The behaviour of Mg and Ca in some types of ultramaphic alkaline melts in the presence of H2O at high temperature and pressure (the experimental data)
V. K. MARKOV1), V.V. NASEDKIN2), N. M. BOEVA2)
1 – Institute of Geosphere Dynamics of Russian Academy of Sciences, Moscow ;
2 - Institute of Geology of Ore deposits, Petrography, Mineralogy and Geochemistry of Russian Academy of Sciences, Moscow
The sample has been investigated, which was presented by olivine-melilite rock from the south-western part of Kugda intrusion, situated in the eastern margin of Maimecha-Kotuiskay alkaline- ultramaphic province (northern part of Siberia platform). The experiments were carried out using high-pressure apparatus of “piston-cylinder” type. The pressure was 5-15 kb and the temperature – 800-12500 C. The water content in the system was 3 and 9 wt.All experiments were carried out in hermetically sealed Pt or Au ampoules. The samples were chemically analyzed using scanning electron microscope (SEM). Besides, the chemical composition of experimental phases was determined by means of more than 40 electron microprobe in 20 experimental samples. F or the phase identification t he X-ray method was used too. The structural peculiarities of rocks and mineral composition were studied by optical methods in polished and thin sections prior and after the experiment . It was shown that at the pressure above 7,5-8 kb the olivine-melilite melt is separated into two melts: pyroxenite and melilite ones. Most part of Mg in this system was concentrated in high temperature phase of diopside type. The remaining melt was depleted of Mg and enriched by Ca. The mineralization vapour practically did not contain Mg. The main components of mineralization vapour were Ca, Na, Si and Fe.
Key words: apparatus of high pressure and temperature, magnesium, calcium, ultramaphic alkaline rocks, diopside, mervinite, liquation.Introduction
Many authors (Smolin 1991 & oth.) reported a special role of Mg in the formation of the upper Earth’s shell ( earth crust, upper mantle). The main geochemical peculiarities of magnesium are the following: Mg has high clark and, correspondingly, the status of rock-forming cation of high chemical activity and with marked halo-oxi-hydrophility. The highest refractory of Mg silicates is also known as compared with other silicates of the rock-forming cations. According to these properties, Mg is concentrated in subcrustal zone in the mantle silicate restite. The system “melt + Mg restite + fluid” can be formed by partial melting of mantle rocks. The most of the authors, who studied mineral equilibriums characterized for P, T conditions of the upper mantle, suggest that H2O content there does not exceed 0,1-0,3% (Mysen & Boettcher 1979, Ringwood 1981). The main phase of the mantle fluid is CO2. The formation of the high-magnesium fluid is not possible in these conditions. This conclusion is supported by the high content of calcium and the low one of magnesium in oceanic sediments near abyssal faults. In order to understand the behavior of Mg in system “melt + crystal + H2O vapour” we decided to study the evolution of Mg melt characteristic for the deep-seated levels of the Earth shell ( 15 – 45 km). It is important to understand, is it possible or not the formation of deep-seated solutions enriched magnesium in this conditions. The olivine-melilite rock was taken for the experiment. As may be seen from the foregoing, the present experiments have been carried out to determine the Mg distribution between melt, solid phase and water vapour at the pressure range of 5-15 kb and the temperature range of 600-12000C with 3 and 9% of H2O.The description of sample
Olivine-melilite rock (kugdite) occurs at the eastern margin of Maimecha-Kotuiskay alkaline- ultramaphic province ( northern part of Siberia platform). The eighteen alkaline-ultramaphic intrusions are present here.One of them, the Kugda intrusion occupies the area near 16 km2. This one consists of the central olivinite core and the north-eastern and south-western margin zones. The southwestern zone is presented by olivine-melilite rock or “kugdite”.
The sample used in the study was presented by kugdite having the following mineral composition (vol.%) : melilite -58, olivine -32, ore mineral - 5 , the remaining were single grains of pyroxene and sphene. (Fig. 1).
Olivine ( Fo 88Fa12 ) was presented by rounded grains up to 0,1-0,3 mm in diameter. Melilite ( Ok 70 Na melilite 25 Gel5 ) formed isometric grains of 0,05-1,0 mm in size. The ore minerals were presented by magnetite and ilmenite intergrowth. The kugdite chemical composition (wt.%) was the following: SiO2 - 40,96; TiO2 – 0,63; Al 2O3- 1,82; FeO – 3,78; CaO – 30,30; MgO- 16,77; MnO-0,00 BaO- 0,43; Na2O – 1,70; K2O – 0,08; P2 O 5 – 0,06; H2 O+ 0,06; H2 O – 0,00; CO2 –0,05, in total – 100,23.The apparatus and experimental methods
The experiments were carried out using high-pressure apparatus of “piston-cylinder” type. The principal scheme is shown on the Fig. 2.
The high pressure camera was made of hard alloy and had a form of a tapered cylinder approximately 40 mm long, 80 mm in diameter and with a bore of 20 mm. The high pressure was created by a hydraulic press. The pyrophyllite was used as a pressure medium. The high pressure apparatus was calibrated according to the phase transformation of KCl-RbCl solid solutions (Livshitz &Larionov, 1970). The accuracy of pressure measurement was 5%.
The working volume of the camera was heated using the graphite heater placed inside the pressure vessel. The furnace was insulated by pyrophyllite from the run and vessel.
