Direct prediction of bioethanol yield in sugar beet pulp using Near Infrared Spectroscopy

• Published in: Bioresource technology 2011-10, Vol.102 (20), p.9542-9549
• Main Authors: Magaña, C., Núñez-Sánchez, N., Fernández-Cabanás, V.M., García, P., Serrano, A., Pérez-Marín, D., Pemán, J.M., Alcalde, E.
• Format: Article
• Language: English
• Subjects: beets; Beta vulgaris - metabolism; bioethanol; Biofuel crops; Biofuel production; Biological and medical sciences; Biotechnology; Calibration; Coefficient of variation; Energy; Ethanol; Ethanol - analysis; Ethanol - metabolism; ethanol production; Ethyl alcohol; First generation bioethanol; Fundamental and applied biological sciences. Psychology; Industrial applications and implications. Economical aspects; least squares; Least-Squares Analysis; Mathematical models; Monochromators; Near Infrared Spectroscopy; prediction; raw materials; raw sugar; Regression; Spectroscopy, Near-Infrared - methods; Standard error; Sugar beet; sugar beet pulp; Sugar beets; Sugar industry
• ISSN: 0960-8524; 1873-2976
• DOI: 10.1016/j.biortech.2011.07.045

► The novel NIR-spectroscopy-based method presented allows the prediction of the potential bioethanol yield of sugar beet. ► Bioethanol yield determination from sugar beet brei samples could be shortened from 64h to 3min using NIRS. ► Direct estimation of sugar beet ethanol yield could be of great interest for sugar industry in the make-decision process. ► NIRS technology might be of great usefulness for the process control for the bioethanol industry. Sugar beets are a raw material for the production of sugar and ethanol. The decision on which end product to pursue could be facilitated by fast and reliable means of predicting the potential ethanol yield from the beets. A Near Infrared (NIR) Spectroscopy-based approach was tested for the direct prediction of the potential bioethanol production from sugar beets. A modified partial least squares (MPLS) regression model was applied to 125 samples, ranging from 21.9 to 31.0gL−1 of bioethanol in sugar beet brei. The samples were analyzed in reflectance mode in a Direct Contact Food Analyser (DCFA) FOSS-NIRSystems 6500 monochromator, with standard error of cross validation (SECV), standard error of prediction (SEP), coefficient of determination (r2) and coefficient of variation (CV) of 0.51, 0.49, 0.91 and 1.9gL−1, respectively. The NIR technique allowed direct prediction of the ethanol yield from sugar beet brei (i.e. the product obtained after sawing beets with a proper machine) in less than 3min.