Further Characterization of the BB Zircon via SIMS and MC-ICP-MS for Li, O, and Hf Isotopic Compositions

In this contribution, we report the results for the characterization of the BB zircon, a newly developed zircon reference material from Sri Lanka, via secondary ion mass spectrometry (SIMS) and multiple-collector inductively coupled plasma-mass spectrometry (MC-ICP-MS). The focus of this work was to further investigate the applicability of the BB zircon as a reference material for micro-beam analysis, including Li, O, and Hf isotopes. The SIMS analyses reveal that BB zircon is characterized by significant localized variations in Li concentration and isotopic ratio, which makes it unsuitable as a lithium isotope reference material. The SIMS-determined δ18O values are 13.81‰ ± 0.39‰ (2SD, BB16) and 13.61‰ ± 0.40‰ (2SD, BB40), which, combined with previous studies, indicates that there is no evidence of conspicuous O isotope heterogeneity within individual BB zircon megacrysts. The mean 176Hf/177Hf ratio of BB16 determined by solution MC-ICP-MS is 0.281669 ± 0.000012 (2SD, n = 29) indistinguishable from results achieved by laser ablation (LA)-MC-ICP-MS. Based on the SIMS and MC-ICP-MS data, BB zircon is proposed as a reference material for the O isotope and Hf isotope determination.

For this purpose, a further comprehensive study of the Li, O, and Hf isotopic characteristics of BB zircons is reported. We investigate its Li isotopic ratios using SIMS for the first time. In addition to extensive testing of Hf isotope composition homogeneity of BB zircon by LA-MC-ICP-MS, we determine the mean Hf isotopic ratio using the solution MC-ICP-MS method for the reliable recommended value. Furthermore, we assess its O isotope homogeneity using SIMS.

Sample Descriptions
Zircon megacrysts BB were collected from a placer deposit of the Ratnapura gemstone field, located in the south-western region of the Sri Lanka Highland Complex. Approximately 300 g of BB zircons, comprising some eighty grains, were acquired and numbered. Cathodoluminescence (CL) images reveal that BB zircons have no zoning or fine oscillatory zoning ( Figure 1). Santos et al. [33] selected several individual crystals (e.g., BB9, BB12, BB17, BB25, and BB39) to conduct detailed U-Pb age, O, and Hf isotopic determinations via TIMS, SIMS, and LA-(MC)-ICP-MS. Additionally, U-Pb LA-MC-ICP-MS and (CA)-TIMS data of five BB zircons (BB38, BB39, BB40, BB41, and BB42) were reported by Lana et al. [32]. In this study, Li and O isotopes of two zircon megacrysts BB16 and BB40 were measured by SIMS, and Hf isotopic measurements were carried out using LA-MC-ICP-MS and solution MC-ICP-MS.

Analytical Methods
For the micro-beam analyses, one large BB40 zircon shard (~6 mm in diameter) and six BB16 zircon shards (five small (0.5-1 mm in diameter) and one large (~3 mm in diameter)) were placed in epoxy mounts together with Plešovice, Penglai, and Qinghu zircon reference materials. The shards were ground away and polished to expose their centers for analysis. In this work, the uncertainties of single analysis are stated as 2 standard errors (2SE), while the uncertainty for the grand mean value is reported as 2 standard deviations (2SD). The external reproducibility of reference materials was not propagated into the uncertainties of single measurements or final grand mean values.

SIMS Li Isotope Analysis
The Cameca IMS 1280HR ion microprobe was used for the Li isotopic measurements of BB zircons at the Institute of Geology and Geophysics, Chinese Academy of Sciences (IGGCAS) in Beijing, following the detailed procedures described by Li et al. [39]. A 20 × 30 µm elliptical spot size was used to traverse BB zircon shards. 6 Li and 7 Li were simultaneously detected. One spot measurement comprised 60 cycles with a total measurement time of about 15 min, including pre-sputtering of 30 s, secondary beam centering of 120 s, and collection for Li isotopic signals of 720 s. Measurements of Li isotopic ratios and concentrations were corrected according to the recommended values of δ 7 Li = 2.1‰ ± 1.0‰ (2SD) and [Li] = 0.86 ± 0.18 µg g −1 for the M257 zircon standard.

