Bioactive Coating With Low Reactive Sources for Application as Multi-nutrient Fertilizer Granules


 Fertilization is essential to provide suitable conditions for plant development and crop productivity, but the environmental cost of fertilizer production is a drawback for achieving a sustainable agriculture. Using unprocessed (raw) nutrient sources such as elemental sulfur (S0) and mineral oxides (ZnO, MnO, CuO) as fertilizers reduces the environmental impact, although they are not readily available to plants. Thus, we developed a polymeric coating material based on gelatinized starch loaded with S0 and oxides, and selected microorganisms – Aspergillus niger and Aciditiobacillus thiooxidans – aiming at a multi-nutrient fertilizer in which the biological components improve the solubility of the low reactive nutrient sources. The acidifying capacity of both microorganisms led to a synergic release and increased the availability of micronutrients and the elemental sulfur oxidation rate. For instance, the polymeric coating composition enabled a sulfate release of up to 76.4 and 71.3% for A. niger and A. thiooxidans, respectively. This innovative system can effectively supply nutrients to plants through the use of cheap and low reactivity nutrient sources with the advantage that it can be applied on currently used fertilizer granules, making easier the adoption by producers.


Introduction
Fertilizers are de ned as sources of macro and micronutrients, which are essential inputs to high crop production. Macronutrients are highly demanded fundamental building blocks of plant structures. On the contrary, micronutrients are required in much smaller quantities by plants, but still limit crop production yields [1][2]. Micronutrients assist on important plant processes and functions, e.g., transport of genetic material, tissue growth, and amino acids synthesis. Moreover, they are constituents of chlorophyll, which is responsible for photosynthesis [2][3][4][5]. Micronutrient losses in soil include leaching of soluble salts (e.g., ZnSO 4 ) and/or precipitation of salts (e.g., interaction of Zn 2+ with phosphate ion, forming zinc phosphate, a poorly soluble salt) [6].
Mineral oxides, such as ZnO, CuO and MnO, display higher elemental contents compared to salts, as well as lower production and environmental costs since they are raw materials. However, these oxides are poorly water-soluble, requiring processes that involve strong inorganic acids and high energy consumption to increase their solubility. This is the typical case of sulfur, a macronutrient often applied as elemental sulfur (Sº), but only absorbed by plants as sulfate anions (SO 4 2− ) [7][8][9]. Moreover, the conversion processes of S° and micronutrients in soil is limited by the soil microbiological activity, including direct metabolization and indirect effects from secreted organic acids [10][11][12].
Thus, an adequate soil enrichment with selected microorganisms can potentially increase the availability of these low reactive nutrient sources. Fungi like Aspergillus niger (A. niger) [13][14] and the bacterium Acidithiobacillus thiooxidans (A. thiooxidans) [15][16] have demonstrated a signi cant potential for improving nutrient availability and e ciency [17]. A. niger produces organic acids and promotes S°o xidation through indirect effects related to local pH reduction and enzyme secretion [18][19]. A. thiooxidans xes CO 2 and grows chemolithoautotrophically using sulfur as an energy source, thereby producing sulphuric acid [15]. Despite the recognized potential of these microorganisms in promoting nutrient solubilization, tailoring the fertilizer structure to maximize the bene ts from microbial activity still remains an unsolved challenge.
Coating granular fertilizers with adequate materials can provide an advantageous support for microorganism vehiculation, promoting cellular protection during application and nutritional support for its reactivation and growth. We have previously demonstrated that maize starch is suitable to disperse high fertilizer particle contents (mineral oxides and elemental sulfur), and forms continuous, welldispersed lms over the fertilizer granules [13][14]. Maize starch is composed of amylose (30 wt.%) and amylopectin (70 wt.%) macromolecules that can act as an energy source for microbial growth. Starch is also natural, biodegradable, abundant and non-toxic, which are desirable features from an agronomic point of view [13][14].
