Fluorine & chlorine

Fluorine and chlorine in alkaline rocks and A-type granites

Eby, G.N. and Charnley, N.

The role of halogens, particularly fluorine, in the petrogenesis of A-type granites and related rocks has been a subject of continuing discussion in the petrologic literature. The determination of the halogen content of a magma is a significant challenge. Whole-rock F and Cl concentrations, even for fine-grained rocks, may not represent the halogen content of the magma. Melt inclusions, when available, can yield the halogen content for a magma at a particular point in its evolution. Halogen containing minerals, most notably biotite and apatite, have been used to estimate the HF and HCl fugacity (in the case of biotite) or directly the F and Cl melt concentrations (using apatite). Here we report the preliminary results from an ongoing study of the halogen content of magmas using the F and Cl content of biotite, apatite, and amphibole to estimate the F and Cl content of the coexisting melt.

Determination of F and Cl in minerals. In minerals, F and Cl are generally determined using the electron microprobe (EMP). Particularly in the case of F, such determinations are far from routine. We have investigated the analytical variables and have come to the following conclusions: (1) F and Cl should be determined using a defocused beam (5 to 10µm) and a low accelerating potential, (2) standardization should be done on a F-containing mineral that is not sensitive to counting times (we use fluorite), (3) F and Cl should be determined at the start of an analytical sequence, (4) the calculated values are sensitive to the correction routine used to reduce the data – ZAF vs PAP vs etc. and (5) a F-standard should be interspersed with unknowns to measure the variability of the F determination. For measurements made on the same EMP, a reasonable analytical error for F is +/- 10% relative and for Cl +/- 5% relative at concentrations several standard deviations above the detection limit.

Estimating F and Cl melt concentrations from mineral chemistry. Based on experimental results, Munoz (1984, 1992) published a set of equations that could be used to estimate the HF and HCl fugacity of vapor and melt co-existing with biotite. The HF and HCl fugacities can be converted to concentrations using the algorithms of Piccoli and Candela (1994). Icenhower and London (1997) experimentally determined F and Cl partition coefficients for biotite/melt equilibria at 640 to 680^oC. The Munoz (1992) and Icenhower and London (1997) models yield similar results for biotites with mg# = 60 to 20. For mg#s <20, there is a significant discrepancy between F values calculated by the two models. The F algorithm of Icenhower and London (1997) was not calibrated for mg#s <20 and this model should not be used for biotites with mg#s <20. In order to have consistent results, in our investigations we have used the model of Munoz (1984, 1992). For Cl, the two models yield similar results. In the case of F the calculations are sensitive to temperature and an independent temperature estimate is required. The Cl calculation is sensitive to both temperature and pressure. For apatite we use the most recent results of Webster et al. (2009).

Currently there is no model relating the distribution of F and Cl between amphibole and melt. We have developed an empirical model for F partitioning between amphibole and melt. Melt F concentrations are estimated using the measured concentrations of F in apatite co-existing with amphibole. Based on these measurements we propose that the following (preliminary) equations can be used to estimate the F concentrations of a melt. The calculation was done for amphibole-apatite pairs that are inferred to have co-crystallized between 950 and 850^oC.

D^Famph/Fmelt = 1.03 + 0.06*mg#

The possible effect of pressure is unknown.

Chilwa alkaline province. The Chilwa alkaline province is located in Malawi near the southern end of the modern-day East African rift system. The province is lithologically diverse and includes carbonatite, basanites and nephelinites, nepheline-sodalite syenites, nepheline syenites, syenites, and granites. All these rocks were emplaced between ca 130 and 120 Ma and are apparently petrogenetically related. The Chilwa granites are classified as A₁ granites (Eby, 1992) and were presumably emplaced in an anorogenic setting. We have analyzed apatites, biotites, and amphiboles from all of the rock-types except the carbonatite, phonolites, and nephelinite/basanites.

Whole-rock F and Cl concentrations are variable. The nephelinites/basanites are uniformly very low in Cl (near or below detection) and have up to 0.6 wt.% F. The more evolved silicate rocks show a range in both F and Cl concentrations of up to 0.8 wt.% F and 0.7 wt.% Cl. Cl/F ratios >1 are found for a number of the silica undersaturated felsic rocks (and some of these do contain sodalite). For the syenites and granites F/Cl ratios are usually less than 1. In the granites and syenites whole-rock F ranges between 0.1 and 0.4 wt.% and whole-rock Cl is usually less than 0.1 wt.%.

