V41E-01 INVITED
A Methodology for Absolute Isotope Composition Measurement
Double spike technique was a well defined method for isotope composition measurement by TIMS of samples which have natural mass fractionation effect, but it is still a problem to define the isotope composition for double spike itself. In this study, we modified the old double spike technique and found that we could use the modified technique to solve the ˇ§trueˇ¨ isotope composition of double spike itself. According the true isotope composition of double spike, we can measure the absolute isotope composition if the sample has natural fractionation effect. A new vector analytical method has been developed in order to obtain the true isotopic composition of a 42Ca-48Ca double spike, and this is achieved by using two different sample-spike mixtures combined with the double spike and the natural Ca data. Because the natural sample, the two mixtures, and the spike should all lie on a single mixing line, we are able to constrain the true isotopic composition of our double spike using this new approach. This method not only can be used in Ca system but also in Ti, Cr, Fe, Ni, Zn, Mo, Ba and Pb systems. The absolute double spike isotopic ratio is important, which can save a lot of time to check different reference standards. Especially for Pb, radiogenic isotope system, the decay systems embodied in three of four naturally occurring isotopes induce difficult to obtain true isotopic ratios for absolute dating.
V41E-02 INVITED
Measuring 226Ra/228Ra in Oceanic Lavas by MC-ICPMS
238U-230Th-226Ra disequilibrium in volcanic rocks provides an important and unique tool to evaluate timescales of recent magmatic processes. Determination of 230Th-226Ra disequilibria requires measurement of U and Th isotopes and concentrations as well as measurement of 226Ra. While measurement of U and Th by ICPMS is now well established, few published studies documenting 226Ra measurement via ICPMS exist. Using 228Ra as an isotope spike we have investigated two ion-counting methods; a 'peak-hopping' routine, where 226Ra and 228Ra are measured in sequence on the central discrete dynode ETP secondary electron multiplier (SEM), and simultaneous measurement of 226Ra and 228Ra on two multiple ion-counter system (MICS) channeltron type detectors mounted on the low end of the collector block. Here we present 226Ra measurement by isotope dilution using the Thermo Fisher NEPTUNE MC-ICPMS. Analysis of external rock standards TML and AThO along with mid-ocean ridge basalt (MORB) and ocean island basalt (OIB) samples show three issues that need to be considered when making precise and accurate Ra measurements: 1) mass bias, 2) background, and 3) relative efficiencies of the detectors when measuring in MICS mode. Due to the absence of an established 226Ra/228Ra standard, we have used U reference material NBL-112A to monitor mass bias. Although Ball et. al., (in press) have shown that U does not serve as an adequate proxy for Th (and thus not likely for Ra either), measurements of rock standards TML and AThO are repeatedly in equilibrium within the uncertainty of the measurements (where total uncertainty includes propagation of the uncertainty in the 226Ra standard used for calibrating the 228Ra spike). For this application, U is an adequate proxy for Ra mass bias at the 1% uncertainly level. The more important issue is the background correction. Because of the extensive chemistry required to separate and purify Ra (typically fg/g level in volcanic rocks), we observe large ambient backgrounds using both ion-counting techniques, which can significantly influence the measured 226Ra/228Ra ratio. Ra off-peak backgrounds need to be measured explicitly and quantitatively corrected. One advantage of using a 'peak-hopping' routine on the central SEM is the optional use of the high abundance sensitivity lens or repelling potential quadrapole (RPQ). This lens virtually eliminates the ambient background and significantly enhances the signal to noise ratio with only a small decrease in Ra ion transmission. Even with the diminished background levels observed using 'peak-hopping' on the SEM with the RPQ, accurate measurement of Ra isotopes requires off-peak background measurement. Finally, when using MICS it is important to account for the relative efficiency of the detectors. Multiple ion counting is, in principle, preferable to 'peak-hopping' because more time is spent counting each individual isotope. However, our results illustrate that proper calibration of detector yields requires dynamic switching of 226Ra between the two ion counters. This negates the inherent advantage of multiple ion counting. Therefore, when considering mass bias, background correction, and detector gain calibration, we conclude that 'peak-hopping' on the central SEM with the RPQ abundance filter is the preferred technique for 226Ra/228Ra isotopic measurement on the Neptune MC-ICPMS.
