Systematic analysis of K-feldspar 40Ar/39Ar step heating results II: relevance of laboratory argon diffusion properties to nature

https://doi.org/10.1016/S0016-7037(01)00846-8Get rights and content

Abstract

We examine a database containing the results of 40Ar/39Ar step-heating experiments performed on 194 basement K-feldspars to recover thermal history information. Qualitative examination of 40Ar/39Ar systematics reveals that about half of the K-feldspars examined are sufficiently well behaved to be suitable for thermal history analysis. Correlation algorithms are developed to quantitatively assess the degree to which age and 39Ar release spectra are compatible with the same volume diffusion process. Upon applying these methods, we find that 65% of all samples yield correlation coefficients in excess of 0.8, whereas roughly 40% give values above 0.9. We further compare the observed correlation behavior with that predicted from the multidiffusion domain model and find good agreement for samples with correlation coefficients above 0.9. In contrast, hydrous phases unstable under in vacuo heating and K-feldspars with highly disturbed age spectra yield poorly correlated age and diffusion properties. The high degree of correlation exhibited by the majority of K-feldspars we have analyzed validates extrapolation of experimentally determined diffusion properties to conditions attending natural Ar loss within the crust. Despite this, a significant number of basement K-feldspars analyzed by the step-heating method yield 40Ar/39Ar systematics that are clearly problematic for thermal history analysis. We numerically explore the effects of low-temperature alteration of K-feldspar on thermochronological analysis and identify a range of conditions under which information is progressively lost. Finally, we demonstrate the insensitivity of thermal history calculations to detailed knowledge of the diffusion mechanism by introducing the heterogeneous diffusion model. We find that the multidiffusion domain approach can successfully recover imposed thermal histories from heterogeneous diffusion-type crystals and conclude that most details of the interpretive model employed are of secondary importance. The only requirement for recovering thermal histories from K-feldspar 40Ar/39Ar step-heating results is that argon loss proceeds by volume diffusion and that laboratory argon release adequately mimics the natural diffusion boundaries and mechanisms—a requirement implicitly met by those samples exhibiting high degrees of correlation.

Introduction

Because tectonic processes characteristically alter the distribution of heat within the crust, the ability to determine time-dependent paleotemperature variations by means of mineral thermochronometry has increasingly been viewed as fundamental data for deciphering the evolution of basement rocks. Thermochronometry via the 40Ar/39Ar method involves modeling of the inhomogeneous internal distributions of radiogenic 40Ar (40Ar∗) produced by diffusion in silicate minerals (McDougall and Harrison, 1999). Such modeling invariably requires experimental calibration of mineral diffusion properties before quantitative thermal history estimates can be produced. Because these calibrations are necessarily performed at higher temperatures and shorter timescales than those relevant to diffusion in nature, the most fundamental issue in thermochronometry is the relevance of laboratory-determined diffusion properties to argon transport within minerals under crustal conditions.

The most promising opportunity to test this fundamental assumption is provided by step-heating experiments performed with K-feldspar. Currently, there are essentially three ways to gather relevant information about argon diffusion in silicate minerals. These include incremental extraction of 40Ar from a bulk sample in 40Ar/39Ar step-heating experiments (Berger and York, 1981), direct mapping of 40Ar/39Ar gradients via laser ablation from an array of regularly spaced melt pits (Onstott et al., 1991), and laser depth profiling (Arnaud and Kelley, 1997). Laser spot mapping permits ∼20-μm scale imaging of isotope distributions along grain surfaces but does not permit study of smaller features that may serve as important diffusion boundaries. Although laser depth profiling (Kelley et al., 1994) can improve the spatial resolution in the near-surface region by two orders of magnitude (i.e., 0.1 μm), analysis of potential diffusion boundaries is limited to original grain boundaries, cleavages, and similar features of large lateral extent exposed at the surface of grains. Step-heating methods therefore represent the only approach capable of providing volume information from the material. Although isotope distributions are not directly imaged, the natural loss mechanism is potentially reproduced.

