Rapid Analysis Method of Selenium Content in Human Hair Using High Precision X-ray Fluorescence Spectroscopy and a Double-Film Thin Sample Technique

To address the challenges of complex sample pretreatment and the difficulty of on-site rapid screening for selenium (Se) detection in human hair, this study established a rapid analytical method based on high precision X-ray fluorescence (HPXRF). This work was motivated by the widespread selenium deficiency in Chinese soils and its subsequent impact on human selenium nutrition through the food chain. The proposed method employs high-sensitivity, low-background-noise HPXRF technology and simplifies sample pretreatment to merely cutting the hair into 1–2 mm segments. A custom-designed double-film thin sample preparation device was introduced to precisely control sample thickness, mass, and distribution, significantly improving measurement parallelism and accuracy. Systematic experiments demonstrated that the double-film thin sample method significantly enhances sample uniformity and signal stability compared to conventional loose sample preparation, increasing the peak-to-background ratio of the Se characteristic peak by approximately 46% and effectively reducing matrix effects. A stable fluorescence signal was obtained with only 125 mg of sample, meeting the analysis requirements for trace biological samples. Samples prepared with this method exhibited good uniformity, enabling highly repeatable results without the need for a rotating platform. For core component selection, a 6 mm² Si-PIN detector was chosen, as it satisfies the sensitivity requirements for trace selenium detection while offering a significant cost-performance advantage. Method validation showed a good linear relationship for Se in the range of 314–2282 μg/kg, with a detection limit of 91 μg/kg. The rapid analysis method for selenium in human hair established in this study is simple, rapid, and reliable, providing a feasible technical approach for the on-site rapid analysis of this element in such samples.