The 176Lu decay constant determined by Lu–Hf and U–Pb isotope systematics of Precambrian mafic intrusions

https://doi.org/10.1016/S0012-821X(04)00012-3Get rights and content

Abstract

At present, there is large uncertainty in the decay constant of 176Lu needed for Lu–Hf isotopic studies. We have determined λ176Lu by cross-calibration of U–Pb and Lu–Hf isotopic systems on mineral fractions from the Proterozoic Karlshamn and Sorkka dolerites in Sweden and Finland. The dolerites crystallized at shallow depths from homogeneous, high-temperature magmas, carry olivine that is nearly 100% unaltered, and show no signs of post-magmatic isotopic disturbance. The Lu and Hf isotopic compositions of plagioclase, olivine, pyroxene, apatite, ilmenite and baddeleyite were determined by multicollector-inductively coupled plasma mass spectrometry (MC-ICPMS). Calibrating the Lu–Hf results against baddeleyite U–Pb dates of 954.1±1.2 and 1256.2±1.4 Ma for the dolerites yields a mean λ176Lu of 1.867±0.008×10−11 year−1. The pristine character of the rocks and the agreement of the λ176Lu values with those from other terrestrial data sets [E. Scherer et al., Science 293 (2001) 683–687] suggest that the true value of λ176Lu lies between 1.86 and 1.87×10−11 year−1. Calibration experiments on extraterrestrial samples give significantly higher (4–6%) values, a discrepancy that may be due to plotting of non-cogenetic samples on the same Lu–Hf isochron diagram, or may have other, as yet undetermined, causes. The result of this study also indicates that the Lu–Hf method is capable for dating the crystallization of mafic rocks. The high 176Lu/177Hf ratio in apatite suggests that intrusive ages can be determined at a precision of a few million years or better.

Introduction

After 25 years of Hf isotopic work on both terrestrial and extraterrestrial samples, it has recently become clear that the decay constant (λ) of 176Lu is not defined to better than 5% (see review by Begemann et al. [2]). Physical determinations of λ176Lu display considerable internal disagreement, with a range of more than 13% for determinations over the past 20 years [3], [4], [5], [6], [7], [8]. Partly for this reason the attention of the isotope earth science community has focused on determination of λ176Lu by geochronologic analysis of samples of known age. So far, age comparison experiments based on meteorites (e.g. [9], [10]) suggest higher values for λ176Lu than those based on terrestrial igneous intrusions dated by U–Pb systematics [1]. There is an urgent need to determine the true λ176Lu, and to discover what causes the scattered values between samples of different origins.

In this study we have investigated two Precambrian dolerites for determination of λ176Lu. The value of the decay constant is calculated from the relationship λ176Lu=ln(m+1)/t, where m equals the slope of the Lu–Hf mineral isochron and t is the age of intrusion constrained by U–Pb baddeleyite dates. In addition to the abundance of baddeleyite, these samples were selected on basis of their extremely fresh mineralogy and textures typical for crystallization from a homogeneous magma. To confirm that the Lu–Hf results fulfill the isochron criteria we determined the Lu–Hf systematics of all main mineral phases in the dolerites. This may be important because, until now, age comparison experiments on terrestrial samples have been limited to Lu–Hf analysis of a few accessory phases, which precludes evaluating whether the result satisfies the isochron requirements. If those samples suffered from slow cooling, magma heterogeneity or later isotopic disturbance might explain the inconsistent results determined from terrestrial and extraterrestrial data sets.

Section snippets

Sample description and analytical procedures

The ∼100 m wide Karlshamn dike in southern Sweden and the ∼50 m thick Sorkka sill in western Finland were sampled for U–Pb and Lu–Hf isotopic work. Samples were taken in quarries, tens of meters from the intrusive contacts, and at least 10 m below ground surface. Both samples are extremely fresh without any signs of metamorphism or alteration, as indicated by the preservation of primary olivine (Fig. 1). These samples have no phenocrysts or xenoliths, and show equigranular textures that are

U–Pb

Six baddeleyite fractions from the Karlshamn dolerite sample appear as concordant to slightly discordant in the U–Pb concordia diagram (Fig. 2). The upper intercept age is 954.2±1.1 Ma with a lower intercept age indistinguishable from zero at −58±190 Ma; MSWD=1.46.

The five single baddeleyite grains from the Sorkka sample are concordant with intercepts at 1256.2±1.4 and 319±288 Ma; MSWD=0.57 (Fig. 2). This result is identical with the age of 1258±10 Ma earlier reported for this dolerite [11]. By

Possible errors in the determined decay constant value from age comparison experiments

A linear regression in the Lu–Hf isochron diagram can be considered a true isochron only if the samples have had identical 176Hf/177Hf ratios at the time of crystallization. For determining λ176Lu by calibrating the Lu–Hf results against baddeleyite U–Pb dates, diffusion of the age-determining elements U, Pb, Lu and Hf must cease at the same time, or within a time interval of insignificant length. The duration of the interval is determined by the difference in closure temperature (Tc) for the

Summary

We have determined the Lu and Hf isotope compositions of clean mineral separates from two Precambrian dolerites from Sweden and Finland. The significance of the results is manifested by distinct linear relationships between analyses plotted in the 176Lu/177Hf vs. 176Hf/177Hf diagram (MSWD ∼0.9 for both samples). Textures and field observations demonstrate that these samples crystallized at shallow depths from high-temperature, homogeneous magmas. Together with experimental diffusion data,

Acknowledgements

We are extremely grateful to Hannu Huhma and Matti Vaasjoki for guidance in the field, and additionally to Olavi Kouvo for the availability of his pioneering baddeleyite separates of Finnish samples. This contribution benefited from critical reviews by Yuri Amelin and an anonymous reviewer. This research was sponsored by The Swedish Foundation for International Cooperation in Research and Higher Education (STINT) as a personal fellowship program to U.S., and by NSF-EAR-0003343.[KF]

References (44)

  • L Heaman et al.

    Paragenesis and U–Pb systematics of baddeleyite (ZrO2)

    Chem. Geol.

    (1993)
  • Y Amelin et al.

    Geochronology of the Voisey’s Bay intrusion, Labrador, Canada, by precision U–Pb dating of coexisting baddeleyite, zircon, and apatite

    Lithos

    (1999)
  • G.H Riley

    Excess Sr87 in pegmatitic phosphates

    Geochim. Cosmochim. Acta

    (1970)
  • D.J Cherniak

    Rare earth element diffusion in apatite

    Geochim. Cosmochim. Acta

    (2000)
  • B.J Giletti et al.

    Strontium diffusion kinetics in plagioclase feldspars

    Geochim. Cosmochim. Acta

    (1994)
  • E Scherer et al.

    Calibration of the lutetium-hafnium clock

    Science

    (2001)
  • J Sato et al.

    The half-life of 176Lu

    Radiochem. Radioanal. Lett.

    (1983)
  • A.P Sguigna et al.

    The half-life of 176Lu by a γ-γ coincidence measurement

    Can. J. Phys.

    (1982)
  • G.F Grinyer et al.

    Half-life of 176Lu

    Phys. Rev. C

    (2003)
  • R.J Gehrke et al.

    Half-live of 176Lu

    Phys. Rev. C

    (1990)
  • P.J Patchett et al.

    Lu–Hf total-rock isochron for the eucrite meteorites

    Nature

    (1980)
  • M Bizzarro et al.

    Early history of Earth’s crust–mantle system inferred from hafnium isotopes in chondrites

    Nature

    (2003)
  • Cited by (2565)

    View all citing articles on Scopus
    View full text