Agronomic Efficiency of Biotite in Soybean and Corn Silage Production

,


Introduction
Potassium (K) is one of the major nutrients for plant growth and development. It plays an important role in agricultural production, besides its part in helping defensive compounds against stress, pests and diseases (Dhillon et al., 2019). In addition, K is fundamental for many metabolic processes, including photosynthesis, protein synthesis and solute transport (Prajapati & Modi, 2012). Insufficient K replacement in the soil can result in nutritional depletion and, consequently, a decline in crop yield.
Based on Brazilian Law 12.890/2013, soil remineralizers (SR) are defined as all mineral materials undergoing size reduction and classification based on mechanical processes; they change soil fertility by adding macro and micronutrients to plants, as well as by improving soil physical or physicochemical properties or its biological activity (Brasil, 2013). Previous studies carried out in tropical Brazilian soils have shown the potential of rock powders, including calxist, acidic volcanic rock, melilitite, olivine, siltstone, tephrite, among others, to be used as SR in agriculture (Ramos et al., 2019;Cunha & Almeida, 2021;Medeiros et al., 2021;Citadin et al., 2022). Although these studies were based on using SR in perennial and annual species, such as oat and barley crops, studies focused on the effect of rock powders deriving from biotite mica on soybean and maize remain scarce. Therefore, if one takes into consideration the importance of adopting alternative fertilizer sources for agriculture, and the relevance of using ground silicate rock powders plant growth, the aim of the current study was to assess the agronomic efficiency and the potential of using the biotite provided by Embu Mineral Company as source of nutrient to soybean (Glycine max (L.) Merrill.] and corn silage (Zea mays L.) crops grown in succession systems, in different soils.

Location and Soil Characterization
The study was conducted from 2020 to 2021, in the greenhouse of the Agronomy School of Federal University of Goias (UFG), Midwestern Brazil, at coordinates 16°40′22″S and 49°15′19″W. The experiment was performed in 9-L capacity plastic pots (or 0.009 m 3 ) filled with soil (experimental unit).
Samples of both soils were collected from the topsoil layer (0.00-0.20 m) in savanna (Cerrado) sites in Goiás State, Brazil. They were air-dried, ground and sieved in 2-mm mesh; subsequently, they were analyzed for selected physical and chemical characteristics (Table 5). Samples were air-dried and sieved in 2-mm mesh for the analysis applied to soil physical and chemical characteristics; then, they were analyzed based on the methods described by Embrapa (2017).

Experimental Design and Treatments
The study was a completely randomized design with four replicates, and seven treatments, for both soil textures: witness, biotite (BE) remineralizer from Embu Mineral Company at four increasing K 2 O rates (30, 60, 120 and 240 kg K 2 O ha -1 ), KCl at 60 kg K 2 O ha -1 , and FMX remineralizer (fine-graded mica schist from Pedreira Araguaia Mineral Company) ( Table 6). Both KCl and FMX were used as reference K 2 O sources. Nutrients, such as N and P, were provided in the form of monoammonium phosphate (MAP), as needed for cultivation. Table 6. Composition of the applied treatments

Soil Incubation and Plant Cultivation
Each pot was filled with a mix of air-dried soils, CaCO 3 (100% CaCO 3 , equivalent to the rate used to obtain 60% base saturation) and specified K 2 O rates. This mix was incubated for 30 days to trigger a reaction with the soil, under greenhouse conditions.
Five soybean seeds (cultivar Brasmax Desafio RR-8473 RSF) were sown in each pot after the incubation period was over. The seedlings were thinned to one plant after emergence. Soybean plants were cultivated for three months. Soil moisture was kept close to 80% of field capacity. This was accomplished by replenishing moisture with deionized water in the upper portion of the pots.
Plant shoot and soil samples were collected from each experimental unit at the end of the soybean experiment and maize silage (cultivar BRS 3046) was grown in succession, based on the same treatments and soil samples previously used in soybean. Five maize silage seeds were sown in each pot and seedlings were, once more, thinned to one plant per pot after their emergence. Maize silage plants were cultivated for 75 days; the process to keep soil moisture was repeated. Plant shoot and soil samples were also collected from each pot after maize silage.

Soil-Plant Sampling and Analysis
Soil samples were air-dried, sieved in 2-mm mesh and analyzed for exchangeable K +, based on methods proposed by Embrapa (2017)-K extraction by using Mehlich-I solution, after collection from each experimental unit. Soybean and maize silage shoots were dried in a forced circulation oven, at 65 °C, until they reached constant weight. The leaves were ground in Willey knife mill (< 40 mesh), packaged, labeled and sent to the laboratory. Total K content was determined based on the methodology proposed by Malavolta et al. (1997).
Soybean grain weight was converted into kilograms per hectare (kg ha -1 ) after harvest in order to determine the soybean grain yield; then, the results were converted sacks per hectare (sc ha -1 ), since the 60 kg sack is the unit for soybean sales in Brazil. Maize silage yield was found by converting the shoot dry mass into tons per hectare (t ha -1 ).

Data Analysis
Data of each crop (soybean and maize silage) were subjected to analysis of variance through F test, at 5% and 1% significance level, in separate; means were compared through Tukey test, at 5% significance level. K rate effects on soils and plants were assessed through polynomial regression analysis. All analyses were performed in Statical Analysis System (SAS) software. Relative efficiency of biotites' remineralizer potential in comparison to reference sources-FMX remineralizer and potassium chloride-was calculated through the following mathematical expression:

Biotite Characterization
Particle size distribution (Table 1) has shown that BE meets the parameters defined by NI N. 5/MAPA, according to which, it is mandatory to have 100% of the particles passing through the 2-mm sieve, 70% or more has to pass through 0.84 mm sieve, and 50% or more has to pass through 0.3 mm sieve.
XRD results (Table 2) showed that BE is composed of oligoclase (35%), biotite (26%), quartz (24%), microcline (14%) and small occurrences of augite (< 1%). According to geochemical parameters established by NI N. 5/2016, free silica limit (as quartz, SiO 2 ) in SR must be lower than 25% (v/v). Thus, BE rocks can be classified as SR, as long as the silica content is below the maximum value determined by the legislation.
Elemental oxide concentration results, as well as trace elements determined through XRF analysis, are summarized in Tables 3 and 4. The BE sample had high silica (> 62%), aluminum oxide (14%) and iron oxide (> 9%) contents. Low CaO (2.24%), MgO (1.96%) and K 2 O (5.31%) contents were also observed in BE. However, the sum of bases reached values higher than 9.5%, which is higher than the minimum requirement established by the legislation (9%); therefore, this number is enough to classify BE rocks as soil remineralizing product.
Potentially toxic elements (Hg, Cd, Pb and As) ( Table 4) recorded values lower than the maximum levels (As: 15, Cd: 10, Hg: 0.1 and Pb: 200 mg kg -1 ). Further elements, such as Cu, Mo and Ni (which do not have a previously established limit) were detected at levels that represent no potential risk for agricultural use. Accordingly, BE rocks are safe to be used in agriculture as SR.