SIMS O Isotope Analysis
The oxygen isotopic compositions of BB zircon were measured on the same Cameca IMS 1280HR ion microprobe at the IGGCAS in Beijing, with the similar analytical procedures reported by Li et al. [28] and Tang et al. [40]. The Gaussian-focused Cs + primary ion beam was used at 10 kv to sputter oxygen ion from BB zircon, achieving an intensity of ~1.5 nA with a spot size of about 20 µm on the sample surface. To compensate for sample charging, a normal-incidence electron gun was used. Moreover, the nuclear magnetic resonance (NMR) controller was used to stabilize the magnetic field. During the analysis, 16 [41]. One spot measurement involved pre-sputtering of 120 s, secondary beam centering of 120 s, and collection of oxygen isotopic signals of 60 s, with a total analytical time of about 3 min. The Penglai zircon was used as the reference material with a recommended δ 18 O VSMOW value of 5.31‰ ± 0.10‰ (2SD) [28]. The measurements of secondary zircon reference material Qinghu gave a grand mean δ 18 O value of 5.37‰ ± 0.43‰ (2SD, n = 27), identical to the recommended value reported in Li et al. [29]

LA-MC-ICP-MS Hf Isotope Analysis
Micro-beam Hf isotopic analyses for BB16 zircon were conducted on a Thermo Scientific Fisher Neptune Plus MC-ICP-MS coupled with a Coherent Geolas Pro 193 nm laser ablation system at the IGGCAS in Beijing (Table 1), which were similar with those reported by Wu et al. [42] and Huang et al. [43]. The LA system was operated using a beam size of 60 µm for BB and SA01 zircons and 44 µm for Mud Tank zircon with a repetition rate of 6 Hz and an energy density of ~4.5 J cm −2 . Helium was used as the carrier gas with a flow rate of 640 mL min −1 . Aiming to achieving higher sensitivity, additional nitrogen was added to the carrier gas with a flow rate of 4 mL min −1 . One spot measurement comprised one block of 200 cycles with an integration time of 0.131 s. The Hf isotopic compositions of the gas blank were not measured because of the extremely low Hf signal. Correction for the isobaric interference of 176 Lu on 176 Hf was performed by measuring the intensity of the interference-free 175 Lu isotope ( 176 Lu/ 175 Lu = 0.02655) assuming βLu = βYb. The mean 173 Yb/ 172 Yb ratio for the individual spot analysis was used to calculate the fractionation coefficient (βYb), and the contribution of 176 Yb to 176 Hf was corrected by applying ratios of 176 Yb/ 172 Yb = 0.588673 and 173 Yb/ 172 Yb = 0.73925. Instrumental mass bias was corrected based on the normalization to 179 Hf/ 177 Hf = 0.7325 using the exponential law [42]. Correction for molecular interferences (e.g., 160 Gd 16 O) was not made due to low light to middle rare earth element contents in zircon. Zircon reference materials Mud Tank and SA01 analyzed during the same session gave grand mean 176 Hf/ 177 Hf values of 0.282507 ± 0.000032 (2SD, n = 15) and 0.282287 ± 0.000020 (2SD, n = 15), consistent with the reported results [23,38]. Table 1. Typical multiple-collector inductively coupled plasma-mass spectrometry (MC-ICP-MS) instrument parameters for Hf isotopic composition analysis of BB zircon.

Solution MC-ICP-MS Hf Isotope Analysis
Seven small shards (0.41-1.36 mg each) of BB16 zircon, without any pretreatment, were digested in a mixture of concentrated HNO3 and HF using stainless steel jacketed Teflon bombs that were placed in an oven at 220 °C for three days. After evaporation, the samples were then re-dissolved in 3 mol L −1 HCl. Separation and purification of the attained solutions for Lu and Hf were carried out by means of ion exchange columns using Ln Spec resin. Solution Hf isotope measurements were performed on a Thermo Fisher Scientific Neptune Plus MC-ICP-MS system at the IGGCAS in Beijing. Details of the procedure have been reported by Yang et al. [44]. Instrumental mass bias was corrected by the measured 179 Hf/ 177 Hf and its natural ratio of 0.7325. The measured 173 Yb and 175 Lu values were used to correct the possible interferences of 176 Yb and 176 Lu on 176 Hf, utilizing 176 Lu/ 175 Lu = 0.02655 and 176 Yb/ 173 Yb = 0.79631 [45]. During the solution Hf isotopic composition analysis, the Alfa Hf solution (JMC14374) was measured and yielded 176 Hf/ 177 Hf values of 0.282193 ± 0.000007 (2SD, n = 6) during the first session and 0.282185 ± 0.000005 (2SD, n = 6) during the second session, which are consistent with reported values in previous studies [42].

