Kamafugite From the Alto Paranaíba Province as a Multi-nutrient Fertilizer

The study evaluated the potential agricultural use of kamafugite as a multi-nutrient fertilizer in the cultivation of Urochloa brizantha. Two experiments were carried out under greenhouse conditions, one using a Ferralsol and the other an Arenosol. The experimental design was a randomized block with four replicates, in a factorial scheme 4 × 2 + 2, comprising four doses of kamafugite (1667, 3334, 6668, and 13336 mg dm), two product forms (filler and powder), and two additional treatments, TSF + KCl and a control treatment without K2O and P2O5 application. The sources were incubated for 60 days on two soil types and, after the incubation period, Urochloa brizantha plants were grown and two consecutive cuts were performed. The kamafugite was efficient in increasing plant dry mass and the levels of P, K, Ca, and Mg accumulated in plants, mainly in the 2nd Urochloa brizantha crop. The granulometry filler was more efficient in the sandy soil and the form powder in the clayey soil. Kamafugite can be used as an alternative source of nutrients for the Urochloa brizantha crop, being a low-cost and a potential source to improve tropical soil fertility over time.


Introduction
Brazilian soils, especially in the Cerrado biome, have low fertility and availability of macronutrients such as phosphorus (P), potassium (K), calcium (Ca), and magnesium (Mg) due to climatic conditions prevailing in the country that resulted in weathering. The soil properties, which predominate the high acidity (pH < 4-5), low cation exchange capacity, low base saturation, high levels of Al 3+ and ions of Fe and Mn are obstacles to agricultural practices. Only after practices like liming and application of mineral fertilizers, which can supply the amount of macro and micronutrients, these soils become suitable for continuous planting of grains, oilseeds, and other crops (Prado, Benites, Polidoro, Gonçalves, & Naumov, 2012).
Among the nutrients, P is one of the major constraints of agricultural production in tropical soils since it presents a complex dynamic in these soils. Under acidic conditions, Fe and Al oxides are preferably positively charged, being able to retain phosphate ions on their surface. Therefore, more phosphate fertilizers than those required by plants are used in order to saturate the components responsible for nutrient fixation. In this context, more than 5 million tons of phosphate fertilizers were delivered to the final consumer in Brazil in 2017 (ANDA, 2017). As observed for P, soil K levels must also be supplied with the use of fertilizers. From the total potassic fertilizers used in Brazil, 95% is imported in the form of potassium chloride (KCl) (IPNI, 2016), representing a big impact on the country's trade balance.
The search for low-cost sources of nutrients associated to the scarcity of fertilizers applicable to agroecological farming systems and the increase dependence on fertilizer imports make the use of agrominerals in agriculture an important alternative to Brazilian soil fertilization. Thus, some specific rocks are nutrient low-release sources, which reduce losses by leaching and fixation, as well as have a long-term action. Kamafugitic rocks are examples that have significant amounts of nutrients in their composition and may increase soil fertility in the medium to long-term, according to their solubility and soil reaction.
Kamafugites constitute a group of rare rocks found in a few locations in the world, e.g., Uganda, Italy, Brazil, and China. According to Junqueira-Brod et al. (2002), these rocks occur in Brazil as extensive spills and deposits, forming two of the largest kamafugitic provinces of the planet, the Igneous Province of Alto Paranaíba (PIAP)-Mata da Corda Group and the Alkaline Province of Goiás (PAGO). Nascimento et al. (2008) affirm that these rocks are potential sources of macronutrients (P, K, Ca and Mg) and can be considered as "multi-nutrient" sources.
In this sense, the use of alternative sources of nutrients, such as kamafugite, is a practice with potential to promote benefits in areas cultivated with Urochloa brizantha cv. Piatã (Syn. Brachiaria brizantha). In general, approximately 80% of the Brazilian tropical pastures are in some degree of degradation, so, the proper management of the fertility of soils cultivated under pasture is a fundamental practice to recover degraded areas. In addition, livestock farming is developed with low investments and the use of agrominerals can be a viable, economical and ecological alternative, given its availability, low production cost and long-term nutrient release.
In this context, the objective of this study was to evaluate the potential of agricultural use of kamafugite, in filler and powder forms, as a source of P, K, Ca, and Mg for Urochloa brizantha cv. Piatã (Syn. Brachiaria brizantha).

