Foreword

Before narrating an entertaining, interesting and scientifically important story of diamonds, I would like to devote a few lines to the people already gone but no less respected and loved by me. One of them was my father. Not only did he give me the instrument for learning the laws of the Universe but inculcated in me a love and an inquisitiveness toward natural sciences. Another one was looking after me for fully a half of my scientific life and directing my studies in the channel of the unknown. Both were great scientists of their time, both made a substantial contribution to the world of science. It is due to them that I eventually became a scientist and was able to write this book.

The origin of diamonds is closely tied to Earth’s general planetary evolution. That is why the substantiation of the proposed diamond and diamond-bearing rocks’ emergence mechanisms is conducted in the context of a modern concept of Earth’s global evolution.

Based on this concept, the book describes the diamond formation exogenous mechanism, according to which diamond-bearing rock melts emerge due to remelting of pelagic oceanic deposits pulled in to great depths through ancient plate subduction zones under the continents. However, only heavy iron ore deposits whose density exceeded average density of continental lithospheric plates (around 3.2 g/cm³) might have immersed to a great depth up to the base of these plates. It is known from the geological record that large volumes of such formations had been deposited only at the end Archean and in the second half of Early Proterozoic. In Archaean, due to high heat flows, the continental plate thickness (together with the continental crust) had not exceeded 60–80 km. In Proterozoic, Earth’s tectonic activity had sharply declined and its further evolution had been proceeding under the lithospheric plate tectonics laws. At that time, plate subduction zones had emerged and the continental lithosphere (together with Earth’s crust) thickness by the end Early Proterozoic had rapidly grown to 200–250 km. Exactly for this reason, melts of the depth diamond-bearing rocks (the kimberlites and lamproites) might have emerged only in the second half of Early Proterozoic about 2.2–1.9 BYa. The fact of the kimberlites, lamproites and their affine rocks emergence from ancient iron rich oceanic deposits is evidenced by their chemical composition. The same is supported by the composition of eclogite xenoliths practically equivalent to the composition of oceanic tholeiite basalts melted out at shallow depths. At that, the kimberlites and carbonatites had formed most likely due to a submergence to great depths and remelting of Earth’s tropical zone carbonate deposits and the lamproites, due to remelting at similar depths of boreal and circum-polar areas’ clayey pelagic deposits. The origin of apatite-bearing sienites in the Khibin and similar massifs is explained by pulling oceanic upwelling zone phosphorites into subduction zones and of diamonds, by carbon reduction, at the expense of exothermal reactions between carbon dioxide gas released at thermal dissociation of the depositional carbonates and organic origin hydrocarbons. It follows thence that the entire carbon of the diamonds had only exogenous origin. The same is testified by isotopic data and by gas-liquid inclusions in this mysterious mineral.

Based on the described formation mechanism of diamond-bearing and their affine rocks, the structure is reviewed of the South African, Yakut and some other diamond-bearing provinces. A positive forecast is made of the diamond-bearing potential in the northeastern Baltic shield and northern Russia platform. The exploration criteria of these minerals in other world regions are formulated.

My special gratitude goes to Dr. Michael V. Gorfunkel who inspired me to write this book and was its guardian angel without whom its coming-into-being would not have been possible.

Introduction

Many scientists traditionally believe that the ore (and not only ore) matter of most endogenous economic minerals comes directly from the mantle or through the mantle matter differentiation and invasion of differentiated magmas in the crust together with water fluids rising from Earth’s depth. However, this assumption is right only in part as the entire matter of the continental crust, including ore elements, had been indeed released in the past from the mantle along with the other rock-forming oxides. In substance, however, this assumption is the “path of least resistance” as it allows us to hide our lack of knowledge of the real mechanisms of local crust enrichment with trace elements in a “black box” of the mantle and to substitute one complex problem with another one, no less complex. Indeed, the entire complexity of the classical approach to explaining the formation causes of local ore and other trace element accumulations in Earth’s crust is in that the concentration of most of them in the mantle is disappearingly low, whereas in commercial deposits it is relatively high and reaches sometimes top-cut grade value. For instance, gold and uranium concentration in the present-day mantle is on the order of 10⁻⁹; mercury and thorium 10⁻⁸; silver, tantalum, tungsten, platinum and lead 10⁻⁷; lithium, niobium, molybdenum and tin 10⁻⁶, etc., whereas in commercial deposits the concentration of these rare elements may rise to fractions and even whole percentages.

