Skip to main content
ARS Home » Research » Publications at this Location » Publication #200679


item Allen, Angela
item Foulk, Jonn
item Gamble, Gary

Submitted to: American Chemical Society Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: 9/30/2006
Publication Date: N/A
Citation: N/A

Interpretive Summary: During the ginning process, sizes of botanical cotton trash become smaller which creates problems in identifying the type of trash entanged within the cotton fibers. Therefore, this study will use a microwave extraction method to separate constituents from various botanical trash samples, and then analyze these constituents through Fourier transform infrared spectroscopic measurements. Principal component analysis was applied to the spectral data in order to identify the types of botanical cotton trash.

Technical Abstract: The size of botanical cotton trash (e.g., leaves, hulls, bracts, stems, and seeds) decreases throughout the ginning process. This physical change causes small trash to remain entangled within the cotton fibers, and also creates problems in identifying the type of trash. As further processing occurs, the quality of cotton is reduced by the presence of this contamination. In this study, a microwave extraction method was used to separate the constitutents associated with various botanical types of cotton trash using a polar solvent (methanol). The modified botanical cotton trash types were used to construct a rapid and accurate method of identifying the types of trash by using Fourier-transform infrared spectroscopic measurements. Using the spectral data, principal component analysis (PCA) was applied in an effort to identify the various types of cotton trash through extracted constituents. Various spectral preprocessing methods (e.g., normalization and derivatives) were examined in order to obtain the best PCA model, which was then used in the prediction of modified botanical cotton trash types. A comparison of the best fit model to experimental data will be shown and discussed.