Introduction to Geochronology/Mass Spectrometry
The majority of age constraints for Neoproterozoic strata have been obtained by U-Pb zircon dating therefore most of this section is concerned with U-Pb analytical methods. The analytical methods used for the ‘isochron geochronometers’ have similarities with the ID-TIMS U-Pb method however the differences are discussed below.
There are two main approaches to U-Pb zircon geochronology: in-situ ‘microbeam’ techniques and isotope dilution thermal ionization mass spectrometry (ID-TIMS). The major difference is that in ID-TIMS geochronology, zircon is dissolved and the U and Pb separated from the other elements prior to analysis whereas in microbeam techniques the zircon is analyzed using either a laser that ablates the sample prior to injection into a plasma or a focussed ion beam that generates secondary ions from a sectioned and polished epoxy mount or thin section.
U-Pb ID-TIMS ID-TIMS analyses of zircon (either as multi-grain fractions, single grains or grain fragments) involves dissolution of the zircon in the presence of tracer isotopes and is called isotope dilution. For U-Pb ID-TIMS analyses the most common tracers are 205Pb and 235U. 205Pb is an artificial isotope that does not occur in nature whereas 235U occurs naturally but an enriched form is used as a tracer. Natural U in the zircon assumed to have a 238U/235U = 137.88 (Steiger and Jager, 1977) therefore the tracer can be used to determine the number of moles of 238U and 235U in a sample. Since the tracer is added before dissolution, the tracer/sample isotope ratios stay constant despite incomplete recovery during chemistry and low ionization efficiency during analysis. Following dissolution the sample undergoes chemical purification using anion exchange chemistry that allows separation of the Zr and REEs from the Pb and U, and Pb and U from one another.
Following purification Pb and U are analysed separately by thermal ionisation mass spectrometry where the ratios of sample isotopes (204Pb, 206Pb, 238U, etc) to the tracer isotopes (205Pb, 235U) can be measured. As the amount of tracer isotope added to the same is known the number of atoms of each naturally occurring isotope in the sample determined based upon the measured ratio of sample isotope/tracer isotope ratio. . After corrections for mass fractionation, the minor contribution of common Pb and U from the reagents, the tracer and labware, the sample 206Pb/207Pb, 206Pb/238U and 207Pb/235U ratios can be determined and 206Pb/207Pb, 206Pb/238U and 207Pb/235U dates calculated. See section 5.2.3 for further discussion of the isotope dilution technique. Optimisation of this technique means that it is now possible to date zircons with <10 pg radiogenic Pb and obtain precision <0.1% on the U/Pb ratio for single grain analyses. It is a very labour intensive technique, each single U/Pb analyses takes several hours of mass-spectrometry and chemical purification in an ultra-clean lab environment, making it time consuming to develop high-n datasets.
U-Pb Microbeam techniques
U-Pb geochronology by microbeam techniques has revolutionized geochronology over the past two decades. The two major techniques are Secondary Ion Mass Spectrometry (SIMS), typified by the SHRIMP (Sensitive High Resolution Ion Microprobe), and Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS). Both of these techniques (collectively termed ‘microbeam’ techniques) offer high-spatial resolution analyses using either a focused ion beam to sputter a volume of zircon (SIMS) or a laser that is used to vaporise a volume of zircon (LA-ICP-MS). Microbeam techniques allow in-situ analysis of very small volumes and thus high-spatial resolution; a typical volume of zircon analyzed by an ion-probe is cylindrical, 20-30 microns in diameter and several microns deep, with somewhat larger volumes for LA-ICPMS (Kosler and Sylvester, 2003). In addition the analyses can be done relatively rapidly (many 10’s analyses per day for LA-ICPMS and SIMS). Furthermore, these techniques allow analysis of remaining mineral for other isotopes/elements of interest (Hf, O, REE’s) and can be made on the same zircon grains in close proximity to the volume analyzed for geochronology.
