Proceedings Article | 20 April 2010
KEYWORDS: Near infrared spectroscopy, Inspection, Glasses, Calibration, Spectrometers, Spectroscopy, Vegetable oil, Quartz, Agriculture, Data acquisition
Sesame seed oil is popular and expensive in Korea and has been often mixed with other less expensive vegetable oils.
The objective of this research is to develop an economical and rapid adulteration determination system for sesame seed
oil mixed with other vegetable oils. A recently developed inspection system consists of a light source, a measuring unit,
a spectrophotometer, fiber optics, and a data acquisition module. A near-infrared transmittance spectroscopic method
was used to develop the prediction model using Partial Least Square (PLS). Sesame seed oil mixed with a range of
concentrations of corn, or perilla, or soybean oil was measured in 8 mm diameter glass tubes. For the model
development, a correlation coefficient value of 0.98 was observed for corn, perilla, and soybean oil mixtures with
standard errors of correlation of 6.32%, 6.16%, and 5.67%, respectively. From the prediction model, the correlation
coefficients of corn oil, perilla oil, and soybean oil were 0.98, 0.97 and 0.98, respectively. The Standard Error of
Prediction (SEP) for corn oil, perilla oil, and soybean oil were 6.52%, 6.89% and 5.88%, respectively. The results
indicated that this system can potentially be used as a rapid non-destructive adulteration analysis tool for sesame seed oil
mixed with other vegetable oils.
