Synthesis of trace element bearing single crystals of Chlor-Apatite (Ca5(PO4)3Cl) using the flux growth method

We present a new strategy on how to synthesize trace-element bearing (REE, Sr) chlorapatites Ca5(PO4)3Cl using the flux growth method. Synthetic apatites were up to several mm long, light blue in colour. The apatites were characterized using XRD, electron microprobe and laser ablation ICP-MS (LA-ICPMS) techniques and contained several hundred μg/g La, Ce, Pr, Sm, Gd and Lu and about 1700 μg/g Sr. The analyses indicate that apatites were homogenous (within the uncertainties) for major and trace elements.


Introduction
Apatite (Ca 5 (PO 4 ) 3 (Cl,F,OH) is an ubiquitous accessory phase in igneous, metamorphic and sedimentary rocks. Natural apatites contain significant amounts of geologically relevant trace elements such as the rare earth elements (REE), high field strengths elements (HFSE) and large ion lithophile elements (LILE). Moreover, apatite is known to contain high concentrations of U and Th so that apatite formation can be established by conventional radioactive element decay dating or its thermal evolution can be reconstructed by investigating "fission tracks" caused by the decay of radioactive elements [1][2][3][4][5]. Furthermore, as human and animal bones consist of apatite, U-series dating of relatively young fossils is a new and exciting area of research in quaternary geosciences (e.g. [6]). To aid reliable analysis of trace element concentrations and isotopic ratios, matrix matched reference materials are needed. Single crystal homogeneous apatites that contain known amounts of trace elements would be ideal.
Moreover, apatite weathering and replacement processes in low-grade metamorphic rocks have been in the focus of research recently both in our institution and elsewhere [7][8][9][10]. This is mainly, as apatite, when equilibrated with or growing from a super-critical fluid in low-grade to high-grade metamorphic rocks, may contain a "geochemical fingerprint", that is a trace element signature from which one might be able to re-construct the composition of the fluid. In order to calibrate such a fingerprint, experiments are needed to investigate the partitioning of trace elements between apatite and fluids in a range of chemical compositions, pressures and temperatures. The experiments in turn need wellcharacterized starting materials, i.e. trace element bearing homogenous single crystals of apatite.
Here we report the high-temperature synthesis of mmsized single crystal chlorapatites (Ca 5 (PO 4 ) 3 Cl) using the so-called flux method. We tried several compositions, temperatures and synthesis routes and here we report on the most successful experiments, both in terms of crystal size as well as in terms of trace element homogeneity.

Experiments
Initial experiments in chemical compositions without trace elements confirmed the validity of previous experimental results [23]. Using the flux growth method pioneered by Prener and others, we could grow idiomorphic apatite single crystals up to ca. 6 mm in size. All experiments were conducted in Pt-crucibles in conventional vertical high-temperature furnaces at atmospheric pressure. The starting material consisted mainly of various mixtures of Ca 3 (PO 4 ) 2 and CaCl 2 , the latter of which acted as the flux. The experiments were heated to a temperature above the liquidus, they were held for a short time, and then slowly cooled to a final run temperature. During the cooling apatite crystals formed from the melt. After quenching, the experimental products were washed in water or diluted HCl for several hours. This effectively removes all the CaCl 2 flux. Table 1 lists experimental run conditions of each individual experiment. Figure 1 shows some representative single crystal apatites grown in our laboratory.

X-ray powder diffraction (XRPD)
For phase characterization an X-ray powder diffraction pattern was recorded using a PHILIPS X´PERT PW 9430 diffractometer with Cu-K α1 radiation and a primary Ge-(111) monochromator of Johansson Type. The operating conditions were 45 kV and 40 mA. Rietveld refinement was performed using the FULLPROF SUITE 2005 [27]. As starting parameters lattice parameters and crystal structural data including isotropic temperature factors for apatite-(CaCl) were taken from the literature [28]. The parameters which were varied for the refinement included the scale factor, the lattice parameters a and c, 4 background parameters, the sample displacement, two asymmetry parameters as well as the shape parameters w and Y of the Thompson-Cox-Hastings pseudo-Voigt profile function. The refinement converged to an R wp = 12.4% (R exp = 9.4%). No significant line broadening could be detected with respect to the Si-640a NIST-Standard which was used to determine the resolution function of the diffractometer. As can be seen from Figure 2 one weak reflection at 25.41°(2θ) remained unexplained which is therefore assumed to belong to an additional unidentified phase. As its intensity is about 0.7% of that of the most intense apatite reflexion we assume that the amount of that phase is about 1% by weight. The results are given in Figure 2 and Table 2 together with recent literature data.
In conclusion our apatite sample can be characterised as nearly pure chlor-apatite with very good crystallinity.

Synthesis of trace element bearing apatites
Once we were satisfied which the flux growth itself (SynCLAP3 and SynCLAP5, see Table 1), we conducted further experiments where the starting material contained a number of geochemically relevant trace elements. However, although we added relatively large amounts of trace elements (e.g., SynCLAP 6, 300 μg/g of each trace element, see Table 3) to the initial starting material mixture, we found that the resulting flux-grown apatites did not contain high concentrations of trace elements (generally well below 10 ppm of each trace element). We believe that the overall low concentrations of trace elements in the synthetic apatite crystals was caused by the fact that most of these trace elements, many of which  are trivalent rare earth elements, are incorporated into apatites by a coupled substitution which involves incorporation of Na + which replaces Ca 2+ or of Si 4+ which replaces P 5+ in the apatite structure. Below we show two possible exchange mechanisms for the incorporation of trivalent rare earth elements (REE) into the apatite structure [29,30].
We believe that the lack of Na + and Si 4+ in apatites grown in SynCLAP 6 strictly limited the incorporation of trivalent trace elements. Consequently, when we added some Na and Si (2 wt.%, SynCLAP 8, see Table 1 for details) to the starting material, we found that the flux-grown apatites contained significant amounts of Si and also significantly higher amounts of trace elements. This shows that incorporation mechanism (1) is more important than mechanism (2). Experiments SynCLAP 9 and 10 were similar to SynCLAP 8. The latter experiments yielded large and trace element bearing apatite but due to high SiO 2 contents of the melt lots of other acicular, needle-like, Ca-silicates formed in the melt. It was difficult to separate apatite crystals from the quench-crystallized matrix after the flux had been washed out. Figure 3 shows typical textures observed in the experiments SynCLAP 8-10.   Consequently, SynCLAP 11 and 12 contained less REE and less Na and Si (Table 3). In conclusion, the apatite single crystal synthesis is best-done following procedures and compositions like in experiment SynCLAP 12. The apatite crystals grown in these experiments are large (see Figure 3), they contain high concentrations of trace elements (Table 4) and the apatite crystals can be easily removed from the matrix.

Trace element concentrations in synthetic apatites
When single crystals are grown from a melt (or flux), trace elements will be incorporated into the crystals. The concentration of the trace elements in the crystals depends on their equilibrium partition coefficients (if equilibrium is attained) and the bulk concentration of the trace element. If diffusion rates of trace elements are low in the crystal (and this is the case for all geologically relevant trace elements in apatite [31][32][33], crystals may be zoned, at least in elements which are compatible, that is elements with a crystal/melt partition coefficient >1. This is due to the fact that the first crystals formed will contain comparatively high concentrations of this compatible trace element and the coexisting melt will be consequently depleted in this element. Crystals that form later, or layers of the crystal which form later during cooling will contain significantly lower concentrations of the trace element. As it is well known that many REE, Sr and other important trace elements are compatible in apatite [29,[34][35][36][37] we were concerned initially that our synthetic apatites may be significantly zoned. However, analytical results using in-house laser ablation ICP-MS techniques [7,36,[38][39][40][41] show that the  Trace element analyses performed using Laser Ablation ICP-MS. Analytical uncertainties are in the order of 15%. *: The individual analysis numbers (e.g., 1-1 stand for crystal 1 analysis 1) correspond with numbers in white circles (SEM images) in Figure 3.
apatites synthesized in SynCLAP12 are rather homogeneous in terms of major and trace elements, surely within the analytical uncertainties. The homogeneity surprised us initially but this is probably due to the fact that the partition coefficients between apatite and CaCl 2 -rich flux are probably very different from the published apatite/silicate melt partition coefficients (e.g., [29]). Moreover, the flux/crystal ratio employed in our study is high which further minimizes potential zoning during crystal growth. Figure 4 shows major and trace element concentrations of some representative apatite crystals from SynCLAP12. In summary, we present an effective procedure to synthesize mm-sized single crystals of chlorapatite that contain a variety of geochemically relevant trace elements. These crystals may be used as starting materials for further experiments or used as reference materials for geochemical analysis. Open access provides opportunities to our colleagues in other parts of the globe, by allowing anyone to view the content free of charge.
Publish with ChemistryCentral and every scientist can read your work free of charge W. Jeffery Hurst, The Hershey Company.
available free of charge to the entire scientific community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours you keep the copyright Submit your manuscript here: http://www.chemistrycentral.com/manuscript/