A Simplified, Accurate and Fast Method for Lithium Isotope Analysis of Rocks and Fluids, and δ7Li Values of Seawater and Rock Reference Materials

We report a new approach to conduct fast and accurate lithium isotope ratio measurements by MC-ICP mass spectrometry after wet chemical sample preparation. In contrast to most previously published methods our MC-ICP-MS set-up did not use a desolvating system to achieve appropriate ion beam intensities and, therefore, was less affected by matrix-induced shifts of the instrumental mass bias. As the total lithium background and build-up in the sample introduction system was low, previous sample residues could be washed out by an extended uptake of the new sample. Elimination of a nitric acid rinse step increased the sample throughput by a factor of two and allowed the instrumental mass bias drift to be tracked more precisely. δ⁷Li values of powdered silicate rock reference materials and seawater obtained in this study revealed good accuracy and an overall analytical uncertainty of typically 0.5‰ (2s). On the basis of a comparison between our lithium isotope data and compiled literature data, we recommend preliminary average δ⁷Li values for seawater (+30.8‰) and several silicate rock reference materials (BHVO-1: +5.0‰; JA-1: +5.6‰; JB-2: +4.8‰). The compilation of published δ⁷Li values for seawater suggests that the observed large lithium isotope differences are due to inter-method and/or interlaboratory bias. Most recently published δ⁷Li values for seawater show little variation and confirm a constant lithium isotope composition (at the sub ‰ level) of seawater in well mixed ocean basins.     

Refinements in BaSO4 to CO2 Preparation and δ18 O Calibration of the Sulfate Reference Materials NBS-127, IAEA SO-5 and IAEA SO-6

Refinements have been made to achieve over 99% yield in the conversion of CO to CO₂ in order to improve the reproducibility and accuracy of δ¹⁸ O measurements in sulfates. BaSO₄ (10-15 mg) was mixed with an identical amount of spectrographic-grade graphite and loaded into a Pt boat. The mixture was gradually heated to 1100 °C to reduce sulfate to CO and CO₂; the former gas was simultaneously converted to CO₂ by a glow discharge between Pt electrodes immersed in a magnetic field (produced by a pair of external neodymium magnets). A small memory effect was noticed during the analysis (less than 0.3‰ per 10‰ difference in δ¹⁸ O between two subsequently analysed samples). The memory effect, however, was suppressed by repetitive preparation of the same specimen. CO₂ produced in this way from sulfate reference samples was analysed on a dual inlet and triple collector mass spectrometer along with CO₂ equilibrated with VSMOW, GISP and SLAP water reference samples. To avoid large departures of measured isotope ratios from ¹⁸O/¹⁶O of the working calibrator we used CO₂ gas prepared from ocean water sulfate for this purpose. The calibrated δ¹⁸ O values (in ‰) obtained in this way for NBS-127, IAEA SO-5 and IAEA SO-6 reference materials were 8.73 ± 0.05, 12.20 ± 0.07 and -10.43 ± 0.12, respectively.     

Laser Ablation and Solution ICP-MS Determination of Rare Earth Elements in USGS BIR-1G, BHVO-2G and BCR-2G Glass Reference Materials

Data are reported for rare earth elements (REE) in three geological glass reference materials (BIR-1G, BHVO-2G and BCR-2G) using a UV (266 nm) laser ablation ICP-MS system and the classical (HF-HClO₄) acid decomposition method, followed by conventional nebulisation ICP-MS. External calibration of laser ablation analyses was performed using NIST SRM reference materials with internal standardisation using ²⁹Si and ⁴⁴Ca. Replicate analyses of reference basaltic glasses yielded an analytical precision of 1-5% (RSD) for all the elements by solution ICP-MS and 1-8% (RSD) by laser ablation ICP-MS. The relative differences between the REE concentrations measured by solution and laser ablation ICP-MS compared with the reference values were generally less than 11 % for most elements. The largest deviations occurred for La determined by solution ICP-MS in BIR-1G. The results of both solution and laser ablation ICP-MS agreed well, generally better than 7%, with the exception of La, Pr and Sm in BIR-1G. The measured REE laser ablation data for BIR-1G, BHVO-2G and BCR-2G agreed with the previously published data on these basaltic reference glasses, within a range of 0-10% for most elements. No significant influences were observed for the predicted spectral interferences on some REE isotopes in the analysis of basaltic glasses.     

