B13A-0884 INVITED
Reporting and measurement of mass-dependent and mass-independent fractionation of mercury isotopes
Hg isotope analysis by MC-ICP-MS is an important new approach for fingerprinting Hg sources and monitoring Hg redox reactions and bioaccumulation, especially with the recent discovery of mass independent Hg isotope fractionation. Unfortunately research groups have adopted different standards, definitions of delta values, and methods of isotopic measurement. We suggest that a single standard, NIST SRM 3133, be adopted for reporting the isotopic variability of Hg isotopes. Isotope ratios should be determined by sample-standard bracketing (SSB) during analysis and reported as permil (‰) deviation from SRM 3133. For the highest precision and accuracy, a Tl internal standard along with SSB should be used to correct instrumental mass bias. Measurement routines should also include on-peak zero corrections and matching of concentration and matrix between the samples and bracketing standard. For samples that display mass-dependent fractionation (MDF), only one delta value needs to be reported (δ202/198Hg). Mass-independent fractionation (MIF) (Jackson et al., 2006; Bergquist et al., 2006; Bergquist and Blum, submitted) requires additional nomenclature, and we suggest reporting MIF as the deviation in isotope ratios from the theoretical mass dependent kinetic isotope fractionation (Δxxx/198Hg)¬. External reproducibility should be monitored by analysis of secondary standards. For studies of MDF, we use an in-house secondary standard solution made from metallic Hg mined from Almaden Spain and obtain a δ202Hg of -0.55 ±0.06‰ (2SD). For studies of MIF, we use NRCC CRM DORM-2 (dogfish muscle) and obtain a mean value of δ202Hg of +0.19 ±0.13‰ (2SD), Δ201Hg of +0.89 ±0.07‰ (2SD) , and Δ199Hg of +1.07 ±0.08‰ (2SD).
B13A-0885
High Precision Isotopic Reference Material Program
Recent developments in thermal ionization and inductively coupled plasma multicollector mass spectrometers have lead to "high precision" isotope ratio measurements with uncertainties approaching a few parts in 106. These new measurement capabilities have revolutionized the study of isotopic variations in nature by increasing the number of elements showing natural variations by almost a factor of two, and new research areas are actively opening up in climate change, health, ecology, geology and forensic studies. Because the isotopic applications are impacting very diverse fields, there is at present little effective coordination between research laboratories over reference materials and the values to apply to those materials. NIST had originally developed the techniques for producing accurate isotopic characterizations, culminating in the NIST Isotopic SRM series. The values on existing materials however are insufficiently precise and, in some cases, may be isotopically heterogeneous. A new generation of isotopic standards is urgently needed and will directly affect the quality and scope of emergent applications and ensure that the results being derived from these diverse fields are comparable. A series of new isotopic reference materials similar to the NIST 3100 single element solution series is being designed for this purpose and twelve elements have been selected as having the most pressing need. In conjunction with other expert users and National Metrology Institutes, an isotopic characterization of the respective 12 selected ampoules from the NIST single element solution series is currently underway. In this presentation the preliminary results of this screening will be discussed as well as the suitability of these materials in terms of homogeneity and purity, long term stability and availability, and isotopic relevance. Approaches to value assignment will also be discussed.
