TITLE

Qualitative and quantitative investigation of chromium-polluted soils by laser-induced breakdown spectroscopy combined with neural networks analysis

AUTHOR(S)
Sirven, J.-B.; Bousquet, B.; Canioni, L.; Sarger, L.; Tellier, S.; Potin-Gautier, M.; Le Hecho, I.
PUB. DATE
June 2006
SOURCE
Analytical & Bioanalytical Chemistry;Jun2006, Vol. 385 Issue 2, p256
SOURCE TYPE
Academic Journal
DOC. TYPE
Article
ABSTRACT
Laser-induced breakdown spectroscopy (LIBS) has been applied to the analysis of three chromium-doped soils. Two chemometric techniques, principal components analysis (PCA) and neural networks analysis (NNA), were used to discriminate the soils on the basis of their LIBS spectra. An excellent rate of correct classification was achieved and a better ability of neural networks to cope with real-world, noisy spectra was demonstrated. Neural networks were then used for measuring chromium concentration in one of the soils. We performed a detailed optimization of the inputs of the network so as to improve its predictive performances and we studied the effect of the presence of matrix-specific information in the inputs examined. Finally the inputs of the network—the spectral intensities—were replaced by the line areas. This provided the best results with a prediction accuracy and precision of about 5% in the determination of chromium concentration and a significant reduction of the data, too.
ACCESSION #
20599056

 

Related Articles

Share

Read the Article

Courtesy of THE LIBRARY OF VIRGINIA

Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics