The application of principal component analysis and non-negative matrix factorization to analyze time-resolved optical waveguide absorption spectroscopy data

Time-resolved optical waveguide absorption spectroscopy (OWAS) makes use of an evanescent field to detect the polarized absorption spectra of sub-monomolecular adlayers. This technique is suitable for the investigation of kinetics at the solid-liquid interface of dyes, pigments, fluorescent molecule...

Full description

Saved in:
Bibliographic Details
Published in:Analytical methods 2013-01, Vol.5 (17), p.4454-4459
Main Authors: Liu, Ping, Zhou, Xia, Li, Yanling, Li, Minqiang, Yu, Daoyang, Liu, Jinhuai
Format: Article
Language:English
Subjects:
Absorption spectroscopy
Dyes
Factorization
Mathematical analysis
Nanoparticles
Optical waveguides
Principal component analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Time-resolved optical waveguide absorption spectroscopy (OWAS) makes use of an evanescent field to detect the polarized absorption spectra of sub-monomolecular adlayers. This technique is suitable for the investigation of kinetics at the solid-liquid interface of dyes, pigments, fluorescent molecules, quantum dots, metallic nanoparticles, and proteins with chromophores. In this work, we demonstrate for the first time the application of non-negative matrix factorization (NMF) to the analysis of time-resolved OWAS data, because NMF prevents the negative factors from occurring, avoids contradicting physical reality, and makes factors more easily interpretable. Meanwhile, principal component analysis (PCA) is researched for the comparison of NMF. Two-factor loadings and scores derived by PCA and NMF are consistent with the results analyzed by the function of time/wavelength-absorbance. Moreover, NMF avoids contradicting physical reality, and makes factors more easily interpretable. We believe that NMF will provide a valuable analysis route to allow processing of increasingly large and complex data sets.
ISSN:1759-9660
1759-9679
DOI:10.1039/c3ay40146d