Tuesday, May 22, 2012

The Hope Diamond: blue by day, red by night*

*for a minute, after exposure to ultraviolet light.

Red phosphorescence of the Hope Diamond. One colorless diamond of the necklace setting presents a blue phosphorescence (copyright: Eloïse Gaillou). 

First of all, I want to explain the reason for this post: I have seen so many inaccurate blog posts and articles about the Hope Diamond that I thought it was time to give a good overview of what we know about the Hope Diamond, mostly from a scientific point of view. I was a postdoc at the Smithsonian Institution in the Department of Mineral Sciences from 2007 until 2011, working with Jeff Post, the curator of the National Gem & Mineral Collection. My research interests during my time at the Smithsonian (research that I'm carrying on at NHM and as research associate at the Smithsonian) were on pink and blue diamonds. Yes... including the Hope Diamond. I have been part of the team of researchers and the lead author of two scientific papers which included the Hope Diamond:

Gaillou E., Post J.E., Butler J.E. (2012) Boron in natural type IIb blue diamonds: chemical and spectroscopic measurements. American Mineralogist, Vol. 97, No 1, 1-18.
The Hope diamond under the ion gun of the ToF-SIMS (copyright: Jim Butler).

Gaillou E., Wang W, Post J.E., King J.M., Butler J.E., Collins A.T., Moses T.M. (2010) The Wittelsbach-Graff and Hope diamonds: not cut from the same rough.Gems & Gemology, Vol. 46, No 2, pp 80-88.
The Wittelsbach-Graff (left, 31.06 ct) and Hope Diamond (right, 45.52 ct) side by side, during a unique night of analyses at the National Museum of Natural History of the Smithsonian Institution (copyright: Smithsonian Institution, photo by Chip Clark).

Here is a general summary of the research I have been conducting with my colleagues on the Hope Diamond:

1- The Hope's boron content.

Background information:

The blue color of the Hope Diamond is due to a rare impurity: boron. Diamonds should be pure carbon in the compact cubic form. But there is no such thing as a perfect diamond, and there are always impurities that replace a few carbon atoms in the diamond structure. The most common one is nitrogen, from a few ppm (or below for some rare diamonds) up to 1000 ppm or more on average, I would say, around 400-500pm. Hydrogen is another important impurity in diamond. Very rarely, boron is also seen. The problem with boron is that it is known that, when it's present in diamonds, it is in a tiny quantity: we are talking about sub-ppm level here. A good way to see if a diamond contains boron is to look at its infrared spectrum, as boron will show very strong absorptions, even at very low content. It is actually the presence of these strong absorptions in the infrared that makes a blue diamond blue: the strong absorptions in the infrared tail off in the visible range, absorbing all the colors of the spectrum, except for the one the further away from the infrared (the closest to the ultraviolet): blue.
FTIR spectrum of a blue (and type IIb) diamond from the Aurora Butterfly of Piece collection. The peaks indicated in blue are due to boron absorption. This diamond contains a "high" amount of boron, which is about 0.49 ppm! (copyright: Eloïse Gaillou).

But adding to the difficulty of detecting boron, we know that when nitrogen is also present, it will "compensate" (or quench, if you want) the boron impurities. So if you have the same amount of boron and nitrogen in your diamond, the boron will be "inactive", not giving rise to any absorption, therefore, not giving any blue color either.
By the way, did I mention that boron also gives diamond the properties of a semiconductor? Boron-containing diamonds have important applications in electronic devices, such as high voltage diodes, transistors, electrochemical electrodes, and superconductors for example. Nowadays, scientists are capable of creating synthetic diamonds with high boron content, or "dope" diamonds (mostly synthetics) with boron (introducing boron atoms inside a diamond). These diamonds contain much more boron than natural ones, and have different physical properties (such as a difference in their phosphorescence as well).

