Feature

Diamond Research Gives Clues to the Formation of the Continents


Placeholder Alt Text
This rough type Ib diamond, from Zimmi in West Africa, near the Sierra Leone-Liberia border, contains a sulphide inclusion with compositions that give clues to how the West African continent formed. Photo by Karen Smit/GIA.

This is a summary of the article "Sulfur Isotopes in Diamonds Reveal Differences in Continent Construction" by Karen V. Smit, Steven B. Shirey, Erik H. Hauri and Richard A. Stern published on 26 April 2019, in Science.

Diamonds, one of the most valuable gemstones, are also very valuable to geoscientists. Mineral inclusions in diamonds are the most direct samples we have from the inaccessible depths of earth. Inclusions in diamonds have given geoscientists information about water in the earth’s interior, the mineralogy of the deep earth and metallic phases in the deep earth. Also, because there is no direct way to determine a diamond’s age from the diamond itself, mineral inclusions trapped within diamonds provide the only way to date diamonds. So, although inclusions in diamonds are often considered to be undesirable in the gem trade, they are extremely valuable scientific samples.

In our study, we measured the sulphur and rhenium-osmium isotopes in sulphide inclusions in diamonds from the Zimmi region of Sierra Leone. The sulphides are tiny minerals, often between 100 and 300 microns across, trapped in the diamonds during growth.

We found that the sulphides recorded two episodes of subduction in the West African continent. Subduction is when the oceanic crust is thrust under another tectonic plate during a collision and into the deep earth. The first subduction event recorded by Zimmi sulphides occurred around three billion years ago and the second was around 650 million years ago.
 

A silver-grey sulphide inclusion in a diamond.
Sulphide inclusions still trapped in the Zimmi diamonds. Sulphides are silver-grey in appearance and are surrounded by blackened fracture systems. These fractures develop during kimberlite ascent because the sulphides expand more than the diamond. Photo by Karen Smit/GIA

3-billion-year-old subduction

The sulphides have isotopic compositions that indicate the sulphur was cycled through the ancient atmosphere prior to the rise of oxygen 2.5 – 2.3 billion years ago (Farquhar et al., 2001). This is indicated by mass-independently fractionated (MIF) sulphur isotopes. Any modern sulphur (or sulphur that did not cycle through the ancient atmosphere) will not have these MIF isotopes1. The presence of MIF sulphur in these sulphide inclusions indicates that they have a surficial origin in the earth’s ancient atmosphere. The sulphides were probably emplaced into the deep earth around 3 billion years ago – at a time that has been well established for subduction and the incorporation of oceanic crust material into the mantle2,3.

650-million-year-old subduction

We also measured the rhenium-osmium isotopes in these sulphide inclusions. Rhenium-osmium is the most widely used technique to date the time of diamond formation4. Zimmi diamonds were found to have 650-million-year-old ages5, an age that overlaps with subduction and collisional mountain building in the region, between 700 and 550 million years ago. Subduction of the oceanic crust and its subsequent dehydration would have introduced carbon-bearing fluids into the deep earth for diamond formation6.

Why do we find this interesting?

Earth’s oldest continents (called cratons) are stabilised by lithospheric mantle keels. The stability of earth’s continents in the face of destructive tectonic activity provided the essential geological backdrop for the emergence of life on our planet. Since this is the only tectonically active, rocky planet that we know, understanding the geology of how our continents formed is a crucial part of discerning what makes earth habitable.

The stability of cratons depends on the underlying mantle keels that are around 150–200 km thick. The processes for how these mantle keels form are still being debated, and there are several different theories for their origin. Some of the models for craton formation involve subduction-style plate tectonics where plates are subducted into the deep earth and stabilise the cratonic keels. Other models do not invoke subduction, and instead, require deeper mantle processes such as melting in mantle plumes or melting at oceanic plateaus.

Luckily these mantle keels have the ideal conditions for diamond formation. The majority of natural diamonds form in these cratonic mantle keels. Diamonds become important samples that can be used to investigate how the stabilising keels below the oldest continents are formed.

Sulphur isotopes in diamonds, combined with their Re-Os ages, can be used to track multiple subduction events during craton growth, even those separated by billions of years. Subduction processes were essential to the growth and modification of the West African craton over a period of 2 billion years.

We compared our results to diamonds from southern Africa and northern Canada. We found that this combined isotopic approach also reveals differences in craton construction worldwide. Similar to the Zimmi diamonds, diamonds from the Jwaneng and Orapa mines in southern Africa also contain MIF sulphur1,7. This indicates that subduction was also an important process in the construction of the cratonic mantle in southern Africa.

But diamonds mined in northern Canada do not show the same sulphur chemistry. Diamonds from the Ekati mine have 3.5-billion-year-old rhenium-osmium ages and do not contain MIF sulphur8,9. This means that the mantle keel in this region originated in some way that did not incorporate surface material. The sulphur in the Canadian diamonds does not tell us how the mantle keel formed, only how it was not formed.

Our work shows that sulphide inclusions in diamonds are a powerful tool to investigate craton construction processes.

View of a sulphide inclusion under an electron microscope.
Sulphide inclusion that has been broken out of a diamond and imaged in a scanning electron microscope. This sulphide has cubo-octahedral morphology and trigons on its surface. Both these features were imposed on the sulphide by the diamond. Image by Karen Smit/GIA


How did we do this?

To characterise the diamonds and image the inclusions, we first laser cut and polished double-sided plates of the diamonds.

