Carbonatite metasomatism of peridotite lithospheric mantle: implications for diamond formation and carbonatite-kimberlite magmatism

Available online at www.sciencedirect.com

ScienceDirect Russian Geology and Geophysics 56 (2015) 280–295 www.elsevier.com/locate/rgg

Carbonatite metasomatism of peridotite lithospheric mantle: implications for diamond formation and carbonatite-kimberlite magmatism N.P. Pokhilenko a,b,*, A.M. Agashev a, K.D. Litasov a,b, L.N. Pokhilenko a a

V.S. Sobolev Institute of Geology and Mineralogy, Siberian Branch of the Russian Academy of Sciences, pr. Akademika Koptyuga 3, Novosibirsk, 630090, Russia b Novosibirsk State University, ul. Pirogova 2, Novosibirsk, 630090, Russia Received 10 July 2014; accepted 25 July 2014

Abstract Mineral inclusions in diamond record its origin at different depths, down to the lower mantle. However, most diamonds entrained with erupting kimberlite magma originate in lithospheric mantle. Lithospheric U-type diamonds crystallize during early metasomatism of reduced (fO2 at the IW oxygen buffer) depleted peridotite in the roots of Precambrian cratons. Evidence of the metasomatic events comes from compositions of garnets in peridotitic xenoliths and inclusions in diamonds. On further interaction with carbonatitic melt, peridotite changes its composition, while diamond no longer forms in a more oxidized environment (fO2 near the CCO buffer). Silicate metasomatism of depleted peridotite (by basanite-like melts) does not induce diamond formation but may participate in generation of group I kimberlite. Low-degree (below 1%) partial melting of metasomatized peridotite produces a kimberlite-carbonatite magmatic assemblage, as in the case of the Snap Lake kimberlite dike. Occasionally, mantle metasomatism may occur as reduction reactions with carbonates and H2O giving rise to hydrocarbon compounds, though the origin of hydrocarbons in the deep mantle remains open to discussion. Melting experiments in carbonate systems show hydrous carbonated melts with low H2O to be the most plausible agents of mantle material transport. An experiment-based model implies melting of carbonates in subducting slabs within the mantle transition zone, leading to formation of carbonatitic diapirs, which can rise through the mantle by buoyancy according to the dissolution-precipitation mechanism. These processes, in turn, can form oxidized channels in the mantle and maintain diamond growth at the back of diapirs by reducing carbon from carbonated melts. When reaching the lithospheric base, such diapirs form a source of kimberlite and related magmas. The primary composition of kimberlite often approaches carbonatite with no more than 10–15% SiO2. © 2015, V.S. Sobolev IGM, Siberian Branch of the RAS. Published by Elsevier B.V. All rights reserved. Keywords: mantle; melting; diamond; peridotite; kimberlite; carbonatite; experiment

Introduction Metasomatism may control the composition trends of lithospheric mantle, diamond formation, and petrogenesis of kimberlitic and other within-plate magmas, as follows from analyses of carbonatite and kimberlite samples and mantle xenoliths in kimberlite pipes. Metasomatism of peridotitic lithospheric mantle has been largely studied (Agashev et al., 2013; Pearson et al., 1995a,b, 2003; Pokhilenko et al., 1999; Shirey et al., 2013; Simon et al., 2003, 2007; Sobolev, 1974). The traditional chemical classification distinguishes carbonatitic and silicate types of mantle metasomatism. Carbonatite metasomatism is most evident in depleted (including diamond-

* Corresponding author. E-mail address: [email protected] (N.P. Pokhilenko)

bearing) dunite-harzburgite in the middle part of mantle section and provides its high enrichments in incompatible elements without notable changes to mineralogy. Silicate metasomatism occurs mostly in lithospheric roots (Agashev et al., 2013) or in the upper lithospheric mantle (Tychkov et al., 2014) and increases percentages of garnet and clinopyroxene in peridotite. Inclusions in diamond reveal its metasomatic origin (Klein-BenDavid et al., 2007; Liu et al., 2009; Navon, 1999), while the chemistry of fluid inclusions records mainly carbonatitic environments of diamond growth (Klein-BenDavid et al., 2009; Zedgenizov et al., 2007). Metasomatic carbonated lherzolite at the base of lithospheric mantle may be responsible, for instance, for the origin of the Snap Lake kimberlite dike system (Agashev et al., 2008; Pokhilenko et al., 2004). Namely, the Snap Lake kimberlite may be originally evolved from carbonatitic magma that produced by low-degree partial melting of metasomatized

1068-7971/$ - see front matter D 201 5, V.S. So bolev IGM, Siberian Branch of the RAS. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.rgg.201 + 5.01.020

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

281

Fig. 1. Representative microphotographs of diamondiferous peridotite xenoliths from Udachnaya kimberlite (Pokhilenko et al., 2014). Ol, olivine; Grt, garnet; a, sample Uv-177/89; b, sample Uv 465/86.

carbonated mantle (Agashev et al., 2001a, 2008). During further ascent and melting progress to higher degrees, such melts acquire a more kimberlitic composition as silicate minerals from lithospehric mantle become assimilated in the melt. This hypothesis has been used in many recent models (Kamenetsky et al., 2009a,b; Sharygin et al., 2014). Experimental studies of peridotite, eclogite, and kimberlite systems with CO2 (Dasgupta and Hirschmann, 2006; Dasgupta et al., 2013; Ghosh et al., 2009, 2014; Litasov, 2011; Litasov et al., 2011a) show the presence of carbonated melt with no more than 5–10% SiO2 at the pressures and temperatures of lithospheric mantle and asthenosphere (Dasgupta et al., 2013; Sharygin et al., 2014; Shatskiy et al., 2015). Other inferences from experimental results mean that carbonates consumed during subduction are recycled mostly within the mantle transition zone and can become a source of plume carbonatitic magma (Grassi and Schmidt, 2011; Litasov et al., 2013b). Many recent melting experiments shed light on liquidus phase relations in kimberlitic systems and melt compositions at pressures to 3–8 GPa (Litasov et al., 2010; Sharygin et al., 2013, 2014; Sokol et al., 2013b; Sokol and Kruk, 2015). In this paper we discuss models of metasomatism in lithospheric mantle roots with implications for diamond formation, origin of kimberlite magma at the lithospheric base, and sources of carbon and hydrogen that come from the mantle transition zone (stagnant slabs) or from the core-mantle boundary. The discussion bases upon earlier published evidence and new analytical data for xenoliths and lithospheric mantle minerals and for kimberlite and carbonatite samples, as well as upon recent high-pressure experimental studies of carbonate-bearing systems.

Materials and data We use a large collection of kimberlites, peridotitic xenoliths and xenocryst minerals of lithospheric mantle analyzed for many years. Some reported results are new and some were published before, such as data on xenoliths (Agashev et al.,

2013; Howarth et al., 2014; Pokhilenko, 2009; Pokhilenko et al., 1993, 1999, 2004, 2014; Shimizu et al., 1999) and on kimberlites and carbonatites from the Siberian craton and the Snap Lake dike (Slave craton, Canada) (Agashev et al., 2000, 2001b, 2008). New results include compositions of diamondrelated high-Cr garnets and clinopyroxenes from coarse granular peridotite xenoliths, including diamond-bearing megacrystalline ones (Fig. 1), found in the Udachnaya kimberlite pipe (Pokhilenko et al., 2014).

Chemistry of high-Cr garnets Major- and trace-element compositions of Cr-rich garnets from coarse granular and megacrystalline peridotitic xenoliths from the Udachnaya kimberlite have been studied (Table 1) with implications for effects of carbonatite metasomatism. Most of the analyzed garnets fall into the dunite-harzburgite field in the Cr2O3–CaO diagram (Sobolev et al., 1969, 1973), while some garnets with greater CaO enrichment correspond to wherlitic compositions (Fig. 2). The Cr2O3 and CaO ranges, respectively, are 6.4–12.4 wt.% and 0.86–7.53 wt.% for harzburgite garnets and 4.5–7.4 and 6.1–7.1 for wherlite varieties. All garnets are highly magnesian (Mg# 81.0–85.4) and strongly depleted in TiO2 (0–0.33 wt.%). Garnets from sheared peridotite (Agashev et al., 2013) have compositions with broadly varying CaO within the lherzolitic trend (Fig. 2). High-Cr garnets have large ranges of incompatible elements and are more or less strongly enriched in LREE, MREE and Nb, but are depleted in HREE and Y. The chondrite-normalized REE patterns of harzburgite garnets (Fig. 3) split into two groups: garnets with (i) highest LREE enrichment and Ce and Pr peaks (Hz1) and those with (ii) MREE enrichment and a Nd peak. The earlier studied compositions of garnets from sheared lherzolite (Agashev et al., 2013) likewise fall into two groups: (i) Lzl 1 garnets with sine-shaped REE spectra (Sm/Er, n > 1) and (ii) garnets with normal REE patterns. Although the two groups overlap in some elements, the spectra differ in both curve shapes and element abundances.

Hz1

42.35 0.03 18.76 6.55 7.34 0.44 22.07 2.80 0.01 100.34 159.3 37.4 317.6 43.4 41.3 0.00 5.09 0.53 0.16 0.22 0.00 0.35 0.78 0.02 0.05 0.01 0.00 0.00 0.00 0.01 0.03 0.16 0.04 0.68 0.17 0.00 0.01 0.02 0.01 0.01

Hz1

41.27 0.04 15.80 9.83 7.34 0.45 20.28 4.31 0.03 99.36 180.3 332.5 431.4 56.7 52.9 0.00 1.67 1.65 2.21 0.39 0.04 0.09 0.53 0.09 0.58 0.17 0.05 0.14 0.02 0.15 0.04 0.27 0.08 0.76 0.16 0.04 0.01 0.01 0.01 0.00

Hz1

41.56 0.06 15.63 10.54 7.43 0.40 22.29 2.09 0.00 100.01 203.9 257.9 389.3 47.9 45.8 0.03 21.92 1.15 1.96 0.48 0.14 2.14 6.23 0.77 2.60 0.36 0.07 0.18 0.02 0.10 0.03 0.13 0.04 0.45 0.10 0.07 0.00 0.01 0.21 0.03

Hz1

41.72 0.02 17.22 8.29 7.43 0.40 21.68 3.04 0.02 99.83 165.9 41.2 364.3 44.8 47.6 0.98 6.79 0.40 2.83 0.27 0.01 0.93 6.88 0.56 1.24 0.21 0.05 0.06 0.01 0.04 0.02 0.08 0.04 0.47 0.15 0.10 0.00 0.04 0.17 0.09

Hz1 41.72 0.02 17.22 8.29 7.43 0.40 21.68 3.04 0.02 99.83 162.4 37.7 374.8 42.7 39.1 0.00 6.86 0.43 2.87 0.27 0.00 1.01 7.19 0.65 1.41 0.18 0.04 0.00 0.01 0.06 0.03 0.13 0.04 0.41 0.15 0.08 0.00 0.05 0.18 0.09

41.96 0.03 18.87 6.42 6.99 0.37 23.45 1.15 0.00 99.22 139.7 86.8 345.5 49.7 58.5 0.98 11.28 0.25 0.42 0.56 0.69 0.40 1.84 0.19 0.63 0.10 0.03 0.07 0.00 0.03 0.01 0.07 0.02 0.40 0.11 0.02 0.01 0.01 0.04 0.04

Hz1

Uv24/78

Note. Oxide contents are in wt.%, element contents are in ppm.

SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO CaO Na2O Total Sc Ti V Co Ni Rb Sr Y Zr Nb Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta Pb Th U

Compo- Uv-440 Uv-489 Uv-500 Uv-709 Uvnent 709-1

41.62 0.03 17.11 8.60 7.46 0.36 23.52 0.86 0.02 99.57 172.9 109.0 344.0 43.5 47.9 0.58 10.85 0.44 3.06 0.55 0.26 0.62 7.31 1.60 5.50 0.36 0.05 0.09 0.02 0.08 0.02 0.05 0.02 0.24 0.08 0.07 0.01 0.02 0.07 0.12

Hz1

Uv21/93

41.39 0.00 13.99 12.46 6.88 0.38 22.51 1.71 0.01 99.32 356.8 19.1 307.7 48.1 68.1 0.50 3.49 0.22 2.04 0.47 0.04 0.23 2.02 0.37 1.78 0.19 0.04 0.09 0.01 0.05 0.01 0.05 0.01 0.22 0.11 0.07 0.00 0.03 0.02 0.05

Hz1

Uv77/93

41.73 0.00 17.90 7.51 7.20 0.35 23.58 0.98 0.01 99.26 104.8 88.3 135.3 15.2 16.6 0.18 4.64 0.67 7.80 0.12 0.05 0.22 1.90 0.49 2.75 0.54 0.12 0.31 0.03 0.15 0.02 0.05 0.01 0.12 0.03 0.20 0.00 0.01 0.02 0.03

Hz1

Uv94/93

Table 1. Chemistry of high-Cr garnets from peridotitic xenoliths in Udachnaya kimberlite

41.35 0.06 15.32 10.97 7.03 0.43 21.82 3.00 0.00 99.99 187.0 354.9 407.4 45.0 46.8 0.00 9.99 1.18 6.65 0.61 0.01 0.63 3.77 0.75 3.74 1.05 0.23 0.43 0.04 0.25 0.06 0.12 0.02 0.28 0.05 0.14 0.02 0.03 0.08 0.06

Hz2 40.83 0.23 15.20 10.33 7.48 0.39 20.19 4.55 0.07 99.26 190.6 1184.0 325.4 42.3 50.7 0.72 2.12 1.16 6.23 0.35 0.37 0.17 1.54 0.51 4.06 1.25 0.26 0.43 0.06 0.30 0.06 0.14 0.05 0.47 0.14 0.17 0.01 0.02 0.02 0.05

Hz2

Uv-433 Uv90/01

41.26 0.21 15.24 10.50 7.00 0.37 21.70 3.12 0.04 99.44 161.8 1379.9 402.8 44.0 47.0 1.10 9.80 1.88 21.32 0.43 0.01 0.38 3.48 0.84 5.67 1.61 0.44 1.11 0.14 0.59 0.09 0.19 0.03 0.29 0.08 0.69 0.02 0.03 0.04 0.07

Hz2 40.80 0.07 14.46 11.74 6.59 0.40 22.33 3.24 0.01 99.65 264.1 376.3 272.3 42.2 48.3 1.91 8.20 2.21 24.02 0.62 0.00 0.41 4.82 1.28 8.35 2.04 0.51 1.41 0.13 0.65 0.09 0.17 0.03 0.32 0.08 0.41 0.02 0.03 0.04 0.09

Hz2

Uv-556 Uv50/91

41.73 0.00 18.39 6.79 7.33 0.50 21.64 2.90 0.01 99.28 440.7 39.3 147.6 44.5 29.0 1.36 16.61 1.37 10.51 0.41 0.06 0.44 6.05 1.66 9.87 2.34 0.61 1.61 0.10 0.66 0.09 0.14 0.02 0.21 0.07 0.13 0.01 0.04 0.02 0.05

Hz2 41.66 0.26 18.50 6.64 7.51 0.47 20.96 3.65 0.09 99.74 153.3 1504.2 183.4 39.2 27.3 1.06 0.79 3.70 29.06 0.24 0.10 0.11 0.84 0.28 2.38 0.81 0.31 0.74 0.11 0.73 0.14 0.47 0.08 0.70 0.13 0.71 0.02 0.02 0.06 0.07

Hz2

UvUv216/86 3/02

39.89 0.33 12.18 13.92 6.45 0.41 18.46 7.53 0.04 99.21 154.9 1712.3 435.2 36.7 49.1 1.02 4.13 5.90 29.21 1.66 0.18 0.68 4.47 1.07 7.85 2.48 0.91 2.72 0.37 1.66 0.25 0.42 0.04 0.38 0.06 0.76 0.15 0.01 0.14 0.20

Hz2

Uv58/91

40.79 0.01 15.40 10.51 7.45 0.48 19.41 6.05 0.02 100.13 159.9 195.0 157.4 39.0 36.4 0.00 9.02 3.52 24.97 0.91 0.00 0.50 5.62 1.61 12.52 4.10 0.91 1.87 0.19 0.73 0.12 0.26 0.04 0.26 0.04 0.57 0.08 0.02 0.07 0.20

Hz2

Uv49/93

41.10 0.01 19.63 4.57 8.07 0.50 19.26 6.13 0.02 99.33 164.3 103.4 201.3 38.1 22.5 1.16 0.31 1.18 0.32 0.17 0.00 0.04 0.37 0.14 1.16 0.21 0.04 0.08 0.01 0.13 0.05 0.22 0.06 0.46 0.09 0.04 0.00 0.01 0.00 0.01

Wrl

0.00 0.02

41.33 <0.01 19.77 4.66 7.68 0.42 19.28 6.16 0.01 99.31 130.6 27.9 327.4 36.2 28.2 1.22 0.05 0.10 0.73 0.37 0.06 0.04 0.28 0.06 0.39 0.12 0.02 0.05 0.01 0.05 0.01 0.03 0.01 0.16 0.03 0.02 0.02

Wrl 41.06 0.02 19.39 5.15 7.42 0.36 19.30 6.34 0.01 99.10 157.7 15.9 337.8 36.7 26.2 0.00 0.16 0.21 0.21 0.21 0.10 0.01 0.02 0.00 0.04 0.02 0.02 0.11 0.02 0.04 0.01 0.08 0.01 0.32 0.09 0.00 0.01 0.02 0.00 0.01

Wrl 40.83 <0.01 18.43 6.21 7.60 0.48 18.78 7.06 0.04 99.45 275.0 367.0 251.7 40.1 31.0 0.00 22.50 1.01 10.81 0.58 0.08 0.85 13.85 3.44 17.08 2.48 0.58 0.83 0.10 0.35 0.05 0.07 0.02 0.19 0.08 0.10 0.00 0.02 0.06 0.11

Wrl

UvUvUvUv126/93 336/89 135/89 68/89

40.59 0.01 17.57 7.41 7.60 0.47 18.40 7.08 0.04 99.20 173.6 8.1 319.8 46.4 36.1 0.00 0.14 0.27 0.27 0.24 0.13 0.00 0.05 0.01 0.09 0.03 0.01 0.08 0.01 0.05 0.01 0.05 0.03 0.37 0.07 0.00 0.00 0.04 0.00 0.02

Wrl

Uv135/89

282 N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

283

Metasomatism of lithospheric mantle Lithospheric mantle peridotite is a residual component remaining either after high-degree melting in the spinel stability field and subsequent subduction to lithospheric mantle depths (Canil, 2004; Simon et al., 2007) or after mantle plume melting in the stability field of garnet (Griffin et al., 2003). Originally it was dunite or harzburgite material with ultimately depleted compositions free from easy-melting components and diamond. Nevertheless, peridotite xenoliths are strongly depleted in magmaphile elements but enriched in incompatible elements, possibly because of interaction with both carbonatite and silicate melts (Agashev et al., 2013). Evidence of carbonatite metasomatism is recorded in chemistry of rocks and minerals in depleted coarse granular peridotite, especially dunite and harzburgite (Pokhilenko et al., 1993; Shimizu et al., 1999). Sheared peridotite xenoliths bear distinct signatures of silicate metasomatism as relatively high percentages of garnet and clinopyroxene and their equilibrium with silicate melts (Agashev et al., 2013). Signatures of mantle metasomatism in mineral chemistry. Some garnets from peridotite xenoliths are distinctly zoned (e.g., samples U-4-76 (Howarth et al., 2014) (Fig. 4) and U-105-89 (Pokhilenko et al., 1999)). Namely, CaO increases from 2 wt.% in the cores (corresponding to harzburgite garnet) to 6–8 wt.% in the rims (typical of lherzolite garnet) at invariable Cr2O3 (Agashev et al., 2013; Pokhilenko et al., 1999). The garnet compositions evolve further toward lower contents of both major oxides in the fields of either

Fig. 2. Garnets in depleted peridoite xenoliths from Udachnaya kimberlite in Cr2O3–CaO diagram. Curves for harzburgite (Hz) groups 1, 2 and wehrlite (Wrl) garnets are according to new data; those for lherzolite (Lzl) garnets groups 1, 2 are after (Agashev et al., 2013). Hz, Lzl, and Wrl mark fields of harzburgite, lherzolite, and wehrlite garnets, respectively (Sobolev et al., 1973).

lherzolite or wehrlite garnets at decreasing Mg# (Fig. 4), which happens during silicate metasomatism. High-Cr garnets from dunite and harzburgite xenoliths have well pronounced sine-shaped REE spectra with Ce, Pr, Nd and Sm peaks (Pokhilenko et al., 2012), which can result uniquely from effects of LREE-rich fluids or melts, such as carbonatite. The origin of these garnet remains a subject of discussions in mantle petrology (Pearson et al., 1995b). According to experi-

Fig. 3. Chondrite-normalized REE patterns in garnets from mantle peridotites. Hz 1, 2, new data; Lzl 1, 2, after (Agashev et al., 2013).

284

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

Fig. 4. Cr2O3–CaO trends in high-Cr garnets under effects of carbonatite (vertical trend) and silicate (inclined trend) metasomatic agents. Composition trends are based on zoning in garnets from peridotitic xenoliths in Udachnaya kimberlite, after (Howarth et al., 2014; Pokhilenko et al., 1999). Compositions of zoned garnet sample (U 4-76) are after (Howarth et al., 2014).

