Chemical Geology 262 (2009) 344–354 Contents lists available at ScienceDirect Chemical Geology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c h e m g e o In situ Pb-isotope analysis of pyrite by laser ablation (multi-collector and quadrupole) ICPMS Jon Woodhead a,⁎, Janet Hergt a,b, Sebastien Meffre b, Ross R. Large b, Leonid Danyushevsky b, Sarah Gilbert b a b School of Earth Sciences, The University of Melbourne, VIC 3010, Australia ARC Centre of Excellence in Ore Deposits, The University of Tasmania, TAS 7001, Australia a r t i c l e i n f o Article history: Received 18 September 2008 Received in revised form 2 February 2009 Accepted 5 February 2009 Editor: R.L. Rudnick Keywords: Sulfide Pyrite Pb-isotope In-situ analysis Laser ablation ICPMS a b s t r a c t Pb-isotope ratios, measured in the mineral pyrite, provide a valuable petrogenetic tool with widespread applicability. In order to interpret complex structural and mineralogical textures, however, a method of insitu analysis is essential. While laser ablation ICPMS is ideally suited to this task, the low melting point of sulfide, the highly variable and often high Pb contents, and the potential presence of relatively radiogenic inclusions introduce a number of analytical problems unique to pyrite Pb-isotope analysis. Here we address these issues using results obtained on two very different analytical systems based around multi-collector and quadrupole ICPMS instruments respectively. We suggest that controlled ablation of pyrite is only achieved at low laser fluence and that, under these conditions, standardisation using silicate reference materials is inappropriate and natural pyrite standards are to be preferred. The inherent variability in Pb (and sometimes U) concentrations in pyrite requires careful selection of detector systems for optimal analysis and in this regard both quadrupole and multi-collector ICPMS instruments can play important and complimentary roles. Multi-collector instruments provide higher precision analyses but detector configurations can prohibit simultaneous measurement of U, Th and Pb. Furthermore, the micrometer-scale variability in Pb concentrations can cause problems for both Faraday cup and ion counting detection systems. In contrast, quadrupole ICPMS systems allow simultaneous measurement of U, Th and Pb, and have more flexible detection systems with many orders of magnitude dynamic range but are unable to produce high precision data. Results are presented for two different analytical systems and demonstrate a very strong dependence of data quality upon signal size. In addition they allow some estimation of the limiting precision obtainable by these methods. Finally, a geological example is provided from the giant Sukhoi Log sedimentary Au deposit of Russia. © 2009 Elsevier B.V. All rights reserved. 1. Introduction The application of radiogenic isotope systems as tracers of both source and process in the field of economic geology is limited by a lack of suitable sulfide minerals (i.e. those with low parent/daughter element ratios enabling preservation of near-initial ratios over time), coupled with the often uncertain relationship between silicate host minerals, which may be more amenable to analysis, and sulfide ores. Studies of Pb-isotope compositions in galena have been and continue to be highly successful (e.g., Carignan et al., 1993; Carr et al., 1995, Ayuso et al., 2004) but the mineral pyrite also holds considerable promise as a phase which is commonly associated with many types of mineralisation and may contain moderate to high quantities of Pb and low U, providing the opportunity to determine near-initial Pb-isotope ratios. Previous Pb-isotope studies of pyrite have largely reported the use of bulk assemblages (e.g., Ho et al., 1994; Olivo et al., 2004) but the ⁎ Corresponding author. E-mail address: jdwood@unimelb.edu.au (J. Woodhead). 0009-2541/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.chemgeo.2009.02.003 typically complex textural relationships found in ore systems ultimately demand a technique with high spatial resolution while still retaining the potential for high levels of precision such that subtle variations can be identified. Two recent studies have demonstrated the feasibility of using laser ablation ICPMS technologies in this regard (Mathez and Waight, 2003; Mathez and Kent, 2007) but these were largely concerned with the application of the results to the petrogenesis of the Bushveld Complex rather optimising an analytical protocol for sulfide analysis. The laser ablation analysis of pyrite for Pb-isotope ratio determination is not without technical difficulties. In particular Pb contents are often highly variable due to micro-inclusions of Pb-rich phases, providing extreme conditions for most signal detection systems, and U-rich domains sometimes occur, potentially requiring correction for radiogenic Pb ingrowth (e.g., Fig. 1). In addition most laser ablation systems couple very efficiently with sulfides and large-scale melting is a common, but generally undesirable consequence leading to the production of large particulates and the resultant possibility for decreased sensitivity, and enhanced matrix effects (e.g. Jackson and 345 J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 Table 1 Analytical conditions for the systems used in this study. Laser Type Wavelength Pulse width Repetition rate Fluence Ablation chamber He gas flow Ar gas flow ICPMS Type RF power Shield torch Dwell time University of Melbourne University of Tasmania Lambda Physik Compex 110 193 nm 25 ns 5 Hz ~ 1 J cm− 2 Helex, low volume (2.5 cm2) New Wave solid state 213 nm 2 ns 10 Hz 1–2 J cm− 2 Custom made, low volume (3.4 cm2) 0.2 l/min 0.95 l/min 0.7 l/min 1.23 l/min Nu Plasma MC-ICPMS 1325 W No Simultaneous detection 208, 207, 206, 204, 202, 200 Agillent 4500 quad ICPMS 1380 W No 202, 204, 206, 208: 40 ms 56, 232, 238: 5 ms System sensitivitya Spot size 55 µm Repetition rate 5 Hz 238 U 50 mV (~3,125,000 cps) 139 La 7 Li 248 ThO/232Th 110 µm 10 Hz 300,000 cps 200,000 cps 150,000 cps 0.15% a Typical system sensitivities were determined at slightly higher fluence (~ 3 J cm− 2) than that used for pyrite analyses to enable efficient ablation of the NIST 612 reference material. Gunther, 2003). Finally, standardisation to correct for mass bias is an important issue requiring careful consideration. In the following text we discuss all of these issues, provide optimum schema for analysis of pyrite and document our experiences using two different analytical systems to achieve these goals. 