Clayey Red Latosol (LV)
Soybean yield (Table 7) in clayey LV soil has presented significant differences based on the F test (26.1), at 14.74% coefficient variation. Soybean yield ranged from 342.1 to 2,439.5 kg ha -1 . According to data by CONAB (2022), soybean 2012/22 harvest season ended in July. The influence of the La Ninã phenomenon on the Southern region and on Mato Grosso do Sul State, which showed significant rainfall decrease, was a determining factor to decrease yield in these regions; consequently, to decrease the total soybean yield in the country. In total 40,950.6 thousand hectares were sown in this crop season; it was 4.5% higher than that of the 2020/21 harvest season. The recorded yield reached 124,047.8 thousand tons; this value is 10.2% lower than that recorded for the 2020/21 season and mean yield reached 3,029 Kg/Ha-this number reflects the water shortage.
Knowing the association between plant growth and development, and genotypes' yield components is essential to define the most productive plant type. Furthermore, knowing the yield components and how they can interfere with soybean final yield can help the positioning of management practices in order to reach higher final yield (Navarro Júnior & Costa, 2002).
Treatments 60 FMX, 60 KCl and 60 kg ha -1 of biotite K 2 O stood out for recording the highest yields. There were no differences in the Tukey test between treatments, except for the witness. Reference standards 60KCl and 60FMX, and biotite dose differences, evidenced differences in comparison to the witness (Table 7). Table 7 highlights that all doses of EMBU remineralizer did not present significant differences in the Tukey test carried out with 60KCl and 60FMX. The biotite remineralizer increased soybean yield up to the dose of 160 kg ha -1 of K 2 O.
Soil K extracted through Mehlich-1 cultivated with soybean (Table 7) presented significant differences in comparison to the treatments. K contents ranged from 26 to 98 mg dm -3 in the LV soil. Soil K content increase was observed up to dose 140 kg ha -1 of biotite remineralizer K 2 O, and it points out K release to the soil/plant system.
With respect to reference K source (FMX) at the same dose (60 kg ha -1 of K 2 O), they did not show significant differences in soil K content and these results are evidence of biotite's potential. FMX showed the highest values and the standards by Souza and Lobato (2004) showed that soil K contents were lower than the proper limits (higher than 50 mg dm -3 in LV) in all treatments. Ribeiro et al. (2010) reported the positive effects of using Alkaline ultramafic rocks and pyroclastic breccias due to the high concentrations of exchangeable K + in the soil after the administration of high dosages.
involved in breathing and photosynthesis (Taiz et al., 2004). However, K is not part of any organic compound that plays structural functions in plants (Faquin, 2005).
As for K reference source (FMX) at the same dose (60 kg ha -1 of K 2 O), they did not present significant differences in soil K content, and this finding shows biotite's potential. KCl, FMX and the dose of 120 kg ha -1 of biotite K 2 O recorded the highest values, but there were no differences between the applied biotite doses. Based on the standards by Souza and Lobato (2004), soil k contents produced by treatments 60 KCl, 60 FMX and the dose of 120 kg ha -1 biotite K 2 O reached the proper levels (higher than 50 mg/dm 3 in LV). Ribeiro et al. (2010) reported the positive effect of using high concentrations of exchangeable K + in the soil after the administration of high doses due to the use of Alkaline ultramafic rocks and pyroclastic breccias.
The highest levels of leaf K were found with 60 FMX and 60 KCl. There were no differences in biotite doses and in reference standards FMX and KCl. Leaf K contents in soybean were close to those found in assays that were lower than the levels referred as proper by Raij et al. (1997) (1.7 to 2.5 dag kg -1 ), Ribeiro et al. (1999) (1.7 dag/kg) and EMBRAPA (2020) (1.8 to 2.5 dag kg -1 ). It is important highlighting that these interpretation criteria are set for soybean in the field and that there is variation in contents and interpretations depending on several factors, on cultivation conditions between them and on cultivars (Fontes, 2016). Results recorded for relative soybean yield efficiency in Red Latosol, of biotite remineralizer potential (%) with remineralizer FMX and potassium chloride (%)-at standard equivalent dose-presented relative efficiency of 94.18% FMX and 97.31% KCl, in comparison to the biotite's relative efficiency (Figure 1). Based on results in the present study, biotite's remineralizer potential allows the release of nutrients that reflect on the expression of soybean culture yield relative efficiency in Red Latosol.