The current geodynamic concepts maintain that the entire matter in the mantle (both upper and lower) had been well stirred over 4 BY of Earth’s tectonic activity by the convection flows and must have a uniform composition. That is why there is no hope today of existence in the mantle of local irregularities with elevated contents of both ore and volatile elements – strong mineralizers such as water, carbon dioxide gas, halogens, etc. Whereas the juvenile (chemically bonded) water content in the quenching glass of fresh oceanic basalts usually does not exceed 0.05%.

Water drastically decreases the melting temperature of silicates, thus, at basalt melt-out it should mostly concentrate exactly in basalt melts. It follows thence that water concentration in the mantle at least does not exceed 5·10⁻⁴. Therefore, the mantle is practically “dry”. Any fluid flows in this geosphere (outside the plate subduction zones) capable of bringing in the crust noticeable amounts of ore, lithophilic and other trace elements are out of the question.

What, then, is an explanation of significant, sometimes outright top-cut grade contents of trace and rare elements in economic mineral deposits? Apparently, only the superposition of two processes, gradual accumulation of ore elements in Earth’s crust and their subsequent concentration in the local volumes of deposits due to the secondary processing (recycling) of Earth’s crust matter. The former process is totally dependent on Earth’s tectonic activity, i.e., on the release rate in its subsurface of endogenous energy. The latter process is connected with the effect of exogenous causes of Earth’s crust rocks destruction and weathering, with the deposition of crust rocks destruction products and often with their new remelting. The main factors controlling the exogenous processes on Earth, beside the Sun radiation, are the evolution of hydrosphere (including the oceans) and atmosphere, both substantially affecting Earth’s climates. Only a simultaneous and total consideration of all these endogenous and exogenous factors may enable us to understand the nature of economic minerals’ accumulation in Earth’s crust.

Everything aforementioned is fully applied to the origins of diamond-bearing kimberlites, lamproites and their affine rocks of the carbonatite and alkali-ultramafic series. Indeed, on the one hand all these rocks are of a clearly depth origin; it is certain that their ancestral magmas had risen from the upper mantle levels. On the other hand, they all are sharply enriched in volatile and lithophilic elements and sometimes even contain notable amounts of complex hydrocarbons (for instance, the Khibin massif of apatite-bearing nephelinites on the Baltic shield). All of this forces us to reconsider long-held concepts of the origin of endogenous commercial minerals, in particular, diamonds. A detailed familiarization with traditional views of the origins of most endogenous commercial mineral deposits as well as of depth diamond-bearing rocks shows that within the framework of “classical” geological approach this problem is conceptually irresolvable. It is necessary to involve in its resolution new ideas based on the modern geological theory.

A theory of the planetary or global Earth’s evolution created and developed by Professor O.G. Sorokhtin was selected as such a theoretical foundation. A component part of this theory is lithospheric plate tectonics; the theory describes general patterns of Earth’s evolution as a planet. It is based on the energy approach, according to which its evolution occurs due to the endogenous processes lowering to the greatest extent the potential energy of Earth itself and of the Earth – Moon system. As its development occurs with the loss of the endogenous heat, the evolution process is irreversible.

The Earth’s evolution theory is based on two departure suppositions. First, it is assumed that Earth had emerged at the expense of a homogenous accretion of a cold proto-planetary gas-dust cloud, and second, that Earth’s core is composed mostly of an alloy of iron and its oxide. At that, it is believed that we know the composition of Earth’s crust and the mantle. Both these assumptions as of today are most substantiated and accepted by most geologists and geophysicists. It appears that these two assumptions are quite sufficient for the construction of a self-consistent Earth’s evolution theory because they include the entire needed information about its primordial structure (a uniform and cold planet) and the reserve of its internal energy.

All these enabled the consideration of the diamond-bearing rock origin issue at a new angle and an attempt at a forecast of a potential diamond-bearing of some world regions. That is exactly why the description of the origin of diamond-bearing kimberlites, their affine rocks and diamonds themselves is accompanied by a more detailed review of major Earth’s evolution theory positions.

The issues of diamonds and diamond-bearing kimberlites’ origin are part of a greater natural phenomenon which is customary to call carbon global cycle. It needs to be noted that as carbon global cycle is traditionally understood its crustal-atmospheric interaction whereas the crustal-mantle carbon transfer is practically unstudied.

