Advanced Search

Indexed by SCI、CA、РЖ、PA、CSA、ZR、etc .

Volume 32 Issue 4
Aug.  2021
Turn off MathJax
Article Contents

Tianlei Zhai, Shengxuan Huang, Shan Qin, Jingjing Niu, Yu Gong. Redox-Induced Destabilization of Dolomite at Earth's Mantle Transition Zone. Journal of Earth Science, 2021, 32(4): 880-886. doi: 10.1007/s12583-021-1410-6
Citation: Tianlei Zhai, Shengxuan Huang, Shan Qin, Jingjing Niu, Yu Gong. Redox-Induced Destabilization of Dolomite at Earth's Mantle Transition Zone. Journal of Earth Science, 2021, 32(4): 880-886. doi: 10.1007/s12583-021-1410-6

Redox-Induced Destabilization of Dolomite at Earth's Mantle Transition Zone

doi: 10.1007/s12583-021-1410-6
More Information
  • Carbonates are considered to be important hosts of oxidized carbon during subduction processes. Here we investigate the redox interactions between dolomite and metallic iron in laser-heated diamond anvil cells up to ~20 GPa. The identification of recovered samples via in-situ synchrotron X-ray diffraction and ex-situ Raman spectroscopy shows that the reaction occurs with the formation of ferropericlase, graphite and hexagonal diamond, while CaCO3 remains stable. The experimental results indicate dolomite and metallic iron phases cannot coexist and demonstrate a possible formation mechanism of ultradeep diamonds via redox reaction between dolomite and iron under the mantle transition zone conditions. The results are significant for understanding carbon transportation during subduction processes and have further implications to the processes in the more complex systems regarding to carbonate-silicate-metal phase relations.
  • 加载中
  • Ballhaus, C., 1995. Is the Upper Mantle Metal-Saturated? Earth and Planetary Science Letters, 132(1/2/3/4): 75-86. https://doi.org/10.1016/0012-821x(95)00047-g doi:  10.1016/0012-821x(95)00047-g
    Bayarjargal, L., Fruhner, C. J., Schrodt, N., et al., 2018. CaCO3 Phase Diagram Studied with Raman Spectroscopy at Pressures up to 50 GPa and High Temperatures and DFT Modeling. Physics of the Earth and Planetary Interiors, 281: 31-45. https://doi.org/10.1016/j.pepi.2018.05.002 doi:  10.1016/j.pepi.2018.05.002
    Brenker, F. E., Vollmer, C., Vincze, L., et al., 2007. Carbonates from the Lower Part of Transition Zone or even the Lower Mantle. Earth and Planetary Science Letters, 260(1/2): 1-9. https://doi.org/10.1016/j.epsl.2007.02.038 doi:  10.1016/j.epsl.2007.02.038
    Bureau, H., Frost, D. J., Bolfan-Casanova, N., et al., 2016. Diamond Growth in Mantle Fluids. Lithos, 265: 4-15. https://doi.org/10.1016/j.lithos.2016.10.004 doi:  10.1016/j.lithos.2016.10.004
    Chen, M., Shu, J. F., Xie, X. D., et al., 2018. Natural Diamond Formation by Self-Redox of Ferromagnesian Carbonate. Proceedings of the National Academy of Sciences of the United States of America, 115(11): 2676-2680. https://doi.org/10.1073/pnas.1720619115 doi:  10.1073/pnas.1720619115
    Dasgupta, R., Hirschmann, M. M., 2006. Melting in the Earth's Deep Upper Mantle Caused by Carbon Dioxide. Nature, 440(7084): 659-662. https://doi.org/10.1038/nature04612 doi:  10.1038/nature04612
    Dasgupta, R., Hirschmann, M. M., 2010. The Deep Carbon Cycle and Melting in Earth's Interior. Earth and Planetary Science Letters, 298(1/2): 1-13. https://doi.org/10.1016/j.epsl.2010.06.039 doi:  10.1016/j.epsl.2010.06.039
    Dorfman, S. M., Badro, J., Nabiei, F., et al., 2018. Carbonate Stability in the Reduced Lower Mantle. Earth and Planetary Science Letters, 489: 84-91. https://doi.org/10.1016/j.epsl.2018.02.035 doi:  10.1016/j.epsl.2018.02.035
    Drewitt, J. W. E., Walter, M. J., Zhang, H. L., et al., 2019. The Fate of Carbonate in Oceanic Crust Subducted into Earth's Lower Mantle. Earth and Planetary Science Letters, 511: 213-222. https://doi.org/10.1016/j.epsl.2019.01.041 doi:  10.1016/j.epsl.2019.01.041
    Dubrovinsky, L., Glazyrin, K., McCammon, C., et al., 2009. Portable Laser-Heating System for Diamond Anvil Cells. Journal of Synchrotron Radiation, 16(6): 737-741. https://doi.org/10.1107/s0909049509039065 doi:  10.1107/s0909049509039065
