diamond’s color zones showed
a very good match between this stone distinction between the colors became blue and yellowish fluorescence.
and a previously tested diamond. The even more pronounced, and the blue
two diamonds were identical in meas- fluorescence of the near-colorless of B-aggregated nitrogen impurities.
urements and weight, as well as their in- bands was clearly visible within the The unique zoning observed in this
frared absorption spectroscopic features, yellowish green fluorescence of the diamond likely reflects changing con-
though the previous grade was Fancy orangy brown section. Yellow phos- ditions during growth. Variations in ni-
brownish greenish yellow. This signifi- phorescence, common in type IaA di- trogen and possibly oxygen content in
cant improvement in color appearance amonds colored by 480 nm visible the earth’s interior—480 nm defects
is attributed to the relatively high con- absorption bands, was also observed in are reportedly related to oxygen impu-
centration of the H4 center produced by the orangy brown area. rities in diamond—could have pro-
the artificial irradiation/annealing treat- Infrared spectroscopy revealed a duced the noteworthy color and
ment confirmed in this study. type Ia diamond with both A and B ag- fluorescence zoning that reveal the
Despite its suspicious spectroscopic gregates. When IR spectra were col- complex growth history.
features, this irradiated diamond lected separately from the
showed some gemological features nor- near-colorless and orangy brown sec- Troy Ardon
mally observed in diamonds with sim- tions, distinct differences were noted.
ilar natural color. This example The orangy brown section contained
demonstrates the value of careful spec- ~33 ppm of A-aggregated nitrogen and Very Large Rough Diamond
tral and gemological analysis, combined ~15 ppm of B-aggregated nitrogen, Throughout history, the discovery of
with a review of the GIA database, in whereas the near-colorless section very large diamond rough has been a
gem identification. showed significantly higher concen- rare occurrence. Not only is it un-
Wuyi Wang, Paul Johnson, and trations: approximately 139 ppm of A likely to survive the journey from
Emiko Yazawa aggregates and 100 ppm of B aggre- deep in the earth to the surface, but
gates. In addition to the higher overall modern extraction methods, includ-
nitrogen content, the near-colorless ing the use of crushers to break large
portion contained a higher proportion rock, can reduce such specimens to
Strong Color Zoning Reflects
Complex Growth Environment
A 2.08 ct emerald-cut diamond re-
cently submitted to the Carlsbad lab- Figure 9. The strong color zoning Figure 11. This DiamondView
oratory received a color grade of Fancy in this Fancy Dark orangy image of the color-zoned diamond
Dark orangy brown. Its strong color brown diamond was visible at shows an even more pronounced
zoning, visible without magnification 10× magnification. distinction between the colors.
(figure 9), consisted of a dark orangy
brown section with a few near-color-
less bands and a near-colorless area.
The orangy brown section showed nu-
merous very small oriented reflective
inclusions, visible in figure 9 only as
dark clouds, that are typically associ-
ated with type IaA diamonds colored
by 480 nm visible absorption bands.
Both the near-colorless side and the
colorless bands displayed very few ori-
ented reflective inclusions.

116 LAB NOTES GEMS & GEMOLOGY SUMMER 2013

smaller pieces. Therefore, the recent various laser excitations. The main spectrometry (LA-ICP-MS) was per-
examination of a 1,138 ct rough dia- emission features were broad bands at formed at the GIA lab in Bangkok.
mond (62.51 × 47.61 × 45.56 mm; fig- 700 and 787 nm. The crystal also dis- The chemical data showed a fairly
ure 12) provided a unique opportunity played a very weak blue fluorescence even composition within the sample
to document the largest single-crystal under long- and short-wave ultravio- except for chromium, which varied
natural diamond ever examined by let radiation, another typical feature. proportionately with the green color
GIA. While it is possible there have The irregular morphology of the zoning (from approximately 300 to
been larger unreported specimens, a 1,138 ct rough was quite interesting, as 9000 ppmw Cr). The chemical data of
search of public sources (e.g., I. Bal- diamond usually occurs in octahedral or the unusual emerald were compared
four, Famous Diamonds, 5th ed., An- dodecahedral forms. But such a mor- to those of reference standards from
tique Collectors Club, 2008) suggests phology is less likely to be maintained known deposits, particularly Panjsher
this could be the second-largest rough as a large crystal forms. Numerous ran- and the Swat Valley, to investigate its
diamond in history—between the domly shaped and distributed inclusions probable origin. The data showed a
3,106 ct Cullinan and the 995 ct Ex- may have contributed to the irregular close match only with the reference
celsior, both of gem quality. crystal growth. These inclusions also emeralds from the Swat Valley, mak-
Known to be from the Democratic caused the gray color appearance. ing that the most likely origin.
Republic of the Congo, this stone dis- The specimen’s strong surface dis- The gemological properties showed
played Raman spectroscopic features solution provided a visual document an RI of 1.589–1.597 and a visible spec-
characteristic of diamond. The sheer of its journey from deep in the earth. trum consistent with emerald. The
size presented a challenge with IR ab- At the high temperatures in the stone showed no reaction to a standard
sorption spectroscopy analysis, as it earth’s interior, diamond can remain 4-watt gemological UV light source.
was too large for the regular sample stable only under very high pressure. Microscopic examination revealed nu-
chamber, beam condenser, or DRIFT As the diamond was transported to merous inclusions of pyrite, carbon-
units. Instead, we used an infrared mi- the surface, the decrease in pressure ates, and reflective thin films oriented
croscope with reflective mode and ob- would have dissolved the stone unless perpendicular to the optic axis (parallel
tained a high-quality absorption it made the journey rapidly. The sur- to the length of the stone), a typical as-
spectrum. The rough was identified as face of the 1,138 ct crystal speaks to semblage of a natural emerald. The
a natural type IaA diamond with very the change that took place during its most unusual observation about this
high nitrogen concentration. We also journey and represents the fine line stone was a prominent green “S” shape
detected a weak hydrogen-related ab- between our experiencing this rare color zone that ran down its length
sorption peak at 3107 cm–1, a com- item and never having known it at all. (figure 13). Emerald’s hexagonal struc-
mon feature in natural diamond. Also John M. King and Wuyi Wang ture would ordinarily lead one to ex-
recorded were strong absorption pect planar or angular color zoning
bands at approximately 3300, 2920, related to the crystal morphology. The
1700, and 1550 cm–1, attributed to mi- Unusual Curved Color Zoning presence of curving zones in a faceted
cron and sub-micron inclusions. Pho- In EMERALD emerald was an anomaly that needed
toluminescence spectra were recorded further examination.
While curved color zoning is typically
at liquid-nitrogen temperature with One possible explanation for the
associated with melt-grown synthet-
ics, particularly flame-fusion products
such as synthetic corundum and syn-
Figure 12. This 1,138 ct single- thetic spinel, an unusual natural Figure 13. This 1.17 ct emerald,
crystal natural diamond (62.51 × emerald with this feature was re- which hosted pyrite and carbon-
47.61 × 45.56 mm) is the largest cently examined in the Carlsbad and ate crystals, displayed unusual
ever examined by GIA. Bangkok laboratories. This 1.17 ct curved zoning.
emerald was obtained for GIA’s per-
manent reference collection (sample
number 100305160993) at a street
market in Kabul, Afghanistan, for its
very unusual inclusion scene. The
stone was represented as being of
Afghan (Panjsher) origin, but the in-
ternal inclusions seemed to resemble
those of emeralds collected from the
Swat Valley of Pakistan. Careful
chemical analysis by laser ablation–
inductively coupled plasma–mass

