< back to main site

Publications

The application of multivariate data analysis techniques in surface analysis.

Lee, J L S; Gilmore, I S (2009) The application of multivariate data analysis techniques in surface analysis. In: Surface Analysis - The Principal Techniques (Second Edition). WileyBlackwell, pp. 563-612. ISBN 9780470017647

Full text not available from this repository.

Abstract

Multivariate analysis is widely used in analytical chemistry to provide identification and quantification for a range of spectroscopic techniques. It has been used for a number of years in surface analysis, most notably in the techniques of secondary ion mass spectrometry (SIMS), X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy. Multivariate analysis has numerous advantages over traditional (manual) analysis. It provides an objective and statistically valid approach using all available information in a data set. The need for manual identification and selection of key peaks and features for analysis is eliminated or significantly reduced, thereby reducing the need for a priori knowledge about the system under study and minimising the potential for bias. By correlating data across a number of variables, an improved signal to noise ratio can be obtained. Multivariate analysis can also be fast and automated.

In this chapter, the basic concepts of multivariate analysis will be outlined, covering the matrix representation of data and its relation to multivariate analysis. We will then review the principle and theory behind many popular multivariate analysis methods. This includes identification using factor analysis methods PCA and MCR, calibration and quantification using regression methods PCR and PLS, and classification and clustering methods using DFA, HCA and ANNs. Throughout the chapter, recent examples from literature will be illustrated, with a focus on applications in surface analysis.

Item Type: Book Chapter/Section
Keywords: multivariate analysis, principal component analysis, multivariate curve resolution, partial least squares, discriminant function analysis, artificial neural networks
Subjects: Nanoscience
Publisher: WileyBlackwell
Last Modified: 02 Feb 2018 13:15
URI: http://eprintspublications.npl.co.uk/id/eprint/4357

Actions (login required)

View Item View Item