Multivariate calibration in inductively coupled plasma mass spectrometry

To date, the existence of isobaric molecular ion interferences resulting from the formation of species such as MO exp + and MCl exp + has been considered to be a significant shortcoming of inductively coupled plasma mass spectrometry (ICP/MS). This problem has been circumvented through the use of two multivariate calibration methods, multiple linear regression (MLR) and principal components regression (PCR). The methods function by utilizing information from multiple, partially selective sensors and have been applied to the exp n Mo exp 16 O exp + / exp n+16 Cd exp + and exp n Zr exp 16 O exp + , exp n Mo exp 16 O exp + , exp n Ru exp 16 O exp + / exp n+16 Cd exp + , exp n+16 In exp + , exp n+16 Sn exp + systems. For the former system, it was possible to determine low levels of Cd in the presence of Mo, by using either MLR or PCR; traditional univariate determinations of Cd at single isotopes were unsatisfactory. Both PCR and MLR were satisfactory for determining Cd in the presence of Zr, Mo, Ru, indium, and Sn; moreover, no major advantages of PCR over MLR were observed. The ICP/MS application of MLR and PCR, and that of other multivariate calibration approaches not considered herein, merits further study. Graphs. 42 ref.--AA