LAB NOTES GEMS & GEMOLOGY SUMMER 2013 117

unusual zoning is that compressional was not a practical approach. While
geological forces caused shearing along the exact cause of this unusual zoning
the emerald crystal’s basal plane. This remains unknown, it is certainly in-
was supported by the appearance of teresting and thought [Link]
pyrite grains, which looked as if single, discovery of such gems demonstrates
brittle pyrite crystals had been sheared that there are complex geological con-
and slightly displaced parallel to the ditions we do not yet understand
basal plane of the beryl host (figure 14). which, when observed, invite further
This apparent lateral shearing move- gemological exploration. Figure 16. The orangy red neptu-
ment throughout the emerald crystal Nathan Renfro, Vincent Pardieu, nite contained dense needles of
could have caused the pyrite grains to and Supharart Sangsawong unknown composition. Field of
separate into aligned tabular fragments view: 1.72 mm.
offset from each other and the beryl
host. Also observed between the pyrite
Rare Faceted NEPTUNITE
grains were planar optical irregulari- Gemstones, Vol. 1, ABC Edition,
An 11.78 ct faceted “black” octagonal
ties. These planes suggested localized Zurich, 1986). Both minerals are Ti-
modified step-cut stone was recently
shear zones that would have signifi- bearing silicates. Benitoite may con-
presented to the Carlsbad laboratory
cantly higher defect concentration, tain whitish inclusions of crossite
for identification (figure 15). Standard
giving rise to the optical nonconfor- (an amphibole) or natrolite (a zeo-
gemological testing revealed an RI of
mity consistent with their location be- lite), so it is possible that the thin
1.69–1.73 and a hydrostatic SG of
tween the tabular pyrite inclusions. needles in this neptunite are com-
3.19. Microscopic examination with a
Under cross-polarized illumination, posed of one of these metamorphic
fiber-optic light source showed the
we observed dense clusters of birefrin- minerals. The main source of neptu-
stone was actually a very dark orangy
gent crystals that were essentially in- nite is San Benito, California, where
red. It was heavily included with frac-
visible under non-polarized lighting. it is found along the walls of natro-
tures and platelets, as well as ran-
The low relief of these dense inclu- lite veins in blueschist (B. Laurs et
domly oriented unidentified thin
sions suggested they were beryl inclu- al., “Benitoite from the New Idria
needles evenly distributed throughout
sions within the beryl host, possibly a District, San Benito County, Califor-
the gem (figure 16). These tests sug-
result of partial recrystallization nia,” Fall 1997 G&G, pp. 166–187).
gested neptunite, an identity con-
caused by dynamic environmental The type locality for this material is
firmed using Raman spectroscopy.
conditions. the Narsarsuk pegmatite in Green-
Neptunite commonly occurs in
Gemology is generally a non-de- land (O.V. Petersen and O. Johnsen,
association with, and as a guest in-
structive, observational science, which “Mineral species first described
clusion in, the rare blue mineral
can be limiting at times. Destructive from Greenland,” Canadian Miner-
benitoite (E.J. Gübelin and J.I.
techniques could have revealed the alogist Special Publication No. 8,
Koivula, Photoatlas of Inclusions in
crystallographic orientation of the 2005, pp. 76–77). Neptunite crystals
pyrite grains with respect to each have been found in sizes up to a few
other and explained the stone’s un- Figure 15. This unusual 11.78 ct inches, but they are rarely faceted
usual structure. Because this was a faceted neptunite appeared to be because of their brittle nature and
unique sample, destructive testing black but was actually a dark relative softness (5–6 on the Mohs
orangy red. scale). The few crystals that are
faceted usually weigh less than 1 ct,
Figure 14. Arrays of tabular offset making this large specimen of par-
pyrite fragments were seen ticular interest. To the best of our
throughout this unusual emerald. knowledge, this is the first faceted
Field of view: 0.86 mm. neptunite identified by GIA.
Amy Cooper and Tara Allen

PHOTO CREDITS:
Chunhui Zhou—1, 2, 3; Jian Xin (Jae) Liao—
4, 7; Paul Johnson—5; Martha Altobelli—6;
Troy Ardon—9, 10, 11; Joshua Balduf—12;
Sasithorn Engniwat—13; Nathan Renfro—14,
16; Robison McMurtry—15

118 LAB NOTES GEMS & GEMOLOGY SUMMER 2013

 Contributing Editors
Emmanuel Fritsch, CNRS, Team 6502, Institut des Matériaux Jean Rouxel (IMN), University of Nantes, France (fritsch@[Link])
Franck Notari, GGTL GemLab–GemTechLab, Geneva, Switzerland ([Link]@[Link])
Kenneth Scarratt, GIA, Bangkok, Thailand ([Link]@[Link])

COLORED STONES AND ORGANIC MATERIALS

Ancient tourmaline and beryl from Afghanistan. A large
private collection of beads and carvings from antique ex-
cavations, assembled in the 1970s by a collector in
Afghanistan, was recently examined. The pieces were
claimed to be found near the ancient city of Bactra (mod-
ern-day Balkh), 300 km northwest of Kabul near the border
with Uzbekistan. Bactra was a center of the Bactrian civi-
lization, which flourished from 2500 to 1500 Bc. The col-
lection contains several thousand drilled beads, mostly
agate, carnelian, and rare blue chalcedony. The next most
abundant materials are turquoise—some as small as 1 mm
in diameter—and lapis lazuli. Less common are garnet (two
types of almandine), rock crystal, amethyst, serpentinite,
and steatite, as well as organic materials identified as
amber, pearl, and mother-of-pearl. Other very interesting
finds are treated rock crystal and milky quartz, both con-
taining blue glass spots on their surfaces, similar to me-
Figure 1: This 9.52 ct pink tourmaline is part of an ex-
dieval sapphire imitations (Fall 2001 GNI, pp. 243–245).
tensive private collection of ornaments claimed to be
The two most interesting pieces, a pink bird (figure 1)
from ancient Afghanistan. Photo by Jaroslav Hyršl.
and a very pale blue bead (figure 2), were studied in detail.
The bird had a light green bottom section, measured 13.9
× 13.6 × 7.6 mm, and weighed 9.52 ct, with a refractive
index close to 1.63. It was strongly pleochroic (pink and stone was weakly pleochroic (very light blue and colorless),
colorless), and it was uniaxial negative in a conoscope. In inert in UV light, and uniaxial negative in a conoscope.
both short- and long-wave UV light the center was inert, These results indicated a very pale aquamarine. The iden-
but a thin triangular zone of light blue fluorescence was tification of both beads was confirmed by their Raman pat-
visible near the surface. These tests confirmed the stone’s terns, collected by A. Gilg (Technical University of
identification as tourmaline. It was originally a “water- Munich).
melon” tourmaline, pink with a green rim, but only a Both beads are possibly among the oldest of their kind.
small part of the green rim had been preserved. Large and historically significant rubellites have been re-
The pale blue bead was set in a heavily corroded metal, ported from the 14th and 16th centuries, and a single beryl
probably bronze. The entire piece measured 20.7 × 12.3 × bead from Nubia has been dated back to Predynastic time
6.7 mm and weighed 20.28 ct, with an RI close to 1.59. The (before 3200 Bc), but few details are known. The first beryl
locality of significance was reportedly the Egyptian emerald
Editors’ note: Interested contributors should send information and illustra-
deposit exploited as early as the Ptolemaic era (after 332 Bc).
tions to Justin Hunter at [Link]@[Link] or GIA, The Robert Both tourmaline and aquamarine are typical pegmatite
Mouawad Campus, 5345 Armada Drive, Carlsbad, CA 92008. minerals. Gem-bearing pegmatites have been mined since
the 1970s in the Nuristan province of eastern Afghanistan
GEMS & GEMOLOGY, VOL. 49, NO. 2, pp. 120–124,
[Link] 2.120 and northern Pakistan—the possible origin of both beads.
© 2013 Gemological Institute of America Jaroslav Hyršl (hyrsl@[Link])
Prague, Czech Republic

