Fundamental and sedimentological controls on luminescence behaviour in quartz and feldspar

Abstract

The optically stimulated luminescence (OSL) characteristics of a suite of quartz and feldspar samples from a range of modern glaciofluvial sediments have been explored to determine the use of OSL as a depositional pathway tracer. Paraglacial and subglacial source material and various glaciofluvial deposits have been analysed from the glacial catchments of Bergsetbreen, Fåbergstølsbreen, and Nigardsbreen as well as the Fåbergstølsgrandane sandur, Jostedalen, Norway. The OSL distribution signatures have been characterised through exploration of sample skewness, kurtosis and overdispersion, and dose distributions of the different depositional settings and source materials are distinct for both quartz and feldspar. Residual ages are greatest for feldspar, indicating significant potential age overestimation where feldspar is used to date glaciofluvial deposits. Sample dose distributions and overdispersion characteristics are driven by source sediment properties, whereas residual ages are controlled by transport and depositional processes. Those transport and depositional processes which result in significant light exposure, also influence dose distributions, and processes that sort sediments least effectively have the highest residual doses. Sample OSL characteristics, transport distance and grain size distributions have been investigated using factor analysis, as a means of predicting sediment source, facies, depositional process and deposit type. Although the depositional processes of the quartz samples can be clearly differentiated based upon OSL characteristics, factor analyses of feldspar and grain size characteristics are inconclusive. The application of quartz OSL to the Norwegian samples was limited by its very poor luminescence sensitivity. Quartz is the preferred mineral for OSL, however, despite the plethora of successful quartz OSL applications, the precise origin of the UV/blue luminescence emission, measured during OSL, remains unclear. The origins of this emission and controls on its intensity were explored using a variety of spectroscopic techniques including photoluminescence, cathodoluminescence, radioluminescence (RL), ionoluminescence (IL) and x-ray excited optical luminescence (XEOL). Exciting sample luminescence at a range of energies enables exploration of the different donor centres responsible for the luminescence emission. Cathodoluminescence and RL emission spectra are similar, comprising broad emissions at 1.5, 2.0 and 2.7 eV (detection in the UV part of the spectrum was not possible for these experiments). Ionoluminescence emission spectra were dominated by the ~ 3.3 eV emission, which is a component of the signal conventionally monitored during OSL. This emission depleted as a function of dose, to the benefit of the red emission (1.8-2.0 eV) for all samples throughout IL, and similar observations were made for the 3.4 eV emission observed from the XEOL emission spectra. The XEOL spectra are dominated by an emission at ~ 3.8 eV, not widely reported for quartz, which has tentatively been attributed to peroxy linkages. Differences between the IL and XEOL emission spectra are interpreted as evidence for the presence of multiple excited states.