Significant Zr isotope variations in single zircon grains recording magma evolution history

Significance Zircon, a common accessory mineral in crustal rocks, records plentiful and critical information on the Earth’s history. The isotopes of its major component, Zr, could be another powerful but unexplored tracer. We apply high-precision, high–spatial-resolution, in situ laser ablation Zr isotope measurements of magmatic zircons in continental arc plutonic rocks. Single zircon grains show impressive internal zoning with lighter Zr isotopes in the core but heavier ones toward the rim, solving a fundamental but controversial issue on how zircon fractionates Zr isotope in evolving magmas. Our results also reveal a strong temperature dependence of Zr isotopic fractionation. The Zr isotope is thus very promising in deciphering the differentiation history of magmatic systems and possibly the continental crust through time.


3
All studied samples are enriched in large ion lithophile elements and depleted in high field strength elements with insignificant Eu anomalies (Table S3). Late Cretaceous igneous rocks from the region show typical depleted mantle-like Sr-Nd isotope features (2). These features suggest the studied samples as part of the juvenile Gangdese arc crust, whose formation was related to the Late Cretaceous oceanic lithosphere subduction.

Text S3 Mass balance calculations for the modeled zircon radius vs. the Zr fraction of melt
The calculations assume 1) sphere shapes of zircon and the melt cell, and 2) Zr is highly incompatible (Kd ~ 0) in major-element minerals (nucleation of tiny zircon grains on the advancing surface of growing major-element minerals are not considered). For more complex situations, a recently published MATLAB software (3) is recommended. Given the same assumptions, comparable results were obtained using this software. The method used in this study is described below.
Parameters used in the calculation include: For mass balance, the total mass of Zr in zircon mzrc should equal to mcell (1−F), which is: 4/3 π r 3 ρ2 Czrc = 4/3 π R 3 ρ1 Cbulk (1−F) After reorganization, the radius of zircon is expressed as: Since R is user defined and Cbulk, Czrc, ρ1, and ρ2 are considered to be constant, the radius of the precipitated zircon would be proportional to the cubic root of 1−F (i.e., the fraction of Zr consumed by zircon crystallization). The final radius of the modeled zircon grain could be adjusted to fit the observed size of zircons by changing the radius of the melt cell R.

Text S4 Analytical methods
Whole-rock geochemical analysis. Rock samples were crushed in a corundum jaw crusher and then powdered down to >200 mesh in an agate mill. Major and trace element compositions were analyzed at the Wuhan SampleSolution Analytical Technology Co., Ltd., Wuhan, China. The major elements were measured by XRF (Rikagu RIX 2100) using fused glass disks. Precision and accuracy are better than 5%, as shown by analyses of USGS basalt and andesite standards (BCR-2, BHVO-2, and AGV-1). Trace elements were analyzed by Agilent 7900 ICP-MS. Sample powders were digested by HF + HNO3 acid in Teflon bombs under high pressures (7-12 MPa).
For most trace elements, the measured concentrations of BCR-2, BHVO-2, and AGV-1 agree with their reference values within a 5% difference. Duplicates of three samples were analyzed to test the analytical reproducibility, which is better than 5% for most elements. isotopes (4). A "wire" signal smoothing device is included in this laser ablation system (5). The carrier and make-up gas flows were optimized by ablating NIST SRM 610 to obtain the maximum signal intensity for 208 Pb, while keeping low ThO/Th and Ca 2+ /Ca + ratios to minimize the matrixinduced interference. The laser beam was set to 32 μm in diameter with a frequency of 6 Hz. The energy density was about 4 J/cm 2 . Standard zircon 91500 (6) was used as an external standard for U-Pb dating. Zircon GJ-1 (7) was measured as the monitor standard. Both NIST SRM 610 and the standard zircons were used for trace element calibration. Reference values for Ti concentrations of zircon 91500 and GJ-1 are from refs. (8) and (9), respectively. Silicon was used for internal standardization to reduce the matrix effect between synthetic glass and natural zircons (10). Each analysis incorporated a background acquisition of 20-30 s (gas blank) followed by 50 s data acquisition from the sample. Data reduction was performed using an Excelbased software ICPMSDataCal (11). Common Pb correction (12) was negligible for most zircons.
Concordia diagrams and weighted averages were produced using Isoplot (ver. 4.15) (13). The obtained weighted average 206 Pb/ 238 U age is 600 ± 10 Ma (2SD, n = 32), consistent with its reference 206 Pb/ 238 U age of 599.8 ± 4.8 Ma (2σ) within analytical uncertainty (7). In-situ U-Pb isotope and Ti concentration data of sample and standard zircons are given in Table S2 and   Table S8, respectively. Analytical Technology Co., Ltd., Wuhan, China. Helium was applied as a carrier gas. Argon was used as the make-up gas and mixed with the carrier gas via a T-connector before entering the ICP. A "wire" signal smoothing device is included in this laser ablation system (5). In order to obtain a high spatial resolution without losing much of the precision, the spot size and frequency of the laser were set to 24 µm and 6 Hz, respectively. The analyzed spots were as close as possible to the spots for Zr isotope analyses. Element compositions were calibrated against various reference materials (NIST SRM 610, BHVO-2G, BCR-2G, and BIR-1G) (11). Standard zircon 91500 and GJ-1 were also analyzed as monitoring standards. Each analysis incorporated a background acquisition of approximately 20-30 s followed by 50 s sample data acquisition. An

In
Excel-based software ICPMSDataCal was used for data reduction (11). The obtained major and trace element data of sample and standard zircons are given in Table S5. Resources, China University of Geosciences (Wuhan). The Faraday collector configuration of the mass spectrometer was composed of an array from L4 to H2 to monitor 89 Y + , 90 Zr + , 91 Zr + , 92 Zr + , 94 Zr + , 95 Mo + , and 96 Zr + . A high-sensitivity combination of X-skimmer and Jet-sample cones was mounted in the Neptune Plus interface. The mass spectrometer was operated in the low mass resolution mode.

In-situ Zr isotope analyses of zircons by fs
The laser ablation was conducted under a helium atmosphere, while argon was mixed into the sample-out line downstream from the ablation chamber before entering the torch of the mass Small dots denote late Cretaceous igneous rocks from the eastern Gangdese arc (17).   Tables   Table S1. Summary of whole-rock compositions, zircon U-Pb ages, Zr isotope compositions, and Ti-in-zircon temperatures. Table S2. LA-ICP-MS U-Pb isotope and Ti concentration data of zircons in plutonic arc rocks from southern Tibet. Table S3. Whole-rock geochemistry of plutonic arc rocks from southern Tibet. Table S4. In-situ fs-LA-MC-ICP-MS Zr isotope compositions of zircons in plutonic arc rocks from southern Tibet. Table S5. In-situ LA-ICP-MS major and trace element profile data of zircons in plutonic arc rocks from southern Tibet.   Table S8. LA-ICP-MS U-Pb isotope and Ti concentration data of standard zircons. Table S9. Summary of the operating parameters for MC-ICP-MS and the femtosecond laser ablation system.