B13A-0886
Uncertainty Estimation for Extreme Isotopic Ratios Measured in Multi-Collector Mass Spectrometers
Isotopic abundances for elements of interest to the earth science community can vary by many orders of magnitude. Commonly used detector systems (Faraday, Daly, SEM) can not encompass the entire dynamic range of interest and making measurements at the limits of the useful range for a single type of detector can also present special difficulties. Therefore, estimating a measurement uncertainty that accurately represents the level of confidence in the results obtained for an extreme ratio (<1:10,000) requires the careful consideration of several components that are not applicable to more conventional isotopic ratio measurement. To be consistent with widely accepted principles for determination of measurement uncertainty, as expressed in the ISO "Guide to the Expression of Uncertainty in Measurement" (GUM), it is necessary to consistently evaluate the contributions of these additional components and provide an uncertainty budget for measurement results. This is of particular importance for measurements of extreme isotopic ratios because many of the uncertainty components are associated with measurement biases. For mass spectrometric measurements of elements with relatively evenly distributed isotopic abundances (e.g. Sr and Nd), multicollector dynamic analysis techniques can reduce the significant uncertainty contributors to measurement variability and uncertainty of the internal normalization ratio. For more extreme ratios such as 234U/238U and 230Th/232Th, not only do uncertainties associated with measurement variability and mass bias corrections have to be considered, but several other components can be significant or even dominant contributors to measurement uncertainty. These components include uncertainties associated with detector inter-calibration, detector linearity, electronic baseline, background corrections, dark noise, dead time, and mass interferences. Although one or several of these parameters may not be significant contributors to the uncertainty for a particular measurement, the relative significance of any individual component can change dramatically depending upon signal intensity. Therefore, a proper uncertainty evaluation will, at least initially, include all of the identifiable components associated with that particular measurement system. The relationship of these various components to the total uncertainty should be presented in an uncertainty budget which provides transparency for the uncertainty estimate and is a useful tool for determining where best to seek improvements in accuracy or precision.
B13A-0887 INVITED
Myths of Isotopic Reference Materials Busted
During the past several years, the determination of the isotopic abundances of elements including H, Li, B, C, N, O, Mg, Si, S, Cl, Ca, Cr, Fe, Cu, Zn, Tl, and Se has substantially increased because of expanded use in hydrology, environmental studies, microbiology, forensic investigations, atmospheric investigations, oceanography, etc. Improvements in instrumentation enable increasingly precise isotope-amount-ratio measurements in these fields, but these improvements in precision commonly do not lead to improvements in accuracy because of the lack or improper use of isotopic reference materials. When properly used, these critically important materials enable any laboratory worldwide to measure the same homogeneous sample and report the same isotopic abundance within analytical uncertainty. For example, for stable isotopic analysis of gaseous hydrogen samples, the agreement among 36 laboratories worldwide before normalization to any hydrogen gas reference material was 11.8 per mill. After normalization to anchors (gaseous H isotopic reference materials) at each end of the delta H-2 scale, the agreement was 0.85 per mill, an improvement of more than an order of magnitude. Consistency of delta C-13 measurements often can be improved by nearly 50 percent by anchoring the delta C-13 scale with two isotopic reference materials differing substantially in C-13 mole fraction, namely NBS 19 calcite and L-SVEC lithium carbonate. Agreement of delta C-13 values of four expert laboratories analyzing USGS40 L- glutamic acid by CF-IRMS methods improved from 0.084 to 0.015 per mill with use of the two scale anchors (NBS 19 and L-SVEC). Solid oxygen isotopic reference materials (IAEA-600 caffeine, IAEA-601 and IAEA-602 benzoic acids, IAEA-NO-3, USGS32, USGS34, and USGS35 nitrates, NBS-127, IAEA-SO-5, and IAEA-SO-6 barium sulfates) are poorly calibrated. Calibrating these solids to the VSMOW-SLAP reference water scale has been very difficult because both the solids and reference waters need to be analyzed simultaneously using state-of-the-art TC/EA methodology. A large difficulty is loading water reference samples into silver capsules with minimal evaporation and no leakage. A technique has been developed to seal 250-nL water samples in smooth walled silver capsules using a specially modified crimping tool, thereby minimizing the aforementioned difficulties and consequent errors when combining and comparing water and solid reference materials. It is anticipated that this new technique will improve the agreement of oxygen isotopic reference materials substantially.
