Carbonatitic zircon � myth or reality: mineralogical-geochemical analysis

 *Savva E.V, **Belyatsky B.V., ***Antonov A.V.

*Institute of Precambrian Geology and Geochronology RAS, St.-Petersburg, Russia

**VNIIOkeangeologia, St.-Petersburg, Russia

***Centre of Isotopic Research, VSEGEI, St.-Petersburg, Russia

Zircon crystallization in a wide range of temperature and physico-chemical conditions determines its wide distribution as an accessory mineral in magmatic and metamorphic rocks of different composition and origin and allows to use it as a reliable geochronometer for dating magmatic processes (Zircon, 2003). It is of common knowledge that in magmatic systems undersaturated in silica, such as alkali-ultramafic and carbonatite melts, with sufficient amount of zirconium, there takes place preferential crystallization of baddeleyite and other zirconium minerals (zirconolite, zirkelite, etc.), zircon being unstable under such conditions should be dissolved (Barker, 2001). Nevertheless, for many carbonatite�alkali-ultramafic massifs zircon is the prevailing accessory mineral occurring both in silicate and carbonatite rocks very often in the form of large, up to first centimeters, crystals of pseudo-octahedral habit (Kukharenko et al. 1965). Recent works on investigation of rare and trace elements distribution in zircon have demonstrated that geochemistry of zircon could be reliable criteria for determination of their host rocks origin (Hoskin, Ireland, 2000; Belousova et al., 2002). Moreover, zircon composition is widely used to define the provenance in sedimentary analysis of sands and sandstones on the base of polycomponent classification and regression analysis (CART), which allows to make wide palaeotectonic reconstructions (Veevers et al., 2006). One of the such reconstruction methods (Burke et al., 2003) is based on mapping of post-collisional zones of crustal extension marked by development of rocks belonging to alkaline complexes and carbonatites (ARC), which for their turn could be distinguished by composition of detrital zircons at area sampling (Veevers, 2007). Successfulness of this method appliance depends entirely on substantiation of classification characteristics � for typical representatives of carbonatite massifs in initial database were taken Mud Tank carbonatite complex (723 Ma) in Central Australia and Kovdor complex (380 Ma), Kola Peninsula. The following characteristics have been assumed as typical for �carbonatite� zircons: average REE content 600-700 ppm at comparatively flat distribution, (Yb/Sm)n: 3�30, Th/U up to 100 and more, 2.3<Lu<20.7 ppm, Ta>0.5 ppm, Hf>0.62 %. We have decided to test plausibility of these characteristics on zircons from carbonatite massifs of different age � from Archean Siilinj�rvi massif, Proterozoic Tiksheozero and Elet�ozero massifs, Paleozoic Lovozero and Khibiny massifs and Mesozoic carbonatitic dykes from Antarctica and India.

We have analyzed more than 100 zircon grains for content of 20 rare and trace elements. Zircon in the studied carbonatites is represented, as a rule, by well faceted large crystals and their particles. The crystals are cracked; the surface very often is rough. The shape of grains is pseudo-octahedral � dipyramidal with prevailing development of steep dipyramid (111) and subordinated development of prism (110), sometimes the zircons are short prismatic. Very often the facets are asymmetric. The color of zircon varies from colorless, pinkish-grey to strongly colored yellowish-brown shades. The crystals are semitransparent due to heterogeneity of its inner structure � abundance of cracks filled with carbonate material, gas-liquid inclusions and cavities, there numerous inclusions of rock-forming minerals such as mica, amphibole, carbonates, apatite, magnetite. The size of the studied zircon grains varies from 1-2 mm to 1x2 cm. In some cases we have marked corroded rim with fine dust-like baddeleyite which develops along the cracks (fig.1). Character of CL images varies even for zircons from the same complex � from absolute absence of any luminescence to bright colors over the whole grain without any regular inner structures. At the same time the back scatted electron images show the traces of zoning. So, morphology of the studied zircons repeats the known features of zircons from the other carbonatite massifs (Kramm et al., 1993; Сlaesson et al., 2000; Frantz et al., 2001; Amelin, Zaitsev, 2002; Chakhmouradian, Williams, 2004; Zozulya et al., 2007).

REE patterns of the studied zircons is characterized by extreme variability � from weakly fractionated flat chondrite-normalised REE distribution with general lower rare elements content to normal fractionated patterns with prominent enrichment in heavy REE and well pronounced negative Eu and positive Ce anomalies, but there are distributions with distinctly increased REE content with positive Eu and weak Ce anomalies (fig.2): (Sm/La)n: 2�738, (Lu/Gd)n: 1�58, Ce/Ce*: 1.2�194.1, Eu/Eu*: 0.7�1.78, (La/Gd)n: 0.0005�0.5, (Pr/Gd)n: 0.01�1.0, (Gd/Yb)n: 0.03�0.9. It is necessary to point out that according characteristic ratios and element contents suggested by (Belousova et al., 2002) for typically carbonatitic zircons in our case there are observed wide variations in (Yb/Sm)n: 0.9�180, Th/U: 0.2�512; [Hf]: 1020�8700 ppm; [Lu]: 1�94 ppm, [Nb]: 3� 2675 ppm; [Ta]: 4.7�68 ppm and only lesser part (about 45%) of the studied zircons could be convincingly classified as carbonatitic. At the same time, variations in zircon composition within the single complex could be compared with general variations and even in some grains there have been observed fluctuations in content of some elements more than one order of magnitude (for example, [Th] � from 4.7 to 270, [Lu] from 1 to 11, [Y] � from 15 to 430 ppm). Zircons from carbonatitic dykes (Antarctica, India) showed the least variations. Crystallization temperatures of the studied zircons by titanium-in geothermometer correspond to a rather narrow interval of 570��850�C. It is noteworthy, that elements which are greatly accumulated in carbonatites (Chakhmouradian, 2006) � REE, HFSE, P, Y, Sr, Ca, in the studied zircons vary widely and do not demonstrate essential enrichment in comparison with zircons from the other rock types.

Thus, independently from the origin of the certain carbonatite-alkali-ultramafic massive � from the primary common melt due to immiscibility of carbonate and silicate melts, in the result of consistent intrusion of independent melts from one developing mantle source or spatially close but essentially different in composition and time of formation sources, mineral composition of carbonatites and first of all mineralogical-geochemical peculiarities of such accessory minerals as zircon, baddeleyite, pyrochlore, perovskite etc. (Chakhmouradian, Williams, 2004), are determined by interaction of carbonate fluid enriched in lithophile elements with carbonate and silicate minerals of the matrix. Evolution of such fluid composition causes considerable changes in mineral geochemistry in the process of their growth, changing of mineral forms and partial replacement of some minerals by the others (for example, the relationship between zircon and baddeleyite), partial dissolving and further recrystallization. Especially well such processes are revealed in old complexes where the later rather low-temperature processes are overprinted on the earlier formed mineral assemblages in the course of tectonic activity in the region. In such cases there could be observed poly-stage mineral systems (for example, zircon) formation of which could take place with considerable time intervals (till dozens and hundreds of years) but its geochemistry will reflect the main evolution stages of massif transformation. In any case, distinguishing the single geochemical type of �carbonatitic zircon� in reflection to genetic link with the host carbonatites seems rather ill-founded, first of all due to the limited data on such unique fluctuation in composition of these zircons.

References:

Amelin Yu., Zaitsev A.N. Precise geochronology of phoscorites and carbonatites: the critical role of U-series disequilibrium in age interpretations. Geochim.Cosmochim.Acta, 2002, 66, 2399-2419.

Barker D.S. Calculated silica activities in carbonatite liquids. Contrib.Mineral.Petrol., 2001, 141, 704-709.

Belousova E.A., Griffin W.L., O�Reilly S.Y., Fisher N.I. Igneous zircon: trace element composition asan indicator of source rock type. Contrib.Mineral.Petrol., 2002, 143, 602-622.

Burke K., Ashwal L.D., Webb S.J. New way to map old sutures using deformed ARCs. Geology, 2003, 31, 391-394.

Chakhmouradian A.R. High-field-strength elements in carbonatitic rocks: geochemistry, crystal chemistry and significance for constraining the sources of carbonatites. Chemical Geology, 2006, 235, 138-160.

Chakhmouradian A.R., Williams T.C. Mineralogy of high-field-strength elements (Ti, Nb, Zr, Ta, Hf) in phoscoritic and carbonatitic rocks of the Kola Peninsula, Russia. In: Wall F., Zaitsev A.N. (eds) Phoscorites and carbonatites from mantle to mine: the key example of the Kola Alkaline Province. Min. Soc. London, 2004, 293-340.

Frants N.А., Savva E.V., Putintseva E.V. Rare-metal minerals (zircon, pirochlore, baddeleyite) from carbonatites of Tiksheozero massif (N. Karelia). Vestnik SPGU, 2001, 4 (31), 76�83.

Hoskin P.W.O., Ireland T.R. Rare earth element chemistry of zircon and its use as a provenance indicator. Geology, 2000, 28, 627-630.

Kramm U., Chernyshev I.V., Grauert B., Kononova V.A., Breker B. Typology and U-Pb zircon systematics from nepheline seyenite from Ilmeny, Ural. Petrology, 1993, 1, 536�549.

Kukharenko, A.A. et al., 1965. The Caledonian Complex of Ultrabasic, Alkaline Rocks and Carbonatites of the Kola Peninsula and northern Karelia. Nedra, Moscow. 550 p. (in Russian).

Veevers J.J. Pan-Gondwanaland post-collisional extension marked by 650-500 Ma alkaline rocks and carbonatites and related detrital zircons: a review. Earth-Science Reviews, 2007, 83, 1-47.

Veevers J.J., Belousova E.A., Saeed A., Sircombe K., Cooper A.F., Read .E. Pan-Gondwanaland detrital zircons from Australia analysed for Hf-isotopes and trace elements reflect an ice-covered Antarctic provenance of 700-500 Ma age, Tdm of 2.0-1.0 Ga, and alkaline affinity. Earth-Science Reviews, 2006, 76, 135-174.