    Farsang, S., Facq, S., Redfern, S. A. T., 2018. Raman Modes of Carbonate Minerals as Pressure and Temperature Gauges up to 6 GPa and 500 ℃. American Mineralogist, 103(12): 1988-1998. https://doi.org/10.2138/am-2018-6442 doi:  10.2138/am-2018-6442
    Fei, Y. W., Zhang, L., Corgne, A., et al., 2007. Spin Transition and Equations of State of (Mg, Fe)O Solid Solutions. Geophysical Research Letters, 34(17): L17307. https://doi.org/10.1029/2007gl030712 doi:  10.1029/2007gl030712
    Fiquet, G., Guyot, F., Kunz, M., et al., 2002. Structural Refinements of Magnesite at very High Pressure. American Mineralogist, 87(8/9): 1261-1265. https://doi.org/10.2138/am-2002-8-927 doi:  10.2138/am-2002-8-927
    Gao, J., Niu, J. J., Qin, S., et al., 2017. Ultradeep Diamonds Originate from Deep Subducted Sedimentary Carbonates. Science China Earth Sciences, 60(2): 207-217. https://doi.org/10.1007/s11430-016-5151-4 doi:  10.1007/s11430-016-5151-4
    Hammersley, A. P., Svensson, S. O., Hanfland, M., et al., 1996. Two-Dimensional Detector Software: From Real Detector to Idealised Image or Two-Theta Scan. High Pressure Research, 14(4/5/6): 235-248. https://doi.org/10.1080/08957959608201408 doi:  10.1080/08957959608201408
    Jephcoat, A. P., Finger, L. W., Cox, D. E., 1992. High Pressure, High Resolution Synchrotron X-Ray Powder Diffraction with a Position-Sensitive Detector. High Pressure Research, 8(5/6): 667-676. https://doi.org/10.1080/08957959208206318 doi:  10.1080/08957959208206318
    Kaminsky, F., Matzel, J., Jacobsen, B., et al., 2016. Isotopic Fractionation of Oxygen and Carbon in Decomposed Lower-Mantle Inclusions in Diamond. Mineralogy and Petrology, 110(2/3): 379-385. https://doi.org/10.1007/s00710-015-0401-7 doi:  10.1007/s00710-015-0401-7
    Kaminsky, F., Wirth, R., Matsyuk, S., et al., 2009. Nyerereite and Nahcolite Inclusions in Diamond: Evidence for Lower-Mantle Carbonatitic Magmas. Mineralogical Magazine, 73(5): 797-816. https://doi.org/10.1180/minmag.2009.073.5.797 doi:  10.1180/minmag.2009.073.5.797
    Li, X. Y., Zhang, Z. G., Lin, J. F., et al., 2018. New High-Pressure Phase of CaCO3 at the Topmost Lower Mantle: Implication for the Deep-Mantle Carbon Transportation. Geophysical Research Letters, 45(3): 1355-1360. https://doi.org/10.1002/2017gl076536 doi:  10.1002/2017gl076536
    Litvin, Y. A., Spivak, A. V., Dubrovinsky, L. S., 2016. Magmatic Evolution of the Material of the Earth's Lower Mantle: Stishovite Paradox and Origin of Superdeep Diamonds (Experiments at 24-26 GPa). Geochemistry International, 54(11): 936-947. https://doi.org/10.1134/s0016702916090032 doi:  10.1134/s0016702916090032
    Liu, J., Dubrovinsky, L., Boffa Ballaran, T., et al., 2007. Equation of State and Thermal Expansivity of LiF and NaF. High Pressure Research, 27(4): 483-489. https://doi.org/10.1080/08957950701684690 doi:  10.1080/08957950701684690
    Maeda, F., Ohtani, E., Kamada, S., et al., 2017. Diamond Formation in the Deep Lower Mantle: A High-Pressure Reaction of MgCO3 and SiO2. Scientific Reports, 7: 40602. https://doi.org/10.1038/srep40602 doi:  10.1038/srep40602
    Mao, H. K., Bassett, W. A., Takahashi, T., 1967. Effect of Pressure on Crystal Structure and Lattice Parameters of Iron up to 300 kbar. Journal of Applied Physics, 38(1): 272-276. https://doi.org/10.1063/1.1708965 doi:  10.1063/1.1708965
    Mao, Z., Armentrout, M., Rainey, E., et al., 2011a. Dolomite Ⅲ: A New Candidate Lower Mantle Carbonate. Geophysical Research Letters, 38(22): L22303. https://doi.org/10.1029/2011gl049519 doi:  10.1029/2011gl049519
    Mao, Z., Lin, J. F., Liu, J., et al., 2011b. Thermal Equation of State of Lower-Mantle Ferropericlase across the Spin Crossover. Geophysical Research Letters, 38(23): L23308. https://doi.org/10.1029/2011gl049915 doi:  10.1029/2011gl049915
    Martinez, I., Zhang, J., Reeder, R. J., 1996. In situ X-Ray Diffraction of Aragonite and Dolomite at High Pressure and High Temperature: Evidence for Dolomite Breakdown to Aragonite and Magnesite. American Mineralogist, 81(5/6): 611-624. https://doi.org/10.2138/am-1996-5-608 doi:  10.2138/am-1996-5-608
