Contrasting Granite Metallogeny through the Zircon Record: A Case Study from Myanmar

Granitoid-hosted mineral deposits are major global sources of a number of economically important metals. The fundamental controls on magma metal fertility are tectonic setting, the nature of source rocks, and magma differentiation. A clearer understanding of these petrogenetic processes has been forged through the accessory mineral zircon, which has considerable potential in metallogenic studies. We present an integrated zircon isotope (U-Pb, Lu-Hf, O) and trace element dataset from the paired Cu-Au (copper) and Sn-W (tin) magmatic belts in Myanmar. Copper arc zircons have juvenile εHf (+7.6 to +11.5) and mantle-like δ18O (5.2–5.5‰), whereas tin belt zircons have low εHf (−7 to −13) and heavier δ18O (6.2–7.7‰). Variations in zircon Hf and U/Yb reaffirm that tin belt magmas contain greater crustal contributions than copper arc rocks. Links between whole-rock Rb/Sr and zircon Eu/Eu* highlight that the latter can monitor magma fractionation in these systems. Zircon Ce/Ce* and Eu/Eu* are sensitive to redox and fractionation respectively, and here are used to evaluate zircon sensitivity to the metallogenic affinity of their host rock. Critical contents of Sn in granitic magmas, which may be required for the development of economic tin deposits, are marked by zircon Eu/Eu* values of ca. ≤0.08.


U-Pb Results
The age for most samples is determined from the intersection of a regression line through uncorrected data, which is anchored at the modern day initial-Pb value ( 207 Pb/ 206 Pb = 0.83 4 ), and the concordia curve. This composition of common Pb has been demonstrated as appropriate for the NordSIM laboratory 5 . This regression approach yields essentially identical results to 204 Pb correction of individual ratios, which are also provided in the Excel file. Due to the near concordance of most data, neither the application nor form of common Pb correction results in significant differences to the calculated ages. LA-ICP-MS U-Pb age data for sample MY76 was previously reported in Gardiner et al. (2016) 6 .

Sample MY71
15 analyses were performed on 10 grains. Analyses from this sample indicate high to extreme U content that is vastly in excess of that in the 91500 zircon standard used for U/Pb calibration. U content correlates with the 207-corrected 238 U/ 206 Pb age, in which apparently older dates correlate with higher U content. This pattern implies a matrix-matching issue and hence in preference we use 207 Pb/ 206 Pb ages. All 15 analyses yield a 207 Pb/ 206 Pb weighted average of 76 ± 9 Ma (MSWD 2.1), interpreted as the best estimate of magmatic crystallization age of the granite (Fig. S1B, Group I).

Sample MY72
13 analyses were performed on 12 grains. Zircon crystals from MY72 indicate variable U concentrations, and eight analyses points have a U content of >4000 ppm (red; Group D). Those below this have a UO2/U ratio which are within the range of the standard run during the session, hence their U/Pb ratio is regarded as robust. One spot is interpreted to have sampled inherited material and yields an older 207 Pb-corrected 238 U/ 206 Pb age of 73 ± 2 Ma (Spot 02, blue; Group X). The remaining four analyses when fitted with a regression from modern day common Pb yield an intersection with the concordia curve of 64 ± 1 Ma (MSWD 2.0), interpreted to reflect the time of magmatic crystallization of the granite ( Fig. S1C; Group I).

Sample MY73
14 analyses were performed on 8 grains. Seven points from sample MY73 have extreme U contents > 2,000 ppm (red). These analyses are outside the UO2/U range of the standard during the session (Group D). One is interpreted to have samples inherited material and yields an older 207-corrected 238 U/ 206 Pb age of 61 ± 1 Ma (Spot 06, blue; Group X). The remaining four analyses when fitted with a regression from a modern day common Pb value yield an intersection with the concordia curve of 58 ± 0.5 Ma (MSWD 0.14), interpreted to reflect the time of magmatic crystallization of the granite ( Fig. S1D; Group I).

Sample MY74
9 analyses were performed on 8 grains. The analyses are concordant to discordant, principally reflecting a mixture between common and radiogenic Pb. One point shows a significant proportion of common Pb (f 207 > 37%), and is excluded from further discussion (Group D). The remaining 8 data points (Group I) yield a regression from common Pb which intersects the concordia curve at 59 ± 0.5 Ma (MSWD 0.54), interpreted as the magmatic crystallization age of the granite ( Fig. S1E; Group I).

