From !khure

Project B1: African Diamond Genesis and Craton Evolution: Exploring new ways to characterise conflict diamonds

French pi: P. Cartigny (with S. Gilder and C. Aubaud)
South African pi: M. de Wit (with S. Richardson and D. Bell)


Project Participants

  • France: Pierre Cartigny, IPG-Paris
  • South Africa: Maarten de Wit, Steve Richardson, Cape Town
  • Germany: Stuart Gilder, Munich; Michael Wiedenbeck, Potsdam
  • Exploration Industry: Mike de Wit, Hielke Jelsma, Kinshasa and Johannesburg.


Diamonds known as conflict diamonds, otherwise known as blood diamonds, originate from the war zones of Africa, and specifically from areas controlled by forces or factions opposed to legitimate and internationally recognized governments, and are used to fund military action in opposition to those governments, or in contravention of the decisions of the United Nations Security Council. On December 1, 2000 the UN General Assembly unanimously adopted a resolution defining the role of conflict diamonds with the intent of cutting-off the sources of funding for rebel forces and to help shorten the wars and prevent their recurrence through breaking the link between the illicit transaction of rough diamonds and armed conflict (from: David Cowley, 2007. What Are Conflict Diamonds?).

In May 2000, the diamond producing/trading industry took the first steps to develop a plan that could halt the trade of conflict diamond by establishing a way that diamonds could be certified, and created the Kimberley Process Certification Scheme (KPCS). Yet, governments want the KPCS to be monitored with more transparency and certainty to identify the place of origination of the diamonds. For example, Canada is a major diamond producer, and being member of the Kimberley Process Certification Scheme and (among others) is particularly concerned in regulating the legal diamond industry. Europe is the strongest partner of most African countries because they are a major donor or aiding development in third world countries and is thus concerned in solving conflicts in this part of the world. The fact that the European Commission assumes duties as Chair of the Kimberley Process for 2007 is a further reason why Europe is particularly concerned in monitoring the Kimberley Process Certification Scheme. The benefits to countries that put an end to trading in conflict diamonds are immense and it could mean better economic development and prosperity throughout Africa. Yet, resourceful and unscrupulous groups still manage to elude the legal barriers and still find ways of infiltrating the diamond centers of the world.

Although this list is not exhaustive and will likely change trough time, the main countries concerned so far by this resolution are: Angola, Sierra Leone, Liberia, Ivory Coast, Democratic Republic of Congo, Republic of Congo, Central African Republic. In terms of geology, these diamonds originate largely, but not exclusively, from two regions, namely the West and Central Africa Shields.

The question asked of scientists is either How can a conflict diamond be distinguished from a legitimate diamond; or [and this is not the same question] how can conflict diamonds be distinguished from legitimate diamonds?

Nobody can answer any of these two questions yet, as no one ever studied conflict diamonds scientifically in earnest; and in particular no one has yet mastered a non- destructive technique that has the potential to be fast enough to be enable examining bulk samples. Answering these questions require representative suites of conflict diamonds to be made available to scientists. It also require every legitimate mine to have its diamond production well characterised (which is not always the case).


Given that some conflict and legitimate diamonds are presently mined within a same region, the method to identify conflict diamonds must be sensitive on the local scale (i.e. the diamond mine) and not to a regional scale only (from one craton or shield to the other).

A large amount of data have been obtained on legitimate diamonds since the last 30 years and one can use these to discuss to potentially recognise diamonds from one legitimate diamond mine from another legitimate diamond mine.

Parameters that can be used to trace the origin of diamonds are:

  1. the physical characteristics (color, shape, resorption, surface features)
  2. the types (eclogitic/peridotitic ratios), nature of the inclusions (silicate, sulfide)
  3. the speciation of nitrogen and its aggregation state (FTIR)
  4. the stable isotope (C, N) compositions (mass spectrometry or SIMS)
  5. the radiogenic isotope compositions of diamond inclusions (TIMS)
  6. the trace element contents of diamond (LA-ICPMS, SIMS)
  7. the magnetic properties (cryomagnetic characterization)
Because diamond characteristics (size, shape, color, plastic deformation, isotope characteristics) are highly variable within a single diamond mine, distinguishing a conflict diamond be distinguished from a legitimate diamond is unlikely.

Of these other methods, 2-6 characterise chemical signatures, whilst the last (7) reflects a physical property of diamonds that has hitherto not been tested.

Using established techniques

Because each mine show some of the parameters 1 to 6 listed above to vary from one mine to the other (a single parameter is not pertinent enough), we can potentially provide a way to chemically identify conflict diamonds using a complete chemical diamond characterisation methodology.

As a first step, we propose to combine parameters 1 to 3 since these are relatively rapid and non-destructive techniques. These techniques will be applied to specimens from a variety of sources with a view of building up a global database of the above characteristics. For this purpose, we have already obtained a pilot sample of diamonds from the Mbuji Mayi in the DRC.

