Luminescence dating is used to estimate the amount of time that has passed since grains of a mineral such as quartz were last exposed to daylight. It is an essential technique for dating geological and archaeological sites, and can date material as old as 500 000 years. However, the measurements contain a degree of variability, the reasons for which are not understood. The RELOS project, which was supported by the European Research Council, set out to uncover the sources of this uncertainty. “This is a big unknown we think needs to be addressed,” says project coordinator Jan-Pieter Buylaert. “In lab experiments, even if we give exactly the same radiation dose to each grain, we still see an unexplained dispersion in the apparent stored charge.” In practice, post hoc justifications by the practitioner are often used to dismiss or explain dispersed results, imbuing luminescence dating with a degree of subjectivity. By identifying the source of the dispersion, Buylaert and his colleagues at the Technical University of Denmark hoped to make the process more robust.
When mineral grains are buried, they gradually accumulate energy from exposure to natural radioactivity in the surrounding matrix. This energy is stored as charge trapped in defects in the crystal structure of the mineral. When the grains are exposed to a bright light, both in nature and in the laboratory, the charge is freed, and the stored energy released as photons. This luminescence allows researchers to quantify the time the grains have been buried. One of the key assumptions in calculating the rate of charge storage is that the grains remain electrically neutral. However the RELOS project found that there is a significant build-up of charge imbalance, with some grains charging negatively, and others positively. “It is the combination of the size of the grain and the range of the radiation that causes this,” explains Buylaert. “But although we could see this effect experimentally, we were unable to make the link to the dispersion observed in natural dose distributions.” A second hypothesis suggested that the size, distribution and geometry of individual grains affect how the energy absorbed from radiation is distributed amongst the grains, and therefore the rate at which charge accumulates in one mineral grain compared to another. Buylaert’s team built complex mathematical models to account for this.
During the course of the project, Buylaert and his colleagues also set out to develop a curve for luminescence response to dose in nature, by measuring grains recovered from the Quaternary terrestrial reference site in the Chinese Loess Plateau. Here, dust has been accumulating at a steady rate for millions of years, and the age of different strata can be cross-checked using signals such as palaeomagnetism (pole reversal) and Milankovitch cycles. Unfortunately this proved not to be possible. “We found many metres of sediment eroded – jumps in ages down the section of many tens of thousands of years. This meant we could not develop a dose response curve at this site, but we were able to link these age gaps to regional and global climate phenomena,” adds Buylaert. These findings were published in ‘Nature Communications’. Buylaert says the geometry models developed under the project will aid his new project on human migration in Central Asia: “The problem will not go away, and we will take any opportunity that arises to reopen our investigations.”
RELOS, luminescence dating, radioactive, age, mineral, archaeological, radiation, charge, dispersion