Abstract It was suggested that alkali–alkaline earth carbonates may have a substantial role in petrological processes relevant to metasomatism and melting of the Earth’s mantle. Because natrite, Na2CO3, Na–Ca carbonate (shortite and/or nyerereite), and calcite, CaCO3, have been recently reported from xenoliths of shallow mantle (110–115 km) origin, we performed experiments on phase relations in the system N a2CO3–CaCO3 at 3 GPa and 800–1300 °C. We found that the system has one intermediate compound, Na2Ca3(CO3)4, at 800 °C, and two intermediate compounds, Na2Ca(CO3)2 and Na2Ca3(CO3)4, at 850 °C. CaCO3 crystals recovered from experiments at 950 and 1000 °C are aragonite and calcite, respectively. Maximum solid solution of CaCO3 in Na2CO3 is 20 mol% at 850 °C. The Na-carbonate–Na2Ca(CO3)2 eutectic locates near 860 °C and 56 mol% Na2CO3. Na2Ca(CO3)2 melts incongruently near 880 °C to produce Na2Ca3(CO3)4 and a liquid containing about 51 mol% Na2CO3. Na2Ca3(CO3)4 disappears above 1000 °C via incongruent melting to calcite and a liquid containing about 43 mol% Na2CO3. At 1050 °C, the liquid, coexisting with Na-carbonate, contains 87 mol% Na2CO3. Na-carbonate remains solid up to 1150 °C and melts at 1200 °C. The N a2CO3 content in the liquid coexisting with calcite decreases to 15 mol% as temperature increases to 1300 °C. Considering the present and previous data, a range of the intermediate compounds on the liquidus of the Na2CO3–CaCO3 join changes as pressure increases in the following sequence: N a2Ca(CO3)2 (0.1 GPa) → Na2Ca(CO3)2, Na2Ca3(CO3)4 (3 GPa) → Na4Ca(CO3)3, Na2Ca3(CO3)4 (6 GPa). Thus, the N a2Ca(CO3)2 nyerereite stability field extends to the shallow mantle pressures. Consequently, findings of nyerereite among daughter phases in the melt inclusions in olivine from the sheared garnet peridotites are consistent with their mantle origin. Keywords Na–Ca carbonates · High-pressure experiment · Nyerereite · Shortite · Raman · Carbonatite
Introduction Alkalis (Na and K) in their natural abundances could account for lowering of the solidus temperatures of carbonated silicate mantle by several 100° (Litasov et al. 2013). Partial melting of carbonated peridotites and eclogites at ≥ 3 GPa yields Na- and K-bearing carbonatite melts (Dasgupta et al. 2005; Ghosh et al. 2009; Brey et al. 2011; Moore 2012; Kiseeva et al. 2013; Thomson et al. 2016). Mass-balance calculations of samples obtained below apparent solidi often produce clear deficits of alkalis (Dasgupta and Hirschmann 2007; * Ivan V. Podborodnikov [email protected] 1
V.S. Sobolev Institute of Geology and Mineralogy, Russian Academy of Science, Siberian Branch, Novosibirsk 630090, Russia
Novosibirsk State University, Novosibirsk 630090, Russia
Ghosh et al. 2009), suggesting the presence of minor alkalirich liquid or solid carbonate phases (Litasov et al. 2013). Identification and precise determination of the compositions of these phases are problematic due to their trace amounts in the natural-like systems. Phase equilibria in simple carbonate systems are, therefore, important to facilitate interpretation of the complex carbonated silicate systems. Particularly, it is essential to know crystal chemistry and Raman spectra of high-pressure alkali-bearing carbonate phases. The T–X high-pressure phase diagrams of the simple carbonate systems are also useful for experimental studies of structure and physicochemical properties of carbonate melts under high-pressures, namely, measurements of density, viscosity, wetting, mobility (infiltration and segregation), element diffusivity, rheology of the melt impregnated rocks, and carbon isotope fractionation. Studies of mechanisms and kinetics of resorption of mantle minerals (diamond, orthopyroxene, etc.) in ascending kimberlite (i.e., alkali-rich
Physics and Chemistry of Minerals
Fig. 1 Phase relations in the system Na2CO3–CaCO3 at 0.1 GPa (a) and 6.0 GPa (b), modified after Cooper et al. (1975) and Shatskiy et al. (2013c). Grey numbers denote composition of corresponding eutectics in mol% Na2CO3. Cal calcite, Arg aragonite,
Na2 CaCO3 solid solution in Na2CO3, Na4Ca Na4Ca(CO3)3, Na2Ca a2Ca(CO3)2, Na2Ca2 Na2Ca2(CO3)3, Na2Ca3 Na2Ca3(CO3)4, Na2Ca4 N Na2Ca4(CO3)5, L liquid, F CO2 fluid
carbonatite) melt also need high-pressure data on phase relations in simple carbonate systems. As part of an investigation of those systems, Na2CO3–CaCO3 is important, given the Na and Ca abundances in the mantle-derived carbonatite melts (Kaminsky et al. 2009; Stoppa et al. 2009; Chen et al. 2013) and in the groundmass of fresh kimberlites (Kamenetsky et al. 2004; Golovin et al. 2017b). The given system is also important in view of the Na and Ca abundances in carbonatite melt inclusions in kimberlite-hosted phenocrysts (Golovin et al. 2007; Kamenetsky et al. 2009, 2013; Mernagh et al. 2011; Abersteiner et al. 2017) and mantle xenoliths derived from depths varying from 110 to 115 km (spinel harzburgite: 3.4–3.5 GPa, 860 °C) (Giuliani et al. 2012) to 140 km (garnet wehrlite: 4.3 GPa, 1060 °C) (Soltys et al. 2016), and 180–240 km (sheared garnet harzburgite and lherzolites: 5.7–7.3 GPa, 1230–1360 °C) (Golovin et al. 2017b, 2018). Phase relations for the join Na2CO3–CaCO3 have been previously determined at 0.1 and 6 GPa (Cooper et al. 1975; Shatskiy et al. 2013c). Although the end members, CaCO3 and Na2CO3, reveal quite high melting temperatures (Li et al. 2017), 400–800 °C higher than the typical shield geotherm (40 mW/m2), the N a2CO3–CaCO3 system forms fusible intermediate compounds (Na–Ca double carbonates) and eutectics, which are several hundreds of degrees lower than melting temperatures of the end members (Fig. 1). Yet, significant difference in intermediate compounds stable at pressures of 0.1 and 6 GPa does not allow any interpolation between these
pressures. Based on the phase relations established at 6 GPa, shortite Na2Ca2(CO3)3 and nyerereite N a2Ca(CO3)2, the double carbonates observed at 0.1 GPa (Cooper et al. 1975), are not stable at the pressure and temperature conditions of sublithospheric mantle (Shatskiy et al. 2013c). At the same time, findings of these minerals in coexistence with aragonite among daughter phases in the melt inclusions in olivine from the sheared garnet peridotites (Golovin et al. 2017b, 2018) may indicate that their stability fields could extend at least to P–T conditions of the shallow mantle. In this paper, experimental results for the join Na2CO3–CaCO3 at 3 GPa are presented and compared with the results at 0.1 GPa (Cooper et al. 1975) and 6 GPa (Shatskiy et al. 2013c). The Na–Ca double carbonates observed in the Na2CO3–CaCO3 system are compared with those in multicomponent systems such as carbonated eclogite (Kiseeva et al. 2013; Thomson et al. 2016), pelite (Grassi and Schmidt 2011), and model carbonatite (Litasov et al. 2013).