Sandy Loam Texture Yellow Latosol (LA)
Soybean yield (Table 8) presented significant differences in the F test carried out in LA (11.21) at variation coefficient of 19.18%. Yield ranging from 683.1 to 2,615.3 kg ha -1 were recorded, with emphasis on treatments 60 kg/ha of Biotite, which recorded the highest yield rates. According to CONAB (2022), yield reached 124,047.8 tons; this value is 10.2% lower than that recorded for the 2002/21 crop season; mean reached yield was 3,029 kg/ha. There were differences in Tukey test between treatments, except for all in comparison to the witness (Table 8).
Biotite undergoes chemical and physical changes after it is added to the soil; these changes influence its ability to provide nutrients, mainly K. Soils present responses to soybean yield similar to FMX and KCl (at higher biotite doses) due to their particular features and properties-they are more yield responsive at the first cultivation after application.
If one takes into account that most potassium silicate minerals have low solubility in water and gradually release nutrients, we can assume that solubilization in the first cultivation year did not happen at the time needed for yield responses in the culture to be equal to those of soluble sources. On the other hand, biotite presented behavior similar to that of FMX.
The remineralizer biotite undergoes chemical and physical changes after it is added to the soil; these changes influence its ability to provide nutrients, mainly K. Soils present responses to soybean yield similar to those of FMX and KCl (at higher doses) due to their particular features and properties-they are more responsive to yield. Soil K extracted through Mehlich-1 cultivated with soybean (Table 8) presented significant differences in treatments K contents ranged from 18 to 33.5 mg dm -3 in LA soil. Treatment 60 KCl recorded the highest contents and 30 kg ha -1 of biotite K 2 O presented the lowest contents. The other treatments did not present differences from one another.
As for reference K source (FMX and KCl), at the same dose (60 kg ha -1 of K 2 O); they did not present significant differences in soil K content and this finding shows biotite's registering potential. KCl presented the lowest values due to its solubility in water. Based on the standards by Souza and Lobato (2004), soil K contents are at adequate levels (higher than 40 mg dm -3 no LA). Ribeiro et al. (2010) reported the positive effect of using Alkaline ultramafic rocks and pyroclastic breccias due to the high concentration of exchangeable K + in the soil after the administration of higher doses. Theodoro et al. (2013) carried out an experiment with five rocks-among them there were some basic rocks-in five cultures (maize, beans, garlic, okra and carrots); they showed that the availability of nutrients in all fractions of Latosols evidenced the interactions among agro-minerals, soil and plants. Reis (2013) tested a treatment in Latosol with micaschist and amphibolite rocks, in millet culture, and reported that these agro-minerals act as the source of these nutrients to plants. They also observed increase in root dry mass (RDM). Duarte et al. (2012) observed that the highest dry matter contents met the highest doses administered to the soil due to the application with silicate rocks in millet culture.
Leaf K contents in soybean (Table 8) showed significant differences in treatments in the F test (11.29) and variation coefficient of 15.58% in LA soil. The highest leaf K contents were found in treatments 60 KCl and 60FMX. There were differences in biotite doses in comparison to the witness. As for FMX, there were no differences between biotite doses and the witness.
Leaf K contents in soybean were close to the adequate levels referred by Raij et al. (1997) (1.7 to 2.5 dag kg -1 ), Ribeiro et al. (1999) (1.7 dag kg -1 ) and EMBRAPA (2020) (1.8 to 2.5 dag kg -1 ). Treatment 60 KCl was the exception, since the other ones recorded number lower than the adequate one. It is important highlighting that these interpretation criteria are set for soybean in the field and that there is variation in contents and in interpretations, depending on several factors, among them one finds cultivation conditions and cultivars (Fontes, 2016). Leaf K relative contents in comparison to the standards (KCl and FMX), adjusted itself in 2 nd degree polynomial regression (Figure 2), and in growing biotite doses.  Note. BE Biotite remineralizer from Embu Mineral Company; FMX: fine-graded mica schist remineralizer from Pedreira Araguaia Company; KCl: commercial potassium chloride.

