Calcite and dolomite in intrusive carbonatites. II. Trace-element variations

Abstract

The composition of calcite and dolomite from several carbonatite complexes (including a large set of petrographically diverse samples from the Aley complex in Canada) was studied by electron-microprobe analysis and laserablation inductively-coupled-plasma mass-spectrometry to identify the extent of substitution of rare-earth and other trace elements in these minerals and the effects of different igneous and postmagmatic processes on their composition. Analysis of the newly acquired and published data shows that the contents of rare-earth elements (REE) and certain REE ratios in magmatic calcite and dolomite are controlled by crystal fractionation of fluorapatite, monazite and, possibly, other minerals. Enrichment in REE observed in some samples (up tõ 2000 ppm in calcite) cannot be accounted for by coupled substitutions involving Na, P or As. At Aley, the REE abundances and chondrite-normalized (La/Yb) cn ratios in carbonates decrease with progressive fractionation. Sequestration of heavy REE from carbonatitic magma by calcic garnet may be responsible for a steeply sloping Bexponential^pattern and lowered Ce/Ce* ratios of calcite from Magnet Cove (USA) and other localities. Alternatively, the low levels of Ce and Mn in these samples could result from preferential removal of these elements by Ce 4+ -and Mn 3+ -bearing minerals (such as cerianite and spinels) at increasing f(O 2 ) in the magma. The distribution of large-ion lithophile elements (LILE = Sr, Ba and Pb) in rock-forming carbonates also shows trends indicative of crystal fractionation effects (e.g., concomitant depletion in Ba + Pb at Aley, or Sr + Ba at Kerimasi), although the phases responsible for these variations cannot be identified unambiguously at present. Overall, element ratios sensitive to the redox state of the magma and its complexing characteristics (Eu/Eu*, Ce/Ce* and Y/Ho) are least variable and in both primary calcite and dolomite, approach the average chondritic values. In consanguineous rocks, calcite invariably has higher REE and LILE levels than dolomite. Hydrothermal reworking of carbonatites does not produce a unique geochemical fingerprint, leading instead to a variety of evolutionary trends that range from light-REE and LILE enrichment (Turiy Mys, Russia) to heavy-REE enrichment and LILE depletion (Bear Lodge, USA). These differences clearly attest to variations in the chemistry of carbonatitic fluids and, consequently, their ability to mobilize specific trace elements from earlier-crystallized minerals. An important telltale indicator of hydrothermal reworking is deviation from the primary, chondrite-like REE ratios (in particular, Y/Ho and Eu/Eu*), accompanied by a variety of other compositional changes depending on the redox state of the fluid (e.g., depletion of carbonates in Mn owing to its oxidation and sequestration by secondary oxides). The effect of supergene processes was studied on a single sample from Bear Lodge, which shows extreme depletion in Mn and Ce (both due to oxidation), coupled with enrichment in Pb and U, po...

Calcite and dolomite in intrusive carbonatites. I. Textural variations

“…An example of discrimination diagram that can be used to distinguish texturally similar carbonatites and metasedimentary rocks from collision zones (data from Demény et al 2004;Chakhmouradian et al 2008;Moore et al 2015;Xu et al 2015;Chakhmouradian et al 2015b). Note the similarity of deformation textures in calcite carbonatite from Eden Lake (Canada) and marble from Miaoya (China), shown in the insets BHybrid^carbonate-silicate magmas may evolve to produce Ca-rich liquids (up to 80 wt.% CaCO 3 ) by immiscibility or crystal fractionation, but in either case, will precipitate silicate minerals before reaching the carbonate liquidus (Lee and Wyllie 1998a, p. 500).…”

“…We consider both these interpretations problematic because (1) the distribution of burbankite and other chadacrysts in the host crystal is not controlled crystallographically, nor confined to cleavage planes or fractures (see the next section); (2) strontianite (SrCO 3 ), which is the most common product of calcite breakdown, does not occur in these parageneses; (3) analogous inclusions are found in minerals, which cannot possibly exsolve carbonates (e.g., Fig. 4e, f); (4) bulk analyses of dolomite grains containing burbankite inclusions give up to several thousand ppm Na and REE, which cannot be realistically a cc om m o da t e d i n t h e st r u c t u r e o f t h i s m i n e r a l (Chakhmouradian et al 2015b); and (5) subsolidus relations in the system CaCO 3 -MgCO 3 (see Phase and compositional relations of significance to rock-forming carbonates in carbonatites) suggest that exsolution of dolomite from calcite intruded by calcite carbonatite, Carb Lake; primary dolomite (Dol1) contains 7.5-11.2 wt.% FeO, 0.7-1.9 wt.% MnO, 0.1-0.3 wt.% SrO, and is associated with potassic-fluoro-magnesio-arfvedsonite (Amp1); at the contact with calcite, Dol1 is mantled by Fe-Mn-Sr-poor dolomite Dol2 (2.3-6.6, 0.3-0.8, 0-0.1 wt.% respective oxides), and Amp1 by fluororichterite (Amp2). f Magnetite-phlogopite-calcite carbonatite intersected by a veinlet of richterite (Rct)-dolomite carbonatite, Aley; both calcite and dolomite contain high levels of Sr, Ba, REE and are interpreted as magmatic (note also the sharp contacts between the two rocks and a flow pattern in the veinlet) should be much more common than vice versa, whereas calcite with ovoid inclusions of dolomite is not as abundant as poikilitic dolomite.…”

“…10b-d), co-existence of fresh and heavily altered mineral grains on a fine spatial scale, and changes in the isotopic composition (C, O and Sr) of carbonates (Kapustin 1987;Schürmann et al 1997;Downes et al 2012;Chakhmouradian et al 2015a). Metasomatic replacement of primary dolomite is probably just as common, but is much more difficult to trace (Chakhmouradian et al 2015b). At advanced stages of alteration, dolomite replaces not only the rock-forming carbonates, but also ferromagnesian silicates (olivine, amphiboles, phlogopite) and magnetite, commonly in intimate association with other secondary minerals (chlorite, serpentine, quartz, goethite, rutile and Mg-Fe carbonates; Fig.…”

“…10e-g). Because dolomite has a smaller capacity for Sr, Ba and REE than its precursor calcite (Chakhmouradian et al 2015b), these elements are often deposited as strontianite, barite, ancylite, REE fluorocarbonates and monazite confined to fractures and grain boundaries. The extent and potential significance of this secondary rare-metal mineralization in carbonatites strongly affected by dolomitization (e.g., Aley in Canada) are yet to be ascertained.…”

“…Most dolomite analyses show≤ 0.6 wt.% SrO at Ca and Fe levels within the experimentally determined range, as discussed above. Despite the existence of synthetic Ca-Sr dolomites (Brice and Chang 1973), the Sr content of carbonatitic dolomite is significantly lower than in cogenetic calcite (Dawson and Hinton 2003;Zaitsev et al 2014;Chakhmouradian et al 2015b). Because of the obvious charge constraints, Na and REE can be incorporated in calcite and dolomite to much lower levels than divalent cations; the mechanisms of Na and REE uptake by these minerals are uncertain (for discussion, see Chakhmouradian et al 2015b).…”