The apatites are all fluorapatites and Cl is at or below the detection limit. To date, all calculated F-melt values fall in a restricted range between 0.248 and 0.303 wt.%. On petrographic grounds apatite is an early crystallizing mineral and is found as inclusions in feldspar and the mafic silicates. There is no apparent relationship between the paragenetic association and the F content. We conclude that the apatites are recording F content through much of the solidification history of the melt and that there is little variation in F. In almost all cases, the calculated melt F concentration is greater than the F content determined in the whole-rock analysis.

The biotites present a somewhat different picture. Calculated F melt concentrations range from 0.1 to 1.0 wt.%. In almost all cases these values exceed the measured whole-rock concentrations. They are also generally greater than the F melt concentrations determined from the apatite chemistry. The biotites appear relatively late in the crystallization history of the magmas and are thus recording F values for the residual melts. Calculated Cl melt concentrations range from 0.02 to 0.2 wt.%. In most cases these values are less than the measured whole-rock Cl concentrations. This observation leads to the conclusion that Cl is a late, probably of hydrothermal origin, addition to the rock, and that the sodalite found in some of the silica undersaturated syenites is of hydrothermal origin.

Taken in total, for this A-type granite province, magmatic F content is in the range of 0.2 to 0.4 wt.%, only becoming more concentrated near the end stage of crystallization, and the magmatic Cl content is generally less than 0.1 wt.%.

Comparison with Argentina A₂ granites. Dalhquist et al. (2010) recently reported biotite chemistry and calculated F and Cl concentrations for A₂-type granites from western Argentina. We have obtained additional data for biotites from associated granite bodies. In all cases, biotite is the principal halogen containing phase and appears to have crystallized through much of the solidification range. All these granites have a largely crustal source with a minor input of mantle material. For the San Blas pluton, which contains late stage mineralization, F melt = 0.27 to 0.54 wt.% and Cl melt = 0.03 to 0.11 wt.%. For the Achala batholith, which has associated U mineralization, F melt = 0.13 to 0.33 wt.% and Cl melt = 0.02 to 0.06 wt.%. These values are in the same range recorded for the A₁-type granites of the Chilwa alkaline province. At least in terms of this limited data set, there seems to be no difference in the F and Cl content of A₁ and A₂ granites. This observation does call into question one of the assumed characteristics of A-type granites derived from the crust, i.e., high F content. Dalhquist et al. (2010) do report that other (none A-type) granites in the same province have lower F contents. Hence, at least at the province level there is a distinction between A-type and none A-type granites.

References

Dalhquist, J. A., Alasino, P. H., Eby, G. N., Galindo, C., Casquet, C., 2010. Fault controlled Carboniferous A-type magmatism in the proto-Andean foreland (Sierras Pampeanas, Argentina): Geochemical constraints and petrogenesis. Lithos 115, 65-81.

Eby, G. N., 1992. Chemical subdivision of the A-type granitoids: petrogenetic and tectonic implications. Geology 20, 641-644.

Icenhower, J., London, D., 1997. Partitioning of fluorine and chlorine between biotite and granitic melt: experimental calibration at 200 MPa H2O. Contributions to Mineralogy and Petrology 127, 17-29.

Munoz, J. L., 1984. F-OH and Cl-OH exchange in micas with applications to hydrothermal ore deposits. In: Bailey, S. W. (Ed.) Reviews in Mineralogy, Micas, vol. 13. Mineralogical Society of America, pp. 469 - 494.

Munoz, J.L., 1992. Calculation of HF and HCl fugacities from biotite compositions: revised equations. Geological Society of America, Abstracts with Programs 24, A221.

Piccoli, P., Candela, P., 1994. Apatite in felsic rocks: a model for the estimation of initial halogen concentrations in the Bishop Tuff (Long Valley) and Tuolumne intrusive suite (Sierra Nevada Batholith) magmas. American Journal of Science 294, 92-135.

Webster, J. D., Tappen, C. M.., Mandeville, C. W., 2009. Partitioning behavior of chlorine and fluorine in the system apatite-melt-fluid. II: Felsic silicate systems at 200 MPa. Geochimica et Cosmochimica Acta 73, 559-581.

Red text represents modifications made to the meeting abstract.

Electronic version of paper

Back to publications.