Sample MY75
8 analyses were performed on 5 grains. Three analysis show very high U content >4,000 ppm and are excluded from age calculations due to either or both matrix matching and common Pb contamination issues (red; Group D). The remaining 5 analyses yield a regression from common Pb which intersects the concordia curve at 72 ± 2 Ma (MSWD 2.8), interpreted as the magmatic crystallization age of the granite ( Fig. S1F; Group I).

Lu-Hf Isotope Results
Initial 176 Hf/ 177 Hf ratios were calculated using the decay constant of Scherer et al.
(2001) 12 . Figure S3 shows an εHf evolution diagram for the Myanmar samples. Two-stage model ages were calculated assuming a 176 Lu/ 177 Hf ratio of 0.015 13 . Figure S3. εHf evolution plot for the Myanmar samples.

Oxygen Isotope Method
The oxygen isotope values in zircon were also measured using the Cameca IMS1280 ion microprobe at NordSIM. The analysis was performed with a ca. 2 nA Cs+ primary ion beam together with a normal incidence, low-energy, electron gun for charge compensation, medium field magnification (c. 80×) and two Faraday detectors (channels L′2 and H′2) at a common mass resolution of c. 2500 allowing simultaneous measurement of 16 O and 18 O. Measurements were performed in pre-programmed chain-analysis mode with automatic field aperture and entrance slit centering on the 16 O signal. The magnetic field was locked using nuclear magnetic resonance regulation for the entire analytical session. Each data-acquisition run comprised a 20 µm× 20 µm pre-sputter to remove the Au layer, followed by the centering steps and 64 s of data integration performed using a non-rastered, c. 10 µm spot. Field aperture centering values were found to be well within those for which no bias has been observed during tests on standard mounts 14  NIST 610 as the reference material and assuming 33.6 wt % Si, indicate that the accuracy was better than 3% for most elements with the exception of P (5%) and Fe (10%). The analytical precision was better than 10% for most elements.
Trace element data was screened for robustness through a range of techniques, including stability of the time resolved signal (a potential indication of inclusion or mixed domain sampling), and comparing the U-Pb isotope ratios in the trace element dataset with those measured using the smaller volume ion microprobe.

Trace Element Results
Rare earth elements (REE) were normalized to the chondritic values of Palme et al.
(2014) 17 . A plot of normalized zircon rare earth element concentrations is shown in Figure S4. Calculation of Ce and Eu anomalies uses the general forms Ce/Ce* = Ce/(La * Pr) 0.5 and Eu/Eu* = Eu/(Sm * Gd) 0.5 , using chondrite-normalized values. A hindrance to the calculation of Ce/Ce* is the often low-to negligible concentration of La in many zircons, a particular problem with some of the copper belt analyses. For these samples, we extrapolated a value for Ce* from Pr and Nd in logarithmic scale. Assessment of this approach for analyses where La was measured showed a good fit between Ce/Ce* calculated using a measured value of La, and that calculated using Ce* estimated from logarithmic extrapolation (Fig. S5). Analytical errors were propagated in quadrature both per analysis level, and at sample mean level.

Bulk Rock Composition
Major and trace element composition of all samples were analyzed at ALS Global Ireland, using ICP-MS (their method ME-MS81d). Samples were prepared through crushing,, milling, and then underwent lithium metaborate fusion. A prepared sample (0.200 g) is added to lithium metaborate ux (0.90 g), mixed well and fused in a furnace at 1000°C. The resulting melt is then cooled and dissolved in 100 mL of 4% HNO 3 / 2% HCl 3 solution. This solution is then analyzed by inductively coupled plasma -mass spectrometry.

Zircon Ce/Ce* versus Eu/Eu*
Plotting Ce/Ce* versus Eu/Eu*, sensitive to redox and fractionation respectively, shows metallogenic affinity for the Myanmar samples in the same sense as the whole-rock (Fe 2 O 3 /FeO) versus (Rb/Sr) diagram of Blevin et al. (1996) 18 . Figure S6 shows the full dataset plotted. We don't yet fully understand how to interpret the scatter of points so for the purposes of discussion we plot median values per sample.