Prediction model of moisture in peanut kernel during hot air drying based on LF-NMR technology

Abstract: Peanut, which has high economic value, is one of the main oil crops in China. Freshly harvested peanuts usually had high moisture contents, which were about 40%-60%. If peanuts with high moisture content did not dry in time, it was easy to fever and mildew. So it is very important for peanuts drying in time. In this paper, three varieties of wet peanuts, Kainong 71, Kainong 8834-9 and Tianfu 3, were dried by hot air, which temperature was 35 ℃ and air velocity was 0.6 m/s. And the moisture contents of peanuts, peanut kernels and peanut shells were monitored during the drying process. The changes of free water, weak-bound water and bound water in peanut kernels during drying were studied by low-field nuclear magnetic resonance (LF-NMR) technology. The common law of internal moisture changes in peanut kernels during drying process was obtained. The mathematical relationship between the total percentage of moisture relaxation peaks in peanut kernels and their moisture contents determined by the national standard method was established. A rapid detection method of peanut moisture contents was proposed in this paper, which had a practical significance for peanut storage and processing. The results showed that regardless of the amount of oleic acid in peanuts and the size of the peanuts, the drying characteristics of the three varieties of peanuts were similar. The moisture contents of the three varieties of peanut and peanut kernel showed different downward trend compared to the peanut shell. The moisture contents of peanut shell first declined rapidly in the early stage (2-4 h), then slowly due to their fiber structure. The LF-NMR relaxation spectra of peanut kernels of the three varieties all had four main peaks, namely, free water peak, weak-bound water peak, bound water peak and oil peak. During the drying process, the free water peak areas of peanut kernels of the three varieties decreased gradually, and almost disappeared at the end of drying. The changing trends of the weak-bound water and bound water peak areas were obvious. The peak areas of oil peaks did not change significantly, indicating that the oil content in peanut kernels remained unchanged during drying process. The relaxation peak area ratios of peanut kernels of the three varieties in low-field NMR relaxation spectra also showed similar trends during drying. The percentages of bound water peak area and oil peak area showed an upward trend. The percentage of free water peak area showed a downward trend, which was very low at the end of drying, indicating that free water almost completely lost at the end of drying. The percentage of weak-bound water peak area firstly rose, then declined at the end of drying, indicating that some weak-bound water was lost. Therefore, free water and weak-bound water were the main kind lost water in the drying process. In addition, it could be judged that if the relaxation peak of free water disappeared or the peak was very small, the moisture contents of peanut kernels were below the safe storage moisture content. The R2 of equation established to predict the moisture content of peanut kernel is 0.888 4. The equation has been validated and the fitting results were good. Therefore, the LF-NMR technology can be used for rapid detection of moisture content in peanut kernels.