Alkali feldspar is the most prevalent potassium-rich silicate of wide crustal occurrence that is sufficiently stable during in vacuo heating to provide a reasonable expectation that natural 40Ar∗ diffusion properties could be reproduced in laboratory step-heating experiments (McDougall and Harrison, 1999). Although isothermal, hydrothermal treatment of hydrous phases is possible, our ability to interpret such experiments is limited to bulk fusion and laser spot analysis. Moreover, the database of argon diffusion measurements in hydrous phases is relatively small Foland 1974, Giletti 1974, Harrison and McDougall 1981, Harrison et al 1985, Baldwin et al 1990, Grove and Harrison 1996. In contrast, many hundreds of diffusion experiments have been performed with K-feldspars sampled from a wide range of geologic environments (Lovera et al., 1997). The microstructural and overall stability of K-feldspar to 1100°C during in vacuo step-heating has been well documented (see Lovera et al., 1993).

The multidiffusion domain (MDD) model has proven capable of describing the full range of laboratory argon diffusion behavior exhibited by basement K-feldspars (Lovera et al., 1997). The leading alternative, the multipath model (Lee, 1995), can only approximate the observed diffusion behavior when its fundamental characteristic (i.e., interaction between diffusion domains) is minimized to such an extent that it becomes indistinguishable from the MDD approach (Arnaud and Kelley, 1997). The MDD model represents argon transport in K-feldspars as being controlled by volume diffusion processes acting on a distribution of length scales. It has been extensively tested and developed since its inception more than a decade ago Lovera et al 1989, Lovera et al 1991, Lovera et al 1993, Lovera et al 1997, Richter et al 1991, Harrison et al 1993, Harrison et al 1994, Quidelleur et al 1997. Although the ability of the MDD model to represent laboratory diffusion properties has become firmly established, its ability to successfully recover basement thermal histories has been repeatedly questioned and remains controversial Parsons et al 1988, Parsons et al 1999, Burgess et al 1992, Foland 1994, Arnaud and Kelley 1997.

Our principal goals in this article are as follows: (1) to identify problematic behavior in K-feldspar 40Ar/39Ar age spectra for thermal history recovery; (2) to examine our database of K-feldspar step-heating experiments to determine the frequency with which misbehaving samples occur; and (3) to quantitatively evaluate the compatibility of 40Ar/39Ar age and laboratory diffusion data from samples within this database to further determine the proportion of basement K-feldspar samples that are likely to be amenable to thermal history analysis. In discussing recent criticisms of the MDD model we also illustrate, by way of numerical simulation, how low-temperature recrystallization or an error in our assumption of the diffusion mechanism might impact calculated thermal histories. Combined with the results from our database samples, these calculations demonstrate the viability of recovering crustal thermal histories through application of the MDD model provided that appropriate samples are analyzed.

Section snippets

Expanded database of K-feldspar 40Ar/39Ar step-heating results

In this section, we examine K-feldspar results obtained between 1994–2000 (Table 1; additional data available from http://oro.ess.ucla.edu/data_repository.html). This database is expanded over that presented in LOVERA et al., (1997), which focused primarily on the ability of the MDD model to represent laboratory argon diffusion properties of K-feldspar. Our primary goal here is to evaluate the self-consistency of natural and laboratory argon diffusion in K-feldspar.

Simple volume diffusion

Correlation between age and laboratory diffusion spectra

The preceding qualitative analysis of our database of step-heating results from basement K-feldspars provides only a general impression of the frequency at which samples yield age spectra compatible with simple volume diffusion theory. We now seek to take our analysis of database samples a step further. Under conditions of monotonic cooling, the MDD model predicts a high degree of correlation between the age spectrum (which typically represents millions to hundreds of millions of years of 40Ar∗

Discussion

Because argon transport in nature cannot be directly observed, speculation regarding what exactly controls the mobility of this trace constituent in geologic materials is necessarily based on circumstantial evidence. Different approaches made on the basis of examination of microscopic and macroscopic properties have been employed to ascertain the validity of extrapolating laboratory properties to natural behavior. We favor a macroscopic approach made on the basis of examination of the