SIMS Li Isotope Composition
Twenty-one analyses were conducted on four small and one large BB16 zircon shards. The four small BB16 zircon shards have consistent δ 7 Li values within analytical uncertainty and give a grand mean of 2.3 ± 2.0‰ (2SD). The large shard has a large δ 7 Li value range from −7.5‰ to −0.6‰. The 7 Li + count rates of BB16 zircon are low and highly variable (579 to 4874 cps/nA), and the calculated Li concentrations range from 0.10 to 0.83 µg g −1 ( Table 2 and Table S1; Figure 2a). The Li concentrations and isotopic compositions of BB40 zircon were measured along four traverses, with each traverse consisting of 19-20 analytical spots (Table 2; Figure 2b). The distance between two spots along the traverse was roughly equal, and visible cracks were avoided. Traverses 1-2 are perpendicular to traverses 3-4. The profile of Li isotopic compositions and concentrations, as revealed by traverses 1 and 2, are quite similar, namely nearly flat for the first nine analytical spots of each traverses, then rise for the subsequent 4-5 analytical spots, and finally descend for the last 5-6 analytical spots (Figure 3a There are many factors that can affect the distribution of Li in zircons. Gao et al. [14] invoked the effect of diffusion to explain the phenomenon that Li contents and Li isotopic ratios are largely variable in zircon rims but homogeneous in zircon cores, which are distinct from change trends of Li contents and isotopic ratios in BB zircons. Sliwinski et al. [46] suggested that lithium in zircon is primarily sequestered within inclusions. However, transmitted light images show that no visible inclusion was detected in BB zircons. It is unclear which factors control the systemic change in Li isotopic compositions and concentrations of BB40 zircon at present. The heterogeneity of Li isotopic compositions and concentrations revealed by this study indicates that BB zircon is unusable as a reference material for micro-beam Li isotopic analysis. Several zircon reference materials used in U-Pb geochronology, including 91500, BR266, TEMORA 2, SA01, Plešovice, Penglai, and Qinghu, have been checked for the homogeneity of Li isotopic compositions, and all of them were shown to have large ranges in δ 7 Li values and Li concentrations [14,38,39], which were ascribed to fast diffusion velocity of Li ion in zircon [14]. At present, only M257 and M127 have been documented to have homogenous Li isotopic compositions and concentrations [31,39]. However, these two zircon reference materials are too small in quantity to be widely used. Accordingly, it is still imperative to find more zircon reference materials with homogenous Li isotopic compositions.
Overall, we conclude from the data set that there is no evidence of oxygen isotope heterogeneity within the BB16 and BB40 zircon crystals. BB16 and BB40 crystals have similar oxygen isotopic compositions with BB9 and BB12 within analytical uncertainty but are still about 0.8‰ to 1‰ heavier in their oxygen isotope compositions than those of the BB25 and BB39 crystals [33]. Although a systematic offset of 0.4‰ for O isotopic compositions between 91500 (the reference material used in Santos et al. [33]) and Penglai (the reference material used in this study) was detected by Santos et al. [33], this cannot explain the huge oxygen isotopic difference (1.07‰) between zircons BB16 and BB25. It is notable that BB09 zircon is 0.76‰ heavier in their oxygen isotope compositions than that of BB25 zircon, even though they were both corrected based on the recommended value of 9.86‰ ± 0.24‰ (2SD) for 91500 [33]. All these results indicate that resolvable oxygen isotopic variations exist among different BB zircon megacrysts, as shown in Table 3 and Figure 5. In view of this, detailed and careful assessments of oxygen isotopic compositions of individual BB zircon megacrysts should be conducted before being used as oxygen reference materials. Details of SIMS oxygen isotope data are shown in Table S2.

