Maize (Zea mays) is an important staple cereal crop grown for food, feed, and biofuel all around the globe (Bhusal et al., 2021). It is the third most important crop grown after rice and wheat (Albahri et al., 2023). Newly harvested maize is often high in moisture content, making it difficult to store for extended periods (Sun, Wang, Ling, & Zhu, 2023), so it is immediately dried to a safe moisture content after harvesting (Wei et al., 2020). Hot air drying is widely used for grain drying because of its easy operation and low cost (Wray & Ramaswamy, 2015). When a sample is dried, heating causes volumetric expansion, while the removal of moisture causes volumetric shrinkage (Irudayaraj, Haghighi, & Stroshine, 1993). The heating and removal of moisture inside the sample are non-uniform during drying, as a result, there is an uneven shrinkage inside the sample, which would lead to drying stress (Takhar, Maier, Campanella, & Chen, 2011). If the stresses exceed the material strength, the maize kernel would crack (Wei et al., 2020). The cracked kernels prove to be unusable as seeds and are susceptible to breakage, mold, and insect infestations (Zhou, Cheng, Zhang, Liu, & Li, 2011). Thus, it is necessary to understand the stresses inside the maize kernel to predict cracking, which would help to understand the crack formation mechanism to reduce maize kernel fissuring.

Due to the small size of maize kernels and the limitations of testing equipment, it is difficult to determine the drying stress inside maize kernels (Wei et al., 2020). With the development of computer technology, computer simulation has been widely applied in maize kernel drying, which can intuitively understand the changes of temperature, moisture content and stress distribution of maize kernels during the drying process (Irudayaraj et al., 1993; Takhar, 2011; Wei et al., 2020). The reasonable development of the geometric mode and mathematical model is the key to the accurate simulation of maize kernel drying. In the many studies, the maize kernel was assumed to be a homogeneous structure to study the stress in maize kernels during drying (Liu, Chen, Wu, Wang, & Wang, 2006; Sun et al., 2023; Wei, Xie, Wang, & Yang, 2019). However, the simulation results are often unsatisfactory because maize kernels have a heterogeneous structure and are composed of four major anatomical components: the germ, a soft endosperm, a hard endosperm, and a pericarp (Gunasekaran, Deshpande, Paulsen, & Shove, 1985). Therefore, in our previous study (Wei et al., 2020), we constructed the multi-component geometric model of maize kernel, moisture-heat transfer and stress-strain mathematical model to study the change of moisture, temperature, and stress of the kernel and predict the kernel cracking. The result showed that the change in moisture and temperature distribution in kernels during drying was accurately described by the model. However, the value of stress may be overestimated because the components of the maize kernels were treated as an elastic materials and an elastic stress-strain model was used to describe the relationship between the stress and strain. The result of the crack prediction was that the cracks were more likely to occur during drying, which was inconsistent with the actual observation that cracks were mainly occurred during the cooling stage (Xian, 2011; Zhao, 2007; Zhu, Cao, & Lian, 1997).

Maize kernel is a viscoelastic material that exhibits both viscous and elastic behavior simultaneously during processing (Mestres, Matencio, & Louis-Alexandre, 1995). On this basis, Sheng, Wang, Li, Mao, & Adhikari, 2014 constructed viscoelastic stress-strain model of the maize kernel and studied the stress of the kernel during the creep and relaxation. Li et al., 2022 described the change process of the stress-strain characteristic of maize kernel with time during threshing based on the viscoelastic stress-strain model. Jia, Sun, & Cao, 2000 and Takhar et al., 2011 considered the viscoelasticity and heterogeneous structure of maize kernels. They constructed a multi-component geometric model of maize kernels and created a viscoelastic stress-strain model to analyse the stress information of maize kernels during the drying process. However, the viscoelastic properties of the all the maize components in their work were the same, which was quite different from the results obtained from the stress relaxation test for each component. The test result showed that the elasticity and viscosity were highest for the pericarp, higher for the hard endosperm, lower for the soft endosperm, and lowest for the germ (Zheng et al., 2023a). The differences between the viscoelastic properties of the four components would result in different responses to an external action. For example, the soft endosperm was more susceptible to damage compared to the hard endosperm and germ during threshing (Wang & Wang, 2019). It was observed by X-ray μCT that there were no drying cracks in the pericarp and germ, and the drying cracks appeared mainly in the endosperm (Fan, Ren, & Yang, 2024). Therefore, it is unreasonable to assume that the viscoelastic properties of the four components are the same to construct the viscoelastic stress-strain mathematical model to study the drying stress and predict the cracking of the components to understand the cracking mechanism. Furthermore, to the best of our knowledge, the occurrence of cracks in maize kernel was judged by comparing the maximum von Mises stress and yield stress. The study shown that the distribution of the stress inside the maize kernel was uneven. The stress in the region near the surface was high, while the stress in the region near the core was low (Takhar et al., 2011). The crack was predicted by comparing the maximum von Mises stress with the yield stress, which would overestimate the degree of cracking.

Thus, this work aimed to (1) develop the viscoelastic stress-strain model of the components considering the heterogeneous structure of the kernel and the difference between the viscoelastic properties of the components, (2) obtain the viscoelastic parameters of the components with different moisture contents at different temperatures, (3) investigate the moisture content distribution, temperature distribution, and stress distribution of the components, and explore the impact of changes in the moisture content and temperature on the stress, (4) to predict the crack formation and degree of the components at different drying stages by comparing the local von Mises stress and corresponding yield stress in each region.