V41E-03 INVITED
"Stable" Isotope Fractionation of Uranium: Implications for Geochemical Cycling and Geochronology
Uranium is the heaviest naturally occurring element. It has three natural radioactive isotopes, 238U, 235U and 234U and is thus widely utilized for geochronology, and two oxidation states, insoluble U(IV) and soluble U(VI). Mass-dependent thermodynamic isotopic fractionation between 235U and 238U (and 234U and 238U), which scales with δM/M2, is not normally considered significant given the small ~1% difference in mass. It is therefore usual to assume that 238U/235U is constant in nature and presently equal to 137.88 throughout the entire solar system. Importantly, isotopic fractionation of the very heavy elements has recently been investigated for mercury and thallium in the context of mass-independent nuclear field shift effects (Schauble 2007, GCA 71, 2170-2189), which do not scale with δM/M2, and are predicted to have permil-level effects on the heavy masses, including uranium. Uranium is thus emerging as a potentially significant element for monitoring biological pathways and redox processes occurring during the transition between the U(IV) and U(VI) oxidation states. We have developed experimental protocols for the precise measurement of 238U/235U and 238U/234U by multiple-collector ICPMS (MC-ICPMS) to investigate "stable" isotope fractionation in uranium. Using a Nu Plasma MC-ICPMS, concentrated solutions are measured at high signal intensity to enable simultaneous data collection on a stable multiple-Faraday array in place of the usual electron multiplier configuration. Using these protocols and a high-purity 233U-236U double-spike to internally monitor instrumental mass fractionation, we are able to resolve variations in 238U/235U and 238U/234U at the 0.4 and 0.3 epsilon level (2σ; 1 ε = 1 part in 10,000), respectively. Terrestrial and meteoritic samples formed in high-temperature environments show no variability in 238U/235U at the 1-ε level. In contrast, measurements for samples formed in low- temperature environments reveal permil-level natural variability in 238U/235U (Stirling et al., EPSL, accepted). Laboratory experiments involving the biologically-mediated reduction of U(VI) to U(IV) generate sizeable shifts in both 238U/235U and 238U/234U towards anomalous values and can constrain the relative importance of mass-dependent versus mass-independent nuclear field shift effects during the "stable" isotope fractionation of uranium. Natural variability in 238U/235U will also impact on the accuracy of the U-series and U-Th-Pb chronometers when applied to samples formed in low-temperature environments, particularly for high-precision methods, as these chronometers currently assume an invariant 238U/235U equal to 137.88.
V41E-04 INVITED
Nuclear Volume-Dependent Fractionation of Uranium Isotopes
Chemical reactions can fractionate isotopes because the magnitudes of equilibrium and rate constants are subtly sensitive to nuclear mass. Geoscientists have exploited this fact to learn about modern environmental processes and past environmental conditions by precisely measuring variations in the isotope compositions of a wide range of elements in natural materials. Here we present evidence from natural terrestrial samples that processes related to ˇ°nuclear volume" rather than ˇ°nuclear mass" significantly fractionate the isotope composition of the heaviest primordial element ¨C uranium. The isotopic composition of U in nature is generally assumed to be invariant. Here, we report variations of the 238U/235U isotope ratio in natural samples (basalts, granites, seawater, corals, black shales, suboxic sediments, ferro-manganese crusts/nodules and BIFs), which span a range of δ238U values of ~ 1.3 ‰, exceeding by far the analytical precision of our method (ˇÖ 0.06‰, 2SD, based on replicate measurements of individual samples). The largest isotope variations found in our survey are between oxidized and reduced depositional environments, with seawater and suboxic sediments falling in between. U isotopes were analyzed with MC-ICP-MS. A mixed 236U-233U isotopic tracer (double spike) was used to correct for isotope fractionation during sample purification and instrumental mass bias. Sediments formed in oxic environments, such as manganese crusts from the Atlantic and Pacific oceans, display δ238U of -0.54 to -0.62 ‰, slightly lighter than that of seawater (-0.41 ‰). However, sediments from reducing environments, such as black shales from the Black Sea (unit I and unit II) and the Cariaco basin, display heavy U isotope compositions with δ238U of up to +0.43 ‰ (0.84 ‰ heavier than seawater). Uranium enrichment in these sediments probably occurred during the reduction of soluble U(VI) (from seawater) to insoluble U(IV). Intriguingly, isotope fractionation in these sediments is opposite in direction to what would be expected from mass-dependent equilibrium or kinetic/diffusive isotope fractionation. Instead, heavy U isotope compositions of reduced U species are expected from first-principles quantum chemical modeling of fractionation driven by nuclear volume. Our modeling results predict that equilibrium nuclear volume and mass-dependent isotope fractionation operate in opposite directions for U, and that the volume effect is of greater magnitude. Combining our observations on natural samples with the results of theoretical modeling strongly indicates that nuclear volume-, rather than mass-dependent isotope fractionation is the dominant process that fractionates the isotope composition of some very heavy elements, such as U, in nature.
