Determination of Drying Characteristics and Physicochemical Properties of Mint (Mentha spicata L.) Leaves Dried in Refractance Window

Abstract

Drying is one of the most common and effective techniques for preserving the quantitative and qualitative characteristics of medicinal plants in the post-harvest phase. Therefore, in this research, the effect of the new refractance window (RW) technology on the kinetics, thermodynamics, greenhouse gasses, color indices, bioactive properties, and percentage of mint leaf essential oil was investigated in five different water temperatures in the form of a completely randomized design. This process was modeled by the methods of mathematical models and artificial neural networks (ANNs) with inputs (drying time and water temperature) and an output (moisture ratio). The results showed that with the increase in temperature, the rate of moisture removal from the samples increased and as a result, the drying time, specific energy consumption, CO2, NOx, enthalpy, and entropy decreased significantly (p < 0.05). In addition, the drying water temperature had a significant effect on the rehydration ratio, color indices, bioactive properties, and essential oil percentage of the samples (p < 0.05). The highest value of rehydration ratio was obtained at 80 °C. By increasing temperature, the main color indices such as b*, a*, L*, and Chroma decreased significantly compared to the control (p < 0.05). However, with the increase in temperature, the overall color changes (ΔE) and L* first had a decreasing trend and then an increasing trend, and this trend was the opposite for the rest of the indicators. The application of drying water temperature from 50 to 70 °C increased antioxidant, phenol content, and flavonoid content, and higher drying temperatures led to a significant decrease in these parameters (p < 0.05). On the other hand, the efficiency of the essential oil of the samples was in the range of 0.82 to 2.01%, and the highest value was obtained at the water temperature of 80 °C. Based on the analysis performed on the modeled data, a perceptron artificial neural network with 2-15-14-1 structure with explanation coefficient (0.9999) and mean square error (8.77 × 10−7) performs better than the mathematical methods for predicting the moisture ratio of mint leaves.