Biodegradable Starch Sachets Reinforced with Montmorillonite for Packing ZnO Nanoparticles: Solubility and Zn2+ Ions Release

Agriculture’s importance in human lives and the economy has directed studies to improve crop production. An essential challenge for improving fertilizer efficacy is reducing losses due to leaching and increasing nutrient supplies. In this context, biodegradable sachets stand out as internal packaging instead of direct insertion into the polymer matrix, facilitating the system processing and making it easier to adapt the soil’s nutritional quantity. Thus, the present work aimed to increase the zinc oxide (ZnO) solubility by obtaining nanoparticles using top-down and bottom-up approaches and packaging them in montmorillonite (MMT) reinforced starch sachets. The different diameters and forms of the ZnO nanoparticles were evaluated to understand the solubility dependence on these parameters. In this way, the top-down process for the attritor milling method allowed the nanoparticles with about 71 nm average diameter and greater homogeneity than the commercial one (approximately 174 nm). The milled ZnO nanoparticles presented better solubility than those synthesized bottom-up processes and the commercial ones, reaching a 90 to 100% solubility plateau in 48 h. Concerning starch sachets, the 1% MMT (w w− 1) insertion in the polymeric matrix promoted increased water vapor barrier and mechanical properties, improving tensile strength. In the solubility test for nanoparticulate ZnO packed in sachets, similar behaviors to free ZnO were observed due to the high affinity of the starch matrix with water. Therefore, starch sachet systems with improved properties from the MMT reinforcement insertion showed as an alternative source of Zn2+ ions to minimize losses during application.


Introduction
Mineral fertilizers have been used to provide nutrients for plants with maximum development in the cultivation process. One of these nutrients is zinc (absorbed in the ionic form Zn 2+ ), responsible for the plant's critical metabolic processes, including longitudinal growth and hormone for obtaining nanometric scale materials. The production of nanoparticles via bottom-up occurs through the molecular properties of self-organization [10]- [12] as in hydrothermal [13,14], precipitation [15]- [17], and polymeric precursors [11,18,19] synthesis. While in the top-down approach, particles are reduced to the nanometer scale [12], [20]- [22] by applying force, such as milling methods [23].
However, the increase in solubility of nanoparticles does not solve the intense losses due to leaching processes during ground fertilizer application. One way to minimize this effect is through conditioning and release systems, such as biopolymer matrices [24]. In addition, different systems such as films [25], fibers [26], hydrogels [27], and capsules [28] increase the release control efficiency during application. In this sense, biodegradable polymers have been used to produce matrices that act in the packaging and subsequent release of fertilizers [29]. Among the promising materials for delivery systems is starch, a natural, lowcost polymer from highly available renewable sources [30]. However, despite several advantages, thermoplastic starch (TPS) is also characterized by hydrophilicity, making it a material with high permeability and low resistance compared to polyolefins. Thus, an alternative to revert such problems is applying reinforcement materials with clays in the polymer matrix [31,32]. Clays from the smectite group are composed of hydrated aluminum silicate minerals, such as montmorillonite (MMT) [33,34]. The MMT with the structure M x (Al 4 − x Mg x )Si 8 O 20 (OH) 4 is the most used clay mineral type for this purpose, presenting a solid interaction with the TPS matrix [31].
Biodegradable sachets may allow internal fertilizer packaging, minimizing processing damage and modulating the desired nutrient amount control. The literature on sachets as soil nutrient releasers is sparse, with only one reference developed by Sciena et al. [35]. The authors produced biodegradable starch/pectin sachets to pack hydroxyapatite (HAP) nanoparticles 24 nm in diameter as a phosphorus source. A solubility improvement of the stored nanoparticles in the sachets was observed with pH decreasing attributed to pectin dissolution from the polymeric matrix. In this sense, the sachets promote protection against fertilizer loss and an additional advantage from controlling the polymeric matrix composition to enhance nutrient solubility. From this perspective, the present work aimed to obtain ZnO nanoparticles with high solubility from top-down and bottom-up approaches followed by packaging in TPS sachets reinforced with MMT, improving the composite mechanical properties to contribute to fertilizer shelf storage and release control.

