Towards extending the luminescence dating range of quartz: exploring the 375 °C quartz TL peak

Abstract

Optically stimulated luminescence (OSL) dating of quartz is widely used to establish an absolute chronology for Quaternary sedimentary deposits. However, its applicability is in general limited to the last ~ 100-150 ka. Therefore extending the range of quartz luminescence dating beyond this limitation is a key challenge. In the quest for extending this limit, other luminescence signals from quartz have been proposed, among which violet stimulated luminescence (VSL) is a promising signal. It is based on the use of a violet stimulation (405 nm) to measure trap deeper than those accessible by blue light. The overall objective of this thesis is to develop and test the applicability of VSL dating to extend the quartz dating range. Attempts to establish an optimised single aliquot regenerative dose (SAR) protocol for VSL dating on four coarse-grained quartz samples from the coastal environment of Sardinia have not all been successful. It is found that the range of the applicability of the SAR VSL protocol is dependent on the natural dose size; using the SAR method for VSL dating remains challenging for samples with large natural doses (i.e. ~250 Gy). Subsequently, the multiple aliquot regenerative dose (MAR) protocol is, for the first time, tested on fine-grained quartz samples from a Chinese loess-palaeosol sequence in Luochuan with reference ages up to ~1400 ka. The natural VSL dose response curve (DRC) saturates at about 900 Gy, which would potentially allow dating at the Luochuan section using VSL up to ca. 300 ka. However, the application of the MAR protocol showed significant age underestimation for samples older than ~100 ka. This MAR VSL age underestimation can clearly be attributed to the different shapes of natural and laboratory DRCs; the natural signal progressively deviate from the laboratory signal beyond ~250 Gy. These observations are, however, contradicting the previous research in the same region, which showed that the MAR DRC using coarse-grained quartz samples from the Luochuan section can reproduce the shape of the natural DRC. It can therefore be concluded that the grain size plays an important role in obtaining reliable ages. A direct comparison of fine- (4-11 µm) and coarse-grained (63-100 µm) quartz VSL data from nine samples from a loess section in southern Germany further confirm these observations. It is shown that there is a systematic tendency for the fine-grained VSL ages towards underestimation with increasing age. The fine-grained MAR DRC starts to deviate from the natural DRC at ~300 Gy and therefore tends to underestimate the reference ages beyond ~100 ka. In addition to the VSL signal, the physical characteristics and applicability of the multiple aliquot methods (MAR and multiple aliquot additive dose (MAAD)) of the quartz isothermal thermoluminescence (ITL) signal measured at 330 °C (ITL330) is systematically investigated by using nine fine-grained quartz samples from the Luochuan section. The natural ITL330 shows that the signal has a theoretical dating range up to ~800 Gy, equivalent to ~230 ka. The comparison of the natural and laboratory DRCs using MAR and MAAD protocols indicates that they start to diverge in shape beyond ~200 Gy, setting an upper limit for reliable ITL330 dating of ~70 ka. However, application of pulsed-irradiation (PI) for the MAAD protocol reveals that the shape of the natural DRC can mostly be reproduced with the PI-MAAD protocol and thus it can provide reliable ages up to natural saturation at ~230 ka. Based on the observations summarised in this doctoral thesis it can be concluded that the natural VSL and ITL330 signals have an extended growth, and the main limitation is the deviation between natural and laboratory generated DRCs beyond a certain dose, which caused a progressive age underestimation. Application of pulsed irradiation increase the reproduction of the extended VSL and ITL330 natural growth. While further investigations will be needed, it appears that this method can be a promising step forward in our attempts to extend the quartz luminescence dating age range.