ABSTRACT A series of quantitative methods—the fundamental parameter (FP) and calibration curve methods—were evaluated for reliable routine analysis of archeological and geochemical silicic samples (e.g., pottery, clay, rock, and soil) using energy‐dispersive X‐ray fluorescence (EDXRF) spectrometry to determine their provenance. Calibration curves for 22 components (Na 2 O, MgO, Al 2 O 3 , SiO 2 , P 2 O 5 , K 2 O, CaO, TiO 2 , MnO, Fe 2 O 3 , V, Cr, Ni, Cu, Zn, Rb, Sr, Y, Zr, Nb, Ba, and Pb) were constructed from 12 geochemical reference materials (RMs). A total of 15 of the 22 components exhibited poor linearity, with correlation coefficients of 0.774–0.987. The accuracy of the calibration curve, standardless FP, and FP methods calibrated with 12 RMs was evaluated by measuring six additional geochemical RMs. Comparison with the recommended values showed that the FP method calibrated with RMs was the most accurate, whereas the calibration curve and standardless FP methods showed comparable, but lower, accuracy. These results indicate that, for EDXRF analysis of archeological and geochemical silicic samples, the FP method calibrated with RMs is more appropriate and less time‐consuming than constructing calibration curves. Furthermore, the number of RMs used to calibrate the FP method was reduced from 12 to 6, and the reliability was evaluated by measuring the same six additional geochemical RMs. The results showed that the FP method calibrated with 12 and 6 RMs achieved generally equivalent accuracy, indicating that the number of RMs can be reduced without compromising reliability, thereby lowering the calibration effort.
Ichikawa et al. (Fri,) studied this question.