Fig. 5. REE trends in garnets from peridotitic lithospheric mantle under effects of carbonatite (1, 2) and silicate (3) metasomatic agents. Based on average REE contents over selected groups of garnets. Arrows show REE trends at different stages of metasomatism.

mental evidence (Canil and Wei, 1992), garnets with Cr2O3 above 4% cannot be a residual phase of peridotite melting. Theoretically, they can form by the reaction Sp + Opx = Ol + Gar under the influence of carbonatite metasomatism (Agashev et al., 2013). The metasomatic agent is strongly fractionated and varies broadly in REE abundances, and broad REE variations are observed in garnet correspondingly. The REE patterns of high-Cr garnets from metasomatized peridotite change as follows (Fig. 5): (i) LREE enrichment by percolating strongly fractionated fluids and melts and peak shifting from Ce to Sm as the agent/rock ratio increases; (ii) LREE depletion and transformation of harzburgite into lherzolite as a result of pyroxene crystallization; (iii) HREE enrichment, which requires reactions with low La/Yb melts (basic) and

Fig. 6. Trace-element chemistry of clinopyroxene from peridotitic xenoliths in Udachnaya kimberlite. DP = deformed (sheared) peridotite (Agashev et al., 2013); GP = granular peridotite, new data and data after (Ionov et al., 2010).

shifting toward lherzolitic Cr2O3-depleted varieties. The presence of zoned garnets with Ti-, Fe-, Ca- and Na-rich rims in sheared peridotite is a prominent feature of silicate metasomatism. Clinopyroxenes in coarse granular harzburgite and lherzolite have high LREE contents, large ranges of incompatible elements and La/Yb (60–1000) ratios, and low Ti/Eu and high (La/Yb)n ratios (Fig. 6). All those features are indicators of carbonatitic metasomatism (Agashev et al., 2013; Pearson et al., 2003) as (Coltorti et al., 1999). In granular peridotite, clinopyroxenes can be a metasomatic late phase, as noted in many publications (e.g., Simon et al., 2007). Modeling of fractional crystallization of LREE-rich clinopyroxene likewise indicates equilibrium with carbonated melt (Shchukina et al., 2013). Clinopyroxenes in sheared peridotite have generally lower contents of incompatible elements and relatively uniform REE patterns with La/Yb from 14 to 48. Thus, the chemistry of garnets and clinopyroxenes with signatures of carbonatitic metasomatism differs in broad ranges of REE and other incompatible elements. This means strong fractionation of carbonatite melts and fluids infiltrated through depleted dunite and harzburgite lithospheric mantle and their affinity with diamond growth environments. Garnets and clinopyroxenes in sheared peridotite are in equilibrium with silicate melt and have rather uniform REE spectra. Signatures of mantle metasomatism in rock chemistry. Lithospheric mantle peridotite is depleted in major elements (Al, Ca, Fe, Ti) and HREE, which confirms its residual origin, but is enriched in incompatible elements (Rb, Sr, Ba, Nd, Ta, U Th and LREE). Signatures of carbonatite metasomatism appear first of all in bulk compositions of granular peridotite xenoliths with low garnet and clinopyroxene percentages. However, silicate metasomatism obscures these signatures in lherzolite containing abundant garnet and clinopyroxene, as well as in sheared peridotite. Composition trends of lithospheric mantle peridotite can be modeled as follows (Fig. 7). First, depleted cratonic harzbur-

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

gite forms as high-degree partial melts are extracted from the mantle (trend 1). Then fluids/melts high in incompatible elements, i.e., proximal to carbonatite compositions, metasomatize the lithospheric mantle (trend 2). Metasomatism at this stage does not change much the major-element chemistry of rocks but provides their enrichment in incompatible elements evident in almost ten-fold increase of La/Yb (Fig. 7). The predicted mixing degree of peridotite and carbonatite with high LREE and 30 wt.% CaO is below 1%. Carbonatite metasomatism continues as highly fractionated residual melts/fluids percolate along fractures and grain boundaries through depleted peridotite further increasing La/Yb ratios (to 100 and more) in rocks at moderate CaO increase (trend 2a). CaO increases only slightly being incorporated into garnet, which is confirmed by the existence of few clinopyroxene grains and zoning in high-Cr garnets. Most of incompatible elements are concentrated in sporadic submicron interstitial phases (carbonate, apatite, perovskite, phlogopite). Finally, silicate metasomatism strongly changes rock chemistry and mineralogy. As a result, the percentages of garnet and clinopyroxene in primary depleted peridotite become notably greater, depending on the metasomatic agent/primary rock ratio, while sheared peridotite approaches the primitive mantle composition (trend 3). Natural objects Carbonatite metasomatism and diamond formation. Carbonatite melts/fluids with broad ranges of incompatible elements appear to be main components responsible for diamond growth in lithospheric mantle, as follows from available data on fluid inclusions in diamonds (Klein-BenDavid et al., 2009, 2014; Zedgenizov et al., 2007). Inclusions in many analyzed diamonds have compositions of magnesian carbonatite which underwent significant fractionation and likewise contributed to the formation of high-Cr garnets with sine-shaped REE spectra (Klein-BenDavid et al., 2009). Numerous finds of magnesite, dolomite, and Sr-Ba carbonate inclusions in diamond (Logvinova et al., 2008, 2011), some in assemblage with phlogopite (Sobolev et al., 2009), support the leading role of carbonatite metasomatism in the origin of peridotite diamond. Most of garnets enclosed in diamond (Stachel and Harris, 2008) show sine-shaped REE spectra ((Sm/Er)n > 1) indicating that diamond growth is maintained by carbonatite metasomatism of depleted peridotite but is impeded by silicate metasomatism. According to available dates of sulfide inclusions in peridotitic diamond, metasomatism and diamond formation began in the Archean, soon after stabilization of cratonic lithosphere, and then repeated in several later events (Stachel and Harris, 2008). Melting of carbonated peridotite and genesis of carbonatite-kimberlite magmas. As noted since long ago, kimberlite and carbonatite magmatism often coexist in space and time and thus must be related (Dawson, 1966). Carbonatites occur among kimberlites in many diamond provinces worldwide: Greenland (Tappe et al., 2009), Arkhangelsk

285

Fig. 7. La/Yb–CaO trends in lithospheric mantle peridotites. Sheared peridotites (Agashev et al., 2013), coarse granular peridotites (Ionov et al., 2010). CH = model composition of depleted cratonic harzburgite. Trends: formation of depleted peridotite residue (1), carbonatite metasomatism (2), continuation of carbonatite metasomatism by strongly fractionated residual melts and fluids (2a), silicate metasomatism (3). 1, sheared peridotites; 2, mixture with carbonatite; 3, primitive mantle; 4, granular peridotites.

(Beard et al., 2000), Yakutia, and Canada (Agashev et al., 2008). We analyzed samples of kimberlite and carbonatite coexisting within the Snap Lake dike system. They make a continuous series of major-element compositions from magnesian carbonatite to typical kimberlite, with systematic variations (Fig. 8) similar to those observed in experimental progressive melting of carbonated lherzolite at 6 GPa (Dalton and Presnall, 1998; Dasgupta and Hirschmann, 2006). The Snap Lake carbonatites and kimberlites have different traceelement compositions, the former being more strongly depleted in Rb, K, Ta and Ti but enriched in U, Sr, P, Zr, Hf, MREE and HREE than the latter. This difference is however at odds with experimentally observed liquation partitioning between immiscible carbonate and silicate liquids (Jones et al., 1995; Veksler et al., 1998). Furthermore, partitioning of elements between the Snap Lake carbonatites and kimberlites differs from that in natural fractionation products of carbonatite and kimberlite magmas (Agashev et al., 2008; Beard et al., 2000), in which incompatible elements are concentrated uniformly in carbonated liquid. The analyzed Snap Lake dike samples were inferred to record a natural succession of melting in deep lithospheric mantle predicted earlier from experiments (Dalton and Presnall, 1998). The source of carbonatite-kimberlite magmas should lie at the base of lithospheric mantle isolated from convection, as their high contents of incompatible elements cannot result from melting of asthenosphere but require metasomatic preenrichment (Agashev et al., 2000, 2008). The presence of diamond enclosing high-Cr garnets with a majorite component in the Snap-Lake kimberlite indicates the source depths about 300 km (Pokhilenko et al., 2004). Peridotite garnets with majorite admixture occur also in kimberlitic diamonds in the Yakutian (Sobolev et al., 2004) and Arkhangelsk (Sobolev et al., 1997) provinces. Enrichment of lithospheric mantle source

286

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

Fig. 8. Carbonatites (open circles) and kimberlites (crosses) of Snap Lake dike system (Agashev et al., 2008), compared with results of experimental melting of carbonated peridotite (black circles) at 6 GPa (Dalton and Presnall, 1998) and with composition of Jericho primary kimberlite (open squares), after (Price et al., 2000), in MgO–CaO–SiO2 diagram.

of kimberlites in incompatible elements and CO2 results from percolation of asthenospheric melts that form metasomatic veins with carbonate, K richterite and Fe-Ti oxides. Melting in the source follows the “vein-plus-wall rock melting” mechanism (Foley, 1992). It begins in the veins with significant contribution of vein minerals and produces carbonatite melts; on further heating, melting reaches degrees above 1% and involves wall rock peridotite to produce progressively more silicate and eventually kimberlitic melt compositions. Thus, the contribution of metasomatic vein assemblage into the resulting melt decreases and that of wall rock peridotite increases as melting progresses, which agrees with numerical modeling (Agashev et al., 2008). Preferential dissolution of orthopyroxene maintains SiO2 increase (Chepurov et al., 2013). Evidence from high-pressure exteriments. Details of experiments with carbonate-bearing (including kimberlite) systems were reported elsewhere (Dasgupta, 2013; Litasov et al., 2011a, 2013a; Sokol and Kruk, 2015). Here we only dwell upon key points relevant to carbonated melts as agents in kimberlite magma generation and as liquid components of chemical or thermochemical plumes that form in the mantle transition zone or at the core boundary. Partial melting of kimberlite systems with low H2O leaves residue with eclogite or grosspydite rather than garnet peridotite compositions (Litasov et al., 2010; Sharygin et al., 2013, 2014). However, wehrlite and lherzolite residue can form at more than 20% melting in hydrous systems (Sokol et al., 2013b; Sokol and Kruk, 2015; Ulmer and Sweeney, 2002). Melting experiments in the peridotite–CO2 system summarized in Fig. 9 show that carbonatite melts with less than 15–20% SiO2 can exist in a large temperature range spanning most of model mantle geotherms from subduction to mantle adiabat or slightly higher. Solidus in this case is controlled by

Fig. 9. Field of carbonatite melt and parameters of carbonate-silicate melt compositions in peridotite–CO2 systems, after (Dasgupta, 2013; Ghosh et al., 2009, 2014; Litasov et al., 2013b). Lines of CO2 and SiO2 contents in melt are for systems with 3–5 wt.% CO2. Geotherms and PT-profiles of subduction are after (Litasov et al., 2011a). Solidus of K-free peridotite–CO2 system (CMASN–CO2) is after (Litasov and Ohtani, 2009).

phase relations in alkali-carbonate systems (Litasov et al., 2013b; Shatskiy et al., 2015) observed also in eclogite and kimberlite systems. These results are of critical importance for melting of metasomatized mantle and related diamond formation and kimberlite melt extraction. Note that water plays an essential role in the generation of carbonatite and kimberlite magmas (Sokol and Kruk, 2015). At H2O above 5–10 wt.% in the system, the melt coexisting with peridotite/eclogite shifts markedly to more silicate compositions. In nature such sources produce lamproite, lamprohyre, and orangeite magmas. Group I kimberlites and carbonatites represent low-H2O sources. Aqueous fluids may be stored in the lithospheric mantle which becomes progressively less soluble in silicates with pressure decrease. Nominally anhydrous minerals likewise may store H2O released on interaction with carbonated melt (Sokol and Kruk, 2015; Sokol et al., 2013a). Although there have been no studies of systems with H2O and CO2 (Foley et al., 2009; Litasov et al., 2011a), the source of group I kimberlite can be inferred to contain small amounts of H2O which would not change much the melt compositions shown in Fig. 9. Thus, the experiments evidence that in many cases the primary kimberlite melt composition may approach carbonatite with SiO2 to 10–15%.

