Huntley et al. (Nature 313, 105-107, 1985) proposed that the time since grains of quartz and feldspar were last exposed to sunlight could be estimated from their photon-stimulated luminescence (PSL) signals and knowledge of the environmental dose rate. They noted the potential to obtain ages from individual sub-samples (aliquots), but a working single-aliquot protocol was not devised for K-feldspars until 1991, and this was later extended to individual sand-sized grains of K-feldspar. The next breakthrough was made in 1997 with the development of the 'Single-Aliquot Regenerative-dose' (SAR) protocol for single aliquots and individual grains of sand-sized quartz. Since then, several improvements have been made to this protocol, including additional validation tests of protocol performance. Most current 'small sample' studies of sand-sized quartz and K-feldspar use the SAR protocol, which has been extended to silt-sized grains, the 'linear modulation' method of photon excitation, and the red PSL and thermoluminescence (TL) emissions from these minerals (until recently, attention has been focussed on their ultraviolet-blue PSL and TL signals).

With the SAR protocol, inter-aliquot normalisation is not required; any sensitivity changes that occur during reconstruction of the burial dose can be tracked and accounted for; the extent to which samples have been adequately bleached by sunlight before deposition can be assessed; and certain types of post-depositional disturbance can be identified. All of these improve the accuracy and precision of the final dose and age estimates. Also, little raw material is required for single-aliquot or single-grain analysis, thus facilitating dating of small samples and those that contain few quartz or K-feldspar grains. Deposits ranging from modern to 0.5 Ma, perhaps as old as 1 Ma, in age may be dated by PSL techniques, with a typical uncertainty on the age of 5-10%.

In my talk, I shall discuss some of the latest developments in 'small sample' PSL dating, with reference to recent applications in the Earth and archaeological sciences. As well as improvements in the accuracy of the PSL age estimates, I shall describe attempts to push the limits of PSL dating beyond 1 Ma and to investigate processes of sediment mixing. An overview will also be provided of the current limitations to PSL dating.