<<

of Strontium in Layered Intrusions: Inconsistency with Predictions of Orthomagmatic Model

A.A.Efimov Institute of Geology and Geochemistry, Ekaterinburg, Russia e-mail: [email protected]

It is assumed that a process of with average SrPl near to 350 ppm. However chamber fractionation is recorded in layered strontium levels corresponding to main strati- intrusions. Primary homogeneous liquid must graphic units are distinguished in the section. In possess some resource of Sr and the main host each of them Sr whole-rock concentrations are , , must extract Sr from it in approximated by linear function of amount of accordance with ctystal/liquid partition coefficient plagioclase with constant SrPl. The doubt arises if which is always much more than 1.0. It must stratigraphic units of the massif have been formed inevitably make residual liquids, late derivates and by fractionating of magma. This doubt is supported plagioclase progressively poorer in Sr. Cryptic by data from some other layered complexes. variation (cryptic layeing) must reflect this In the more than 6000 m of visible width expected tendency. section of the Freetown Layered Complex (Sierra As established for Uralian complexes of Leone) composition of main mineral phases varies ophiolitic and "zoned" type, whole-rock Sr at a very narrow rate. In all the section correlation concentrations within natural series, usually plagioclase amount vs whole rock Sr content is layered or taxitic, have linear correlation with close to ideal linear. All the rocks are as if they are amount of plagioclase which has hereat constant Sr a mixture of femic practically containing concentation (SrPl) not depending on amount of no Sr and plagioclase with constant SrPl (650 ppm). plagioclase (Efimov, 1989). Hence Sr concentration Cryptic variation is absent (Umeji, 1983). W.P. in this mineral phase was determined not by Sr Meurer and A.E. Boudreau (1996) studied the concentation in given volume but by level of Middle Banded Series of the Stillwater Complex chemical potential of Sr which was common for the (Montana, USA) in very detail. In spite of great series. Consequently, layered were variations in mineral proportions there are no equipotential, open Sr-systems; their parts (layers, distinct stratigraphic or lateral trends of mineral domains) were in thermodynamic communication composition in the section. Cryptic variation is also each with other. Sr was redistributed in proportion absent. Whole rock Sr concentrations are strictly of plagioclase amount in each given point. correlated with plagioclase amount; almost ideal Unfortunately systematic stidies on Sr geo- linear trend demonstrates constancy of SrPl (about chemistry for layered intrusions such that by 200 ppm). Authors consider these facts as the result A.Morse (1982) are very rare. of significant modification of rocks by Ioko-Dovyren layered complex (North Near- postcumulative processes which obscured their Baikal region, East Siberia) forms a subvertical primary genetic features. The special study by lens-like body 28 km in lenght and up to 3.5 km in A.Morse (1982) is dedicated to geochemistry of Sr width. It is assumed as a product of fractional in Kiglapait Complex (Labrador, Canada). The PCS crystallization of picrite-like magma in primarily (per cent solidified) vs Sr plot demonstrates the subhorizontal . Whole-rock Sr absence of any distinct trend of Sr in the section. content increases upsection from 30-40 to 300 ppm Average whole rock Sr content is 330 ppm; values and in general correlates with plagioclase amount. for the Lower Zone vary ±100 ppm and for the Sr content in lower -"plagiodunite" zone is Upper Zone ±150 ppm around this figure. Lower very low. The next mainly troctolitic zone (SrPl Zone plagioclase contains 532 ppm; SrPl increases about 250 ppm) is connected by a sharp 200-m to 680 ppm in the Upper Zone before decreasing to gradient with overlying zone composed mainly of 310 ppm at the top. There is a very distinct linear olivine gabbro (SrPl about 350 ppm). Values of SrPl trend (r = 0.93) in the lower part (90% of the up to 500 ppm are fixed in zone of "upper gabbros" section) which composes mainly of ; it which is not well expressed. There are no demonstrates here constancy of SrPl (about 450 indications of upsection decreasing whole-rock Sr ppm) at any plagioclase amount. In the Kivakka content and SrPl. In rough approximation all the Complex (North Karelia, Russia; Koptev- gabbroic series forms single linear (r > 0.9) trend