Product Characterization
The kamafugitic rock is originated in the region of Alto Paranaíba, county of Lagoa Formosa, Minas Gerais, Brazil (18°53′11″ S; 46°19′30″ W). These rocks are classified as mafic to ultramafic, with ultrapotassic affinity (K 2 O/Na 2 O molar ratio > 3), whose occurrence is commonly associated with the continental rifts. Such condition allows the crystallization of K-rich minerals such as phlogopite, leucite, and kalsilite (Le Maitre, 2002;Brod et al., 2000;Sahama, 1974). Due to the phosphate mineralization in practically the entire region of occurrence of the material rock, 1.5 to 2.5% of P 2 O 5 are found, with occasional occurrences that can reach contents of up to 25% P 2 O 5 .
The kamafugite used in the present study is an alkaline volcanic rock, containing leucite, kalsilite, pyroxene, olivine, perovskite, and rarely phlogopite. The highest levels of phosphate occur associated with facies with swarms of white venules of calcite and apatite that give a breached aspect to the rock. The powder and filler kamafugite forms were obtained by crushing and milling processes. Thus, the product in filler form is obtained by passing 100% of particles in 0.3 mm sieve, containing 4.0 and 1.2% of total and available K 2 O, respectively, as well as 3.1% of total P 2 O 5 , 1.94% of available P 2 O 5 , 1.4% of CaO, and 0.7% of MgO. The product was also applied as a powder, obtained by passing 100% of particles in 2.0 mm sieve, presenting 4.0% of total K 2 O, 1.2% of available K 2 O, 3.3% of total P 2 O 5 , 1.97% of available P 2 O 5 , 1.7% of CaO, and 0.5% of MgO.
Both product forms were characterized according to macronutrient contents, determined on a flame photometer according to the method described by MAPA (2017). In addition to kamafugite, triple superphosphate (TSF) and potassium chloride (KCl), containing 45% of total P 2 O 5 and 60% of total K 2 O, respectively, were used as standard sources. The chemical characteristics of nutrient sources used in this study were determined according to methodology described in Silva (2009).

Experiment Conduction
Two experiments were conducted in the period of 12/22/2017 to 10/19/2018, under greenhouse conditions, at the Institute of Agricultural Science at Federal University of Uberlândia, Umuarama Campus, Minas Gerais, Brazil, at the geographical coordinates 18°52′40″ S; 48°15′20″ W. Each experiment was conducted using samples of two soil types collected at 0-20 cm depth in the state of Minas Gerais. These soils were classified according to World Reference Base (WRB) for Soil Resources (FAO, 2015) as Ferralsol (clayey soil) and as Arenosol (sandy soil) with 84% and 14% of clay, respectively. The clayey soil presented pH equal to 4.5 (1:2.5 soil sample/water ratio), 0.2 mg dm -3 of P, 19.6 mg dm -3 of K (both P and K were measured by Mehlich-1 method), 0.2 cmol c dm -3 of Ca 2+ , and 0.1 cmol c dm -3 of Mg 2+ (Ca 2+ and Mg 2+ were measured by AAS on 1.0 mol L -1 KCl extracts added with lanthanum oxide). The sandy soil presented pH equal to 4.4, 3.1 mg dm -3 of P, 11.7 mg dm -3 of K, 0.2 cmol c dm -3 of Ca 2+ , and 0.1 cmol c dm -3 of Mg 2+ , measured using the same methods as used for clayey soil.
The experiment followed a completely randomized design with four replicates, comprising 96 total plots in factorial scheme 4 × 2 + 2. Four doses of kamafugite (1667, 3334, 6668, and 13336 mg dm -3 ) were applied resulting in four P 2 O 5 doses (63, 126, 252, and 504 mg dm -3 of P 2 O 5 ) and four K 2 O doses (20, 40, 80, and 160 mg dm -3 of K 2 O) in two product forms (filler and powder), calculated as a function of the total K 2 O and P 2 O 5 contents present in the kamafugite. In addition, a treatment including TSF + KCl (252 mg dm -3 of P 2 O 5 and 80 mg dm -3 of K 2 O) was used as standard treatment, as well as a control treatment without P 2 O 5 and K 2 O application. In order to maintain humidity, around 80% of the field capacity value for each soil, deionized water was added to pots. Soil moisture was rigorously controlled by daily weighing of the plastic containers, replacing the volume lost through evapotranspiration with deionized water.
The treatments were incubated for 60 days in 5 dm -3 of the two soil types (clayey and sandy soils). The soil moisture was controlled by daily weighing of the plastic containers, replacing the volume lost through evapotranspiration with deionized water to maintain humidity around 80% of the field capacity value for each soil. All treatments received equal doses of CaCO 3 and MgCO 3 to correct the soil acidity and increase the bases saturation to 70%. After the incubation period, 200 mg kg -1 of N were added to samples through the ammonium sulfate ((NH 4 ) 2 SO 4 ). Micronutrients were also supplied adding 1.5, 5.0, 0.5, and 0.05 mg dm -3 of Cu, Zn, B, and Mo through CuSO 4 ·5H 2 O, ZnSO 4 ·7H 2 O, H 3 BO 3 and (NH 4 ) 6 Mo 7 O 24 ·4H 2 O, respectively. Ten seeds of Urochloa brizantha cv. Piatã (Syn. Brachiaria brizantha) were sown per pot, at a depth of 2 cm. After the emergence of the seedlings, thinning was carried out, maintaining six plants per pot.
At 15 and 30 DAS (days after sowing), ammonium sulfate was applied providing 50 mg dm -3 of N. Daily irrigation was done as well as periodic observations to evaluate the possible incidence of pests and plant diseases in Urochloa brizantha cultivation.