120 GEM NEWS INTERNATIONAL GEMS & GEMOLOGY SUMMER 2013

Figure 2: This aquamarine bead, reported to be from Figure 4. Under magnification, distinct color con-
Afghanistan, weighs 20.28 ct and is likely one of the centrations are visible within the pores and cavities
oldest of its kind. Photo by Jaroslav Hyršl. of this “soft coral.” The apparent use of a coloring
agent only on the surface, evidenced by the enamel-
like appearance within the trenches and the white-
Orangy pink coated “soft coral.” Dyeing white coral to im- ness of the ridges, suggests these beads were coated,
itate popular colors such as red, orange, and pink is a not dyed. Photomicrograph by Gagan Choudhary;
widely known practice. But a bead strand examined at the magnified 48×.
Gem Testing Laboratory in Jaipur, India, revealed a color-
ing method unfamiliar to gemologists (figure 3). producing an enamel-like appearance in the trenches while
The strand weighed approximately 380 ct and contained the ridges appeared white. This evidence suggested that the
spherical beads ranging from 6 to 15 mm in diameter. These beads were coated with paint rather than dyed.
were identified as coral belonging to the species Melithaea Although the microscopic features were sufficient for
ocracea of the order Alcyonacea (e.g., M.c. Pederson, Gem identification, further tests confirmed those findings.
and Ornamental Materials of Organic Origin, Elsevier But- Under UV light, the beads displayed bright orange fluores-
terworth-Heinemann, Oxford, 2004, pp. 192–218). Their sur- cence (more strongly under short-wave), a feature com-
face showed deep pores and dull luster. The presence of monly associated with pink-orange dyes. No typical
pores was sufficient to identify the beads as soft coral, which absorption spectra were visible under a desk-model spec-
is also characterized by a spongy, brittle, rough appearance. troscope or recorded with a UV-Vis-NIR spectrometer. This
Although their identification was straightforward, the is because the beads’ dull luster, opacity, and rough, uneven
beads’ appearance caused doubt regarding their color origin. surfaces neither reflect light (in diffused reflectance mode)
When viewed under magnification (figure 4), distinct color nor absorb it (in transmission mode). Weak and broad ab-
concentrations were visible within the surface pores and sorption bands were displayed at approximately 540 and
cavities. The coloring agent was restricted to the surface, 680 nm, and the presence of calcium carbonate (aragonite)
was confirmed by Raman spectroscopy.
Such “soft coral” is commonly dyed red and the pores
Figure 3. This string of orangy pink “soft coral” beads are filled with a polymer to enhance polish and luster. But
(6–15 mm in diameter) displayed a rough surface and this specimen’s color and luster were unusual. The lab also
large pores. Photo by Gagan Choudhary. received similar coral strands in pink and black, suggesting
this dye method can produce a variety of colors.
Gagan Choudhary (gagan@[Link])
Gem Testing Laboratory, Jaipur, India

Strands of tiny akoya keshi pearls. How small can a pearl be?
At the American Gem Trade Association show in Tucson,
the Japanese Pearl Exporting co. presented some akoya keshi
pearls only 0.7–0.9 mm in diameter (figure 5). At this size, it
is difficult to distinguish individual pearls when looking at a
strand. According to the seller, the single strand shown in
figure 5 contains more than 300,000 keshi pearls, which
would take one person four years to drill and assemble.
Traditional tools are used to drill these tiny pearls. A
thick wooden board is soaked overnight to soften it, and
each pearl is embedded in the board to keep it stable. A

GEM NEWS INTERNATIONAL GEMS & GEMOLOGY SUMMER 2013 121

Figure 5. This strand contains more than 300,000
akoya keshi pearls, each one only 0.7–0.9 mm in di-
ameter. Photo by Eric Welch. Figure 7. These yellowish green kyanite rough crystals
range from 1.4 to 5.1 grams, while the cushion cabo-
chon weighs 4.84 ct. Photo by Don Mengason.
very sharp, thin drill is used to bore a hole in each pearl.
By pulling a thread attached to the drill, the worker can
keep rotating the drill until it reaches the desired depth. exhibited the following properties: an RI of 1.719–1.731,
“Keshi,” the Japanese word for poppy seeds, form as by- a hydrostatic SG of 3.70, and no reaction to short- or
products of the akoya culturing process. They account for less long-wave UV radiation. These physical properties were
than 0.5% of all annual akoya production, which has dimin- consistent with those published for kyanite. This iden-
ished in recent years. Akoya keshi pearls come in various tification was confirmed by Raman spectroscopy.
sizes, shapes, colors, and lusters. Strands are often paired with Microscopic examination revealed several transparent
colored stones such as ruby and sapphire (figure 6). crystals, fine particles, numerous fractures, and cleavages.
Tao Hsu Trapped in several cracks was an epigenetic reddish brown
GIA, Carlsbad mineral residue. It is also notable that a small percentage of
the rough crystals examined showed narrow blue color zones.
Green kyanite. At the Riverpark Inn show in Tucson, gem To determine the cause of the yellowish green color, we
dealer Tom Schneider (San Diego, california) exhibited a collected a visible spectrum using a UV-Vis-NIR scanning
parcel of kyanite displaying an unusual saturated yellowish spectrophotometer (figure 8). This revealed prominent fea-
green color (figure 7). Mr. Schneider said he purchased sev- tures located in the visible region at 432 and 445 nm. These
eral kilograms of the material in Arusha, Tanzania, its features have been attributed to Fe3+ in kyanite (R.G. Burns,
stated country of origin. While greenish blue, blue, and or- Mineralogical Applications of Crystal Field Theory, 2nd ed.,
ange kyanite from Tanzania have been previously docu- 1993, cambridge University Press). LA-IcP-MS measure-
mented (Winter 2004 GNI, pp. 341–342; Summer 2009 ments collected on the cushion cabochon confirmed a high
GNI, pp. 146–147), this would be the first vibrantly colored concentration of iron—more than 20,400 ppmw, consistent
yellowish green material from there. Yellowish green kyan-
ite from Brazil has been previously reported (Winter 2001
GNI, pp. 337–338), but it had a light color and lacked the Figure 8. Visible spectroscopy revealed dominant
strong saturation of the material from Tanzania. absorption features at 432 and 445 nm, attributed
A cushion cabochon fashioned from one of the rough to Fe3+ and consistent with the very high levels of
crystals was examined in the carlsbad laboratory and iron measured in this yellowish green kyanite.

Figure 6. These tiny akoya keshi pearl strands are VISIBLE SPECTRUM
ABSORPTION COEFFICIENT (cm-1)

paired with ruby and sapphire. Photo by Eric Welch.