V41E-05
Preparation and Application of new Synthetic Uranium Isotopic Reference Materials at IRMM
Several new sets of synthetic isotope reference materials for mass spectrometer calibration have been developed at IRMM based on proven methods of purifying and mixing highly enriched oxides. In the first stage a set of 10 mixtures of 233U, 235U and 238U was made in which the 235U:238U ratios were kept at 1:1 and the 233U/235U ratios varied from 1.0 to 10-6 (IRMM-074). This set is ideal for checking the linearity response of detectors used in isotope mass spectrometry. For an absolute calibration of 235U/238U ratio measurements two types of synthetic mixtures were prepared. First the IRMM-3020-3090 set of primary isotopic mixes with 20%,35%, 50%, 75% and 90% 235U enrichment was made. Secondly also a double spike IRMM-3636 with a 233U/236U ratio of 1:1 was prepared which allows internal mass fractionation correction for 235U/238U ratio measurements. The 234U abundance of this double spike material is low enough to allow an accurate and precise correction, even within measurements of close to equilibrium uranium samples. For all synthetic isotope mixtures certification is done based on weights, purity and isotopic enrichment of the highly enriched starting materials. However, a careful verification e.g. by mass spectrometry is necessary. This paper describes the preparation techniques, verification measurements using TIMS and some applications for synthetic isotope reference materials produced by IRMM.
V41E-06
EARTHTIME: Isotopic Tracers and Optimized Solutions for High-Precision U-Pb ID-TIMS Geochronology
High-precision U-Pb ID-TIMS geochronology presents several analytical challenges. Measuring <10-100 pg radiogenic Pb in accessory phases, e.g. zircon, requires detector(s) with large dynamic range and linear response. Typical analyses involve a few tens of counts per second (cps) of 204Pb and 105 to >106 cps for 206Pb. This usually requires ‘peak jumping' on an ion counting Daly or SEM, or at high 206Pb count rates exceeding the limit of the Daly/SEM, mixed ion counting - Faraday detection methods. Double-spiked uranium measurements (as oxide) are often less challenging and measured on Faraday detectors, although low U, small and/or old samples often requires ion counting. At present, interlaboratory biases exceed analytical uncertainties, and originate from spike calibration, detector characteristics, and ion- counting protocols. In an attempt to eliminate most to all of this bias we have prepared a community tracer solution and series of optimized solutions for monitoring long-term mass spectrometer performance. As part of the EARTHTIME Initiative a quantity of mixed 205Pb-233U-235U tracer has been prepared, aiming to effectively eliminate interlaboratory bias due to tracer calibration. This tracer solution is calibrated against three mixed U/Pb gravimetric reference solutions (high purity NBS and CRM metal derivatives). Interlab tracer U/Pb ratio calibration indicates agreement to better than 300 ppm. Four synthetic U/Pb solutions have also been prepared with concordant (at ca. 10, 100, 500 and 2000 Ma) isotopic compositions and elemental ratios. These solutions are in sufficient quantities to permit use by all interested labs for many years. These solutions will provide a means of assessing agreement between different detection modes, long-term reproducibility and inter-laboratory agreement without the complexity of sample dissolution, ion exchange chemistry, etc. Single detector (Daly/SEM) ion counting underpins the majority of high-precision U/Pb dates; therefore detector linearity over a large dynamic range and long-term ‘stability' require regular verification. Single element reference materials (e.g. SRM U500; the IRMM 073/074 series) can be used for calibrating ion counting systems but they are not always appropriate for determining the response effect of other elements (e.g., Pb). We believe the new tracers and standard solutions are essential for facilitating assessment and improvement of long-term accuracy of high-precision U/Pb dates.