    Martirosyan, N. S., Litasov, K. D., Lobanov, S. S., et al., 2019a. The Mg-Carbonate-Fe Interaction: Implication for the Fate of Subducted Carbonates and Formation of Diamond in the Lower Mantle. Geoscience Frontiers, 10(4): 1449-1458. https://doi.org/10.1016/j.gsf.2018.10.003 doi:  10.1016/j.gsf.2018.10.003
    Martirosyan, N. S., Shatskiy, A., Chanyshev, A. D., et al., 2019b. Effect of Water on the Magnesite-Iron Interaction, with Implications for the Fate of Carbonates in the Deep Mantle. Lithos, 326/327: 435-445. https://doi.org/10.1016/j.lithos.2019.01.004 doi:  10.1016/j.lithos.2019.01.004
    Martirosyan, N. S., Litasov, K. D., Shatskiy, A. F., et al., 2015a. Reactions of Iron with Calcium Carbonate at 6 GPa and 1 273-1 873 K: Implications for Carbonate Reduction in the Deep Mantle. Russian Geology and Geophysics, 56(9): 1322-1331. https://doi.org/10.1016/j.rgg.2015.08.008 doi:  10.1016/j.rgg.2015.08.008
    Martirosyan, N. S., Litasov, K. D., Shatskiy, A., et al., 2015b. The Reactions between Iron and Magnesite at 6 GPa and 1 273-1 873 K: Implication to Reduction of Subducted Carbonate in the Deep Mantle. Journal of Mineralogical and Petrological Sciences, 110(2): 49-59. https://doi.org/10.2465/jmps.141003a doi:  10.2465/jmps.141003a
    Martirosyan, N. S., Yoshino, T., Shatskiy, A., et al., 2016. The CaCO3-Fe Interaction: Kinetic Approach for Carbonate Subduction to the Deep Earth's Mantle. Physics of the Earth and Planetary Interiors, 259: 1-9. https://doi.org/10.1016/j.pepi.2016.08.008 doi:  10.1016/j.pepi.2016.08.008
    Merlini, M., Cerantola, V., Gatta, G. D., et al., 2017. Dolomite-Ⅳ: Candidate Structure for a Carbonate in the Earth's Lower Mantle. American Mineralogist, 102(8): 1763-1766. https://doi.org/10.2138/am-2017-6161 doi:  10.2138/am-2017-6161
    Merlini, M., Crichton, W. A., Hanfland, M., et al., 2012. Structures of Dolomite at Ultrahigh Pressure and Their Influence on the Deep Carbon Cycle. Proceedings of the National Academy of Sciences of the United States of America, 109(34): 13509-13514. https://doi.org/10.1073/pnas.1201336109 doi:  10.1073/pnas.1201336109
    Németh, P., Garvie, L. A. J., Aoki, T., et al., 2014. Lonsdaleite is Faulted and Twinned Cubic Diamond and does not Exist as a Discrete Material. Nature Communications, 5: 5447. https://doi.org/10.1038/ncomms6447 doi:  10.1038/ncomms6447
    Nestola, F., Korolev, N., Kopylova, M., et al., 2018. CaSiO3 Perovskite in Diamond Indicates the Recycling of Oceanic Crust into the Lower Mantle. Nature, 555(7695): 237-241. https://doi.org/10.1038/nature25972 doi:  10.1038/nature25972
    Oganov, A. R., Hemley, R. J., Hazen, R. M., et al., 2013. Structure, Bonding, and Mineralogy of Carbon at Extreme Conditions. Reviews in Mineralogy and Geochemistry, 75(1): 47-77. https://doi.org/10.2138/rmg.2013.75.3 doi:  10.2138/rmg.2013.75.3
    Palyanov, Y. N., Bataleva, Y. V., Sokol, A. G., et al., 2013. Mantle-Slab Interaction and Redox Mechanism of Diamond Formation. Proceedings of the National Academy of Sciences of the United States of America, 110(51): 20408-20413. https://doi.org/10.1073/pnas.1313340110 doi:  10.1073/pnas.1313340110
    Patterson, J. R., Kudryavtsev, A., Vohra, Y. K., 2002. X-Ray Diffraction and Nanoindentation Studies of Nanocrystalline Graphite at High Pressures. Applied Physics Letters, 81(11): 2073-2075. https://doi.org/10.1063/1.1508169 doi:  10.1063/1.1508169
    Pócsik, I., Hundhausen, M., Koós, M., et al., 1998. Origin of the D Peak in the Raman Spectrum of Microcrystalline Graphite. Journal of Non- Crystalline Solids, 227-230(2): 1083-1086. https://doi.org/10.1016/s0022-3093(98)00349-4 doi:  10.1016/s0022-3093(98)00349-4
    Reich, S., Thomsen, C., 2004. Raman Spectroscopy of Graphite. Philosophical Transactions Series A, Mathematical, Physical, and Engineering Sciences, 362(1824): 2271-2288. https://doi.org/10.1098/rsta.2004.1454 doi:  10.1098/rsta.2004.1454
    Rohrbach, A., Ballhaus, C., Golla-Schindler, U., et al., 2007. Metal Saturation in the Upper Mantle. Nature, 449(7161): 456-458. https://doi.org/10.1038/nature06183 doi:  10.1038/nature06183