Methods All experiments were run in the ‘Discoverer-1500’ DIA-type apparatus in the V.S. Sobolev Institute of Geology and Mineralogy SB RAS in Novosibirsk, Russia. “Fujilloy TN-05” 26-mm tungsten carbide cubes with truncation edge length of 12 mm were employed as Kawai-cell anvils. We used semi-sintered ZrO2 ceramics (OZ-8C, MinoYogyo Co., Ltd)
Physics and Chemistry of Minerals
as a pressure medium shaped as a 20.5 mm octahedron with ground edges and corners. Pyrophyllite gaskets, 4.0 mm in both width and thickness were used to seal the compressed volume and support the anvil flanks. The cell assembly design and performance have been previously reported (Shatskiy et al. 2013b, 2018). The sample temperature was monitored via a W-Re3%/25% thermocouple. No correction for the effect of pressure on the thermocouple electromotive force was applied. Details of pressure calibration are reported by (Shatskiy et al. 2018). Deviation of pressure from the desirable value during heating to 900 °C in the given cell and press load of 3 MN did not exceed ± 0.1 GPa, as confirmed by comparison of pressure calibration at 25 and 900 °C. The cell assembly contains four graphite cassettes (i.e., multiple sample holders). Each cassette contains four holes 1.0 mm in diameter and depth. To study the present system, we used eight holes with different sample compositions shown in Table 1. The remaining holes were employed to study alternative carbonate systems. Starting materials were prepared by blending reagent grade Na2CO3 and CaCO3 in an agate mortar with acetone and loaded as a powder into graphite cassettes. Since Na2CO3 is hygroscopic material, special attention for sample preparation was paid to minimize the amount of the moisture in the sample absorbed from the atmosphere. The loaded cassettes were dried at 300 °C for 1–2 h. Prepared assemblies were stored at 200 °C in a vacuum for ≥ 12 h prior to experiment. All experiments were conducted at 15–35% outside humidity. An absence of any features in the O-H stretching region of Raman spectra (3000–4000 cm−1) proves that no measurable amount of H 2O is present in the recovered samples. The experiments were performed by compression to a load of 3.0 MN (corresponding to a pressure of 3 GPa) and then heating to a target temperature at a rate of 50–100 °C/ min. The temperature was maintained within 2.0 °C of the desired value, using a temperature control mode at a constant press load. The thermal modeling software by Hernlund et al. (2006) suggests that the temperature gradient across individual samples and sample charge should be less than 5 and 9 °C/mm at 800 and 1300 °C, respectively. The experiments were terminated by turning off the power, resulting in a temperature drop to < 100 °C in 10–20 s, followed by slow decompression. In this study, falling-sphere experiments were conducted to determine the melting point of Na2CO3 at 3 GPa. 200-micron-size Pt sphere was placed near the top of sample. At 3 GPa pressure, the sample was heated to the target temperature, held for 5 min, and than quenched. The recovered samples were ground to reveal the Pt spheres. After the experiments were completed, the recovered graphite cassettes were cut using a low-speed diamond saw to get vertical cross-sections of samples. The obtained
specimens were mounted in a plexiglas holder with epoxy and polished in low-viscosity oil using 400(37)-, 1000(13)-, and 1500(9)-mesh(µm) sandpaper. The sample surface was cleaned using an oil spray between each step of polishing. The final polishing was done on a satin cloth with 3 µm diamond paste. We used petroleum benzene to remove the oil after polishing. The clean samples were stored in petroleum benzene, prior to carbon coating and loading into a scanning electron microscope (Shatskiy et al. 2017). Recovered samples were examined using a MIRA 3 LMU scanning electron microscope (Tescan Orsay Holding) coupled with an INCA energy-dispersive X‑ray microanalysis system 450 equipped with the liquid nitrogen-free Large area EDS X-Max-80 Silicon Drift Detector (Oxford Instruments Nanoanalysis Ltd) at IGM SB RAS. Energy-dispersive X‑ray spectra (EDS) were collected using an electron beam-rastering method, in which the stage is stationary while the electron beam moves over the surface area, with dimensions 5–100 µm (for mineral phases) and 50–500 µm (for a quenched melt) at 20 kV accelerating voltage and 1.5 nA beam current. Live counting time for X-ray spectra was 30 s (Lavrent’ev et al. 2015). The correctness of the EDS measurements was confirmed under the same conditions using post-experimental samples with known compositions and homogeneous textures obtained below the solidus and above the liquidus (Table 1). Raman measurements were performed using a Horiba Jobin Yvon LabRAM HR800 Raman microspectrometer with the 514 nm line of an Ar-ion laser at IGM SB RAS (Shatskiy et al. 2015a). The Raman spectra of calcite and aragonite (De La Pierre et al. 2014) were used for identification of CaCO3 polymorphs.