Bottom-up Methods
Hydrothermal (HT-200°C/3 h) For the ZnO synthesis via the hydrothermal route, 0.3 mol of zinc acetate in 50 mL was initially added and solubilized at room temperature for 1 h under magnetic stirring in a Teflon® flask. Then, 50 mL of sodium hydroxide (5 mol L − 1 ) was dropped into the solution, forming a white precipitate. Subsequently, it was taken to hydrothermal treatment at 200°C for 3 h, centrifuged, washed until pH 7, and dried in a circulation oven at 50°C for 24 h. Precipitation (P-25°C/1 h) First, a potassium hydroxide (50 mL, 6 mol L − 1 ) solution in deionized water was prepared for pH adjustment to 14. Then zinc nitrate hexahydrate (3.68 g) was added to the above solution with continuous stirring for 1 h. The pH 14 remained constant throughout this process, dripping with potassium hydroxide. Next, the white precipitate formed was washed with distilled water and centrifuged until pH 7. Subsequently, dried in a circulation oven at 60°C for 24 h. Polymeric Precursors Method (PPM-600°C/2 h) Initially, citric acid (20 g) and 50 mL of deionized water were added into a 300 mL beaker and stirred continuously until complete dissolution. Next, zinc acetate (5.20 g) was dissolved in 20 mL of deionized water mixed with the citric acid solution. This procedure was followed by adding ammonium hydroxide (18 mL) with continuous stirring and heating at 60ºC until the solution volume reached 50 mL. Then, ethylene glycol (12 mL) was added to the reaction. The system temperature was raised to 90°C to evaporate as much water as possible. The beaker was then placed in an oven, and the gel was calcined at 300°C for 2 h. Subsequently, this material was agglomerated, and the resulting product was then heat-treated in an oven at 600°C for 2 h.

Top-Down Method -Milling (M-16 h/2000 rpm)
The technique to mechanically reduce the particle size consisted of filling the attritor mill vessel (Netzsch PE 075) in 60% of its volume with ZnO:water at a ratio of 1:2 (w w − 1 ). Thus, 56 g of commercial ZnO was poured into the vessel into 112 mL of distilled water. Zirconium spheres (3 mm diameter) were used as a milling medium. Milling was performed with a rotation of 2000 rpm for 16 h. After the milling process, the material was separated from the milling medium and taken to the circulation oven for drying at 80ºC for 24 h.

Preparation of Starch:MMT Sachets
The formulation of starch films that compose the sachet matrix was prepared according to Moreira et al. [36], using food corn starch. In this step, MMT concentrations of 1, 2, 4, and 10% (w w − 1 ) were evaluated concerning the amount of starch. Next, the MMT was homogenized with glycerol (12 g) and water (5.6 g) in a tip ultrasound (Branson 500) for 1 min with 30% frequency intensity and then homogenized with the dry mixture containing 28 g of starch and 0.28 g of stearic acid. Next, the obtained mixture was processed in a mixing chamber with roller rotors coupled with a Haake rheometer, processed under a rotation speed equal to 160 rpm and a temperature of 130ºC for 4 min. Then, the starch films were obtained from the plastic material in a thermopress at 140°C for 10 min with 10 tons of applied force.
The selected films were molded to obtain the sachet for storing the ZnO. In this step, 2 cm x 2 cm squares were cut. Then, a heat press was heated to 140°C, and the three edges of the sachets were sealed by heating for 3 s. Finally, 50 mg of ZnO was introduced internally, and the last edge was sealed in the same condition as the previous ones.

Characterizations
X-ray diffraction (XRD) measurements were performed on a Shimadzu XRD 6000 diffractometer using Cu Kα radiation (λ = 1.5488 Å) with 30 kV of operation voltage and 30 mA of current. The sample morphology was evaluated using a Field Emission Scanning Electron Microscope (SEM) Zeiss SIGMA model with a cannon mission operating at current extraction voltages of 2 to 10 kV. The zeta potential and dynamic light scattering (DLS) were measured using a Malvern Zeta sizer nano-Zs model. The specific surface areas (SSA) were measured by nitrogen adsorption at 77 K (Micromeritics ASAP − 2020) and calculated by the Brunauer-Emmett-Teller (BET) method. The film permeabilities to water vapor were characterized in triplicate based on the ASTM E96 standard. The assembled system was left to rest in an oven with external ventilation at a temperature of 35°C. The system mass was measured at intervals of 2 h for 2 days. The samples' mechanical responses in a universal mechanical testing machine (EMIC DL2000) were performed following the ASTM D638 standard, traction speed of 5 mm min − 1 , and load cells of 50 kgf in a controlled humidity.

