Sustainable release of LNT and calcium ion of CSL-gel
To assess the release property of loaded drugs in CSL-gel, we measured the release rate of LNT at the corresponding time according to the regression equation of LNT (y = 0.0028x + 0.0266, R2 = 0.9998, Figure S1 ). As shown in Figure 2a and2b , CSL-gel maintained a stable and cumulative release of LNT from the first day to the 14 days after soaking. However, a substantial amount of LNT was released from SL-gel on the first day and the LNT release slightly increased with the extension of soaking time and reached the maximum amount on the eighth day after soaking. At the ninth day, a limited amount of LNT was released following the extension of the soaking. Furthermore, we used Korsmeyer-Peppas and Higuchi models to determine the release of LNT. Figure 2c showed that a clear linear correlation (R2=0.989) between the cumulative release rate and the logarithm of time from CSL-gel in the Korsmeyer-Peppas model fitting curve was observed. Similarly, a high linear correlation between the cumulative release rate and the square root of time during the release of LNT from the CSL-gel (R2=0.9823) was also observed in the Higuchi model fitting curve (Figure 2d ). However, no linear correlation between the cumulative release rate and the square root of time (R2=0.8683) and between the cumulative release rate and the logarithm of time (R2=0.933) of SL-gel was observed. The corresponding dispersion coefficient (n) of SL-hydrogel was 0.3089 (n < 0.43), indicating that the release of LNT from SL-gel under these conditions followed Fickian diffusion (Table 3 ). In contrast, the dispersion coefficient (n) of CSL-gel was 0.6305 (0.43 < n < 0.85), which indicates the CSL-gel followed non-Fick diffusion mechanism (Table 3 ). These findings demonstrated that the chitosan shell controls the release of LNT from the CSL-gel, resulting in the cumulative and sustainable release of LNT.
Soil is a complex system in which the properties, such as pH, temperature, and ion concentration, are dynamic due to the impact of external factors such as fertilization, irrigation, and rainfall[31]. The drug release from hydrogel will be influenced by different soil environments. To assess the release rate of LNT at different soil properties, we measured the release of LNT from hydrogel at different pH, temperatures, and ion concentrations. As shown in Figure 2e , the cumulative LNT release rate of CSL-gel highly increased following the increased temperature, which is caused by the high swelling of the hydrogel at a high temperature. Similarly, the corresponding release rate of LNT from CSL-gel largely increased from 7.54% to 41.35% when the pH increased from 3 to 9 (Figure 2f ). This observation follows the fact that at low pH most of the carboxyl groups on alginate are deionized and form strong hydrogen bonds with hydroxyl groups, resulting in a tight shrinkage of the calcium alginate pellet that prevents the LNT release from the CSL-gel. Interestingly, we found that ion concentration present positively regulated the release LNT from CSL-gel. As shown in Figure 2e , the cumulative release rate of LNT highly increased from 33.7% to 99.5% following the increased Na+ concentration from 0 to 0.8 M. The ion exchange between Na+ and Ca2+ strengthens the electrostatic repulsion between -COO-, which ultimately increases the swelling of the hydrogel to stimulate the release of LNT [32].