Show simple item record

dc.contributor.authorLoudermilk, James Brian
dc.date.accessioned2014-03-04T03:21:29Z
dc.date.available2014-03-04T03:21:29Z
dc.date.issued2008-05
dc.identifier.otherloudermilk_james_b_200805_phd
dc.identifier.urihttp://purl.galileo.usg.edu/uga_etd/loudermilk_james_b_200805_phd
dc.identifier.urihttp://hdl.handle.net/10724/24683
dc.description.abstractDuring the cotton harvest, other parts of the cotton plants in addition to the cotton fibers are collected. This plant debris causes a number of problems in cotton processing, including decreased yarn quality and increased number of yarn breakages during spinning. Because these debris break down in size physically after harvest, human visual identification of and discrimination among these contaminants are difficult or impossible, but identification of the contaminant type is desirable for investigations of process breakdowns and efficiency monitoring of cleaning processes. To achieve contaminant identification and discrimination, a number of projects that explore spectral comparison in general and develop solutions to the specific problem of cotton contaminant discrimination is reported. Chemometric methods for contaminant class discrimination have been developed for and applied to the FT-IR attenuated total reflection (ATR) spectra of cotton contaminants. Novel voting scheme algorithms were developed to improve spectral identification by library searching for a USDA spectral library of cotton contaminants. Improvements in contaminant identification were also realized via partial least squares discriminant analysis (PLS-DA). Quantitative analysis of cotton contaminant mixtures was achieved with the use of partial least squares (PLS) regression and a novel error correction algorithm that was developed. This work also reports the development of a mixture generator algorithm to generate sets of mixtures representative of mixture spaces of arbitrary dimensions. Finally, the spectral differences caused by the use of different FT-IR spectrometers and ATR accessories were investigated by measuring spectra of a polyethylene terephthalate film with the use of several different spectrometers and accessories. The spectra were compared before and after corrections for depth of penetration and anomalous dispersion effects. The results show that these correction methods do not always achieve the goal of increased spectral similarity.
dc.languageeng
dc.publisheruga
dc.rightspublic
dc.subjectChemometrics
dc.subjectFT-IR
dc.subjectATR
dc.subjectcotton
dc.subjectcontaminants
dc.subjectspectral library
dc.subjectspectral search system
dc.subjectPLS
dc.subjectPLS-DA
dc.subjectvoting scheme algorithms
dc.subjectquantitative analysis
dc.subjectcalibration
dc.subjectmixtures
dc.subjectexperimental design
dc.subjectdepth of penetration
dc.subjectanomalous dispersion
dc.titleFT-IR spectrometry and chemometrics for spectral discrimination of cotton contaminants
dc.typeDissertation
dc.description.degreePhD
dc.description.departmentChemistry
dc.description.majorChemistry
dc.description.advisorJames A. de Haseth
dc.description.committeeJames A. de Haseth
dc.description.committeeDavid S. Himmelsbach
dc.description.committeeLionel A. Carreira
dc.description.committeeJames L. Anderson


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record