Earth’s crust carbonates tie up about 3.91·10²³ g CO₂ and about 1.95·10²² g of organic carbon (C_org). Its substantial fraction settles in the form of deposits on the seafloor and on continental slopes. Further on, the deposits are pulled in subduction zones and transferred by the mantle convection in Earth’s divergent areas. For instance, in rift zones is observed a wide range of hydrocarbon gases’ emanations, methane (CH₄) and ethane (C₂H₆) up to propane (C₃H₈) and butane (C₄H₁₀). In and of themselves, complex hydrocarbons in a free state are unstable under high PT conditions and tend to decompose into simpler ones, up to methane (CH₄). This is a proof that the generation of the listed compounds occurs in near-surface and low-temperature environments of the medium and not due to their release from the deep mantle. A question arises: how do they get there? A study of this issue showed that carbon transfer from the subduction zones into the rift systems might occur through the transformation of its phase states and the formation of specific compounds (metal carbides) in depth of the mantle. As a result of the crustal-mantle cycle manifestation, carbon is subjected sequentially to a multistage transformation from chemogenous state into biogenic and back as well as to the submersion in the mantle on the levels of its convective stirring and to release on the surface through the rift zones. There, the metal carbide decomposition and the formation of a wide range of hydrocarbon gases occur. The process of carbon depth transfer process described in the book is closely tied with its crustal-atmospheric branch as the primary supplier of carbon is carbon dioxide gas and the organic matter. Together, these two branches form the global carbon cycle in nature. Geometrically, it may be imaged as the number 8, not 0 as is currently accepted.

This research was funded by the state assignment of IO RAS, theme 0149-2019-0005.

Chapter 1
Major Parameters of Diamond-Bearing and Affine Rocks

The bed-rocks of diamond-bearing rocks are, as is well known, kimberlites and lamproites. Those are depth magmatic rocks usually encountered only on the ancient continental platforms and forming subvolcanic bodies – blowpipes (diatremes) or magma-bringing dykes. Affine but somewhat less depth rocks are carbonatites and alkali-ultramafic rocks of a quite wide composition range. However, their common feature is low silica contents and relatively elevated concentration of magnesium. This enables attributing all these formations with ultramafic rocks. As opposed to the classical ultramafic rocks of the peridotite series mantle origin, kimberlites, lamproites and alkali-ultramafic rocks are enriched in the titanium, alkalis (first of all, potassium), phosphorus, rare lithophilic and volatile elements including water and carbon dioxide (especially carbonatites).

The chemistry and geochemistry of diamond-bearing kimberlites and lamproites is described in many articles and monographs [1–4]. There are also numerous descriptions of the alkali-ultramafic rocks [5–7] and carbonatites [8–10]. For this reason, there is no need to analyze here in detail the chemical composition of these exotic rocks. Attention must be paid only to the geochemical specifics of kimberlites as most typical representatives of this rock class.

Analyzing specifics of kimberlites’ chemical composition, I.P. Ilunin with colleagues [3] noted that the SiO₂/MgO and MgO/FeO ratios in kimberlites correspond with the dunite and peridotite values whereas the Al₂O₃ and Na₂O concentrations are notably lower than in basalts. On the other hand, in contents of some rare elements the kimberlites are close to alkaline basalts. Nevertheless, it is emphasized that no mixing of the peridotites with alkali-basaltoid matter allows to come with the kimberlitic composition. That is because any notable addition of a basaltoid matter to the peridotite will unavoidably result in an increase of the silica, aluminum and sodium concentrations, and at insignificant addition will not occur contents of rare elements typical for the kimberlites.

Also important is that compared to the mantle ultramafic rocks (peridotite and lherzolite), kimberlites are substantially enriched in titanium, potassium and phosphorus. At that, usually enrichment of the kimberlites with rare elements correlates with elevated phosphorus concentrations. To an even greater extent such correlation shows up in carbonatites [3].