Experimental results The results of the experiments are summarized in Table 1. Selected backscattered electron (BSE) images of subsolidus and supersolidus samples in the system Na2CO3–CaCO3 are shown in Figs. 2 and 3, respectively. The subsolidus samples are represented by homogeneous aggregates of carbonate phases, with grain size up to 100 µm (Fig. 2). In non-stoichiometric mixtures annealed at 800–850 °C for 17–18 h, the limiting reagents, i.e., N a2CO3 at X(Na2CO3) ≤ 20 mol% and CaCO3 at X(Na2CO3) ≥ 30 mol%, have been consumed completely to form double Na–Ca carbonates (Fig. 2a–d; Table 1). After annealing of the stoichiometric mixture, X(Na2CO3) = 50 mol%, at 850 °C for 18 h, both reagents, Na2CO3 and CaCO3, were completely reacted to form an Na2Ca(CO3)2 compound (Fig. 2f; Table 1). This suggests that reactions have gone to completion and equilibrium has been achieved. In supersolidus runs, the carbonate crystals grew to larger size (up to 300 µm) in the lower-temperature
Table 1 Compositions of the run products in the system Na2CO3–CaCO3 at 3 GPa
90 85 20 10 1100 5 min 100 1150 5 min 100 1175 1 h 30 20 10 5 1200 1 h 10 5 1200 5 min 100 1300 1 h 5
Bulk Na2#: i initial Na2CO3 concentration in the system in mol%, m balk composition measured using EDS by rastering the electron beam over sample cross-section, – phase was not established in the run products, × no data, + phase is present, but the composition is not defined; t run duration, b.d. below detection limit. Standard deviations for the last significant digits are given in parentheses, wherever the number of measurement is more than one. The number of analyses is given in lowercase after parentheses. Bulk–bulk composition measured by rastering an electron beam over a surface area of post-experimental samples
sample side, while melt segregated at the higher-temperature sample side. The melt quenched to a dendritic aggregate of carbonate crystals (Fig. 3). Phase relations established in the system N a2CO3–CaCO3 at 3 GPa are illustrated in Fig. 4. At 800 °C, the system has one intermediate compound, N a2Ca3(CO3)4, which appears in coexistence wit h Na-carbonate at X(Na2CO3) = 30–75 mol% (Fig. 2a, b) and in coexistence with aragonite at X(Na2CO3) = 10–20 mol% (Fig. 2c, d). At 850 °C, an additional intermediate compound, N a2Ca(CO3)2, appears in coexistence with Na-carbonate at X(Na2CO3) = 60 and 75 mol% (Fig. 2e, f) and with N a 2Ca 3(CO 3) 4 at X(Na2CO3) = 30 and 40 mol% (Fig. 2h). Measurable amounts of Ca in Na2CO3 determined by electron microprobe suggest dissolution of the C aCO3 component in the Na-carbonate as a solid solution under given conditions (Table 1). The maximum CaCO3 solubility in sodium carbonate of about 20 mol% was established at 850 °C (Table 1). The BSE images reveal exsolution textures in Na-carbonate (Fig. 2f). Ca-carbonate recovered from experiments was identified by Raman spectroscopy as aragonite at T ≤ 950 °C and calcite at T ≥ 1000 °C (Fig. 5c, d). The Na2CO3 solubility in CaCO3 does not exceed the detection limit of EDS employed in our study (i.e., < 0.5 mol%) (Table 1). Therefore, it is suggested that there is very little, if any, solid solution of Na2CO3 in aragonite and calcite at 3 GPa. The first melt was established at 900 °C in the compositional range X(Na2CO3) = 30–75 mol%. At X(Na2CO3) = 60 and 75 mol%, the melt contains 57–59 mol% Na 2CO 3
and coexists with Na-carbonate (Fig. 3a; Table 1). At X(Na2CO3) = 30 and 48 mol%, the melt contains about 50 mol% Na2CO3 and coexists with Na2Ca3(CO3)4 (Fig. 3b; Table 1). Thus, N a2Ca(CO3)2 melts incongruently to produce Na2Ca3(CO3)4 and a liquid (Fig. 4). Interpolation of obtained experimental data points suggests that the eutectic between Na-carbonate and N a 2Ca(CO 3) 2 occurs near 56 mol% Na2CO3, whereas the peritectic between Na2Ca(CO3)2 and Na2Ca3(CO3)4 occurs near 51 mol% Na2CO3 (Fig. 4). As temperature increases to 1050 °C, the Na2CO3 content in the liquid coexisting with Na-carbonate, increases to 87 mol%, while the C aCO3 solubility in N a2CO3 decreases to 7 mol% (Fig. 4; Table 1). At 1000 °C and X(Na2CO3) = 10 and 20 mol%, the calcite + Na2Ca3(CO3)4 assemblage appears at the lower-temperature capsule side, while calcite and liquid coexist at the higher-temperature side (Fig. 3e, f). At X(Na2CO3) = 30 and 40 mol%, a liquid coexists with Na2Ca3(CO3)4 (Fig. 3c, d). As temperature increases to 1050 °C, Na2Ca3(CO3)4 disappears via incongruent melting to produce calcite and a liquid containing 31–34 mol% Na2CO3 (Figs. 3h, i, 4). The Na2CO3 content in a liquid coexisting with calcite decreases to 15 mol% as temperature increases to 1300 °C (Figs. 3j, 4). Based on our falling-sphere experiments conducted at 3 GPa, Na2CO3 remains solid up to 1150 °C (Fig. 3k) and melts at 1200 °C (Fig. 3l). The established melting point, 1175 ± 25 °C, is lower by about 340 °C than that of C aCO3, 1515 ± 15 °C, (Shatskiy et al. 2018) and by about 100 °C than the N a2CO3 melting point, 1269 ± 40 °C, determined by
Physics and Chemistry of Minerals
Fig. 2 Representative BSE micrographs of sample cross-sections from subsolidus experiments in the Na2CO3–CaCO3 system at 3 GPa. Cal calcite, Arg aragonite, Na2 CaCO3 solid solution in Na2CO3, Na2Ca Na2Ca(CO3)2, Na2Ca3 Na2Ca3(CO3)4
monitoring resistance change during sample heating (Li 2015; Li et al. 2017) (Fig. 6).
Discussion The Na2CO3–CaCO3 binary versus pressure The alkali–alkali earth carbonate systems are known to form fusible intermediate compounds, eutectics or
peritectics, which are several hundreds of degrees lower than melting curves of pure end-members (Shatskiy et al. 2015b). At 6 GPa, the Na2CO3–CaCO3 system has three intermediate compounds N a 4Ca(CO 3) 3, Na 2Ca 3(CO 3) 4, and N a 2 Ca 4 (CO 3 ) 5 and three eutectics: at 70 mol% Na2CO3 and 1200 °C, 50 mol% Na2CO3 and 1200 °C, and 21 mol% Na2CO3 and 1300 °C (Figs. 1b, 7). As pressure decreases to 3 GPa, the number of intermediate compounds decreases to two: N a2Ca(CO3)2 and Na2Ca3(CO3)4. The melting changes to incongruent with peritectics at
Physics and Chemistry of Minerals
43 mol% Na2CO3 and 1000 °C and 51 mol% N a2CO3 and 880 °C (Figs. 4, 7). The Na-carbonate–Na2Ca(CO3)2 eutectic locates at 56 mol% and 860 °C (Fig. 7). At 0.1 GPa, the low-temperature Na2Ca2(CO3)3 compound decomposes at 400 °C to produce calcite and N a2Ca(CO3)2, which melts congruently at 817 °C (Fig. 1a). The eutectics locate at 76.8 mol% Na2CO3 and 725 °C and 44.5 mol% N a2CO3 and 813 °C (Fig. 7). The Na-carbonate + liquid region unnaturally becomes thinned above 1050 °C at 3 GPa. This shape may be owing the α –β phase transition in N a2CO3 (Shatskiy et al. 2013a).
