Graphite pseudomorphs after diamonds: An

Lithos 236–237 (2015) 16–26

Contents lists available at ScienceDirect

Lithos journal homepage: www.elsevier.com/locate/lithos

Graphite pseudomorphs after diamonds: An experimental study of graphite morphology and the role of H2O in the graphitisation process Andrey V. Korsakov a,b,⁎, Egor I. Zhimulev a, Denis S. Mikhailenko a, Sergey P. Demin c, Olga A. Kozmenko a a b c

V.S. Sobolev Institute of Geology and Mineralogy of the Siberian Branch of the RAS, 3, Ac. Koptyuga Ave., Novosibirsk 630090, Russian Federation Novosibirsk State University, Pirogova St. 2, Novosibirsk 630090, Russian Federation JV Tairus, 3, Ac. Koptyuga Ave., Novosibirsk 630090, Russia

a r t i c l e

i n f o

Article history: Received 28 May 2014 Accepted 19 August 2015 Available online 3 September 2015 Keywords: Diamond Graphite Metastable graphite Pseudomorphs UHPM

a b s t r a c t Experiments at 1 atmosphere and 2.0–2.5 GPa over a range of temperatures of 1400–2100 °C have been carried out to investigate the diamond-to-graphite transformation in “dry” and “wet” systems. Internal and external morphologies of graphite pseudomorphs after diamonds were studied by Raman spectroscopy, reflected light microscopy and scanning electron microscopy. In a “dry” system, the results show that at 2.0–2.5 GPa, graphite pseudomorphs preserve the finest details of external morphology of original diamond crystals, whereas at 1 atmosphere only the general outline of diamond crystals can be recognised on these pseudomorphs. In all experimental runs at P = 2.0–2.5 GPa under various temperatures, the growth of oriented graphite crystallites was observed only on the {111} diamond faces, while randomly oriented graphite crystals were observed on the {100} and especially {110} diamond faces. In a “wet” system, we were unable to reproduce graphite pseudomorphs after diamond. However, newly formed large graphite crystals were found on the partly dissolved diamonds. The diamond crystal form has been changed from that of a sharp octahedron to octahedron having rounded edges and corners. Flat-bottomed negatively oriented trigons were formed on the octahedral {111} faces. Large graphite crystals tend to be concentrated at {100} and {110} surfaces. Rare single euhedral graphite flakes occur on the {111} faces. Visible growth spirals on {001} faces of graphite crystals appear on all crystals. The {111} faces of diamond crystal are partly covered by translucent graphite coat. The diamond-to-graphite transformation in the “wet” system occurs via coupled dissolution–precipitation processes. All morphological features (e.g., negatively oriented trigons, rounded edges) described for partly dissolved diamonds are easily recognised. None of these features have been detected so far for partly graphitised metamorphic diamond crystals. Our results suggest that in geological environments, the presence of volatiles (in particular, H2O) definitely cannot be considered as favourable for the diamond-to-graphite transformation and, thus, show that this hypothesis seems highly unlikely. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Complete graphite pseudomorphs after diamonds described from Beni-Bousera, northern Morocco, (El Atrassi et al., 2011; Pearson et al., 1989, 1995; Slodkevich, 1982) and from Ronda, southern Spain (Davies et al., 1992), include the following: (1) sharp-edged octahedra, with or without rounded fibrous graphite coats (the most abundant form); (2) rhombicuboctahedra exhibiting well-formed cube and octahedral faces; (3) contact twins or macles, with and without re-entrant angles, most of which are coated; and (4) irregular to rounded masses whose surface morphologies resemble framesite. Findings of graphite cuboids from Maksutov complex, Ural Mountains, Russia, are also

⁎ Corresponding author. E-mail address: [email protected] (A.V. Korsakov).

http://dx.doi.org/10.1016/j.lithos.2015.08.012 0024-4937/© 2015 Elsevier B.V. All rights reserved.

interpreted as graphite pseudomorphs after diamonds (Leech & Ernst, 1998), although so far no trace of UHPM events were recognised for this complex. In contrast to kimberlites, where black and/or graphite-coated diamonds rarely occur (Grenville-Wells, 1952), diamonds in the UHPM rocks often coexist with graphites (Dobrzhinetskaya et al., 2009; Janak et al., 2013; Katayama et al., 2000; Korsakov & Shatsky, 2004; Korsakov et al., 2010; Massonne et al., 1998; Naemura et al., 2011; Ogasawara et al., 2000; Perraki et al., 2006, 2009; Zhang et al., 1997; Zhu & Ogasawara, 2002). Metastable growth of graphite in the diamond stability field was documented for ultra-high-pressure metamorphic rocks from the Kokchetav massif (Korsakov & Shatsky, 2004; Korsakov et al., 2010). Cuboid is predominant morphology of metamorphic diamonds; however, perfect octahedral crystals occur in clinozoisite gneisses and quartz–tourmaline–muscovite metasomatic rocks (Korsakov et al., 2002, 2009; Lavrova et al., 1999; Shimizu &

A.V. Korsakov et al. / Lithos 236–237 (2015) 16–26

Ogasawara, 2013). Therefore, the presence of graphite cuboids in UHPM rocks is predominantly interpreted as partial graphitisation of diamonds in terms of solid-state reaction or dissolution–precipitation processes (Dobrzhinetskaya et al., 2009; Janak et al., 2013; Massonne et al., 1998; Ogasawara et al., 2000; Zhang et al., 1997; Zhu & Ogasawara, 2002). However, it is very difficult to find P–T conditions at which the diamond dissolution and graphite precipitation reactions take place simultaneously (Putnis & Putnis, 2007). In all previous experimental studies (Arima & Kozai, 2008; Chou & Anderson, 2009; Fedortchouk & Canil, 2009; Fedortchouk et al., 2011; Khokhryakov & Pal'yanov, 2007; Kozai & Arima, 2005), only dissolution of diamond was investigated in detail, but none of those publications reported the precipitation of graphite. The presence of very thin film of graphite was reported by Kozai and Arima (2005), but no dissolution features were documented in those runs. Unlike the case of experiments, diamond graphitisation in natural samples is often attributed to the effect of an “omnipotent” fluid, which would drastically change the situation and promote diamond conversion to graphite within a relatively short time, even under unsuitable P–T conditions (Bostick et al., 2003; Dobrzhinetskaya et al., 2009; Leech & Ernst, 1998). This study aims (i) to present the peculiar features of graphite pseudomorphs after diamond, which allow to distinguish such pseudomorphs from graphite coating around diamond that formed due to metastable growth, and (ii) to test the possible catalytic effect of H2O on the diamond-to-graphite transformation. Additionally, in this paper, we substantiate that graphite pseudomorphs after diamond is hard to produce even in the presence of a significant amount of fluid.

2. Experiment In this study, we used synthetic diamond crystals of cubooctahedron form with {111}, {100} and {110} faces. They were synthesised in the Fe–Ni–C system at P = 5.5–6 GPa and T = 1350–1450 °C, for details to refer Chepurov et al. (1997). All the crystals were perfectly faceted, with sharp edges and smooth faces, containing inclusions of sizes less than 5 μm. Natural diamonds selected from Udachnaya and Diavik mines had octahedron, cuboid or irregular morphologies. The size of diamond crystals was 0.5–2.5 mm.

