Original paper

Morphology of diamonds as a possible indicator of their genesis

Evdokimov, M. D.; Ladygina, M. Y.; Nesterov, A. R.

Neues Jahrbuch für Mineralogie - Abhandlungen Band 176 Heft 2 (2001), p. 153 - 177

35 references

published: Jun 22, 2001

BibTeX file

ArtNo. ESP154017602003, Price: 29.00 €

Download preview PDF Buy as PDF


Relationships between the morphology of diamond crystals and conditions of their crystallization have been the subject of intensive scientific debate for over two centuries. The hypothesis ascribing formation of rounded diamonds to the processes of crystal growth was initially suggested by Hauy (1801), and further developed by Koksharov (1869). Rose & Sadebeck (1877) and, subsequently, Eremeev (1896) modified this hypothesis in accordance with the contemporary level of mineralogical knowledge. Fersmann & Goldschmidt (1911) proposed an alternative concept in which formation of rounded crystal shapes was attributed primarily to dissolution (corrosion) of diamonds. Genetic implications of the crystal morphology of diamonds were also discussed by Williams (1932), Shafranovskii (1948) and Kukharenko (1954, 1955). Interestingly, the advocates of both hypotheses suggested that diamond crystals grew in a liquid, and one of the key elements in their debate was a role of convection in the processes of diamond growth and dissolution (Ansheles 1954). To our knowledge, none of the studies published prior to the 1950-es attempted to explain the occurrence of plane-faced diamonds in eclogite xenoliths from South-African kimberlite diatremes. Also, such "unorthodox" hypotheses as growth of diamonds in a plastic medium (Sutton 1928) were generally neglected. The early discussions on the morphology of diamond led to the conclusion that macroscopic and optical studies could not provide sufficient criteria for distinction between growth- and dissolution-induced morphological features. For instance, polygonal pits and microlamellar sculpture on {111} faces could result from either of these processes. Perhaps, the most disputable aspect of diamond morphology is the origin of rounded crystal forms, i. e. so-called dodecahedroids, octahedroids and hexahedroids (sensu Orlov 1977). Estimations of Kukharenko (1954) indicate that more than 75 % of all diamond crystals found in India, South Africa and South America are curve-faced. However, in some kimberlite provinces (e. g. South Yakutia), "normal" octahedral crystals strongly predominate over curve-faced diamonds. If we assume that rounded shapes are characteristic of plutonic crystallization conditions, the origin of plane-faced octahedral crystals becomes unclear. It is noteworthy that views of some scholars engaged in the discussion of diamond morphology were often misconstrued by their opponents, as well as sometimes by their disciples. Professor A. A. Kukharenko was traditionally thought to be an adherent of the dissolution hypothesis. However, from his works, it is obvious that he recognized the importance of both dissolution and growth for crystallization of diamond in geological systems. Shortly before the publication of his monograph on the morphology of diamonds from the Urals, he wrote (Kukharenko 1954, translation ours): "We distinguish: I. Growth forms 1. Growth forms related to rapid crystallization from highly supersaturated systems: octahedra with coarse lamellar sculpture, clusters of octahedra, skeletal crystals; 2. Growth forms pertaining to lower degrees of supersaturation: regular octahedra. Both these types are characteristic of relatively rapid crystallization in near-surface conditions. 3. Forms pertaining to slow growth from systems approaching saturation: combinatory polyhedra such as octahedroids. Crystals of type 3 are characteristic of slowly cooling plutonic bodies. II. Dissolution forms 1. Forms resulting from slow dissolution in weakly undersaturated systems: octahedroids. 2. Ultimate forms of extensive dissolution in undersaturated systems near the equilibrium: dodecahedroids. Both these types seem to be typical mainly for plutonic bodies. 3. Forms related to rapid dissolution and combustion of diamonds: trisoctahedra with curved and coarsely sculpted faces. 4. Corrosion forms arising from extensive dissolution and burning: irregularly shaped grains with porous and cavernous surfaces. III. Regeneration forms: rounded crystals such as dodecahedroids with tile-like sculpture on their faces, most conspicuous near the edges and vertices of crystals (about g3)." The above classification was further developed by Kukharenko in his monograph. It was based on the descriptions, goniometric measurements and sketches of more than 200 crystals presented in his atlas. Here, we should also mention that drawings and gnomonic projections of 289 crystals studied by Fersmann & Goldschmidt (1911) are presented as an attachment to their monograph. Meticulous examination of numerous crystals led Kukharenko to the conclusion that it is competition of growth and dissolution processes that ultimately defines the shape of diamond crystals. Note that, almost five decades ago, this author could not be aware of several factors serving as crucial elements of our present-day views on the genesis of diamonds. One of the most important among these factors is the nature of inclusions in diamonds, i.e. their morphology, distribution and chemical composition. It is well known that such inclusions can be classed into two major groups, i.e. those derived respectively from eclogitic or ultramafic assemblages. The eclogitic inclusions are represented predominantly by rounded grains of silicate minerals (e.g. pyrope-almandine garnets, Fe-enriched olivine) and such oxide phases as coesite and rutile. Inclusions derived from ultramafic rocks are represented by Cr-rich pyrope, diopside, enstatite and spinels, as well as magnesian ilmenite and sulfides (Sobolev 1974, Shemanina et al. 1980). Another factor crucially important for the genetic interpretation of diamond morphology is the ancient age of diamonds, approaching 2,500 M.a. (Mal'kov 1980). Thirdly, it has been experimentally proven that diamonds may crystallize under metastable conditions, e.g. from a gaseous medium. These experimental data were used by Nikishov (1984) in his model of kimberlite formation. Also important are the observations of zoning in diamonds (Beskrovanov 1992). In the present work, we attempted to establish differences between the morphology of growth and dissolution features on a submicroscopic level. In our discussion of crystal morphology and genesis, we adhere to the conventional view that the evolutionary history of diamonds can be described in terms of three successive stages: (i) eclogitic (nucleation and early growth), (ii) asthenospheric (generation of kimberlitic magmas), and (iii) kimberlitic. Evidently, the morphology of diamonds in placer deposits is also affected by such supergene factors as mechanical wear and chemical corrosion. However, in the present work, we shall discuss exclusively the role of hypogene factors.


macroscopiccrystallizationdiamondsSouth-African kimberlitepolyhedraxenolithsequilibriumgeneration of kimberlitic magmas