Submitted to: Sensors and Actuators B: Chemical
Publication Type: Peer reviewed journal
Publication Acceptance Date: 4/2/2013
Publication Date: 4/28/2013
Citation: Lee, H., Cho, B., Kim, M.S., Lee, W., Tewari, J., Bae, H., Sohn, S., Chi, H. 2013. Prediction of crude protein and oil content of soybeans using Raman spectroscopy. Sensors and Actuators B: Chemical. 185:694-700. Interpretive Summary: Spectroscopic technologies can potentially provide a means to rapidly and nondestructively assess a variety of food components. A collaborative study was conducted to develop an optimal prediction model for determining the protein and oil contents of soybean seeds using Raman spectroscopy. The oil and protein contents of soybean seeds are typically determined chemically and the analyses require destructive sample preparations. Raman spectral measurements coupled with numerical analyses allowed predictions of the oil and protein contents of soybean seeds with 91.6% and 87.2% accuracy, respectively. This study demonstrated that the Raman techniques can be applied to the prediction of soybean crude protein and oil contents. The methods and results presented in this investigation are useful to food technologists, food engineers, and food processing industries.
Technical Abstract: While conventional chemical analysis methods for food nutrients require time-consuming, labor-intensive, and invasive pretreatment procedures, Raman spectroscopy can be used to measure a variety of food components rapidly and non-destructively and does not require supervision from experts. The purpose of this study was to develop an optimal prediction model for determining the protein and oil content of soybeans using dispersive Raman spectroscopy. The crude oil content of soybeans is typically determined chemically using the Soxhlet extraction method and crude protein content is determined using the semimicro-Kjeldahl method and an auto protein analyzer. In the present study, Raman spectra were measured in the 200 cm-1 –1800 cm-1 spectral range and partial least squares analysis (PLS) was used to develop the optimal model for predicting the crude protein and oil content of soybeans. The results of the PLS model that used the effective wavenumber regions selected by intermediate PLS (iPLS) method were better than those of models that used the entire wavenumber. The Rp2 and SEP of the optimal PLS model for crude protein were 0.916 and 0.636%, respectively. The Rp2 and SEP for crude oil content were 0.872 and 0.759%, respectively. The PLS prediction result was slightly better for protein than for oil. The conventional Raman techniques investigated in this study can be applied to the prediction of soybean crude protein and oil content.