Plant and Soil Analysis
Two Urochloa brizantha plant cuts were performed approximately at 4 cm above ground level at 40 and 80 DAS. After the second plant cut, soil samples were also collected. The plant samples were dried in an oven at 65 ºC until reaching constant weight and weighed to obtain dry matter of the aerial part (DM). Afterwards, the samples were ground and submitted to nitric-perchloric digestion (Sarruge & Haag, 1974). After that, the nutrients concentrations in the plant were analysed according to Silva (2009). Soil chemical analyses were performed according to Teixeira, Donagemma, Fontana, and Teixeira (2017) (soil available P, K, Ca and Mg).
The contents of nutrients in the aerial part were converted to accumulated values using DM values, obtaining values in mg pot -1 . Using the nutrient accumulated values, agronomic efficiency index (AEI) of the kamafugite relative to TSF + KCl was calculated using the equation proposed by Fageria, Santos, and Moraes (2010): (1) The kamafugite DM refers to the plant DM with the application of the kamafugite (80 mg dm -3 of K 2 O and 252 mg dm -3 of P 2 O 5 ), TSF + KCl DM refers to the plant DM with the application of TSF + KCl (252 mg dm -3 of P 2 O 5 + 80 mg dm -3 of K 2 O) and the control DM refers to the plant DM in the additional control treatment without P 2 O 5 and K 2 O application.

Statistical Analysis
Results of each variable were submitted to variance analysis according to Tukey's test at 0.05 of significance, as well as polynomial regression analysis using the SISVAR statistical program (Ferreira, 2014). The treatments were compared to control treatment and to standard treatment (TSF + KCl) using the Dunnet's test at 0.05 of significance (Silva & Azevedo, 2009).