Dvornikov et al., 2001) whole rock Sr compensate viscous gaps of layered mass by concentrations are in a satisfactory correlation with diffusion along pressure (chemical potential) plagioclase amount and vary in gabbroic part of the gradient in order to diminish a contrast between section mainly from 250 to 370 ppm at approximate host and gaps. (4) Experimental data of special average level about 300 ppm. They do not detect importance for An-Fo system (Kushiro, Yoder, any stratigraphic direction or upsection decreasing. 1966) precludes orthomagmatic origin of : The average trend is close to linear; one can assume troctolitic liquid produces aluminous (but the average SrPl to be about 400 ppm. never anorthite-olivine association) which never Thus for all these examples Sr distribution occurs in natural sequences as well as spinel does not subordinate to predictions of ortho- cumulates. These data are usually simply ignored magmatic model. Consequently some other process (Efimov, 1985). (5) At last, the marine geo- is recorded in considered sections. physicists community came to the conclusion that The 1st important conclusion is: there are no there are no great magma chambers postulated by indications to the tendency of decreasing Sr petrologists in modern spreading zones such as upsection predicted by orthomagmatic crystal- mid-ocean ridges (Kent et al., 1993). This fact is in lization of a magma basin. The 2nd important a sharp contradiction with cumulative interp- conclusion is: at least in a part of layered retations of 3rd layer of modern oceans and complexes in all the section or in some significant gabbroic member of the ophiolite suite. stratigraphic units of the section whole-rock Thus Sr distribution data together other concentration of Sr is a linear function of amount of independent evidence lead to a conclusion that plagioclase containing constant concentration of Sr. orthomagmatic evolution of at least some of layered This most important fact can not be plausibly intrusions seems to be doubtful. However this interpreted by fractionating of a magmatic liquid. conclusion does not answer question about nature However the same fact can have a fully correct of primary substance of these geologic objects. We thermodynamic interpretations: constant concen- can not discuss here this global problem. A tration of a component in a phase which does not hypothesis which connects ophiolitic gabbroic depend on phase amount symbolizes equality of rocks with decompression of high pressure basic chemical potential of this component at the moment rocks of the upper (Efimov, 1984) seems to of generation of all the sequence. Hence the be rather consequent and acceptable for this case. following conclusion: all the great number of layers was in state of chemical equilibrium and composed References a single whole, a some equipotential system (a Efimov A.A., 1985. Gabbro-ultramafic complexes reservoir, a kind of thermo- and barostate) where all of the Urals and the ophiolite problem. the layers were formed simultaneously but not one Moskva, Nauka, 232 p. (In Russian) after another as predicted by the classical Efimov A.A., 1985. Nature of troctolite. Doklady interpretation. Akademii Nauk of the USSR. Vol. 281, p. There are another independent evidence 1419-1423. (In Russian) that confirm these conclusions. Efimov A.A., 1989. Metamorphic differentiation: A (1) Phenomena of state (plastic) flow mechanism of segregations in gabbroic rocks which makes dynamo- formation in gabbro of the Ioko-Dovyren metamorphic nature of layering and structural massif (North Near-Baikal region). pattern of massifs very likely. (2) Precise micro- Geokhimiya. No. 3, p. 1042-1046. (In probe studies do not confirm existence of two Russian) generation of crystals, cumulative and inter- Efimov A.A., Efimova L.P., Mayegov V.I., 1989. cumulative. The degree of idiomirphism can not be Strontium in plagioclase of Uralian gabbros: convincing. (3) Phenomena of metamorphic Petrogenetic and applied aspects. differentiation. The petrology forbids existence of Geokhimiya, No. 11, p. 1541-1553. (In plagioclase liquids but in Ioko-Dovyren Complex Russian) extension structures of "horsetail" type are filled by Kent G.M., Harding A.J., Orcutt J.A., 1993. anorthosite consisting of the same plagioclase as in Distribution of magma beneath the East the host troctolite (Efimov, 1989). In Stillwater Pacific Rise between the Clipperton Complex, discordant bodies of anorthosite and Transform and the 9o17'N Deval from leucocratic troctolite are interpreted as result of forward modeling of common depth point replacement (Meurer et al., 1997) but indication of data. J. Geophhys. Res. Vol.98. B, p. 13945- the last process are not very convincing. There are 13969. reasons to interpret them as fill segregations that Koptev-Dvornikov E.V., Kireev B.S. et al., 2001. J.Petrology. Vol. 37, p. 583-607. Distribution of kumulative parageneses, Meurer W.P., Klaber S., Boudreau A.E., 1997. rock-forming and unimportant minerals in Discordant bodies from olivine-bearing the vertical section of the Kivakka intrusion zones III and IV of the Stillwater Complex, (Olanga group of intrusions, North Karelia). Montana – evidence for postcumulus fluid Petrologiya. Vol. 9, p. 3-27. (In Russian) migration and reaction in layered intrusions. Kushiro I., Yoder H.S.,1966. Anorthite- Contrib. Mineral. Petrol. Vol. 130, p. 81-92. and anorthite- reactions and their Morse S.A., 1982. Kiglapait geochemistry V: bearing on the -eclogite Strontium. Geochim. et Cosmochim. Acta. transformation. J. Petrology. Vol.7, p. 337- Vol. 46, p. 223-234. 362. Umeji A.C.1983. Geochemistry and of Meurer W.P., Boudreau A.E., 1996. Petrology and the Freetown layered basic igneous complex mineral compositions of the Middle Banded of Sierra Leone. Chem. Geology. Vol. 39, p. series of the Stillwater Complex, Montana. 17-38.