Clayey Red Latosol (LV)
Maize silage yield (Table 9) in clayey LV soil did not show significant differences in the F test (11.76) and variation coefficient of 38.32%. Maize silage yield ranged from 3.9 to 38.5 tons per hectare. According to CONAB (2022), mean maize grain yield in Brazil was 4,366 kg ha -1 in the 2020/2021 crop season. Maize crop season in Brazil is basically divided in first and second season, and the yield of maize grown in the first crop season was 5,687 kg ha -1 in 2020/2021; it was higher than the second crop season, which reached 4,050 kg ha -1 (CONAB, 2022). It is worth recalling that cultivation in the first crop season is performed in the best sowing times, and it meets the meteorological parameters and physiological aspects inherent to maize plants; these parameters optimize these plants' yield capacity. Treatments 120 kg ha -1 of biotite K 2 O presented yield equivalent to that of tested standards (KCl and FMX), as observed in Table 9. It is essential that doses of 120 BE of remineralizer by EMBU (Table 9) did not present significant differences in the Tukey test applied to standards FMX and KCl. The increasing biotite doses led to increased maize silage yield until dose 150 kg ha -1 de K 2 O.
Soil K extracted through Mehlich-1 and cultivated with silage maize (Table 9) did not present significant differences among treatments (F tests of 2.05 and VC = 55.46). K contents ranged from 22 to 72 mg dm -3 in LV soil. Increased soil K contents were significant up to dose 240 kg/ha of biotite K 2 O; these results indicate the release of K into the soil/plant system. Biotite doses did not show significant differences in the Tukey test applied to 60 KCl and 60 FMX (Table 9). The biotite doses led to increase in soil K contents until the dose of 240 kg ha -1 of K 2 O. With respect to the reference K source (FMX) at the same dose (60 kg ha -1 of K 2 O), it did not present significant differences in soil K content, a fact that showed biotite's potential. Based on standards by Souza and Lobato (2004); soil K contents of 240 BE and 60 FMX met the adequate levels (higher than 50 mg dm -3 in LV). Ribeiro et al. (2010) reported the positive effect of using Alkaline ultramafic rocks and pyroclastic breccias, at high exchangeable K + concentrations on soils after the administration of high doses. Theodoro et al. (2013) carried out an experiment with five rocks, among them, basic rocks, in five cultures, namely: maize, beans, garlic, okra and carrots; they found that nutrients' availability in Latosol, in all fractions, has shown interaction among agro-minerals, soil and plant. Reis (2013) tested a treatment in Latosol with micaschist and amphibolite rocks, in maize culture, and reported that these agro-minerals act as source of these nutrients to plants; they increased root dry mass (RDM). Duarte et al. (2012) observed that the highest dry mass contents were proportional to the highest doses of silicate rocks applied in the soil.
As for leaf K content in silage maize (Table), it was possible observing significant differences in treatments subjected to F test of 4.64 and variation coefficient of 21.64% in LV soil. Leak K contents in silage maize were below the levels referred to as adequate by Raij et al. (1997) (1.7 to 5.5 dag kg -1 ) and Ribeiro et al. (1999)  It is important to point out that these interpretation criteria were set for maize, in general, in the field; furthermore, there is a whole variety of contents and interpretations that depend on several factors, among them one finds cultivation conditions and cultivars (Fontes, 2016). If one takes into account the criteria by Sousa & Lobato (2004), treatments 30 BE, 60 BE, 120 BE, 240 BE, 60 FMX and 60 KCl are at adequate level, despite the low K contents in the soil. These outcomes may have resulted from leaf sampling, which was carried out at flowering, and from soil collection, which took place at the end of the maize cultivation time. It is also essential highlighting that the soil presented very low K contents before the experiment was installed.  Silage maize yield (Table 10) did not present significant differences in the F test in LA (1.77) and variation coefficient 58.8%. Yield ranging from 10.5 to 26.1 tons per hectare were recorded, with emphasis on treatments 240 kg/ha of Biotite and KCl, which accounted for yield higher than 20 t ha -1 . Mean yield was close to the means recorded for Goiás State, and for Brazil, based on CONAB (2022). The growing biotite doses led to soybean yield increase until the dose of 240 kg ha -1 of K 2 O.
