A new understanding of luminescence processes in feldspar using novel site-selective spectroscopic techniques

Metastable states in solids are widely used for dosimetry and photonic applications. Feldspar, a ubiquitous naturally occurring aluminosilicate, consists of many defects and impurities; some of these transform into metastable states by capturing electrons or holes, when exposed to ionizing radiation.

These metastable states can have lifetimes of millions of years rendering feldspar useful for luminescence geochronology. In this dating technique, the dose-dependent concentration of the metastable states (generated by ionizing radiation) is measured via optically stimulated luminescence (OSL) or infrared stimulated luminescence (IRSL) signals.

These signals are generated by charge transfer across the metastable states, followed by electron-hole recombination resulting in the emission of light. Despite many decades of research, the luminescence mechanisms and the associated defect system in feldspar are poorly understood; for example, the defect responsible for the main dosimetric trap (i.e. principal trap) and its physical characteristics are still unknown.

This lack of knowledge may largely be attributed to the inherent physical processes involved in OSL and IRSL generation. The OSL/IRSL technique is not ideal for characterizing the principal trap (e.g. optical trap depth, electronic states, number of defects and their concentration, etc.) as it involves both electron and hole sites as well as the charge transport dynamics, making any interpretation of the electron-trapping state ambiguous.

Therefore, it is desirable to use site-selective methods that can directly probe the principal trap without involving any hole sites in the emission process. The main purpose of this Ph.D. research is to advance our current understanding of the luminescence processes in feldspar and the associated defect system using site-selective multi-spectroscopic techniques.

This work shows that there are two principal traps in K-Na feldspar. These traps emit Stokes-shifted infrared photoluminescence (IRPL) bands centered at 1.41 eV (880 nm) and 1.30 eV (955 nm). The two trapping centers have similar electron capture cross-sections and excited-to-ground state relaxation lifetimes, but different trap depths and excited-state energies.

These results suggest that the 1.41 eV and 1.30 eV emission centers consist of the same defect that resides in two different sites and, thus, experiences different crystal fields. Cathodoluminescence (CL) microscopy explores the question on the spatial variability of the two principal traps and their link to feldspar composition.

CL investigations suggest that the two emission centers (i.e. the two traps) vary spatially even within a single-grain of feldspar and their relative emission peak intensity (1.30 eV vs. 1.41 eV) shows a correlation with the K-Na content. This work sheds new light on the long-standing issues of estimation of trap depth in feldspar, and whether there are single or multiple traps giving rise to the OSL/IRSL signals.

This Ph.D. research also establishes a link between the IRPL emission bands (1.41 eV and 1.30 eV) and the OSL/IRSL phenomenon. Tracking of changes in IRPL (i.e. trapped electron population) due to IRSL (i.e. electron and hole populations) shows that a) both the 1.41 eV and 1.30 eV centers participate in IRSL, and b) only a fraction of the principal trap population participates in the IRSL at a given measurement temperature.

A comparison of thermal depletion of IRSL and IRPL signals suggests that the trapped electrons in the principal trap are quite stable up to about 400 ºC. The decrease in IRSL because of preheating to 300- 400 ºC occurs due to the depletion of holes; the holes are used up during the TL measurement (i.e., preheating) prior to the IRSL measurement.

Furthermore, it is observed that the electron trapping probability in the principal trap is both a function of its electron capture cross-section and its distance to the nearest hole. This new understanding is anticipated to play a crucial role in the development of mathematical models of luminescence phenomena involving metastable states.

Finally, the test of the potential of IRPL in sediment dating suggests that IRPL can be successfully adapted to a SAR protocol; it recovers accurate equivalent doses from 100 to 300 Gy (age range 20-128 ka) without a fading correction. In terms of practical utility, a new measurement facility for detecting infrared photoluminescence (IRPL) at 1.41 eV (880 nm) and 1.30 eV (955 nm) for routine dosimetric measurements has been developed.

Furthermore, a dose measurement protocol, i.e. coupled IRPL-IRSL SAR protocol, is developed to measure natural doses in feldspar using IRPL. This work establishes a fundamentally different dating technique based only on trapped electrons, compared to the existing OSL and IRSL dating techniques.