News Registration Abstract submission Deadlines Excursions Accommodation Organizing committee
First circular Second circular Abstracts Seminar History Program Travel Contact us
Первый циркуляр
Второй циркуляр
Оформление тезисов
Важные даты
Обратная связь

Тезисы международной конференции

Рудный потенциал щелочного, кимберлитового

 и карбонатитового магматизма

Abstracts of International conference

Ore potential of alkaline, kimberlite

and carbonatite magmatism


Stability and breakdown of Ca13CO3 melt combined with formation of 13C -diamond in static experiments up to 40 GPa and 4000 K

Spivak A.V.*, Litvin Yu.A.*, Dubrovinsky L.S.**

* Institution of the Russian Academy if Science Institute of Experimental Mineralogy RAS, Chernogolovka, Russia, spivak@iem.ac.ru

** Bayerisches Geoinstitut, University Bayreuth, Bayreuth, Germany


Carbonate melts have a crucial role for diamond forming carbonatite medium: (1) they are effective solvents of silicates, aluminum silicates, oxides; (2) carbonate melts are the basis of completely miscible carbonate-silicate and carbonate-silicate-oxide melts. Solid carbon (diamond and metastable graphite) is very soluble in the atomic form in these melts at the PT conditions of diamond stability [Spivak et al., 2008]. Currently, information about the phase state of carbonates is contradictory for the conditions of the transition zone and lower mantle. The melt of CaCO3 undergoes decomposition by the reaction of CaCO3 = CaO + CO2 in the shock-wave experiments at 3200 - 3500 K and 80 GPa [Ivanov AB, Deutsch A., 2002]. However, the direct evidence of decomposition reaction of carbonate melt at the maximum degree of compression or under the influence of high residual temperature after short-term relief to the normal pressure could not be obtained in the shock-wave experiment. Nevertheless, the available experimental data and the calculated equations of state of CaCO3 and its decomposition products CaO and CO2 summarized in a phase diagram of CaCO3. Congruent melting curve of CaCO3 was studied in a static experiment up to 7 GPa, and extrapolated on the basis of the equation of state [Irving A.J., Wyllie P.J., 1973]. In the interval 10 - 30 GPa the congruent melting field CaCO3 is limited by the melting curve of CaCO3 at 2100 - 2500 K and the proposed curve of decomposition of CaCO3 melt to CaO and CO2 at high temperatures 3500 - 3800 K. Meanwhile, accurate understanding of physical and chemical behavior of CaCO3 and other carbonates is of fundamental significance for a number of important problems of mineralogy and geochemistry of the transition zone and lower mantle. Especially, genesis of super deep diamonds, as well as the origin and evolution of carbonate-containing melts (carbonatites , kimberlites, etc.) are among the problems

The goal of this work is experimental study of the phase state of CaCO3 at static pressure up to 40 GPa (generated in the apparatus with diamond anvils) and temperatures up to 4000 K, obtained by laser heating of a strongly compressed sample (for details see [Eremets M., 1996 ]). The PT-parameters of the experiment is consistent with the physical conditions of formation of super deep diamonds, which overlap with the possible field of congruent melting of Ca-carbonate. An important feature of this work is the use of isotope individual carbonate Ca13CO3 that excludes the possibility of distortion of results of 12C of the anvils (natural diamond) is unexpectedly involved into products of the experiments.

Starting materials were chemical reagents of Ca13CO3 (prepared on the basis of the isotope 13C). The experimental sample consists of two layers of powder Ca13CO3, and a thin layer of platinum powder between them. The sample is placed into the hole of 150 µm diameter (filled by an inert gas neon) in a metallic gasquet of rhenium. The gasquet is clamped between the diamond anvils with a working surface 350 µm. Nd: YLF infrared laser (wavelength 1064 nm) was used for heating. The duration of heat was about 5 minutes. The pressure in the sample is determined by the shift of the ruby luminescence. A ruby grain of ~ 5 µm size is placed inside the sample. During lowering the heating temperature to the room value, the sample has remained under strong compression inside Re gasquet.  Products of experiments were studied using micro-Raman spectroscopy. The system LabRam with He-Nd-laser (exciting wavelength 632 nm) was used for register Raman spectra. Spectra from different parts of the sample collected at the gradual decompression, as well as after quenching, when the sample is completely extracted from the working holes in rhenium gasquet.

Experimental samples are marked (Fig. 3) with the visible zone of calcium carbonate melting (about 50 microns) in experiments at 20-22 GPa and 3500 K (Fig. 1). The sample in the experiment at 11 GPa and 3500 K are not visualy different from them, but their Raman spectra include a peak of graphite. The samples were not heated along the perimeter, and these areas were used for comparative estimation. Raman spectra of samples under pressure were collected before and after heating. Also, the spectra of Ca12CO3 and C13CO3, graphite on the basis of the isotopes 12C and 13C were obtained for the comparison.