15

10
432 nm
445 nm
5

350 400 450 500 550 600 650 700 750 800

WAVELENGTH (nm)

122 GEM NEWS INTERNATIONAL GEMS & GEMOLOGY SUMMER 2013

with the saturated yellowish green color.
This new supply of vibrant yellowish green material
from Tanzania presents an inexpensive option for unusual
ornamental gem materials.
Nathan Renfro and Andy Shen
GIA, Carlsbad

Banded mimetite. Mimetite is a member of the apatite

group and a lead chloro-arsenate, known to collectors as
one of the rarest faceted stones. Some yellow transparent
specimens weighing up to 1 ct are known from old finds
in Tsumeb, Namibia. Small orange translucent stones
under 0.5 ct have been faceted from chinese mimetite, and Figure 10. The sphalerite variety schalenblende has char-
a new type of lapidary mimetite (figure 9), likely from Bo- acteristics similar to those of banded mimetite. This
livia but purchased in Tucson without a reported locality, specimen weighs 106.20 ct. Photo by Jaroslav Hyršl.
was identified recently.
The cabochon measured 40.5 × 31.7 × 7.2 mm and
weighed 177.75 ct, with a very high luster. Its lapidary work UV. The mimetite was identified by powder X-ray diffraction
was of low quality, but the stone had a very nice agate-like by John Attard of San Diego, but its exact origin could not
structure, with alternating yellow-brown and dark brown be determined. Thin veins of mimetite deposited from hot
layers. It closely resembled two unusual ornamental stones: springs are known to have been mined from a deposit in
the colloidal variety of sphalerite (figure 10) known by the Lomitas, located in La Paz Department, Bolivia.
German name schalenblende, and the colloidal cassiterite
variety called “wood tin.” Schalenblende is known from Jaroslav Hyršl
many localities worldwide, mainly Poland but also Bolivia,
Turquoise-rock crystal composite. Recently examined was
where it is very difficult to visually distinguish from
a faceted rock crystal backed by a slice of turquoise—an
mimetite. Wood tin is known to be from Bolivia and Mexico.
obvious but unusual combination of gem materials. The
The RI of all three minerals is very high, over the refrac-
21.58 ct oval measured 20.02 × 15.03 × 9.05 mm and was
tometer limit. Loose stones can be easily distinguished by
easily identified as a composite due to its color and appear-
their hardness and specific gravity, and the hardness of
ance. From the top, it appeared light greenish blue (figure
mimetite and schalenblende is the same, about 3.5 on the
11, left); from the sides, it exhibited a colorless top and
Mohs scale. Wood tin is much harder, about 6.5. The SG of
bluish green slice at the base (figure 11, right).
the mimetite was 6.31, considerably higher than the 3.8–4.2
Further tests confirmed the specimen was made of rock
of a typical schalenblende. The SG of wood tin can vary from
crystal and turquoise, based on its colorless and greenish
5.2 to 6.6, likely depending on the presence of thin chal-
blue components. The colorless portion displayed an RI of
cedony layers. Mimetite and wood tin are also inert in UV
1.543–1.552, with birefringence of 0.009, while the greenish
light, while schalenblende is usually yellow in long-wave
blue layer displayed a vague shadow edge at around 1.61.
Under magnification, the colorless portion was clean and
Figure 9. This mimetite cabochon, likely from Bolivia, free of inclusions, while the greenish blue base showed some
weighs 177.75 ct and has an agate-like structure of whitish cloudy patches, consistent with those in turquoise
yellow-brown and dark brown layers. Photo by and similar materials. As expected, the junction plane con-
Jaroslav Hyršl. tained numerous trapped, flattened gas bubbles without
color. Under long- and short-wave UV light, a white to light
yellowish glow was visible only along the junction plane,
indicative of glue. Under a desk-model spectroscope, the
specimen showed a weak absorption at about 430 nm.
FTIR spectra taken by orienting the colorless portion dis-
played peaks at approximately 3595, 3480, 3380, 3305, and
3198 cm–1—a pattern typically associated with natural rock
crystal (e.g., Summer 2011 GNI, pp. 146–147). Raman analy-
sis of the greenish blue base using a 532 nm laser in the 200–
2000 cm–1 region revealed a strong peak at about 1040 cm–1,
with a number of weaker peaks at approximately 238, 334,
427, 482, 554, 596, 645, and 810 cm–1. These peaks are con-
sistent with those of turquoise in both the lab’s database and
the RRUFF database of Raman spectra. No polymer-related

GEM NEWS INTERNATIONAL GEMS & GEMOLOGY SUMMER 2013 123

 Figure 11. This 21.58 ct
composite was found to
have a top section of
natural rock crystal
and a base of natural
turquoise. Photos by
Gagan Choudhary.

peaks were present. In addition, qualitative energy-disper- were distinctly lower than the values for Larimar. Semi-
sive X-ray fluorescence (EDXRF) analysis revealed the pres- quantitative EDXRF chemical analysis gave a composition
ence of Al, P, Fe, and cu, consistent with turquoise. As of 95% Al2O3 and 4% SiO2, with traces of SO3, K2O, caO,
mentioned, a desk-model spectroscope revealed a weak ab- Sc2O3, cr2O3, Fe2O3, and Ga2O3. In contrast, the average
sorption at around 430 nm, and the same was confirmed composition of pectolite consists of 54.23% SiO2,
using UV-Vis spectroscopy, which displayed a sharp peak at 33.74% caO, 9.32% Na2O, and 2.71% H2O.
approximately 429 nm (Fe3+) and a broad absorption at about Although the imitation material does indeed resemble
680 nm (cu2+), suggesting natural rather than dyed color. Larimar, standard gemological tests such as microscopy,
While analysis using both classical and spectroscopic RI, and SG should be sufficient to distinguish it.
techniques identified this as a composite of natural rock crys- Lore Kiefert
tal and turquoise, the reason for its creation is still unclear. Gübelin Gem Lab Ltd.
Gagan Choudhary Lucerne, Switzerland
Peter Groenenboom
AEL Arnhem
SYNTHETICS AND SIMULANTS
Arnhem, Netherlands
Imitation Larimar. Pectolite is a mineral of the wollastonite
group with the chemical formula Naca2Si3O8(OH). It has an ERRATA
RI of 1.599–1.628, with a typical spot RI reading of about
1.60 and a specific gravity of 2.81. The light blue variety of 1. The Spring 2013 “About the cover” on the masthead
this mineral, known by the trade name “Larimar,” has only page should have listed the photographer as Jian Xin (Jae) Liao.
been reported from the Dominican Republic (R.E. Woodruff 2. The Lab Notes figure 7 photo was by Paul Johnson.
and E. Fritsch, “Blue pectolite from the Dominican Repub- 3. In the GNI entry on dumortierite-quartz rock, two
lic,” Winter 1989 G&G, pp. 216–225; figure 1). lines were omitted from the bottom left column on p. 60.
We recently acquired a 3.91 ct drop-shaped cabochon, The complete version is available in the Spring 2013 issue
purchased in India as a Larimar imitation, that consisted at [Link]/gems-gemology.
of a ceramic material (figure 12, left). According to the ven-
dor, the material was manufactured in china. close mi-
croscopic examination showed white gas bubbles and blue Figure 13. Microscopic observation reveals dark
color concentration (figure 13). blue color concentrations and whitish gas bubbles.
The cabochon’s spot RI reading of 1.53 and SG of 1.87 Photomicrograph by Lore Kiefert; magnified 40×.

Figure 12. This 3.91 ct ceramic imitation of Larimar

(left) is shown with Larimar from the Dominican
Republic. Photo by Lore Kiefert.

