Characterization of Four Copper Materials for Application as Reference Materials for High Precision Copper Isotope Analysis by Laser Ablation Inductively Coupled Plasma Multi-Collector Mass Spectrometry

Abstract

Laser ablation inductively coupled plasma multi-collector mass spectrometry (LA-MC-ICP-MS) allows rapid, in situ, highly precise measurements of Cu isotope ratios of native Cu and Cu-bearing minerals. However, the National Institute of Standards and Technology Cu-metal isotope standard NIST SRM976 that is commonly used to calibrate LA-MC-ICP-MS Cu isotope measurements of native Cu is no longer available. We have investigated the suitability of four Cu metal materials, SSC-1, SSC-3 and SSC-4 (cathode Cu metal rods) and CUPD-1 (Cu anode sawings), originally developed by the Canada Centre for Mineral and Energy Technology (CANMET) as certified reference materials for trace element analysis, as Cu isotope reference materials for LA-MC-ICP-MS analysis and solution nebulization (SN) of Cu. The Cu isotopic composition and homogeneity of these four materials were characterised by SN- and LA-MC-ICP-MS, and are reported for the first time. The bulk Cu isotopic compositions, expressed as δ⁶⁵Cu_SRM976 in per mil (‰) relative to NIST SRM976 with combined uncertainties (U, k = 2), of SSC-1, SSC-3 and SSC-4, determined utilizing SN-MC-ICP-MS, are identical within analytical uncertainty at 0.03 ± 0.07‰ (n = 29), 0.04 ± 0.04‰ (n = 28), and 0.05 ± 0.08‰ (n = 29), respectively; the composition of CUPD-1 is 2.14 ± 0.08‰ (n = 28). The compositions are 0.01 ± 0.07‰ (n = 29), 0.04 ± 0.06‰ (n = 29), 0.03 ± 0.06‰ (n = 28) and 2.15 ± 0.06‰ (n = 28), respectively, relative to the European Reference Material ERM®-AE633 Cu isotope standard. The Cu isotope homogeneity of the four new reference materials was assessed by determining whether multiple individual in situ Cu isotope measurements made by LA-MC-ICP-MS analysis (43 µm spot size), using each of the other three reference materials as a calibrator, approximate a single normal distribution. We also investigate whether there are statistically significant differences between the mean δ⁶⁵Cu values of three independent data sets for each of the Cu isotope reference materials using one-way analysis of variance (ANOVA). Normality tests (graphical assessment of normal distribution quantile-quantile plots, and the Shapiro-Wilk, Jarque-Bera and reduced chi-squared statistic tests) show that: 1) the Cu isotope data acquired on SSC-1, SSC-3, SSC-4 and CUPD-1 do not depart significantly from a normal distribution, 2) the scatter of the Cu isotope data is due to analytical uncertainty with 95% confidence, and 3) there are no other significant sources of scatter; e.g. heterogeneity of the reference materials. The results of one-way ANOVA reveal that the mean difference of the δ⁶⁵Cu value for each of the reference materials SSC-1, SSC-3, SSC-4 and CUPD-1 is statistically not significant at the 0.05 level. The mean δ⁶⁵Cu_SRM976 values with combined uncertainties (U, k = 2) of SSC-1, SSC-3, SSC-4 and CUPD-1, determined by LA-MC-ICP-MS using each of the other three reference materials as a calibration standard, are 0.03 ± 0.09‰ (n = 132), 0.05 ± 0.09‰ (n = 154), 0.03 ± 0.09‰ (n = 144) and 2.14 ± 0.10‰ (n = 106), respectively. These values are in agreement with those determined by SN-MC-ICP-MS analysis at the 95% confidence level and have excellent precision (2 s.d. ≤ 0.10‰). These results suggest that SSC-1, SSC-3, SSC-4 and CUPD-1 can be considered isotopically homogeneous at a spatial resolution of 43 μm, and they are suitable reference materials for calibration and quality control of in situ and solution nebulization Cu isotope analyses of Cu.

Keywords: LA-MC-ICP-MS; copper isotopes; homogeneity; native copper; reference materials.

FIGURE 1

Cu isotope determinations of certified reference material ERM®-AE647. The determined δ⁶⁵Cu values, measured against NIST SRM976, were consistent during one analytical session in February 2019, and agree within uncertainty with the certified δ⁶⁵Cu_SRM976 value and precision from Moeller et al. (2012) (as indicated by a solid black line and shaded area). The determined δ⁶⁵Cu values against ERM®-AE633 were consistent over three analytical sessions, measured from September 2018 to February 2019 and also agree well within the uncertainty with the certified δ⁶⁵Cu_AE633 values and precision from Moeller et al. (2012) (as indicated by a solid green line and shaded area). Precision of a single measurement is shown at 2 standard errors (number of measurement cycles, n = 40).

FIGURE 2

(A) SN-MC-ICP-MS Cu isotope determinations of reference material SSC-1, SSC-3, SSC-4 and CUPD-1 against NIST SRM976; (B) SN-MC-ICP-MS Cu isotope determinations of reference material SSC-1, SSC-3, SSC-4 and CUPD-1 against ERM®-AE633. Measurement precision is indicated by dashed lines. For a single measurement, the within-run precision was 0.03% (2 standard error; number of measurement cycles n = 40).

FIGURE 3

δ⁶⁵Cu_SRM976 values of reference materials SSC-1, SSC-3, SSC-4, CUPD-1 and ERM®-AE647 determined using calibration standard NIST SRM976 vs. using ERM®-AE633.

FIGURE 4

(A) Raw time-resolved ⁶⁵Cu/⁶³Cu isotope ratios (green line) and ⁶⁶Zn/⁶⁴Zn ratios (pink line) from LA-MC-ICP-MS analysis of reference material SSC-1 (DE06B22 analysis). Zn was added to the sample carrier gas as a dry aerosol. The relatively stable Cu isotope ratios indicate that substantial ablation time-dependent fractionation of Cu isotope ratios during the course of a ns laser ablation spot analysis was largely avoided by aerosol filtering. (B) Time resolved ⁶⁵Cu/⁶³Cu ratios corrected (blue line) for mass bias from LA-MC-ICP-MS analysis of reference material SSC-1 (DE06B22 analysis). The relatively stable Cu isotope ratios are entirely independent of the Cu signals (red line), and laser induced isotope fractionation was largely avoided by aerosol filtering.

FIGURE 5

Normal quantile-quantile plots for SSC-1, SSC-3, SSC-4, and CUPD1 show that the quantile data points for the determined δ⁶⁵Cu values for each of the reference materials do not seriously deviate from the 1:1 line and have linear correlation coefficients close to 1.

FIGURE 6

Results of Cu isotope ratio determinations of reference materials SSC-1 (A), SSC-3 (B), SSC-4 (C), and CUPD-1 (D) by LA-MC-ICP-MS using each of the other three reference materials as a calibration standard. For a single measurement, the within-run precision is 2 standard error. For comparison, δ⁶⁵Cu values and measurement precision obtained by SN-MC-ICP-MS is indicated by solid black lines and shaded areas.

FIGURE 7

Cu isotope compositions of SSC-1, SSC-3, SSC-4 and CUPD-1 determined by LA-MC-ICP-MS vs. SN-MC-ICP-MS. The Cu isotope data overlap well the 1:1 relationship within the uncertainty, indicating that the measured Cu isotope composition of SSC-1, SSC-3, SSC-4 and CUPD-1 by LA analysis agree with those by SN analysis within the analytical uncertainties, and that precise and accurate Cu isotope measurements by LA-MC-ICP-MS using these reference materials as calibration standards were achieved.

FIGURE 8

Cu isotope composition of native copper sample NMC 12864 determined by LA-MC-ICP-MS using reference materials SSC-1, SSC-3, SSC-4 and CUPD-1 as calibration standards.