Chemometrics and metrology
Multivariate calibration for ICP-AES
Inductively Coupled Plasma – Emission Spectrometry (ICP-AES) is well established as a powerful multielement technique, but can suffer from both spectral and non-spectral interferences. The nature of the interferences is often complex; hence it is not always been possible to apply the required corrections for accurate analysis. Many attempts have been made to circumvent spectral interferences in ICP-AES. The most common being optimisation of line selection using ICP wavelength tables, however, this approach is limited in that satisfactory appraisal of spectral interferences requires comprehensive ICP tables.
A key area for data enhancement is through the use of multivariate calibration. The advantages of such techniques are that lengthy sample pre-treatment steps to remove interferences are avoided, full spectrum modelling is possible, and simultaneous calibration of all the matrix components is achieved. In the present work training and test data-sets have been designed using the theory of orthogonal arrays, and the data acquired using simultaneous ICP-AES. Multivariate calibration has been approached in two ways. First, integrated peak area data has been acquired for 200 separate analytical lines of 18 elements at 17 levels, and multivariate calibration has been performed using Multiple Linear Regression (MLR), Partial Least Squares (PLS), Principal Component Regression (PCR), and a multiple modelling approach. Second, investigations have been made into ways of utilising the raw ICP-AES spectrum in multivariate calibration using Principal Components.
Publications
Griffiths, M.L., Svozil, D., Worsfold, P., Denham, S. and Evans, E.H. (2002). Variable reduction algorithm for atomic emission spectra: application to multivariate calibration and quantitative analysis of industrial samples. Journal of Analytical Atomic Spectrometry 17, 800-812.
Griffiths, M.L., Svozil, D., Worsfold, P., Evans, E.H. (2006). The application of piecewise direct standardisation with variable selection to the correction of drift in inductively coupled atomic emission spectrometry. Journal of Analytical Atomic Spectrometry 21, 1045-1052.