124 GEM NEWS INTERNATIONAL GEMS & GEMOLOGY SUMMER 2013

The Young Generation—Tom + He rejects traditional gem cuts such featuring an original gemstone carv-
Jutta Munsteiner as the round brilliant, which focuses ing paired with a diamond. They
on the reflecting quality of a dia- incorporated a proprietary round cut
By Wilhem Lindemann, 225 pp., mond and is far too complex for called “Spirit,” as well as the square
illus., publ. by Arnoldsche Art human sight. For this reason he “Context,” which resembles a mod-
Publishers, Stuttgart, Germany, adopted a cutting style that is asym- ern French cut.
2012. US$85.00 metrical, inviting the viewer’s atten- The photos exhibit a precision
With this remarkable bilingual vol- tion inside the crystal rather than on that complements the Munsteiners’
ume, Wilhem Lindemann presents its reflective qualities. cutting style. The rendering of color
the works of husband and wife Tom The next section takes the reader has not been artificially enhanced,
and Jutta Munsteiner, master gem on an inspirational journey through which is often a treacherous tempta-
cutters and jewelers. Second-genera- the work of Tom Munsteiner, whose tion. Captions generally mention the
tion gem cutter Tom Munsteiner fol- most memorable creations include carat weight of the gemstones, which
lowed in his father’s footsteps. “Magic Eye,” “Ritmo,” and “Spirit of is important since neither scale nor
Beginning in the 1960s and continu- Nature.” The “Magic Eye” pieces dimension are indicated. Given the
ing through the mid-1990s, Bernd have a sphere carved into the back of impressive size of certain objects,
Munsteiner reinvented gemstone the gems at the interface of their axes this information helps the reader
cutting with his unique edged, multi- to produce reflections on the face. appreciate the quality and craftsman-
faceted gem carvings. Working near “Ritmo” features geometric notches ship involved. Numerous illustra-
Idar-Oberstein, the mecca of the on the reverse side that re-create ran- tions show the variation of cutting,
German lapidary and colored stone dom asymmetrical imaginary struc- within each series and from one
trades, he cut an abundance of “redis- tures. Of particular interest to gemol- series to another. The many photos
covered” gemstones from Brazil’s ogists is the “Spirit of Nature” series serve to distinguish the couple’s
pegmatites, particularly tourmalines, of gem carvings, which accentuates respective pieces: Tom’s unique gem
aquamarines, amethysts, quartzes, the inclusions in a wide range of carvings and Jutta’s minimalist jew-
and topazes. materials. Some of the most striking elry designed with a sharp, powerful
Tom and Jutta Munsteiner were pieces are an oligoclase with red style.
classically trained as apprentices after hematite platelets, an elongated The text speaks to the Mun -
showing early artistic promise. aquamarine prism with a helicoidal steiners’ inspiration, but also
Today, they exhibit at venues such as inclusion running through the length describes how the gems were cut—a
the Tucson, Las Vegas, and Basel of the gemstone, and rock crystal helpful feature since some of the
shows. Visiting the couple’s booth, carvings displaying phantom crystals. gems were cut from the back to create
one can only be struck by the This section is one of the book’s a sense of volume and depth. Without
strength and originality of their cre- strongest. those details, such techniques would
ations. Whether gem carving or The third part focuses on Jutta’s be difficult to comprehend.
sculpture, each work exhibits a fine jewelry designs, which feature her This magnificent opus would
combination of jewelry and art. husband’s gem carvings. Her inspira- have benefited from the Mun -
In the first part of the book, tion features sinuous curves, as evi- steiners’ original sketches, which
Lindemann retraces the evolution of denced by the “Personalities” neck- would have made the volume livelier
lapidary art, from late third millen- piece series. The imposing “Jeanne and more personal. Some basic tech-
nium BC cylinder seals to contempo- d’Arc” necklace shows a ridged, elon- nical explanation of cut and lapidary
rary art movements such as gated rutilated quartz mounted on a work, being the couple’s craft and
Bauhaus. Tom Munsteiner’s influ- custom-made snake chain. expertise, would have also provided
ences, dating as far back as the The fourth part illustrates Tom some foundation. But these criti-
Romantic era, have inspired him to and Jutta’s collaboration through cisms are minor considering the
envision the entire crystal as a land- images of their classic jewelry, such quality of gemstones and illustra-
scape, with its inclusions and flaws. as the “Twins” series of rings, each tions presented.

BOOK REVIEWS GEMS & GEMOLOGY SUMMER 2013 S1

 The illuminating text and the ing charts on birthstones, ring sizes, Nicholas II in 1917. The author
striking originality and craftsman- hallmarks, and more. Overall, this details in chronological order the
ship of the Munsteiners’ work make volume would make a nice addition jewelry and objects d’art manufac-
this volume a real treat for students, to any gemological library if cost is tured in St. Petersburg during this
gemologists, gem cutters, and jewelry not an issue. period. The book contains numerous
enthusiasts. It offers an incomparable JANA MIYAHIRA-SMITH photos of the jewels and family por-
insight into Europe’s contemporary Gemological Institute of America traits from that era
and cutting-edge creations Carlsbad, California GIA LIBRARY STAFF
DELPHINE A. LEBLANC
Hoboken, New Jersey

Splendour and Science of Diamond Street: The Hidden

Pearls World of Hatton Garden
Dictionary of Gemstones & By Rachel Lichtenstein, 364 pp., publ.
By Dona Mary Dirlam and Robert
Jewelry Weldon, Eds., 139 pp., publ. by the by Hamish Hamilton, London, 2012.
By Akira Chikayama, 652 pp., illus., Gemological Institute of America, £20.00
publ. by IMACBC Co., Tokyo, 2013, Carlsbad, California, 2013, US$89.00 Hatton Garden is a famous area in
US$100.00 The book delivers a broad overview London where jewelry has been
The late Akira Chikayama was a of the world of natural and cultured manufactured since medieval times,
noted gemologist who considered the pearls. The well-researched and lav- and it continues to be known as a
Dictionary of Gemstones & Jewelry ishly illustrated presentation covers fine jewelry and diamond center.
his personal “world heritage.” the history of pearls, world localities, The author reveals the history of the
Indeed, the very scope of this dictio- classification, treatments, identifica- jewelry trade there and offers stories
nary is impressive. Entries include tion, fashion, and the future of the about the people, culture, and lore of
gems, minerals, sources, synthetics, industry. One chapter focuses on the this fascinating district.
treatments, jewelry findings, tools story of pearls in India, while anoth- GIA LIBRARY STAFF
used in gem cutting and jewelry er includes results from a GIA
making, famous gemologists, scien- research study on natural pearls from
tists, mineralogists, and artists—the the Arabian Gulf.
list goes on. GIA LIBRARY STAFF
A massive undertaking to begin Gemological Institute of America
with, the book was translated from Carlsbad, California
Japanese to English. But it did con-
tain some minor errors. In the listing
for the Searcy diamond, for example,
there appears a photo of the Sancy. In Jewels from Imperial St.
the entry for “abraded culet,” the Petersburg
author likened it to an open culet,
which I do not agree with. And see- By Ulla Tillander-Godenhielm, 295
ing entries on Gübelin, Kunz, and pp., illus. publ. by Unicorn Press,
Shipley made me wonder why Dana, London, 2012. US$80.00
Liddicoat, and Sinkankas were not Geared toward the jewelry historian,
included. this thoroughly researched work is a
But as a general reference, the window into the world of the
book serves its purpose well. The Russian imperial Romanovs, starting
appendices alone are useful, covering with the reign of Empress Elizabeth
a wide range of subjects and provid- (1741–1762), and ending with

S2 BOOK REVIEWS GEMS & GEMOLOGY SUMMER 2012

2013
 DIAMONDS
Abenteuer Jakutien: Auf den Spuren der Diamanten
[Adventures in Yakutia: On the track of the diamonds].
K. Erler and H. Vollstädt, Lapis, Vol. 38, No. 4, 2013, pp.
18–25, 54 [in German].
This article starts with a brief history of Yakutian diamond
production, from Mikhail V. Lomonosov’s theory of diamond
deposits in 1761 to the discovery of the first kimberlites in
1954 to present-day operations. The bulk of the article then
presents information the authors gathered during two visits
REVIEW BOARD to Yakutia, focusing on Mirny.
Edward R. Blomgren The report contains a map and a list of data on the most
Asheville, North Carolina important mines. It is lavishly illustrated with images of
Mirny—the mine, the town, the museum—and impressive
Jo Ellen Cole
Vista, California diamond specimens.
RT
Edward Johnson
GIA, London
China market: The new frontier. Diamond World, Vol. 40, No. 2,
Michele Kelley Jan-Feb. 2013, p. 126-132
Monmouth Beach, New Jersey China, the world’s second-largest jewelry consumer after the
Guy Lalous United States, is also the second-largest jewelry manufactur-
Academy for Mineralogy, Antwerp, Belgium ing center after India. There is enormous potential for dia-
Kyaw Soe Moe mond jewelry in China, where the government has supported
GIA, New York the industry through major policy decisions: reducing the
Keith A. Mychaluk VAT (value-added tax) on diamonds, abolishing import
Calgary, Alberta, Canada duties, and setting up the Shanghai Diamonds Exchange for
James E. Shigley trading in rough and polished diamonds. The government has
GIA, Carlsbad signed multibillion dollar resources-for-infrastructure deals
with various African governments, securing its supply of
Russell Shor
GIA, Carlsbad rough diamonds. These efforts have drawn a huge flow of cap-
ital investment from diamond companies in Tel Aviv and
Jennifer Stone-Sundberg
Antwerp. Skilled diamond craftspeople have been brought in
Portland, Oregon
from Surat, India, to train Chinese workers.
Rolf Tatje
Duisburg, Germany
Dennis A. Zwigart
State College, Pennsylvania
This section is designed to provide as complete a record as
practical of the recent literature on gems and gemology. Articles
are selected for abstracting solely at the discretion of the section
editors and their abstractors, and space limitations may require
that we include only those articles that we feel will be of greatest
interest to our readership.
Requests for reprints of articles abstracted must be addressed to
the author or publisher of the original material.
The abstractor of each article is identified by his or her initials at
the end of each abstract. Guest abstractors are identified by their
full names. Opinions expressed in an abstract belong to the abstrac-
tor and in no way reflect the position of Gems & Gemology or GIA.
© 2013 Gemological Institute of America

S3 GEMOLOGICAL ABSTRACTS GEMS & GEMOLOGY SUMMER 2013

 Overall, China has cheaper labor costs than India, and with high accuracy and symmetry. The new software
its workforce provides better finishing of jewelry. Much of requires a recent PC with a specific graphics card.
the work in Chinese jewelry factories is automated, and EOSFancy can be used manually or via a transfer from a
the level of technology and infrastructure is higher than rough scanner. The system can be used for both prebrut-
in India. India produces every size and type of rough, ing and finishing, and it is the ideal complement to laser
while China mostly manufactures sawables. While India prebruting. With EOSFancy, a lasered girdle can be fin-
holds the advantage in manufacturing, China is a major ished perpendicular to the table with a result closer to
consumer of polished diamonds. China is also better at the end dimensions, with a shorter processing time. The
jewelry manufacturing. But there is no original Chinese development is now in testing phase at a limited num-
brand of international jewelry, and country needs better ber of Antwerp-based diamond companies. A commer-
communication with the outside world. cial version is expected to be released by HRD during
GL 2013.
GL
The Ellendale diamond field: Exploration history, discovery,
geology, and mining. A. L. Ahmat, The Australian Infrared microspectroscopy of natural Argyle pink dia-
Gemmologist, Vol. 24, No. 12, 2012, pp. 280–288 mond. K. S. Byrne, J. D. Anstie, J. Chapman, and
The Ellendale diamond field in West Kimberley is one of A.N. Luiten, Diamond and Related Materials, Vol.
only three hard-rock diamond mine locations in 23, 2012, pp. 125–129, [Link]
Australia. Though not the first Australian diamond mond.2012.01.032
mine, Ellendale was the country’s first hard-rock One of the most distinctive characteristics of natural
deposit. It holds a special place in world diamond histo- pink diamond is the lamellae, or “graining,” in which
ry as it led in November 1976 to the recognition of a the color resides. These features are believed to have
new host-rock for diamond, olivine lamproite. Up until formed under plastic deformation. A great deal of
that time, commercial-sized diamonds were considered research has focused on color centers in diamond, partic-
to be sourced only from kimberlite. The Ellendale lam- ularly the N-V center. This article investigates the crys-
proites are geologically very young, only 22 Ma (million talline defect responsible for pink coloration in natural
of years) old. Within several years of the initial discov- diamond. Infrared microspectroscopic images were
ery, some 46 lamproite pipes were found at Ellendale. By obtained to estimate the spatial distribution of B-centers
1980, 38 of these pipes had been assessed for their dia- in natural pink Argyle diamonds. The spatial distribu-
mond content. tion of the nitrogen B-center (N–V–N3) was anti-correlat-
More than two decades later, geologists from the ed with the intensity of pink coloration. The authors
Kimberley Diamond Company (KDC) recognized eluvial believe this is the first such observation. IR spectral fea-
diamond enrichment over these pipes. After a lengthy tures were not influenced by exposure to UV illumina-
legal battle, they wrested the Ellendale mining lease tion, in contrast to the visible change in coloration
from the Ashton Joint Venture and commenced mining under the influence of UV radiation.
there in May 2002. Ellendale is recognized as a source of GL
high-value fancy yellow diamonds. These high priced
stones have been marketed through a special deal with Internal texture and syngenetic inclusions in carbonado.
Tiffany & Co since 2009. But the future of mining there F. V. Kaminsky, R. Wirth, and L. Morales, The
is tenuous. Ellendale 4 was closed in 2009, and the high Canadian Mineralogist, Vol. 51, No. 1, pp. 39–55,
Australian dollar, combined with dwindling reserves, [Link]
may jeopardize the survival of Ellendale 9. Carbonado is a black or dark brown cryptocrystalline
GL aggregate of diamond. Its age ranges from 2.6 to 3.8 Ga
(billions of years), according to lead isotopic composi-
EOSFancy: Automatic bruting of fancy shapes. T. Gevers tion. Carbonados have been found only in alluvial placer
and J. Vandeloo, De Belgische Diamantnijverheid, deposits, such as those in Brazil and Central Africa, and
Jan/Feb/Mar 2013 Vol. 83, No. 1, pp. 14–16. never in kimberlite. Many hypotheses, including non-
More than eight years ago, Belgium’s Scientific Research kimberlitic, crustal origin of carbonado were proposed
Centre for Diamonds (WTCOD) developed an automatic by previous researchers. Analysis of internal texture and
bruter for round diamonds. The bruting machine was syngenetic inclusions is important to understanding the
based on a new principle: the application of a grinding genesis.
wheel, rotating in a cooling liquid, to a rotating dia- Carbonado samples were collected from alluvial
mond. WTCOD has now expanded the scope of the EOS placer deposits of the Macaubas River basin in the
bruting machine to fancy shapes by linking the feed Brazilian state of Minas Gerais and the Ubangi River
movement of the disc to the rotating position of the dia- basin in the Central African Republic. They were dark
mond. With EOSFancy, one can brute any convex shape or nearly black, with a smooth and shiny surface. The

GEMOLOGICAL ABSTRACTS GEMS & GEMOLOGY SUMMER 2013 S4

samples contained pores ranging from a few hundred er. Many also display a strong green long-wave UV fluores-
nanometers to a few micrometers. Secondary electron cence, supposedly due to uranyl ions.
images revealed a typical polycrystalline structure, The beauty of the agates and their fluorescence are
which showed fine-grained texture with irregularly illustrated by a range of color photos.
shaped diamond grains averaging 5 to 15 µm. Grain RT
boundaries were in irregular, zigzag patterns. Numerous
pores, up to 20 µm, were found inside grain boundaries. Montana moss agate: Collectible chalcedony from the
They were usually filled with postgenetic minerals such Yellowstone River. B. Britt, Rock & Gem, April,
as kaolinite, flurencite, and quartz. Dislocation lines 2013, Vol. 43, No. 4, pp. 34–37.
were also observed within some diamond grains. The The expansion of various mining projects in Canada’s
pole figures suggested that crystallographic orientations North (Northwest Territories, Yukon, and Nunavut) is set
of these grains were quasi-random. to lead the country in economic growth over the next two
Syngenetic inclusions ranging from 10 to 50 nm years. Spinoff activity that accompanies mine-related
were identified using the combined FIB/TEM technique: prosperity is also expected to increase based on pending
garnet, apatite (including fluorapatite), phlogopite, silica, mines in the assessment phases or preproduction stages,
Ca-Mg-Sr- and Ca-Ba-carbonates, halides (sylvite, KCl, thus further boosting employment and local economic
and bismocolite, BiOCl), native nickel and metal alloys development. Production in these areas is expected to last
(Fe-Ni, Cr-Fe-Mn, and Pb-As-Mo), oxides (FeO, Fe-Sn-O, well into the coming decades. Environmental groups and
TiO2, SnO2, and PbO2), Fe-sulfides, and fluid inclusions. local residents are also interested in the impact on the
An intergrowth of fluorapatite, phlogopite, and silica Canadian North ecosystem as well as the long-term
was also observed. Almandine-pyrope garnet inclusions effects on the towns and industry when the boom is over.
with silica-enriched amorphous rims were confirmed. ERB
Fluid inclusions contained Si, Al, Fe, O, Ti, Ca, S, Cl, K,
and carbonate. These fluid inclusions suggested the car- Neufund: Blaue “Wolkenachate” von Weierbach bei Idar-
bonados were formed in a chloride-rich environment. Oberstein, Rheinland-Pfalz [New find: Blue “cloud
Crystal inclusions were not pressure-indicator minerals. agates” from Weierbach near Idar-Oberstein,
Instead, they were stable in a wide range, from crust to Rhineland-Palatinate]. K. Schäfer, Lapis, Vol. 37,
lower mantle, except bismocolite, which could only be No. 6, 2012, pp. 25–29 [in German].
formed in the crust. Although this whole inclusion The agate deposits that formed the basis of the lapidary
assemblage was possibly of crustal origin, conclusive and cutting industry at Idar-Oberstein until the 19th cen-
research is still needed. The authors propose that these tury are mostly depleted. What reaches the market today
samples were crystallized and grown in one stage, which is the result of rockhound activities. This article reports
was followed by sintering. This hypothesis could be con- on a remarkable find from 2012 along a road construction
firmed by comparison with one-stage growth process of site at Weierbach, near Idar-Oberstein. Found in a small
a polycrystalline diamondite. outcrop of volcanic rocks, the nodules showed diameters
KSM of 2–12 cm (0.8–4.8 in.). While agates from Idar-Oberstein
generally have a fine banding and pastel tones, most of
these agates were not banded but showed cloud-like
structures and striking sky blue and red colors. The arti-
GEM LOCALITIES cle contains numerous images of the splendid agates and
Achate von der Teufelskanzel bei Oberthal im Saarland the deposit.
[Agates of the Teufelskanzel near Oberthal, RT
Saarland, Germany]. K Schäfer, Lapis, Vol. 38, No. 5,
2013, pp. 14–21, 54 [in German]. Wittichenite from the Cattle Grid Pit, Mount Gunson
The “Teufelskanzel” (Devil’s Pulpit), a rhyolitic promon- Mine, South Australia. R. Noble
tory at Oberthal about 30 km (20 miles) southwest of Idar- [noblemin@[Link]], The Mineralogical
Oberstein, is one of the many small agate deposits of the Record, March-April, 2013, Vol. 44, pp. 133–143.
Nahe-Saar Basin. A failed first attempt to mine the agates In 1981, large, attractive wittichenite crystals found at the
in 1882 was followed by rather chaotic rockhound activi- Cattle Grid Pit in South Australia were mistaken for chal-
ties and the equally unsuccessful decision by the commu- cocite. Then in 2008, they were correctly identified by the
nity of Oberthal to take over all mining activities. Finally, Smithsonian Institution. Electron microprobe analysis of
digging for agates was generally prohibited, a ban that is this species shows Cu, Bi, and S (weight percentages of
still in force. 38.67%, 41.70%, and 19.90%, respectively).
The agates generally formed as thunder eggs. They The Cattle Grid wittichenite is brittle, with strongly
show a very beautiful coloration and often bizarre, pic- conchoidal fractures and a relatively low hardness. The
turesque forms that inspire the imagination of the behold- colors vary from highly lustrous silvery white to bronze to

