Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Analytical Methods for Classifying Wheat.

Author
item Hareland, Gary

Submitted to: The Nordic Cereals Proceedings
Publication Type: Proceedings
Publication Acceptance Date: April 10, 1996
Publication Date: N/A

Interpretive Summary: Particle size measurements were determined on samples of ground wheat by laser diffraction analysis. Ground wheat particles were distributed unevenly over a wide range for all wheat samples tested, but distinct peaks occurred within two narrow size ranges. Within these two narrow size ranges, hard wheat particles were concentrated in the large particle size range, and soft wheat particles were concentrated in the small size range. From these results, hard red spring and hard red winter wheat classes were different from each other. Soft wheat classes were not different from each other but were different from the hard wheat classes. Near infrared (NIR) spectroscopy was calibrated against the laser diffraction measurements. NIR calibrations were used to predict particle size measurements for the two narrow size ranges with high accuracy.

Technical Abstract: Particle size measurements from 211 samples of Udy Cyclone milled wheat were obtained by laser diffraction analysis. Ground wheat particles were distributed unevenly over a wide size range for all wheat samples tested, but distinct peaks occurred within the 10-41 micrometer and 41-146 micrometer size ranges. Hard wheat contained a high percent volume of particles within 41-146 micrometers. In contrast, soft wheat contained a high percent volume of particles within 10-41 micrometers. Hard red spring and hard red winter wheat classes were significantly different for both size ranges. Soft red winter and soft white wheat classes were not different from each other but were different from the hard red spring and hard red winter wheat classes. The particle size measurements were compared with infrared spectral changes at 1680 nanometers and 2230 nanometers. Strong correlations occurred between near infrared spectra and percent volume of particles within 10-41 micrometers (r = -0.85) and 41-146 micrometers (r = 0.80), but, weak correlations occurred between near infrared spectra and mean particle diameter (r = 0.49) and median particle diameter (r = 0.70). Particle size measurements were predicted by near infrared spectroscopy from modified partial least squares regression models. Percent volume of particles within 10-41 micrometers and 41-146 micrometers and median particle diameter were predicted with higher accuracy (r**2 = 0.94, 0.88, and 0.88, respectively) than mean particle diameter (r**2 = 0.77).

Last Modified: 11/1/2014
Footer Content Back to Top of Page