Mantle redox state and material transport The redox state of past and present mantle and magma, which controls fluid compositions and P-T conditions of solidus (Foley, 2008, 2011; Foley et al., 2009; Frost and

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

McCammon, 2008; Kadik, 1986; Litasov et al., 2011a, 2014; Sokol et al., 2009, 2010; Stagno et al., 2013; Taylor and Green, 1988), is a key factor in mineral formation. Upper mantle mass transport occurs mostly by fluids in reducing conditions and by silicate (in presence of H2O) or carbonate (in presence of CO2) melts in relatively cold (< 100 °C above solidus) oxidized mantle at rather low temperatures. The redox state can be expressed via fugacity of volatiles, most often oxygen (fO2), calculated on the basis of thermodynamic mineral equilibrium, with participation of transition elements, or estimated experimentally using electrochemical cells (Kadik et al., 1989), or inferred from Fe3+/ΣFe in minerals measured by Fe K-edge XANES (Yaxley et al., 2012). Oxygen fugacity in peridotite xenoliths from kimberlite decreases progressively with depths (Creighton et al., 2009; Frost and McCammon, 2008), while the redox profiles of depleted peridotite (dunite and harzburgite) remain poorly constrained. Estimates of fO2 for xenoliths from the Udachnaya kimberlite (Fig. 10) correspond to reducing conditions and approach the iron-wüstite (IW) oxygen buffer at the 200 km depth. Similar fO2 behavior was reported for peridotite beneath the Kaapvaal craton (Woodland and Koch, 2003). Sheared and granular peridotites generally represent a more oxidized mantle with fO2 two or three log units 2–3 higher than depleted peridotite (Fig. 10), i.e., they formed under the effect of a more oxidized asthenospheric melt or fluid. Note that they fall in the stability field of diamond (rather than carbonates) in the P–fO2 or T–fO2 diagrams. However, diamond coexists with carbonated or carbonate-silicate melts at fO2 2–3 log units below the solid-phase EMOG oxygen buffer (Stagno et al., 2013) (Fig. 10). Therefore, melts responsible for metasomatism of peridotite can be carbonatitic even in reducing subcratonic mantle. An alternative model implies a reducing fluid or melt. According to equations of state calculated for real gases and their mixtures, H2O and CH4 predominate in fluids of the C–O–H system (Belonoshko and Saxena, 1992; Zhang and Duan, 2009) at P-T-fO2 parameters for peridotite xenoliths from kimberlite (Goncharov et al., 2012; Litasov et al., 2014; Yaxley et al., 2012). The weak point of this model is the lack of constraints on solubility of silicate components in reduced CH4-H2O fluids. This solubility was earlier suggested to be negligible at 2–3 GPa (Taylor and Green, 1988), but more recent qualitative analysis has shown that the silicate component dissolved in a reduced fluid can reach 10–20% (Litasov et al., 2014). Dissolved silicates can form compounds with other components of the fluid and thus change its composition. Nevertheless, some fluid inclusions in diamond and other minerals bear traces of reduction by mantle fluids/melts, though such finds are very rare: e.g., hydrocarbons found in garnets and diamonds (Garanin et al., 2011; Kulakova et al., 1982; Pokhilenko et al., 1994; Tomilenko et al., 2001, 2009). Note that natural diamonds most likely grow in carbonated and carbonated-hydrous environments (Litvin, 2009; Pal’yanov et al., 1999, 2002; Palyanov et al., 2013; Sokol et al., 2001), while their nucleation and growth are very slow (Sokol et al., 2009) and hardly probable in reducing conditions at

287

Fig. 10. fO2 as a function of pressure for Udachnaya xenoliths, after (Goncharov et al., 2012; Ionov et al., 2010; Pokhilenko et al., 2008; Yaxley et al., 2012). 1, depleted garnet peridotite; 2, sheared garnet peridotite; 3, coarse granular garnet lherzolite; 4, spinel peridotite. Fields correspond to garnet (Grt) and spinel (Sp) lherzolite, according to analysis of volatiles in minerals from (Pokhilenko et al., 2008). Oxygen buffers (abbreviated as IW = iron-wüstite, WM = wüstite-magnetite, EMOD/G = enstatite-magnesite-olivine-diamond/graphite, D/GCO = diamond/graphite-CO2) are shown relative to 40 mW/m2 geotherm, after (Goncharov et al., 2012; Yaxley et al., 2012). Arrows show fO2 trends for depleted peridotites corresponding to reduced ancient mantle and for sheared peridotite corresponding to effects of more oxidized material on peridotite of craton roots.

high ratios of silicate/carbonate (and H2O) under pressures and temperatures of cratonic shields. Thus, fO2 estimates for natural samples, experimental evidence of diamond nucleation and growth, as well as phase relations in systems with C-O-H fluids indicate that carbonatite or carbonate-silicate melts control diamond formation and metasomatism of peridotite. The redox state is a critical factor also in deeper sublithospheric mantle. Disproportionation of FeO and energetically favorable incorporation of Fe3+ into the silicate structure above 6–8 GPa liberate metallic iron (Fe-Ni alloy), which should be at least 1 wt.% in the lower mantle provided that its bulk oxygen content is the same as in the upper mantle (Frost and McCammon, 2008). Therefore, fO2 in the greatest part of mantle is at the IW buffer or 1–2 log units below it due to the lack of wüstite (e.g., an assemblage of iron–olivine–enstatite) or to the formation of Fe-Mg solid solutions. In these conditions, the C-O-H fluid composition should change from CH4-H2O to pure H2O in the lower mantle (Frost and McCammon, 2008) and probably be far from the field of carbonatite magma generation. Aqueous or methane-aqueous fluids remain poorly studied as agents of mantle mass transport but are expected to be inferior to carbonatite melts. The reason is that silicates are very highly soluble (to 70–90 wt.%) in aqueous fluids at the mantle geotherm (Shatskiy et al., 2009, 2013a) and such melts cannot migrate through almost nonporous plastic mantle below 250–300 km. Migration by the dissolution-precipitation mechanism (see below) requires low solubility of silicates in

288

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

the solvent melt. They are carbonatite melts with small amounts of H2O and alkalis that satisfy this requirement (Fig. 9) and most likely provide mass transport, at least within the transition zone and asthenospheric upper mantle. The hypothesis is supported by high H2O solubility in minerals of the transition zone, where water is bound in wadsleyite and ringwoodite. The transition zone hardly contains up to 0.5 wt.% H2O; storage of H2O is more likely restricted to the lithospheric base in which water is less often bound with mantle minerals than in deeper mantle and the respective magma reservoirs may form with different concentrations of H2O and CO2. Subduction of oxidized material disturbs the mantle redox state. Many slabs bear large amounts of carbonates which can sink below the zones of island arc magma generation. The fO2 difference between the slab and the ambient mantle in the transition zone may reach 5–7 log units (Frost and McCammon, 2008; Litasov et al., 2011b). However, reduced mantle does not impede migration of carbonated melts produced by slab melting in the transition zone.

Carbon sources, migration of carbonated melts in the mantle and metasomatism of craton roots Carbon isotope compositions of both lithospheric and lower-mantle diamonds include primary carbon with δ13C about 0‰ and subducted organic carbon with negative δ13C to –25‰ (Cartigny, 2005). Correspondingly, there are two carbon sources for primary carbonate, carbonate-silicate, or reduced silicate melts with H2O and CH4 fluids. The key points in this respect are the mechanisms of melt migration in the upper and lower mantle and delivery of lower mantle diamonds to the lithospheric base. Melting of subducted carbonates in transition zone. Insights into the mass transport mechanisms can come from melting experiments in carbonated systems. Low-degree melts of carbonated peridotite and eclogite systems with small amounts of alkalis have alkali-carbonatite compositions with low SiO2 (Dasgupta and Hirschmann, 2006; Dasgupta et al., 2004, 2013; Grassi and Schmidt, 2011; Kiseeva et al., 2012,

2013; Litasov and Ohtani, 2009, 2010; Rohrbach and Schmidt, 2011). However, rigorous constraints on the melt compositions and solidus of such systems are impossible because of small melt fractions and related problems with preparation of samples and EMPA measurements. A workable approach in this case may consist in the use of systems compositionally proximal to low-degree peridotite and eclogite melts. The only study of this kind that existed till recently (Sweeney, 1994) was performed at pressures to 5 GPa, but above the solidus and with H2O in the system. Experimental studies of Na- and K-bearing carbonate systems at pressures to 21 GPa (Litasov et al., 2013b) allowed us to estimate the true mantle solidus (Fig. 9) and to analyze the compositon of stable alkali carbonates, as well as the coexisting melts. Both systems contain aragonite and magnesite as main phases (Fig. 11). Magnesite is a near-liquidus phase, along with silicates, and low-degree melts are thus enriched in alkalis and Ca. Aragonite in the sodic system bears up to 7 wt.% Na2O, 1 wt.% K2O, and 8 wt.% MgO. Most of Na being concentrated in aragonite, other alkali carbonates in the system occur in small percentages and only at low temperatures. Some alkali carbonates observed in experiments are new phases (Figs. 11 and 12), such as K2Mg2(CO3)3 (K2Mg2) found in several runs at 6.5–10.5 GPa, K2Ca4(CO3)5, and K2Mg(CO3)2 (K2Mg) (Fig. 12). The latter is stable both in potassic and sodic carbonatites over the whole pressure range, but it becomes a main phase only at 10.5–21.0 GPa, where its temperature stability increases as well (to 1350 °C at 21 GPa, Fig. 12). K2Ca4(CO3)5 is likewise present at all pressures but becomes main only at 21 GPa, with its temperature stability in K-carbonatite increasing to 1250 °C. This phase forms a series of compositions toward aragonite and toward K2Ca2(CO3)3 till possibly stoichiometric K2Ca3(CO3)4 (Fig. 12). Its Na analog (Na2Ca3(CO3)4) was synthesized and described in (Gavryushkin et al., 2014; Shatskiy et al., 2013b). The K2Ca4(CO3)5 phase coexists with Na-aragonite in the Na system and becomes dominant over the latter at 21 GPa. Other phases in the sodic system are Na2Mg2(CO3)2 at 3 GPa and 750 °C, with shortite (Na2Ca2(CO3)3) or nyererite (Na2Ca(CO3)) compositions at 10 GPa and 1100 °C or 15 GPa and 1100 °C, respectively. The low-degree carbonatite

Fig. 11. BSE images of Na- and K-bearing carbonatite samples, after (Litasov et al., 2013b). a, K-carbonatite at 21 GPa and 1300 °C; b, Na-carbonatite at 21 GPa and 1200 °C. Phases: K-Ca, K2Ca4(CO3)5; K-Mg, K2Mg(CO3)2; Mst, magnesite; Na-Ag, Na-aragonite, Ca-perovskite; Fp, ferropericlase. Scale = 0.5 mm.

