Studies On Rapid Detection Of Ingredients Of Soybeans, Soymilk And Tofu By Nir And Soymilk Coagulated Technology

The detection of ingredients of soybeans, soymilk and tofu and soymilk coagulated technology were two critical control points during tofu processing. Currently, the detection of soybeans, soymilk and tofu components still use conventional methods. Because of the cumbersome process, time-consuming and short shelf life of tofu, it was necessary to establish a rapid detection method. Soymilk coagulated technology is not only related to tofu quality, but also related to automation of soymilk coagulation. Therefore, this paper studied the rapid detection of protein, fat and water content in soybeans, soymilk and tofu by near infrared spectroscopy. And we established a mathematical model which was about effect of composite coagulant, protein concentration and coagulated temperature on coagulated time. The results are as follows:(1) With the mode of integrating sphere diffuse, the spectra of 83 different soybeans simples were obtained by the Fourier transform near-infrared spectrometer(FT-NIR). Combined with the results of chemical analysis, the calibration models of the three components were established separately. The calibration models had a best prediction performance when the spectra of the protein was preprocessed by first order derivative + minus a straight line and fat and moisture were preprocessed by first order derivative + multiplicative scatter correction(MSC). The determination coefficients(R2)for the protein, fat and moisture content were 0.9146, 0.9229 and 0.9631, respectively, and the root mean square errors of cross-validation(RMSECV)were 0.746, 0.415 and 0.349, respectively. External validation of the model showed there was no significant difference between chemical values and model predictions, which indicated that the calibration models could be used to detect protein, fat and moisture content of soybeans.(2)With the mode of integrating sphere diffuse, the spectra of 60 soymilk simples were obtained by FT-NIR. Combined with the results of chemical analysis, the calibration models of the three components were established separately. The calibration models had a best prediction performance when the spectra of the protein, fat and soluble solids were preprocessed by constant offset elimination, first order derivative and standard normal variate transformation(SNV), respectively. The R2 for the protein, fat and soluble solids content were 0.9664, 0.9500 and 0.9507, respectively, and RMSECV were 0.0769, 0.0874 and 0.316, respectively. External validation of the model showed there was no significant difference between chemical values and model predictions, which indicated that the calibration models could be used to detect protein, fat and soluble solids content of soymilk.(3) With transmission mode, the spectra of 60 tofu simples were obtained by FT-NIR. Combined with the results of chemical analysis, the calibration models of the three components were established separately. The calibration models had a best prediction performance when the spectra of the protein was preprocessed by second order derivative, and fat and moisture were preprocessed by standard normal variate transformation(SNV). The R2 for the protein, fat and moisture content were 0.9364, 0.9227 and 0.9302, respectively, and RMSECV were 0.276, 0.285 and 0.584, respectively. External validation of the model showed there was no significant difference between chemical values and model predictions, which indicated that the calibration models could be used to detect protein, fat and moisture content of tofu.(4) To improve the quality of pack tofu and to slow coagulation of soymilk, Gluconate-δ- lactone(GDL)and calcium chloride(CaCl2)were compounded to improve the quality of tofu. Then on this basis, the effect of soymilk protein concentration, coagulated temperatures and GDL /Ca Cl2 ratio on coagulated time was analyzed. Box-Behnken design(BBD)and response surface methodology(RSM)were adopted to optimize coagulated conditions. The results showed that in the conditions of 4.0% of protein concentration, 85 ℃of coagulated temperature, the quality of tofu with GDL/CaCl2 ratio 8:2 was better than other ratio. With response surface methodology we established a mathematical model of coagulating time which was reasonable and reliable after test. And it could accurately predict coagulated time of soymilk.