Intermediate compounds in the Na2CO3–CaCO3 join versus pressure Considering the present and previous data (Cooper et al. 1975; Shatskiy et al. 2013c), intermediate compounds on the liquidus of the C aCO 3–Na 2CO 3 system change as pressure increases in the following sequence: Na 2Ca(CO 3) 2 (0.1 GPa) → Na 2Ca(CO 3) 2, Na 2Ca 3(CO 3) 4 (3 GPa) → Na4Ca(CO3)3, Na2Ca3(CO3)4 (6 GPa) (Fig. 7). At 3 GPa, the “low-pressure” N a2Ca(CO3)2 compound appears as temperature increases to 850 °C (Fig. 4). The Raman spectrum of Na2Ca(CO3)2 (Fig. 5b) is similar to that synthesized at 1 atm and 400 °C (Böttcher and Reutel 1996) and at 0.1 GPa and 450 °C (Golovin et al. 2017a). Recently the crystal structure of N a2Ca(CO3)2, synthetic analogue of nyerereite grown at 0.1 GPa and 450 °C, was fully solved: P21ca, a = 10.0713(5) Å, b = 8.7220(2) Å, c = 12.2460(4) Å (Gavryushkin et al. 2016). The experimental data suggests that the temperature interval of Na 2Ca(CO 3) 2 stability narrows from 400 to 817 °C at 0.1 GPa (Cooper et al. 1975) (Fig. 1a) to 825–875 °C at 3 GPa (Fig. 4), whereas at 6 GPa N a2Ca(CO3)2 disappears (Shatskiy et al. 2013c) (Fig. 1b). Although at 0.1 GPa, Na2Ca(CO3)2 melts congruently (Fig. 1a), at 3 GPa it disappears via incongruent melting, producing Na2Ca3(CO3)4 and Na-rich carbonate melt (Fig. 4). The Raman spectrum of “high-pressure” Na2Ca3(CO3)4 compound synthesized at 3 GPa (Fig. 5a) is the same as that recovered from experiments at 6 GPa (Shatskiy et al. 2015a). The crystal structure of the Na2Ca3(CO3)4 compound synthesized at 6 GPa was recently solved: P1n1, Z = 8, a = 31.4421(8) Å, b = 8.1960(2) Å, c = 7.4360(2) Å, and β = 89.923(2)° (Gavryushkin et al. 2014). It was found that at 6 GPa, Na2Ca3(CO3)4 melts congruently just above 1300 °C (Fig. 1b) (Shatskiy et al. 2013c), while at 3 GPa, it melts incongruently to calcite plus liquid at 1000 °C (Fig. 4). The Na 4Ca(CO 3) 3 compound appears as a liquidus phase at 6 GPa, where it melts congruently above 1300 °C (Fig. 7). This phase has a cubic structure (space group Ia3d, a = 14.5770(5) Å) (Rashchenko et al., in preparation).
The Raman spectrum of Na4Ca(CO3)3 has been previously reported (Shatskiy et al. 2015a). In addition, the system has low-temperature phases, which undergo subsolidus breakdown and does not appear on the liquidus. The N a2Ca2(CO3)3 compound, synthetic analogue of mineral shortite, forms at 0.1 GPa as temperature decreases below 400 °C according to the reaction: Na2Ca(CO3)2 (nyerereite) + CaCO3 (calcite) = Na2Ca2(CO3)3 (shortite) (Cooper et al. 1975) (Fig. 1a). Na2Ca2(CO3)3 crystallizes in the orthorhombic unit cell a = 4.9720(9) Å, b = 11.068(3) Å, and c = 7.1271(14) Å in space group Amm2 (Dickens et al. 1971; Song et al. 2017). According to the present study, the Na2Ca2(CO3)3 compound is not stable at 3 GPa and ≥ 800 °C (Fig. 4). However, based on our preliminary data on the ternary Na–Ca–Mg carbonate system, the partial Mg substitution for Ca stabilizes the N a2(Ca0.95–0.76Mg0.05–0.24)2(CO3)3 compound even at 3 GPa and 800–900 °C. The Na2Ca4(CO3)5 compound forms at 6 GPa below 1175 °C (Shatskiy et al. 2013c). Based on the present results and those reported by Shatskiy et al. (2013c), the a2Ca3(CO3)4 and Na2Ca4(CO3)5 compound decomposes to N aragonite as pressure decreases from 6 to 3 GPa at 800 °C or as temperature increases from 1050 to 1100 °C at 6 GPa. According to our preliminary data on the Na–Ca–Mg carbonate system, the partial Mg substitution for Ca stabilizes Na2(Ca0.97Mg0.03)4(CO3)5 even at 3 GPa as low temperature (700 °C) subsolidus phase. The crystal structure of the Na2Ca4(CO3)5 compound synthesized at 6 GPa and 1050 °C was recently solved: P63mc; a = 10.3740(1) Å, c = 6.2594(1) Å (Rashchenko et al. 2017).
Pressure‑dependent subsolidus reactions in the Na2CO3–CaCO3 binary According to the phase diagrams presented at Figs. 1 and 4, equilibrium subsolidus associations in the Na2CO3–CaCO3 binary are: At 6 GPa (Shatskiy et al. 2013c) • • • • •
Physics and Chemistry of Minerals ◂Fig. 3 BSE images of sample cross-sections from supersolidus
experiments in the N a2CO3–CaCO3 system at 3 GPa. Cal calcite, Na2 CaCO3 solid solution in Na2CO3, Na2Ca3 Na2Ca3(CO3)4, L quenched liquid, HT high-temperature side, LT low-temperature side. The numbers in parentheses indicate N a2CO3 concentration (mol%) in corresponding carbonate phase
conditions (Table 2). All obtained reaction volumes are positive, supporting the possibility of the discussed solid-phase transformations upon release of pressure. The latter suggest that association of Na2CO3, Na2Ca(CO3)2, Na2Ca2(CO3)2, and CaCO3 reported for inclusions in superdeep diamonds (Kaminsky et al. 2009), kimberlitic olivines (Golovin et al. 2018), and unique carbonate nodules of Udachnaya-East kimberlite pipe (Kamenetsky et al. 2006) may actually represent breakdown products of high-pressure minerals identical to synthetic phases described here.