V41E-07
High-Precision Double-Spike Sr Isotope Measurements: Applications to Geochemistry and Cosmochemistry
Strontium isotope systematics consist of the stable 88Sr/86Sr ratio, the radiogenic 87Sr/86Sr ratio and the 84Sr/86Sr ratio which is stable in terrestrial materials and may vary due to nuclear- synthetic processes in some meteorites [1]. Growing interest in natural mass dependent 88Sr/86Sr variations indicate that terrestrial samples vary by ~500 ppm and have been measured to a precision of 50-70 ppm by MC-ICP-MS [2, 3]. However, the precision is large relative to the likely variation in many terrestrial systems. Furthermore, 84Sr/86Sr ratios are poorly determined due to spectral interferences and therefore MC-ICP-MS techniques are not appropriate for high-precision analyses of all the Sr isotope ratios. Recent advances in TIMS allow 87Sr/86Sr and 84Sr/86Sr ratios to be measured to an external precision of ~5 and 60 ppm respectively with instrumental fractionation being corrected by internal normalisation to a constant 88Sr/86Sr ratio. Following on from the pioneering work of [4] we use an 84Sr- 87Sr double-spike coupled with TIMS analyses. Optimal spiking, extended ion collection times and a 25-fold improvement in the determination of the spike composition compared to [4] allow us to measure 87Sr/86Sr and 88Sr/86Sr ratios to better than 10 ppm external precision. The double-spike determination consists of an un-spiked and spiked measurement, with spiking prior to chemical separation because column separation produces a consistent and resolvable light fractionation effect of 21 to 93 ppm. The un-spiked run allows us to assess any non-mass dependent fractionation effects. Double-spike deconvolution is performed in 87Sr denominator space. Our preliminary high-precision data allows us to resolve small (25 ppm) mass-dependent shifts in seawater 88Sr/86Sr, which may reflect short-timescale Sr fluxes in the oceans. References [1] D.A. Papanastassiou, G.J. Wasserburg, Geophys. Res. Lett. 5 (1978) 361-376. [2] J. Fietzke, A. Eisenhauer, Geochem. Geophys. Geosyst., 7, (2006) Q08009, doi:10.1029/2006GC001243. [3] G.M. Nowell, et al., Geochim. Cosmochim. Acta. 71 (2007) A725. [4] P.J. Patchett, Nature, 283 (1980b) 438-441.
V41E-08
A New Multi Collector Isotope Ratio Mass Spectrometer
With the goal of improving the sensitivity of isotope ratio measurements, particularly for actinides, a new magnetic sector mass spectrometer that utilizes up to seven full-sized discrete dynode electron multipliers operating simultaneously has been designed, constructed and is in the early stages of testing. The design is based on a newly developed ion dispersion lens that enables the mass dispersed individual isotope beams to be separated by 35 mm; this allows a full-sized discrete dynode pulse counting multiplier to be used for each beam. The ion dispersion lens (US patents 6,297,501 and pending) is a two element electrostatic 90 degree sector device that causes the beam-to-beam dispersion to increase faster than the intra-beam dispersion. Each of the multipliers is housed in an isolated case and is equipped with a deflector/condenser lens at the entrance to optimize pulse generation. The instrument includes a 9-sample filament cartridge mounted on a micro-manipulator X-Y stage that enables adjustment of the filament position with 10 micron resolution within the ion lens. Initial testing has shown that the instrument is performing as predicted by the ion optics model of the design.