The temperatures were usually measured with chromel-alumel thermocouples. The thermocouple was placed at the center of the furnace through two holes drilled along of the axis of the cylindrical heater. The thermocouple junction was placed in contact with a run capsule. The maximum gradient observed did not exceed 2oC. The temperature uncertainly was 5oC.
The experiments were carried out at the temperature range from 600 to 1200 0 C. There were two variants of experiment: an approach to equilibrium “from above” and “from below”. In first case, the temperature was increased up to 12000 C, followed by a two hours exposure and after that, the sharp decrease of the temperature up to predetermined value was done with an exposure in 5 hours. In the second case, the temperature was increased to predetermined value with the followed exposure in 7-8 hours. About 40 mg of. finely crushed kugdite and an appropriate amount of distilled water were encapsulated in a welded, thin-wall platinum tubing. The water contents were selected 3 and 9 %(wt).
The capsule content was controlled by heating at 120oC during 2 hours. The capsules were weighted before and after heating, and only that capsules were used in the experiments, which weight did not change after that procedure. To insure that water did not leak out during the runs, capsules were also weighted before and after run. The capsules were made of two kinds of material: platinum and aurum.
There was some loss of iron from the sample into the Pt container during a run, but this has been monitored by microprobe analyses of samples after the runs.
We showed that the loss of iron for the experimental procedures was usually less than 30-35% of the total iron presented and there was only a small change in the oxidation state.
The run products were picked out carefully and examinated under optical microscope.Experimental results
The results of kugdite experimental study with 3 and 9% H2O are presented in Fig 3, a-b. The system with 3% H2O. The heavy lines (1) and (11) mark the kugdite liquidus and solidus, correspondingly. The most striking features of the melting curves in figure 3 are the changes in slope occurring where they cross the line of the transition “melilite mervinite” (111). This line has the extension to the field of liquid state. Above this line, all runs are characterized by the presence of two glasses of different composition as the phases of quenching, below this line only one glass is present.
All runs below the equilibrium line of “melilite mervinite” were characterized by Ca melilite and olivine presence. The subsolidus, run at < 7,5 kb, led to the yield of original mineral assemblage of the kugdite ( Ol+Ak+Fl)+spinel (Sp).
At the pressure higher than 7,5-8,5 kbar, melilite was disappeared, mervinite and diopside were the main crystalline phases. As it was explained above ,there were two variants of running the experiment, if the pressure was constant. The first variant: an approximation to equilibrium from high to low temperatures and the second variant from low to high temperatures. In the first case, we observed the separation of sample into two parts : the lower and the upper. In runs with 3% H2O, the low part of the sample consisted of the glass and mervinite crystals, the upper from diopside crystals. The boundary between upper and lower parts was harp. The subsolidus run at 7,5 kb and higher showed the disappearance of spinel and appearance of garnet. The chemical composition of the phases synthesized in 3% H2O system is shown in Table 1. The pyroxene with higher Mg content was crystallized at higher pressure.
The system with 9% H2O. The increasing of water content in the system led to the decreasing of temperature range for crystallization. The synthesized crystals as a rule had more regular form. The fluid phase as drop-liked bodies were presented in the upper part of capsules practically in all runs. The phases obtained at high pressures and temperatures are listed in Table 2. In runs with 9% H2O, upper part of sample consists of mervinite melt, below one from diopside melt.Conclusion
Scanning electron microscope was used to determine the average chemical composition of different zones in experimental samples including fluid phase. Scanning was produced along 3-4 profiles by crossing the sample in transverse directions ( Table 3 ).
As the experiments at high pressure and temperature were shown, there were two kind of mineral differentiation. At the pressure above 7 kb, ultramaphic alkaline melt was separated into 2 liquids: pyroxenite liquid and mervinite one. At the lower pressure, usual crystal differentiation was present. In both cases, Mg had a tendency to enter into highly temperature phases: diopside, olivine. The remaining part of the system became enriched with Ca. The fluid phase contained Mg in small proportion, rather as traces. The preliminary analysis of experimental data leads to the conclusion that deep-seated subcrustal ultramaphic magmas can not produce large volumes of Mg solutions, which could be the source of magnesite occurrences.
Kononova V.A. 1976. Jacupirangite-urtite series of alkaline rocks.Nauka, Moscow . 214
Livschiz L.D., Larionov L.V. 1970. The determination pressure drop in high pressure camera . Izvestija of USSR Academia Nauk. Physics of Earth. 2, 75-78.
Markov V.K., Nasedkin V.V. & Riabinin Yu.N. 1973. Liquation in ultramaphic –alkaline melt. Izvestija of USSR Academia Nauk. Physics of Earth. 1, 3-8.
Markov V.K., Nasedkin V.V.& Borisovsky S.E. Phisico-chemical conditions of mica crystallization in ultramaphic –alkaline melts In : V.P. Petrov (Ed); High-Mg raw materials”,Nauka. Moscow . 259-261
Mysen B.O., Boettcher A.L. 1979 Melting of a hydrous mantle. Mir. Moscow . 110
Ringwood A.E., 1981. Composition and petrology of the Earth’s mantle. Nedra, Moscow , 583
Smolin P.P. ,1983. Mineragenetic history, problems of evalution of resources and rational application of magnesite, brucite and talc . In V.P. Petrov (Ed.); High-Mg raw materials, Nauka. Moscow . 11-60.
Yoder H.S.,Jr. 1973. Melilite stability and paragenesis. Geophysical labarotary Carnegie Inst. Of Washington, 1622, 140-172