B13A-0888
Method-Dependent Variations in Oxygen Isotope Compositions Obtained for Structural Carbonate in Bone Bioapatite
The carbon and oxygen isotope compositions of structural carbonate contained in bioapatite can be obtained by reaction with ortho-phosphoric acid at various temperatures and reaction times, using off-line or automated sample preparation, and continuous-flow or dual-inlet isotope-ratio mass spectrometry (IRMS). Here, we compare the isotopic compositions obtained for structural carbonate in bone bioapatite using (1) conventional off- line gas extraction (25°C) and dual-inlet IRMS, (2) GasBench automated sampling (50°C) and continuous-flow IRMS, and (3) MultiPrep automated sampling (50°C, 90°C) and dual-inlet IRMS. On average, the stable carbon isotope compositions obtained for the same sample using different methods are within ±0.33 per mil (n=29). The reproducibility of oxygen isotope compositions using the different methods is much poorer (±3.18 per mil; n=29). The differences among these methods were most pronounced for samples analyzed using the MultiPrep at 90°C and dual-inlet IRMS. In the latter case, some samples consistently had extremely low oxygen isotope compositions, and also yielded a contaminant gas containing masses 47, 48, and 49, which was not separated from the normal reaction product (carbon dioxide) during cryogenic processing. Normal and anomalous samples do not systematically differ in their crystallinity indices, C/P ratios, gas yields, or total organic content. Anomalous samples have an additional peak in their deconvoluted FTIR spectra at 866 cm-1 and their ignition products lack β-TCP. They also contain slightly more Al and Si cations, and slightly fewer Na cations. It is not clear how these subtle structural and chemical differences relate to the production of the contaminant gas.
B13A-0889
Memory Effects in Compound-Specific D/H Analysis by GC-P-IRMS
Compound-specific D/H measurements of lipids often encounter huge ranges in D/H ratio. Moreover, culture experiments using D-enriched water to study fractionations often extend the working range to 1000 permil or more. While seeking improved methods for standardization of such measurements, we observed that the D/H ratio of each peak in a chromatogram is subtly influenced by that of the preceding peak. Such 'memory' effects have not been previously reported, but can have a significant impact on certain types of measurements. Here we describe experiments that fully characterize this memory phenomenon. To quantitatively evaluate memory effects, we analyzed ethyl palmitate (C16E) with D/H values ranging between - 230 and +800 permil co-injected with n-propyl palmitate (C16P, D/H = -225 permil) and with methane reference gas (D/H = -148 permil) at variable time separations. Five factors are observed to affect the memory effects between two successive peaks: 1) difference in D/H; 2) difference in relative abundance; 3) time separation; 4) age and condition of pyrolysis reactor; 5) difference in D/H between peak and background. For a time separation of 100s, approximately 2.4 to 4.3% of the measured hydrogen in C16P appears to derive from the preceding peak (C16E). This factor decreases to 1.2% if C16P is replaced by methane reference gas that is injected downstream of the GC column. Memory effects decrease exponentially with increasing peak separation with a decay half-time of 433s, and increase as the pyrolysis reactor ages. Our results can be explained quantitatively by two pools of exchangeable H in the system, one in the GC column and one in the pyrolysis reactor. The latter appears to increase in size as the reactor ages. The size of the memory effects make them significant for measurements in which successive peaks differ in D/H by more than 100 permil. Culture experiments using D-enriched waters are particularly susceptible, and will be systematically biased by these effects. Also, the widely used normalization procedure based on n-alkanes with alternating D/H around -50 and -200 permil can be shown to derive almost entirely from memory effects, rather than from true scale compression. A simple strategy to separately assess isotopic memory and scale compression is currently lacking.