Under normal conditions, Raman spectra Ca12CO3 have characteristic bands at 155, 280, 712 and 1086 cm-1 [Rutt H.N., Nicola J.H., 1974]. The spectra of Ca13CO3 contain the main band at 153, 278, 710, 1086 cm-1, and the spectra of graphite on the basis of the isotope 13C - the band at 1278 and 1535 cm-1. Calcite on the basis of the isotope 13C transforms into aragonite in experiments at high pressures and temperatures, as evidenced  the spectra which have the characteristic bands of aragonite 703 and 197 cm-1 (according to their displacement due to the presence of the isotope 13C). The most intense band at 1086 cm-1 is characteristic for the two polymorphic modifications of calcium carbonate and can not be used as a distinctive feature of them.

Graphite on the basis of the isotope 12C was found in the central parts with a diameter of 50 microns after heating to 3000 K at 20-22 GPa. Raman spectra of the experimental samples contain bands of graphite 12C, including a broad band with a peak in the range at 1580-1585 cm-1, known as “band G”, and a broad band with a peak in the range at 1340-1355 cm-1 - "band D”. The position and width of the G band are determined by the perfection of graphite structure and the position of the D band is identified with various types of violations of the order in the structure of graphite. Because Ca13CO3 was used as the starting material, the formation of graphite 12C by carbon 13C from this carbonate should be entirely avoided entirely. Source of carbon for graphite 12C could be just the carbon from diamond anvils. It can not exclude that in the experiments described in [Bayarjargal L. et al., 2010], the same occurred. Thus, the effect of the congruent melting of  Ca13CO3 is determined in experiments at 3000 - 3500 K and 20 GPa. However, the Raman spectra of the sample obtained at 11 GPa and 3500 K, contain a broad band G with a maximum in the range at 1528-1537 cm-1, "echo" of the band at 1580 cm-1 and D band with a peak in the range at 1275-1285 cm -1. The Raman spectra of the sample obtained at 30-40 GPa and 4000 K contain band with a maximum at 1280 cm-1. In those cases we can meet the formation of carbon phases (diamond and graphite), with a source of carbon from the calcium carbonate on the basis of 13C isotope. The reality of this process is supported by RT-conditions of the experiment, which refers to the corresponding field on the carbon phase diagram. This fact may testify of the reality of Ca13CO3 decomposition process on a two-step mechanism, which was discussed in [Bayarjargal L. et al., 2010]. Characteristic bands of graphite were not observed in Raman spectra of "unheated" parts of the sample.

As a result of the experiments, it is showed that calcium carbonate melts congruently at 20-22 GPa and 3500 K. The experimental data are in aggrement with the preliminary phase diagram of CaCO3 [Ivanov AB, Deutsch A., 2002], constructed on the basis of shock experiments and thermodynamic calculations. The data confirm the fact of congruent melting of CaCO3 (aragonite) at 20 - 22 GPa and 3500 K. This means that the field of congruent melting of calcium carbonate is quite wide, extending from 2300 to 3500 - 3800 K at 20-22 GPa. However, the results of experiments at 11 and 30-40 GPa and 4000 K demonstrate the possibility of high-temperature phase boundary of the decomposition of CaCO3 to CaO melt and compressed CO2 fluid phase.

CaCO3 is a representative inclusion in diamonds from transition zone and lower mantle. Existence of a broad field of CaCO3 congruent melting allows us to consider deep carbonate melts based on CaCO3 as a possible parental media for super deep diamonds.


Support: МК-913.2011.5, Program RAS №02, 10-05-00654, 11-05-00401, НШ-3654-2011-5, FPP 2011-1.3.1-151-006_6.



Bayarjargal L., T.G. Shumilova, A. Friedrich, B. Winkler Diamond formation from CaCO3 at high pressure and temperature // Eur. J. Miner. 2010. V. 22. P. 29-34.

Bundy F.P., W.A. Basset, M.S. Weathers, R.J. Hemley, H.K. Mao, A.F. Goncharov  The pressure-temperature phase and transformation diagram for carbon; updated through 1994 // Carbon 1996. V. 34. № 2. Р. 141-153.

Eremets M.  High Pressure Experimental Methods. New York. Oxford University Press Inc. 1996.  390 p.

Irving A.J., P.J. Wyllie Melting relationships in CaO-CO2 and MgO-CO2 to 33 kbar // Earth Planet. Sci. Lett. 1973 V. 20. Р. 220-225.

Ivanov A.B., A. Deutsch The phase diagram of CaCO3 in relation to shock compression and decomposition // Phys. Earth Planet. Inter. 2002. V. 129. P. 131-143.

Rutt H.N., J.H. Nicola Raman spectra of carbonates of calcite structure // J. Phys. C: Solid State Phys. 1974. 7. 4522-4528.

Spivak A.V., Yu.A. Litvin, A.V. Shushkanova, V.Yu. Litvin, A.A. Shiryaev Diamond formation in carbonate-silicate-sulfide-carbon melts: Raman- and IR-spectroscopy // Eur. J. Mineral. 2008. 20. P. 341-347.