Skip to main content
ARS Home » Research » Publications at this Location » Publication #136999
Title: APPLICATION OF PARTIAL LEAST SQUARES REGRESSION TO NEAR-INFRARED REFLECTANCE SPECTROSCOPIC DETERMINATION OF SHIVE CONTENT IN FLAX

Author
item SOHN, MI RYEONG - VISITING SCIENTIST
item BARTON II, FRANKLIN
item MORRISON III, WILEY
item ARCHIBALD, DOUGLAS - PENN. STATE UNIV.

Submitted to: Journal of Applied Spectroscopy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/8/2003
Publication Date: 5/1/2003
Citation: SOHN, M., BARTON II, F.E., MORRISON III, W.H., ARCHIBALD, D.D. APPLICATION OF PARTIAL LEAST SQUARES REGRESSION TO NEAR-INFRARED REFLECTANCE SPECTROSCOPIC DETERMINATION OF SHIVE CONTENT IN FLAX. JOURNAL OF APPLIED SPECTROSCOPY. VOL. 57. ISS. 5. P. 551-556. 2003.

Interpretive Summary: Flax fiber can be used in many products ranging from fine linen to adding strength and reduced weight in composites. The products in which flax is used can depend on the amount of trash or shive (the core portion of the flax stem) in a lot of flax fiber. Currently, graders and buyers determine trash content visually. Using a sample set with a shive content ranging from 0 to 100%, we developed a robust standard calibration model using near infrared spectroscopy that has a precision of 2% when applied to flax processed by several different methods. This study suggests that an inexpensive instrument can be developed to provide rapid, accurate analysis of trash in flax fiber. Having an accurate value for trash will eliminate the guesswork in determining the suitability of a flax shipment for a given product and thus, reduce production costs.

Technical Abstract: Shive, the non-fiberous core portion of the stem, in flax fiber after retting is related with fiber quality. The objective of this study is to develop a standard calibration model for determining shive content in retted flax by using near infrared reflectance spectroscopy. Calibration samples were prepared by manually mixing pure, ground shive and pure, ground fiber from flax retted by 3 different methods (water-, dew- and enzyme retting) to provide a wide range of shive content from 0 to 100%. Partial least squares (PLS) regression was used to generate a calibration model and spectral data were processed using various pre treatments such as a multiplicative scatter correction (MSC), normalization, derivatives, and Martens' Uncertainty option to improve the calibration model. Calibration model developed with a single sample set resulted in a standard error of 1.8% with one factor. Best algorithm was produced from 1st derivative processing of the spectral data. MSC was not effective processing for this model. However, a big bias was observed when independent sample sets were applied to this calibration model to predict shive content in flax fiber. Calibration model developed using combination sample set showed slightly higher standard error and number of factors compared to one for single sample set, but this model was sufficiently accurate to apply to each sample set. The best algorithm for combination sample set was generated from a second derivative followed by MSC processing of spectral data and from Martens' Uncertainty option, it resulted in a standard error of 2.3% with 2 factors. The value of the digital 2nd derivative centered at 1674nm for these spectral data was highly correlated to shive content of flax and could form the basis for a simple, low cost sensor for the shive or fiber content in retted flax.