Fundamental to the microbeam U-Pb zircon methods is use of a primary standard against which the U/Pb ratio of the unknown zircon is calibrated. For SIMS techniques this calibration involves the analyses of standard zircon to develop a calibration curve for a known U/Pb ratio (which is determined via ID-TIMS analyses) and against which analyses of unknown zircons can be compared. This is achieved through analytical sessions where a standard zircon is repeatedly analysed interspersed with analyses of unknown zircons (this is termed sample-standard bracketing). During the course of a single analytical session, the measured ratio/date of a standard can drift by several percent. For LA-ICPMS the approach is somewhat similar in that sample-standard bracketing is employed in order to determine the inter-elemental fractionation which is then applied to the unknown zircons. In both SIMS and LA-ICPMS techniques the 207Pb/206Pb ratio is a direct measurement, for SIMS mass-dependent fractionation appears to be minimal and the measured ratio is commonly used whereas in LA-ICPMS analyses mass dependent fractionation is quantifiable and is corrected for either using sample-standard bracketing and/or using a solution with known 205Tl/203Tl ratio to correct for mass bias on Pb isotopic ratios. For further details of microbeam techniques see Ireland and Williams (Ireland and Williams, 2003) for a review of SIMS U-Pb geochronology and Kosler and Sylvester (Kosler and Sylvester, 2003) for a review of LA-ICPMS geochronology
The benefit of the high spatial resolution provided by microbeam techniques is a tradeoff in that the precision of individual spot analyses using LA-ICPMS and SIMS is lower than ID-TIMS by approximately an order of magnitude (Ireland and Williams, 2003; Kosler and Sylvester, 2003). In-situ techniques are without question essential tools for characterizing complex (zoned) zircons from volcanic and metamorphic rocks and for characterizing detrital populations which in some cases can provide robust estimates of the minimum age of a sequence.
Isochron techniques involve analysis of multiple samples assumed to be the same age, have a spread in parent/daughter ratio, and have remained closed systems. In order to maximize the likelyhood that the samples are the same age and have the same initial isotopic composition it is preferable to design an appropriate sampling strategy. For example, Re-Os geochronology is often attempted on laminated organic rich shales so it is essential to have multiple samples from the same restricted stratigraphic interval (e.g., (Kendall et al., 2004), to avoid samples that span an amount of time much larger than uncertainties. In some cases, such as working with core samples, this is not always possible (Kendall et al., 2006; Schaefer and Burgess, 2003). This could be a complicating factor if there is temporal variation in the initial isotopic composition of the daughter element, and/or represents a significant amount of time, especially in condensed sections. In U-Pb in carbonates, different cement domains can have very different U-Pb ratios but is difficult to know a priori whether the cements are all exactly the same age and have the same initial ratios. In both of these cases, deviations from the assumption of the same age and initial ratio lead to increased scatter and uncertainties in calculated dates.
Sample dissolution and purification techniques are similar to the procedures for U-Pb ID-TIMS. Prior to isotope ratio mass spectrometry samples undergo dissolution and chemical purification. For multi-element systems (such as Re-Os, U-Pb and Lu-Hf) isotopic tracers are added prior to dissolution for the isotope dilution (see above) whereas for systems where only daughter isotopes are measured (i.e., Pb-Pb), direct measurements of the isotope ratios are made. The isotopic composition is determined via thermal ionization mass-spectrometry although it is also possible to use solution mode ICPMS for most elements.
The accuracy and precision of isochron techniques is largely controlled by a spread in initial parent/daughter ratio, closed system behaviour, and an identical initial isotopic composition for all samples. For precipitates such as carbonates and phosphates there is often no significant detrital input however this is not the case for organic-rich shales targeted for Re-Os where there is potential that significant concentrations of initial Os from multiple sources. This has been demonstrated in several studies (Creaser et al., 2002; Kendall et al., 2004) however it is possible to limit the detrital Os contribution by selective dissolution of the organic component using a CrO3-H2SO4 dissolution approach. Kendall et al (Kendall et al., 2004) compared two dissolution methods (aqua regia vs. CrO3-H2SO4 dissolution) on greenschist facies organic rich shale from the Old Fort Point Formation in Western Canada. Both dissolution techniques were used on the same powders, however the aqua regia method yielded scattered data and a resulting “isochron” regression with an MSWD of 65 and a large “age” uncertainty (9%) in comparison to the CrO3-H2SO4 dissolution method which yielded an isochron with much less scatter (MSWD = 1.2) and a relatively low uncertainty (0.8% 2) (Kendall et al., 2004).