    Rohrbach, A., Schmidt, M. W., 2011. Redox Freezing and Melting in the Earth's Deep Mantle Resulting from Carbon-Iron Redox Coupling. Nature, 472(7342): 209-212. https://doi.org/10.1038/nature09899 doi:  10.1038/nature09899
    Santillán, J., Williams, Q., Knittle, E., 2003. Dolomite-Ⅱ: A High-Pressure Polymorph of CaMg(CO3)2. Geophysical Research Letters, 30(2): 1054. https://doi.org/10.1029/2002gl016018 doi:  10.1029/2002gl016018
    Sato, K., Katsura, T., 2001. Experimental Investigation on Dolomite Dissociation into Aragonite+Magnesite up to 8.5 GPa. Earth and Planetary Science Letters, 184(2): 529-534. https://doi.org/10.1016/S0012-821X(00)00346-0 doi:  10.1016/S0012-821X(00)00346-0
    Schindler, T. L., Vohra, Y. K., 1995. A Micro-Raman Investigation of High-Pressure Quenched Graphite. Journal of Physics: Condensed Matter, 7(47): L637-L642. https://doi.org/10.1088/0953-8984/7/47/001 doi:  10.1088/0953-8984/7/47/001
    Smith, D. C., Godard, G., 2009. UV and VIS Raman Spectra of Natural Lonsdaleites: Towards a Recognised Standard. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 73(3): 428-435. https://doi.org/10.1016/j.saa.2008.10.025 doi:  10.1016/j.saa.2008.10.025
    Stagno, V., Frost, D. J., McCammon, C. A., et al., 2015. The Oxygen Fugacity at which Graphite or Diamond Forms from Carbonate-Bearing Melts in Eclogitic Rocks. Contributions to Mineralogy and Petrology, 169(2): 1-18. https://doi.org/10.1007/s00410-015-1111-1 doi:  10.1007/s00410-015-1111-1
    Sung, J., 2000. Graphite→Diamond Transition under High Pressure: A Kinetics Approach. Journal of Materials Science, 35(23): 6041-6054. https://doi.org/10.1023/a:1026779802263 doi:  10.1023/a:1026779802263
    Thomson, A. R., Kohn, S. C., Bulanova, G. P., et al., 2014. Origin of Sub-Lithospheric Diamonds from the Juina-5 Kimberlite (Brazil): Constraints from Carbon Isotopes and Inclusion Compositions. Contributions to Mineralogy and Petrology, 168(6): 1-29. https://doi.org/10.1007/s00410-014-1081-8 doi:  10.1007/s00410-014-1081-8
    Thomson, A. R., Walter, M. J., Kohn, S. C., et al., 2016. Slab Melting as a Barrier to Deep Carbon Subduction. Nature, 529(7584): 76-79. https://doi.org/10.1038/nature16174 doi:  10.1038/nature16174
    Toby, B. H., 2001. EXPGUI, a Graphical User Interface for GSAS. Journal of Applied Crystallography, 34(2): 210-213. https://doi.org/10.1107/s0021889801002242 doi:  10.1107/s0021889801002242
    van der Hilst, R. D., Widiyantoro, S., Engdahl, E. R., 1997. Evidence for Deep Mantle Circulation from Global Tomography. Nature, 386(6625): 578-584. https://doi.org/10.1038/386578a0 doi:  10.1038/386578a0
    van Westrenen, W., Li, J., Fei, Y. W., et al., 2005. Thermoelastic Properties of (Mg0.64Fe0.36)O Ferropericlase Based on in situ X-Ray Diffraction to 26.7 GPa and 2 173 K. Physics of the Earth and Planetary Interiors, 151(1/2): 163-176. https://doi.org/10.1016/j.pepi.2005.03.001 doi:  10.1016/j.pepi.2005.03.001
    Walter, M. J., Kohn, S. C., Araujo, D., et al., 2011. Deep Mantle Cycling of Oceanic Crust: Evidence from Diamonds and Their Mineral Inclusions. Science, 334(6052): 54-57. https://doi.org/10.1126/science.1209300 doi:  10.1126/science.1209300
    Yang, J. S., Shen, T. T., Zhang, C., et al., 2019. Preface: Introduction of IGCP 649 Project-Diamonds and Recycled Mantle. Journal of Earth Science, 30(3): 429-430. https://doi.org/10.1007/s12583-019-1229-6 doi:  10.1007/s12583-019-1229-6
    Zhang, X., Yang, S. Y., Zhao, H., et al., 2019. Effect of Beam Current and Diameter on Electron Probe Microanalysis of Carbonate Minerals. Journal of Earth Science, 30(4): 834-842. https://doi.org/10.1007/s12583-017-0939-x doi:  10.1007/s12583-017-0939-x
    Zhu, F., Li, J., Liu, J. C., et al., 2019. Kinetic Control on the Depth Distribution of Superdeep Diamonds. Geophysical Research Letters, 46(4): 1984-1992. https://doi.org/10.1029/2018gl080740 doi:  10.1029/2018gl080740
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(3)  / Tables(1)