Zn 2+ Ions Solubility and Release Assays
The ZnO nanoparticles solubility evaluation was performed by adding 50 mg of samples in 50 mL of 2% (w w − 1 ) citric acid extractive solution (in triplicate). The flasks were reserved at room temperature, and the aliquots were removed at different intervals, from 5 min to 120 h. Each aliquot was separated after centrifugation at 8000 rpm for 10 min at 25ºC. The Zn 2+ ions in the solution were determined using a Flame Atomic Absorption Spectrophotometer (FAAS, Perkin Elmer, PinAAcle 900T).
For the Zn 2+ ions release, the sachets were immersed in 500 mL of extractive solution, in quintuplicate, at room temperature. The times for removing aliquots ranged from 24 to 720 h. The determination of Zn 2+ ions was performed by removing 5 mL aliquots and subsequent centrifugation for reading in the FAAS.

ZnO Nanoparticles
XRD technic performed the structural characterization for the ZnO samples obtained from bottom-up and top-down approaches. All the diffractograms in Fig. 1 showed peaks referring to the ZnO wurtzite structure crystallographic planes, according to JCPDS card n° 361,451. Additional planes were not detected, indicating the absence of contamination from secondary phases. This result can be attributed to the possible reduction in particle size diameters subjected to the milling process [37].
The Raman spectroscopy analysis showed that all materials obtained present the same optical vibrational behavior with peaks at 97 cm − 1 and 439 cm − 1 , attributed to the E 2 low and E 2 high modes corresponding to Zn-O vibration. Such results observed for ZnO nanoparticles agree with those reported in the literature [9]. Additionally, the sample M-16 h/ 2000 rpm (Fig. 2a) also exhibits broader peaks, reinforcing that the milling process minimized the particle sizes, as suggested by the XRD results (Fig. 1). Figure 3 shows SEM-FEG images of ZnO nanoparticle samples obtained by bottom-up and top-down routes. Commercial ZnO (Fig. 3a) had a size range of 174 ± 95 nm, indicating particles with well-defined edges. However, when compared with ZnO M-16 h/ 2000 rpm (Fig. 3b), the particles are smaller than the commercial one, at 71 nm ± 41, with a less defined shape. These images verified the milled range of 36 nm ± 6 and 30 nm ± 8, in the form of nanoplates and nanometric hedgehogs, respectively. However, in the image of PPM-600ºC/2 h, nanometric particles with more significant heterogeneity were observed than the other chemical syntheses, with a size of 46 nm ± 14 and without a defined shape. Chen et al. [38] observed the same result after evaluating the temperature increase from 400 to 600ºC by polymeric precursor synthesis to obtain CuO/ZnO/ZrO catalyst nanoparticles. Table 1 presents the N 2 sorption/desorption results by the BET method, showing the decrease in particle size effects on the surface textural characteristics and the Zeta Potential stability results. All SSA nanoparticle values significantly increased compared to commercial ZnO one (3.1 m² g − 1 ). The increase in area and pore size for ZnO M-16 h/2000 rpm (18.5 m² g − 1 and 140.2 nm) results from the particles' decrease in size and greater homogeneity (174 nm ± 95 to 71 nm ± 41) due to processing that caused collisions during the milling process. On the other hand, ZnO (P-25ºC/1 h) had the highest area and pore size among the syntheses, with 12.8 m² g − 1 and 206.7 nm, respectively. This result can be attributed to this no-heat treatment route, promoting pore closure and particle growth. Furthermore, the internal channels between the nanoplates that formed hedgehogs (Fig. 3d) can also explain the pore size behavior.