Table 1 Experimental results. Grf—thin translucent film of graphite, Gr—graphite aggregate, ND and SD—natural and synthetic diamond crystal, respectively. Run n0

Time

P, GPa

T, °C

Diamond

Phases

“Dry” experiments LP1 10 min LP2 20 min LP3 1h LP4 1h LP5 12 h LP6 12 h 5-29-07 20 min 5-32-07 23 min 5-35-07 5 min 5-40-07 20 min 5-48-07 30 min 5-4-08 5 min 5-10-08 2h 5-13-08 24 h 4-8-08 5 min

10–4 10–4 10–4 10–4 10–4 10–4 2.0 2.0 2.0 2.0 2.0 2.0 2.5 2.5 2.0

1500 1800 2000 1500 1500 1800 1800 1800 2050 2050 2050 2050 1250 1250 2150

2 SD 2 SD 2 SD 2 SD 2 SD 2 SD 2 SD 2 SD 2 SD 2 SD 2 ND 2 ND +2 SD 2 ND +2 SD 2 ND +2 SD 2 SD

MgO + Grf MgO + Gr MgO + Gr MgO + Grf MgO + Grf MgO + Gr MgO + Gr MgO + Gr MgO + Gr MgO + Gr MgO + Gr MgO + Gr MgO + Grf MgO + Gr MgO + Gr

“Wet” experiments 5-22-08 1h 5-29-08 15 min 4-4-08 1h 4-5-08 1h 4-9-08 1h 4-1-14 2h

2.5 2.5 2.5 2.5 2.5 2.0

1550 1620 1600 1550 1560 1470

2 SD 2 SD 2 SD 3 SD 2 SD 2 ND+ 1 SD

Ca(OH)2 + Gr Ca(OH)2 + Gr Ca(OH)2 + Gr Ca(OH)2 + Gr Ca(OH)2 + Gr Ca(OH)2 + Gr

17

High-pressure experiments on diamond graphitisation were carried out using a “split-sphere” (BARS) type multi-anvil apparatus at P–T range 2.0–2.5 GPa and T = 1400–2000 °C (Table 1). A high-pressure cell in the form of a tetragonal prism 20 × 20 × 23 mm3 in size, with cut edges and apexes, containing a graphite heater with 11 mm in diameter was used. The temperature was measured using a PtRh6/PtRh30 thermocouple. Details on the pressure and temperature calibration are given by Chepurov et al. (1997, 2012, 2013). HT experiments (1800– 2100 °C) on diamond graphitisation in “dry” systems were performed without Pt capsules, whereas for 1400–1600 °C, Pt capsules were used. Several crystals were placed into MgO powder with 1–2 μm grain size that subsequently was pressed into solid tablets. These tablets were placed inside graphite heater. Similar MgO tablets were used in diamond graphitisation experiments with CO–CO2 atmosphere and in ambient pressure experiments to prevent breakdown of graphite pseudomorphs. For volatile-bearing systems (“wet”), Pt capsules with 6 mm in diameter and with wall width of 0.5 mm thick were utilised. Several diamond crystals were placed inside Ca(OH)2 and pressed into solid tablets (Table 1). Reagents MgO and Ca(OH)2 with a purity of 99.9% were used as starting materials. After treatment, graphite/diamond crystals were extracted from MgO and Ca(OH)2 by heating in acid (to prevent the destruction of the pseudomorphs) and then were studied by Raman spectroscopy and scanning electron microscopy (SEM) methods. The graphitised diamond crystals were mounted into epoxy (Petropoxy 154). They were cut with high-speed diamond saw with thin diamond blade (thickness 0.2 mm and diameter 130 mm). Then crystals were polished with special diamond lap (diamond size 20 μm). Further details of diamond polishing can be found elsewhere (De Corte et al., 2000; Korsakov & Shatsky, 2004; Korsakov et al., 2010). Final polishing was performed using soft lap and 1 μm diamond paste. Grain size of artificial diamond particles from diamond-bearing materials at all stages of sample preparation was always less than that of original diamond crystals and diamonds relics (up to 300 μm in size) found in cores of graphitised diamond, and thus, the latter ones cannot be regarded as a result of contamination of experimental products by preparation material. Raman spectra were obtained using a Horiba LabRam HR800 system. Excitation wavelength 532 nm was used with a power of 10 mW. Scattered light was dispersed using a holographic grating with 1800 grooves per millimetre. Spectral resolution was about 3.0 cm–1. An Olympus BH-2 microscope with a 100 × objective allowed collecting scattered light from spots as small as 1 μm in diameter. Raman spectra were recorded from 100 to 4000 cm–1. Similarly to Naemura et al. (2011), band position, band height, band area and band width (i.e., full width at half maximum, FWHM) were determined after baseline correction and curve fitting with the Voigt function by applying data analysis software Fityk 0.8.9. The study of diamond faces and graphite layers with back-scattered electron imaging (BSE) was performed on a scanning electron microscope (SEM) JEOL SM-6100 using 20 kV accelerating voltage, probe current over a range of 10–12–10–6 A, and maximum magnification up to 20000. Reflected optical microscopy (ROM) provides important information about the internal morphology of graphite pseudomorphs after diamond. An Olympus BH-2 microscope with 10, 20, 50 and 100 × objectives was used. 3. Results All experiments with temperature below 1500 °C and different duration (up to 12 hours) reveal that only a thin translucent layer of graphite, defined by Raman spectroscopy, appears on diamond crystals in the first 10 minutes and no changes in thickness or morphology of graphite coatings were detected at long-term experiments (Table 1).

18


3.1. “Dry” experiments Graphite pseudomorphs after diamond, obtained at 1 atmosphere, are very fragile, and only general outlines of preexisting diamond crystals are recognisable (Fig. 1a, b). They consist of very fine-grained aggregates of graphite, and it is not possible to verify the orientation of graphite crystals with respect to diamond crystal (Fig. 1a, b). Graphite pseudomorphs obtained at high-pressure runs are solid and consist of relatively large graphite crystals up to 20 μm in size (Fig. 1c–f). There is preferred orientation of graphite crystals on octahedral plane of diamonds, {001} plane of graphite crystal is parallel to (111) plane of diamond crystal. However, randomly oriented graphite crystals occur on {100} and especially on {110} planes of diamond crystals (Fig. 1e–f). In order to examine an influence of diamond morphology on graphite pseudomorphs, cuboid and complex diamond crystals were used. The original diamond morphology can be seen on Fig. 2a, c, e.