Calcium Isotopes (δ44/40Ca) in MPI-DING Reference Glasses, USGS Rock Powders and Various Rocks: Evidence for Ca Isotope Fractionation in Terrestrial Silicates

We report δ^44/40Ca_{(SRM 915a)} values for eight fused MPI‐DING glasses and the respective original powders, six USGS igneous rock reference materials, the U‐Th disequilibria reference material TML, IAEA‐CO1 (Carrara marble) and several igneous rocks (komatiites and carbonatites). Sample selection was guided by three considerations: (1) to address the need for information values on reference materials that are widely available in support of interlaboratory comparison studies; (2) support the development of in situ laser ablation and ion microprobe techniques, which require isotopically homogenous reference samples for ablation; and (3) provide Ca isotope values on a wider range of igneous and metamorphic rock types than is currently available in the scientific literature. Calcium isotope ratios were measured by thermal ionisation mass spectrometry in two laboratories (IFM‐GEOMAR and Saskatchewan Isotope Laboratory) using ⁴³Ca/⁴⁸Ca‐ and ⁴²Ca/⁴³Ca‐double spike techniques and reported relative to the calcium carbonate reference material NIST SRM 915a. The measurement uncertainty in both laboratories was better than 0.2‰ at the 95% confidence level. The impact of different preparation methods on the δ^44/40Ca_{(SRM 915a)} values was found to be negligible. Except for ML3‐B, the original powders and the respective MPI‐DING glasses showed identical δ^44/40Ca_{(SRM 915a)} values; therefore, possible variations in the Ca isotope compositions resulting from the fusion process are excluded. Individual analyses of different glass fragments indicated that the glasses are well homogenised on the mm scale with respect to Ca. The range of δ^44/40Ca_{(SRM 915a)} values in the igneous rocks studied was larger than previously observed, mostly owing to the inclusion of ultramafic rocks from ophiolite sections. In particular, the dunite DTS‐1 (1.49 ± 0.06‰) and the peridotite PCC‐1 (1.14 ± 0.07‰) are enriched in ⁴⁴Ca relative to volcanic rocks (0.8 ± 0.1‰). The Carrara marble (1.32 ± 0.06‰) was also found to be enriched in ⁴⁴Ca relative to the values of assumed precursor carbonates (< 0.8‰). These findings suggest that the isotopes of Ca are susceptible to fractionation at high temperatures by, as yet, unidentified igneous and metamorphic processes.     

A Protocol for the Determination of the Rare Earth Elements at Picomole Level in Rocks by ICP-MS: Results on Geological Reference Materials USGS PCC-1 and DTS-1

Rare earth element analyses are widely used in geology, environmental science and archaeology. Over the past decade inductively coupled plasma-mass spectrometry has become an important source of rare earth data on geological material. However, ICP-MS analysis of rock samples without pre-concentration can be problematic because of complex sample matrices that can generate significant molecular isobaric interferences on rare earth peaks and which need to be corrected. Such problems are exacerbated for ultramafic rocks because the low levels of rare earth elements demand more concentrated solutions in order to maintain signals above background levels. These high solid loads result in intra-run changes in instrument sensitivity which need to be monitored. Pre-concentration chemistries have been developed in order to avoid high solid loads but these are time-consuming and must offer quantitative recoveries or use a yield tracer. Here, we describe an alternative method for rare earth element analysis by ICP-MS, which involves no pre-concentration and is, therefore, able to deliver data rapidly. Our approach is to apply an external correction procedure, based on the analysis of a reference material closely matched in composition to the unknown samples, which allows correction for both interferences and variations in instrument sensitivity. Testing this method, we obtained accurate rare earth element results for basaltic rocks with a precision of about 2% (1s). We demonstrate that the method is also applicable to ultramafic rocks with abundances at ultra-trace (ng g⁻¹) level and present data for twelve separate dissolutions of the peridotite USGS PCC-1 and four separate dissolutions of the dunite DTS-1 reference materials. The repeatability of the data is between 3% and 9% (1s).     