2. Analytical instrumentation and methods Table 1 provides a summary of key instrumental parameters for the analytical systems used in this study. All analyses conducted at the University of Melbourne were performed on a Nu Plasma MC-ICPMS instrument. For solution analyses, Pb was separated on Eichrom™ SrResin and introduced to the mass spectrometer using an Aridus desolvation unit, equipped with a Glass Expansion™ OpalMist Teflon nebuliser operating at an uptake rate of ~ 30 µl min− 1 under free aspiration. Analyses were corrected for mass bias effects using the modified thallium doping technique described in Woodhead (2002), and the SRM 981 values noted therein as reference. In-situ analyses at the University of Melbourne utilised a HelEx ablation system, constructed around a Lambda Physik™ Compex 110 excimer laser operated with ArF providing an output wavelength of 193 nm. The system has been described in some detail in a number of previous publications (e.g. Eggins et al., 1998; Woodhead et al., 2004; Eggins et al., 2005; Woodhead et al., 2005). The laser output energy and power density were adjusted as described below and the laser was typically operated at a repetition rate of 5 Hz, with spot sizes ranging from 90 to 200 µm, depending upon pyrite Pb concentration. Data were collected in time-resolved mode but all data deconvolution was Fig. 1. Selected elemental maps for a pyrite grain from the Sukhoi Log sediment-hosted Au deposit (Russia), with scales in counts per second. Image size is approximately 1.3 × 1.7 mm. The maps were produced at the University of Tasmania using a laser ablation ICPMS method similar to that employed by Woodhead et al. (2007). The shape of the pyrite grain is clearly delineated in the Fe concentration plot (top). Note that the Pb concentration is quite heterogeneous and reaches some very high values (~1e+ 8 cps). It is also clear that this particular grain contains some areas with significant U content. Both these features must be accommodated in any successful analytical protocol. 346 J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 undertaken offline in the Iolite software package (Hellstrom et al., 2008) which allows for detailed visualisation of ion beam intensities versus time and integration of specific portions of the ablated signal. In this way it was possible to avoid any complications due to surface Pb contamination from sample polishing (surficial contamination at some level is almost unavoidable during polishing of sulfides). Hg interferences on the 204Pb isotope were corrected using an initial 60 s ‘on peak’ baseline followed by subtraction of any sample-derived Hg signal by monitoring the 202Hg isotope and peak-stripping. All analyses conducted at the University of Tasmania were performed on an Agilent 4500 ICPMS coupled to a New Wave 213 nm solid state laser in a custom made, low volume (3.4 cm3), barrel-shaped chamber in a He atmosphere. The laser was operated at 5 or 10 Hz using spot sizes between 15 and 110 µm. Data were collected in time-resolved mode with 30 s gas blank measurement followed by 60 s analysis time typically drilling at around ~1 µm s− 1. Data deconvolution was performed using custom made excel-based spreadsheets. Surface contamination was eliminated by pre-ablating all spots at the same spot size as used for analysis but a lower repetition rate (typically 5 s at 1 Hz). Hg interferences were stripped using the same techniques as described above for the multicollector instrument. Where necessary, corrections for radiogenic Pb ingrowth resulting from in situ decay of U and Th within the pyrite or included minerals were undertaken using measured U/Pb, Th/Pb and Pbisotope ratios and the known or estimated age of the deposit in question. 3. Some observations regarding power density When compared to many silicates, sulfides have relatively low melting points (e.g. that of pyrite is ~1180 °C; Hurlbut et al., 1985). As a consequence it is frequently observed in the laser ablation of sulfides that extensive melting occurs with significant melt and condensate accumulation around the ablation site: this is most readily observed as a dark halo around the ablation site in reflected light but is also apparent on SEM images (Fig. 2a and b). These observations clearly demonstrate that, under such conditions, there is considerable production of melt droplets often of a relatively large (up to micrometer) size. Experience with silicate and metal laser ablation strongly suggests that this is an undesirable trait since sample accumulation around the ablation pit will not only result in decreased sensitivity (as less sample is delivered to the plasma) but also any large melt droplets that do enter the plasma are likely to induce matrix effects and promote isotopic fractionation (e.g. Jackson and Gunther, 2003). Although such effects may be less noticeable when conducting relatively low-precision quadrupole ICPMS measurements, in order to obtain the highest precision and accuracy isotope ratios by MC-ICPMS in particular it is clearly preferable to minimise any such phenomena. To this end we performed a variety of ablation experiments by varying the laser power density on the sample and observing the effects using optical and SEM imaging. Similar experiments were performed on both the excimer Ar–F gas (193 nm) laser at Melbourne and the solid state Nd:YAG (213 nm) laser at the University of Tasmania. On the excimer laser by using a low laser output energy, in combination with a 25% transmission beam splitter we were able to reduce melting effects almost entirely while still maintaining controlled ablation. For typical silicate analysis the Melbourne excimer laser system is generally operated at a fluence of ~ 5 J cm− 2 or slightly less whereas we estimate the ideal situation for sulfide ablation to be in the region of ~ 1 J cm− 2. Using these conditions, melt/condensate droplets external to the ablation site are greatly reduced, and thus sample transport to the ICPMS and system sensitivity are optimised. Intriguingly, during these experiments, we observed considerable structure in the ablation pits themselves; in particular, a variety of conelike structures typically appear (Fig. 2c to f). Although not widely reported, almost identical structures have been observed previously during the ablation of both polyamides and organic crystals (e.g. Dyer et al., 1986) and have been termed ‘trulli’ after a form of traditional dwelling with a cone-shaped roof from Apulia, Italy (Kampmeier et al., 1997). Little is known of their origin but it is thought that small imperfections in the sample surface may initially act to shield the underlying sample from the incoming laser pulses and in this way surface irregularities begin to develop. In materials that are highly reflective further laser pulses tend to exacerbate such surface imperfections since glancing impacts on the sides of the cone are insufficient to ablate the material. Certainly the enhanced ablation we observe around the base of such structures, producing ‘moats’ lends some weight to this hypothesis. We have noticed, on occasion, however, that trulli may extend beyond the sample surface (e.g. Fig. 2f) which suggests that they may, under some circumstances, be partly constructional in origin, presumably by adhesion of condensate from the ablation plume. Although the investigation of the exact origin of this unusual phenomenon is beyond the scope of this study, it is universally agreed in the existing literature that such features are only present at values of laser fluence just above the ablation threshold (Dyer et al., 1986; Kampmeier et al., 1997). As a result, the appearance of trulli, and lack of significant ‘blackening’ around the ablation pit (the latter when viewed in reflected light) can both be used to assess optimal levels of laser power density for sulfide ablation. In this way efficient transport to the mass spectrometer can be readily achieved. These findings are entirely consistent with the recent work of Wohlgemuth-Ueberwasser et al. (2007) who note that fluence above 3 J cm− 2 produces more melting and promotes elemental fractionation in sulfides and Jackson and Gunther (2003) who suggest that higher fluence promotes incomplete vaporisation and ionisation of large particles in the plasma. Trulli-like structures were also observed using the University of Tasmania system. In this case we estimate that optimum fluence for sulfide ablation will be in the range 1–2 J cm− 2. At low fluence the count rates increase rapidly with small changes in fluence so that analyses are strongly affected by small fluctuations in the laser power and at high fluence, deposition and blackening occurs around the ablation site, suggesting that melting, and deposition may be occurring. Optimum ablation conditions were therefore chosen at the lowest point where large increases in fluence make relatively small differences to the count rates. 4. Standardisation and data quality There has been much debate in the literature as to the viability of the commonly used NIST glasses for calibration of samples with variable matrices during LA-ICPMS. In the case of sulfides, whereas some workers have used NIST glasses for external calibration of sulfide ablation data (e.g. Halter et al., 2004; Sylvester, 2008) there is also a broad view that, in order to obtain the most accurate elemental data, some form of matrix-matched material is to be preferred and, indeed a number of attempts have been made to produce sulfide reference materials (e.g., Wilson et al., 2002; Wohlgemuth-Ueberwasser et al., 2007; Danyushevsky et al., in press). While the corresponding case for matrix effects during isotopic analysis by laser ablation is less welldeveloped (e.g., see Norman et al., 2006 for an example), we have argued above that optimal ablation of sulfide materials is best performed under conditions of low laser fluence and, under these circumstances, calibration with a reference material of radically different matrix is not a viable option. For sulfide ablation we reduce the laser power density by a factor of around x5 compared to our normal ablation conditions. Using these laser output energies it is simply not possible to ablate NIST glass efficiently since it is close to or below its ablation threshold. Rather than attempt to produce a synthetic reference material, we chose instead to characterise a number of natural sulfides for use as potential standards, an approach which proved relatively straightforward. As a result of this work two pyrites were identified as likely candidates for J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 laser ablation studies: 700380, is a nodular pyrite from drill-hole 208 (depth 602.7 m) from the Jaguar deposit of western Australia and 110028 is a fine-grained massive sulfide from the CSA Mine in Cobar, New South Wales. Jaguar is an Archean volcanic hosted massive sulfide deposit located 4 km south of the Teutonic Mine in central Western Australia at 28.40 S, 121.15E (Barley, 1992). The pyrite nodule (2 cm in diameter) was initially chosen because it has a high Pb concentration but this also proved to be a perfect primary calibration standard with both homogeneous Pbisotope composition and relatively homogeneous Pb content (1000– 4000 ppm: Table 2), producing large and constant ion beams under most ablation conditions. The CSA mine at Cobar is a Devonian Cu–Pb–Zn deposit in central New South Wales, located at 31.41S, 148.80E (Giles and 347 Marshall, 2004). This pyrite sample was chosen because its Pb isotopic composition is both homogeneous and contrasts with the Jaguar Deposit pyrite but, in terms of Pb concentration, it is much more variable (0.2– 2000 ppm: Table 2). As a result it produces analyses more typical of ‘realworld’ pyrites and therefore can be used as a robust secondary standard. Tables 2 and 3 contain representative trace element analyses from the University of Tasmania quadrupole ICPMS and the solution based Pbisotope analyses of both materials using the Melbourne MC-ICPMS system, respectively. Fig. 3a shows the raw laser ablation MC-ICPMS Pb-isotope data (corrected for Hg interference on 204Pb but not mass bias) obtained on the 700380 pyrite—our current primary calibration material—for eight Fig. 2. Optimisation of the Melbourne laser ablation system. SEM images of the ablation process in pyrites using the 193 nm excimer laser. All ablation pits are ~ 100 μm in diamater. (a) and (b) demonstrate the results of sulfide ablation using laser fluence values appropriate for silicate analysis. This results in substantial melting and production of a large dark area around the ablation pit, often visible in reflected light. Closer inspection reveals this area to be littered with melt droplets often approaching 1 μm in size. (c) and (d) show the effects of lowering the laser fluence to just above the sulfide ablation threshold. Here melt production is greatly reduced, with far less droplet accumulation surrounding the pit, resulting in higher sensitivity (note that none of these surfaces were cleaned prior to SEM study). Ablation under these conditions is also characterised by the appearance of ‘trulli’ structures within the ablation pit itself. (e) and (f) demonstrate that, in some cases, trulli appear to extend beyond the rim of the ablation pit, leading to the conclusion that, at least in part, they may be constructional in nature. The fact that significant ‘moats’ appear around the base of individual trulli is, however, certainly consistent with the conventional interpretation of their formation. See text for discussion. 