The uncertainty of UV-Vis spectrophotometric and FAAS analysis for the determination of iron in pharmaceutical products

UV-Vis spectrophotometric (UV-Vis) and flame atomic absorption spectrometric (FAAS) analysis for iron determination in a pharmaceutical product were compared in terms of uncertainty budgets. Both methods are selective for iron; no interference due to other components present in the matrix was found. The results for the UV-Vis and FAAS methods were 11.4 +/- 0.5 mg and 11.9 +/- 0.8 mg, respectively, of iron per tablet (at a 95% confidence level). In both methods, the main uncertainty contribution is that due to the calibration function's non-linearity (0.32 mg and 0.46 mg or 42% and 31% of the standard uncertainty in the case of UV-Vis and FAAS, respectively). This finding encourages us to take uncertainty due to non-linearity into account in uncertainty estimations, even for highly linear methods like UV-Vis. In the ranking of uncertainty contributors, non-linearity is followed by instrument drift in the FAAS method and the uncertainty of volumetric measurements in UV-Vis. In particular, pipetting contributes about 16% of the uncertainty for UV-Vis spectrometry and 10% for FAAS, which is somewhat larger than that usually assumed. The uncertainty contributions of the two methods are compared and discussed in terms of experimental conditions.