Principal component analysis Pyrolysis

Statistical manipulations on the USDA database (cluster analysis, principal component analysis with varimax rotation e.g., Everitt, 1980) revealed subsets of represent ve species, as idealized in Fig 2b, but with dif ent variables (orthogonal principal components) than traditional fractions as measured by USDA. A set cf species from each orthogonal subset appears in Table 1. The Latin names, and where available, the common names of the biomass species are given. The extractives ranges are ash content, 4 to 17% protein content, 5 to 14% polyphenol, 3 to 11% and oil content, 1 to 4%. However no species contains extremes of all 4 variables. Nor can species be found, retaining native compositions, at extremes of just one extractive composition, while the other fractions are present at constant levels. Thus we use orthogonal but non-intuitive compositions in this work, then rank pyrolysis effects in terms of traditional extractives content to get an understanding of their impact on biomass pyrolysis. 

Prasad, S. Pierce, K.M. Schmidt, H. Rao, J.V. Guth, R. Bader, S. Synovec, R.E. Smith, G.B. Eiceman, G.A., Constituents with independence from growth temperature for bacteria using pyrolysis-gas chromatography/dififerential mobdity spectrometry with analysis of variance and principal component analysis. Analyst 2008,133, 760-767. 

Principal components analysis is a well-established multivariate statistical technique that can be used to identify correlations within large data sets and to reduce the number of dimensions required to display the variation within the data. A new set of axes, principal components (PCs), are constructed, each of which accounts for the maximum variation not accounted for by previous principal components. Thus, a plot of the first two PCs displays the best two-dimensional representation of the total variance within the data. With pyrolysis mass spectra, principal components analysis is used essentially as a data reduction technique prior to performing canonical variates analysis, although information obtained from principal components plots can be used to identify atypical samples or outliers within the data and as a test for reproducibihty. 

Olive oil is an ideal candidate for multivariate analysis. For economic reasons, the labelling of olive oils is frequently falsified (Collins, 1993 Firestone and Reina, 1987 Firestone, Carson and Reina, 1988 Firestone et al., 1985 Li-Chan, 1994 Simpkins and Harrison, 1995b Zamora, Navarro and Hidalgo, 1994), so there is a need for easy and cheap methods for identification. This chapter concentrates on the application of multivariate methods to nuclear magnetic resonance (NMR) and pyrolysis mass spectrometry (PyMS) data. It provides a brief introduction to principal components analysis (PCA), principal components regression (PCR), partial least squares regression (PLS) and the use of artificial neural networks (ANNs), then moves on to variable selection and its application to olive oil data. 

Mass spectrometry and chemometric methods cover very diverse fields Different origin of enzymes can be disclosed with LC-MS and multivariate analysis [45], Pyrolysis mass spectrometry and chemometrics have been applied for quality control of paints [46] and food analysis [47], Olive oils can be classified by analyzing volatile organic hydrocarbons (of benzene type) with headspace-mass spectrometry and CA as well as PC A [48], Differentiation and classification of wines can similarly be solved with headspace-mass spectrometry using unsupervised and supervised principal component analyses (SIMCA = soft independent modeling of class analogy) [49], Early prediction of wheat quality is possible using mass spectrometry and multivariate data analysis [50],... 

Wet granulation and direct compression are two methods used to manufacture tablets in the pharmaceutical industry. Zomer et al. used pyrolysis-gas chromatography-mass-spectrometry coupled with SVM classification to discriminate between the two tablet production methods.Mass spectra data were submitted to a PCA analysis, and the first principal components were used as input for SVM models having linear, polynomial, and Gaussian RBF kernels. SVM classifiers with polynomial and RBF kernels performed better in prediction than discriminant analysis. 

---