S4 GEMOLOGICAL ABSTRACTS GEMS & GEMOLOGY SUMMER 2013

a deep purple-black. The crystals are predominantly tabu- between mine sites was revealed by a non-destructive test
lar-prismatic in habit, but some show myriad crystal faces that preserved the integrity of the sample.
and deeply striated prisms. The Miller indices for these JEC
crystal forms, and their intricate and complicated crystal
morphology, are discussed in depth. The yellow color center and trapped electrons in beryl. L.
The article contains 16 large photos of wittichenite, Andersson, [loandersson@[Link]], The
some crystals weighing more than 75 grams, as well as the Canadian Mineralogist, 2013, Vol. 51, No. 1, pp.
mine site and the copper localization stratigraphy. The 15–25, [Link]
number of wittichenite crystals larger than 2 cm in maxi- Investigations continue into the cause of yellow color in
mum dimension is estimated to be about 30, and these are beryl. The generally accepted explanation is that it is relat-
held in only three collections (two institutional and one ed to the charge-transfer between Fe3+ ions substituting for
private). Unfortunately, the entire production of copper octahedral Al3+ ions in the crystal and the surrounding
sulfide concentrate from the Cattle Grid Pit is now owned oxygen ions.
by a private smelter, where the bismuth-containing wit- Using electron paramagnetic resonance (EPR), an
tichenite is considered an undesirable impurity. absorption technique that uses magnetic fields to measure
ERB transition energies in atomic defect centers, this study
contradicts the accepted interpretation, proposing an alter-
nate model.
Although a strong EPR signal from octahedral Fe3+ ions
INSTRUMENTS AND TECHNIQUES can be found in beryl of all colors, including colorless, the
Determining the geographical origins of natural emeralds study determined that Fe3+ ions substituting for Al3+ at the
through nondestructive chemical fingerprinting. D. octahedral position cannot be the cause of the yellow
P. Cronin and A. M. Rendle. The Journal of color. There is, however, a signal unique to yellow beryl
Gemmology, Vol. 33, No. 1-4, 2012, pp. 1–13. that comes from Fe3+ ions at a tetrahedral site.
The technology to identify a gemstone’s geologic and geo- The author proposes a simple model describing the oxi-
graphic origins has improved in the last decade to the point dation and reduction processes involving iron electrons
where accurately pinpointing the locality has become a trapped in the crystal structure to explain the creation and
reality. Using emeralds from different mines in the decay of the yellow color. This study discusses reinterpre-
Cordillera Oriental region of Colombia, some of them only tations of other experimental findings that lead to different
a few miles apart, scientists can now discern minute differ- explanations for beryl’s colors
ences in chemical composition and determine each sam- ERB
ple’s point of origin by studying the chemical “finger-
prints” in their chromophores. The chromophores are trace
elements of vanadium, nickel, or magnesium and vary geo-
graphically as a result of interstitial fluid migration prior to
SYNTHETICS AND SIMULANTS
precipitation. Blue spinel crystals in the MgAl2O4-CoAl2O4 series: Part I.
After recognizing that within-site chemical homo- Flux growth and chemical characterization. V.
geneity and cross-site chemical heterogeneity depended on d'Ippolito, G. B. Andreozzi ([Link]@uniro-
the amount of chromophores in the crystal structure, the [Link]), F. Bosi, and U. Hålenius, American
authors sought to find out if chemical heterogeneity Mineralogist, Vol. 97, No. 11-12, 2012, pp.
between emerald mining localities was chemically unique 1828–1833, [Link]
and statistically provable based on chromophore compo- Synthetic CoAl2O4 spinel, a high-temperature oxide
nents alone. Included in the study were emerald samples (melting point of 1955°C), is the most stable compound of
from the Muzo, Cosquez, and Guali mines of Colombia, a family of spinel-structured oxides obtained from CO3O4
plus the Campo Verdes and Carnaíba mines in Brazil, and at progressive increase of Al contents. High-quality spinel
the Chantete mine in Zambia. Consideration for sample single crystals with compositions closely corresponding
inclusion in this study involved the history of the host rock to the solid-solution series spinel sensu stricto (MgAl2O4)-
chemistry on the skeletal crystal structure, the occurrence cobalt spinel (CoAl2O4) were produced by flux growth
of trace elements that link emeralds to their point of origin, method, with Na2B4O7 as the flux. Low cooling rates and
the subsurface geology, fluid-rock interactions and the geo- linear temperature profiles were applied in the thermal
chemical processes leading to emerald formation. interval 1200–800°C, followed by rapid cooling. Thermal
The study concluded that element-based heterogeneity runs were performed in a reducing atmosphere. Selected
exists between the six emerald mining sites. Through the crystals were investigated by SEM/EDS X-ray mapping to
use of an Amray scanning electron microscope (SEM) cou- check for compositional homogeneity and by electron-
pled to an EDAX energy-dispersive X-ray spectrometer microprobe analysis to obtain the chemical formula.
(EDX), statistically significant chemical heterogeneity Crystals were found to be chemically homogeneous

GEMOLOGICAL ABSTRACTS GEMS & GEMOLOGY SUMMER 2013 S5

and entirely representing the MgAl2O4-CoAl2O4 solid- the square top of a pyramidal substrate, were observed in
solution series, with the latter component ranging from a crystal grown on type A substrate with 20° dislocations.
7% to 100%. The saturation of the vivid blue color However, these dislocations were diverted toward the
increased with total cobalt content, but no shift in color edges rather than the top. Photoluminescence image
hue was observed along the series. The unit-cell parame- showed that the inclined faces effectively diverted the
ter a increased from 8.084 to 8.105 Å along the solid-solu- direction of threading dislocations from {100} to {110}
tion series. The observed linear increase of the a parame- direction. Absence of sharp edges in these pyramidal sub-
ter was seemingly due to the higher CoAl2O4 component. strates also helped in reducing dislocations. The lateral
This was a premature and misleading conclusion, howev- surfaces had the highest growth rate and possessed “fish
er, because structural changes in spinel may depend on scale” morphology in type A grown crystals, a morpholo-
both chemical composition and inversion parameter, gy not observed in type B crystals. A maximum thickness
which in turn is a function of thermal history. The com- of 2.2 mm could be achieved with a dislocation of 40° on
position of crystal products does not correspond to the type B substrate. A crystal 1.70 mm thick containing very
composition of the starting oxide mixture: cobalt low defects was achieved using type B substrate with 20°
enriched in the crystals. A tentative explanation of this or 30° dislocations.
behavior is suggested by considering possible ionic poten- KSM
tial as well crystal field stabilization effects.
GL Zircon et geikielite artificiellement étoilés du Sri Lanka
[Zircon and geikielite from Sri Lanka with artificial
Growth strategy for controlling dislocation densities and stars]. J.-P. Gauthier, B. Rondeau, and T. Prada,
crystal morphologies of single crystal diamond by Revue de l'Association Française de Gemmologie,
using pyramidal-shape substrates. A. Tallaire, J. No. 180, 2012, pp. 5–11 [in French].
Achard, O. Brinza, V Mille, M Naamoun, F Silva, Two asteriated cabochons acquired in Sri Lanka were
and A. Gicquel, Diamond and Related Materials, identified as natural zircon and geikielite. But the six-
Vol. 33, pp. 71–77, [Link] rayed star had not been observed in natural star zircons,
mond.2013.01.006 and the eleven-rayed star of the geikielite was crystallo-
A single synthetic diamond crystal can be used as an elec- graphically “impossible.” Microscopic observation clearly
tronic device because of its properties, such as a high showed both had been created by a fine scratching of the
breakdown field and thermal conductivity. However, cabochon domes.
defects, such as dislocation bundles, which were formed In both cases, the surfaces appeared to have been pro-
during growth process, can cause current leakage, higher duced by short parallel scratches. The center of each star
bandgap, and birefringence. Defect-free single diamond was a white spot, and their branches were short, some-
crystal that is large enough to be used as an electronic times missing or undulating and hatched. The effect is
device is difficult to grow using current methods. For formed mainly by diffraction at the scratches rather than
example, the “threading dislocations” can be formed in a reflection from needles or hollow tubes, like in natural
CVD diamond using (001)-oriented substrate. The authors stones. When the focus of the microscope is lowered
demonstrated that these defects could be reduced by mod- some millimeters from above the cabochon surface
ifying substrate’s shape and orientations. through below the surface, the star gets finer and is
Two type Ib HPHT-grown, pyramidal substrates, type sharpest below the surface, in contrast to natural stars,
A and type B, were used to grow single CVD crystals in where the sharpest star seems to hover above the surface.
this research. The {100} directions in type A substrate and The features described above, along with some other
{110} directions in type B were polished into inclined faces characteristics mentioned by Steinbach (2011), allow a
with 20°, 30°, and 40° angles. Both possessed a square top safe distinction of scratched star stones from their natural
measuring 200 x 200 µm2. Crystals were grown on these counterparts mostly by simple observation with the
substrates using plasma-assisted CVD technique with naked eye.
power density of 100 W/cm3 and methane concentration RT
of 5% at 850°C, achieving a growth rate of 13 microns per
hour.
After removal of the substrate, the fully grown CVD
crystals showed high transparency. Nitrogen-related
defects were not detected. The inclined faces in type A
substrate helped prevent twin sectors from reaching the
top face. The growth rates of the lateral and top faces
were 32.5 and 13.5 µm per hour, respectively. When
thickness reached about 490 microns, the inclined faces
started to disappear. The dislocations, originating from

S6 GEMOLOGICAL ABSTRACTS GEMS & GEMOLOGY SUMMER 2013