Submitted to: Proceedings of SPIE
Publication Type: Proceedings
Publication Acceptance Date: April 14, 2011
Publication Date: May 12, 2011
Citation: Liu, Y., Thibodeaux, D.P., Gamble, G.R. 2011. Development of simple algorithm for direct and rapid determination of cotton maturity from FTIR spectroscopy. Proceedings of SPIE. 80321:1-9. Interpretive Summary: Immature fibers were prone to entanglement formation during mechanical processing and also to alter the desired color appearance in dyed products. In cotton industry, such devices as image analysis and AFIS have been developed to measure cotton maturity, in which the obtained results could not be verified. As an approach, we developed a simple three-IR-band algorithm for determining fiber maturity directly. The result was well validated by independent measurements. This procedure has the potential to be a rapid, routine, and accurate measuring technique, because it needs minimal sample preparation and is based on structural differences between immature and mature fibers.
Technical Abstract: Fourier transform infrared (FTIR) spectra of seed and lint cottons were collected to explore the potential for the discrimination of immature cottons from mature ones and also for the determination of actual cotton maturity. Spectral features of immature and mature cottons revealed large differences in the 1200-900 cm-1 region, and such spectral distinctions formed the basis on which to develop simple three-band ratio algorithm for classification analysis. Next, an additional formula was created to assess the degree of cotton fiber maturity by converting the three-band ratios into an appropriate FTIR maturity (MIR) index. Furthermore, the MIR index was compared with parameters derived from traditional image analysis (IA) and advanced fiber information system (AFIS) measurements. Results indicated strong correlations (R2 > 0.89) between MIR and MAFIS and between MIR and MIA among either International Cotton Calibration (ICC) standards or selected cotton maturity references. On the other hand, low correlations between the pairs were observed among regular cotton fibers, which likely resulted from the heterogeneous distribution of structural, physical, and chemical characteristics in cotton fibers and subsequent different sampling specimens for individual and independent measurement.