Herein, we developed a bioactive coating material based on gelatinized starch containing balanced contents of S 0 and oxides as sources of micronutrients (ZnO, MnO and CuO). This system has been used as an encapsulating matrix for the acidifying fungus Aspergillus niger or the oxidizing bacteria Acidithiobacillus thiooxidans to increase the micronutrient availability of low reactivity sources. The lmforming ability of the composition was very adequate to coat monoammonium phosphate (MAP) granules, which was an ideal vehicle for the targeted composition. Solubilization and soil incubation assays indicated that the supporting fertilizer (MAP) did not affect the sulfur and micronutrients solubilization neither its release rate was affected by the coating. Our results indicate that this composite coating is a simple strategy to deliver multiple nutrients with minimal pre-processing, saving energy and avoiding byproduct generation. wt.%), and copper oxide (CuO Mineral , 28 wt.%) were supplied by Heringer Fertilizers (Brazil). All materials were used as received. Monoammonium phosphate (MAP) granules (diameter ranging between 2.8 and 3.2 mm) were kindly provided by Adubos Vera Cruz, Ltd.

Microorganisms
Aspergillus niger C (BRMCTAA 82), a lamentous fungus, was obtained from Embrapa Food Technology collection (Rio de Janeiro, RJ, Brazil). Spore suspensions of Aspergillus niger (A. niger) were kept at -18°C in a solution of water with glycerol (20 wt.%) and NaCl (0.9 wt.%). Spores were germinated on Petri dishes containing potato dextrose agar at 30°C. After 96 h, a suspension of grown spores was harvested by adding distilled water. The spore concentration was determined using a Neubauer chamber.
The bacterium Acidithiobacillus thiooxidans was obtained from São Paulo State University (Araraquara-SP, Brazil). Suspensions of Acidithiobacillus thiooxidans (A. thiooxidans) were kept at 4°C in a medium 9K [15]. The bacterium was reactivated on 9K medium in an orbital shaker incubator at 30°C and 150 rpm for 10 days. The nal medium was vacuum ltered using Whatman No. 1 lter paper and, subsequently, the cellular concentration was determined using a turbidimetric method.

Coating of fertilizer granules with micronutrients and microorganisms
Biocoating loaded with each microorganism was prepared by maize starch (St) gelatinization. This process followed the methodology described by Klaic et al. (2018), which consisted in the dispersion of starch (8 wt.%) and glycerol (4 wt.%) in distilled water (250 mL) [13][14]. The gelatinization process was carried out by keeping the dispersed starch at about 90 ºC for 30 min under stirring until a sticky St paste gel was formed. Elemental sulfur (256 ± 32 nm) and the mineral oxide mix (ZnO, MnO and CuO, average particle sizes of 446 ± 46, 325 ± 55 and 302 ± 51 nm, respectively), both sources of micronutrients, were transferred to the St gel at a proportion of 1:2 between starch (33.6 wt.%) and mix particulate materials (66.4 wt.%), and their dispersion was accomplished by vigorous agitation for 15 min. In addition, ZnO, MnO, CuO and Sº were added at a ratio of 6.4, 7.6, 17.7 and 35.5 wt.%, respectively. As the mineral oxides have impurities, the oxide content added to the matrix was previously calculated to obtain a nal micronutrient concentration (Zn 2+ , Mn 2+ and Cu 2+ ) of around 3.75 wt.% and 35 wt.% for Sº. After dispersing the particulate materials in the St gel, the temperature was reduced to 30 ºC for incorporating the A. niger spores (or A. thiooxidans cells). The fungi spore (or bacterial cell) suspensions were prepared as described in Section Microorganisms to achieve a concentration of 1x10 8 fungi spores (or bacterial cells) per gram of material.
The coating process was carried out by dispersing the biopolymer over the MAP granules using a metal turntable coater with 25 cm side shields, rotating at 30 rpm, under air ow heated at 40 ºC [20]. Table 1 summarizes the materials and codes of the coated and uncoated granule samples. Coated MAP granules without A. niger or A. thiooxidans (here denoted as MAP mix ) were also prepared for comparison purposes.