348 Table 2 Laser ablation-ICPMS trace element data (ppm), normalised to Fe. Source Cr Mn Co Ni Cu 700380 Jaguar deposit, western Australia Ti 14 25 17 5.2 4.8 6.2 4.5 4.4 6.4 4.9 5.9 7.1 5.9 0.8 1.0 2.7 b 0.61 b 2.22 b 2.80 b 2.15 b 2.06 b 2.35 b 2.59 b 2.55 b 2.50 b 3.13 5 7 12 3 3 3 3 3 3 4 4 3 3 261 178 167 109 44 111 423 381 371 375 355 314 339 179 177 164 220 151 158 165 169 195 185 166 157 181 19,482 13,546 18,048 8851 445 2428 2143 416 38,201 3410 4387 2845 11,656 110028 CSA Mine, Cobar, New South Wales 7.9 5.9 6.2 5.3 5.5 8.6 4.7 14 298 7.6 b 2.43 b 2.34 b 2.19 b 2.49 b 2.30 b 2.50 b 2.20 b 1.8 b 1.1 b 1.9 12 3 2 15 b0.8 4 3 3 33 4 33 1.2 75 4.4 0.5 0.9 22 6.3 34 0.9 Analytical methods after Danyushevsky et al. (in press). 1.0 3.2 3.9 3.2 b 0.9 1.9 1.4 1.4 1.3 1.4 Zn 5.8 3.2 3.2 57 1.2 5.3 3.5 83 61 3.9 As Se Zr Mo Ag Cd 285 2488 296 85 20 275 61 11 58 88 213 35 145 1624 1416 1449 1548 1230 1511 1833 1690 1722 1668 1693 1603 1726 17 18 20 22 8.8 25 18 13 22 27 16 20 20 0.3 0.8 2.8 0.01 0.1 0.2 b 0.05 0.1 b 0.04 b 0.04 b 0.04 0.1 b 0.05 0.1 0.2 0.2 b0.1 b0.4 b0.4 b0.3 b0.3 b0.5 b0.4 b0.5 b0.4 b0.5 109 96 85 115 40 92 128 76 114 101 92 79 141 1.3 9.1 1.5 0.8 b 1.0 b 1.6 b 1.2 b 1.5 b 1.6 b 1.7 b 1.1 b 1.3 1.3 4 b 0.7 b1 14,419 4 b 1.2 724 301 4229 15 175 211 112 1711 33 237 141 302 1052 94 42 b4 17 b5 6.9 5.8 7.7 6.2 14 9.0 1.5 0.2 0.9 b 0.06 b 0.05 0.8 3.2 1.1 1.8 2.2 b0.6 b0.4 b0.5 b0.5 b0.4 b0.4 b0.4 b0.2 b0.1 b0.1 b 0.1 b 0.1 b 0.1 7 b 0.1 0.1 b 0.1 0.3 6.1 0.1 b 1.3 b 1.5 b 1.2 20 b 1.2 b 1.3 b 1.3 0.5 7.3 b 0.1 Sn Sb 48 212 113 48 15 76 45 14 26 47 87 39 33 1065 1022 987 1126 674 953 1183 1005 1134 1120 1119 1002 1115 0.2 2.2 0.1 2.2 b 0.1 0.2 b 0.1 2.6 1.7 0.3 8.2 16 3.9 66 b 0.08 9.7 3.9 6.6 37 4.0 Te 3.2 3.1 2.7 4.0 1.9 3.9 6.5 2.2 3.1 3.7 2.7 5.5 3.3 b1.5 b1.1 b0.8 b1.2 b1.0 b1.2 b1.1 b0.1 0.1 0.1 Ba La W Au 3.8 6.2 4.6 2.8 1.3 3.3 2.5 1.9 1.7 2.8 4.4 2.2 2.2 0.06 0.14 0.21 0.03 0.07 0.06 0.04 b 0.01 b 0.01 0.02 0.09 0.04 0.02 0.2 0.1 0.1 0.2 0.1 0.2 0.2 0.1 0.2 0.2 0.2 0.2 0.1 5.1 4.9 4.0 5.7 3.4 4.6 6.0 4.4 5.2 5.0 4.1 4.6 4.9 b0.16 b0.12 b0.13 b0.09 b0.11 b0.15 b0.16 b0.04 b0.1 b0.1 b 0.01 b 0.01 b 0.02 b 0.02 b 0.01 b 0.01 b 0.01 0.01 0.27 0.18 b0.13 b0.08 b0.08 b0.08 b0.04 b0.06 b0.05 b0.02 1.2 b0.02 b 0.1 0.1 b 0.07 1.1 b 0.1 0.1 b 0.05 0.2 1.0 0.01 Tl Pb Bi 33 25 35 24 31 30 23 32 29 28 29 33 28 2570 2902 2720 2537 1141 3389 2528 1432 3091 3850 1500 2940 2835 189 206 175 229 130 195 219 179 210 209 181 192 202 0.1 0.2 b0.03 0.1 b0.04 0.7 0.0 0.1 10.5 0.1 143 215 83 2070 0.2 137 67 117 785 82 0.04 b0.02 b0.02 0.21 b0.02 b0.03 0.04 1.4 9.3 0.03 Th U 0.01 0.07 0.06 0.01 b 0.01 b 0.01 b 0.01 b 0.01 b 0.01 0.02 b 0.01 b 0.02 b 0.01 0.01 0.02 0.05 0.01 b0.01 b0.02 b0.01 b0.01 b0.01 b0.01 b0.01 b0.01 b0.02 0.11 b 0.01 0.04 b 0.02 b 0.01 0.08 0.19 0.05 0.11 0.25 0.11 0.01 0.03 b0.01 b0.01 0.10 0.09 0.03 0.04 0.09 J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 Sample 349 J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 Table 3 Solution MC-ICPMS Pb-isotope ratios for our two reference pyrites, corrected for mass bias using Tl-normalisation methods as documented in Woodhead (2002). Pb/206Pb Sample 700380 110028 Aliquot 1 Aliquot 2 Aliquot 3 Mean and standard deviation Aliquot 1 Aliquot 2 Aliquot 3 Mean and standard deviation 207 2 s.e. 208 Pb/206Pb 1.087757 1.087830 1.087857 1.087815 0.862775 0.862894 0.862853 0.862841 0.000014 0.000014 0.000014 0.000052 0.000009 0.000010 0.000009 0.000060 2.483143 2.483347 2.483453 2.483315 2.112382 2.112713 2.112636 2.112577 analytical sessions conducted over a period of 2 years. Although mean values for each session can differ considerably as a result of variable mass spectrometer parameters (e.g. cleanliness of cones, torch position, gas flows) the data within any one analytical session show a remarkable degree of consistency with levels of reproducibility comparable to typical thermal ionisation mass spectrometry, which is often considered the benchmark analytical method for Pb-isotope 2 s.e. 206 Pb/204Pb 2 s.e. 207 Pb/204Pb 2 s.e. 208 Pb/204Pb 0.000025 0.000037 0.000035 0.000157 0.000023 0.000026 0.000022 0.000173 13.3589 13.3606 13.3598 13.3598 18.1165 18.1160 18.1150 18.1158 0.0006 0.0007 0.0008 0.0009 0.0007 0.0006 0.0009 0.0007 14.5310 14.5340 14.5336 14.5329 15.6305 15.6324 15.6307 15.6312 0.0007 0.0007 0.0009 0.0017 0.0006 0.0005 0.0008 0.0011 33.1722 33.1793 33.1792 33.1769 38.2689 38.2747 38.2707 38.2715 2 s.e. 0.0015 0.0016 0.0019 0.0041 0.0015 0.0012 0.0018 0.0030 analysis (illustrated in the figure). We interpret this high level of consistency as a direct result of our attempts to achieve stable and reproducible laser ablation conditions (see above) although we also readily admit that we have not conducted any extensive experiments under conditions of higher fluence with which to compare these data simply because we view this as being potentially detrimental to our ICPMS detection systems. Certainly, however, there is evidence in the Fig. 3. (a) Raw (corrected for Hg interference but not mass bias) Pb-isotope ratios obtained for the 700380 pyrite during 8 laser ablation MC-ICPMS analytical sessions conducted over a period of 2 years. Separate sessions are distinguished using alternate open and closed symbols, and separated by vertical hatched lines. Numbers represent percent standard deviation for any particular session. Although there is a significant range in values depending upon specific ICPMS conditions on any given occasion, variation within any given analytical session is small and comparable to what might be observed during conventional TIMS Pb-isotope analysis (conventional TIMS and double spike-TIMS/solution MC-ICPMS uncertainties provided for comparison). (b) Raw (corrected for Hg interference but not mass bias or drift) Pb-isotope ratios obtained for the 700380 pyrite during 16 quadrupoleICPMS analytical sessions, covering a similar time period. Note different scale. The precision of individual measurements is ~ 10 times larger than on the multi-collector instrument but there nevertheless remains a high degree of internal consistency within individual analytical sessions. 