Introduction

The capacity of water and nutrient absorption by plant leaves has been well-known for more than three centuries (^{Fernández and Eichert, 2009}), making foliar fertilization a common strategy. In certain cases, foliar fertilization is justified, for example, i) when soil conditions limit the nutrient availability for roots; (ii) when there is a high likelihood of the nutrients applied on soil will be lost; and (iii) when plant demand is higher than the soil supply capacity (^{Fernández et al., 2013}). Zinc, Cu and Mn fulfill several functions in plants metabolism, such as enzymatic activation, reduction in oxidation radicals, and formation of reproductive organs and photosynthesis processes (^{Marschner and Marschner, 2012}; ^{Cakmak, 2000}; ^{Kirkby and Römheld, 2007}); therefore, deficiencies of these elements limit crop yields worldwide.

Micronutrient sources are generally labeled under inorganic sources, synthetic chelates, organic complexes, and silicate oxides. The inorganic sources include metallic salts, such as sulfate, chloride and nitrate (soluble in water), oxide, carbonate and phosphate (insoluble in water), and oxysulfate (intermediate solubility) (^{Lopes, 1999}). Due to their high nutrient contents, commercialization of micronutrients as oxides and carbonates is increasing in Brazil. At the beginning, these sources were used to supply micronutrients through the soil. However, finely ground oxides and carbonates maintained in suspension by adding additives created a new class of fluid fertilizer named as concentrated suspension (CS). The use of CS for foliar fertilization was limited in the past since the legislation allowed only water-soluble sources to be registered as foliar fertilizers. However, the recent shift in the legislation (^{MAPA, 2018}) allows registration of foliar fertilizers by declaring the content of water-soluble nutrient (for soluble fertilizer) or the total content (for nutrients present in CSs), creating a new scenario for commercialization of these products as foliar fertilizers.

Nevertheless, the capacity of plant leaves to absorb insoluble sources of micronutrients remains unclear, especially considering the unknown particle size and shape of the products, which affect penetration into leaf mesophyll. Preliminary studies have indicated an effective absorption of nanomaterials (particles < 100 nm) by plant leaves (^{Eichert et al., 2008}; ^{Prasad et al., 2012}). The correct characterization and development of foliar fertilizers, which improve the use of nutrients by plants, is one of the most important challenges for the fertilizer industry (^{Fernández et al., 2013}). The efficacy of new products for foliar fertilization must be accurately evaluated before their widespread usage in field crops.

This study aims to characterize commercial CSs to support the regulation of these products as foliar fertilizers and provide support to future studies on the capacity of plant leaves to absorb nutrients from these sources.

Materials and Methods

For this study, 13 commercial micronutrient sources (five of Mn, five of Zn and three of Cu) were selected, all in CS forms and belonged to five enterprises of the Brazilian fertilizer industry. Table 1 lists the nutrient concentrations as well as the densities of suspensions reported on the labels. The products were directly purchased from local stores and all were checked regarding their shelf life. Before the measurements, the samples were homogenized using an oscillating stirring table orbital shaker at 80 rpm for 24 h in their original flasks. Right after shaking, an aliquot of each fertilizer was transferred to a volumetric flask and diluted in deionized water up to a 100 mg L^–1 concentration of each nutrient, taking into account the concentration prescribed by the manufacturer.

The mean particle size was measured using the dynamic light scattering (DLS) technique according to the PCC-1 protocol (^{Hackley and Clogston, 2011}). The analyses were performed using the Zetasizer Nano Range facility using water as a dispersant (viscosity of 0.8872 cP and refractive index of 1.33), with an acquisition time of 60 s and a frequency of 398,000 counts per seconds (kcps). For the measurements, an aliquot of 1 mL of each fertilizer was transferred to a polystyrene cuvette and directly loaded into the equipment.