Article Metrics

Article views(31) PDF downloads(5) Cited by()

Related
Proportional views

Redox-Induced Destabilization of Dolomite at Earth's Mantle Transition Zone

doi: 10.1007/s12583-021-1410-6

Abstract: Carbonates are considered to be important hosts of oxidized carbon during subduction processes. Here we investigate the redox interactions between dolomite and metallic iron in laser-heated diamond anvil cells up to ~20 GPa. The identification of recovered samples via in-situ synchrotron X-ray diffraction and ex-situ Raman spectroscopy shows that the reaction occurs with the formation of ferropericlase, graphite and hexagonal diamond, while CaCO3 remains stable. The experimental results indicate dolomite and metallic iron phases cannot coexist and demonstrate a possible formation mechanism of ultradeep diamonds via redox reaction between dolomite and iron under the mantle transition zone conditions. The results are significant for understanding carbon transportation during subduction processes and have further implications to the processes in the more complex systems regarding to carbonate-silicate-metal phase relations.

Tianlei Zhai, Shengxuan Huang, Shan Qin, Jingjing Niu, Yu Gong. Redox-Induced Destabilization of Dolomite at Earth's Mantle Transition Zone. Journal of Earth Science, 2021, 32(4): 880-886. doi: 10.1007/s12583-021-1410-6
Citation: Tianlei Zhai, Shengxuan Huang, Shan Qin, Jingjing Niu, Yu Gong. Redox-Induced Destabilization of Dolomite at Earth's Mantle Transition Zone. Journal of Earth Science, 2021, 32(4): 880-886. doi: 10.1007/s12583-021-1410-6
  • Carbon is a fundamental volatile element in our life and global environment. The carbon cycles between Earth's interior and its surface modulate Earth's atmosphere and climate, so it is crucial to explore how carbon cycles through exterior and interior reservoirs (Dasgupta and Hirschmann, 2010). Carbonates, the most abundant carbon-bearing minerals in the Earth's crust, can be subducted down into the deep Earth. Seismic, geological and geochemical studies have extended the subduction processes to the mantle transition zone and even to the core-mantle boundary (Thomson et al., 2014; van der Hilst et al., 1997), during which carbonates undergo melting, decomposition, structural transitions, interactions with the surrounding mantle, etc. Thus, carbonates during subduction can provide unique insights into the geochemistry, mineralogy and geodynamics of the deep Earth (Walter et al., 2011). Calcite, magnesite and dolomite are believed to be predominant carriers for oxidized carbon during subduction. Calcite has been suggested to undergo a series of phase transitions at mantle conditions (Bayarjargal et al., 2018). In contrast, magnesite is stable up to at least ~82 GPa and ~2 000 K (Fiquet et al., 2002). Dolomite is also considered as a potential carbon-carrier due to its relatively high stability (Merlini et al., 2017, 2012; Mao et al., 2011a). At Earth's surface conditions, dolomite CaMg(CO3)2 adopts a rhombohedral structure (R$ \overline{3} $). Dolomite will transform to dolomite Ⅱ and dolomite Ⅲ above ~17 and ~35 GPa, respectively, then dolomite Ⅳ will appear at the lowermost mantle conditions (Merlini et al., 2017, 2012; Mao et al., 2011a).

    The occurrence of carbonates as syngenetic inclusions in ultradeep diamonds suggests that the formation of these diamonds might be related to subducted carbonates (Yang et al., 2019). The coexistence of carbonates with deep Earth typical minerals such as ferropericlase, bridgmanite (brd) and CaSiO3-perovskite (CaPv) within those diamonds implies the origin of ultradeep diamonds might even reach the lower mantle (Kaminsky et al., 2009; Brenker et al., 2007). Numerous studies focused on the reactions between carbonates and mantle silicates (Stagno et al., 2015) or silica (Drewitt et al., 2019; Li et al., 2018; Maeda et al., 2017), which would produce diamond at mantle conditions, accompanied by the formation of CaPv and brd. The diamonds might also be formed by a self-oxidation-reduction mechanism (Chen et al., 2018). Furthermore, the redox interactions between carbonates and iron could take place according to previous studies (Martirosyan et al., 2019a, 2016, 2015a, b; Dorfman et al., 2018; Gao et al., 2017; Palyanov et al., 2013; Rohrbach and Schmidt, 2011), which would lead to the formation of diamonds.

    The redox buffer in the silicate mantle is believed to change from Fe2+/Fe3+ to Fe0/Fe2+ at depths > 250 km as the experiments and thermodynamic calculations predicted (Rohrbach and Schmidt, 2011; Ballhaus, 1995), arising from the disproportionation of Fe2+ in silicates such as bridgmanite, pyroxene and garnet. Thus, the redox interactions will occur because the iron-saturated mantle provides a relatively reduced environment conducive to carbon reduction. Carbonate-iron reactions could account for the presence of carbonate, ferropericlase, iron carbide and other typical mineral inclusions within ultradeep diamond.

    Redox interactions between carbonates and iron are recently proposed as an important mechanism for ultradeep diamond formation (Rohrbach and Schmidt, 2011). The MgCO3-Fe system has been extensively studied at high pressure and high temperature conditions from the upper mantle to the deep lower mantle, which presents a typical redox pathway for diamond formation (Martirosyan et al., 2019a, 2015b; Zhu et al., 2019; Gao et al., 2017; Palyanov et al., 2013). Palyanov et al. (2013) conducted redox-gradient experiments between Mg-carbonate and iron at upper mantle conditions in multianvil apparatus, and they presented a spatial distribution for diamond formation near carbonate-Fe boundary. However, the reduced carbon observed by Martirosyan et al. (2015b) was graphite rather than diamond under similar conditions. The depth of diamond formation via the reaction between MgCO3 and Fe was experimentally extended to mantle transition zone (Gao et al., 2017). Zhu et al. (2019) demonstrated that the reaction between MgCO3 and Fe was kinetically feasible in cold slabs subducted into the mantle transition zone and lower mantle. Martirosyan et al. (2019a) further suggested that this reaction could even take place in the lowermost mantle. The CaCO3-Fe interaction was investigated at conditions from upper mantle to upper mantle transition zone, and the data revealed the formation of carbide, Ca-wüstite and graphite, while no diamond was found (Martirosyan et al., 2016, 2015a). The reaction between dolomite and iron showed that under deep lower mantle conditions, the MgCO3 component decomposed whereas the CaCO3 remained stable (Dorfman et al., 2018). Until now, several aforementioned studies have confirmed this carbonate-iron interaction mechanism, but available experimental data are scarce at conditions related to the possible source depths of some ultradeep diamonds, i.e., the mantle transition zone (Gao et al., 2017).