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

melts are enriched in alkalis and Ca because dolomite (at low pressures) and magnesite are liquidus phases (Fig. 12). We infer from this experimental evidence that alkali carbonates can sink into the mantle transition zone, at least in the case of cold subduction. Most of carbonates should melt in the transition zone, given that the solidus curve of carbonate rocks slopes at progressively lower angles above 10 GPa, while magnesite can remain stable as deep as the lower mantle. Long-lasting subduction of carbonates to the transition zone and their melting may lead to (Fig. 13) segregation of a carbonatite diaipir and its rise by buoyancy to the lithosperic base (a) or upward percolation of carbonated melt (b) and its reduction by reaction with the ambient mantle (c) (Rohrbach and Schmidt, 2011). This or that scenario can realize in each specific case depending on the budget of carbonate supplied into the transition zone and the buffer capacity of the redox reservoirs. Dissolution-precipitation rise of carbonatite diapirs (a) appears to be the most realistic scenario (Litasov et al., 2013a; Shatskiy et al., 2013a). Indeed, the infiltration of melt released by slab melting into the weakly porous mantle is three orders of magnitude slower (Hammouda and Laporte, 2000) than the consumption of carbonates by subduction. As a result, melt segregates along the slab-mantle boundary, and the first carbonatite diapir oxidizes all reduced mantle above it on its way from 550 km to 200–250 km. We estimate that 0.1– 0.2 wt.% free Fe in the transition zone and upper mantle below 250 km can reduce no more than 30% of carbon in a carbonatite diapir 500–1000 m in diameter, and the following diapir will rise without significant redox reactions. Diapirism in deep mantle may be a workable mechanism for transport of both carbonate and silicate melts (Weinberg and Podlad-

289

Fig. 12. Melts and carbonate phases in Ca–(Mg + Fe)–(Na + K) diagram. Letters N and K mark primary compositions (Litasov et al., 2013a,b). Arrow shows temperature-dependent melt composition trend.

chikov, 1994). The rise of carbonatite diapirs to the lithospheric base, where melt becomes channeled along fractures or impregnates pores, is a plausible mechanism that may give rise to a magma source producing kimberlite and other aklali igneous rocks and be responsible also for diamond formation. Plumes rising from the core boundary and lower mantle material transport. The origin and paths of plume melts in the lower mantle remain unknown, and all related discussions (concerning melts at these depths or the very possibility of their infiltration by outgassing, e.g., in the Archean) may be

Fig. 13. a, “Big mantle wedge” and melting of a stagnant or downgoing slab; b, stages of formation and ascent of a carbonatite diapir: first carbonatite melt becomes rapidly extracted from slab and accumulates along the slab-mantle boundary; then it penetrates into overlying mantle and forms a diapir; diapir rises by dissolution-precipitation mechanism. CCO and IW are, respectively, graphite and iron-wüstite oxygen buffers. After (Litasov et al., 2013a,b; Shatskiy et al., 2013a).

290

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

only speculative. The uncertainty comes primarily from the lack of knowledge on H2O solubility in lower mantle minerals. According to widely accepted data, the amounts of H2O in Mg- perovskite, even with Al and Fe impurities, are vanishing (Bolfan-Casanova, 2005; Bolfan-Casanova et al., 2003). There is, however some reports of H2O concentrations in Mgperovskite reaching 0.3 wt.% (Litasov et al., 2003). Being weakly soluble in the lower mantle, water may be an important agent in migration of carbon-bearing liquids. The change of fluids from CH4-H2O to a more oxidized composition with H2O (to 100%) on pressure increase in the transition zone and lower mantle, at oxygen fugacity remaining within the IW buffer, prompts that carbonate, carbonate-hydrous or carbonate-hydrous-silicate melts may act in the lower mantle, while carbonatitic melt or solid carbonates may be stable in association with free iron (Stagno and Frost, 2010). Although there is no data on solubility of silicates in carbonate-hydrous media at the lower mantle pressures and temperatures, it appears reasonable to infer that a carbonatite melt with low H2O may be the most probable candidate for dissolution-precipitation migration. This model is consistent with close relationship between lower mantle inclusions in diamond (Harte and Richardson, 2012) and carbonatitic media recorded in the presence of carbonate phases (Brenker et al., 2007; Kaminsky et al., 2009, 2013) as well as in the compositions of lower mantle silicates (Walter et al., 2008, 2011; Zedgenizov et al., 2014) with their trace element patterns corresponding to equilibrium with carbonated melts. Plume vs. vein metasomatism of continental lithosphere. It remains to discuss the extent of metasomatism in subcontinental lithospheric mantle. Metasomatism may be of a regional scale corresponding to the effect of mantle plumes or even superplumes covering no less than 1000 km (Howarth et al., 2014; Sharygin et al., 2014) or be restricted to local weak zones in the lithosphere related to vein systems within tens of meters around kimberlite pipes (Malkovets et al., 2007). However, irrespective of the scale, it causes the same effects: early metasomatism of depleted peridotite induces diamond formation while its further progress leads to refertilization and spread of sheared peridotite (Fig. 14) and reduction of the diamond stability field (Figs. 14 and 15). Note that metasomatism preceding kimberlite magmatism has to be distinguished from that in the Archean responsible for the origin of diamonds (Shirey and Richardson, 2011; Shirey et al., 2013). The arguments for the plume metasomatism model are as follows. Kimberlite fields, including those in the Siberian craton, are associated with large igneous provinces and with long-lasting mantle plumes rising from the core boundary (Haggerty, 1994; Torsvik et al., 2010). During the respective magmatic events, the Siberian craton was located above the low-velocity zone corresponding to the African superplume (Torsvik et al., 2010). The echo of carbonatite magmatism (metasomatism of lithosphere) is found now in basalts from Canary Islands. Relation with mantle plumes shows up in similarity of Sr and Nd isotope ratios and diagnostic trace-element ratios of group I kimberlites to ocean island basalts (Becker and le Roex, 2006; Maas et al., 2005), as well as in

their high 3He/4He ratios (Sumino et al., 2006; Tachibana et al., 2006). Sheared peridotite likewise can record processes at the boundary of a large plume with continental lithosphere. Intense and broad deformation common to zones of plume margins (d’Acremont et al., 2003; Cloetingh et al., 2013; Kohlstedt and Holtzman, 2009) is hardly possible in local zones of vein systems. Furthermore, deformed peridotite has higher basalt-type Fe, Ti, Al, Na, and K enrichments than the granular variety (Agashev et al., 2013). On the other hand, the model of vein metasomatism accounts well for compositional dissimilarity of xenoliths in proximal kimberlite pipes and for concordant crystallization of high-Cr garnet and diamond; both garnet and diamond in this case result from metasomatic reactions of chromite-bearing depleted peridotite with CH4-H2O fluid/melt (Malkovets et al., 2007). Note that the silicate and carbonate liquids are immiscible and produce, respectively, diamond and subcalcic high-Cr garnet at the account of reaction with chromite (Rege et al., 2008, 2010). The plume model or intermediate scenarios between the two appear more realistic as the presence of carbonatite melts is confirmed by many facts while crystallization of diamond on reaction with reduced carbon-bearing melts has no experimental support.

Conclusions Metasomatism of depleted peridotite lithospheric mantle has implications for diamond formation, as well as for sources of kimberlite magmatism at the lithospheric base and deep sources of carbon and hydrogen entrained with small or large plumes or superplumes from the mantle transition zone (stagnant slabs) or core boundary. Most of U-type lithospheric diamonds have crystallized during early carbonatite metasomatism of reduced depleted peridotites in Precambrian cratons, which is recorded in the chemistry of fluid and mineral inclusions in diamond and in the chemistry and mineralogy of peridotitic xenoliths. Further progress of metasomatism in a more oxidized environment changed mineral percentages in rocks and impeded crystallization of diamond. Subsequent silicate metasomatism of depleted peridotite (by basanite-like melts) did not induce diamond formation but had links with deformation at the lithospheric base and origin of a heterogeneous source of carbonatite-kimberlite magmas. More rarely metasomatism consists in reduction reactions with carbonates and H2O and formation of hydrocarbon compounds, while the mantle origin of the latter as components of the liquid phase remains open to discussion. The available melting experiments in carbonated systems show that low-H2O carbonatitic melts are most likely principal agents of mantle mass transport. Subduction of carbonates into the transition zone and their melting lead to segregation of carbonatite diapirs which rise by buoyancy to the lithospheric base. In each specific case, interaction with reduced mantle

Fig. 14. Metasomatism of lithospheric mantle in cratonic roots. a, Plume model, with metasomatism involving a large subcratonic region. Model agrees with magmatism in Siberian craton and formation of Siberian Trap Province, maintained by African superplume (Sharygin et al., 2015); b, vein model, with metasomatism of local weak zones beneath kimberlites. Model agrees with trace-element and isotopic patterns of xenoliths (Malkovets et al., 2007). Arrows show melt/fluid flows. Yellow rhombs mark diamond occurrences. Pyroxenites and eclogites are not shown for simplicity. LAB = lithosphere–asthenospere boundary.

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295 291

292

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

Fig. 15. PT-conditions at which diamondiferous depleted and barren sheared peridotites formed, respectively, by more or less strong interaction with metasomatizing melt (Sharygin et al., 2015). Panel b shows zone where sheared peridotite substitutes for depleted variety reducing the diamond formation domain. PT-conditions of peridotite are given for Udachnaya kimberlite after (Agashev et al., 2013; Boyd et al., 1997; Goncharov et al., 2012; Ionov et al., 2010). Graphite-diamond transition curve is after (Day, 2012).

depends on the budget of carbonate supplied into the transition zone and on the buffer capacity of redox reservoirs. According to our modeling results, the rise of carbonatite diapirs by dissolution and precipitation appears to be the most plausible mechanism. The earliest rising diapir oxidizes all overlying reduced mantle on its way, for instance, from 550 km to 200–250 km. The ascent of carbonatite diapirs to the lithospheric base, where free melt migration gives way to percolation along fractures or infiltration into pores, can maintain generation of kimberlite and other alkali magmas, as well as diamond formation. Experiments at 3–8 GPa show that the primary kimberlite magma may have its composition close to carbonatite with SiO2 under 10–15%. The lower mantle mass transfer remains poorly constrained, but participation of carbonated melts there can be inferred from data on mineral inclusions in superdeep diamonds. The study was carried out as part of Integration Projects 97, 115 and 59P (2012–2014) of the Siberian Branch of the Russian Academy of Sciences. More support came from grant 14.B25.31.0032 from the Ministry of Education and Science of the Russian Federation and grant 12-05-01043a from the Russian Foundation for Basic Research.