(9) We roughly estimated the volume effects of the reactions (1–9) using unit cell volumes of involved phases at ambient
Appearance of Na–Ca carbonates under mantle P–T conditions has been previously revealed in several studies on phase relationships in multicomponent systems such as carbonated eclogite (Kiseeva et al. 2013; Thomson et al. 2016), pelite (Grassi and Schmidt 2011), and model carbonatite (Litasov et al. 2013). It is, therefore, important to compare those data with phase relations in pure carbonate systems. Litasov et al. (2013) studied phase relations in a model Na-rich and K-rich carbonatite at 3–21 GPa. They found that over that entire pressure range, Na is mainly hosted by Carich crystalline carbonate in both systems. The EDS analyzes of this compound revealed 10–12 mol% Na2CO3 + K2CO3, K# = 7–30, and Ca# = 78–95, where K# = 100·K/(K + Na) and Ca# = 100·Ca/(Ca + Mg + Fe). Based on the apparent similarity of the Raman spectra, the authors identified this carbonate as aragonite, whereas significant broadening of the main CO32− Raman band near 1090 cm−1 may indicate a disordered structure. Kiseeva et al. (2013) examined phase relations in carbonated eclogite (GAl-cc and Volga-cc) from 9 to 21 GPa. They found that the subsolidus carbonates are represented by magnesite and nearly pure aragonite to 9 GPa in Volga-cc and to 13 GPa in GAl-cc. At higher pressures the subsolidus carbonates were represented by magnesite and Na–Ca carbonate with composition similar to that reported by Litasov et al. (2013) (10–15 mol% N a2CO3 + K2CO3, K# = 3–13, Ca# = 76–81). This carbonate was also identified as aragonite based on the Raman spectra. According to the present and previous studies of the binary and ternary carbonate systems at 3 and 6 GPa, CaCO3 aragonite does not dissolve Na as well as K, Mg, and Fe (Shatskiy et al. 2014, 2016a, b, 2018). Instead, Na enters the Na–Ca double carbonates, Na2Ca3(CO3)4 and Na2Ca4(CO3)5. The Raman spectra of these compounds exhibit the main bands in the same spectral region as aragonite (Shatskiy et al. 2015a), which could lead to misinterpretation. The deficiency of Na relative to that required by the Na2Ca4(CO3)5 stoichiometry is presumably due to migration
Physics and Chemistry of Minerals
natural Udachnaya-East kimberlite system at 6.5 GPa and 900 °C (Sharygin et al. 2015). In the DG2 composition (dry pelite) at 22 and 23.5 GPa and temperature ≤ 1400 °C, a carbonate with an approximate stoichiometry N a 2Ca 2(CO 3) 3 and almost no Mg, Fe, K has been identified (Grassi and Schmidt 2011). This carbonate resembles stoichiometry of shortite, which is not stable at 3 GPa and ≥ 800 °C (Fig. 4) and at 6 GPa and ≥ 900 °C (Fig. 1b). In the Volga-cc composition (eclogite) at 21 GPa and 1200 °C, a carbonate with an empirical formula (Na 0.89K 0.11) 2(Ca 0.68Mg 0.24Fe 0.08) (CO3)2 was detected (Kiseeva et al. 2013). This carbonate resembles stoichiometry of nyerereite, which is stable at 3 GPa and 850 °C (Fig. 4), but decomposes as pressure increases to 6 GPa (Fig. 1b) according to the reaction: Na2Ca(CO3)2 = Na4Ca(CO3)3 + Na2Ca3(CO3)4. Appearance of carbonates with the same stoichiometries as shortite and nyerereite at pressures ≥ 21–22 GPa may suggest the existence of high-pressure polymorphs of Na2Ca2(CO3)3 shortite and Na2Ca(CO3)2 nyerereite. Yet, considering available data, we also cannot exclude that the stability field of shortite extends to high pressures and temperatures. Fig. 4 Phase relations in the Na2CO3–CaCO3 system at 3 GPa. Open and grey circles indicate composition of solid phases and melt measured by EDS. Grey numbers denote composition of corresponding eutectic and peritectics in mol% Na2CO3. Arg aragonite, Cal calcite, Na2 CaCO3 solid solution in Na2CO3, Na2Ca Na2Ca(CO3)2, Na2Ca3 Na2Ca3(CO3)4, L liquid
of this element under a stationary electron beam for both WDS and EDS analyses (Dasgupta et al. 2006; Dasgupta and Hirschmann 2007; Shatskiy et al. 2013c). Grassi and Schmidt (2011) determined the phase relationships in carbonated pelites (DG2 and AM) from 5.5 to 23.5 GPa. They found that breakdown of clinopyroxene at ≥ 16 GPa causes Na–Ca carbonate (16–20 mol% Na 2CO 3 + K 2CO 3, K# = 1–7, Ca# = 73–84) to replace aragonite. Similar tendency was established in a carbonated MORB composition. It was found that dissolution of Na-poor pyroxene components into coexisting garnet at pressures just above 13 GPa results in redistribution of Na from silicate (clinopyroxene) to Na–Ca carbonate (Na0.97K0.03)2(Ca0.86Mg0.11Fe0.03)4(CO3)5, lowering the solidus by ~ 200 °C and yielding a Na-rich carbonatite melt (Thomson et al. 2016). The composition of this carbonate is very similar to the Na–Ca carbonate in Grassi and Schmidt (2011) study and to the N a 2(Ca ≥ 0.87Mg ≤ 0.13) 4(CO 3) 5 compound established in the Na2CO3–CaMg(CO3)2 system (Shatskiy et al. 2016b). The Na–Ca carbonate compound with Raman spectra similar to N a 2Ca4(CO3)5 was also detected among the run products from the experiment in
Possible participation of Na–Ca carbonates in melting reactions in the shallow mantle The amount of Na2O is the dominant parameter determining the solidus temperature of peridotite at pressures of ≤ 3 GPa, because Na2O becomes less compatible with decreasing pressure owing decreasing jadeite component in clinopyroxene (Walter and Presnall 1994; Hirschmann 2000). Moore (2012) studied melting phase relations at 3 GPa in an analogue system Na2O–CaO–MgO–Al2O3–SiO2–CO2 (NCMAS-CO2) approximating a pyrolite mantle variously enriched with Na2O. At ≥ 1200 °C, carbonate minerals were completely consumed by the melting reactions yielding silicate-bearing dolomite melt with N a2O varying from 0.82 to 16.64 mol%1 as bulk N a2O increases from 0.34 to 7.90 mol%. Subsolidus assemblage established at 1100 °C includes olivine, orthopyroxene, clinopyroxene, garnet, and dolomite. Despite an apparent absence of Na-bearing carbonate phases, convergence of mass balance requires presence of Na2CO3 in the subsolidus carbonate component in concentration, 100·Na2CO3/(Na2CO3 + (Ca,Mg)CO3), varying from 1.6 to 44.9 mol% over studied range of bulk compositions. In this compositional range, the carbonate assemblage includes dolomite, eitelite [Na2Mg(CO3)2], and Na–Ca carbonate. The latter is represented by N a2Ca3(CO3)2 at