Biotite undergoes chemical and physical changes after it is added to the soil and these changes influence its nutrient-availability capacity, mainly K availability. Soils presented soybean yield response similar to that of FMX and KCl at higher biotite doses due to their particular characteristics and properties -they were more responsive to yield.
Assumingly, most potassium silicate minerals present low solubility in water and gradually release nutrients. Hence, it is possible saying that solubilization, at the first cultivation year, did not happen at the time necessary to achieve yield response in this culture, similar to that of other soluble sources. On the other hand, biotite presented behavior similar to that of FMX, and it is a quite positive and promising outcome in the remineralizer category.
Soil K extracted through Mehlich-1 cultivated with silage maize (Table 10) did not present significant differences among treatments (F tests of 1.05 and variation coefficient of 40.33%). K contents ranged from 29 to 50 mg dm -3 in LA soil. All soil K contents were within the class lower than adequate, based on interpretation criteria by CFSG (1988), except for treatment 60 KCl.
With respect to reference K source (FMX) at the same dose (60 kg ha -1 of K 2 O), they did not present significant differences in soil K contents; this finding points towards biotite's registration potential in the remineralizer category. Based on standards by Souza and Lobato (2004), soil K contents are at levels lower than the adequate one (higher than 40 mg dm -3 in LA), except for 60 FMX and 60 KCl. Ribeiro et al. (2010) reported the positive effect of using Alkaline ultramafic rocks and pyroclastic breccias with high exchangeable K + concentrations on soil after the administration of high doses. Theodoro et al. (2013) carried out an experiment with five ricks, among them basic rocks, in five cultures, namely: maize, beans, garlic, okra and carrots, and found nutrient availability in Latosol fractions; this outcome highlights interaction among agro-minerals, soil and plants. Reis (2013) tested a treatment in Latosol with micaschist and amphibolite rocks, in millet culture, and reported that these agro-minerals act as source of these nutrients to plants; furthermore, they have increased root dry mass (RDM).
Leaf K content in silage maize (Table 10) showed significant differences in treatments in the F test of 5.93 and variation coefficient of 24.86% in LA soil. The highest leaf K content was recorded for treatment 60 KCl. There were no differences in levels referred to as adequate by Raij et al. (1997) (1.7 to 5.5 dag kg -1 ), Ribeiro et al. (1999) (1.75 to 22.5 dag kg -1 ). However, based on the criteria by Sousa and Lobato (2004) (1.3 to 3.0 dag kg -1 ), all treatments were at adequate levels, except for 30 BE. The culture's nutritional management can be improved by the application of leaf analysis as operational tool (Brockley, 2001). The growing biotite doses led to increase in leaf K content until the dose of 180 kg ha -1 of K 2 O.
According to criteria by Sousa and Lobato (2004), treatments 60 BE, 120 BE, 240 BE and 60 FMX were at adequate levels, despite the low soil K contents. These outcomes may have resulted from leaf sampling, which was carried out at flowering; moreover, soil samples were collected at the end of maize cultivation time. It is important highlighting that the behavior of these biotite doses in comparison to the remineralizer registered in MAPA (FMX). Its performance was similar. Note. Means followed by equal letters in the rows did not differ in the Tukey test, at 5% probability level. ns , * and **: not significant, at 5% and 1% significance levels in the F test, respectively. BE: Biotite remineralizer from Embu Mineral Company; FMX: fine-graded mica schist remineralizer from Pedreira Araguaia Mineral Company; KCl: commercial potassium chloride.

Conclusions
Yield data point towards K release in the soil, and it is absorbed by plants during the tests, a factor that reflects on yield increase due to biotite application. Product biotite presented behavior in the soil similar to that of FMX and, in some cases, to that of KCl. Biotite has the potential to act as Potassium source in soybean and maize crops.