B13A-0890
High Spatial Resolution Isotopic Abundance Measurements by Secondary Ion Mass Spectrometry: Status and Prospects
Secondary Ion Mass Spectrometry, SIMS or ion microprobe analysis, has become an important tool for geochemistry because of its ability study the distributions of elemental and isotopic abundances in situ on polished samples with high (typically a few microns to sub-micron) spatial resolution. In addition, SIMS exhibits high sensitivity for a wide range of elements (H to Pu) so that isotope analyses can sometimes be performed for elements that comprise only trace quantities of some mineral phase (e.g., Pb in zircon) or on major and/or minor elements in very small samples (e.g., presolar dust grains). Offsetting these positive attributes are analytical difficulties due to the complexity of the sputtering source of analyte ions: (1) relatively efficient production of molecular ion species (especially from a complex matrix such as most natural minerals) that cause interferences at the same nominal mass as atomic ions of interest, and (2) quantitation problems caused by variations in the ionization efficiencies of different elements and/or isotopes depending upon the chemical state of the sample surface during sputtering--the so-called "matrix effects". Despite the availability of high mass resolution instruments (e.g., SHRIMP II/RG, CAMECA 1270/1280/NanoSIMS), the molecular ion interferences effectively limit the region of the mass table that can be investigated in most samples to isotope systems at Ni or lighter or at Os or heavier. The matrix effects and the sensitivity of instrumental mass discrimination to the physical state of the sample surface can hamper reproducibility and have contributed to a view that SIMS analyses, especially for so- called stable isotopes, are most appropriate for extraterrestrial samples which are often small, rare, and can exhibit large magnitude isotopic effects. Recent improvements in instrumentation and technique have extended the scope of SIMS isotopic analyses and applications now range from geochronology to paleoclimatology to volcanology to biogeochemistry and cosmochemistry. Multiple collector (static magnetic field) measurements at high mass resolving power have enabled high precision (sub-permil) for several stable isotopes systems (e.g., C, O, Mg, S). Applied to geochronology, the multiple collector approach permits very rapid survey of zircon Pb-Pb ages to identify candidate Hadean grains for further detailed analysis. Ion imaging has been used to correlate isotope compositions with biochemistry (e.g., FISH-SIMS) or to search for especially rare samples among larger populations (e.g., supernova grains of Stardust). For favorable sample geometries with lateral homogeneity, SIMS isotope analyses may be conducted in depth-profiling mode which brings spatial resolution into the tens of nm range. Applications of this approach include experimental petrology, thermochronology, and isotopic analyses of shallowly-implanted solar wind ions. New approaches to removal of molecular ion interferences include reverse- geometry instrumentation and accelerator-based SIMS. There always exists trade-offs between microanalysis and trace analysis on the one hand, and high precision on the other. In this contribution, I will review current status for isotope precision and accuracy of SIMS for applications in stable and radiogenic isotopes as a function of spatial scale. A discussion of current limits and future prospects for improvement in understanding matrix effects will be given. Examples from ion imaging/ depth profiling/ geochronology and cosmochemistry will be provided.
B13A-0891 INVITED
Advances in laser ablation MC-ICPMS isotopic analysis of rock materials
Laser ablation multiple-collector inductively coupled plasma-source mass spectrometry (LA-MC-ICPMS) is a rapid method for obtaining high-precision isotope ratio measurements in geological samples. The method has been used with success for measuring isotope ratios of numerous elements, including Pb, Hf, Mg, Si, and Fe in terrestrial and extraterrestrial samples. It fills the gap between the highest precision obtainable with acid digestion together with MC-ICPMS and thermal ionization mass spectrometry (TIMS) and the maximum spatial resolution afforded by secondary ion mass spectrometry (SIMS). Matrix effects have been shown to be negligible for Pb isotopic analysis by LA-MC-ICPMS (Simon et al., 2007). Glass standards NBS 610, 612, and 614 have Pb/matrix ratios spanning two orders of magnitude. Our sample-standard bracketing laser ablation technique gives accurate and precise 208Pb/206Pb and 207Pb/206Pb for these glasses. The accuracy is superior to that obtained when using Tl to correct for mass fractionation. Accuracy and precision (± 0.2 ‰) for Pb in feldspars is comparable to that for double-spike TIMS. Data like these have been used to distinguish distinct sources of magmas in the Long Valley silicic magma system. LA-MC-ICPMS analyses of Mg isotope ratios in calcium-aluminum-rich inclusions (CAIs) from carbonaceous chondrite meteorites have revealed a wealth of new information about the history of these objects. A byproduct of this work has been recognition of the importance of different mass fractionation laws among three isotopes of a given element. Kinetic and equilibrium processes define distinct fractionation laws. Reservoir effects can further modify these laws. The result is that the linear coefficient β that relates the logarithms of the ratios n2/n1 and n3/n1 (ni refers to the number of atoms of isotope i) of isotopes with masses m3 > m2 > m1 is not unique. Rather, it is process dependent. In the case of Mg, this coefficient ranges from 0.521 for single-step equilibrium processes to 0.510 or even lower for kinetic processes. Rayleigh fractionation involving a kinetic process with a single-step β of 0.510 produces an effective β of 0.512. Such differences in fractionation laws can be crucial for determining excesses or deficits in isotopes relative to mass fractionation. Contrary to some assertions, Si isotope ratios can be measured with high accuracy and precision using 193 nm excimer lasers with nanosecond pulse widths (Shahar and Young, 2007). Silicon isotope ratios in CAIs measured by 193 nm LA-MC-ICPMS have been combined with Mg isotope ratios to constrain the astrophysical environments in which these oldest solar system materials formed. Accuracy of the measurements was determined using gravimetric standards of various matrix compositions. The results establish that matrix effects for Si are below detection at the ± 0.2 ‰ precision of the laser ablation technique. High mass resolving power (m/Δ m ~ 9000) is necessary to obtain accurate Si isotope ratios by laser ablation. High-precision LA-MC-ICPMS measurements of 176Hf/177Hf in zircons can be obtained by normalizing to 179Hf/177Hf assuming an exponential fractionation law and no mass-dependent Hf, Lu, or Yb stable isotope fractionation. With corrections for interfering 176Lu and 176Yb precision for this method can be on the order of 0.3 epsilon (0.03 ‰). The approach has been used to infer the existence of continental crust on Earth 4.4 billion years before present (Harrison et al., 2005).
B13A-0892
MPI-DING Reference Glasses for in-situ Isotope Analysis and the GeoReM Database
Microanalytical techniques, such as LA-ICPMS, LA-MC-ICPMS, SIMS, have become important tools for in-situ isotope analysis. Yet, the lack of widely available geological reference glasses has made interlaboratory comparisons and assessment of analytical accuracy difficult or even impossible in many cases. Jochum et al. (2000) prepared eight geological MPI-DING glasses of different natural composition (from ultramafic to highly siliceous composition) for the purpose of providing reference materials for in-situ microanalytical work. Recently, reference values for up to 74 major and trace elements have been determined following the recommendations of the IAG (Jochum et al., 2006). The isotopic composition of H, Li, B, O, Ca, Sr, Nd, Hf and Pb was determined in the MPI-DING reference glasses by high-precision mass spectrometric (e.g., TIMS, MC-ICPMS) using large sample amounts and by microanalytical (LA-ICPMS, LA-MC-ICPMS, SIMS) techniques. The precision of the bulk techniques is significantly higher than that of the microanalytical techniques. However, the microanalytical techniques have the advantage of high spatial resolution and fast sample preparation. The boron isotope composition was determined using TIMS and LA-MC-ICPMS. LA-MC-ICPMS data for the MPI- DING glasses agree with the TIMS data at the 2 σ level. The MPI-DING glasses range from δ 7Li of 2.0 to 17.1 ‰ LSVEC. The uncertainties of MC-ICPMS measurements compared to SIMS data are smaller, owing to the larger amount of Li measured. Some small local