    Here we investigate the interactions between dolomite and metallic iron utilizing laser-heated diamond anvil cells (DACs) combined with synchrotron radiation X-ray diffraction (XRD) and Raman spectroscopy under conditions relevant to mantle transition zone. The presented results are significant for interpreting the change and transportation of carbon-hosts and the reduction mechanism of dolomite during deep subduction processes from the Earth's surface to mantle transition zone.

  • Electron probe microanalysis (EPMA) was carried out on JEOL JXA-8100 EPMA device equipped with four wavelength dispersive spectrometer in EPMA Laboratory of School of Earth and Space Sciences, Peking University. The beam current and acceleration voltage were set at 5 nA and 15 kV (Zhang et al., 2019). The quantitative analysis of 11 routine oxides demonstrate that the chemical composition of natural dolomite is Ca(Mg0.964Fe0.028Mn0.008)(CO3)2. The natural dolomite was analyzed by powder XRD and the results show that the sample was a single dolomite phase. The iron powder sample purchased in Alfa Aesar was 99.9% purity. The mixtures of dolomite and iron powder (molar ratio of 1 : 1) were treated as initial materials. Equal proportions of iron in two experimental runs created the same oxygen fugacity (fO2) environment. Two experimental runs were conducted for comparison in the present study.

  • Symmetric-type DACs with diamond anvils of 300 or 400 μm in-diameter were employed in high pressure experiments. Holes of 150 and 200 μm in-diameter were drilled in two rhenium gaskets with a pre-indented thickness of ~38 and ~40 μm, respectively. The sample mixture was compressed into thin plates of ~70 μm in-diameter and ~25 μm in-thickness and then loaded into the chambers in a sandwiched configuration by two LiF plates with 7–8 μm in-thickness. The LiF plates could act as thermal insulator and pressure-transmitting medium. The LiF plates were also used as pressure indicator on the basis of the (111) and (200) typical peaks via the equation of state (EOS) (Liu et al., 2007). The pressure errors were from the fitting error for LiF in this work and the fitting error for referenced LiF EOS (Liu et al., 2007). The estimated pressure error was less than 1 GPa. A portable laser heating system with 1 064 nm laser radiation was employed to achieve high temperature (Dubrovinsky et al., 2009). The frequency and period were set at 50 kHz and 0.02 ms. The laser beam was focused to a spot with ~20 μm in-diameter. In prior to the XRD tests, the mixtures were firstly compressed to reach the pressure of interest at room temperature and then heated by laser scanning for ~0.5 h. The temperature was 1 300–1 500 K (±200 K).

    In-situ XRD measurements were performed at 4W2 beamline in Beijing Synchrotron Radiation Facility (BSRF). The monochromatic beam with a wave length of 0.619 9 Å was used with the beam spot size of 30×8 μm2. The patterns were recorded utilizing a charge coupled device (CCD) detector and the exposure time was 500 or 700 s. A CeO2 standard was employed to calibrate the detector-to-sample distance and the geometry parameters for radial integration of bidimensional data (Jephcoat et al., 1992). Unidimensional diffraction spectra were obtained by integrating in Fit2D software for the collected images (Hammersley et al., 1996). Le-Bail Method installed in GSAS+EXPGUI program was utilized to refine the lattice parameters of reaction products (Toby, 2001).

  • After pressure was released, in order to avoid the interference of diamond anvil, we took the recovered sample out of the DACs to perform ex-situ Raman experiments. The Raman measurements were carried out on Renishaw-1000 Laser-Raman spectrometer in Environmental Mineralogy Laboratory of Peking University, China. Diode-pumped solid-state laser with a wave length of 532 nm was selected as the excitation source. A Leica microscope equipped with 50× objective was utilized to focus the beam in a spot with ~1.5 μm in-diameter. A piece of monocrystalline silicon was employed to calibrate the spectra and the back scattered light was recorded in scattering geometry via a CCD detector with ±1 cm-1 resolution. The wave number range was 800–1 800 cm-1 and the exposure times for per frame were 180 s. Peakfit software was utilized to fit Raman spectra.

  • As mentioned above, two runs of experiments were conducted with dolomite and iron as starting materials in the present study. High-pressure diffraction data are obtained from ambient pressure to 20.2 GPa for run-1 as well as from ambient pressure to 15.8 GPa for run-2. In both two experiments, the peaks of sample mixtures at room pressure are fine assigned to the dolomite Ⅰ phase (R$ \overline{3} $) and body-centered cubic (bcc) phase of iron (Im$ \overline{3} $m). Above ~13 GPa, a new phase appears, which is indexed to a hexagonal-close-packed (hcp) phase of iron (P63/mmc)), in agreement with previous reports (Mao et al., 1967). In run-1, another transformation (dolomite Ⅰ to dolomite Ⅱ (P$ \overline{1} $)) is observed at ~19 GPa, as the [104] peak of dolomite Ⅰ splits (Santillán et al., 2003). There is no evidence of dolomite-iron interaction at room temperature, because the XRD results reveal only the existence of starting materials (Fig. 1).

    Figure 1.  Representative XRD patterns at 20.2 GPa (a) and 15.8 GPa (b) before heating and after quenching. The characteristic peaks are labeled by hkl. The crosses (×) show the unindexed peaks. The star (*) shows the initial appearance of dolomite Ⅱ. Abbreviations: dolomite Ⅰ and Ⅱ (D); hcp-iron (Fe); LiF (L); ferropericlase (Fp).

