A two-way procedure for background correction of chromatographic/spectroscopic data by congruence analysis and least-squares fit of the zero-component regions: comparison with double-centering

Abstract Liang, Y.-Z., Kvalheim, O.M., Rahmani, A. and Brereton, R.G., 1993. A two-way procedure for background correction of chromatographic/spectroscopic data by congruence analysis and least-squares fit of the zero-component regions: comparison with double-centering. Chemometrics and Intelligent Laboratory Systems , 18: 265–279. A new procedure for detecting and correcting for baseline offset/drift and spectral background in hyphenated chromatographic data is developed. The procedure consists of several distinct steps: first, the major principal components in the zero-component chromatographic regions are extracted before the appearance of the first eluting chemical constituent and after elution of the last chemical constituent in a peak cluster. Comparison of the loading patterns of the first principal component in the two zero-component regions by means of congruence analysis is used to reveal the presence of a constant spectral background and/or systematic baseline offset or drift. If baseline drift is revealed, the baseline for the whole chromatogram is estimated by means of a least-squares fit of the data from the two zero-component regions with retention time as ‘independent’ variable. A background-corrected chromatogram is finally obtained by subtracting the estimated spectral background and the estimated baseline from the original data. The new procedure is conceptually similar to double-centering of the data. However, as shown in this work, double-centering destroys the positivity of the data and introduces an artifical ‘chemical’ rank that complicates the resolution of the data. The new method is tested and compared with double-centering using experimental multicomponent data.

[1]  Desire L. Massart,et al.  Effect of the scan time on methods based on evolving factor analysis in liquid chromatography , 1992 .

[2]  Yizeng Liang,et al.  White, grey and black multicomponent systems: A classification of mixture problems and methods for their quantitative analysis , 1993 .

[3]  Olav M. Kvalheim,et al.  A general-purpose program for multivariate data analysis , 1987 .

[4]  A comparison of the heuristic evoling latent projections and evolving factor analysis methods for peak purity control in liquid chromatography with photodiode array detection , 1992 .

[5]  The relationship of closure, mean centering and matrix rank interpretation , 1992 .

[6]  H. R. Keller,et al.  Heuristic evolving latent projections: resolving two-way multicomponent data. 2. Detection and resolution of minor constituents , 1992 .

[7]  H. R. Keller,et al.  Evolving factor analysis in the presence of heteroscedastic noise , 1992 .

[8]  R. Brereton,et al.  INVESTIGATION OF THE ALLOMERIZATION REACTION OF CHLOROPHYLL a: USE OF DIODE ARRAY HPLC, MASS SPECTROMETRY AND CHEMOMETRIC FACTOR ANALYSIS FOR THE DETECTION OF EARLY PRODUCTS , 1994 .

[9]  Paul J. Gemperline,et al.  Target transformation factor analysis with linear inequality constraints applied to spectroscopic-chromatographic data , 1986 .

[10]  H. R. Keller,et al.  Artefacts in evolving factor analysis-based methods for peak purity control in liquid chromatography with diode-array detection , 1992 .

[11]  Marcel Maeder,et al.  Evolving factor analysis, a new multivariate technique in chromatography , 1988 .

[12]  Yizeng Liang,et al.  Heuristic evolving latent projections: resolving two-way multicomponent data. 1. Selectivity, latent-projective graph, datascope, local rank, and unique resolution , 1992 .

[13]  R. Brereton,et al.  Resolution of strongly overlapping two‐way multicomponent data by means of heuristic evolving latent projections , 1993 .

[14]  Paul J. Lewi,et al.  Spectral map analysis: Factorial analysis of contrasts, especially from log ratios , 1989 .

[15]  G. Guiochon,et al.  Bias and nonlinearity of ultraviolet calibration curves measured using diode-array detectors , 1989 .

[16]  O. Kvalheim.  Oil—source correlation by the combined use of principal component modelling, analysis of variance and a coefficient of congruence , 1987 .