Skip to main content
ARS Home » Research » Publications at this Location » Publication #119605
Title: QUALITY EVALUATION OF BARLEY BY NEAR INFRARED SPECTROSCOPY

Author
item Kays, Sandra
item Barton Ii, Franklin

Submitted to: United States-Japan Cooperative Program in Natural Resources
Publication Type: Proceedings
Publication Acceptance Date: 11/19/2000
Publication Date: 11/19/2000
Citation: Kays, S.E., Barton II, F.E. 2000. Quality evaluation of barley by near infrared spectroscopy. 29th Annual meeting of the proceedings of the United States Japan Cooperative Program in Natural Resources. pp. 1-4

Interpretive Summary:

Technical Abstract: The dietary fiber content of barley can vary widely among cultivars and is a critical quality parameter of cultivars to be used for foods and in the brewing industry. Near-infrared (NIR) spectroscopy was investigated as a screening tool to evaluate the quality of barley, in terms of total dietary fiber content. Near-infrared transmission spectra of barley grains from 56 cultivars, grown in Japan, were obtained using a Foss Grainspec.Xact Rice Analyzer (850-1050 nm) at the National Food Research Institute. Near-infrared reflectance spectra of milled barley were obtained using an NIRSystems 6500 monochromator (1100-2500 nm) at the USDA-ARS-RRC . Total dietary fiber was determined for each cultivar by the conventional AOAC enzymatic-gravimetric method (Method 991.43). The range in total dietary fiber was 5.3-19.7% and the standard error of the AOAC method was 0.37%. Using multivariate data analysis, to perform partial least squares analysis sand cross validation, NIR models were developed for the prediction of tota dietary fiber in barley grains and in milled barley samples. For NIR transmission spectroscopy with polished grains the standard error of cross validation (1.10%) and coefficient of determination (0.79) indicated sufficient accuracy of the model for selecting or rejecting high total dietary fiber cultivars. For NIR reflectance spectroscopy of milled barley samples, developing the model with second derivative processing, the standard error of cross-validation (0.67%) and multiple coefficient of determination (0.93) indicated a high degree of precision in the prediction of total dietary fiber. The reduced accuracy of transmission spectroscopy may be due, partly, to less information available for dietary fiber in the wavelength region used.