Solution and LA-MC-ICP-MS Hf Isotope Data
A total of eighty-four Hf isotope measurements by LA-MC-ICP-MS were undertaken to investigate the homogeneity of Hf isotopes on six BB16 zircon shards (Table 4). They show very low 176 Yb/ 177 Hf ratios between 0.000603 to 0.002679. There is no visible correlation between the measured 176 Hf/ 177 Hf and 176 Yb/ 177 Hf ratios (Figure 6), suggesting an accurate correction of isobaric interferences of 176 Yb on 176 Hf. Fourteen Lu-Hf isotopic analyses on BB16 zircon shard1 were conducted, and the measured 176 Hf⁄ 177 Hf values range from 0.281650 ± 0.000015 (2SE) to 0.281705 ± 0.000017 (2SE), with a grand mean of 0.281673 ± 0.000025 (2SD, n = 14). Likewise, 14 random Lu-Hf isotopic measurements were carried out on BB16 zircon shard 2, 3, 4, 5, and 6, respectively, and the results are listed in Table  4. As shown in Figure 7b,c, all the eighty-four measured 176 Hf/ 177 Hf ratios form a Gaussian distribution and give a grand mean of 0.281672 ± 0.000025 (2SD; six shards). Although the LA-MC-ICP-MS measurements have documented the homogeneity of Hf isotopic compositions, it is notable that no solution Hf isotope analysis has been carried out in previous studies. In this study, seven aliquots of BB16 zircons were dissolved for the chemical purification of Hf. Results of solution Hf isotope analyses by MC-ICP-MS are listed in Table 4. Twenty-nine MC-ICP-MS measurements were conducted on the seven aliquots of purified Hf solution in two sessions, which resulted in 176 Hf/ 177 Hf values of 0.281659 ± 0.000010 (2SE) to 0.281684 ± 0.000008 (2SE). In session 1, the measured 176 Hf/ 177 Hf values form a grand mean of 0.281670 ± 0.000012 (2SD, n = 14). Session 2 comprised fifteen measurements on the same seven aliquots of Session 1 and achieved a grand mean 176 Hf/ 177 Hf ratio of 0.281669 ± 0.000012 (2SD, n = 15). Overall, the grand mean for all twenty-nine solution MC-ICP-MS measurements is 0.281669 ± 0.000012 (2SD, n = 29; Figure 7a). The obtained 176 Hf/ 177 Hf isotopic ratios in the two sessions are identical within analytical uncertainty. Therefore, the mean 176 Hf/ 177 Hf ratio of 0.281669 ± 0.000012 (2SD) determined by solution MC-ICP-MS measurements is taken to be the best estimate of the Hf isotope compositions of BB zircon. Complete data are given in Table S3.
The results of LA-MC-ICP-MS analyses are consistent with the value of the solution MC-ICPMS results within analytical uncertainty, and indistinguishable within uncertainty from the average value of 0.281676 ± 0.000009 (2SD, n = 16) reported by Santos et al. [33]. Therefore, the BB16 zircon shards are fairly homogeneous in Hf isotopes at the 60 × 60 µm sampling size and appear to lack any significant intra-and inter-shard variations. For individual analyses, see Table S3. Previous studies have also carried out many LA-MC-ICP-MS Hf isotopic measurements on other BB zircon megacrysts, and they also yielded very low 176 Yb/ 177 Hf ratios and identical 176 Hf/ 177 Hf ratios of 0.281668 ± 0.000029 (2SD) to 0.281684 ± 0.000016 (2SD) [33]. This signifies that all the BB zircon megacrysts have comparable Hf isotopic compositions. Compared to other widely-used zircon reference materials (Figure 8), BB zircons have relatively low 176 Yb/ 177 Hf ratios, and thus, they can be used as a reference material to adjust for inter-laboratory bias of the measured 176 Hf/ 177 Hf ratios, as suggested by Fisher et al. [47].

Conclusions
Combining our results with the earlier study by Santos et al. [33] indicates that no systematic dispersion of O isotopic compositions within single zircon megacrysts is detectable at the analytical precision of the SIMS analyses (Figures 3c and 4). However, detectable variations have been revealed among different zircon megacrysts, as shown in Table 3 and Figure 5. We strongly suggest that each BB zircon needs to be independently assessed for O isotope compositions before it can be used as a reference material.
Individual BB zircons show significant heterogeneity of Li isotopic compositions and concentrations, and thus, BB zircon cannot be used as a reference material for micro-beam Li isotopic determinations.
Beyond providing a detailed characterization via SIMS, this study conducts the testing of Hf isotope composition homogeneity of BB16 zircon by solution MC-ICP-MS. The result clearly indicates that the recommended 176 Hf/ 177 Hf value is 0.281669 ± 0.000012 (2SD), which is in good agreement with the statistical mean of LA-MC-ICP-MS analysis in this work and previous work by Santos et al. [33]. The O isotopic compositions of BB16 zircon were documented to be homogenous by SIMS analyses. Therefore, we propose that BB16 zircon is a suitable reference material for in situ Hf and O isotopic measurements of zircon.