Applying electron spin resonance (ESR) dating to sedimentary quartz poses a significant challenge due to the difficulty in bleaching the signals. The aim of this PhD thesis is to contribute to the improvement of a reliable ESR dating technique by quantifying and correcting ESR residual signals caused by insufficient bleaching prior to sediment deposition. Particularly within the realm of archaeology, there is a pressing need for a geochronological method capable of dating quartz and other materials well beyond the constraints of the already-established dating methods, e.g. luminescence dating. The initial study investigates the reliability of quartz ESR dating, focusing on the Ti centre. Loess from the Luochuan site on the Chinese Loess Plateau with a known reference age was used. By the use of the single-aliquot regenerative dose (SAR) protocol the equivalent dose (De) was determined for a sequence of five samples. For the two youngest samples the ESR ages were higher than the anticipated ages, indicating an incomplete signal resetting before sedimentation, whereas the oldest samples showed an approx. 20 % underestimation, indicating thermal signal loss with time. In the latter case, the apparent age could be successfully corrected for the signal loss. For the remaining samples the corrected ages were in satisfactory agreement with the expected ages. The second study focuses on the ESR residual signals caused by incomplete signal resetting before sedimentation. For this purpose, a series of early Holocene fluvial sediments with known optically stimulated luminescence (OSL) ages were analysed. Applying the SAR protocol to determine the residual doses for both the aluminium (Al) and titanium (Ti) impurity centres. It was revealed, that all investigated samples carried a considerable residual dose. To rule out methodological problems, the SAR protocol was tested for accuracy. The test revealed that the signal originating from the lithium-compensated Ti centre (Ti-Li) and a signal which originates from both the Ti-Li and the hydrogen-compensated Ti centre (Ti-H), referred to as Ti-mix, showed good results whereas the signal from the Al centre and the Ti-H signal showed that the heating and annealing steps throughout the process alter the samples ESR characteristics rendering the SAR protocol inappropriate to use in this two cases. Our findings suggest the necessity of conducting more direct comparisons between luminescence and ESR equivalent doses, with the subtraction of residual doses obtained from the difference being essential for obtaining reliable ESR ages if needed. This procedure was not used in previous work, as it was either assumed that the Ti centres were all completely bleached or, as long as the determined ages of the individual centres matched, complete bleaching was assumed. In order to apply the knowledge gained in the course of this work, the transition age from the Early Stone Age (ESA) to the Middle Stone Age (MSA) in archaeological sites near the Victoria Falls, Zambia, was dated since this technological change in stone tool making has been poorly understood by now. The combined advantages of both, single-grain OSL and single aliquot ESR dating on quartz, the good bleachability of the signals and the extended age range was used. For the same set of young samples, derived from sandy deposits bearing Stone tools, we applied both methods and found large differences in OSL and ESR equivalent doses. We estimated the mean ESR residual age by discerning the difference between OSL ages and the apparent ESR ages. Specifically emphasizing the SAR protocol, we successfully determined the mean ESR residual age for the Ti and Al centre, encompassing the non-bleachable signal component for the latter. We successfully determined the average residual age of ESR for both Ti and Al centre, incorporating the non-bleachable signal component for the latter. The size of these residual ages, ranging from 209 ± 13 ka to 268 ± 39 ka and 695 ± 23 ka to 742 ± 118 ka for the Ti centre and Al centre, which cannot be understated and must be taken into consideration. Consequently, the apparent ESR ages were adjusted by subtracting the residual age. By this we were able to get persistent residual subtracted ESR ages, which are within a 2- uncertainty when compared to the OSL ages. Eventually, we successfully pinpointed the end of the Early Stone Age at 590 ± 86 ka, establishing a maximum age for the transition to the Middle Stone Age in this specific region of south-central Africa. It should be noted that the residual ages depend on the respective geological setting, the climate and the deposition history of the sediments. Factors such as the transport route, the efficiency of bleaching by sunlight and possible rearrangements have a significant influence on whether a sediment grain was completely bleached before deposition or whether a residual signal or a residual age remains. In order to recognize and evaluate such influences, it is necessary to examine the deposition environment in detail, for example by analysing the grain size distribution, sediment structures or the composition of the sediment. Only through such sedimentological and geomorphological contextualization can ages obtained in this way be reliably interpreted. Nevertheless, we think that this approach gives a good example for the combined use of two related dating methods by carefully investigate the signal properties to unveil ESR residual signals. The age we published adds to the sparsely dated ESA/MSA transition in this geographic region.