Isotope Dilution MC-ICP-MS Rare Earth Element Analysis of Geochemical Reference Materials NIST SRM 610, NIST SRM 612, NIST SRM 614, BHVO-2G, BHVO-2, BCR-2G, JB-2, WS-E, W-2, AGV-1 and AGV-2

We present data for the concentrations of eleven rare earth elements (La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Er, Yb, Lu) in eleven international geochemical reference materials obtained by isotope dilution multi-collector inductively coupled plasma-mass spectrometry (MC-ICP-MS). We have analysed both rock powders and synthetic silicate glasses, and the latter provide precise data to support the use of these as reference materials for in situ trace element determination techniques. Our data also provide precise measurements of the abundance of mono-isotopic Pr in both glasses and powders, which allows more accurate constraints on the anomalous redox-related behaviour of Ce during geochemical processes. All materials were analysed in replicate providing data that typically reproduce to better than one percent. Sm/Nd ratios in all these materials also reproduce to better than 0.2% and are accurate to < 0.2% and can thus be used as calibrants for Sm-Nd geochronology. Our analyses agree well with existing data on these reference materials. In particular, for NIST SRM 610, USGS BHVO-2, AGV-1 and AGV-2, our measured REE abundances are typically within < 2% (and mostly 1%) of REE concentrations previously determined by isotope dilution analysis and thermal ionisation mass spectrometry, consistent with the higher degree of precision and accuracy obtained from isotope dilution techniques. Close agreement of results between basaltic glass reference materials USGS BHVO-2G and BCR-2G and the BHVO-2 and BCR-2 powders from which they were created suggests that little fractionation, concentration or dilution of REE contents occurred during glass manufacture.     

Facies analysis and correlation of high-order sequences in middle–outer ramp successions: variations in exported carbonate on basin-wide δ13Ccarb (Kimmeridgian, NE Spain)

Carbonate mud that accumulated in the deep parts of a late Kimmeridgian carbonate ramp (Iberian Basin, NE Spain) was partly derived by resedimentation from shallow water production areas. High-frequency sea-level changes, probably driven by climatic changes in tune with precession and short-eccentricity cycles, affected carbonate production and the amount of exported sediment. Facies analysis and correlation of three outcrops located in middle and outer ramp settings allows a comparison of high-order sequences (bundles of beds and sets of bundles) across a ramp transect and an assessment of the carbonate factory. Analysis of the storm deposits found in middle ramp settings identifies deepening to shallowing high-frequency cycles based on the level of exported carbonate. In outer ramp areas, many of the bundles exhibit a thinning trend, indicating a progressive decrease of carbonate production and hence, carbonate export during periods of high-frequency sea-level rise. δ¹³C_carb values show a gradual increase through the studied long-term transgressive interval ranging from 1·5‰ to 2·8‰. Within this long-term evolutionary trend, short-term δ¹³C_carb fluctuations occur that correspond with some of the high-order cycles defined from sedimentary facies analysis. These short-term δ¹³C_carb shifts are interpreted as shifts in carbonate export from shallow reef regions to the outer ramp. A consequence of this study is that variation in δ¹³C_carb can be used for correlation in outer ramp successions, at least on a basin-wide scale.     

Determination of High Field Strength Elements, Rb, Sr, Mo, Sb, Cs, Tl and Bi at ng g−1 Levels in Geological Reference Materials by Magnetic Sector ICP-MS after HF/HClO4 High Pressure Digestion

Six low abundance rock reference materials (basalt BIR-1, dunite DTS-1, dolerite DNC-1, peridotite PCC-1, serpentine UB-N and basalt TAFAHI) have been analysed for high field strength elements (Zr, Nb, Hf, Ta, Th and U), Rb, Sr, Mo, Sb, Cs, Tl and Bi at ng g⁻¹ levels (in rock) by magnetic sector inductively coupled plasma-mass spectrometry after HF/HClO₄ high pressure decomposition. The adopted method uses only indium as an internal standard. Detection limits were found to be in the range of 0.08 to 16.2 pg ml⁻¹ in solution (equivalent to 0.08 to 16.2 ng g⁻¹ in rock). Our data for high field strength elements, Rb, Sr, Mo, Sb, Cs, Tl and Bi for the six selected low abundance geological reference materials show general agreement with previously published data. Our Ta values in DTS-1 and PCC-1 (1.3 and 0.5 ng g⁻¹) are lower than in previously published studies, providing smooth primitive mantle distribution patterns. Lower values were also found for Tl in BIR-1, DTS-1 and PCC-1 (2, 0.4 and 0.8 ng g⁻¹). Compared with quadrupole ICP-MS studies, the proposed magnetic sector ICP-MS method can generally provide better detection limits, so that the measurement of high field strength elements, Rb, Sr, Mo, Sb, Cs, Tl and Bi at ng g⁻¹ levels can be achieved without pre-concentration, ion exchange separation or other specialised techniques.