Location: Peanut Research
Title: Nondestructive NIR reflectance spectroscopic method for rapid fatty acid analysis of peanut seeds Authors
Submitted to: Peanut Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: March 1, 2011
Publication Date: July 24, 2011
Repository URL: http://dx.doi.org/10.3146/PS10-3.1
Citation: Sundaram, J., Kandala, V.C., Butts, C.L., Chen, C.Y., Sobolev, V.S. 2011. Nondestructive NIR reflectance spectroscopic method for rapid fatty acid analysis of peanut seeds. Peanut Science. 38(2):85-92. Interpretive Summary: Non-destructive methods to measure the fatty acid composition in peanut seed are needed to assist plant breeders in selecting seed of new hybrids with the desired characteristics. A method measuring the reflectance of light in the Near-Infrared (NIR) spectrum was developed to measure the fatty acid composition in peanut seed. The reflectance of light over the wavelength of 400 to 2500 nm was measured on a peanut kernel, then the oil was extracted and analyzed to determine the total fatty acid content and the proportion of oleic, linoleic, and palmitic fatty acids. The reflectance data was compared to the chemical composition data and a calibration was developed. The calibration then allows scientists to measure the NIR reflectance and estimate the fatty acid composition of a single peanut kernel without destroying the seed. This will allow a peanut breeder to test a single seed, and plant the seed that has the desired fatty acid composition and continue to develop that particular peanut variety.
Technical Abstract: NIR reflectance spectroscopy was used to analyze the fatty acid concentration present in breeder's peanut seeds samples, rapidly and nondestructively. Absorbance spectra were collected in the wavelength range from 400 nm to 2500 nm using a NIR spectrometer. Fatty acids, oleic, linoleic and palmitic acids were converted to their corresponding methyl esters and their concentrations were measured using a gas chromatograph (GC). Partial least square (PLS) analysis was performed on a calibration set, and models were developed for prediction of fatty acid concentrations. The best model was selected based on the coefficient of determination (R2), Root Mean Square Error of Prediction (RMSEP), residual percent deviation (RPD) and correlation coefficient percentage between the GC measured values and the NIR predicted values. The NIR reflectance model developed yielded RPD values of three and above for prediction of the three fatty acids, indicating that this method would be suitable for fatty acid predictions in peanut kernel seeds.