The Zeta Potential (Table 1) shows the stability of the particles in aqueous media. Commercial ZnO presented a positive charge equivalent to + 27 mV, similar to the milled ZnO to reinforce the results observed in the structural analyses ( Figs. 1 and 2), which indicated peaks broadening associated with decreasing nanoparticle sizes.
The samples HT-200ºC/3 h and P-25ºC/1 h presented well-defined homogeneous particles, thickness sizes in the   (Fig. 4b). This result can be attributed to the nanoparticles' size diminishing during milling, increasing suspension stability. Regarding the synthetic routes, the HT-200ºC/3 h (Fig. 4c) presented a cluster distribution range predominantly between 164 and 825 nm, mean of 417 nm ± 215. A similar result was observed with cluster heterogeneity for PPM-600ºC/2 h particles in Fig. 4e (91 to 825 nm, mean 377 nm ± 266).
On the other hand, ZnO P-25ºC/1 h nanoparticles (Fig. 4d) showed distribution homogeneity in the region of 459 to 1280 nm (mean 764 nm ± 302) compared to the previous synthesis (PPM-600ºC/2 h). The higher average size value can be attributed to the lowest Zeta potential value (|11.3| eV), indicating the agglomeration effect due to the hedgehog-like shape (Fig. 3). The cluster formation, displayed in the histogram, correlates with low Zeta potential values for HT-200ºC/3 h and by PPM-600ºC/2 h syntheses, with a value of |17| mV. Thus, the lower degree of the agglomeration process is intensified by the load repulsion, and the potential values agree with the literature [41]. Comparing ZnO obtained by the milling method to the others proved superior since the particles showed the best behavior in the solution.
The Zn 2+ ions released test from ZnO-free particles in a 2% citric acid (w w − 1 ) extractive medium (Fig. 5). The release of the ion from the commercial oxide was gradual, over 120 h, with full release only at the end of this period. ZnO M-16 h/2000 rpm reached a plateau of 90 and 100% Zn 2+ in just 48 h, indicating that the reduction in particle size by the milling process and the higher dispersion capacity sample ZnO M-16 h/2000 rpm with a value of + 28 mV. Thus, the relative values between commercial and milled ZnO indicate the absence of the charged surface. A similar result was observed by Adhikari et al. [39], that the value of pure zinc oxide found was + 27 mV at pH 7. The zeta potential values for the synthetic hydrothermal, precipitation and polymeric precursors samples were + 17.1 mV ± 1.4, + 11.3 mV ± 4, and + 17.9 mV ± 0.9, respectively. Such results were found for the syntheses due to the morphologies (Fig. 3) that allowed particle agglomeration and, consequently, smaller hydration sphere formations. The more significant reduction in the ZnO potential for P-25ºC/1 h may be due to the formation of caused hedgehog-like forms, reducing the hydration sphere. Kavitha et al. [40] demonstrated that ZnO potentials for the precipitation synthesis method presented a value equivalent to + 17 mV, the same value range found in the present work. Figure 4 shows the DLS technique's stability and dispersion capacity of ZnO samples. For commercial ZnO, clusters from 300 to 5000 nm and a mean value of 1777 nm ± 1570 were observed, indicating particle heterogeneity and agglomeration (174 nm ± 95), verified by SEM (Fig. 3a). The milling method for obtaining ZnO nanoparticles (M-16 h/2000 rpm) during the heat treatment, favoring the particles' agglomerate formation. In addition, the agglomeration dispersion in aqueous suspension (Fig. 4) also contributed to the low solubility of this material. Thus, the milled ZnO proved to be more efficient in increasing solubility due to decreasing particle size, increasing material reactivity, and better dispersion in aqueous media. The results were similar to those reported by Vale and Alcarde [42], confirming the release of zinc ions above 50% by the same extractive medium.

