TZACK@GWDG.DE
Although experimental determinations of trace element partitioning (D) are important to calibrate partitioning as a function of known parameters such as T, P and bulk composition, most such determinations are done at near-liquidus conditions where crystal growth conditions are optimal, and allow only one mineral phase to grow. In-situ trace element measurements in natural systems are a useful addition to experimental results if samples can be shown to be in equilibrium. They have the advantage that several mineral phases can be analysed in the same sample, so that difficulties in choosing between different experimental data sets can be avoided.
Suitable equilibrium conditions for this kind of analysis were found during investigations of garnet pyroxenites from Kakanui (Zack, 1995). Microprobe measurements show that all coexisting phases in each sample are homogenous and unzoned. The xenoliths consist of clinopyroxene (CPX), amphibole (AMP) and garnet (GRT) with widespread minor amounts of ilmenite and spinel. All phases are in mutual contact with predominant poikiloblastic textures indicating a magmatic origin. Recrystallization (such as exsolution lamellae of GRT from CPX), although common at other localities, could not be observed at Kakanui. Phases show pronounced differences in Mg# between samples (e.g. GRT = 68-45) and a strong correlation of the Mg# between coexisting phases exists. Trace element analyses for GRT and AMP by Laser Ablation Microprobe (LAM) support the former petrographic and petrologic observation of equilibrium conditions and allow the compilation of a partition coefficient data set (Table 1).
This data set is consistent with recently published experimental results for DAMP/L (LaTourette et al., 1995) and DGRT/L (Hauri et al., 1995) and also includes the first reliable values
of DAMP/L for Sc and V. The DAMP/L for Sc (0,75) differs significantly from DCPX/L for Sc (1.31) and could be a potential tracer for recognizing the influence of AMP rather than CPX in fractionation series of volcanic rocks, with Sc being enriched by AMP fractionation and depleted by CPX fractionation.
Kakanui minerals may prove to be ideal candidates for trace element standards for in-situ analyses.
Hart, S.R. & Dunn, T., Contr. Miner. Petrol. 113, 1-8 (1993).
Hauri, E.H., Wagner, T.P. & Grove, T.L., Chem. Geol. 117, 149-166 (1994).
Jenner, G.A., Foley, S.F., Jackson, S.E., Green, T.H., Fryer, B.J. & Longerich, H.P., Geochim. cosmochim. Acta 58, 5099-5103 (1994).
LaTourrette, T., Hervig, R.L. & Holloway, J.R., Earth planet. Sci Lett. 135, 13-30 (1995).
Zack, T., Diploma thesis, Universität Göttingen (1995).
Table 1.
DCPX/L1 DAMP/CPX s (%) DAMP/L DGRT/CPX s (%) DGRT/L
Th 0.0122 0.94 12 0.011 0.045 0.0005
U 0.01032 1.01 11 0.010 0.22 0.0023
Nb 0.0077 22.11 16 0.17 <0.06 <0.0005
Ta 0.0193 8.32 29 0.16 <0.4 <0.008
La 0.0536 1.27 3 0.068 0.004 0.0002
Ce 0.0858 1.14 5 0.10 0.019 25 0.0016
Pr (0.14) 1.10 10 0.15 0.039 21 0.0055
Sr 0.1283 4.50 17 0.58 0.007 116 0.0009
Nd 0.1873 1.08 7 0.20 0.088 22 0.016
Sm 0.291 0.98 15 0.29 0.37 2 0.11
Zr 0.1234 0.59 9 0.073 0.77 48 0.10
Hf 0.256 0.54 14 0.14 0.29 45 0.074
Eu (0.32) 1.08 5 0.35 0.66 18 0.21
Gd (0.40) 1.07 6 0.43 1.35 10 0.54
Tb (0.42) 1.08 8 0.45 2.17 8 0.91
Dy 0.442 0.98 21 0.44 3.48 20 1.54
Y 0.467 1.05 23 0.49 6.23 23 2.91
Er 0.387 1.02 29 0.40 8.57 38 3.32
Yb 0.43 1.41 29 0.61 19.36 9 8.32
Sc 1.31 0.58 9 0.75 2.29 48 3.00
V 3.1 0.94 13 2.92 0.36 23 1.11
1 Hart and Dunn (1993), 2 Hauri et al., (1994), 3 Jenner et al., (1994)