Concluding remarks

The high degree of correlation exhibited between measured age and log(r/ro) spectra determined for the majority of K-feldspars we have analyzed validates extrapolation of experimentally determined diffusion properties to conditions attending natural argon loss within the crust. Our experimental results and numerical analysis indicate that the MDD model, although potentially not a sensitive probe of K-feldspar microstructure, can be a robust tool to recover thermal histories. Despite this, a

Acknowledgements

This research was conducted under the auspices of Department of Energy grant DE-FG-03 to 89ER14049. We thank the numerous analysts involved in generating the 40Ar/39Ar data reported in this article and the assistance of Nyssa Roeth, Ainslie Harrison, and Natalı́ Degrati for their help in compiling the database.

Associate editor: F. J. Ryerson

References (54)

  • O.M. Lovera et al.

    Systematic analysis of K-feldspar 40Ar/39Ar step heating resultsI. Significance of activation energy determinations

    Geochim. Cosmochim. Acta

    (1997)
  • K.I. Mahon et al.

    The thermal and cementation histories of a sandstone petroleum reservoir, Elk Hills, California.Part I: 40Ar/39Ar thermal history results

    Chem. Geol.

    (1998)
  • T.C. Onstott et al.

    Laser microprobe measurement of chlorine and argon zonation in biotite

    Chem. Geol.

    (1991)
  • I. Parsons et al.

    Argon-loss by alkali feldspar

    Geochim. Cosmochim. Acta

    (1988)
  • P.W. Reiners et al.

    Helium diffusion and (U-Th)/He thermochronometry of titanite

    Geochim. Cosmochim. Acta

    (1999)
  • F.M. Richter et al.

    Tibetan tectonics from 40Ar/39Ar analysis of a single K-feldspar sample

    Earth Planet. Sci. Lett.

    (1991)
  • T. Tagami et al.

    Natural long-term annealing of the zircon fission-track system in Vienna Basin deep borehole samples; constraints upon the partial annealing zone and closure temperature

    Chem. Geol.

    (1996)
  • G. Turner et al.

    Excess argon, crustal fluids and apparent isochrons from crushing K-feldspar

    Earth. Planet. Sci. Lett.

    (1992)
  • I.M. Villa

    Multipath Ar transport deduced from isothermal heating experiments

    Earth. Planet. Sci. Lett.

    (1994)
  • I.M. Villa et al.

    Geochronology and excess Ar geochemistry of the Lhotse Nup Leucogranite, Nepal Himalaya

    J. Volcanol. Geotherm. Res.

    (1990)
  • A.C. Warnock et al.

    Hump-shaped 40Ar/39Ar age spectra in K-feldspar and evidence for Cretaceous authigenesis in the Fountain Formation near Eldorado Springs, Colorado

    Earth. Planet. Sci. Lett.

    (1999)
  • J.A. Wartho

    Apparent argon diffusive loss 40Ar/39Ar age spectra amphiboles

    Earth. Planet. Sci. Lett.

    (1995)
  • J.A. Wartho et al.

    Direct measurement of Ar diffusion profiles in a gem-quality Madagascar K-feldspar using the ultra-violet laser ablation microprobe (UVLAMP)

    Earth. Planet. Sci. Lett.

    (1999)
  • P.K. Zeitler et al.

    Saddle-shaped 40Ar/39Ar age spectra from young, microstructurally complex potassium-feldspars

    Geochim. Cosmochim. Acta

    (1986)
  • S.L. Baldwin et al.

    Diffusion of 40Ar in metamorphic hornblende

    Cont. Mineral. Petrol.

    (1990)
  • W.L. Brown

    Feldspar in igneous rock

  • J. Crank

    The Mathematics of Diffusion

    (1975)
  • Cited by (135)

    View all citing articles on Scopus
    View full text