The measurements, the results:

Our goal was to measure the active and total (active + inactive) boron content in several blue diamonds... including the Hope diamond. Measuring the active boron is easy: we use a Fourier-transform infrared spectrometer (FTIR). To measure the total boron content... not that easy! Especially in a non-destructive way. But we developed a method that was close enough to non-destructive: it removed only a few nanometers in thickness of materials (we are talking about only hundreds of atomic layers) by 50 µm x 50 µm on the surface, using a time-of-flight secondary ion mass spectrometer (ToF-SIMS). That is why were are know to have "drilled" into the Hope diamond, as reported in the documentary "The Mystery of the Hope diamond". These experiments have also been reported in the New York Times: "For scientists, Hope diamond's blue may offer geology lesson".

The Hope Diamond contains in average 0.36 ppm (atomic) of active boron (as seen by FTIR). It is usually accepted that natural blue diamonds contain less than 0.5ppm of boron, so, that matched what was already known. The surprise came from the local chemical analyses obtained with the ToF-SIMS, measuring the total boron content. Remember: we are actually counting the number of atoms of boron among the atoms of carbon with this method, while FTIR gives us the average active boron content. We did a total of 8 measurements, in 5 different areas. And here is what we did not expected: some areas of the Hope Diamond were "boron-rich", containing up to 8 ppm of boron, while other parts were boron depleted, with boron content below the detection limit of the machine (about 0.15 ppm). The distribution of boron inside the Hope Diamond (and inside other blue diamonds as a matter of fact) is heterogeneous. This means that, during growth, the boron content of the environment either fluctuated or the environment itself made it more or less difficult for the boron present to be incorporated inside the diamond. The variables of pressure, temperature, oxidizing / reducing conditions, compatibility etc. are not at all constrained for the boron / diamond couple.
I am here cleaning the Hope diamond before introducing it into the ToF-SIMS (copyright: Eloïse Gaillou & Jeff Post).

Mounting the Hope Diamond for the ToF-SIMS experiment. We conducted the analyses on the culet facet (on the back of the diamond). A conducting film was placed around the culet facet of the Hope Diamond (copyright: Eloïse Gaillou & Jeff Post).

Close-up view of the Hope Diamond mounted, ready to be introduced inside the ToF-SIMS (copyright: Eloïse Gaillou).

My colleague, Detlef Rost putting the Hope Diamond inside the ToF-SIMS (copyright: Eloïse Gaillou).
View of the Hope Diamond through the window of the ToF-SIMS (copyright: Jim Butler).

Detlef Rost getting ready to analyze the boron content of the Hope Diamond (copyright: Eloïse Gaillou).

2- The red glow.

Background information.

First, let's make the difference between fluorescence and phosphorescence. Fluorescence defines the emission of light by a material (such as a mineral) while exposed under a source of light, such as ultraviolet (UV) light. Have you even noticed your white shirt or your teeth (made partially of the mineral apatite) glowing in a club? Clubs and bars often have what they call a black light, which is basically UV light, and some materials will react to that by "glowing" while exposed to the light. That's what we call fluorescence.
Phosphorescence defines the emission of light by a material after exposure to light. I'm sure you have all seen the "glow in the dark" stars that you can glue to your ceiling, or the hands of some watches, they are exposed to light, but glow after the light is turned off. That is phosphorescence.

Hope Diamond (in the black box) almost ready to have its phosphorescence analyzed. The white box with a red and blue button is the UV lamp; next to it (on its left), the small spectrometer with which the light emitted is registered.

The Hope Diamond with the phosphorescence probe, which is composed of seven fiber optics (six to conduct the UV light, one to collect the light emitted by the sample).

Phosphorescence of blue diamonds.

Blue diamonds never show fluorescence (never emit light while under exposure to UV). But most blue diamonds show a very short (1s or so) whitish-blue phosphorescence. Some look like they don't display any phosphorescence. Rarely, some will show a long lasting red phosphorescence. That is the case of the Hope Diamond.
The Hope diamond: left, under normal lighting conditions. Right, after exposure to UV light (copyright: Smithsonian Institution. Photos by Chip Clark).

We can now not only estimate the phosphorescence, but measure it with a spectrometer. What we noticed is that all blue diamonds present a phosphorescence (sometimes extremely weak, and  not visible to the eye) with two components: one in the blue and one in the red (see figure below). The blue phosphorescence doesn't last, while the red, when intense enough to begin with, decreases very slowly over time. Depending on the diamond, the relative intensity of these two components varies.
Phosphorescence spectra, over time, of a blue diamond (beginning the acquisition just after the UV light is turned off). As you can see in this figure, even the Hope diamond present a blue component,
which is much weaker than the red one, that is why we only see a red glow in the dark. Look at the time scale as well.