Results and Discussion
The effects of kamafugite on Urochloa brizantha plant DM after consecutive cuts, as compared with TSF + KCl, are shown in Figure 1. In general, the application of both kamafugite forms resulted in higher DM values than the control treatment, mainly after the application of the highest doses (Figure 1). At 40 DAS, the application of 6668 mg dm -3 of the product in the filler form in sandy soil provided the highest DM values, which were close to the DM values of the standard treatment TSF + KCl (Figure 1b). At 80 DAS in clayey soil, at the highest doses of kamafugite, the dry mass values were higher than the standard treatment TSF + KCl, regardless the product form (Figure 1c). After the second growth, these values ranged, on average for both kamafugite forms, from 10 to 18 g pot -1 in clayey soil and from 11 to 16 g pot -1 in sandy soil. This result demonstrates the good residual effect of the product when compared to the first cut, increasing the plant dry matter production over time.   Vol. 12, No. 10;2020 solubility in water and good solubility in weak acids present slower release of P to soil, decreasing their soil fixation potential. The use of slow-release nutrient source is especially important to grasses of the genus Urochloa, which can promote soil acidification and, consequently, a higher reactivity of natural fertilizer during its development (Dias et al., 2015).
In clayey soil, the application of kamafugite at doses of 6668 and 13336 mg dm -3 of P 2 O 5 resulted in soil P contents considered appropriated and high, respectively, according to Sousa and Lobato (2002) (Figure 3e). In sandy soil, the P content increased from very low to adequate and high following the application of increases doses of kamafugite in the powder and filler form, respectively (Sousa & Lobato, 2002) (Figure 3f). These results are especially important in tropical agroecosystems where Urochloa brizantha plants are largely grown, since the low soil P availability is a major constraint to agricultural production in acid soils of tropical and subtropical regions (Ramos, Faquin, Rodrigues, Silva, & Boldrin, 2009).
In clayey soil, both forms of kamafugite applied increased the accumulated K values in the aerial part of Urochloa brizantha plants harvested at 40 DAS in relation to the control, yet lower than the standard treatment TSF + KCl. Only on this soil, the product in the filler form was more efficient than the powder form (Figure 4a). At 40 DAS, as the dose of kamafugite applied to sandy soil through both product forms soil was increased, increase in accumulated K values were also obtained (Figures 4a and 4b). At 80 DAS, the application of the highest dose of kamafugite (13336 mg dm -3 ) increased plant accumulated K values in both soils (Figures 4C and 4d). At the highest K 2 O dose applied through kamafugite in clayey soil, accumulated K values were higher than the standard treatment TSF + KCl (Figure 4c). At the same dose (13336 mg dm -3 of kamafugite) in sandy soil, accumulated K values do not differ statistically from the standard treatment ( Figure 4d). Those results show the efficacy of kamafugite in providing K to plants when compared to more soluble sources, with higher nutrient concentration.
The application of soluble K-sources, as KCl, results in a low K retention in soil's cation-exchange capacity since the element presents a small hydrated ion load and size (Takahashi, Dahlgren, Kanno, Nanzyo, & Takahashi, 2018). The K present in the kamafugite is released to soil in a gradual form, avoiding leaching losses and, consequently, being better utilized by plants. Thus, the kamafugite can be used as a sustainable and low-cost source of K for long term crops, as this rock releases K 2 O slowly and reduces the demand for more expensive chemical fertilizers after consecutive growth cycles. Martins et al. (2014) also highlighted the potential of ultrapotassic rocks as alternative sources of K for agriculture in addition to conventional sources (KCl).
The observed increases in plant K contents following the application of the kamafugite are related to the chemical composition of this source, since it has 4 dag dm -3 of K 2 O. The kamafugite K-content can be attributed to the ultrapotassic nature of these rocks, which is related to the excess of K 2 O in relation to Na 2 O. Such condition allows the crystallization of potassium rich minerals such as phlogopite, leucite, and kalsilite.  Due to the high K mobility in soil, the exchangeable K content does not efficiently represent the availability of this nutrient in soils with different mineralogical, physical, and chemical characteristics (Meurer & Anghinoni, 1993). In addition, because they are rocks with a gradual release of nutrients, it is expected that with a longer incubation time than the one used in this study, greater K-release to soil would occur. Bakken, Gautneb, Sveistrup, and Myhr (2000), studying the use of biotite and carbonatite as sources of K over three years in pasture cultivation, observed that in the last year of the study, without the addition of any K fertilizer, the effects of carbonatite and biotite were comparable to the KCl. These authors concluded that increasing the evaluation period may result in gains comparable to those obtained with conventional sources of nutrients.
In general, the application of the highest kamafugite doses resulted in increases in the accumulated Ca and Mg values in Urochloa brizantha plants harvested at 40 DAS and 80 DAS, with better results observed in clayey soil ( Table 1). The finer grain size of the product (filler) showed to be more efficient in increasing Ca and Mg contents in plants than the powder form (Table 1).