The kimberlite geochemistry specifics could have been visually manifested at their comparison with samples of undepleted mantle matter. However, to our great regret, we are never dealing with fresh samples of the mantle rocks wherein the rare elements’ contents would have been preserved undistorted. Instability of the dispersed elements’ direct determinations in the mantle rocks brought in to Earth’s surface is caused by the fact that in the process they practically always experienced a very strong influence from metamorphogenic factors which substantially distorts their primary composition in the rare element domain. Thus, if we use ultramafic rock samples from ophiolite nappes, we should take into account that their matter was at least twice subjected to hydrothermal actions. The first time, at the time of the oceanic crust formation due to its hydration by the oceanic water saturated with alkalis and other easily dissolvable elements. The second time, in the process of this crust obduction (the ophiolite nappe) on continental margins due to the action of overheated and mineralized water coming from plate subduction zones. No less distorted turns out the primary composition of dispersed elements in ultramafic xenoliths within the kimberlites themselves. It is caused by two reasons. First, it is quite likely that these xenoliths are fragments of the ancient oceanic crust pulled in the past geological epochs under the continent plates. Second, due to the fact that over the extended time of a close contact with kimberlite melts in these samples could have occurred (and have occurred) substantial metamorphic alterations [11–13]. In most cases, such alterations should have been boiled down to ultramafic rocks contamination with dispersed elements coming from mineralized hydrotherms or from the kimberlite magmas saturated with volatile components.

Nevertheless, comparisons of kimberlites and lherzolite xenoliths average chemical composition [14] are quite demonstrative. These comparisons show that kimberlites are somewhat impoverished in such major petrogenic elements as Si, Mg, Na, Cr, and Ni whereas their contents of Al, Fe, Mn and some ore elements (Co, Zn) almost correspond with their concentrations in lherzolites. But the most typical feature of kimberlite rocks is their clear enrichment with dispersed elements. This is concerning especially of lithophilic and rare earth elements. Ya. Muramatsu’s determinations showed that kimberlites are enriched with carbon 150-fold, phosphorus, 25-fold, alkali (K, Rb, Sc), 24–68-fold, light rare earth elements (La–Eu), 30 to 200-fold and radioactive elements Th and U, respectively, 80- and 60-fold.

It is difficult to verify these estimates for all trace elements by independent determinations but for some of them it was possible to accomplish. Accepting the quoted Muramatsu’s concentrations of K, U and Th in the mantle as the genuine ones, it is easy to calculate that the total depth heat flow generated by them and coming from the mantle had to reach 4.8·10²⁰ erg/s. Added to this mantle flow must be heat generated by radioactive elements concentrated in the continental crust, which amounts approximately is 0.9·10²⁰ erg/s. In this case, total radiogenic heat generation in Earth must have reached 5.7·10²⁰ erg/s. However, another exceptionally powerful source of the heat energy is operating within the present-day Earth. This is the process of the mantle matter gravity differentiation resulting in separation within the planet’s central parts of a high-density oxide-ferric core and in the initiation of convection currents in the mantle. Inclusion of this energy source (around 3·10²⁰ erg/s) would have made total Earth’s heat loss in the considered case equal to 8.7·10²⁰ erg/s. However, actual heat loss by Earth is only half of this and is equal approximately to 4.2–4.3·10²⁰ erg/s [15–19].

The energy estimates quoted above indicate that accepted by Muramatsu [14] concentrations of a part of trace elements (K, U, and Th) in the mantle matter are clearly overvalued whereas the enrichment factor of kimberlite rocks with the same elements, substantially underestimated. More correct analysis of Earth’s energy balance accounting for the energy of tidal interaction of Earth with Moon dispersed in mantle (close to 0.1·10²⁰ erg/s), for the potassium content and K/U and K/Th ratios in the continental crust and in lunar rocks enabled us to determine that Earth’s mantle currently contains no greater than 0.012% K, 2.6·10⁻⁷% U and 7·10⁻⁷% Th [19]. Assuming these concentration values, we come up with the kimberlite enrichment with potassium reaches not 24 but 87-fold. For uranium and thorium, the values are even greater: respectively, 1,200 and 2,300-fold (instead of 62- and 80-fold).

The above example of independent radioactive elements’ mantle concentrations estimates is begging for a general conclusion that for some other trace elements the extent of kimberlite rocks enrichment, compared with their Clarke contents in the mantle matter may turn out substantially greater than determined by Muramatsu [14].

Our estimates (see below) give the mantle content of about 110 g/t of carbon dioxide and no greater than 0.05% water. According to J. Dawson [1], the kimberlites contain around 3.3–7.1% CO₂ and 5.9–18.7% H₂O. Therefore, the kimberlites are enriched in these volatile compounds respectively 300–650 and 120–370-fold.

It is, however, noteworthy that in kimberlite minerals [1] and even within the diamond crystals [20, 21] are often encountered inclusions of gaseous and liquid hydrocarbons and even alcohols and more complex organic compounds.