References Agashev, A.M., Orihashi, U., Watanabe, T., Pokhilenko, N.P., Serenko, V.P., 2000. Isotope-geochemical features of the Siberian Platform kimberlires in connection with the problem of their origin. Geologiya i Geofizika (Russian Geology and Geophysics) 41 (1), 90–99 (87–97). Agashev, A.M., Pokhilenko, N.P., McDonald, J.A., Takazawa, E., Vavilov, M.A., Sobolev, N.V., Watanabe, T., 2001a. A unique kimberlite–carbonatite primary association in the Snap Lake dike system, Slave Craton: evidence from geochemical and isotopic studies, in: A Tale of Two

Cratons, The Slave-Kaapvaal Workshop, Merrickville, Ontario, Canada, pp. 46–48. Agashev, A.M., Watanabe, T., Budaev, D.A., Pokhilenko, N.P., Fomin, A.S., Maehara, K., Maeda, J., 2001b. Geochemistry of kimberlites from the Nakyn field, Siberia: Evidence for unique source composition. Geology 29, 267–270. Agashev, A., Pokhilenko, N., Takazawa,E., McDonald, J., Vavilov, M., Watanabe, T., Sobolev, N., 2008. Primary melting sequence of a deep (>250 km) lithospheric mantle as recorded in the geochemistry of kimberlite–carbonatite assemblages, Snap Lake dyke system, Canada. Chem. Geol. 255, 317–328. Agashev, A.M., Ionov, D.A., Pokhilenko, N.P., Golovin, A.V., Cherepanova, Y., Sharygin, I.S., 2013. Metasomatism in lithospheric mantle roots: Constraints from whole-rock and mineral chemical composition of deformed peridotite xenoliths from kimberlite pipe Udachnaya. Lithos 160–161, 201–215. Beard, A., Downes, H., Hegner, E., Sablukov, S., 2000. Geochemistry and mineralogy of kimberlites from the Arkhangelsk Region, NW Russia: evidence for transitional kimberlite magma types. Lithos 51, 47–73. Becker, M., le Roex, A.P., 2006. Geochemistry of South African on- and off-craton, Group I and Group II kimberlites: Petrogenesis and source region evolution. J. Petrol. 47, 673–703. Belonoshko, A.B., Saxena, S.K., 1992. A unified equation of state for fluids of C-H-O-N-S-Ar composition and their mixtures up to very high-temperatures and pressures. Geochim. Cosmochim. Acta 56, 3611–3626. Bolfan-Casanova, N., 2005. Water in the Earth’s mantle. Mineral. Mag. 69, 229–257. Bolfan-Casanova, N., Keppler, H., Rubie, D.C., 2003. Water partitioning at 660 km depth and evidence for very low water solubility in magnesium silicate perovskite. Geophys. Res. Lett. 30, 1905, doi: 1910.1029/2003g l017182. Boyd, F.R., Pokhilenko, N.P., Pearson, D.G., Mertzman, S.A., Sobolev, N.V., Finger, L.W., 1997. Composition of the Siberian cratonic mantle: evidence from Udachnaya peridotite xenoliths. Contrib. Mineral. Petrol. 128, 228–246. Brenker, F.E., Vollmer, C., Vincze, L., Vekemans, B., Szymanski, A., Janssens, K., Szaloki, I., Nasdala, L., Joswig, W., Kaminsky, F., 2007. Carbonates from the lower part of transition zone or even the lower mantle. Earth Planet. Sci. Lett. 260, 1–9.

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295 Canil, D., 2004. Mildly incompatible elements in peridotites and the origins of mantle lithosphere. Lithos 77, 375–393. Canil, D., Wei, K., 1992. Constraints on the origin of mantle-derived low Ca garnets. Contrib. Mineral. Petrol. 109, 421–430. Cartigny, P., 2005. Stable isotopes and the origin of diamond. Elements 1, 79–84. Chepurov, A.I., Zhimulev, E.I., Agafonov, L.V., Sonin, V.M., Chepurov, A.A., Tomilenko, A.A., 2013. The stability of ortho- and clinopyroxenes, olivine, and garnet in kimberlitic magma. Russian Geology and Geophysics (Geologiya i Geofizika) 54 (4), 406–415 (533–544). Cloetingh, S., Burov, E., Francois, T., 2013. Thermo-mechanical controls on intra-plate deformation and the role of plume-folding interactions in continental topography. Gondwana Res. 24, 815–837. Coltorti, M., Bonadiman, C., Hinton, R., Siena, F., Upton, B., 1999. Carbonatite metasomatism of the oceanic upper mantle: evidence from clinopyroxenes and glasses in ultramafic xenoliths of Grande Comore, Indian Ocean. J. Petrol. 40, 133–165. Creighton, S., Stachel, T., Matveev, S., Höfer, H., McCammon, C., Luth, R.W., 2009. Oxidation of the Kaapvaal lithospheric mantle driven by metasomatism. Contrib. Mineral. Petrol. 157, 491–504. d’Acremont, E., Leroy, S., Burov, E.B., 2003. Numerical modelling of a mantle plume: the plume head–lithosphere interaction in the formation of an oceanic large igneous province. Earth Planet. Sci. Lett. 206, 379–396. Dalton, J.A., Presnall, D.C., 1998. Carbonatitic melts along the solidus of model lherzolite in the system CaO–MgO–Al2O3–SiO2–CO2 from 3 to 7 GPa. Contrib. Mineral. Petrol. 131, 123–135. Dasgupta, R., 2013. Ingassing, storage, and outgassing of terrestrial carbon through geologic time. Rev. Mineral. Geochem. 75, 183–229. Dasgupta, R., Hirschmann, M.M., 2006. Melting in the Earth’s deep upper mantle caused by carbon dioxide. Nature 440, 659–662. Dasgupta, R., Hirschmann, M.M., Withers, A.C., 2004. Deep global cycling of carbon constrained by the solidus of anhydrous, carbonated eclogite under upper mantle conditions. Earth Planet. Sci. Lett. 227, 73–85. Dasgupta, R., Mallik, A., Tsuno, K., Withers, A.C., Hirth, G., Hirschmann, M.M., 2013. Carbon-dioxide-rich silicate melt in the Earth’s upper mantle. Nature 493, 211–215. Dawson, J.B., 1966. The kimberlite–carbonatite relationship, in: Naidu, P.R.J. (Ed.), Papers and Proc. 4th General Meet., Int. Mineral. Assoc., IMA Volume, Mineral. Soc. India, pp. 1–4. Day, H.W., 2012. A revised diamond-graphite transition curve. Am. Mineral. 97, 52–62. Foley, S.F., 1992. Vein plus wall rock melting mechanism in lithosphere and the origin of potassic magmas. Lithos 28, 435–453. Foley, S.F., 2008. Rejuvenation and erosion of the cratonic lithosphere. Nature Geosci. 1, 503–510. Foley, S.F., 2011. A reappraisal of redox melting in the Earth’s mantle as a function of tectonic setting and time. J. Petrol. 52, 1363–1391. Foley, S.F., Yaxley, G.M., Rosenthal, A., Buhre, S., Kiseeva, E.S., Rapp, R.P., Jacob, D.E., 2009. The composition of near-solidus melts of peridotite in the presence of CO2 and H2O between 40 and 60 kbar. Lithos 112, 274–283. Frost, D.J., McCammon, C.A., 2008. The redox state of Earth’s mantle. Annu. Rev. Earth Planet. Sci. 36, 389–420. Garanin, V., Biller, A.Y., Skvortsova, V., Bovkun, A., Bondarenko, G., 2011. Polyphase hydrocarbon inclusions in garnet from the Mir diamondiferous pipe. Moscow Univer. Geol. Bull. 66, 116–125. Gavryushkin, P.N., Bakakin, V.V., Bolotina, N.B., Shatskiy, A.F., Seryotkin, Y.V., Litasov, K.D., 2014. Synthesis and crystal structure of new carbonate Ca3Na2(CO3)4 homeotypic with orthoborates M3Ln2(BO3)4 (M = Ca, Sr, and Ba). Crystal Growth & Design 14, 4610–4616. Ghosh, S., Ohtani, E., Litasov, K.D., Terasaki, H., 2009. Solidus of carbonated peridotite from 10 to 20 GPa and origin of magnesiocarbonatite melt in the Earth’s deep mantle. Chem. Geol. 262, 17–28. Ghosh, S., Litasov, K., Ohtani, E., 2014. Phase relations and melting of carbonated peridotite between 10 and 20 GPa: a proxy for alkali- and CO2-rich silicate melts in the deep mantle. Contrib. Mineral. Petrol. 167, 1–23. Goncharov, A., Ionov, D., Doucet, L., Pokhilenko, L., 2012. Thermal state, oxygen fugacity and C-O-H fluid speciation in cratonic lithospheric

293

mantle: New data on peridotite xenoliths from the Udachnaya kimberlite, Siberia. Earth Planet. Sci. Lett. 357, 99–110. Grassi, D., Schmidt, M.W., 2011. The melting of carbonated pelites from 70 to 700 km depth. J. Petrol. 52, 765–789. Griffin, W., O’Reilly, S., Abe, N., Aulbach, S., Davies, R., Pearson, N., Doyle, B., Kivi, K., 2003. The origin and evolution of Archean lithospheric mantle. Precambrian Res. 127, 19–41. Haggerty, S.E., 1994. Superkimberlites: A geodynamic diamond window to the Earth’s core. Earth Planet. Sci. Lett. 122, 57–69. Hammouda, T., Laporte, D., 2000. Ultrafast mantle impregnation by carbonatite melts. Geol. 28, 283–285. Harte, B., Richardson, S., 2012. Mineral inclusions in diamonds track the evolution of a Mesozoic subducted slab beneath West Gondwanaland. Gondwana Res. 21, 236–245. Howarth, G.H., Barry, P.H., Pernet-Fisher, J.F., Baziotis, I.P., Pokhilenko, N.P., Pokhilenko, L.N., Bodnar, R.J., Taylor, L.A., Agashev, A.M., 2014. Superplume metasomatism: Evidence from Siberian mantle xenoliths. Lithos 184–185, 209–224. Ionov, D.A., Doucet, L.S., Ashchepkov, I.V., 2010. Composition of the lithospheric mantle in the Siberian craton: New constraints from fresh peridotites in the Udachnaya-East kimberlite. J. Petrol. 51, 2177–2210. Jones, J.H., Walker, D., Pickett, D.A., Murrell, M.T., Beattie, P., 1995. Experimental investigations of the partitioning of Nb, Mo, Ba, Ce, Pb, Ra, Th, Pa, and U between immiscible carbonate and silicate liquids. Geochim. Cosmochim. Acta 59, 1307–1320. Kadik, A.A., 1986. Fractionation of volatile components during melting of the upper mantle. Geologiya i Geofizika (Soviet Geology and Geophysics) 27 (7), 70–73 (57–61). Kadik, A.A., Sobolev, N.V., Zharkova, E.V., Pokhilenko, N.P., 1989. Formation of diamond-bearing peridotite xenoliths from Udachnaya kimberlite (Yakutia): Redox conditions. Geokhimiya 27, 1120–1136. Kamenetsky, V.S., Kamenetsky, M.B., Sobolev, A.V., Golovin, A.V., Sharygin, V.V., Pokhilenko, N.P., Sobolev, N.V., 2009a. Can pyroxenes be liquidus minerals in the kimberlite magma? Lithos 112, 213–222. Kamenetsky, V.S., Kamenetsky, M.B., Weiss, Y., Navon, O., Nielsen, T.F.D., Mernagh, T.P., 2009b. How unique is the Udachnaya-East kimberlite? Comparison with kimberlites from the Slave Craton (Canada) and SW Greenland. Lithos 112, 334–346. Kaminsky, F., Wirth, R., Matsyuk, S., Schreiber, A., Thomas, R., 2009. Nyerereite and nahcolite inclusions in diamond: evidence for lower-mantle carbonatitic magmas. Mineral. Mag. 73, 797–816. Kaminsky, F.V., Wirth, R., Schreiber, A., 2013. Carbonatitic inclusions in deep mantle diamond from Juina, Brazil: new minerals in the carbonatehalide association. Can. Mineral. 51, 669–688. Kiseeva, E.S., Yaxley, G.M., Hermann, J., Litasov, K.D., Rosenthal, A., Kamenetsky, V.S., 2012. An experimental study of carbonated eclogite at 3.5–5.5 GPa: Implications for silicate and carbonate metasomatism in the cratonic mantle. J. Petrol. 53, 727–759. Kiseeva, E.S., Litasov, K.D., Yaxley, G.M., Ohtani, E., Kamenetsky, V.S., 2013. Melting and phase relations of carbonated eclogite at 9–21 GPa and the petrogenesis of alkali-rich melts in the deep mantle. J. Petrol. 54, 1555–1583. Klein-Ben David, O., Izraeli, E.S., Hauri, E., Navon, O., 2007. Fluid inclusions in diamonds from the Diavik mine, Canada and the evolution of diamond-forming fluids. Geochim. Cosmochim. Acta 71, 723–744. Klein-Ben David, O., Logvinova, A.M., Schrauder, M., Spetius, Z.V., Weiss, Y., Hauri, E.H., Kaminsky, F.V., Sobolev, N.V., Navon, O., 2009. High-Mg carbonatitic microinclusions in some Yakutian diamonds—a new type of diamond-forming fluid. Lithos 112, 648–659. Klein-BenDavid, O., Pearson, D.G., Nowell, G.M., Ottley, C., McNeill, J.C.R., Logvinova, A., Sobolev, N.V., 2014. The sources and time-integrated evolution of diamond-forming fluids—Trace elements and isotopic evidence. Geochim. Cosmochim. Acta 125, 146–169. Kohlstedt, D.L., Holtzman, B.K., 2009. Shearing melt out of the Earth: An experimentalist’s perspective on the influence of deformation on melt extraction. Ann. Rev. Earth Planet. Sci. 37, 561–593. Kulakova, I.I., Ogloblina, A.I., Rudenko, A.P., Florovskaya, V.N., Botkunov, A.I., Skvortsova, V.L., 1982. Polycyclic aromatic hydrocarbons in