After experimental runs 5-48-07 and 5-4-08, even very fine details of original diamond crystals can be recognised on graphitised diamonds with complex morphology (Fig. 2). It is difficult to estimate the real size of graphite crystals on {111} planes of diamond because the graphite is predominantly oriented along the {111} planes (Fig. 2d, f). However, the study of the cross section of graphite pseudomorphs reveals that octahedral sectors are occupied by large graphite crystals (up to 2 mm in diameter and thickness up to 0.2 mm Fig. 3). Graphite crystals up to 50 μm in diameter are observed on surfaces {110} of original diamond crystals (Fig. 1f), while in cross sections of pseudomorphs, the sectors of {110} are occupied by coarse-grained graphites with orientation of (001) parallel to one of octahedron faces (Fig. 3d). At higher magnifications, the large grain of graphites appears not homogeneous with respect to reflectance dichroism (Fig. 3e–f). The fine-grained graphites can be recognised as inclusions in large graphite crystals (Fig. 3f). Cuboid diamond crystals are covered by randomly oriented graphite crystals, but some minor depressions found on

Fig. 1. Graphite pseudomorphs after diamonds obtained in “dry” experiments. (a, b) Completely graphitised octahedral diamond crystal at P = 1 atmosphere, T = 2000, t = 60 min (run LP3). (c, d) Partly graphitised octahedral diamond crystal at P = 2.0 GPa, T = 1800°C, t = 20 min (run 5-29-07). (e, f) Partly graphitised cub-octahedral diamond crystal at P = 2.0 GPa, T = 2050°C, t = 20 min (run 5-40-07).


19

Fig. 2. Graphite pseudomorphs after diamonds with different starting shape of diamond crystals obtained in “dry” experiments. (a, b) Original and graphitised cuboid diamond crystal (experimental run 5-4-08). (c, d) Original and graphitised complex morphology diamond crystal (experimental run 5-48-07). (e, f) Original and graphitised complex diamond crystal (experimental run 5-48-07).

original crystals can still be seen. The graphite crystal size on {100} of diamond varies from 5 to 20 μm (Fig. 1d). 3.2. “Wet” experiments: influence of H2O on mechanism of diamond graphitisation In all “wet” experiments, diamonds were found in the lower part of the ampule, indicating that liquid (with density lower than diamond density) was formed in the ampule during experiments. It is worth noting that before experiments all diamond crystals were of sharp octahedral and cub-octahedral shape, with flat plane and sharp edges. After experiments, octahedral diamonds have shown rounded edges and corners (Fig. 4a). Many small negatively oriented trigons develop on the {111} (Figs. 4a and 5a), and they are considered as diagnostic feature of diamond dissolution in H2O-rich fluid (Fedortchouk et al., 2011; Khokhryakov & Pal'yanov, 2007). The trigons have flat bottoms parallel to the {111} face with flat walls (Fig. 5b and d). Similar trigons were obtained on diamond after

dissolution experiments with H 2O-rich fluid at 1150–1350 °C and 1.0 GPa by Fedortchouk et al. (2011), but no newly formed graphite crystals were found in their experiments. Thin translucent graphite layers cover almost all diamond crystals; however, most negatively oriented trigons show no traces graphite (Fig. 4). Abundant euhedral graphite crystals (up to 200 μm in diameter and ~10 μm thick) and their rosette-like intergrowths occur along the edges and corners (Figs. 4–6). The orientation of graphite crystals with respect to diamond is almost random. Predominantly, the (001) face of graphite crystals is inclined by different degree to diamond surfaces (Figs. 4 and 5). However, in some rare case, the (001) face of graphite crystals is subparallel to (111) diamond face (Fig. 6b). Almost all graphite crystals have clear evidence for predominant spiral growth mechanism formation (Fig. 5). The centre of the spiral lies roughly at the centre of the (001) face. Steps at the centre of the spiral appear to be rounded (Fig. 5c and f). The step spacing in the centre of the spiral is on average ~ 1 μm. The step spacing on the outer portion of the spiral is about ~10 μm (Fig. 5f).

20


Fig. 3. Reflected light photomicrographs (crossed polarisers) of graphite replacing diamond crystal, obtained at P = 2.0 GPa, T = 2050°C, t = 20 min run 5-40-07 (Fig. 1e). (a) An overview of partly graphitised cub-octahedral diamond crystal. Octahedron sectors are traced with red colour. The large graphite crystals with the same crystallographic orientations display similar reflectance dichroism. Yellow rectangulars indicate the locations of photomicrographs (b–f). (b) Relics of diamond (Dia) surrounded by coarse-grained graphite (Gr-L) with interstitial fine-grained “vein-like” graphite (Gr-S). (c) The coarse-grained graphite replacing diamond. The diamond octahedron, representing relict of original diamond crystal, can be recognised and the (001) of graphite crystal is parallel to the (111) diamond relict. (d) Graphite crystals with different orientations appearing in (110) sector, which is marked by yellow line. (e, f) Morphology of the large deformed graphite crystals. The large graphite crystals always have “inclusions” of fine-grained graphites with other crystallographic orientations. Locally the finegrained graphite looks like a “vein,” and it is very similar to “vein-like” graphites presented on (b), but at other scale. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

During experiments, diamond dissolution and precipitation of graphite occur nearly simultaneously because otherwise instead of euhedral graphite crystals, we should have quenched carbon, which is typically observed in experiments on diamond growth from nonmetallic system (Sokol et al., 2001). Thus, this process can be considered as solvent-mediated diamond-to-graphite phase transformation, but this mechanism does not produce pseudomorphs of graphite after diamond. After graphite on the diamond was removed using a mixture of K2Cr2O7 + H2SO4, we observed unusual morphological features similar to snowflakes (Fig. 7) on the {100} and {110} surfaces of diamond crystals (Fig. 7b, c). Similar features were documented by Sokol et al. (2001), and they were interpreted as prints of joint diamond–graphite growth (see their Fig. 5b, c).

3.3. Raman spectroscopic study of graphite All graphite pseudomorphs after diamond were examined by Raman spectroscopy. First-order Raman spectra were analysed from 1100 to 1800 cm–1. Representative Raman spectra of graphite are shown in Fig. 8. They have two bands in the first-order region, at ~ 1360 (D1) and ~1580 (G) cm–1, with a shoulder at ~1620 (D2) cm–1. The intensity ratio of the G band to the D1 band strongly decreases when the angle between the “c” crystal axis and the optical axis of the laser beam increases from 0 to 90° (Fig. 9). Similar results were documented by Wang et al. (1989). Only the broad G band is observed in the Raman spectra of translucent graphite films, occurring on {111} diamond faces (Fig. 8c magenta and green spectra). The D1 band overlaps with the diamond


21

band, and its intensity cannot be quantified (Fig. 8c). The Raman spectra collected at different points of graphite pseudomorphs after diamond obtained at 1 atmosphere are very similar to each other (Fig. 8a), irrespective of the original diamond faces {100}, {110} or {111}. In general, the D1/G ratio is higher for graphite pseudomorphs obtained at 1 atmosphere (Table 2). The variations of D1/G ratio are likely controlled by the orientation of the incident laser to the “c” plane of graphite crystals rather than by the order–disorder pattern (Table 2). Contrary to Pantea et al. (2002b), no additional Raman bands were found in our study. All graphite crystals composing the pseudomorphs have the same order–disorder pattern independently from their size and morphology. However, the second order bands in Fig. 8a are significantly broader and more symmetrical than those in Fig. 8b and c. Unfortunately, in case of macrocrystalline graphite, Raman spectroscopy technique faces with instrumental limitations and should not be used for the estimation of crystallographic parameters (Badenhorst, 2014). Translucent graphite coatings have very weak Raman signal (Fig. 8c); however, in rare cases, both graphite and diamond bands can be recognised in Raman spectra (Fig. 8c). 4. Discussion and concluding remarks