Phosphorescence spectra over time of the Hope Diamond. After exposure to UV light, the Hope Diamond exhibit an intense red phosphorescence, that fades slowly, and disappears after a minute and a half.

The phenomenon of phosphorescence in blue diamonds is not well understood. We are trying to give an explanation in our article, but more experiments are needed. Both emissions in the blue and red most likely are created through a process called "donor-acceptor pair recombination"  involving distant neutral donors (most likely N-related) and neutral boron acceptors. We will keep you posted when we will know more about this process. Give us a year or two! We will need to do experiments at very low temperatures (about -340°F or -200°C), for which we have to use liquid nitrogen to cool the diamonds down. But we won't conduct such an experiment on the Hope Diamond!

From left to right: Jeff Post, Jim Butler, and Eloïse Gaillou. We are here acquiring a phosphorescence spectrum of the Hope Diamond (copyright: Smithsonian Channel).

3- Boron in blue diamonds: coming from the ocean?

The recent history of the Hope Diamond.

Let me talk about the history of the Hope Diamond. Of course, I could mentioned that it was first own by a French dealer, Tavernier, who purchased it in India where it was mined, in the 17th century. The diamond, which was roughly cut, and was about 115ct was sold to King Louis the XIV of France, who made it recut in a gorgeous 69 ct stone, taking the name of "Diamant bleu de la couronne" (or French Blue). Much later on, the diamond was stolen during the French revolution (1792). We now know that the 45.52 ct stone that appeared on the English market and which was bought by Henry Philip Hope, was the recut of the French Blue. The Hope Diamond was eventually bought by Cartier who sold it to Evalyn Walsh McLean in 1910 (by the way,  Cartier created the myth of the curse). At her death, Winston bought it, and eventually donated it to the Smithsonian institution in 1958, where it will be forever (hopefully). But you can learn much more about the history of the Hope Diamond, and all mysteries that are still not solved, when the French Blue was stolen for example, by reading books and articles on the subject. The Smithsonian has a good reference list on the matter, and a good time line at on this page.

The Tavernier Blue, ancestor of the Hope diamond. It was acquired by Jean-Baptiste Tavernie in India around 1650. 

Gouache version of the Order of the Golden Fleece by Monney for Horovitz & Farges (2008). The Golden Fleece was made in 1749 and worn by King Louis XV. The French Blue is the major component of this piece (copyright Horovitz & Farges).

The Hope Diamond, in the setting as owned by Henry Philip Hope. Extracted (and copyright) from Goddard Orpen (1890 - Stories about Famous Precious Stones).

Evalyn Walsh McLean wearing the Hope Diamond, mostly likely not long after she acquired it from Cartier (in 1911).

The Hope Diamond, in its Cartier setting, own by the Smithsonian Institution since 1958 (copyright: Smithsonian Institution).

 Temporary Winston setting for the Hope Diamond done in 2010 (Embracing Hope), in which the Hope Diamond was displayed for a year. It is now back in its original setting. Below: worn by Hilary Rhoda (copyright for both pictures: Smithsonian Institution).

The ancient history of the Hope Diamond.

This will be more of a discussion than actual facts here...
I won't be able to tell you exactly when the crystal from which the Tavernier (then the French Blue and eventually the Hope)Diamond began to grow. For that, we would need a diamond with specific inclusions (such as sulfides) to do some chemistry on them to be able to date it. Oh, by the way, in the process, we would have to break the diamond in order to remove the inclusions. Ok, so the exact when doesn't really matter. Most likely, the Hope Diamond crystal (as I'm going to call it) began to grow when the cratonic structure (this very old and thick crust) was already in place, in what is now India (there is a good online document about the geology and the resources, including diamonds, of India, click here). That might have happened between 2.5 and 3.5 billion years ago (Ga) approximately. It is known that diamonds are preferentially formed underneath these old cratons that are distributed all over the world: it's the Clifford's rule.

