Research Paper Infrared stimulated luminescence measurements of single grains of K-rich feldspar for isochron dating Bo Li a , Sheng-Hua Li a, * , Geoffrey A.T. Duller b , Ann G. Wintle b a Department of Earth Sciences, The University of Hong Kong, Pokfulam Road, Hong Kong, China b Institute of Geography and Earth Sciences, Aberystwyth University, Aberystwyth SY23 3DB, UK article info Article history: Received 4 February 2009 Received in revised form 2 February 2010 Accepted 11 February 2010 Available online 23 February 2010 Keywords: K-feldspar IRSL Single grain Luminescence Recuperation Isochron Dating abstract This paper explores the use of single grain luminescence measurements in isochron dating of K-rich feldspars. The thermal stability of individual feldspar grains was investigated using pulse annealing methods, which appears to distinguish between K-rich and Na-rich feldspars. A good isochron fit was obtained using synthetic aliquots produced from the single grain data set and the age obtained based on an assumed K content of 13  1% was in good agreement with that obtained using single aliquot measurements (and with other age control). Ó 2010 Elsevier B.V. All rights reserved. 1. Introduction A method of dating has been proposed recently in which the infrared stimulated luminescence (IRSL) of potassium-rich feldspar is measured (Li et al., 2007, 2008a,b), the isochron IRSL (iIRSL) method. In this method, the equivalent doses (D e ) for grains of different diameter were determined using the IRSL signal. Isochrons were constructed by plotting the D e values as a function of the internal dose rates. The method is similar to the subtraction method proposed by Fleming and Stoneham (1973) for thermoluminescence signals from quartz and feldspar and to more recent isochron methods (e.g. Mejdahl, 1983; Clark, 1994; Zhao and Li, 2002). Any isochron dating method is critically dependent on the accuracy and precision of the data, and the iIRSL method is no exception. The IRSL measurements were made on aliquots made up of several thousand grains that had been extracted by heavy liquid separation (selection of grains lighter than 2.58 g cm 3 ) after sieving to obtain the relevant grain sizes; the age was determined from the slope of the best-fit line. It was assumed that this prepa- ration procedure resulted in potassium-rich grains, enabling the internal dose rate to be estimated using an assumed content of 13  1% (Huntley and Baril, 1997; Zhao and Li, 2005; Li et al., 2008a). The accuracy and precision of the D e values obtained for a single grain size fraction will depend upon whether the IRSL signals are indeed mainly from KF or from some other feldspars (e.g. sodium- rich feldspar, NaF) that are present. In the work reported here, single grain measurements have been used to investigate variations in thermal stability and other lumi- nescence characteristics at a smaller observation scale and assess the potential for improving precision in isochron dating by the use of synthetic aliquots. These aspects are investigated by making IRSL measurements on single grains. The presence of NaF grains in the lighter than 2.58 g cm 3 fraction is determined by pulse annealing experiments, as previously used by Tso et al. (1996) to determine the relative thermal stability of KF and NaF. The impact of the presence of NaF grains on the D e obtained using a multiple grain aliquot is ascertained by measuring the D e values for several hundred indi- vidual grains. The potential of using single grain measurements and appropriate rejection criteria to isolate KF grains, and therefore improve the precision of D e determination, is also explored. 2. Samples and experimental procedures Two sedimentary sand samples from desert areas in Northern China were selected for this study. Sm1 was from below the * Corresponding author. E-mail address: shli@hku.hk (S.-H. Li). Contents lists available at ScienceDirect Quaternary Geochronology journal homepage: www.elsevier.com/locate/quageo 1871-1014/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.quageo.2010.02.003 Quaternary Geochronology 6 (2011) 71e81