1 Concentration of Na2O in the melt was corrected using a mass balance approach.
Physics and Chemistry of Minerals
Fig. 5 Representative unpolarized Raman spectra of recovered Na2Ca3(CO3)4 (a), Na2Ca(CO3)2 (b), calcite (c), and aragonite (d) at ambient conditions. The synthesis conditions are shown at the left-upper side of each spectrum
700 °C or N a2(Ca0.95−0.76Mg0.05−0.24)2(CO3)3 at 800–850 °C (our unpublished data). The subsolidus assemblage established at 1100 °C and bulk Na2O = 0.34 mol% in the run #62 in Moore (2012) study corresponds to dolomite-bearing garnet lherzolite. Mass balance calculations for this experiment revealed 85% deficit of N a2O indicating the presence of minor Na-rich
carbonate phases. Based on the present and Podborodnikov et al. (under review) results all Na–Ca double carbonates and eitelite melt above 900–1000 °C at 3 GPa. Consequently, the missing phase has to be Na-rich dolomite melt. Thus, at 3 GPa, the solidus temperature of the NCMASCO2 peridotite and near solidus melt composition could be controlled by the dolomite + eitelite + Na–Ca carbonate
Physics and Chemistry of Minerals
Fig. 6 Comparison of N a2CO3 melting temperatures established in the falling-sphere experiments (FS) and using electrical conductivity measurements (ECM). TS—this study, S13—(Shatskiy et al. 2013a), L15—(Li 2015), L17—(Li et al. 2017)
assemblage, which start to melt between 850 and 900 °C and yield Na-dolomite melt with about 50 mol% Na2CO3.
Implications Kimberlites, lamproites and alkali basalts carry debris of the underlying lithospheric mantle. Many of these xenoliths show chemical, textural, and mineralogical evidence for metasomatism by fluids or melts within the mantle (Menzies and Hawkesworth 1986; Green and Wallace 1988). Recently Giuliani et al. (2012) reported Na-rich carbonate melts preserved in primary multiphase inclusions hosted by metasomatic ilmenite grains contained in a spinel harzburgite from the Bultfontein kimberlite (Kimberley, South Africa). Mineral thermometry indicates that the spinel harzburgite crystallized at ~ 860 °C, which corresponds to pressures between 3.4 and 3.5 GPa (~ 110–115 km) on a 40 mW/m2 geotherm. This temperature corresponds to the lower stability limit of Na–Ca and Na–Ca–Mg carbonate melts as it follows from the present results and our preliminary data on the Na–Ca–Mg carbonate system. According to our experimental data, the specific feature of such near solidus carbonate melts is extremely high Na2CO3 content, about 50 mol%. The minimum melting temperatures and melt compositions versus pressure for the N a2CO3–CaCO3 join are summarized in Fig. 8. These data suggest that the Na–Ca carbonate melts may occur under suitable geothermal conditions.
Fig. 7 Comparison of T–X phase diagrams for the N a2CO3–CaCO3 join at 0.1 GPa (Cooper et al. 1975), 3 GPa (this study), and 6 GPa (Shatskiy et al. 2013c). Melting point of CaCO3 at 0.1 GPa—(Wyllie and Tuttle 1960) and at 3 GPa—(Shatskiy et al. 2018). Cal calcite, Arg aragonite, Na2 CaCO3 solid solution in N a2CO3, Na4Ca Na4Ca(CO3)3, Na2Ca Na2Ca(CO3)2, Na2Ca2 Na2Ca2(CO3)3, Na2Ca3 Na2Ca3(CO3)4, Na2Ca4 Na2Ca4(CO3)5, L liquid
According to the phase relations established at 6 GPa, shortite Na2Ca2(CO3)3 and nyerereite N a2Ca(CO3)2, the double carbonates observed at 0.1 GPa (Cooper et al. 1975), are not stable in the mantle (Shatskiy et al. 2013c) (Fig. 1b). The present study, however, revealed that the nyerereite stability field extends to the shallow mantle pressures of at least 3 GPa (Fig. 4). Although N a2Ca2(CO3)3 shortite is not stable at 3 GPa and ≥ 800 °C (Fig. 4), partial Mg substitution for Ca stabilizes shortite as a liquidus phase even at
Physics and Chemistry of Minerals Table 2 Calculation of volume effects for solid-phase reactions occurring in the Na2CO3– CaCO3 binary upon pressure decrease from 6 to 3 GPa, and from 3 to 0.1 GPa
1—(Dusek et al. 2003), 2—Rashchenko et al. (in preparation), 3—(Gavryushkin et al. 2016), 4—(Dickens et al. 1971), 5—(Gavryushkin et al. 2014), 6—(Rashchenko et al. 2017), 7—(Maslen et al. 1995), 8— (Antao and Hassan 2009)
The mass-balance calculations for these experiments [e.g., (Dasgupta and Hirschmann 2007; Ghosh et al. 2009; Moore 2012)] indicate missing alkaline carbonatite melt or subsolidus alkali-bearing carbonates. Acknowledgements The authors are very grateful to Robert Luth and an anonymous reviewer for constructive comments and helpful suggestions and Taku Tsuchiya for editorial handling. We thank Kathryn Moore and an anonymous reviewer for critical reading of an early draft of the manuscript. This study was financially supported by the Russian Foundation for Basic Research (Project No 17-05-00501). KL thanks for partial support from the state assignment Project No. 0330-2016-0006.
Fig. 8 Minimum melting temperatures versus pressure for the Na2CO3–CaCO3 join. The numbers indicate compositions of eutectic and peritectic melts coexisting with Ca-carbonate (upper line) or with Na–Ca double carbonates (lower line). The data at 0.1 GPa are from (Cooper et al. 1975), at 3 GPa (this study), and at 6 GPa (Shatskiy et al. 2013c). Geotherms for continental regions are from (Pollack and Chapman 1977)
3 GPa (Podborodnikov et al. in preparation). Consequently findings of nyerereite and shortite among daughter phases in the melt inclusions in olivine from the sheared garnet peridotites (Golovin et al. 2017b) are consistent with their mantle origin. We would additionally emphasize an importance of the studies on the pure carbonate systems or systems enriched by carbonate components to investigate the behavior of incipient melt fractions, which can be missed during experimental studies of natural systems with low CO2 and alkali contents.
