Near-infrared transmittance spectroscopy: a potential tool for non-destructive determination of oil content in groundnuts

Citation
Jb. Misra et al., Near-infrared transmittance spectroscopy: a potential tool for non-destructive determination of oil content in groundnuts, J SCI FOOD, 80(2), 2000, pp. 237-240
Citations number
11
Categorie Soggetti
Agricultural Chemistry
Journal title
JOURNAL OF THE SCIENCE OF FOOD AND AGRICULTURE
ISSN journal
00225142 → ACNP
Volume
80
Issue
2
Year of publication
2000
Pages
237 - 240
Database
ISI
SICI code
0022-5142(20000115)80:2<237:NTSAPT>2.0.ZU;2-F
Abstract
The oil content of 64 samples of groundnut kernels, representing 47 genotyp es, was determined by the conventional Soxhlet extraction procedure (Oil(SO X)). Th, values of Oil(SOX) ranged from 403 to 536 g kg(-1). The optical de nsities (ODs) of these samples were determined at 12 wavelengths (918, 928, 940, 950, 968, 975, 985, 998, 1010, 1023, 1037 and 1045nm) in the near-inf rared (NIR) region using a food composition analyser (essentially a filter- based NIR spectrophotometer). The instrument also recorded the temperatures of the sample (Temp(S)) and the air (Temp(A)) surrounding it. A sample hol der (75mm x 150mm; optical path length 25mm) was used for optical density m easurement. The data obtained were subjected to multiple linear regression analysis using the ODs at 12 wavelengths, Temps and TempA as the independen t (predictor) variables, and Oil(SOX) as the dependent variable. The multip le linear regression equation comprising 14 predictors showed a significant relationship between predicted values of oil content (Oil(NIR)) and Oil(SO X). The standard error of calibration and the coefficient of determination for calibration (R-c(2)) were 3.54 and 0.821 respectively, while the standa rd error of prediction and the coefficient of determination for prediction were 5.82 and 0.865 respectively. The ratio of standard error of calibratio n and standard error of prediction was 0.608. The differences between Oil(S OX) and Oil(NIR) were less than +/- 20g kg(-1) for samples having oil conte nts in the range from 480 to 510g kg(-1). However, for samples having Oil(S OX) lower than 480g kg(-1) or higher than 510g kg(-1), differences greater than +/- 20g kg(-1) were observed. There exists scope for further refining the regression equation by using a larger number of samples for generating optical data. The results demonstrated the potential of NIR transmittance s pectroscopy for determining the oil content of groundnuts in a non-destruct ive manner. (C) 2000 Society of Chemical Industry.