350 J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 literature for variable isotopic fractionation during ablation of copper with different laser pulse energies (Jackson and Gunther, 2003) and thus we prefer to avoid conditions of extensive sample melting. Fig. 3b illustrates 700380 data for the quadrupole instrument analysed over a similar period. These display within-run errors ~ 10 times greater than the MC-ICPMS data as expected for a single-collector instrument but, nevertheless, within any one analytical session, data exhibit a high degree of consistency. In our current methodology, analyses of this reference pyrite throughout an analytical session are then used to determine mass bias and hence make a so called ‘external’ correction for this effect during processing of unknowns, using either an average value for the entire analytical session (usually the case) or a calibration line versus time if fractionation factors have drifted significantly during the session (which we rarely observe). Using this procedure, corrected isotope ratios are obtained and, as a result, these within-session variations in the analysis of the calibration material represent a first-order control on the quality of unknown analyses. As an illustration of this methodology using the MCICPMS instrument, Fig. 4 shows data for our secondary standard, which closely approaches many pyrites in terms of Pb content and variability. These data, now corrected for both Hg interference and mass bias, cover the same analytical period as the calibration standards shown in Fig. 3a. Levels of accuracy,1 that is the degree to which the data approach the assumed true values (in this case the Tl-normalised bulk-pyrite solution MC-ICPMS data of Table 3) are typically around ~1500 ppm for ratios involving 204Pb, dropping to ~300 ppm for 208Pb/206Pb and ~900 ppm for 207 Pb/206Pb ratios. Throughout the period over which the work described here was conducted, these fractionation and drift-corrected Pb-isotope analyses show a relative standard deviation of 0.13 and 0.15% for 206Pb/ 204 Pb and 207Pb/204Pb ratios and 0.09% and 0.03% for 207Pb/206Pb and 208 Pb/206Pb ratios respectively. Such reproducibility is only a factor of ~2– 3 larger that of conventional (i.e. not double or triple spiked) Pb isotope analysis of bulk samples by thermal ionisation mass spectrometry, which is remarkable considering the minimal sample preparation involved and small amount of sample material consumed. For most pyrite analyses, within-run analytical precision is highly dependent upon Pb content. Fig. 5 shows typical 2 s.e. within-run error estimates for analyses obtained using both quadrupole and MCICPMS instruments across a wide range of signal sizes. No attempt is made here to compare in detail the sensitivities of the two systems since both were coupled to quite different lasers and thus there are too many variables involved to make any meaningful comparison (although an approximate indication of relative sensitivities for silicate analysis can be obtained from Table 1). These diagrams illustrate the very strong relationship between signal size and withinrun precision which is common to both analytical systems. In addition they provide an estimate of the limiting precision of each technique, given maximum signal size. For MC-ICPMS, this is around 0.005 (2 s. e.) for the 206Pb/204Pb ratio and 0.001 for the 207Pb/206Pb ratios whereas, for quadrupole ICPMS, these figures are ~0.2 and ~0.005 respectively. For determination of the 207Pb/206Pb ratio the MC-ICPMS limiting precision is thus ~ 10 times better than the quadrupole data while for the 206Pb/204Pb ratio this is a factor of ~ 40, reflecting the difficulties inherent in measurement of the 204Pb signal and accurate correction for 204Hg interference using a less sensitive single collector instrument with sequential rather than simultaneous signal detection. We do note, however, that when using quadrupole (and indeed MC-) ICPMS it is possible to improve these figures by pooling the results of multiple analyses. There is an obvious loss of spatial resolution using 1 Accuracy is defined here as the difference between measured and fractionation corrected ratios for this pyrite and the assumed true ratio as defined by solution analysis. For a group of analyses, accuracy (usually expressed in ppm) can be calculated as A = [ΣA2i /(N − 1)]1/2 where Ai = 1,000,000 × (Ri′ − Rtr)/Rtr, R′ is the measured and corrected ratio for a standard, and Rtr is the true ratio. After Platzner et al. (1997), page 187. such a method; however, if different pyrite generations are clearly visible optically, this technique can improve precision considerably. The pooling of the analyses also allows possible outliers to be identified and can be performed quickly due to the rapidity of the technique (1–2 min per analysis) but, of course, must be undertaken with a degree of caution in terms of the interpretation of the resulting ‘average’ analyses. 5. Optimal analytical strategies and an example from Sukhoi Log Based upon our observations in analysing many thousands of grains, it is clear that there is no one ideal analytical strategy for in-situ Pb-isotope determination of pyrites. Published (e.g. Large et al., 2007) and unpublished laser ablation data obtained at the University of Tasmania from a variety of ore deposit types suggest that pyrites can be highly variable in Pb content. Although most ore deposits contain pyrite with around 100 ppm Pb, VHMS, SEDEX, lode gold, IOCG and epithermal Au deposits may be more enriched (50–10,000 ppm) and porphyry Cu–Au and skarn systems less so (0.1–50 ppm). A rather more limited dataset of analyses for pyrite not associated with ore deposits suggests that somewhat lower Pb concentrations (0– 10 ppm) prevail in these types of sample, although diagenetic pyrite in black shales can contain Pb up to 500 ppm. Studies of metamorphosed ore deposits generally show that the paragenetically early pyrite, with colloform, nodular or spongy textures, contains more Pb than later recrystallised euhedral pyrite. In addition it is frequently observed that individual pyrite crystals can show Pb concentrations which vary by several orders of magnitude on the scale of micrometers, largely due to the presence of microscopic galena inclusions, as noted above (see Fig. 1, and Table 