Online measurement of moisture content, moisture distribution, and state of water in corn kernels during microwave vacuum drying using novel smart NMR/MRI detection system

Moisture content is unevenly distributed and hard to measure when agricultural products are dried using microwave drying. A low-field nuclear magnetic resonance/imaging (NMR/MRI) and microwave vacuum drying (MVD) combination equipment was developed. The residual moisture content, distribution, and state of water (free, immobilized, and bound) in fresh corn kernels during MVD were quickly measured in real time. NMR results indicated that the amplitude of free and immobilized water decreased very rapidly at the early stage of MVD, while the amplitude of bound water experienced a similar rapid decrease at the last stage. MRI results indicated that the moisture content was always distributed unevenly during MVD, especially at the early stage. The moisture distribution tended to become uniform when drying progressed and the bound water became dominant. The residual moisture content of corn kernels and integral (total) amplitude of NMR were found to fit well with a linear model (R2 > 0.991, P < 0.01), which can be applied to predict the residual moisture content with error less than 0.01 g moisture/g solid during MVD. This work shows that an NMR/MRI smart detection system is capable of measuring and quantifying residual moisture content, moisture distribution, and state of water in foods online and in real time during drying.