Fig. 4. Optical images of graphite-coated diamond crystal obtained in Diamond-Ca(OH)2 system at 1450 °C and 2.0 GPa (run 5-22-08). (a) Overview of diamond crystal with rounded edges and corners. Original diamond crystal had flat plane and sharp edges similar to Fig. 1c. The yellow colour of original diamond crystal can be seen in negatively oriented trigons. The thin graphite film on {111} face of diamond crystal is translucent. It has light grey colour. (b) Details of diamond {111} face and graphite rosettes morphologies. The graphite rosettes are concentrated at the former {100} diamond face, however, rare graphite rosettes occur on {111} diamond face. The graphite crystals are randomly oriented with respect to diamond crystal. However, in rare cases, the graphite face {001} is subparallel to {111} face of diamond crystal (Fig. 6b). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The coexistence of diamond and graphite in UHPM rocks was documented almost for all diamond-bearing complexes (De Corte et al., 2000; Frezzotti et al., 2011; Janak et al., 2013; Katayama et al., 2000; Korsakov et al., 2002; Massonne, 1999; Mposkos & Kostopoulos, 2001; Perraki et al., 2006; Shatsky et al., 1995; Smith & Godard, 2013; Van Roermund et al., 2002; Vrijmoed et al., 2008). These findings are generally interpreted as partial graphitisation of original diamond crystals (Dobrzhinetskaya et al., 2003; Janak et al., 2013; Massonne, 1998; Ogasawara et al., 2000). However, as it was shown recently (Khokhryakov et al., 2009; Korsakov & Shatsky, 2004; Korsakov et al., 2010; Okada et al., 2004; Pal'yanov et al., 2002; Sokol et al., 2001; Yamaoka et al., 2000, 2002), graphite may crystallise metastably in the diamond stability field. It should be noted that for some UHPM complexes, peak metamorphic conditions are estimated to be no more

Fig. 5. Reflected light images of graphite-coated diamonds obtained in Diamond-Ca(OH)2 system at 1550 °C and 2.5 GPa (run 5-22-08). (a and d) Diamond with thin translucent graphite layer, and abundant graphite crystals at the corners and edges (run 5-22-08). (b and e) Distribution of negatively oriented trigons and graphite thin layer on (111) diamond surfaces. Diamond and graphite can be distinguished by differences in reflective index. In reflected light graphite (light grey colour) appears slightly brighter than diamond (dark grey colour). It is worth noting that bottom of negatively oriented trigons never have any traces of graphite. (c and f) Representative images of the large graphite crystals and graphite rosettes grew by spiral growth mechanism. The growth spirals on {001} faces are observed. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

22


Fig. 6. SEM images of graphite-coated diamond crystals presented in Fig. 5. (a–c) Overview of diamond crystals (rung 5-22-08). (b) Graphite crystal (001) face is oriented subparallel to (111) face of diamond crystal. (d) Graphite crystals with various crystallographic orientations relative to diamond surfaces. It is worth noting that similar morphology of joint diamond– graphite growth was documented by (Sokol et al., 2001) see their Fig. 5b-c.

than 800 °C (Janak et al., 2013), or even lower P − T conditions were reported by Frezzotti et al. (2011). At such low temperature, the graphitisation processes were not observed in experiments, even on the contact of diamond hosts with fluid inclusions (Nechaev & Khokhryakov, 2013). Furthermore, well-ordered graphites coexist

with disordered graphite, although the order–disorder patterns in most cases were verified only by Raman spectroscopy (De Corte et al., 2000; Frezzotti et al., 2011; Janak et al., 2013; Katayama et al., 2000; Korsakov et al., 2002; Massonne, 1999; Mposkos & Kostopoulos, 2001; Perraki et al., 2006; Smith & Godard, 2013; Van Roermund et al., 2002;

Fig. 7. SEM images of diamond crystals after dissolution of graphite coating. (a and d) Overview of partly dissolved diamond crystals with rounded edges and corners and negatively oriented trigons on the {111}. (b–e) details of diamond morphology at octahedral corners (“snowflakes”) and {100} diamond surface. (c and f) Snowflakes have step-like surface, similar to prints of joint diamond–graphite growth described by (Sokol et al., 2001).


23

Fig. 8. Representative Raman spectra of graphite crystals obtained in “dry” 1 atmosphere and 2 GPa (a and b, respectively) and “wet” (c) experiments. Details of D1, D2 and G bands are presented on insert in b. Difference in the orientation of {001} plane of the analysed graphite crystal could be responsible for the small peak at ~1360 cm–1. LG—large graphite crystals grown by spiral growth mechanism; SG—translucent graphite film. Orientation ∥ (001) ⊥ (001), (001)—basal graphitic planes (001) are parallel, perpendicular and inclined to the incident laser beam, respectively.

Vrijmoed et al., 2008). Due to a small size of UHPM diamonds and coexisting graphite crystals, it is difficult to control the orientation of these graphite crystals. Therefore, we believe that the findings of the D1 band do not always testify for the disordered nature of graphites. The recent study of industrial and natural graphite by Badenhorst (2014) reveals that Raman spectroscopic study alone cannot be used for identification of order–disorder pattern of graphite. The details of solid-state diamond transformation to graphite have been summarised elsewhere (e.g., Pantea et al., 2002a,b, 2004; Qian et al. 2001, 2004) and will not be reiterated here. The Raman spectroscopic study (Pantea et al., 2002a,b and references therein) reveals substantial differences in the spectra of graphite grown on the {100} and {111} of diamond faces (see their Fig. 4a, b). In the spectra of graphite formed on the {100} face, the 1578 cm–1 G peak was always accompanied by a D2 shoulder centred at 1610 cm–1. For the {111} face, Raman spectra characteristics also vary from one location to another, but the G peak is sharp and almost always more intense than the D1 peak (Pantea et al., 2002b). They explain the difference in Raman spectra by the presence of areas of disoriented graphite and oriented graphite. Graphitisation of {100} and especially {110} diamond sectors proceeds

by a different mechanism (Pantea et al., 2002b) and most probably with a different rate, thus affecting the morphology of graphite aggregates (Fig. 1b–e). In our study, we did not find any difference between Raman spectra of graphite grown on the {100} or {111} faces. Thereto, the difference of ratio D1/G bands due to different orientation (e.g., basal graphitic planes are parallel to the incident laser beam or basal planes are perpendicular to the laser beam) was not so high in our study (Figs. 8 and 9). Purely water or H2O-rich fluid is considered to be one of the most important natural reagents, which stimulates diamond-to-graphite transformation (Dobrzhinetskaya et al., 2003). According to experiments of Qian et al. (2001) at T = 1470 K and P = 2.0 GPa, the discrepancy between the effectiveness of diamond-to-graphite phase transformation in “dry” and “wet” may indicate the catalytic effect of water. Our results indicate that diamond dissolution and graphite precipitation are closely related processes. The presence of well-formed spirals on the graphite crystals indicates their growth by the spiral mechanism (Fig. 5c). They are also suggestive of growth from a fluid phase at relatively low supersaturation (Sungawa, 1984), rather than by a solid-state transformation. In addition, the spirals indicate that