2). Together, the high and inherent variability in Pb concentrations impose severe constraints upon any analytical method. In our studies we have considered two different analytical approaches using multi-collector and quadrupole ICPMS instruments respectively; each has its advantages and disadvantages. Ultimately MC-ICPMS instruments are capable of producing more precise data than quadrupole mass spectrometers, due to flat-topped peaks, simultaneous detection of all relevant ion beams eliminating noise associated with so-called ‘plasma flicker,’ and the generally higher sensitivity of sector-based instruments. Despite this, the extremely variable and often unstable signals observed during pyrite ablation make the use of multi ion counting an unattractive proposition with the real possibility of damage to the detectors. In contrast, Faraday cup detection systems can potentially cope with relatively large signals (10–50 V per channel on the commonly available mass spectrometers) but have a much reduced dynamic range compared to the electron multiplier units employed on quadrupole-ICP instruments and, in addition, can have problems with rapidly changing beam sizes if socalled tau corrections are not correctly implemented. Tau effects result from the relatively slow response of the Faraday cup preamplifiers to changes in signal intensity (which can be a few seconds) compared to ion counters. Although algorithms are usually incorporated into the instrument software to calibrate this response, problems sometimes occur with the most typical symptom being an apparent change in isotope ratio correlated with rapid change in signal intensity (e.g. Hirata et al., 2003; Paul et al., 2005). The veracity of the tau correction is easily monitored by measuring an isotope pair simultaneously in a Faraday cup-ion counter configuration and opening/closing the laser shutter, thereby simulating rapidly changing signal intensity. Measured isotope ratios should remain constant throughout this process. In contrast to multi-collector instruments, quadrupole ICP systems are incapable of such high precision and the accuracy of measurements can be compromised when detectors switch mode (e.g. from pulse to analog detection), or if they become saturated with high count-rates. The likely occurrence of such scenarios is exacerbated by the rapidly changing signals often observed during sulfide ablation. Nevertheless, the use of quadrupole ICPMS instruments does confer a J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 351 Fig. 4. Pb isotope ratios for the 110028 pyrite, obtained using the multi-collector instrument, and corrected for Hg interference on 204Pb and mass bias. Levels of accuracy and precision are noted on each panel, while the dashed line represents the ratio determined by high-precision solution MC-ICPMS. Within-run errors are highly variable and strongly controlled by signal size (see later section). 352 J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 Fig. 5. Typical Pb-isotope analyses by MC- and quadrupole-ICPMS (most analyses of 30–60 s duration) comparing within-run precision (quoted as 2 s.e.) relative to total ion beam intensity (note different scales on both axes). All plots show the strong dependence of within-run precision on signal size, although those of the multi-collector instrument attain lower absolute values for 2 s.e. (see arrows for estimate) at high signal size. number of advantages. In particular, their detection systems are optimised for use over a very large dynamic range (typically 8–9 orders of magnitude) and thus have the ability to analyse minerals with a wide variety of Pb contents. Furthermore, as noted above, pyrites can contain significant contents of U, which can be attributed to micro-inclusions of other minerals such as zircon, monazite and uraninite (Fig. 1). In such cases it may be necessary to apply an age correction to the Pb-isotope data obtained and this is not possible unless U, Th, and Pb contents are measured concurrently with Pbisotope ratios. Although some MC-ICPMS instruments do have specialized collector blocks which are designed for static multicollection of U–Th–Pb isotope ratios (e.g., Simonetti et al., 2005) these are generally inappropriate for pyrite ablation studies since they incorporate ion counter detection for some or all of these ion beams. Clearly in these cases the advantages of simultaneous U, Th, and Pb determination by quadrupole ICPMS outweigh the disadvantages of lower precision on isotope ratio determination. Thus, one might consider the following strategy for optimal analysis. Under circumstances in which large (N200 μm) grains are available, with moderate Pb contents (hundreds to thousands of ppm) MC-ICPMS will usually produce high-quality data comparable to TIMS or solution-based MC-ICPMS analyses. In some circum- stances, however, it will be more advantageous to utilise quadrupoleICPMS analysis, for example in samples where very large ranges in Pb content are to be expected, or where Pb contents are very low (tens of ppm) and so the Faraday collectors employed on MC-instruments are inappropriate detection systems. In addition some pyrites will have significant uranium concentrations and thus require an age correction to the Pb-isotope data, a feature which can only be assessed with reconnaissance trace element analyses or elemental mapping techniques (e.g., Fig. 1). In this case the availability of simultaneously determined U, Th, and Pb concentration data will prove invaluable, even at the expense of relatively poorer precision on the isotope ratio determination. In rare cases pyrite will prove intractable to both MC- or quadrupole-ICPMS analysis, most often as a result of extremely low Pb concentrations, possibly in conjunction with significant levels of sample Hg. In such cases we would advise the use of a micro-drilling technique coupled with low-blank Pb chemistry and analysis using high-precision TIMS or solution MCICPMS methods. As an example of the utility of pyrite Pb-isotope analysis we may consider the analyses of pyrites from the giant Sukhoi Log sedimentary Au deposit in Russia. Petrogenetic interpretations of this dataset have been published elsewhere (Meffre et al., 2008) but here we J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 simply consider the data themselves. Sukhoi Log is one of Russia's largest gold deposits and contains textural, structural and chemical evidence for multiple geological processes including diagenesis, metamorphism and post-metamorphic