Resolution of multicomponent peaks by orthogonal projection approach, positive matrix factorization and alternating least squares

The application of orthogonal projection approach (OPA), alternating least squares (ALS), and positive matrix factorization (PMF) to resolve HPLC-DAD data into individual concentration profiles and spectra is discussed. OPA w as initially described as a purity method but the inclusion of an ALS proce dure allows its application as a curve resolution method. PMF is a least sq uare approach to factor analysis that in this study has been used as a tool to tackle the problem of curve resolution. OPA, ALS and PMF have been appl ied using a single matrix (two-way data) or an augmented matrix containing several data matrices simultaneously. The results obtained with the differe nt resolution methods are compared and evaluated using measures of dissimil arity between the real and the estimated spectra. The study is performed in three data subsets, obtained by segmentation of the original data matrix. Within each data subset, there is a reduced number of species present which makes the resolution easier. (C) 2000 Elsevier Science B.V. All rights res erved.