Age calibration of the Fish Canyon sanidine ⁴⁰Ar/³⁹Ar dating standard using primary K–Ar standards

Abstract

The ⁴⁰Ar/³⁹Ar dating technique requires the use of neutron fluence monitors (standards). Precise calibrations of these standards are crucial to decrease the uncertainties associated with ⁴⁰Ar/³⁹Ar dates. Optimal calibration of ⁴⁰Ar/³⁹Ar standards should be based on K/Ar standards having independent isotope dilution measurements of ⁴⁰K and ⁴⁰Ar*, based on independent isotope tracers (spikes) because this offers the possibility to eliminate random interlaboratory errors. In this study, we calibrate the widely used Fish Canyon sanidine (FCs) standard based on four primary K/Ar standards (GA-1550, Hb3gr, NL-25, and GHC-305) on which K and Ar* concentrations have been determined in different labs with independently calibrated tracers. We obtained a mean age of 28.03 ± 0.08 Ma (1σ; neglecting uncertainties of the ⁴⁰K decay constants) for FCs, based on the decay constant recommended by Steiger and Jäger [Steiger R.H., Jäger. E. 1977. Subcommission on geochronology: convention of the use of decay constants in geo- and cosmochronology. Earth Planet. Sci. Lett. 36, 359–362.]. This age corresponds to a mean ⁴⁰Ar*/⁴⁰K value of (1.6407 ± 0.0047) × 10⁻³. We also discuss several criteria that prevent the use of previous calibrations of FCs based on other primary standards (LP-6, SB-3 and MMhb-1). The age of FCs obtained in this study is based on the ⁴⁰K decay constants of Steiger and Jäger (1977) but we anticipate the imminent need for revision of the value and precision of the ⁴⁰K decay constants (representing the main source of uncertainties in ⁴⁰Ar/³⁹Ar dating). The ⁴⁰Ar*/⁴⁰K result of FCs obtained in this study allows therefore a rapid calibration of the age of FCs with uncertainties at the 0.29% level but perhaps more importantly this value is independent of any particular value of the ⁴⁰K decay constants and may be used in the future in conjunction with revised decay constants.

Introduction

The ⁴⁰Ar/³⁹Ar dating method is a relative technique whose calibration is based on the ages of neutron fluence monitors (standards). Ideally, the age of the standards is determined by K/Ar dating (e.g. Turner et al., 1971) or other methods such as astronomical calibration (e.g. Renne et al., 1994). However, some minerals, not suitable for K/Ar dating due to the difficulty of quantitatively extracting ⁴⁰Ar* (e.g. sanidine, Webb and McDougall, 1967, McDowell, 1983), are known to provide more precise and reproducible ages when measured by ⁴⁰Ar/³⁹Ar dating (e.g. Fish Canyon sanidine (FCs); Renne et al., 1998, Alder Creek sanidine (ACs); Nomade et al., 2005) especially at the single grain level. Use of these “secondary” standards is also often justified by the fact that their age and composition should be comparable to the unknown and therefore minimize the range of isotopic ratios to be measured (e.g. Renne et al., 1997). These minerals could be therefore used as secondary [or even higher order; e.g. ACs (Nomade et al., 2005)] standards but their ages should be initially calibrated with primary standards. Ultimately, accuracy and precision on the age of an unknown sample is constrained by the precision and accuracy on the age of the standards.

One of the most widely used standards in ⁴⁰Ar/³⁹Ar geochronology is FCs, mainly because of its high homogeneity and typically superior reproducibility (e.g. Renne et al., 1998, Lanphere and Baadsgaard, 2001, Spell and McDougall, 2003). However, different interlaboratory experiments based on different standards and different analytical conditions provide ages for FCs varying over ∼2% (e.g. see Fig. 2 in Dazé et al., 2003) seriously undermining the precision (and accuracy) of ⁴⁰Ar/³⁹Ar dating technique. Most recent studies involving FCs as a fluence monitor refer to an age of 28.02 ± 0.16¹ Ma (relative to the decay constants of Steiger and Jäger, 1977) as determined by Renne et al. (1998). This age is based on calibration with a primary standard (GA-1550) showing excellent reproducibility at the single grain level (Renne et al., 1998). In addition the K concentration of GA-1550 has been measured by isotope dilution (Renne et al., 1998) and the ⁴⁰Ar* concentration has been determined by the ‘first principles’ K/Ar dating (i.e. using calibration against a known volume of argon air) (McDougall and Roksandic, 1974).

Spell and McDougall (2003) also performed an intercalibration between FCs and GA-1550, obtaining an age of 28.10 ± 0.04 Ma for FCs based on their revised age of 98.5 ± 0.8 Ma for GA-1550, or 28.10 ± 0.15 if the error is recalculated to include errors on K and Ar* concentrations of the primary standard. The revised age for GA-1550 was based on a re-evaluation of flame photometry data for the K concentration. Although the ages reported by Renne et al. (1998) and Spell and McDougall (2003) for FCs based on calibration with the same primary standard (GA-1550) agree within the stated errors at the 1σ level, it is noteworthy that the intercalibration factors themselves ($R_{\text{FCs}}^{\text{GA-1550}}$) are 3.5957 ± 0.0038 and 3.575 ± 0.005, respectively, and do not agree at the 2σ level.

Lanphere and Baadsgaard (2001) used another primary standard, SB-3, to determine an age of 27.57 ± 0.09 Ma for FCs. This age is at the young end of the spectrum of ⁴⁰Ar/³⁹Ar ages in use today, and is quite clearly distinct within stated errors from the values derived from the other first-principles standard, GA-1550.

The main purpose of this paper is to clarify the absolute ⁴⁰Ar*/⁴⁰K ratio of FCs via intercalibration with other primary standards whose ⁴⁰Ar* and ⁴⁰K concentrations were measured independently in different laboratories using different ultimate calibrations. In this way, non-systematic errors in e.g. tracer calibrations are averaged, at least to some extent. We restricted our choice of primary standards to those whose K concentration have been determined by isotope dilution (due to its typically superior reproducibility and accuracy over flame photometry measurements) and whose ⁴⁰Ar* concentrations have been measured using ‘first principles’ (Lanphere and Dalrymple, 2000) or can be related quantitatively to first-principles standards. Thus this paper reports new intercalibration data for FCs against four such primary standards. Three of these are used as interlaboratory standards (Hb3gr hornblende, NL-25 hornblende and GA-1550 biotite) and the fourth is an intralaboratory standard formerly used in the UC Berkeley K–Ar lab (GHC-305 biotite). We discuss previous intercalibrations between FCs and MMhb-1, SB-3 biotite and LP-6 standards and we evaluate the homogeneity of the NL-25 hornblende at the single grain level which has not been addressed before.