influences. Mineralisation is intimately associated with multiple pyrite generations and Large et al. (2007) have documented the paragenesis and trace element geochemistry of these different pyrite phases. These authors have identified an early stratiform arsenian pyrite containing structurallybound (invisible) gold from 0.5 to 12 ppm Au, and later coarsergrained pyrites relatively depleted in invisible gold, formed through late diagenesis and metamorphism. Fig. 6 shows the results of our combined quadrupole ICPMS and MC-ICPMS analyses of the Sukhoi Log pyrites and illustrates many of the pertinent features discussed above. Firstly we note that analyses from both multi-collector and quadrupole instruments are highly consistent, giving confidence in the analytical protocols at both the University of Melbourne and CODES laboratory at the University of Tasmania. Levels of precision approaching conventional TIMS analyses can be obtained from single spots on the MC-ICPMS. Similar levels of precision on quadrupole ICPMS analyses can be achieved by pooling multiple analyses, as is the case in this study. Clear distinctions in isotopic composition are noted between different pyrite morphologies and generations. These variations are unrelated to radiogenic ingrowth since the pyrites analysed in this case contain very little U and hence are believed to represent primary compositions. The Pb isotope data, when viewed in the context of mineralogical texture, provide compelling evidence for a protracted history of pyrite formation, as proposed by Large et al. (2007), involving multiple sources of Pb throughout the history of the Sukhoi Log deposit. Combining these data with U–Th–Pb geochronology of zircon and monazite from the deposit, Meffre et al. (2008) were able to define a pre-deformation phase of Au-bearing pyrite growth, formation of Pb-rich pyrite during sedimentation or early diagenesis, and Pb bearing pyrite with free gold inclusions, throughout late Neoproterozoic deformation and subsequent metamorphic and hydrothermal events in the Palaeozoic. Early stratiform pyrites contain Pb with a rather distinctive signature compared to the rest of the deposit which Meffre et al. (2008) suggest 353 hints at an old organic rich shale source. Elucidation of the details of these processes would not have been possible without the ability to perform in-situ analyses and thus investigate complex textural and structural relationships between successive multiple pyrite generations. For further petrogenetic and geochronological interpretation readers are referred to Large et al. (2007) and Meffre et al. (2008). 6. Conclusions In-situ Pb isotope analysis of pyrite by laser ablation ICPMS can be a valuable tool in the investigation of the often protracted and complex paragenetic processes occurring in mineralised systems. Such analyses present some unique analytical challenges, however. Laser ablation protocols should ideally be optimised to produce the least possible sample melting during ablation, typically by lowering laser power densities considerably compared to those used for silicate ablation. Under such circumstances intriguing textures are observed in the ablation pits, which have been termed trulli by some authors. Although their exact origin is unclear, trulli provide a useful indication of laser fluence just above the ablation threshold and which we believe is ideal for pyrite analysis. Using natural pyrite reference materials it is possible to obtain insitu Pb-isotope data of a quality approaching conventional TIMS data, based upon bulk analyses. Within-run precision is strongly dependent upon signal size, with multi-collector instruments providing better precision than quadrupole instruments. Natural pyrites are, however, characterised by highly variable and often high Pb contents and, in some cases, significant U contents: optimal Pb isotope analysis requires careful attention to these characteristics. Under ideal circumstances, where pyrites contain hundreds to thousands of ppm Pb, MC-ICPMS analyses provide highly precise and accurate data. Quadrupole ICPMS data can be utilised where Pb concentrations are inappropriate for Faraday cup detection or where age corrections for radiogenic ingrowth may be required. In such cases, pooling of data from multiple spot analyses can help to improve analytical precision markedly but this ‘averaging’ of data points must always be undertaken with care. Fig. 6. A 208Pb/204Pb vs. 206Pb/204Pb plot for pyrites from the Sukhoi Log stratiform Au deposit of Russian (data replotted after Meffre et al., 2008 to distinguish quadrupole, open symbols, from MC-ICPMS, closed symbol, data). Errors shown are 95% confidence. Quadrupole analyses represent 4 to 6 pooled analyses. Four different pyrite types, distinguished on textural and temporal characteristics show distinct isotopic signatures. Each pyrite type is indicated by a different symbol and shaded field; there is also a strong suggestion in the data (not explored by Meffre et al., 2008), that the large euhedral pyrites may contain two distinct populations (separated in this figure by dashed line). Note that these data are plotted uncorrected for any radiogenic ingrowth since the magnitude of this correction at Sukhoi Log is mostly small (b 0.2 difference in the 206Pb/204Pb ratio) compared to the isotopic differences observed between the various pyrite types (see Meffre et al., 2008 for further analytical details and the text for discussion. 354 J. Woodhead et al. / Chemical Geology 262 (2009) 344–354 Acknowledgements Mike Shelley is thanked for his insights into the origins of trulli structures and Chad Paton for programming assistance with the Iolite code. We thank Adam Kent and one anonymous reviewer for the detailed comments which have helped to focus our thoughts and improve the manuscript significantly. This work was funded by the ARC CODES Centre of Excellence in Ore Deposits. References Ayuso, R.A., Kelley, K.D., Leach, D.L., Young, L.E., Slack, J.F., Wandless, J.F., Lyon, A.M., Dillingham, J.L., 2004. Origin of the Red Dog Zn–Pb–Ag Deposits, Brooks Range, Alaska: evidence from regional Pb and Sr sources. Economic Geology 99, 1533–1553. Barley, M.E., 1992. A review of Archean volcanic-hosted massive sulfide and sulfate mineralization in Western Australia. Economic Geology 87 (3), 855–872. Carignan, J., Machado, N., Gariepy, C., 1993. Pb isotope composition of Ni–Cu–Pb ore deposits in an Archean greenstone bet; evidence for Proterozoic remobilisation in the Pontiac Subprovince