24


Fig. 9. Representative Raman spectra of single graphite crystal in different orientations (a) basal graphitic planes are parallel to the incident laser beam, (b) basal planes are perpendicular to the laser beam, obtained in “wet” experiments.

during growth process the crystals were unrestricted by an interface with another solid (i.e., they grew into a fluid-filled space). The spiral grown graphite is quite rare and it was documented in Precambrian marbles by Rakovan and Jaszczak (2002). Similar features were also reported by Korsakov et al. (2010) on graphite coating surrounding diamond crystals in UHPM rocks from the Kokchetav massif (see their Fig. 7c–d). Therefore, this phenomenon cannot be considered as solidstate mechanism of diamond-to-graphite phase transformation, even if a fluid phase is presented. The dissolution–reprecipitation mechanism also preserves the morphology and transfers crystallographic information from parent to product by epitaxial nucleation (Cardew & Davey,

1985; Putnis & Putnis, 2007). Continuation of the transformation reaction depends on keeping open fluid transport pathways to the reaction interface between parent and product solids (Putnis & Putnis, 2007). The porosity must be generated in the product phase. The generation of porosity depends on two factors: the relative molar volumes of the two solid phases and, more importantly, on the relative solubility of the two phases in the fluid. Graphite has larger molar volume than diamond, and it is highly unlikely that porosity will appear due to diamond-to-graphite transformation. The solubility of diamond and graphite in COH fluids is very low (Palyanov & Sokol, 2009; Sokol & Pal'yanov, 2008); thus, the reaction cannot be proceeded by this mechanism beyond a few layers, and the parent phase is effectively armoured from further interaction with the fluid (Putnis & Putnis, 2007). No dissolution features were found on the surfaces of metamorphic diamonds (De Corte et al., 2000; Korsakov & Shatsky, 2004; Korsakov et al., 2005, 2010; Perraki et al., 2009; Shatsky et al., 1998). Korsakov et al. (2010) reported that after dissolution of graphite coating around metamorphic diamonds, only growth features were observed on all diamond surfaces. Negatively oriented trigons on diamond crystals in our “wet” experiments without even very fine graphite layer occur on {111}, indicating that these trigons serve as a source of carbon for crystallisation of graphite and that there was no precipitation of graphite during the quenching (Figs. 4 and 5a, b and d, e). Similar features for diamond crystals were described for diamond crystals by Fedortchouk et al. (2011), who studied the dissolution of diamond crystals in H2O and CO2 rich fluids. It is worth noting that they did not find any evidence for graphite formation in their experiments. Their results support the idea that recrystalization of diamond to graphite is highly unlikely in natural samples. Recent experimental study of diamond graphitisation on the wall of fluid inclusions reveals that there is no graphitisation of diamond below 800 °C even in the presence of fluid (Nechaev & Khokhryakov, 2013). These results corroborate very well with TEM study of fluid inclusions in UHPM diamonds from the Kokchetav and Erzgebirge (Dobrzhinetskaya et al., 2009; Hwang et al., 2001, 2005, 2006). Up to now, there is no available information about the presence of poorly ordered graphite on the walls of fluid inclusions in metamorphic diamond. However, the poorly ordered graphite on the wall of fluid inclusions was reported for the variety V diamond (Orlov's

Table 2 Textural types of graphites obtained in the experimental runs, and parameters obtained from their Raman spectra. Type—morphological type of graphite: FG = very fine-grained graphite crystals (Fig. 1b), LG = large euhedral graphite crystals (Fig. 4b) and SG = translucent graphite film (Fig. 4b). Orientation ∥ (001) ⊥ (001), (001)—basal graphitic planes (001) are parallel, perpendicular and inclined to the incident laser beam, respectively; n.i.—not identified orientation due to small size of graphite crystals. The R2 = area ratio of D1 / (G + D1 + D2) and calculated temperature [T*(°C) = 641 − 445*R2] after (Beyssac et al., 2002). It is worth noting that temperature estimates obtained by graphite geobarometer (Beyssac et al., 2002) are significantly low that the temperature of the experiments Table 1. Run number

Orientation

Diamond face

Type

G position

FWHM

D1 position

LP3 LP3 LP3 LP3 LP3 LP3 5-22-08 5-22-08 5-22-08 5-22-08 5-22-08 5-22-08 5-22-08 5-22-08 5-22-08 5-22-08 5-32-07 5-32-07 5-32-07 5-32-07 5-32-07 5-32-07

n.i. n.i. n.i. n.i. n.i. n.i. ⊥ (001) (001) ∥ (001) (001) (001) (001) (001) (001) n.i. n.i. (001) (001) (001) (001) (001) (001)

{100} {111} {110} {111} {100} {110} {100} {100} {100} {100} {111} {111} {111} {110} {111} {111} {111} {110} {100} {100} {110} {111}

FG FG FG FG FG FG LG LG LG LG LG LG LG LG SG SG LG LG LG LG LG LG

1579.6 1579.9 1580.1 1580.0 1578.4 1580.0 1578.9 1577.4 1580.0 1579.9 1579.0 1578.7 1578.7 1579.5 1605.3 1580.8 1581.1 1581.1 1581.9 1580.7 1576.5 1580.1

33.08 30.67 28.67 26.32 39.47 37.36 12.27 17.52 13.67 12.65 22.89 21.41 18.31 21.59 279.26 25.85 15.25 19.00 14.41 21.47 28.29 12.63

1346.4 1345.9 1345.9 1346.9 1349.6 1351.4 1352.5 1350.3 1354.9 1353.0 1350.9 1352.9 1352.5 1355.4 – – 1357.6 1350.9 1356.8 1352.0 1354.9 1363.5

FWHM 50.16 48.19 48.13 48.29 68.04 87.49 42.96 88.91 41.52 45.11 51.26 46.05 44.91 67.13 – – 43.25 44.97 48.31 54.30 103.16 145.72

R2

T*(°C)

0.52 0.48 0.44 0.40 0.54 0.58 0.00 0.52 0.28 0.08 0.46 0.26 0.25 0.31 – – 0.14 0.06 0.15 0.20 0.55 0.10

