Elsevier

Radiation Measurements

Volume 42, Issue 10, November 2007, Pages 1618-1626
Radiation Measurements

Testing the reliability of ESR dating of optically exposed buried quartz sediments

https://doi.org/10.1016/j.radmeas.2007.09.005Get rights and content

Abstract

Optical dating of quartz by optically stimulated luminescence has a time range that is generally less than about 500 ka, due to relatively rapid saturation of the available luminescence defects in quartz. We test here a new method, electron spin resonance (ESR) optical dating of quartz, in which radiation-sensitive defects at aluminum and titanium atoms on silicon sites give rise to signals which can only be detected near liquid nitrogen temperature and which have a much higher capacity to absorb radiation dose before saturating than optical luminescence-detected signals. Our results show this method yields agreement with independent age control out to about 2.5 million years, extending here the dating range of optically exposed quartz in sediments in along-shore sediments (aeolian and waterlain) by a factor of about 5. Three sites in along-shore lacustrine and marine aeolian environments yielded very good agreement with independent age control. Details of single saturating exponential fitting in relation to agreement with expected burial doses and annealing of Al and Ti signals provide additional data to consider the best approaches to the dating method. Furthermore, we propose a new criterion for ESR optical dating: both the Al signal and Ti signal ages must agree to insure accurate burial ages. Moreover, when Al signal ages are lower than Ti signal ages, then the Al signal may be taken as the minimum burial age.

Introduction

Electron spin resonance (ESR) dating is based on the measurement of radiation-sensitive ESR signals which can begin to form after crystallization or can regrow after being reset to zero or reduced to a residual level by exposure during depositional processes. In ESR optical dating, characteristic ESR signals in detrital quartz grains are reduced or zeroed by daylight exposure, and then regrow during burial of the quartz in sediments due to exposure to natural radiation. The method resembles the better-known techniques of thermoluminescence (TL) and optically stimulated luminescence dating (Prescott and Robertson, 1997, Wintle, 1997), but has the potential to date much older deposits, possibly up to more than 5 Ma, due mainly to a slower growth to saturation of the signals compared with OSL or TL.

Walther and Zilles (1994) showed that a signal associated with aluminum (Al) defects can be reduced to between 40% and 60% of its initial value by extensive bleaching. Yoshida (1996) showed that the titanium (Ti) signal is fully zeroed after 10–20 days of sunlight, while the Al signal can only be reduced to a nearly constant residual value after more than about 100 days of sunlight. Further experiments by Toyoda et al. (2000) confirmed that the Ti–Li signal (used in the present study) can be fully zeroed, and concluded that when it and one of the two other types of Ti signals (Ti–H and Ti–Na) give concordant results, then the age obtained is reliable. However, in many cases these other signals are small or non-existent in our experience.

Laurent et al., 1994, Laurent et al., 1998 showed that aluminum ESR signals in detrital quartz grains could be used to date daylight-exposed detrital sediments with ages up to >8Ma, based on ESR ages in good agreement with biostratigraphic age estimates of alluvial sands in the Somme Basin of France. However, it still remained to be seen if ESR optical dates agreed with other independent methods of dating.

Previous studies have concluded that the dates obtained by this method are generally consistent with other estimates of age and therefore appear to be reliable. It remains to compare ESR ages with those obtained by other accepted dating methods over the multi-million year time range. Here we have carried out ESR analyses on three types of samples: (1) a dune sand from Cape Kidnappers, New Zealand whose age can be firmly established by Ar40/Ar39 and fission track ages on an associated volcanic stratum, (2) nearshore lacustrine sands from ancient Lake Bungunnia in the Murray Basin of Australia whose ages are constrained by paleomagnetic analyses, and (3) along-shore lacustrine sands in conglomerate from the archaeological site of ‘Ubediya in the Jordan Valley of northern Israel whose age is constrained by a combination of paleomagnetic, stratigraphic and biostratigraphic data.

In this paper, rather than attempting to determine the age of sites, we have assumed that the chronological age is well established by the other age criteria available at each site. We can then test the agreement between expected and observed values for the equivalent dose (DE) using various treatments of the data. The purpose of this exercise is to identify the optimal treatment of ESR data that gives the best agreement between expected and observed DE. This would then provide us with the appropriate techniques to use for ESR optical dating where the age is otherwise unknown.

Section snippets

Sample material and independent age controls

To yield accurate dates, ESR optical dating of quartz requires that the sediment sample being dated has been fully exposed to direct sunlight for periods of tens to hundreds of days. We believe that along-shore environments, where sands are generally reworked for decades of residence time in the near shore environment, provide one of the best testing grounds for this method. Both aeolian sands and waterlain sands from these environments are very likely to have received many days of accumulated

Cape Kidnappers

Sample CKDS was collected by one of us (PS) and Dave Huntley (Simon Fraser University, Physics Department) who kindly provided the sample. The external γ dose rate was determined by Huntley using a portable γ-ray spectrometer. The cosmic dose rate was zero due to the thick overburden (>100m) in the cliff exposure which has been eroded landward in the very recent geological past. The β particle dose was determined by instrumental neutron activation analysis (INAA) of the bulk sediment for U, Th

Results

We tested the effects of bleaching and thermal annealing on the CKDS sample. We first compared the decay of the Al and Ti signals during light exposure in the natural sample and one that had received an additional γ dose of 6000 Gy from the Co60 source (Fig. 2). In both samples, the Ti signal was reduced to zero after about 30 h (equivalent to 20 days of 12 h of sunlight per day). After about 80 h of SOL2 exposure, the Al signal in the natural sample was reduced to a constant residual value equal

Discussion

The analysis of the dose response curves shows that added doses well above Do in almost every case yield reliable DE values. Thus we have only weak evidence that this may be used to gain accurate age results, namely the better agreement obtained when omitting doses above Do for sample Au 9. This example suggests that in some cases Do may be useful to define the upper limit for added doses. The increased error in DE in this case may also be a consequence of having insufficient added dose points

Conclusions

Overall we find excellent promise for ESR optical dating of sediments in quartz-bearing, sand sized sediments deposited by wind and water in coastal settings. In most of the cases we have studied the Al (bleachable portion only) and Ti signals yield DE values in good agreement with values expected from the known age of samples. Further studies are needed to better understand the questions of whether added dose range can influence correct DE determination for the Ti signal. Despite this question

Acknowledgments

We thank the Natural Sciences and Engineering Research Council of Canada (NSERC), and a Premier's Research Excellence Awards of Ontario for financial support to WJR. We thank the Department of Earth Sciences, Melbourne University, Mike Sandiford (University of Melbourne) and Derek Feibel for logistical support in collecting the Australian samples. We thank K. Chancellor-Maddison for assistance with the ESR measurements of the Australian samples.

References (26)

  • Bar-Yosef, O., Goren-Inbar, N., 1993. The Lithic Assemblages of ‘Ubediya. A Lower Palaeolithic Site in the Jordan...
  • Bourdon, B., Henderson, G.M., Lundstrom, C., Turner, S.P., 2003. U-series Geochemistry. Reviews in Mineralogy and...
  • S.C. Cande et al.

    Revised calibration of the geomagnetic polarity timescale for the Late cretaceous and Cenozoic

    J. Geophys. Res.

    (1995)
  • Cited by (85)

    View all citing articles on Scopus
    1

    Current address: Stiftung Preußische Schlösser und Gärten, Berlin-Brandenburg, Abteilung Restaurierung, Postfach 601462, 14414 Potsdam, Germany.

    View full text