    Representative diffraction patterns before heating and after quenching for run-1 and run-2 are plotted in Fig. 1. As shown in Fig. 1, there are four intense peaks after quenching in two runs, while the peaks of iron and dolomite disappear. We refine the multiphase assemblages via the known crystal structures and EOS for identification of new diffraction lines. In addition to two peaks of cubic LiF, the remaining two diffraction lines with high intensity belong to a cubic periclase (MgO) phase. Since the volume of this phase is larger than that of pure MgO at equivalent pressure, this phase should be identified as a ferropericlase ((Mg, Fe)O) phase (Fig. 2). The lattice parameters refined by Le-Bail method of (Mg, Fe)O are listed in Table 1. Comparison of the previously proposed unit cell parameters on (Mg1-xFex)O (Fei et al., 2007) with the refined data of (Mg, Fe)O in the present study reveals that the composition of (Mg, Fe)O in run-1 is close to (Mg0.80Fe0.20)O (Fig. 2). Therefore, this oxide should be called ferropericlase because of the magnesium-rich content. In run-2, the Fe content of the (Mg, Fe)O solid solution is slightly lower than that of run-1, the composition is determined as (Mg0.85Fe0.15)O by interpolation method (Fig. 2).

    Figure 2.  The compositional determination of ferropericlase by interpolation method. The solid curves show the EOS of (Mg1-xFex)O (x=0, 0.2, 0.39, 0.58) from Fei et al. (2007) and van Westrenen et al. (2005). The solid circles show pressure based on the LiF pressure calibration from Liu et al. (2007) (the error bar of ±1 GPa).

    Run-1 Run-2
    Pressure (GPa) 20.2 (1) 15.8 (1)
    Temperature (K) 1 500 (200) 1 300 (200)
    Reactants Dolomite, iron Dolomite, iron
    Products (Mg0.80Fe0.20)O, graphite, hexagonal diamond, aragonite (Mg0.85Fe0.15)O, graphite, aragonite
    V3) of (Mg, Fe)O 68.49 (1) 69.65 (1)

    Table 1.  Experimental conditions, results and the lattice parameters of (Mg, Fe)O

  • It is widely accepted that Raman test is a powerful tool to distinguish different forms of carbon (e.g., carbonates, graphite and diamond), thus the recovered samples are further characterized by Raman measurements. Figure 3 shows the representative spectra of the recovered samples. Two obvious Raman bands at ~1 340 and ~1 580 cm-1 are assigned to disorder-induced mode (D band) and G band of graphite, respectively, consistent with previously reported computational and experimental results (Reich and Thomsen, 2004; Schindler and Vohra, 1995). The broad and intense D bands of the quenched samples agree well with the submicron graphite observed by Pócsik et al. (1998). The study also presents evidence for the hexagonal diamond formation on the basis of a characteristic Raman peak around 1 324 cm-1 in run-1. The hexagonal diamond (also known as lonsdaleite, a close-packed tetrahedral coordinated sp3 configuration) is one of the C polymorphs, which is considered as the intermediate phase from graphite to diamond (Smith and Godard, 2009). Besides, the peak at 1 084.7 cm-1 (ν1) attributes to the symmetrical stretching mode of [CO3]2- group in two runs. Previous studies have confirmed laser annealed dolomite decomposes into aragonite+ magnesite under conditions from the upper mantle to the topmost lower mantle (Merlini et al., 2012; Mao et al., 2011a; Sato and Katsura, 2001; Martinez et al., 1996). In view of the present studied pressure-temperature range and the wave numbers of ν1 for several carbonates in previous studies (Farsang et al., 2018), we conclude that this peak belongs to aragonite. Note that there are no peaks of aragonite in XRD patterns. It may be due to relatively low diffraction intensity of aragonite compared with cubic ferropericlase and LiF.

    Figure 3.  Selected Raman spectra of the recovered samples in run-1 ((a), (b) and (c)) and run-2 ((d), (e) and (f)). Abbreviations: A. Aragonite; D. disorder-induced mode for graphite; G. graphite; H. hexagonal diamond.

    Our results above show that ferropericlase, graphite, hexagonal diamond and aragonite are observed in run-1 products, while in run-2, the same products are identified except hexagonal diamond.

  • The reaction of dolomite reduction at Earth's mantle transition zone can be presented as

    The (Mg, Fe)O phase is known as an indicator for the formation of diamonds in the Earth's lower mantle. Ferropericlase with about 20% Fe is the second most abundant mineral in the lower mantle, thus it plays a significant role in our insights into the deep Earth (Mao et al., 2011b). Nevertheless, our results demonstrate that ferropericlase can be formed at mantle transition zone pressure via the reactions between carbonates and metallic iron rather than needing lower mantle conditions, consistent with previous experimental data (Thomson et al., 2016). The statistical results of the Mg/(Mg+Fe) ratio of natural wüstite-periclase solid solution vary greatly from 0.36 to 0.90 (Litvin et al., 2016), and our experimental results are included in this range. Considering the previous experimental studies on redox reactions in other carbonates-iron series, we can discuss the possible effects of fO2 on the magnesium content (Mg#) of (Mg, Fe)O. Palyanov et al. (2013) proposed that high fO2 contributed to Mg concentration in magnesiowüstite. In our study, the initial carbonates/iron molar ratio of 1 : 1, compared with a higher proportion of iron (1 : 2, 1 : 3 even saturation of iron) in previous studies, indicates that the fO2 in our experiments is relatively high, which may contribute to the formation of magnesium- rich ferropericlase.