410 427 445 464 401 382 641 409 518 605 437 525 528 501 – – 579 613 576 552 396 597


classification) from placers of the northeastern Siberian Platform (Tomilenko et al., 2001). Graphite inclusions in diamonds appearing as results of some heating event in their history were described by (Howell et al., 2013), but no trace of fluid was found to be associated with these graphite inclusions. The peak metamorphic conditions for the Kokchetav and Erzgebirge were estimated as high as 1000–1100 °C and 6–8 GPa (Massonne, 2003; Mikhno & Korsakov, 2012, 2013; Mikhno et al., 2013; Ogasawara et al., 2002; Okamoto et al., 2000). The lack of graphitisation on the wall of fluid inclusions in metamorphic diamond even for these rocks implies that for other metamorphic complexes, where peak P − T parameters are lower, diamond-to-graphite solvent-mediated phase transformation is highly unlikely. Most probably the crystallisation of poorly ordered graphite on external diamond surface results from the cooling of COH fluids, which were trapped together with diamond in UHPM conditions. In our “wet” experiments, findings of morphological features similar to joint growth of diamond and metastable graphite (Sokol et al., 2001) remain rather enigmatic. There is no report on diamond growth in graphite stability field at 2.0 GPa. Therefore, we assume that these features are likely dissolution features, but further investigation is required for better understanding of this phenomenon. Diagnostic features of graphitised diamonds can be summarised as follow: • Complete graphite pseudomorphs after diamonds obtained at 1 atmosphere consist of very fine-grained aggregate of graphite. They are very fragile and can hardly survive in natural samples. Only general outlines of original diamond morphology can be identified for such pseudomorphs. • Graphite pseudomorphs obtained at 2.0 GPa are very solid. They consist of relatively coarse-grained graphite aggregates. Oriented graphite crystallites are observed on {111} diamond faces at different degree of graphitisation. It should be noted that even at complete graphitisation, the tiny surface morphology of original diamond crystals can be recognised. Randomly oriented graphite crystallites occur on {100} and {110} diamond surfaces at different degree of graphitisation, but generally the crystal shape and some relatively large elements of original diamond crystals morphology can be identified. The size of newly formed graphite crystals does not exceed 100 μm, even at very high temperature. The graphitisation of {100} and especially {110} diamond sectors proceeds by a different mechanism (Pantea et al., 2002b) and also most probably by a different rate, as reflected by morphology of graphite aggregates. • Two coupled dissolution–precipitation processes occur at 2.0 GPa in H2O-saturated conditions. Therefore, these phenomena cannot be considered as solid-state graphitisation of the diamond crystals. As a result of these processes, negatively oriented trigons appear on {111} planes of diamond crystals, and originally flat crystals are transformed into rounded ones. Crystallisation of graphite occurs at relatively low supersaturation by means of spiral growth mechanism.

Our observations provide important clue that metastable graphite (i.e., graphite formed in the diamond stability field) can be distinguished from graphite formed by partial graphitisation of diamond when using criteria listed above.

Acknowledgments This study was supported by grant of Russian Science Foundation (RSF 15-17-30012). Two anonymous reviewers are thanked for their comments that helped to improve the manuscript. Prof. Takao Hirajima is thanked for his editorial handling of the manuscript.

25

References Arima, M., Kozai, Y., 2008. Diamond dissolution rates in kimberlitic melts at 1300–1500°C in the graphite stability field. European Journal of Mineralogy 20, 357–364. Badenhorst, H., 2014. Microstructure of natural graphite flakes revealed by oxidation: limitations of xrd and Raman techniques for crystallinity estimates. Carbon 66, 674–690. Beyssac, O., Goffé, B., Chopin, C., Rouzaud, J., 2002. Raman spectra of carbonaceous material in metasediments: a new geothermometer. Journal of Metamorphic Geology 20, 859–871. Bostick, B.C., Jones, R., Ernst, W., Chen, C., Leech, M.L., Beane, R.J., 2003. Low-temperature microdiamond aggregates in the Maksyutov metamorphic complex, South Ural Mountains, Russia. American Mineralogist 88, 1709–1717. Cardew, P.T., Davey, R.J., 1985. The kinetics of solvent-mediated phase transformations. Proceedings of the Royal Society of London A 398, 415–428. Chepurov, A.I., Fedorov, I.I., Sonin, V.M., 1997. An Experimental Modeling of DiamondForming Processes. SB RAS, UIGGM, Novosibirsk. Chepurov, A., Tomilenko, A., Zhimulev, E., Sonin, V., Chepurov, A., Kovyazin, S., Timina, T.Y., Surkov, N., 2012. The conservation of an aqueous fluid in inclusions in minerals and their interstices at high pressures and temperatures during the decomposition of antigorite. Russian Geology and Geophysics 53, 234–246. Chepurov, A., Zhimulev, E., Agafonov, L., Sonin, V., Chepurov, A., Tomilenko, A., 2013. The stability of ortho- and clinopyroxenes, olivine, and garnet in kimberlitic magma. Russian Geology and Geophysics 54, 406–415. Chou, I.M., Anderson, A.J., 2009. Diamond dissolution and the production of methane and other carbon-bearing species in hydrothermal diamond-anvil cells. Geochimica et Cosmochimica Acta 73, 6360–6366. Davies, G.R., Nixon, P.H., Pearson, D.G., Obata, M., 1992. The tectonic implications of graphitised diamonds from the Ronda peridotite massif, S. Spain. Geology 21, 471–474. De Corte, K., Korsakov, A., Taylor, W.R., Cartigny, P., Ader, M., De Paepe, P., 2000. Diamond growth during ultrahigh-pressure metamorphism of the Kokchetav massif, northern Kazakhstan. The Island Arc 9, 284–303. Dobrzhinetskaya, L.F., Green, H.W., Bozhilov, K.N., Mitchell, T.E., Dickerson, R.M., 2003. Crystallization environment of Kazakhstan microdiamonds: evidence from their nanometric inclusions and mineral associations. Journal of Metamorphic Geology 21, 425–437. Dobrzhinetskaya, L.F., Wirth, R., Rhede, D., Liu, Z., Green, H.W., 2009. Phlogopite and quartz lamellae in diamond-bearing diopside from marbles of the Kokchetav massif, Kazakhstan: exsolution or replacement reaction? Journal of Metamorphic Geology 27, 607–620. El Atrassi, F., Brunet, F., Bouybaouene, M., Chopin, C., Chazot, G., 2011. Melting textures and microdiamonds preserved in graphite pseudomorphs from the beni bousera peridotite massif, morocco. European Journal of Mineralogy 23, 157–168. Fedortchouk, Y., Canil, D., 2009. Diamond oxidation at atmospheric pressure: development of surface features and the effect of oxygen fugacity. European Journal of Mineralogy 21, 623–635. Fedortchouk, Y., Manghnani, M.H., Hushur, A., Shiryaev, A., Nestola, F., 2011. An atomic force microscopy study of diamond dissolution features: the effect of H2O and CO2 in the fluid on diamond morphology. American Mineralogist 96, 1768–1775. Frezzotti, M., Selverstone, J., Sharp, Z., Compagnoni, R., 2011. Carbonate dissolution during subduction revealed by diamond-bearing rocks from the alps. Nature Geoscience 4, 703–706. Grenville-Wells, H., 1952. The graphitization of diamond and the nature of cliftonite. Mineralogical Magazine 29, 803–816. Howell, D., Griffin, W.L., Piazolo, S., Say, J.M., Stern, R.A., Stachel, T., Nasdala, L., Rabeau, J.R., Pearson, N.J., O'Reilly, S.Y., 2013. A spectroscopic and carbon-isotope study of mixedhabit diamonds: impurity characteristics and growth environment. American Mineralogist 98, 66–77. Hwang, S.L., Shen, P., Chu, H.T., Yui, T.F., Lin, C.C., 2001. Genesis of microdiamonds from melt and associated multiphase inclusions in garnet of ultrahigh-pressure gneiss from Erzgebirge, Germany. Earth and Planetary Science Letters 188, 9–15. Hwang, S.L., Shen, P., Chu, H.T., Yui, T.F., Liou, J.G., Sobolev, N.V., Shatsky, V.S., 2005. Crustderived potassic fluid in metamorphic microdiamond. Earth and Planetary Science Letters 231, 295–306. Hwang, S.L., Chu, H.T., Yui, T.F., Shen, P., Schertl, H.P., Liou, J.G., Sobolev, N.V., 2006. Nanometer-size P/K-rich silica glass (former melt) inclusions in microdiamond from the gneisses of Kokchetav and Erzgebirge massifs: diversified characteristics of the formation media of metamorphic microdiamond in UHP rocks due to hostrock buffering. Earth and Planetary Science Letters 243, 94–106. Janak, M., Krogh Ravna, E., Kullerud, K., Yoshida, K., Milovsky, R., Hirajima, T., 2013. Discovery of diamond in the tromsø nappe, Scandinavian caledonides (N. Norway). Journal of Metamorphic Geology 31, 691–703. Katayama, I., Zayachkovsky, A.A., Maruyama, S., 2000. Prograde P-T records from inclusions in zircons from ultrahigh-pressure rocks of the Kokchetav massif, northern Kazakhstan. The Island Arc 9, 417–427. Khokhryakov, A.F., Pal'yanov, Y.N., 2007. The evolution of diamond morphology in the process of dissolution: experimental data. American Mineralogist 92, 909–917. Khokhryakov, A.F., Nechaev, D.V., Sokol, A.G., Palyanov, Y.N., 2009. Formation of various types of graphite inclusions in diamond: experimental data. Lithos 112, 683–689. Korsakov, A.V., Shatsky, V.S., 2004. Origin of graphite-coated diamonds from the UHP metamorphic rocks. Doklady Earth Sciences 399, 1160–1163. Korsakov, A.V., Shatsky, V.S., Sobolev, N.V., Zayachkovsky, A.A., 2002. Garnet-biotiteclinozoisite gneisses: a new type of diamondiferous metamorphic rocks of the Kokchetav massif. European Journal of Mineralogy 14, 915–929.

