Blue-light stimulated luminescence dating of gypsum in loess deposits

Liu, Yijing , Wang, Leibin , Zhao, Hui , Deng, Huizhen , Xiao, Fuan

2026-06-01 RADIATION MEASUREMENTS 2026   194(卷), null(期), (null页)

查看原文

Gypsum veins in loess deposits are valuable paleoclimate archives in arid and semi-arid regions. Although previous thermoluminescence (TL) studies on gypsum have demonstrated its potential for luminescence dating, the blue-light stimulated luminescence (BLSL) characteristics of gypsum in such settings remain unexplored. We applied a low-temperature BLSL dating approach to investigate gypsum veins from the Jiuzhoutai loess section on the Chinese Loess Plateau. Determination of the equivalent dose (D-e) was performed using the single-aliquot regenerative (SAR) protocol. A series of experimental conditions, including tests of preheat temperature, stimulation temperature, OSL bleaching temperature, and dose recovery, were used to evaluate the feasibility of the SAR-BLSL protocol for D-e determination of the gypsum in the loess deposits. Pulse-annealing and anomalous fading tests were used to assess signal stability. Pulse annealing experiments yielded parameters of E approximate to 1.87 eV and the frequency factor (s) of 1.16 & times; 10(17) sec(-1), corresponding to an electron lifetime of similar to 6.05 & times; 10(8) y, at an ambient temperature of 10 degrees C for first-order kinetics assumption. This indicates potential thermal stability at least on million-year timescales. Nevertheless, measurable anomalous fading was observed (g(2d) = 9.35 +/- 1.20%) that warrants further evaluation for age correction. TL signal loss tests confirmed the efficient removal of traps associated with the similar to 280 degrees C TL peak, while also inducing an unstable, photo-transferred TL peak at similar to 300 degrees C that is most pronounced during early stimulation. Consistently, D-e(t) analysis shows a plateau at similar to 18-22 s (309 +/- 5 Gy) that agrees with the ITL-derived D-e (316 +/- 4 Gy) within uncertainty, supporting use of the late-decay interval to better isolate the slow component. Dose-response curves are well fitted by a single-exponential function. Regenerative doses up to 4000 Gy constrain D-0 similar to 600-700 Gy, and the average D-e results yielded a value of 144 +/- 6 Gy. After applying an experimental g value-based fading correction, the BLSL age of gypsum was calculated as 256 +/- 80 ka. We conclude that BLSL dating is applicable to the dating of gypsum within loess deposits, but routine tests of signal stability and cautious interpretation of BLSL ages are required because of potential fading.