So, here we are, in the Earth's mantle, between 2.5 and 3.5 Ga ago, at least 150 km below the surface of the Earth. I will here make a big simplification, but if you want to learn more about diamond formation, you can check out for example the special issue of the magazine Elements on diamonds. There are two main environments in which diamonds grow: in a peridotite, this olivine-rich rock which makes most of our Earth's mantle, or in an eclogite, which was originally a subducted plate that went down in the Earth's mantle. So the carbon making the diamonds is either of mantle origin or a sub-surface origin. And we can tell that by looking at the carbon isotopes.
Now, what about boron? Boron is not an element that is in high concentration deep inside the Earth. But at the same time, the boron content is not high in diamonds either. And as we don't know the partition coefficient between diamond and boron (at what rate, what concentration and under which conditions boron can be incorporated inside the diamond structure), we just don't know if we need a lot of boron or not to incorporate less than 1ppm inside the diamond structure. Do we need to have boron brought back from the surface, through subduction, to create blue diamond (oceanic plates are boron-rich)? To have the answer to this question, we would need to analyze the boron isotopes of blue diamonds. But the problem is that we already have some big trouble analyzing the main stable isotope of boron of diamond (11B about 80.1% of naturally occurring boron atoms), because of its low concentration (remember, less than 1ppm). Right now, there is no technique precise enough to be able to measure the other stable boron isotope (10B, about 19.9% of the naturally occurring  boron atoms) in an accurate enough way. Indeed, to determine if the boron is mantle or surface derived, it's the small variation between those two components that we need to measure... So... for the time being, this will remain a mystery... until we develop a new method on a new machine to conduct such experiment.

We can just propose that the most likely hypothesis is that boron comes from the surface, where it is the most abundant. That doesn't mean that the carbon has to have the same origin though...

We have now this blue rough diamond, bigger than 115 ct (115 ct was the weight after the first cut of the crystal, nobody knows what the weight of the rough crystal was) still deep inside the Earth. It is thanks to a very particular  type of magma that diamonds can be brought back to the surface, without transforming into graphite (graphite is the stable form of carbon at depths of less than 100-150km): the kimberlitic magma. This magma that has an "exotic" composition forms deep inside the Earth's mantle (deeper than 150km). On its way to the surface, the kimberlitic magma will tear off rocks that are on its way, including most likely some diamond-rich rocks if the eruption occurs in the middle of a craton. Another essential characteristic of this magma, is that it is really violent, and climbs back from the mantle to the Earth's surface at a speed of about 10 to 30 meter per second.... at least! Why is it essential for diamond? Well... if the magma is too slow, the diamond would have the time to transform into graphite...
But fortunately, our big blue got carried quickly to the surface of the Earth, and as the magma cooled down quickly, with diamonds as xenocrysts in it. In India, there were several episods of kimberlite emplacements, several of them occurred over 1Ga ago. Since then, the kimberlite had the time to be weathered, and diamonds washed off the kimberlite, to be deposited as alluvial minerals in a river.
Even if we don't know where the Hope original crystal was found, it is most likely that it was discovered in an alluvial deposit, as it was the only kind of mining that was done in the 17th century in India.

One possible model for the formation of the Hope Diamond. It might be that the origin of boron (and carbon) comes from surface material that were subducted and push down inside the Earth. How deep? We don't know. But we know that subducted material can go down below the transition zone (TZ), and can be brought back up to the surface.


  1. Thanks for this wonderful job.

  2. Just completed watching Mystery of the Hope Diamond - this is indeed an excellent added discussion. Look forward to learning of the results of your further research on the Hope in the next year or two. Thank you

  3. Very detailed.
    Did you mean to say that you think the donor element in the red phosperesence phenomenon is Nitrogen, or currently unidentified?

  4. The phosphorescence (blue & red) is currently not fully explained yet. It is most likely related to boron, interacting with something else (nitrogen, vacancy, both?). Stay tuned!

  5. thank you for this wonderfully insightful report . I had just caught the show on tv and was looking for what the unidentified element was. I will stay tuned

  6. I have a blue diamond with very strong red fluorescence to.