26


Korsakov, A.V., Vandenabeele, P., Theunissen, K., 2005. Discrimination of metamorphic diamond populations by Raman spectroscopy (Kokchetav, Kazakhstan). Spectrochimica Acta Part A 61, 2378–2385. Korsakov, A.V., Travin, A.V., Yudin, D.S., Marschall, H.R., 2009. 40Ar/39Ar Dating of tourmaline from metamorphic rocks of the Kokchetav Massif, Kazakhstan. Doklady Earth Sciences 424, 168–170. Korsakov, A.V., Perraki, M., Zedgenizov, D.A., Bindi, L., Vandenabeele, P., Suzuki, A., Kagi, H., 2010. Diamond–graphite relationships in ultrahigh-pressure metamorphic rocks from the Kokchetav Massif, Northern Kazakhstan. Journal of Petrology 51, 763–783. Kozai, Y., Arima, M., 2005. Experimental study on diamond dissolution in kimberlitic and lamproitic melts at 1300–1420°C and 1 GPa with controlled oxygen partial pressure. American Mineralogist 90, 1759–1766. Lavrova, L.D., Pechnikov, V.A., Pleshakov, M.A., Nadajdina, E.D., Shukolyukov, Y.A., 1999. A new genetic type of diamond deposit. Scientific World, Moscow. Leech, M.L., Ernst, W.G., 1998. Graphite pseudomorphs after diamond? A carbon isotope and spectroscopic study of graphite cuboids from the Maksyutov Complex, south Ural Mountains, Russia. Geochimica et Cosmochimica Acta 62, 2143–2154. Massonne, H.J., 1998. Whiteschists from the Kokchetav Massif, Kazakhstan, and the Dora Maira Massif, W Alps—a comparison of their metamorphic and metasomatic evolution. Int Workshop UHP Metamorphism Exhumation, Dec 3–6, Stanford University, pp. A204–A205. Massonne, H.J., 1999. A new occurrence of microdiamonds in quartzfeldspathic rocks of the Saxonian Erzgebirge, Germany and their metamorphic evolution. Proceedings of 7th International Kimberlite Conference, pp. 533–539. Massonne, H.J., 2003. A comparison of the evolution of diamondiferous quartz-rich rocks from the Saxonian Erzgebirge and the Kokchetav massif: are so- called diamondiferous gneisses magmatic rocks? Earth and Planetary Science Letters 216, 347–364. Massonne, H.J., Bernhardt, H.J., Dettmar, D., Kessler, E., Medenbach, O., Westphal, T., 1998. Simple identification and quantification of microdiamonds in rock thin-sections. European Journal of Mineralogy 10, 497–504. Mikhno, A., Korsakov, A., 2012. Prograde zonation in ultrapotassic clinopyroxene from ultrahigh-pressure garnet-clinopyroxene rocks from the Kumdy-Kol mine (Kokchetav Massif, Kazakhstan). Doklady Earth Sciences 447, 1333–1337. Mikhno, A.O., Korsakov, A.V., 2013. K2O prograde zoning pattern in clinopyroxene from the Kokchetav diamond-grade metamorphic rocks: missing part of metamorphic history and location of second critical end point for calc-silicate system. Gondwana Research 23, 920–930. Mikhno, A.O., Schmidt, U., Korsakov, A.V., 2013. Origin of k-cymrite and kokchetavite in the polyphase mineral inclusions from kokchetav uhp calc-silicate rocks: evidence from confocal Raman imaging. European Journal of Mineralogy 25, 807–816. Mposkos, E.D., Kostopoulos, D.K., 2001. Diamond, former coesite and supersilicic garnet in metasedimentary rocks from the Greek Rhodope: a new ultrahigh-pressure metamorphic province established. Earth and Planetary Science Letters 192, 497–506. Naemura, K., Ikuta, D., Kagi, H., Odake, S., Ueda, T., Ohi, S., Kobayashi, T., Svojtka, M., Hirajima, T., 2011. Diamond and other possible ultradeep evidence discovered in the orogenic spinel-garnet peridotite from the moldanubian zone of the bohemian massif, Czech Republic. Ultrahigh-Pressure Metamorphism 25, 77–124. Nechaev, D., Khokhryakov, A., 2013. Formation of epigenetic graphite inclusions in diamond crystals: experimental data. Russian Geology and Geophysics 54, 399–405. Ogasawara, Y., Ohta, M., Fukasawa, K., Katayama, I., Maruyama, S., 2000. Diamond-bearing and diamond-free metacarbonate rocks from Kumdy-Kol in the Kokchetav massif, northern Kazakhstan. The Island Arc 9, 400–416. Ogasawara, Y., Fukasawa, K., Maruyama, S., 2002. Coesite exsolution from supersilicic titanite in UHP marble from the Kokchetav massif, northern Kazakhstan. American Mineralogist 87, 454–461. Okada, T., Utsumi, W., Kaneko, H., Turkevich, V., Hamaya, N., Shimomura, O., 2004. Kinetics of the graphite–diamond transformation in aqueous fluid determined by in-situ X-ray diffractions at high pressures and temperatures. Physics and Chemistry of Minerals 31, 261–268. Okamoto, K., Liou, J.G., Ogasawara, Y., 2000. Petrology of the diamond-grade eclogite in the Kokchetav Massif, northern Kazakhstan. The Island Arc 9, 379–399. Pal'yanov, Y.N., Sokol, A., Borzdov, M., Khokhryakov, A., 2002. Fluid-bearing alkaline carbonate melts as the medium for the formation of diamonds in the Earth's mantle: an experimental study. Lithos 60, 145–159. Palyanov, Y.N., Sokol, A.G., 2009. The effect of composition of mantle fluids/melts on diamond formation processes. Lithos 112, 690–700. Pantea, C., Gubicza, J., Ungar, T., Voronin, G., Zerda, T., 2002a. Dislocation density and graphitization of diamond crystals. Physical Review B 66, 094106. Pantea, C., Qian, J., Voronin, G.A., Zerda, T.W., 2002b. High pressure study of graphitization of diamond crystals. Journal of Applied Physics 91, 1957–1962.