Furthermore, physical and chemical characterization, as well as bioactivation experiments in liquid culture medium and soil, were carried out on all materials. The experimental methodologies, including the analytical methods for the determination of micronutrients and solubilized phosphorus, were described in detail in the Supplementary Material.  Figure 1 shows the interface of MAP granules with and without the bioactive coating layer. The SEM image of uncoated MAP exhibits a rough surface (Fig. 1a) Figure 2 shows the temporal pro les of pH, solubilized P, Zn 2+ , Mn 2+ and Cu 2+ , and of elemental sulfur (Sº) oxidation for the controls and A. niger-loaded samples in liquid nutrient medium. All samples evaluated and their nomenclatures were described in Table 1. This experiment provided the minimum conditions required for microorganism germination and fungal growth, evaluating the potential of the bioactive coating in promoting micronutrient solubilization and elemental sulfur oxidation. These results showed that the A. niger spores (MAP Mix+A.niger ) germinated and proliferated throughout the bioactive coating, indicating that the starch coating was e cient to maintain the A. niger spores viable after the coating process. Moreover, the starch coating did not delay the phosphorus (P) release, since all granules reached above 80% of P solubilized in 3 days, as shown in Fig. 2a. This result was expected because MAP has a high solubility, and the maize starch coating does not impose a physical barrier to the phosphate solubilization. In addition, it is important to highlight that after the 3th day a slight decreasing in P solubilized was observed for MAP Mix+A.niger . This reduction suggests that the microorganisms consumed an amount of P in their metabolic process like cell reproduction and/or production of new molecules/metabolites [21]. Figure 2b presents the temporal pro le of pH, and the results showed a lower pH reduction for the Mix (control treatment), remaining virtually the same pH value from the 6th to 12th day. It is possible to note that MAP (uncoated) and MAP Mix (without microorganism) also showed a slight pH decrease until the 3th day of incubation, with pH stabilized at around 5.5. However, the MAP Mix+A.niger granules showed a signi cant pH reduction from 7 to 3, which is associated with the ability of A. niger to produce acidity as organic acids [13,[22][23]. niger acidi cation is mainly due to the production of organic acids, such as citric acid, oxalic acid and gluconic acid, which present an indirect effect on the elemental sulfur oxidation, since low pH and some enzymes (such as sulfatase) intensify the oxidation of S 0 [18][19]24]. The biological oxidation process that converts S° to SO 4 2− is essential for plants, because Sº is absorbed only in the sulfate form [25][26][27]. A similar set of experiments was carried out using the bioactive coating loaded with A. thiooxidans. Figure 3 shows the temporal pro les of pH, and solubilized P, Zn 2+ , Mn 2+ and Cu 2+ , as well as of Sº oxidation in liquid nutrient medium for the control samples and that containing A. thiooxidans. The results showed that the A. thiooxidans cells (MAP Mix+A.thio ) proliferated within the bioactive coating, given that elemental sulfur was oxidized with simultaneous pH reduction. This shows that the starch coating was e cient in keeping the A. thiooxidans cells viable after the coating process, as noted for the A. niger spores (Fig. 2). As seen for MAP Mix+A.niger (Fig. 2a), the A. thiooxidans/starch coating was not effective in delaying the P release, as shown in Fig. 3a. In addition, a slight decrease in P solubilized was observed for MAP Mix+A.thio after the 3th day of incubation, suggesting that A. thiooxidans consumed the solubilized P in its metabolic process at some extent. Figure 3b displays the temporal pro le of pH, and the results showed that MAP Mix+A.thio increased the pH of the medium from 2.8 to 4.0. This effect is most likely due to the buffering capacity of MAP and the consequent alkalization of the medium provided by high solubility of MAP, in contrast to that seen in Fig. 2b. It is interesting to emphasize that the acidi cation mechanisms used by A. niger (fungus) to produce acidity is different from that used by A. thiooxidans (A. niger produces organic acids by a secondary metabolism, while A. thiooxidans produces SO 4 2− by consuming S 0 as an energy source) [15,16,28]. Figure 3c shows the temporal pro les of elemental sulfur (Sº) oxidation for the controls and A. thiooxidans-loaded samples. Uncoated MAP and Mix showed elemental sulfur oxidation below 1% over the 12 days of incubation, while MAP Mix provided a S 0 oxidation of 5%. By contrast, the bioactive coating with A. thiooxidans MAP Mix+A.thio , provided higher oxidation rates, reaching 100% of Sº oxidation.