294

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295

diamond-related minerals: a probable formation mechanism. Dokl. Akad. Nauk SSSR 267, 1458–1461. Litasov, K.D., Ohtani, E., 2009. Solidus and phase relations of carbonated peridotite in the system CaO–Al2O3–MgO–SiO2–Na2O–CO2 to the lower mantle depths. Phys. Earth Planet. Inter. 177, 46–58. Litasov, K., Ohtani, E., 2010. The solidus of carbonated eclogite in the system CaO–Al2O3–MgO–SiO2–Na2O–CO2 to 32 GPa and carbonatite liquid in the deep mantle. Earth Planet. Sci. Lett. 295, 115–126. Litasov, K., Ohtani, E., Langenhorst, F., Yurimoto H., Kubo, T., Kondo, T., 2003. Water solubility in Mg-perovskites, and water storage capacity in the lower mantle. Earth Planet. Sci. Lett. 211, 189–203. Litasov, K.D., Sharygin, I.S., Shatsky, A.F., Ohtani, E., Pokhilenko, N.P., 2010. Experimental constraints on the role of chloride in the origin and evolution of kimberlitic magma. Dokl. Earth Sci. 435 (2), 1641–1646. Litasov, K.D., Shatskiy, A.F., Pokhilenko, N.P., 2011a. Phase realtions and melting in systems peridotite–H2O–CO2 and eclogite–H2O–CO2 at pressures up to 3–27 GPa. Dokl. Earth Sci. 437 (2), 498-502. Litasov, K.D., Goncharov, A.F., Hemley, R.J., 2011b. Crossover from melting to dissociation of CO2 under pressure: Implications for the lower mantle. Earth Planet. Sci. Lett. 309, 318–323. Litasov, K.D., Shatskiy, A., Ohtani, E., 2013a. Earth’s mantle melting in the presence of C-O-H-bearing fluid, in: Karato, S. (Ed.), Physics and Chemistry of Deep Earth, Wiley-Blackwell, pp. 38–65. Litasov, K.D., Shatskiy, A., Ohtani, E., Yaxley, G.M., 2013b. The solidus of alkaline carbonatite in the deep mantle. Geology 41, 79–82. Litasov, K.D., Shatskiy, A., Ohtani, E., 2014. Melting and subsolidus phase relations in peridotite and eclogite systems with reduced COH fluid at 3–16 GPa. Earth Planet. Sci. Lett. 391, 87–99. Litvin, Iu.A., 2009. Physicochemical conditions of diamond formation in the mantle matter: experimental studies. Russian Geology and Geophysics (Geologiya i Geofizika) 50 (12), 1188–1200 (1530–1546). Liu, Y., Taylor, L.A., Sarbadhikari, A.B., Valley, J.W., Ushikubo, T., Spicuzza, M.J., Kita, N., Ketcham, R.A., Carlson, W., Shatsky, V., 2009. Metasomatic origin of diamonds in the world’s largest diamondiferous eclogite. Lithos 112, 1014–1024. Logvinova, A.M., Wirth, R., Fedorova, E.N., Sobolev, N.V., 2008. Nanometre-sized mineral and fluid inclusions in cloudy Siberian diamonds: new insights on diamond formation. European J. Miner. 20, 317–331. Logvinova, A.M., Wirth, R., Tomilenko, A.A., Afanasiev, V.P., Sobolev, N.V., 2011. The phase composition of crystal-fluid nanoinclusions in alluvial diamonds in the northeastern Siberian Platform. Russian Geology and Geophyics (Geologiya i Geofizika) 52 (11), 1286–1297 (1634–1648). Maas, R., Kamenetsky, M.B., Sobolev, A.V., Kamenetsky, V.S., Sobolev, N.V., 2005. Sr, Nd, and Pb isotope evidence for a mantle origin of alkali chlorides and carbonates in the Udachnaya kimberlite, Siberia. Geology 33, 549–552. Malkovets, V.G., Griffin, W.L., O’Reilly, S.Y., Wood, B.J., 2007. Diamond, subcalcic garnet, and mantle metasomatism: Kimberlite sampling patterns define the link. Geology 35, 339–342. Navon, O., 1999. Diamond formation in the Earth’s mantle, in: Proceedings of the 7th International Kimberlite Conference. Red Roof Design, Cape Town, pp. 584–604. Pal’yanov, Y.N., Sokol, A.G., Borzdov, Y.M., Khokhryakov, A.F., Sobolev, N.V., 1999. Diamond formation from mantle carbonate fluids. Nature 400, 417–418. Pal’yanov, Y.N., Sokol, A.G., Borzdov, Y.M., Khokhryakov, A.F., Sobolev, N.V., 2002. Diamond formation through carbonate-silicate interaction. Am. Mineral. 87, 1009–1013. Palyanov, Y.N., Bataleva, Y.V., Sokol, A.G., Borzdov, Y.M., Kupriyanov, I.N., Reutsky, V.N., Sobolev, N.V., 2013. Mantle–slab interaction and redox mechanism of diamond formation. Proc. Nat. Acad. Sci. 110, 20,408–20,413. Pearson, D., Canil, D., Shirey, S., 2003. Mantle samples included in volcanic rocks: xenoliths and diamonds. Treatise on Geochemistry 2, 171–275. Pearson, D., Shirey, S., Carlson, R., Boyd, F.R., Pokhilenko, N., Shimizu, N., 1995a. Re-Os, Sm-Nd, and Rb-Sr isotope evidence for thick Archaean lithospheric mantle beneath the Siberian craton modified by multistage metasomatism. Geochim. Cosmochim. Acta 59, 959–977.

Pearson. D., Snyder. G., Shirey. S., Taylor. L., Carlson. R., Sobolev. N., 1995b. Archaean Re-Os age for Siberian eclogites and constraints on Archaean tectonics. Nature 374, 711–713. Pokhilenko, N., 2009. Polymict breccia xenoliths: Evidence for the complex character of kimberlite formation. Lithos 112, 934–941. Pokhilenko, N.P., Sobolev, N.V., Boyd, F.R., Pearson, D.D., Shimizu, N., 1993. Megacryslalline pyrope peridotites in the lithosphere of the Siberian Platform: mineralogy, geochemical peculiarities and the problem of their origin. Geologiya i Geofizika (Russian Geology and Geophysics) 34 (1), 71–84 (56–67). Pokhilenko, L.N., Fedorov, I.I., Pokhilenko, N.P., Tomilenko, A.A., 1994. Fluid regime of formation of mantle rocks according to data of chromatographic analysis and thermodynamics calculations. Geologiya i Geofizika (Russian Geology and Geophysics) 35 (4), 67–70 (60–63). Pokhilenko, N., Sobolev, N., Kuligin, S., Shimizu, N., 1999. Peculiarities of distribution of pyroxenite paragenesis garnets in Yakutian kimberlites and some aspects of the evolution of the Siberian craton lithospheric mantle, in: Proc. 7th International Kimberlite Conference. Red Roof Design, Cape Town, pp. 689–698. Pokhilenko, N., Sobolev, N., Reutsky, V., Hall, A., Taylor, L., 2004. Crystalline inclusions and C isotope ratios in diamonds from the Snap Lake/King Lake kimberlite dyke system: evidence of ultradeep and enriched lithospheric mantle. Lithos 77, 57–67. Pokhilenko, L.N., Pokhilenko, N.P., Fedorov, I.I., Tomilenko, A.A., Usova, L.V., Fomina, L.N., 2008. Fluid regime peculiarities of the lithoshere mantle of the Siberian platform, in: Vladykin, N.V. (Ed.), Deep-seated Magmatism, its Sources and Plumes. Proc. VIII International Workshop, Russian Academy of Sciences. Irkutsk, pp. 122–136. Pokhilenko, L.N., Malkovets, V.G., Agashev, A.M., Griffin, W., 2012. The geochemical features of the garnets from peridotites of Udachnaya pipe (Yakutia), in: Earth and Evolution, Proc. 22nd Goldschmidt Conf. Montreal, Canada. Pokhilenko, L.N., Mal’kovets, V.G., Kuz’min, D.V., Pokhilenko, N.P., 2014. New data on the mineralogy of megacrystalline pyrope peridotite from the Udachnaya kimberlite pipe, Siberian Craton, Yakutian diamondiferous province. Dokl. Earth Sci. 454, 179–184. Rege, S., Griffin, W., Kurat, G., Jackson, S., Pearson, N., O’Reilly, S.Y., 2008. Trace-element geochemistry of diamondite: Crystallisation of diamond from kimberlite–carbonatite melts. Lithos 106, 39–54. Rege, S., Griffin, W.L., Pearson, N.J., Araujo, D., Zedgenizov, D., O’Reilly, S.Y., 2010. Trace-element patterns of fibrous and monocrystalline diamonds: Insights into mantle fluids. Lithos 118, 313–337. Rohrbach, A., Schmidt, M.W., 2011. Redox freezing and melting in the Earth’s deep mantle resulting from carbon-iron redox coupling. Nature 472, 209–212. Sharygin, I.S., Litasov, K.D., Shatsky, A.F., Golovin, A.V., Ohtani, E., Pokhilenko, N.P., 2013. Melting of kimberlite of the Udachnaya-East pipe: experimental study at 3–6.5 GPa and 900–1500 °C. Dokl. Earth Sci. 448 (2), 200–205. Sharygin, I.S., Litasov, K.D., Shatskiy, A., Golovin, A.V., Ohtani, E., Pokhilenko, N.P., 2015. Melting phase relations of the Udachnaya-East Group I kimberlite at 3.0–6.5 GPa: experimental evidence for alkali carbonatite composition of primary kimberlite melt and implications for mantle plumes. Gondwana Res., doi:10.1016/j.gr.2014.10.005. Shatskiy, A., Litasov, K.D., Matsuzaki, T., Shinoda, K., Yamazaki, D., Yoneda, A., Ito, E., Katsura, T., 2009. Single crystal growth of wadsleyite. Am. Mineral. 94, 1130–1136. Shatskiy, A., Litasov, K.D., Borzdov, Y.M., Katsura, T., Yamazaki, D., Ohtani, E., 2013a. Silicate diffusion in alkali-carbonatite and hydrous melts at 16.5 and 24 GPa: Implication for the melt transport by dissolution-precipitation in the transition zone and uppermost lower mantle. Phys. Earth Planet. Inter. 225, 1–11. Shatskiy, A., Sharygin, I.S., Litasov, K.D., Borzdov, Y.M., Palyanov, Y.N., Ohtani, E., 2013b. New experimental data on phase relations for the system Na2CO3–CaCO3 at 6 GPa and 900–1400 °C. Am. Mineral. 98, 2164–2171. Shatskiy, A.F., Litasov, K.D., Palyanov, Yu.N., 2015. Phase relations in carbonate systems at pressures and temperatures of lithospheric mantle:

N.P. Pokhilenko et al. / Russian Geology and Geophysics 56 (2015) 280–295 review of experimental data. Russian Geology and Geophysics (Geologiya i Geofizika) 56 (1–2), 113–142 (149–187). Shchukina, E., Agashev, A., Golovin, N., Pokhilenko, N., 2013. Evidence of mantle metasomatism in garnet peridotites from V. Grib kimberlite pipe (Arkhangelsk region, Russia), EGU General Assembly Conference Abstracts, p. 9456. Shimizu, N., Pokhilenko, N., Boyd, F., Pearson, D., 1999. Trace element characteristics of garnet dunites/harzburgites, host rocks for Siberian peridotitic diamonds, in: Proc. VIIth Int. Kimberlite Conf. Red Roof Design, Cape Town, pp. 773–782. Shirey, S.B., Richardson, S.H., 2011. Start of the Wilson cycle at 3 Ga shown by diamonds from subcontinental mantle. Science 333, 434–436. Shirey, S.B., Cartigny, P., Frost, D.J., Keshav, S., Nestola, F., Nimis, P., Pearson, D.G., Sobolev, N.V., Walter, M.J., 2013. Diamonds and the geology of mantle carbon. Rev. Mineral. Geochem. 355–421. Simon, N.S., Irvine, G.J., Davies, G.R., Pearson, D.G., Carlson, R.W., 2003. The origin of garnet and clinopyroxene in “depleted” Kaapvaal peridotites. Lithos 71, 289–322. Simon, N.S., Carlson, R.W., Pearson, D.G., Davies, G.R., 2007. The origin and evolution of the Kaapvaal cratonic lithospheric mantle. J. Petrol. 48, 589–625. Sobolev, N.V., 1974. Mantle Xenoliths in Kimberlites: Implications for Upper Mantle Compositions [in Russian]. Nauka, Novosibirsk. Sobolev, N.V., Lavrentiev, Iu.G., Pospelova, L.N., Sobolev, E.V., 1969. Cr-pyropes from Yakutian diamonds. Dokl. Akad. Nauk SSSR 189 (1), 162–165. Sobolev, N., Lavrent’ev, Y.G., Pokhilenko, N., Usova, L., 1973. Chrome-rich garnets from the kimberlites of Yakutia and their parageneses. Contrib. Mineral. Petrol. 40, 39–52. Sobolev, N.V., Efimova, E.S., Reymers, L.F., Zakharchenko, O.D., Makhin, A.I., Usova, L.V., 1997. Mineral inclusions in diamonds of the Arkhangelsk kimberlite province. Geologiya i Geofizika (Russian Geology and Geophysics) 38 (2), 358–370 (379–393). Sobolev, N.V., Logvinova, A.M., Zedgenizov, D.A., Yefimova, E.S., Seryotkin, Y.V., Floss, K., Taylor, L.A., 2004. Mineral inclusions in microdiamonds and macrodiamonds from Yakutian kimberlites: a comparative study. Lithos 77, 225–242. Sobolev, N.V., Logvinova, A.M., Efimova, E.S., 2009. Syngenetic phlogopite inclusions in kimberlite-hosted diamonds: implications for role of volatiles in diamond formation. Russian Geology and Geophysics (Geologiya i Geofizika) 50 (12), 1234–1252 (1588–1606). Sokol, A.G., Kruk, A.N., 2015. Conditions of kimberlite magma generation: experimental constraints. Russian Geology and Geophysics (Geologiya i Geofizika) 56 (1–2), 245–259 (316–336). Sokol, A.G., Pal’yanov, Y.N., Pal’yanova, G.A., Khokhryakov, A.F., Borzdov, Y.M., 2001. Diamond and graphite crystallization from C-O-H fluids under high pressure and high temperature conditions. Diamond Relat. Mater. 10, 2131–2136. Sokol, A.G., Palyanova, G.A., Palyanov, Y.N., Tomilenko, A.A., Melenevsky, V.N., 2009. Fluid regime and diamond formation in the reduced mantle: Experimental constraints. Geochim. Cosmochim. Acta 73, 5820– 5834. Sokol, A.G., Palyanov, Y.N., Kupriyanov, I.N., Litasov, K.D., Polovinka, M.P., 2010. Effect of oxygen fugacity on the H2O storage capacity of forsterite in the carbon-saturated systems. Geochim. Cosmochim. Acta 74, 4793–4806. Sokol, A.G., Kupriyanov, I.N., Palyanov, Y.N., 2013a. Partitioning of H2O between olivine and carbonate–silicate melts at 6.3 GPa and 1400 °C: Implications for kimberlite formation. Earth Planet. Sci. Lett. 383, 58–67. Sokol, A.G., Kupriyanov, I.N., Palyanov, Y.N., Kruk, A.N., Sobolev, N.V., 2013b. Melting experiments on the Udachnaya kimberlite at 6.3–7.5 GPa: Implications for the role of H2O in magma generation and formation of hydrous olivine. Geochim. Cosmochim. Acta 101, 133–155. Stachel, T., Harris, J., 2008. The origin of cratonic diamonds–constraints from mineral inclusions. Ore Geol. Rev. 34, 5–32. Stagno, V., Frost, D.J., 2010. Carbon speciation in the asthenosphere: Experimental measurements of the redox conditions at which carbonate-

295

bearing melts coexist with graphite or diamond in peridotite assemblages. Earth Planet. Sci. Lett. 300, 72–84. Stagno, V., Ojwang, D.O., McCammon, C.A., Frost, D.J., 2013. The oxidation state of the mantle and the extraction of carbon from Earth’s interior. Nature 493, 84–88. Sumino, H., Kaneoka, I., Matsufuji, K., Sobolev, A.V., 2006. Deep mantle origin of kimberlite magmas revealed by neon isotopes. Geophys. Res. Lett. 33, L16318 http://dx.doi.org/16310.11029/12006GL027144. Sweeney, R.J., 1994. Carbonatite melt compositions in the Earth’s mantle. Earth Planet. Sci. Lett. 128, 259–270. Tachibana, Y., Kaneoka, I., Gaffney, A., Upton, B., 2006. Ocean-island basalt-like source of kimberlite magmas from West Greenland revealed by high 3He/4He ratios. Geology 34, 273–276. Tappe, S., Steenfelt, A., Heaman, L.M., Simonetti, A., 2009. The newly discovered Jurassic Tikiusaaq carbonatite-aillikite occurrence, West Greenland, and some remarks on carbonatite–kimberlite relationships. Lithos 112, 385–399. Taylor, W.R., Green, D.H., 1988. Measurement of reduced peridotite C-O-H solidus and implications for redox melting of the mantle. Nature 332, 349–352. Tomilenko, A.A., Shatskiy, V.S., Ragozin, A.L., Shebanin, A.P., 2001. Variation in the fluid phase composition in the process of natural diamond crystallization. Dokl. Earth Sci. 379 (5), 571–574. Tomilenko, A.A., Kovyazin, S.V., Pokhilenko, L.N., Sobolev, N.V., 2009. Primary hydrocarbon inclusions in garnet of diamondiferous eclogite from the Udachnaia kimberlite pipe, Yakutia. Dokl. Earth Sci. 426 (4), 695–698. Torsvik, T.H., Burke, K., Steinberger, B., Webb, S.J., Ashwal, L.D., 2010. Diamonds sampled by plumes from the core-mantle boundary. Nature 466, 352–355. Tychkov, N.S., Agashev, A.M., Malygina, E.V., Nikolenko, E.I., Pokhilenko, N.P., 2014. Thermal perturbations in the lithospheric mantle as evidenced from P–T equilibrium conditions of xenoliths from the Udachnaya kimberlite pipe. Dokl. Earth Sci. 454 (1), 84–88. Ulmer, P., Sweeney, R.J., 2002. Generation and differentiation of group II kimberlites: Constraints from a high-pressure experimental study to 10 GPa. Geochim. Cosmochim. Acta 66, 2139–2153. Veksler, I., Petibon, C., Jenner, G., Dorfman, A., Dingwell, D., 1998. Trace element partitioning in immiscible silicate–carbonate liquid systems: an initial experimental study using a centrifuge autoclave. J. Petrol. 39, 2095–2104. Walter, M.J., Bulanova, G.P., Armstrong, L.S., Keshav, S., Blundy, J.D., Gudfinnsson, G., Lord, O.T., Lennie, A.R., Clark, S.M., Smith, C.B., Gobbo, L., 2008. Primary carbonatite melt from deeply subducted oceanic crust. Nature 454, 622–630. Walter, M., Kohn, S., Araujo, D., Bulanova, G., Smith, C., Gaillou, E., Wang, J., Steele, A., Shirey, S., 2011. Deep mantle cycling of oceanic crust: evidence from diamonds and their mineral inclusions. Science 334, 54–57. Weinberg, R.F., Podladchikov, Y., 1994. Diapiric ascent of magmas through power law crust and mantle. J. Geophys. Res. Solid Earth (1978–2012) 99, 9543–9559. Woodland, A.B., Koch, M., 2003. Variation in oxygen fugacity with depth in the upper mantle beneath the Kaapvaal craton, Southern Africa. Earth Planet. Sci. Lett. 214, 295–310. Yaxley, G.M., Berry, A.J., Kamenetsky, V.S., Woodland, A.B., Golovin, A.V., 2012. An oxygen fugacity profile through the Siberian Craton—Fe K-edge XANES determinations of Fe3+/ΣFe in garnets in peridotite xenoliths from the Udachnaya East kimberlite. Lithos 140, 142–151. Zedgenizov, D.A., Rege, S., Griffin, W.L., Kagi, H., Shatsky, V.S., 2007. Composition of trapped fluids in cuboid fibrous diamonds from the Udachnaya kimberlite: LAM-ICPMS analysis. Chem. Geol. 240, 151–162. Zedgenizov, D.A., Shatskiy, A., Ragozin, A.L., Kagi, H., Shatsky, V.S., 2014. Merwinite in diamond from Sao Luiz, Brazil: A new mineral of the Ca-rich mantle environment. Am. Mineral. 99, 547–550. Zhang, C., Duan, Z.H., 2009. A model for C-O-H fluid in the Earth’s mantle. Geochim. Cosmochim. Acta 73, 2089–2102.