Pantea, C., Gubicza, J., Ungár, T., Voronin, G., Nam, N., Zerda, T., 2004. High-pressure effect on dislocation density in nanosize diamond crystals. Diamond and Related Materials 13, 1753–1756. Pearson, D., Davies, G., Nixon, P., Milledge, H., 1989. Graphitized diamonds from a peridotite massif in morocco and implications for anomalous diamond occurrences. Nature 338, 60–62. Pearson, D.G., Davies, G.R., Nixon, P.H., 1995. Orogenic ultramafic rocks of UHP (diamond facies) origin. Cambridge University Press, Cambridge, pp. 456–510. Perraki, M., Proyer, A., Mposkos, E., Kaindl, R., Hoinkes, G., 2006. Raman microspectroscopy on diamond, graphite and other carbon polymorphs from the ultrahigh-pressure metamorphic Kimi Complex of the Rhodope Metamorphic Province, NE Greece. Earth and Planetary Science Letters 241, 672–685. Perraki, M., Korsakov, A.V., Smith, D.C., Mposkos, E., 2009. Raman spectroscopic and microscopic criteria for the distinction of microdiamonds in ultrahigh-pressure metamorphic rocks from diamond in sample preparation materials. American Mineralogist 94, 546–556. Putnis, A., Putnis, C.V., 2007. The mechanism of reequilibration of solids in the presence of a fluid phase. Journal of Solid State Chemistry 180, 1783–1786. Qian, J., Pantea, C., Voronin, G., Zerda, T.W., 2001. Partial graphitization of diamond crystals under high- pressure and high-temperature conditions. Journal of Applied Physics 90, 1632–1637. Qian, J., Pantea, C., Huang, J., Zerda, T.W., Zhao, Y., 2004. Graphitization of diamond powders of different sizes at high pressure–high temperature. Carbon 42, 2691–2697. Rakovan, J., Jaszczak, J.A., 2002. Multiple length scale growth spirals on metamorphic graphite {001} surfaces studied by atomic force microscopy. American Mineralogist 87, 17–24. Shatsky, V.S., Sobolev, N.V., Vavilov, M.A., 1995. Diamond-bearing Metamorphic Rocks of the Kokchetav Massif (Northern Kazakhstan). Cambridge University Press, Cambridge, pp. 427–455. Shatsky, V.S., Rylov, G.M., Efimova, E.S., Corte, K.D., Sobolev, N.V., 1998. Morphology and real structure of microdiamonds from metamorphic rocks of the Kokchetav massif, kimberlites, and alluvial placers. Geologiya i Geofizika 39, 942–956. Shimizu, R., Ogasawara, Y., 2013. Diversity of potassium-bearing tourmalines in diamondiferous kokchetav uhp metamorphic rocks: a geochemical recorder from peak to retrograde metamorphic stages. Journal of Asian Earth Sciences 63, 39–55. Slodkevich, V.V., 1982. Graphite paramorphs after diamond. Proceedings of the Russian Mineralogical Society 111, 13–33 (in Russian). Smith, D., Godard, G., 2013. A Raman spectroscopic study of diamond and disordered sp3carbon in the coesite-bearing straumen eclogite pod, Norway. Journal of Metamorphic Geology 31, 19–33. Sokol, A., Pal'yanov, Y.N., 2008. Diamond formation in the system MgO-SiO2-H2O-C at 7.5 GPa and 1,600 C. Contributions to Mineralogy and Petrology 155, 33–43. Sokol, A., Pal'Yanov, Y.N., Pal'Yanova, G., Khokhryakov, A., Borzdov, Y.M., 2001. Diamond and graphite crystallization from C-O-H fluids under high pressure and high temperature conditions. Diamond and Related Materials 10, 2131–2136. Sungawa, I., 1984. Materials Science of the Earth's Interior. Chapter Growth of Crystals in Nature. Terra Scientific Publishing Co, Tokyo. Tomilenko, A., Ragozin, A., Shatskii, V., Shebanin, A., 2001. Variation in the fluid phase composition in the process of natural diamond crystallization. Doklady Earth Sciences 379, 571–574. Van Roermund, H., Carswell, D.A., Drury, M.R., Heijboer, T.C., 2002. Microdiamonds in megacrystic garnet websterite pod from Bardane on the island of Fjortoft, western Norway: evidence for diamond formation in mantle rocks during deep continental subduction. Geology 30, 959–962. Vrijmoed, J., Smith, D.C., van Roermund, H., 2008. Raman confirmation of microdiamond in the Svartberget Fe-Ti type garnet peridotite, Western Gneiss Complex, Western Norway. Terra Nova 20, 295–301. Wang, X., Liou, J.G., Mao, H.K., 1989. Coesite-bearing eclogites from the Dabie mountains, central China. Geology 17, 1085–1088. Yamaoka, S., Kumar, M., Akaishi, M., Kanda, H., 2000. Reaction between carbon and water under diamond-stable high pressure and high temperature conditions. Diamond and Related Materials 9, 1480–1486. Yamaoka, S., Kumar, M.D.S., Kanda, H., Akaishi, M., 2002. Thermal decomposition of glucose and diamond formation under diamond-stable high pressure–high temperature conditions. Diamond and Related Materials 11, 118–124. Zhang, R.Y., Liou, J.G., Ernst, W.G., Coleman, R.G., Sobolev, N.V., Shatsky, V.S., 1997. Metamorphic evolution of diamond-bearing and associated rocks from Kokchetav massif, northern Kazakhstan. Journal of Metamorphic Geology 15, 479–496. Zhu, Y.F., Ogasawara, Y., 2002. Carbon recycled into the deep Earth: evidenced by dolomite dissociation in subduction-zone rocks. Geology 30, 947–950.