Introduction Industrial and medical applications of ionizing radiation, as well as unstable political situation worldwide, which may result in military releases of radioactivity, increase a risk of uncontrolled exposures of people to ionizing radiation. Retrospective dosimetry allowing for fast triage of victims is crucial for rescue actions. Previous studies showed that smartphones’ screens are promising for dosimetry based on electron paramagnetic resonance (EPR). A variety of EPR line-shapes in different screens, regarding background signals (BG) and radiation-induced signals (RIS), various sensitivities to interfering factors like UV light and temperature, impose serious limitations on this method. This study focuses on classification of screen glasses, taking into account their elemental compositions, EPR properties (sensitivity to UV and temperature), in order to formulate practical recommendations for dosimetry. Methods EPR spectra of 45 screens, unirradiated and irradiated with X-rays, were measured. Elemental composition of the glasses was determined using Energy Dispersive Spectroscopy. Effects of UV on samples’ EPR spectra were checked. Annealing at 200 °C enabled to evaluate effects of heating on BG and RIS. A self-written program based on c-means algorithm was used to find intercorrelations between elemental compositions and EPR features (types) of the glasses. Results and discussion Our spectra-differentiating algorithm resulted in identification of five types of glasses correlated with their elemental composition, sensitivity to X-rays, to UV and high temperature. Glasses labelled as type III and V were recommended for dosimetry due to their resistance to UV and undergoing temperature-caused bleaching of RIS without affecting their BG signals; a feature which enables reconstruction of the original BG from an irradiated sample – a key step in retrospective dosimetry. The introduced categorization of screen glasses, based on chosen features of their EPR spectra, is a simple and practical method for evaluation of their applicability in retrospective dosimetry following radiation accidents, e.g. for triage of exposed people.
Marciniak et al. (2025) studied this question.