Abersteiner A, Giuliani A, Kamenetsky VS, Phillips D (2017) Petrographic and melt-inclusion constraints on the petrogenesis of a magmaclast from the Venetia kimberlite cluster, South Africa. Chem Geol 455:331–341 Antao SM, Hassan I (2009) The orthorhombic structure of CaCO3, SrCO3, PbCO3 and B aCO3: linear structural trends. Can Mineral 47(5):1245–1255 Böttcher ME, Reutel C (1996) The Raman spectrum of alphaNa2Ca(CO3)2. J Raman Spectrosc 27(11):859–861 Brey GP, Bulatov VK, Girnis AV (2011) Melting of K-rich carbonated peridotite at 6–10 GPa and the stability of K-phases in the upper mantle. Chem Geol 281(3–4):333–342 Chen W, Kamenetsky VS, Simonetti A (2013) Evidence for the alkaline nature of parental carbonatite melts at Oka complex in Canada. Nat Commun 4:2687 Cooper AF, Gittins J, Tuttle OF (1975) The system Na2CO3-K2CO3-CaCO3 at 1 kbar and its significance in carbonatite petrogenesis. Am J Sci 275(5):534–560 Dasgupta R, Hirschmann MM (2007) Effect of variable carbonate concentration on the solidus of mantle peridotite. Am Miner 92(2–3):370–379 Dasgupta R, Hirschmann MM, Dellas N (2005) The effect of bulk composition on the solidus of carbonated eclogite from partial melting experiments at 3 GPa. Contrib Miner Petrol 149(3):288–305
Dasgupta R, Hirschmann MM, Stalker K (2006) Immiscible transition from carbonate-rich to silicate-rich melts in the 3 GPa melting interval of eclogite plus CO2 and genesis of silica-undersaturated ocean island lavas. J Petrol 47(4):647–671 De La Pierre M, Carteret C, Maschio L, André E, Orlando R, Dovesi R (2014) The Raman spectrum of CaCO3 polymorphs calcite and aragonite: a combined experimental and computational study. J Chem Phys 140(16):164509 Dickens B, Hyman A, Brown WE (1971) Crystal structure of Ca2Na2(CO3)3 (shortite). J Res Natl Bur Stand Sect A Phys Chem A 75(2):129–140 Dusek M, Chapuis G, Meyer M, Petricek V (2003) Sodium carbonate revisited. Acta Crystallogr Sect B Struct Sci 59:337–352 Gavryushkin PN, Bakakin VV, Bolotina NB, Shatskiy AF, Seryotkin YV, Litasov KD (2014) Synthesis and crystal structure of new carbonate Ca3Na2(CO3)4 homeotypic with orthoborates M3Ln2(BO3)4 (M = Ca, Sr, and Ba). Cryst Growth Des 14(9):4610–4616 Gavryushkin PN, Thomas VG, Bolotina NB, Bakakin VV, Golovin AV, Seryotkin YV, Fursenko DA, Litasov KD (2016) Hydrothermal synthesis and structure solution of Na2Ca(CO3)2: “Synthetic Analogue” of mineral nyerereite. Cryst Growth Des 16(4):1893–1902 Ghosh S, Ohtani E, Litasov KD, 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 Giuliani A, Kamenetsky VS, Phillips D, Kendrick MA, Wyatt BA, Goemann K (2012) Nature of alkali-carbonate fluids in the subcontinental lithospheric mantle. Geology 40(11):967–970 Golovin AV, Sharygin VV, Pokhilenko NP (2007) Melt inclusions in olivine phenocrysts in unaltered kimberlites from the UdachnayaEast pipe, Yakutia: some aspects of kimberlite magma evolution during late crystallization stages. Petrology 15(2):168–183 Golovin A, Korsakov A, Gavryushkin P, Zaitsev A, Thomas V, Moine B (2017a) Raman spectra of nyerereite, gregoryite, and synthetic pure Na2Ca(CO3)2: diversity and application for the study micro inclusions. J Raman Spectrosc 48(11):1559–1565 Golovin AV, Sharygin IS, Korsakov AV (2017b) Origin of alkaline carbonates in kimberlites of the Siberian craton: evidence from melt inclusions in mantle olivine of the Udachnaya-East pipe. Chem Geol 455:357–375 Golovin A, Sharygin I, Kamenetsky V, Korsakov A, Yaxley G (2018) Alkali-carbonate melts from the base of cratonic lithospheric mantle: links to kimberlites. Chem Geol 483(4):261–274 Grassi D, Schmidt MW (2011) The melting of carbonated pelites from 70 to 700 km depth. J Petrol 52(4):765–789 Green DH, Wallace ME (1988) Mantle metasomatism by ephemeral carbonatite melts. Nature 336(6198):459–462 Hernlund J, Leinenweber K, Locke D, Tyburczy JA (2006) A numerical model for steady-state temperature distributions in solid-medium high-pressure cell assemblies. Am Miner 91(2–3):295–305 Hirschmann MM (2000) Mantle solidus: experimental constraints and the effects of peridotite composition. Geochem Geophy Geosyst 1(10):2000GC000070 Kamenetsky MB, Sobolev AV, Kamenetsky VS, Maas R, Danyushevsky LV, Thomas R, Pokhilenko NP, Sobolev NV (2004) Kimberlite melts rich in alkali chlorides and carbonates: a potent metasomatic agent in the mantle. Geology 32(10):845–848 Kamenetsky VS, Sharygin VV, Kamenetsky MB, Golovin AV (2006) Chloride-carbonate nodules in kimberlites from the udachnaya pipe: alternative approach to the evolution of kimberlite magmas. Geochem Int 44(9):935–940 Kamenetsky VS, Kamenetsky MB, Weiss Y, Navon O, Nielsen TFD, Mernagh TP (2009) How unique is the Udachnaya-East kimberlite? Comparison with kimberlites from the Slave Craton (Canada) and SW Greenland. Lithos 112:334–346