Accounting for all these factors is making even more acute the problem of the kimberlites’ origin and of determining the mechanism of so great enrichment of these rocks with lithophilic elements with simultaneously of the silica contents in them. At that, a question needs to be answered of where hydrocarbons in inclusions, with specific for them negative isotopic shifts for the carbon, are coming from.

By definition [4], lamproites are a community of high-magnesium potassium alkaline rocks saturated or slightly undersaturated with the silica and with low contents of aluminum and calcium. The lamproite composition, compared with the kimberlites, is distinct in much broader variability. However, always typical for them are the highest concentrations of potassium (up to 7–10% K₂O), rubidium (up to 300–500 g/t) and barium (up to 5,000–10,000 g/t), elevated content of strontium (up to 1,000–4,000 g/t) and light rare earth’s elements (up to 300–600 g/t of La, up to 600–1,000 g/t of Ce, up to 250–500 g/t of Nd and up to 15–30 g/t of Sm). In the magnesium content (between 8 and 24% MgO), lamproites occupy an intermediate position between the mantle matter and basalts. On the other hand, they are enriched with the uranium (1 to 10 g/t) and thorium (12 to 150 g/t) respectively 400- to 4,000-fold and 1,700- to 20,000-fold compared with the mantle. The lamproites contain substantially smaller amounts of carbon dioxide and water than the kimberlites but compared with the mantle matter these rocks are enriched in carbon 20 to 600-fold and in water, 25- to 70-fold.

Carbonatites and alkaline-ultramafic formations, although not belonging with diamond-bearing rocks, are certainly affine formations of reasonably similar genesis. Their specifics, in brief, are as follows. Quite typical for the carbonatites are wide ranges in the composition of major petrogenic oxides [9]. The silica concentration in these rocks varies practically between 0% and 30–40%, ferrous oxides (in toto), 10–12% to 20–25%, magnesium oxide, 12% to 20%, etc. However, the main distinctive feature of the carbonatites is drastically elevated calcium oxide content, 12% to 50%, and carbon dioxide content, 3–8% to 30–40%. Commonly observed in the carbonatites are elevated contents of sulfur (around 0.3–0.5%), fluorine (around 0.3%) and chlorine (up to 1–3%). Same as the kimberlites, the carbonatites are substantially enriched in light rare earth’s elements but especially in strontium and rare earths: tantalum and niobium and also uranium and thorium.

The alkaline-ultramafic rocks are known for their undersaturation with the silica (its concentration usually does not exceed 40%) and saturation with alkali, especially sodium (up to 5–6% Na₂O and up to 1.5–2% K₂O). In the contents of the alumina, iron, magnesium and calcium, these rocks are close to the alkali series of basalts. However, the main distinction of the alkali-ultramafic rocks from alkali basalts is a typical set and elevated concentration in them of lithophilic elements and rare metals: phosphorus, niobium, tantalus, zirconium, rare earths, strontium, barium and radioactive elements. In some cases (for instance, in the Khibin massif apatite-bearing sienites) are encountered notable amounts (up to 150 cm³ per a kilogram of rock) hydrocarbon gases and even heavy hydrocarbons, up to C₁₉ [22].

Thus, with all variety and specifics of the rock association under review (kimberlites, lamproites, carbonatites and alkali-ultramafic rocks), they are brought close together by the depth (subcrustal) origins, low or moderate silica contents and a drastic enrichment with dispersed lithophilic elements, rare metals and volatile compounds (up to hydrocarbon inclusions). Besides, all these rocks are encountered only in the ancient, Early Pre-Cambrian continental crust.

It is important to include here also quite interesting data regarding the kimberlites’ and carbonatites’ isotopy. It is noted in the monograph by J. Dawson [1] that δD for serpentine and phlogopite (which contain a large mass of kimberlites’ bonded water) varies within a range of 89 to 102‰ and δ¹⁸O values vary between –1.08 and +12.2‰ by SMOW standard. Carbon δ¹³C isotope composition in the kimberlites’ carbonate phase for the rock as a whole as well as for the carbonatites varies within a range of + 1 +2‰ to –8‰, and in some cases to –20 –25‰ by the PDB standard whereas variation of δ¹⁸O oxygen ranges within +6 to +24 +25‰ SMOW [1]. The included data of δ¹³C and δ¹⁸0 broad variations in a carbonate matter of kimberlites and carbonatites also indicate heterogeneity of the sources of matter involved in their formation.