Rhenium-osmium isotopes

We broke the diamond plates using a small hammer and a steel cracker to release the sulphide inclusions. The inclusions are tiny and weighed between 3 and 162 micrograms. After breaking each inclusion out of the diamond, we imaged it and analysed its major element composition using a scanning electron microscope.

In a clean laboratory, we then dissolved the sulphide in some acid along with a known amount of tracer solution known as an isotopic spike. Chemical procedures were then carried out to separate the rhenium and osmium into different acidic solutions. The solutions that contained rhenium and osmium were each dried out on a hotplate.

The dry salt of osmium was placed on a metallic filament, which was then placed in a thermal ionisation mass spectrometer (TIMS). The filament was heated to produce ions that were then accelerated through the instrument and the different osmium isotopes were measured on a very sensitive detector.

The dried salt of rhenium was taken up into a dilute acid solution. That solution was introduced into a multi-collector inductively coupled mass spectrometer (MC-ICP-MS). The solution was ionised in a plasma and those ions were accelerated through the instrument, where the two different rhenium isotopes were measured simultaneously on very sensitive detectors.

The amount of rhenium and osmium that we measured in the sulphides is in the femtogram to picogram range - where 1 femtogram is 10-15 g or 1 part per quadrillion (ppq) and 1 picogram is 10-12 g or 1 part per trillion (ppt). Because we were measuring such tiny amounts, any small amount of contamination could have ruined the sample. This is why it was essential to carry out this work in a clean laboratory with dedicated supplies for diamond inclusion work.

Rhenium-osmium analyses were carried out using chemistry labs and a Thermo-Fisher Triton instrument at the Department of Terrestrial Magnetism at the Carnegie Institution for Science.

Sulphur isotopes

Sulphide inclusions were not broken out of the diamonds for sulphur isotope measurements. Instead, we polished the diamond plates down further with a diamond scaife to expose the sulphide inclusions. It took a few hours to a day to expose each inclusion.

Diamond plates containing the exposed sulphide inclusions were placed in a secondary ion mass spectrometer (SIMS). The samples were bombarded with a caesium ion beam to create small pits in the sulphides. The material released from the sulphides was then accelerated through the instrument and the four different sulphur isotopes were measured simultaneously on very sensitive detectors.

Sulphur isotope measurements were carried out using a Cameca IMS 1280 instrument at the University of Alberta. Preliminary measurements were carried out using a Cameca NanoSIMS at the Department of Terrestrial Magnetism at the Carnegie Institution for Science.

About the diamonds themselves

The diamonds used in this study are from the Zimmi alluvial deposit in Sierra Leone. The Zimmi mining area is located south of the town Zimmi, near the Liberia-Sierra Leone border. The Zimmi locality is known for producing yellow diamonds with abundant sulphide inclusions.

The Zimmi diamonds have nitrogen impurities in the rare isolated form that classifies them as Type Ib diamonds. Type Ib diamonds are exceptionally rare among natural diamonds and account for less than 0.1% of worldwide natural diamonds.

The Authors

Karen V. Smit has been a research scientist at the Gemological Institute of America (GIA) since 2014. She is interested in using diamonds and their mantle host rocks to understand diamond formation processes in the deep earth. Her research focuses on the origin of the earth’s continents, how plate tectonic processes influence diamond stability and, most recently, on using the spectroscopic features of natural diamonds to distinguish them from lab-grown and treated diamonds.

Steven B. Shirey has been a staff member at the Department of Terrestrial Magnetism of the Carnegie Institution for Science for 34 years. Shirey's research focuses on the evolution of the earth's continents, especially as a function of mantle evolution. His interest in diamonds began more than 20 years ago as a way to examine continent formation and plate tectonics from the deepest possible perspective in the mantle.

Erik H. Hauri (now deceased) was a staff member at the Department of Terrestrial Magnetism of the Carnegie Institution for Science for 24 years. He passed away in 2018. Hauri was interested in the geochemical cycles of volatiles on the earth and moon and their relationship to planetary dynamics. Hauri's remarkable expertise in secondary ion mass spectrometry led to breakthroughs in understanding the water content of the moon, and the analysis of stable isotopes in diamonds and their mineral inclusions.

Richard A. Stern is a research scientist and ion probe facility manager at the University of Alberta. His research over the last 25 years has focused on the development and application of secondary ion mass spectrometry to a broad range of topics in geochemistry and geochronology, most recently emphasising light stable isotopes.

 

References

1. Farquhar et al., 2002
https://science.sciencemag.org/content/298/5602/2369

2. Barth et al., 2002
https://www.sciencedirect.com/science/article/pii/S0301926802001110

3. Aulbach et al., 2019
https://academic.oup.com/petrology/advance-article-abstract/doi/10.1093/petrology/egz011/5364029

4. Pearson et al., 1998
https://www.sciencedirect.com/science/article/pii/S0012821X98000922

5. Smit et al., 2016
https://www.sciencedirect.com/science/article/pii/S0301926816300882

6. Smit et al., 2019
https://www.sciencedirect.com/science/article/abs/pii/S0009254119301895

7. Thomassot et al., 2009
https://www.sciencedirect.com/science/article/pii/S0012821X09001447?via%3Dihub

8. Westerlund et al., 2006
https://link.springer.com/article/10.1007%2Fs00410-006-0101-8

9. Cartigny et al., 2009
https://www.sciencedirect.com/science/article/pii/S0024493709002497?via%3Dihub

Karen V. Smit is a research scientist at GIA.