    In our experiments, the pressure-temperature conditions are set in diamond thermodynamic stability field, but the reduced carbon is mainly in the form of graphite. Combined with previous experimental studies, we propose the following reasons. It is generally accepted that in the process of carbonates reduction, the carbon-host is firstly crystallized in the form of graphite and then transformed into diamond. Relatively low temperature and short heating duration in our experiments compared with the conditions from Palyanov et al. (2013) are unfavorable to the crystallization and growth of diamond. The intense and broad D band in Raman spectra and the relative intensity between G and D bands may be related to the grain size and crystallinity of graphite (Patterson et al., 2002). It can be concluded from the Raman spectra that the crystallinity of graphite is low. Studies have demonstrated that a kinetic barrier needs overcoming to initiate the transition from graphite to diamond (Sung, 2000). The higher the crystallinity of graphite is, the lower the kinetic barrier will be, and the pressure-temperature conditions required for the diamond formation can be more easily achieved. Therefore, the carbon in reaction products is mainly in the form of graphite rather than diamond. In addition to graphite, another form of elemental carbon, hexagonal diamond, can be identified in the Raman spectra. Hexagonal diamond is proposed as faulted and twined cubic diamond (Németh et al., 2014), which can be preserved after quenching (Oganov et al., 2013).

    In two experiments, no iron carbide is detected in the recovered samples, which can be interpreted by insufficient iron in the initial reactants. Based on the 'redox freezing' model (i.e., carbonate melts are reduced to immobile diamond or graphite), it will produce (Mg, Fe)O and diamond in the complete interaction between magnesite and Fe-metal, and the iron carbide is an intermediate product of the reaction (Rohrbach and Schmidt, 2011; Rohrbach et al., 2007). Therefore, in the MgCO3-iron system, the iron carbide cannot coexist with the carbonate, and it will be completely consumed when the carbonate is excessive (Martirosyan et al., 2015b).

    We demonstrate that dolomite-Fe redox reactions destabilize the MgCO3 component under pressure-temperature conditions of the mantle transition zone, while CaCO3 is preserved in the form of aragonite phase. Dorfman et al. (2018) revealed experimentally dolomite-Fe interactions under the deep lower mantle conditions. However, we believe that subducted dolomite will be thermodynamically unstable when subducting to the reducing mantle transition zone condition (Rohrbach and Schmidt, 2011; Rohrbach et al., 2007).

    We conclude that MgCO3 is more sensitive than CaCO3 to redox breakdown under the conditions studied. CaCO3 thus may be preserved within carbonate-rich sediments in subducted cold slabs, in agreement with the kinetically slow reaction mechanism reported previously under mantle transition zone conditions (Martirosyan et al., 2016). Calcium-bearing silicates are believed to be one of the most abundant minerals and the dominant calcium-hosts in the Earth, and CaCO3 is a candidate calcium source for the formation of CaPv reported in ultradeep diamond inclusions (Nestola et al., 2018). Thus, the residual CaCO3 in our experimental study may continue to interact with (Mg, Fe)SiO3 to produce CaSiO3 in the silicate mantle. The reactions provide an explanation for the observation that CaCO3 and CaSiO3 can be captured as inclusions in diamonds (Kaminsky et al., 2016, 2009; Brenker et al., 2007). Additionally, further mineralogical analysis of ultradeep diamond inclusions implies that the potential reactions between CaCO3 and SiO2 in subducting slabs can provide C-rich fluids contributing to diamond formation (Dasgupta and Hirschmann, 2006). The CO2 released by CaCO3-SiO2 reaction and the water possibly present in the mantle transition zone, will reduce the melting temperature and then lead to the formation of the C-H-O-bearing fluids or melts required for diamond growth (Li et al., 2018).

    In the natural geological settings, in addition to temperature and fO2, the factors that affect carbonate-iron redox reaction include sulfur fugacity, alkali metals, fluids, pressure, pH, etc. For instance, the addition of alkali metals will greatly lower the temperature of partial melting of carbonates, so that diamond can be formed effectively by the reactions between alkaline carbonates and reduced mantle domains (Thomson et al., 2016). The presence of fluids can promote diamond growth and help diamond capture syngenetic inclusions (Martirosyan et al., 2019b; Bureau et al., 2016). As a result, in the realistic and complex geological settings, the processes and products of slab- mantle interaction will be more complicated. Therefore, the mechanism of carbonates-iron reactions in more complex systems remains to be studied.

  • In the present study, we have conducted laser-heated DAC experiments to investigate the redox interaction in the dolomite-iron system under pressure-temperature conditions reaching those of the mantle transition zone. Our results show that the reaction proceeds with the formation of ferropericlase, its composition is determined as (Mg0.80Fe0.20)O in run-1 and (Mg0.85Fe0.15)O in run-2. Carbon in the form of graphite is observed in two runs, and another form of carbon, hexagonal diamond, is identified in run-1. Note that CaCO3 remains stable after quenching. The results indicate dolomite and metallic iron phases cannot coexist and demonstrate a possible formation mechanism of ultradeep diamond via redox reaction between dolomite and iron at those conditions. The presented reaction can be regarded as an intermediate process in subducted slabs in contact with reduced mantle domains. Our results are significant for understanding the change of carbon-hosts, carbon transportation and the reduction mechanism of dolomite at mantle transition zone conditions, and they have further implications for the processes in the more complex systems regarding to carbonate-silicate-metal phase relations.

Reference (57)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return