Portable X-ray fluorescence analysis of water: thin film and water thickness considerations

Water is requisite for life and essential for many industries. Increasingly, global water supplies include inorganic contaminants leading to millions of deaths annually. High analytical cost and a lack of field portable methods have stymied the evaluation of contaminated water. By comparison, portable X-ray fluorescence (PXRF) spectrometry has emerged as a method suitable for low cost, rapid analysis for many matrices yet few studies have evaluated liquids via PXRF. Herein, a novel means of assessing PXRF analytical performance for liquid matrices was evaluated on 1,440 samples comprised of three different standards (Cd, Cr, Pb) featuring three different film types (Kapton, Mylar, and Prolene) at five different liquid depths (4.29, 8.59, 17.18, 25.77 and 30.06 mm), and with four different concentrations (1,000, 500, 250, 125 mu g/g). To adjust the PXRF values for a liquid matrix, regression models were fitted using PXRF reported values as the predictor and the true standard concentration values as the target. Results indicated that prior to statistical adjustment to PXRF reported values, increased liquid depth as well as Mylar or Kapton film provided optimal predictive accuracy. However, after PXRF adjustment (linear for Cr and Cd, quadratic for Pb), a depth of 4.29 mm and any of the three film types provided quality elemental predictions. After PXRF adjustment, the size of the mean of PXRF difference with the known standard concentration vs. the true standard concentration values became much smaller compared to the prior adjustment difference. Additionally, the size of the difference was usually smaller for the larger depth (25.77 and 30.06 mm).