Materials characterization and incubation in liquid culture medium
Therefore, A. thiooxidans is more e cient than A. niger in oxidizing elemental sulfur due to its inherent oxidative metabolism [15][16]. Figures 3d-3f  cations. Although A. thiobacillus oxidized all elemental sulfur (Fig. 3c), the buffering effect of MAP on the medium may have impaired the solubilization of oxides [29].
The results showed that, despite A. thiooxidans displays the best oxidizing e ciency onto Sº, A. niger promotes better micronutrient availability from the low reactive mineral sources. This is due to the kind of acidity produced by A. niger, e.g., organic acids that dissolve heavy metals by forming soluble metal complexes and chelates [30][31]. Moreover, other compounds that have at least two hydrophilic reactive groups (such as phenol derivatives) may also have been excreted into the culture medium during microbial cultivation and helped solubilize the mineral oxides [30]. Other mechanisms, such as enzymatic reduction of highly oxidized metal compounds may also play a role in the oxide solubilization process, although the enzymatic effect is low in comparison with the acidi cation due to organic acids [31]. Figure 4 shows the temporal pro les of pH, Sº oxidation and Zn 2+ , Mn 2+ and Cu 2+ solubilization in the soil incubation experiments, which were performed to evaluate the effects of the bioactive coating under more real conditions. It is possible to notice in Fig. 4a that the control treatment (Mix) had a minimal pH variation remaining at 6.3 after 42 days of incubation. On the other hand, the treatments with the MAP granules showed a similar behavior with increasing pH in 3 days, except for MAP Mix+A.thio . MAP solubilization basi es the soil and acts as a buffer, reaching a pH 7.1. For MAP Mix+A.thio , pH increased after 7 days of incubation, but signi cantly reduced after 42 days ( Table 2). The increased acidi cation found with MAP Mix+A.niger and MAP Mix+A.thio was led by oxidation of Sº to SO 4 2− and organic acid production, as aforementioned. and Cu 2+ of 3.8, 4.1 and 3.4%, respectively. This indicates that the particulate dispersion process was not su cient to result in a signi cant oxide solubilization due to the low solubility of oxides in water. This effect points out the role of the microorganisms loaded in the bioactive coatings, creating a synergistic effect that signi cantly enhanced the oxide solubilization, as well as increased microorganism viability in soil. Noteworthy, these fertilizers are frequently applied in arid soils, which typically display low biological activity. suggesting a precipitation effect. This can be associated with the increase in pH, leading to the precipitation of metals (especially Cu 2+ ) and explains their low solubilization. On the other hand, the organic acids (produced by A. niger) chelate the cations, reducing precipitation and therefore increasing their solubilization, which are de ned as acidolysis and complexolysis mechanisms, respectively. To con rm this hypothesis, Table 3 shows the solubilization of the respective oxides (ZnO, MnO, CuO) in different organic acids (citric and oxalic acid) and inorganic acids (sulfuric and hydrochloric acid). The experiment showed that organic acids (citric and oxalic) were more e cient in solubilizing oxides when compared to sulfuric and hydrochloric acids under similar conditions.   [13,24] Finally, the soil incubation test provides a more adverse environmental condition for microorganisms when compared to the liquid system. It was noted that the bioactive coating was also e cient in keeping the microorganisms viable to act as promoters of micronutrient solubilization and S° oxidation. This shows the potential of the coating strategy developed herein and opens new opportunities for the development of new class of fertilizer coatings using microorganisms.

Conclusions
The developed coating formulation, comprising a starch matrix, S 0 and oxides (ZnO, MnO, CuO) was effective in encapsulating A. niger or A. thiooxidans cells, and had good adhesion on MAP granules, leading to an ideal vehicle for micronutrients -over other fertilizers. There was a synergy between the nutrients and microorganisms incorporated in the biocoatings, resulting in system acidi cation and increasing both micronutrient availability and elemental sulfur oxidation. This innovative system can ensure the supply of multiple nutrients to plants by using low reactivity, inexpensive sources through a more environmentally-friendly management. In addition, the biocoating process can also incorporate (or encapsulate) other microorganisms of agronomic interest to be applied on different granular fertilizers.