of Canada. Economic Geology 88, 709–715. Carr, G.R., Dean, J.A., Suppel, D.W., Heithersay, P.S., 1995. Precise lead isotope fingerprinting of hydrothermal activity associated with Ordovician to Carboniferous metallogenic events in the Lachlan Fold Belt of New South Wales. Economic Geology 90, 1467–1505. Danyushevsky, L., Robinson, P., Gilbert, S., Norman, M., Large, R., McGoldrick, P., Shelley, M., in press. A technique for routine quantitative multi-element analysis of sulphide minerals by laser ablation ICP-MS geochemistry, exploration, environment, analysis. Dyer, P.E., Jenkins, S.D., Sidhu, J., 1986. Development and origin of conical structures on XeCl laser ablated polyimide. Applied Physics Letters 49, 453–455. Eggins, S.M., Kinsley, L.P.J., Shelley, J.M.G., 1998. Deposition and element fractionation processes during atmospheric pressure laser sampling for analysis by ICP-MS. Applied Surface Science 127–129, 278–286. Eggins, S.M., Grun, R., MczCulloch, M.T., Pike, A.W.G., Chappell, J., Kinsley, L., Mortimer, G., Shelley, M., Murray-Wallace, C.V., Spotl, C., Taylor, L., 2005. In situ U-series dating by laser ablation multi-collector ICPMS: new prospects for Quaternary geochronology. Quaternary Science Reviews 24, 2523–2538. Giles, A.D., Marshall, B., 2004. Genetic significance of fluid inclusions in the CSA Cu–Pb– Zn deposit, Cobar, Australia. Ore Geology Reviews 24 (3–4), 241–266. Halter, W.E., Pettke, T., Heinrich, C.A., 2004. Laser-ablation ICP-MS analysis of silicate and sulfide melt inclusions in an andesitic complex I: analytical approach and data evaluation. Contributions to Mineralogy and Petrology 147, 385–396. Hellstrom, J., Paton, C., Woodhead, J., Hergt, J., 2008. Iolite: software for spatially resolved LA-(quadrupole and MC) ICPMS analysis. Mineralogical Association of Canada short course series 40, 343–348. Hirata, T., Hayano, Y., Ohno, T., 2003. Improvements in precision of isotopic ratio measurements using laser ablation-multiple collector-ICP-mass spectrometry: reduction of changes in measured isotopic ratios. Journal of Analytical Atomic Spectrometry 18, 1283–1288. Ho, S.E., McNaughton, N.J., Groves, D.I., 1994. Criteria for determining initial lead isotopic compositions of pyrite in Archaean lode–gold deposits: a case study at Victory, Kambalda, Western Australia. Chemical Geology 111, 57–84. Hurlbut, C.S., Klein, Cornelis, 1985. Manual of Mineralogy, 20th Ed. John Wiley and Sons, New York, pp. 285–286. Jackson, S.E., Gunther, D., 2003. The nature and sources of laser induced isotopic fractionation in laser ablation multicollector inductively coupled mass spectrometry. Journal of Analytical Atomic Spectrometry 18, 205–212. Kampmeier, J., Dreisewerd, K., Schurenberg, M., Strupat, K., 1997. Investigations of 2,5DHB and succinic acid as matrices for IR and UV MALDI. Part 1. UV and IR laser ablation in the MALDI process. International Journal of Mass Spectrometry and Ion Processes 169/170, 31–41. Large, R.R., Maslennikov, V.V., Robert, F., Danyushevsky, L.V., Scott, R.J., Chang, Z., 2007. Multi-stage sedimentary and metamorphic origin of pyrite and gold in the giant Sukhoi Log deposit, Lena Goldfield, Russia. Economic Geology 102, 1233–1267. Mathez, E.A., Waight, T.E., 2003. Lead isotopic disequilibrium between sulfide and plagioclase in the Bushveld Complex and the chemical evolution of large layered intrusions. Geochimica et Cosmochimica Acta 67, 1875–1888. Mathez, E.A., Kent, A.J.R., 2007. Variable initial Pb isotopic compositions of rocks associated with the UG2 chromitite, eastern Bushveld Complex. Geochimica et Cosmochimica Acta 71, 5514–5527. Meffre, S., Large, R.R., Scott, R., Woodhead, J., Chang, Z., Gilbert, S.E., Danyushevsky, L.V., Maslennikov, V., Hergt, J.M., 2008. Age and pyrite Pb-isotopic composition of the giant Sukhoi Log sediment-hosted gold deposit, Russia. Geochimica et Cosmochimica Acta 72, 2377–2391. Norman, M.D., McCulloch, M.T., O, ' Neill, H.St.C., Yaxley, G.M., 2006. Magnesium isotopic analysis of olivine by laser ablation multi-collector ICP-MS: composition dependent matrix effects and a comparison of the Earth and Moon. Journal of Analytical Atomic Spectrometry 21, 50–54. Olivo, G.R., Isnard, H., Williams-Jones, A.E., Gariépy, C., 2004. Pb isotope compositions of pyrite from the C quartz–tourmaline vein of the Siscoe gold deposit, Val D'Or, Quebec: constraints on the origin and age of gold mineralization. Economic Geology 102,137–146. Paul, B., Woodhead, J.D., Hergt, J.M., 2005. Improved in situ isotope analysis of low-Pb materials using LA-MC-ICPMS with parallel ion-counter and faraday detection. Journal of Analytical Atomic Spectrometry 20, 1350–1357. Platzner, I.T., Habfast, K., Walder, A.J., Goetz, A., 1997. Modern Isotope Ratio Mass Spectrometry. John Wiley and Sons, U.K. 514pp. Simonetti, A., Heaman, L., Hartlaub, R.P., Creaser, R.A., MacHattie, T.G., Böhm, C., 2005. U–Pb zircon dating by laser ablation MC-ICP-MS using a new multiple ion counting Faraday collector array. Journal of Analytical Atomic Spectrometry 20, 677–686. Sylvester, P.J., 2008. Matrix effects in laser ablation ICPMS. Mineralogical Association of Canada short course series 40, 67–78. Wilson, S.A., Ridley, W.I., Koenig, A.E., 2002. Development of sulfide calibration standards for the laser ablation inductively coupled plasma mass spectrometry technique. Journal of Analytical Atomic Spectrometry 17, 4006–4009. Wohlgemuth-Ueberwasser, C.C., Ballhaus, C., Berndt, J., Stotter nee Paliulionyte, V., Meisel, T., 2007. Synthesis of PGE sulfide standards for laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Contributions to Mineralogy and Petrology 154, 607–617. Woodhead, J.D., 2002. A simple method for obtaining highly accurate Pb-isotope data by MC-ICPMS. Journal of Analytical Atomic Spectrometry 17, 1381–1385. Woodhead, J., Hergt, J., Shelley, M., Eggins, S., Kemp, R., 2004. Zircon Hf-isotope analysis with an excimer laser, depth profiling, ablation of complex geometries, and concomitant age estimation. Chemical Geology 209, 121–135. Woodhead, J.D., Swearer, S., Hergt, J., Maas, R., 2005. In situ Sr-isotope analysis of carbonates by LA-MC-ICP-MS: interference corrections, high spatial resolution and an example from otolith studies. Journal of Analytical Atomic Spectrometry 20, 22–27. Woodhead, J., Hellstrom, J., Hergt, J., Greig, A., Maas, R., 2007. Isotopic and elemental imaging of geological materials by laser ablation Inductively Coupled Plasma mass spectrometry. Journal of Geostandards and Geoanalytical Research 31, 331–343.