heterogeneity in Li isotopes could be observed using SIMS. Highly precise Sr and Nd isotope analyses were performed for each MPI-DING glass using TIMS. Recently, Sr isotopes have also been determined by LA-ICPMS in the low-Sr (30 - 300 ppm) glasses. The 87Sr/86Sr ratios for KL2-G (0.7038 ± 0.0002), ML3B-G (0.7035 ± 0.0003) and GOR128-G (0.7067 ± 0.0007) agree within 0.01 - 0.04 % with the TIMS data. Lead isotope ratios have been determined by TIMS, MC-ICPMS, solution ICPMS and LA-ICPMS. LA-ICPMS is able to determine 208Pb/206Pb and 207Pb/206Pb ratios in the low-Pb (1 - 10 ppm) MPI-DING glasses with a precision of about 0.05 - 0.2%. Possible heterogeneities of Pb isotopes are found by high-precision TIMS and MC-ICPMS measurements using different glass chips. However, the differences are lower than the reproducibility obtained from microanalytical techniques. Because the amount of data for reference materials has significantly increased in recent years, there is a need for a database that contains published analytical data together with important metadata (e.g., uncertainty, analytical technique, laboratory) in a form in which it can be accessed readily by geochemists. Jochum et al. (2005) have therefore developed the GeoReM database (http://georem.mpch-mainz.gwdg.de) for reference materials and isotopic standards of geochemical, mineralogical and environmental interest. At present, GeoReM contains ca. 1500 reference materials (including the MPI-DING glasses) and 13500 analyses from about 2100 publications.
B13A-0893
A Laser-Extraction Technique for Oxygen Isotope Analysis of Diatom Frustules
Biogenic opal in the form of diatom frustules is abundant in lacustrine and marine environments throughout the Cenozoic. Oxygen isotope ratios preserved in the SiO2 of diatom frustules provide a valuable archive of paleoclimatic data, such as changes in temperature and precipitation/evaporation. However, oxygen isotope analysis of diatom silica is complicated by the presence of >15 wt. % H2 O that must be removed prior to analysis for accurate δ18Odiatom measurements to be obtained. Various techniques have been used to remove the hydrous silica including vacuum dehydration, isotope exchange, and stepwise fluorination. Here we present a laser-extraction technique for oxygen isotope analysis of biogenic opal that is rapid and requires far less material than other techniques. Batches of approximately twenty 1 to 2 mg aliquots of pure diatoms are loaded in a laser chamber and pre-fluorinated with F2 gas, similarly to previously published stepwise fluorination techniques to remove the hydrous portion of biogenic silica. The remaining SiO2 is then volatilized in the presence of BrF5 with a CO2 laser to produce O2 gas. The δ18Odiatom values produced through the laser-extraction technique increase with increasing pre-fluorination times as the hydrous portion of the silica is removed. A plateau with a constant, reproducible δ18O value is obtained once all of the hydrous silica is removed. The diatoms samples analyzed in this study are from sediments collected from the Valles Caldera in New Mexico, USA, and consist of a coarse (>50 micron) and fine (<50 micron) size samples from different locations in the caldera. Two different coarse grained samples were analyzed using the laser-extraction technique and had δ18Odiatom values that range from 31.2 to 32.4 ‰ and 31.6 to 32.5 ‰. In these samples values plateaued at 32.3 (±0.1) ‰ and 32.2 (±0.2) ‰. A finer grained sample, also from the Valles Caldera, had δ18Odiatom values that ranged from 28.5 to 29.0 ‰ and plateaued at 28.7 ‰ (±0.2). Additional diatom samples used in this study came from a variety of modern and ancient lacustrine and marine sources and were used to compare the laser extraction technique with other techniques. δ18Odiatom values obtained with laser extraction are 2 to 4 ‰ higher than δ18Odiatom values obtained using conventional vacuum dehydration of the same samples, suggesting that there is more complete removal of the hydrous material with the laser extraction technique or less exchange of the hydrous component during dehydration. Reproducibility of δ18Odiatom values obtained by the laser-extraction technique is ± 0.2 ‰. We propose that the laser-extraction technique for obtaining oxygen isotope values from diatom silica has advantages over other techniques in terms of sample size, removal of the hydrous material, and analytical procedure.