Biodegradable Sachets
The starch films' morphological characteristics are shown by SEM analysis in Fig. 6. A smooth surface was observed for the pure TPS film, with starch granule absence attributed to the effective plasticization process. However, all films with 1 to 10% MMT showed roughness (Fig. 6b-g). This result agrees with Zakaria et al. [43], which observed made the particles more soluble. The method via hydrothermal synthesis showed a low release percentage between 40 and 50% during the first 16 h, gradually increasing until reaching an average value of 87% ±5 in 120 h. The ZnO obtained via precipitation was similar to hydrothermal for the first 24 h, releasing about 80% ±4 at the end. However, this route registered a slower release behavior than ZnO via hydrothermal treatment. The results verified for HT-200ºC/3 h and P-25ºC/1 h can be attributed to the low ZnO reactivity due to the agglomeration effect of the obtained nanoparticles. Furthermore, the plate-shaped nanostructures (Fig. 3c,d) formed micrometric hedgehog-like forms, which minimize the solubilization effect. For the ZnO nanoparticles (PPM-600ºC/2 h), 71% ±1 release occurred in the first 5 h, remaining in the plateau between 60 and 70% until the end of 120 h. This behavior can be attributed to the characteristic of this synthetic route due to the elimination of organic material  Table 2 show the mechanical tensile behavior of TPS films. The TPS film with 1% MMT showed increased stress (3.8 MPa) and deformation (60% ±0.8) compared to pure TPS, with stress and deformation equivalent to 2.5 MPa and 38%, respectively. This increase is due to the transport of loads applied to the reinforcement elements (MMT), indicating the high homogeneity and distribution in the matrix [45]. Furthermore, the morphology of the MMT particles (in plates) tends to orient the load in the same direction in which it is tensioned, which favors better resistance to the composite [46]. The films with the addition of MMT from 2%, on the other hand, showed an increase in strain and reduced stress. This result can be attributed to the excess reinforcement load applied to the polymer matrix. In addition, the clay's ability to exfoliate between the polymer chains decreases the degree of organization, increasing the chain mobility, which favors the increased deformation ( Table 2) [29].
The film chosen to produce biodegradable sachets had an MMT concentration of 1% from the presented results. In addition, this film showed the best performance as a barrier to water vapor permeability (Fig. 7a) and superior mechanical strength (Fig. 7b), attributed to better dispersion of the nanoparticulate material in the polymeric starch matrix.
The Zn 2+ ions released assays from the nanoparticulate ZnO obtained by milling method packaged in pure TPS and 1% MMT sachets were performed in an extractive medium of 2% citric acid (w w − 1 ). In Fig. 8, the results showed the Zn 2+ availability since the initial 24 h, with similar behavior between both sachets. The pure starch sachet released 82% ±12 of Zn 2+ ions, while the one reinforced with 1% MMT released 66% ±11 in the first 24 h. The free ZnO particles (Fig. 5) released 81% ±4 of Zn 2+ ions in 24 h (Fig. 5b), similar to that stored in the sachets. After 720 h (Fig. 8), the sachets presented a similar release percentage, 85% ±2 (TPS) and 83% ±1 (TPS + 1% MMT), maintaining the expected availability of free ZnO nanoparticles plateau. Therefore, the release behavior of both free ZnO and ZnO packaged in the sachets occurred similarly. The high solubility of the packed ZnO can be attributed to the hydrophilicity characteristic of the starch that compounds the sachet matrix, a behavior also observed in the WVP results a rougher increase in potato starch matrix films reinforced with calcium bentonite (1 to 20%).
The permeability of the films to water vapor was analyzed using the WVP test. Figure 7a showed that most films with MMT concentrations had permeability below what was found for the pure starch film. Therefore, the reinforcement material addition was carried out, among other possibilities, to reduce the polymer matrix permeability since starch provides a high hydrophilic characteristic. Among them, the film with 1% MMT showed the best permeability (12.31 g mm m − 2 h − 1 kPa − 1 ), below 16.03 g mm m − 2 h − 1 kPa − 1 , corresponding to pure TPS, indicating good clay mineral dispersion in the polymeric structure. This reduction can be explained by the tortuous path produced in the polymeric structure by the MMT plates, which makes it difficult and increases the way to be completed by the water molecules permeating the entire matrix [44]. The other concentrations did not significantly affect the result, presenting permeability values between 12.31 and 16.03 g mm m − 2 h − 1 kPa − 1 . ( Fig. 7a). The absence of particular behavior for ZnO packaging in TPS 1% MMT sachets is due to the low reinforcement material concentration.
In this way, the starch sachet may preserve and minimize dispersion losses, maintaining the property of fast solubility desired for fertilizers in the soil, and avoiding the mineralizing process in non-soluble forms.

Conclusion
Pure ZnO nanoparticles with distinct morphologies were obtained from top-down and bottom-up approaches. The topdown approach (milling method) produced nanoparticles that allowed higher solubility gain than the commercial and bottom-up samples. The milling process M-16 h/2000 rpm presented the best result as the increase in solubility, and this behavior was attributed to the nanoparticles' size decreasing after the milling process. The TPS composite with the insertion of 1% MMT in the polymeric film matrix presented the best result for permeability barrier and mechanical resistance ascribed to the MMT homogeneous distribution. The Zn 2+ release from ZnO nanoparticles packaged with pure starch and reinforced with 1% MMT sachets presented a similar performance, preserving the intrinsic characteristics of the free oxide and maintaining the solubility. The packaging in sachets and the rapid release of ZnO nanoparticles as a source of Zn 2+ ions enables the system to minimize losses due to leaching and air dispersion, providing adequate crop supplies. manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted