Combined in-situ determination of halogen (F, Cl) content in igneous and detrital apatite by SEM-EDS and LA-Q-ICPMS: A potential new provenance tool

Apatite is an accessory mineral that is widely present in most igneous and metamorphic rocks, and is a common detrital component in clastic rocks. As apatite typically incorporates U at sufficient concentrations to enable thermochronometry (e.g., U-Pb, fission track, and (U-Th)/He), it is commonly used in detrital provenance studies. Apatite also incorporates a wide range of trace elements at concentrations readily detectable by routine laser ablation-quadrupole-inductively coupled plasma-mass spectrometry (LA-Q-ICPMS) analysis, which can also yield diagnostic provenance information (in particular Sr-Y-REE). However, detrital thermochronometry is of limited use in the case of source terrane(s) with similar crystallisation or cooling histories, while trace elements may yield non-diagnostic provenance information when their source rocks are lithologically similar (e.g., cogenetic granitoids). Here we test a complementary provenance approach which exploits F and Cl substitution into the stoichiometric Z-site in the apatite lattice. Halogen abundance in igneous apatite typically reflects the source melt composition and its subsequent differentiation history, offering potential for provenance discrimination. We apply this method to apatite from bedrock samples of five late- to post-tectonic granitoids in the Grampian and Moine terranes of the Scottish Caledonides, which yield zircon U-Pb ages of ca. 425-398 Ma. Emplacement over this relatively short time window makes it difficult to discriminate between sediment sourced from these plutons using radiometric dating alone. Apatite F content was measured using an energy dispersive X-ray spectrometer coupled to a scanning electron microscope (SEM-EDS), while Cl and trace element (Sr-Y-REE) abundances were determined by LA-Q-ICPMS. We show that the sampled granitoids can be divided into three groups based on their apatite trace element contents and into three separate groupings based on their F and Cl abundances; all five plutons can thus be discriminated from each other using their integrated trace elements and F-Cl systematics. The granitoid groups identified by their apatite F and Cl contents correlate with those depicted by their modal mineralogy as defined in Quartz-Alkali Feldspar-Plagioclase (QAP) space, with granitoids richer in K-Feldspar (and thus more evolved) yielding apatites with higher F contents. As a pilot study, we also analysed apatite from modern river sediment in the Spey River catchment of the Grampian terrane, for which U-Pb age data were also acquired. The Spey detrital apatites yield U-Pb ages that are indistinguishable at the 2 sigma-level, but their trace element abundances allow three bedrock types to be distinguished: I-type granitoids, S-type granitoids and medium- to high-grade metamorphic rocks. The detrital apatite F and Cl data do not permit such straightforward discrimination, as apatites from S-type granitoid and metamorphic rocks have comparably low Cl contents. However, the detrital apatites with I-type granitoid affinity display significant dispersion in halogen contents (yielding the lowest F abundances with correspondingly the highest Cl contents), which identifies the presence of several more "mafic" I-type granitoid bodies in the Spey catchment. Apatite F and Cl abundances thus have potential for detailed characterisation of the individual components of granitoid suites, including delineating I-type granitoids or the presence of less evolved magmatic bodies among detrital apatites.