The study of carbon and nitrogen stable isotopes, nitrogen contents (Parameter 4) can be undertaken using (compared to SIMS measurements) well established, rather rapid, accurrate, yet destructive techniques. If successful in tracing the origin of diamonds, this destructive technique would not be an appropriate one anymore because then the sample must be preserved, but SIMS will be. Thus in the first instance, the method will allow to fastly expand the database to conflict and yet unstudied legitimate diamond mines.

The strengh of using parameters 1 to 4 is that these can already use the significant amount of data obtained during the last 30 yeras on legitimate worldwide diamond productions.

Developing new techniques

There are potentially several new non-destructive ways to characterize both the physical and chemical properties of diamonds that are briefly summerized below.

Chemical fingerprinting using trace elements

Ongoing with this work will be to explore new routes to reliable non-destructive chemical fingerprinting of diamonds of parameters 6. We believe this can be done as briefly described below:

Trace element geochemistry of diamond is relatively well known, and unfortunately to date the bulk or trace element compositions of diamonds are of little value to reveal their origin (Griffith et al., 2006). Moreover the geochemical methods are largely destructive, thus obliterating the diamond sample. Yet the geochemical signatures of individual diamonds, both in the form of trace element and isotopic compositions, have the potential for geographically localizing the origin of a crystal of unknown provenance. Through the use of modern micro analytical techniques it is possible to conduct such measurements in a nearly non-destructive fashion. For this strategy to be applied to the trade in conflict diamonds there are three geochemical questions which must be addressed:

  1. Are diamonds, in general, homogeneous in their compositions such that a single measurement from any surface is representative of an entire crystal?
  2. Do diamonds from a common source have one or more geochemical traits which are common to all crystals from that geographic unit (e.g., single diamond pipe or single cratonic region)?
  3. Do diamonds of different geographic origins differ significantly in one or more such characteristics?
If the answer to each of these three issues is “yes”, then the potential exists to constrain or even localize the source of individual crystals. The first step in this research initiative must therefore be to address all three of these issues.

Due to its high data acquisition rate, ability to measure many elements during a single analysis and its good to excellent limits of detection for many elements, the preferred analytical technique for this initial phase is laser ablation- inductively coupled plasma mass spectrometry (LA-ICPMS) trace element analysis. This approach has, however, two disadvantages: (1) the sensitivity is such that a single analysis will generally consume some 20,000 μm3 of sample during a single analysis and (2) the analytical precision obtained by LA-ICPMS is generally not adequate for isotope ratio measurements. Nonetheless, this analytical method will be able to answer, at least in the case of trace elements, whether geochemical fingerprinting can contribute towards localizing the geographic origin of a single crystal of unknown provenance.

Strategy, Preliminary Phase at GFZ-Potsdam, Germany

As a first step, simple test measurements by LA-ICPMS need to be conducted on diamond crystals in order to establish that no unanticipated issues impact the trace element determination in diamond (e.g., adequate optical coupling between laser and sample, stable signal from mass spectrometer, able to focus laser reproducibly on sample surface). If no technical problems are encountered then the next step would be to conduct measurements on a collection of crystals of know origins. (1) Multiple analyses need to be conducted on each crystal to asses whether crystal heterogeneity is a common problem. (2) Two or more crystals from the same pipe need to be analysed to asses if they form a homogeneous population for at least some geochemical properties. (3) Samples from multiple diamond pipes from the same geographic / cratonic region need to be analyzed to assess whether any trace element characteristics are common from across an entire province. (4) Crystals from two or more unrelated geographic sources need to be measured and these data need to be assessed for whether any parameters can distinguish between the different groups. Thus, the minimum study material for this initial phase of research would be approximately:

  • 2 random crystals from diamond source “A” of geographic domain “X”
  • 2 random crystals from diamond source “B” of geographic domain “X”
  • 2 random crystals from diamond source “C” of geographic domain “Y”
  • 2 random crystals from diamond source “D” of geographic domain “Y”

Each of the crystals would need to be measured twice by LA-ICPMS for a broad spectrum of trace elements. The data from each duplicate would be used to assess whether trace element zoning is a common or rare feature in diamonds. If intra-crystal variations in concentrations are rare then a principal component analysis of the data would be necessary to asses which element signatures (concentrations or ratios), if any, are characteristic of a given geographic domain and which are also distinctive between domains.</p>

If the results of the LA-ICPMS work indicate that trace elements offer a potential means of constraining the geographic origin of a crystal of unknown provenance then similar work using 13C/12C determinations by secondary ion mass spectrometry (SIMS) should also be undertaken and these data should be assessed along a similar line. In the mean time, conventionnal destructive methods at IPG-Paris will be used to characterise diamonds.

Strategy, Rough Outline of Advanced Phases at GFZ-Potsdam, Germany

Assuming that the preliminary phase of this work is consistent with the use of trace element / isotope signatures for constraining geographic origin then a further work would include:

  • Expanding the data base to cover all geographic domains of economic importance and assessing these data to establish which geochemical characteristics are distinctive / unique to which deposits.
  • Conducting a set of blind tests to demonstrate the reliability of this approach.
  • Developing and certifying a suite of diamond reference materials for the absolute determination of the key characteristics.
  • Developing and validating SIMS-based analytical protocols for routine use.
If this approach is to become a standard tool for investigating the trade in conflict diamonds then SIMS will need to become the primary analytical technique. This is because SIMS offers a sensitivity which is a factor of 10 to 50 higher than that of LA-ICPMS, resulting in a concomitant reduction in the amount of sample being consumed. For example, in the case of 13C/12C determinations on diamond the crater produced during SIMS analysis will be on the order of 10μm in diameter and <1μm in depth. In the case of trace element analyses at sub-ppm concentration levels the volume consumed during a SIMS analysis may be of the order of 5 to10 times larger. The disadvantages of SIMS are that it is a significantly slower analytical method and it is not well suited for screening large numbers of elements. SIMS also requires a well polished sample surface on which to conduct the analysis and the crystals, unlike in the case of LA-ICPMS analyses, must be embedded in some form of medium to produce a large (cm-sized), flat surface.

Physical fingerprinting: Can diamonds be characterized magnetically?

We propose to test a new method with the intention to characterise diamonds non destructively using their physical properties tied to a magnetic „fingerprint‟, described briefly below. Since this is a new untried method we describe our strategy briefly below. This work will be carried in the Geophysics Department at the Maximillian University, Munich, Germany, under the leadership of Prof S Gilder.

A new non-destructive route to test magnetic properties of diamonds
Diamond geochemistry is relatively well known, and unfortunately the bulk or trace element compositions of diamonds are of little value to reveal their origin. Geochemical methods are largely destructive, thus obliterating the diamond sample. Moreover, bulk geochemical analyses may be blind to the growth mechanism or strain state acting when the diamonds formed. Such processes are fundamentally important in controlling the way in which inclusions grow and are oriented with respect to the diamond‟s crystallography. Thus, to potentially learn something about the genesis of diamonds, a technique must be sought that is non-destructive and can yield information concerning their strain state and inclusion geometry. Magnetism holds much promise in these respects knowing that a small proportion of the inclusions are magnetic phases such as iron sulfides.

We propose to study the magnetic signatures of individual diamonds in two principal ways: (1) to characterize their low temperature (down to liquid helium, or 4 K) behavior and (2) to measure their magnetic anisotropy. A well-known example of the former is the Verwey transition in magnetite that occurs at around 120 K. The exact temperature and shape of the Verwey transition in magnetic moment vs. temperature space depend on the oxidation state, grain size, composition and internal stress of the magnetite. Oxidation smears out the transition while small additions of aluminum, titanium, etc., will lower the temperature of the transition. When cooled through the Verwey transition in the presence or absence of an external field, large, multi-domain magnetite grains exhibit different responses than small single domain grains. Thus, much can be learned from the response of the magnetic moment at low-temperatures. Even a few nanometer-sized grains produce enough signal to be measurable. Note that the magnetic properties of iron sulfides such as pyrrhotite also undergo important changes at low temperatures.

Solid inclusions are virtually never spherical in diamonds. This means that the solid inclusions inherently have a shape-anisotropy. Moreover, the inclusions may be aligned according to the growth pattern of the diamond, such as along zones or certain crystallographic axes. A quick and non-invasive way to quantify the geometrical alignment of an assemblage of magnetic inclusions is to study the magnetic anisotropy of the diamonds. This can be done in two ways. One is via the anisotropy of magnetic susceptibility. This method yields the eigenvalues of the susceptibility tensor and thus defines the shape of the anisotropy ellipsoid. Another way is via the anisotropy of magnetic remanence. This method is much more time consuming (one and a half hours/sample), but potentially more revealing because the technique can be applied under different imposed field conditions. What this means is that the anisotropies of specific magnetic species and the specific grain size distribution of those species can be analyzed. If a diamond underwent a multiple growth history, with each growth period having a slightly different magnetic composition or characteristic grain size, these differences can be studied individually and thus the generations of growth can be better understood.

The paleomagnetic laboratory at LMU is currently equipped with a Kappabridge to measure the anisotropy of magnetic susceptibility. We have a 2G Inc. three axis superconducting magnetometer together with coil systems that can measure the anisotropy of magnetic remanence. Our proposal also coincides with the arrival of a second 2G Inc. three axis superconducting magnetometer to LMU. This magnetometer will be fit with a low temperature insert and a pulse magnetizer that can measure the full magnetic vector of materials to 4 K.

We have already obtained a pilot sample of 34 diamonds from an operating mine in the Central African Republic (CAR). Work on this set of diamonds is ongoing. Once magnetically tested, this sample will be distributed to the group in IPGP and GFZ-Potsdam for further chemical analyses.

Capacity Building

South African PhD candidates will be linked to the projects in Munich and Potsdam at the earliest opportunity in 2008.

Diamond supplies

Crucial to all the stated projects is a steady supply of diamonds form different mines and regions. To facilitate this we have established relationships with offices of two diamond exploration companies based in Kinshaha (Dr Mike de Wit) and Johannesburg (Dr Hielke Jelsma). These contacts have assured their interest in the project, and will endeavour to provide a unspecified number of diamonds, and will also help in persuading other active companies into partnership with this !Khure program.