Physics and Chemistry of Minerals Kamenetsky VS, Grütter H, Kamenetsky MB, Gömann K (2013) Parental carbonatitic melt of the Koala kimberlite (Canada): constraints from melt inclusions in olivine and Cr-spinel, and groundmass carbonate. Chem Geol 353:96–111 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(5):797–816 Kiseeva ES, Litasov KD, Yaxley GM, Ohtani E, Kamenetsky VS (2013) Melting phase relations of carbonated eclogite at 9–21 GPa and alkali-rich melts in the deep mantle. J Petrol 54(8):1555–1583 Lavrent’ev YG, Karmanov N, Usova L (2015) Electron probe microanalysis of minerals: microanalyzer or scanning electron microscope? Russ Geol Geophys 56(8):1154–1161 Li Z (2015) Melting and structural transformations of carbonates and hydrous phases in Earth’s mantle, Doctor of Philosophy (Geology). The University of Michigan, United States Li Z, Li J, Lange R, Liu J, Militzer B (2017) Determination of calcium carbonate and sodium carbonate melting curves up to Earth’s transition zone pressures with implications for the deep carbon cycle. Earth Planet Sci Lett 457:395–402 Litasov KD, Shatskiy A, Ohtani E, Yaxley GM (2013) The solidus of alkaline carbonatite in the deep mantle. Geology 41(1):79–82 Maslen E, Streltsov V, Streltsova N, Ishizawa N (1995) Electron density and optical anisotropy in rhombohedral carbonates. III. Synchrotron X-ray studies of CaCO3, MgCO3 and M nCO3. Acta Crystallogr Sect B Struct Sci 51(6):929–939 Menzies M, Hawkesworth C (1986) Mantle metasomatism. Academic Press, London Mernagh TP, Kamenetsky VS, Kamenetsky MB (2011) A Raman microprobe study of melt inclusions in kimberlites from Siberia, Canada, SW Greenland and South Africa. Spectrochim Acta Part A Mol Biomol Spectrosc 80(1):82–87 Moore KR (2012) Experimental study in the Na2O–CaO–MgO–Al2O3– SiO2–CO2 system at 3 GPa: the effect of sodium on mantle melting to carbonate-rich liquids and implications for the petrogenesis of silicocarbonatites. Mineral Mag 76(2):285–309 Pollack HN, Chapman DS (1977) On the regional variation of heat flow, geotherms, and lithospheric thickness. Tectonophysics 38:279–296 Rashchenko SV, Bakakin VV, Shatskiy AF, Gavryushkin PN, Seryotkin YV, Litasov KD (2017) Noncentrosymmetric Na2Ca4(CO3)5 carbonate of “M13M23XY3Z” structural type and affinity between borate and carbonate structures for design of new optical materials. Cryst Growth Des 17(11):6079–6084 Sharygin I, Litasov K, Shatskiy A, Golovin A, Ohtani E, Pokhilenko N (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 melts and implications for mantle plumes. Gondwana Res 28:1391–1414 Shatskiy A, Gavryushkin PN, Sharygin IS, Litasov KD, Kupriyanov IN, Higo Y, Borzdov YM, Funakoshi K, Palyanov YN, Ohtani E (2013a) Melting and subsolidus phase relations in the system Na2CO3-MgCO3+-H2O at 6 GPa and the stability of Na2Mg(CO3)2 in the upper mantle. Am Miner 98(11–12):2172–2182 Shatskiy A, Sharygin IS, Gavryushkin PN, Litasov KD, Borzdov YM, Shcherbakova AV, Higo Y, Funakoshi K, Palyanov YN, Ohtani E (2013b) The system K2CO3–MgCO3 at 6 GPa and 900–1450 °C. Am Mineral 98(8–9):1593–1603 Shatskiy A, Sharygin IS, Litasov KD, Borzdov YM, Palyanov YN, Ohtani E (2013c) New experimental data on phase relations for the system Na2CO3–CaCO3 at 6 GPa and 900–1400 °C. Am Miner 98(11–12):2164–2171 Shatskiy A, Borzdov YM, Litasov KD, Kupriyanov IN, Ohtani E, Palyanov YN (2014) Phase relations in the system FeCO3–CaCO3 at
Physics and Chemistry of Minerals 6 GPa and 900–1700 °C and its relation to the system CaCO3– FeCO3–MgCO3. Am Miner 99(4):773–785 Shatskiy A, Gavryushkin PN, Litasov KD, Koroleva ON, Kupriyanov IN, Borzdov YM, Sharygin IS, Funakoshi K, Palyanov YN, Ohtani E (2015a) Na-Ca carbonates synthesized under uppermantle conditions: Raman spectroscopic and X-ray diffraction studies. Eur J Mineral 27:175–184 Shatskiy AF, Litasov KD, Palyanov YN (2015b) Phase relations in carbonate systems at pressures and temperatures of lithospheric mantle: review of experimental data. Russ Geol Geophys 56:113–142 Shatskiy A, Litasov KD, Palyanov YN, Ohtani E (2016a) Phase relations on the K2CO3–CaCO3–MgCO3 join at 6 GPa and 900– 1400 °C: implication for incipient melting in carbonated mantle domains. Am Miner 101(2):437–447 Shatskiy A, Litasov KD, Sharygin IS, Egonin IA, Mironov AM, Palyanov YN, Ohtani E (2016b) The system N a2CO3–CaCO3–MgCO3 at 6 GPa and 900–1250 °C and its relation to the partial melting of carbonated mantle. High Press Res 36(1):23–41 Shatskiy A, Podborodnikov IV, Arefiev AV, Litasov KD, Chanyshev AD, Sharygin IS, Karmanov NS, Ohtani E (2017) Effect of alkalis on the reaction of clinopyroxene with Mg-carbonate at 6 GPa: Implications for partial melting of carbonated lherzolite. Am Miner 102(9):1934–1946
Shatskiy A, Podborodnikov IV, Arefiev AV, Minin DA, Chanyshev AD, Litasov KD (2018) Revision of the CaCO3–MgCO3 phase diagram at 3 and 6 GPa. Am Miner 103(3):441–452 Soltys A, Giuliani A, Phillips D, Kamenetsky VS, Maas R, Woodhead J, Rodemann T (2016) In-situ assimilation of mantle minerals by kimberlitic magmas—direct evidence from a garnet wehrlite xenolith entrained in the Bultfontein kimberlite (Kimberley, South Africa. Lithos 256:182–196 Song Y, Luo M, Zhao D, Peng G, Lin C, Ye N (2017) Explorations of new UV nonlinear optical materials in the N a2CO3–CaCO3 system. J Mater Chem C 5(34):8758–8764 Stoppa F, Jones A, Sharygin V (2009) Nyerereite from carbonatite rocks at Vulture volcano: implications for mantle metasomatism and petrogenesis of alkali carbonate melts Research Article. Open Geosci 1(2):131–151 Thomson AR, Walter MJ, Kohn SC, Brooker RA (2016) Slab melting as a barrier to deep carbon subduction. Nature 529:76–79 Walter M, Presnall DC (1994) Melting behavior of simplified lherzolite in the system CaO–MgO–Al2O3–SiO2–Na2O from 7 to 35 kbar. J Petrol 35(2):329–359 Wyllie PJ, Tuttle OF (1960) The system CaO–CO2–H2O and the origin of carbonatites. J Petrol 1(1):1–46