The aggregation potential was measured by the zeta potential using the same equipment. The zeta potential indicates the difference of electrostatic potential between a particle and the surrounding environment. The zeta potential gives the intensity in which the charges contrast on the particle surface, that is, the higher the value of the zeta potential (both positive and negative), the greater the difference between charges. For a modulus value higher than 30 mV, the repulsion becomes dominant, avoiding aggregation and providing stability to the material (^{Heurtault et al., 2003}). An aliquot of 0.5 mL of each fertilizer was transferred to a capillary cuvette, and directly loaded into the equipment. The analyses were carried out using water as dispersant (viscosity of 0.8872 cP, electric dispersion constant of 78.5 and refractive index of 1.33), a frequency of 1,008,000 readings per second (kcps) and an acquisition time of 12 s.

The shapes and sizes of the particles were evaluated using the scanning electron microscopy (SEM) analysis. The fertilizers were dripped onto a carbon plate coated with an aluminum foil and then loaded into the Magellan 400 L equipment. Approximately eight images were taken of each fertilizer with magnification between 50 and 350 thousand times.

The arsenic, cadmium, lead, chromium, mercury, and selenium concentrations were determined according to the SW-846-3051a methodology of the United States Environmental Protection Agency (^{USEPA, 1998}). Samples were digested using a 3:1 mixture of nitric and hydrochloric acids (65 % and 37 %, respectively) in a microwave closed system, and quantified by inductively coupled plasma atomic emission spectrometry (ICP-OES). The program used for digestion consisted of heating the sample to 175 °C for 5 min, keeping the temperature at 175 °C for 10 min then allowing 10 min for cooling.

Results and Discussion

Scanning Electron Microscopy

The results of the SEM analysis showed that the particle shapes varied broadly among the products. Fertilizers containing Cu presented spherical or cylindrical forms (Figure 1), while the Mn fertilizers demonstrated larger aggregates with lamellar structures (Figure 2). The shape of the particles was more homogeneous for Zn fertilizers, close to parallelepipeds or cylinders (Figure 3). Differences in particle shapes observed among the products may be related to the mineral used as raw materials to produce the fertilizer, changes during processing, and degrees of hardness during the milling process.

Figure 1 Scanning electron microscope images of fertilizers containing CuO. A, B, and C correspond to the fertilizers used in Tables 1 to 4. 

Figure 2 Scanning electron microscope images of fertilizers containing MnCO₃. D, E, F, G, and H correspond to the fertilizers used in Tables 1 to 4. 

Figure 3 Scanning electron microscope images of fertilizers containing ZnO. I, J, K, L, and M correspond to the fertilizers used in Tables 1 to 4. 

All fertilizers presented a high heterogeneity in terms of particle size. For Cu fertilizers, the smallest particles were observed in A, and most were smaller than 200 nm, while B and C presented conglomerates with larger sizes, that is, between 270 and 950 nm (Figure 1). Mn fertilizers showed particles between 200 and 400 nm in F and H, which was different from the others (D, E and G) in which the intense aggregation rate built larger structural units with diameters between 500 and 1300 nm (Figure 2). The highest particle size heterogeneity was observed in all Zn fertilizers, which ranged from tiny particles smaller than 100 nm to larger ones close to 500 nm (Figure 3).

Differently from the DLS technique, which evaluates the behavior of a compound possibly formed by the aggregation of several particles, the SEM technique allows individual visualization of the particles, even those aggregated to other structures. This makes these analyses complementary since DLS indicates the behavior of the structure in the dispersion at the moment of the analysis, while SEM exhibits particles that could be dispersed before absorption by the leaves.

Conclusion

The average hydrodynamic diameter of the CSs with Cu, Mn, and Zn were 315 ± 55, 378 ± 184 and 435 ± 107 nm, respectively, revealing that these fertilizers do not fit in the nanomaterial or nanofertilizer specifications (< 100 nm). The zeta potential of the CSs varied from –20 to –30 mV, suggesting possible aggregation of the particles during handling and application. The scanning electron microscope images indicated a large variety in sizes and shapes of particles. Only Zn fertilizers presented some particle sizes smaller than 100 nm.

The CSs presented concentrations of arsenic, cadmium, lead, chromium, and mercury within the limits allowed by the Brazilian legislation. More studies are required to quantify effectively the potential absorption of these fertilizers by plant leaves to provide support to the current Brazilian legislation of fertilizers.