ORGANOMINERAL FERTILIZERS FROM OILSEED PIES

Objective: The present study aimed to characterize oilseed cakes and phosphate sources and select promising formulations in terms of phosphorus supply. Theoretical framework: The availability of phosphates with sufficient quality for industrial processing has a reduced longevity, while materials with low phosphorus content are little used. The association between organic materials and marginal phosphorus sources has been reported as promising in the use of these materials. Methodology : Samples of castor bean (TM), crambe (TC), sunflower (TG) and jatropha (TPM), simple superphosphate (SS) and reactive natural phosphate (FNR) cakes were obtained and characterized. 48 formulations were produced, derived from a mixture of cakes, SS and FNR, in addition to isolated forms, with a nitrogen:phosphorus (N:P) ratio of 1:1 to 1:4, totaling 61 treatments. In a randomized block design, they were incubated in two soils with contrasting textures, and samples were collected at 15, 45, 75 and 105 days to determine the phosphorus content with a sodium bicarbonate solution (pH 8.5). Equations were adjusted for each formulation evaluated, and groupings were carried out to select promising mixtures. Results and conclusion: Although the cakes did not contain significant amounts, mainly of phosphorus and potassium, no toxic elements were observed, a pattern different from that observed for FNR, which contains considerable levels of cadmium. The N:P ratio 1:4 proved to be superior, providing a higher P content in the soil. Increases in P were observed in quantity and linearly throughout the evaluated period depending on the proposed mixtures in comparison to the control treatment and the application of isolated sources. Research implications: The association of nutrients from organic and mineral sources has shown promise in supplying the nutrient phosphorus, minimizing the effects of weathered soils on nutrient management. Originality/value: The study is relevant as it seeks solutions to the problem of nutritional management of tropical soils, aggravated by the high external dependence on nutrients.


INTRODUCTION
Agricultural systems need to maintain adequate levels of productivity and depend on soil fertility.In these systems, nutrient replacement via the addition of fertilizers occurs at an average rate of 43% of the nutrients required by the crops (Fresco, 2003;Reetz, 2017).
Phosphorus (P), one of the nutrients most needed by crops, is associated with energy transfer, enzyme composition, nucleic acids and phospholipids, and its deficiency contributes to a lower development of plants (Taiz et al., 2017).
The interaction of P with the soil colloids, especially the more weathered ones, which have predominance of Fe and Al oxides in their composition (Oliveira et al., 2020;Santos et al., 2020, Hou et al., 2020), and have high capacity for specific adsorption of P, reduce the availability of the nutrient.Phosphate fertilizing for these soils requires technologies that can circumvent this problem, as phosphate fertilizers are mostly imported products, which increase production costs and increase external resource dependency.

THEORETICAL GROUNDS
Strategies for improving the efficiency of phosphate fertilization include the use of organic additives that protect the phosphate ion from interactions with the soil.The organic material favors the production of hydroxylates, which compete in a direct manner with the sites of adsorption, besides promoting an increase in the pH of the medium, diminishing the interactions of P with the oxides.With the release of compounds with high affinity to Al and Fe, such as SO4 2-and F -, releasing the phosphate ions adsorbed to the soil solution (Iyamuremye,1996; Novais & Smyth, 1999).Depending on the pH of the medium, most functional groups present in lignocellulosic materials can dissociate and interact with metal cations, as in this state they are negatively charged.Amino groups (-NH2) present in proteins, present positive charges when protonated, interacting with anions (Ribeiro, 2012).
The increase in the demand for phosphates has given an impulse to the agro-mineral industry to develop new technologies to increase the productivity and quality of fertilizers, besides encouraging methods that make it possible to make use of minerals of lower quality, which represents an increase in the economic use and in the longevity of the reserves.This incentive remains in place in the National Fertilizer Plan 2022-2050 that provides for the search for alternative sources in the field of plant nutrition, with citation of organomineral fertilizers, as well as associated technologies (Araújo et al., 2020a;Caligaris et al., 2022).
Organomineral fertilizers are defined as "the product resulting from the physical mixing or combination of mineral and organic fertilizers", the set of norms relating to the production of this type of fertilizer being regulated by Decree 4.954 of January 14, 2004(Brazil, 2004) and by Normative Instruction SDA/MAPA 25 of July 28, 2009(Brazil, 2009), which deals with the general aspects on the theme.
The target of numerous research aimed at producing more efficient and less costly fertilizers (Magalhães et al., 2017), organomineral fertilizers are associated with the proper disposal of agro-industrial waste (Antille et al., 2017;Ferreira et al., 2015;Magalhães., 2017).
They can be cited as beneficial effects of organomineral fertilizers, the increase in the cation exchange capacity (CTC) of the soil, the availability of nutrients, improvement in soil aeration, increase water retention capacity, promote improvements in soil aggregation, favor microbial activity, act in the complexation of heavy metals and provide lower pH variations (Cardoso et al., 2017), by the addition of the organic component, since these additions act as precursors of humic acids (Xie et al., 2012).
Materials composed of cellulose, hemicellulose and lignin, contain carboxyl, hydroxyl, phosphate, ether and amino groups, to which the superabsorbent characteristic is attributed (Ibrahim et al., 2010).This capacity is associated with water retention and slow release of nutrients as observed by Xie et al. (2012), in fertilizers formulated with wheat straw.
There is a range of materials from which it is possible to produce organo-mineral fertilizers, such as waste from the bioenergy industry, such as waste oils from the oil extraction process, which are produced on a large scale, and have considerable nutrient content and are attractive for use as organo-mineral fertilizers.
Depending on the type of extractor used, the P values obtained may not reflect on the content of this element in the soil.The quantification of the P content of natural phosphates (FN) can be underestimated when using Bray I or Olsen extractors or overestimated with the use of Bray II or Mehlich I extractors, depending on the type of extractor used (Zapata & Roy, 2004).
Studies related to the efficiency in the use of fertilizers are supported by instability in the international scenario, such as the recent conflicts in Eastern Europe, a region of high importance for the global supply of raw materials used in the production of fertilizers (Osaki, 2022), an event that demonstrated Brazilian vulnerability in the theme (Caligaris et al., 2022), collaborating so that technological innovations can reduce, at least partially, the dependence on these inputs.
Brazil is one of the largest consumers of NPK fertilizers on the world stage, since it is also one of the main suppliers of food (CONAB, 2023;Rodrigues et al., 2015), and faces variations in fertilizer prices due to international fluctuations, mainly oil price and supply of raw materials (Colle & Alvim, 2016).
From the above, the present study aims to select among the proposed formulations, the promising ones in the supply of phosphorus, as well as the viability of granulation of the formulations obtained from oilseeds pies (castor, crambe, sunflower and jatropha), simple superphosphate and reactive natural phosphate as sources of phosphorus.

Chemical characterization of oilseed pies and phosphorus sources
For the study, pies of oilseeds (castor bean, crambe, sunflower and jatropha plant) and simple superphosphate and reactive natural phosphate (Figure ).In these materials, a sample was taken for the quantification of the total element contents according to Usepa 3050B.Determination of P was performed by colorimetry, K by flame emission photometry, and the others by atomic absorption spectrophotometry (Ca, Mg, Fe, Cu, Mn, Ni, Zn, Pb and Cd).N has been determined by combustion (Dumas method) as Table .

Fertilizer formulation
The castor bean, crambe, sunflower and jatropha pies, as well as the phosphorus sources, were processed so that they could be reached at the appropriate granulometry for the formulation, through grinding and sieving in a 35 mesh mesh.
From the proportions of nitrogen and phosphorus (N:P) pre-established at 1:1, 1:2, 1:3 and 1:4 (N:P), and from the chemical characterization of the pies and phosphorus sources, the amounts of the components resulting in the formulations presented in Table 1, which are summarized in three main groups, namely: pies x reactive natural phosphate (FNR), pies x simple superphosphate (SS) and pies x FNR x SS.

Soil fertilizer assessment
For this study, a random block design was established, considering 61 treatments derived from four oilseed pies, two sources of phosphorus and four N:P proportions, applied alone and associated, plus a control (Table 1).The study was conducted in a greenhouse.
The tests were mounted with soil samples collected from the surface layer (0-20 cm) of two distinct soils Planossolo (sandy texture in the surface layer) and Argissolo, which had a clay texture.Samples were characterized for their granulometry and fertility attributes (Donagemma et al., 2011) (Table 2).After characterization, the material was dried and sieved and packaged in plastic containers with a capacity of 450 ml.Fertilization with incorporation into the soil, equivalent to 40 kg of N ha -1 , was simulated, with phosphorus doses corresponding to one to four times the N dose as proposed (Table 1).
Soil humidity was maintained at about 70% of field capacity, with irrigation interval of 3 to 5 days according to the evaluation of the experimental units by weighing.During the evaluation period, data were collected on temperature and relative humidity, from the home environment of vegetation Figure 2. The average temperature and humidity in the period were 25.9 °C and 74.5%, respectively.
In the experimental units, the samples were collected at 15, 45, 75, and 105 days after the application of the treatments, with the help of a plastic sampler introduced to the soil in order to collect a representative sample, covering the whole depth of the vessel.One sampler was used for each experimental unit to avoid contamination.
After each collection, the samples were dried and stored for determination of the labile phosphorus contents, using the extraction proposed by Olsen et al. (1954) according to de Sims (2009).The determination was performed by spectrophotometry (Donagemma et al., 2011).
The analysis consisted of the extraction with sodium bicarbonate solution (NaHCO3) of 0,5 mol L -1 at pH 8,5.Samples of 2 g of TFSA were weighed in Falcon ® -type tubes, and then 30 mL of extractor solution (NaHCO3 0.5 mol L-1 pH 8.5) was added, proceeding with slight manual agitation for 5 seconds and then remaining at rest for 16 hours.An aliquot of 5 mL of the supernatant was removed and the concentration of phosphorus determined by spectrophotometry according to Donagemma et al. (2011).

Statistical analyzes of data
The data were evaluated for homogeneity and normality, followed by regression analysis, and grouping analysis to select promising formulations according to the highest availability of P measured, using R software (R Core Team, 2022).

Chemical assessment of oilseeds pies and phosphorus sources
In Table the total content of the elements quantified in single superphosphate (SS), reactive natural phosphate (FNR), castor bean (TM) pies, sunflower (TG), crambe (TC) and jatropha (TPM) is presented.The percentage of phosphorus quantified in the NRF (26.1 ± 6.12%), considering the margin of error, is in line with the product label (29% of P2O5 total).
Considering the primary macronutrients present in the pies, for TM the highest concentration of nitrogen (N) and potassium (K) was observed, and for CT the highest concentration of phosphorus (P).Considering micronutrients, it is worth highlighting that the highest concentration of manganese (Mn) was verified in the castor bean pie.
The elements cadmium (Cd) in reactive natural phosphate (FNR) and lead (Pb) in simple superphosphate are observed in higher concentrations in the selected source group, however, concentrations do not exceed the limit established by SDA Normative Instruction No. 27, June 5, 2006(Brazil, 2006).This characterization is relevant, as fertilizers are a potential pathway for introducing trace elements into the environment (Valle, 2012).

Selection of organomineral fertilizers with oilseed pies and phosphorus sources
After incubation of the soil with the treatments presented in Table 1in general, lower levels of phosphorus (P) available in the clay-textured samples (Figure ) to the detriment of those with a sandy texture (Figure ) due to the increased adsorption capacity of that soil.
For the clay-textured samples, the pies formulations associated with reactive natural phosphate (NRF) and simple superphosphate (SS) provided higher levels of P available, however, some mixtures, such as crambe pie (CT) with NRF ( For samples with a clay texture (Figure ), it is observed that the formulations without simple superphosphate (SS), presented linear or decreasing pattern in the first evaluations, contrasting with the formulations with SS, in which the increase in phosphorus content was observed in the same period.
The application of pies in an isolated manner provided the increase of the P only after 45 days of evaluation.The FNR associated with the pies, especially TM and TG, brought about an increase in the levels of phosphorus in the soil from the 50 days of evaluation, being the highest value in the 75 days, while the association with TC and TPM brought forward the release of P to 15 days after the start of measurements, being more responsive to the N:P ratio.
As for the release of P in the sandy texture samples, a linear pattern was observed throughout the experimental period for formulations containing pies associated with FNR, which can be observed by analyzing the Figure (DEFG), as well as in treatments where pies have been applied in isolation (Figure (C).
The pies + SS formulations provided increased P release during the trial period up to 45 days and pies + SS + FNR showed increasing P concentration in soil up to 75 days.After the maximum value for the two groups considered, there was a reduction in the concentration of P in the soil.Increases in the levels of available P are most evident in formulations containing pies + SS + FNR (Figure 4LMNO).
The application of SS in isolation (Figure 4B) provided a significant increase in the availability of P at 15 days, however a dramatic reduction is observed at 45 days.
For the highest dose of SS, a maximum value of P was observed in the sandy texture samples compared to clay.This difference can be explained by the higher buffering capacity provided by materials with higher clay contents.The same was observed in the general comparison, where the levels of P quantified in the sandy texture samples were higher than those of clay texture.
The formulations evaluated for the P available at different times in the contrasting samples were organized into three groups, namely: formulations containing NRF (Figure 1 13 formulations containing SS (Figure 2); and formulations containing FNR + SS (Figure 3).Clusters were built for each group (keeping the treatment under control and the pies), from which it was possible to establish a hierarchy of the formulations as to the supply of P in the soil.

DISCUSSION
The values of macronutrients (NPK) usually found in pies mentioned in the literature have in their composition values of N ranging from 3 to 6%, P on average 1% and K from 1 to 1.5% (Pereira et al., 2015;Potes et al., 2012;Souza et al., 2009), values in agreement with the materials evaluated (Table ).Although the nutrient content of the pies is compatible with the literature, supplementation with nutrients from other sources without maintaining as an alternative to increase their usability, for example, composing organomineral fertilizers, which must present minimum values of nutrients.
The relative efficiency of urea for castor beans, turnips and jatropha, considering the supply of N, was reported in the studies by Carvalho & Costa (2009), which found values of 46%, 57% and 53% respectively in clay textured soil, and 80%, 86% and 89% in sandy textured soil, in experiments carried out with corn (Zea mays L.) in pot experiments.Through the analysis of the results, the authors demonstrate how the texture of the soil can influence the efficiency of these products.
For the formulations evaluated in this study, considering P as the supplied element, a similar behavior to that cited by Carvalho & Costa (2009) was found, in which the highest concentrations of P were quantified in the sandy texture samples, indicating greater comparative efficiency between soils.This pattern is due to the differences in interaction that P with soil materials of different granulometries.Soils with a high clay content have affinity for the phosphate ion and have specific interaction sites that may render the PO4 -4 anion unavailable (Wiel et al., 2016).
The elements cadmium (Cd) in reactive natural phosphate (FNR) and lead (Pb) in single superphosphate as Table do not exceed the limit established by SDA Normative Instruction No. 27, June 05, 2006(Brazil, 2006), that for the case of fertilizer exclusively supplying phosphorus, the maximum value of Cd and Pb (in mg kg -1 ) are limited to 4 and 20 times per guaranteed percentage point of P2O59 respectively.The presence of trace elements in the pies depends on the conditions of the soil where they were grown.Traces of Cd were detected in the pies used, however, the highest concentration is 30 times lower than the limit defined by SDA Normative Instruction No. 7, April 2016(Brazil, 2016), considering organic fertilizers and soil conditioners.
Phosphatic fertilizers naturally show great variability in their composition due to the diversity of the minerals from which they are obtained.Trace elements are the undesirable components in the composition of fertilizers, which, although in low concentrations, generate concern since they are toxic.
Care regarding the content of trace elements in fertilizers and improvers is important as they consist of a pathway for the introduction of these elements into the environment (Valle, 2012), in addition to mining, industrial activity and waste disposal.These elements can be harmful to organisms if present in high concentrations in the soil (Antoniadis et al., 2019).
The accumulation of toxic elements in plant biomass occurs as described by Brannvall et al. (2015), which report high soil concentrations of P (506 mg kg -1), As (2.7 mg kg-1), Cd (0.8 mg kg-1) and Pb (12.1 mg kg-1) as a result of the addition of 2% of sewage treatment plant biosolid, with high accumulation of As (4.3 mg kgyyyyyyyyy-14) and Cd (0.3 mg) -1) by a mixture of species used in revegetation of soils with low natural fertility.Antille et al. (2017) observed no changes in trace element concentrations after organomineral application.
The behavior of P after incubation in soil as observed in Figure and Figure demonstrates the interactions between different formulations and with contrasting soils.The maintenance of higher P content in the soil for longer is due to the organic component in the organomineral formulation, by shielding common adsorption sites in weathered soils (Meurer, 2010), and by adding exchange sites in these soils (Liu, 2009).
The differences in P available over time reflect the behavior of the organic component in the organomineral formulation.For Pies + SS there was an increase in the levels of P quantified up to 45 days because the source of P is of high solubility, while for Pie + SS + FNR based formulations increasing concentrations of P up to 75 days due to the lower solubility of natural phosphate were observed.
The buffering effect of the organic component can be better visualized in the isolated application of SS in the application on the clay texture samples (Figure A) sandy (Figure 4B), where mainly for the sandy texture, there was an increase at 15 days with subsequent sharp reduction, which shows the absence of the effect of the presence of the organic material.This behavior of P is reinforced by Araújo et al. (2020b) who observed greater accumulation of biomass and nutrients in millet grown with organomineral fertilizer in sandy soil.Cardoso et al. (2017) found a 22% increase in potato production with the use of organomineral fertilizer compared to mineral in the rainy season.The authors suggest that the organic component may have led to the gradual release of the nutrients, favoring the best results.These results collaborate with that observed from the inoculation of organomineral formulations in soil, where the organic component has delayed the maximum release of P in the soil.
Assessing the availability of phosphorus in the soil as a result of the application of increasing doses of this nutrient via mineral and organomineral fertilizer, Souza et al. (2013), observed after incubation for 60 days, higher levels of the nutrient in the treatments that received organomineral fertilizer, using as extractor, exchanger resin and Melich-1, which contradicts the finding of mineral fertilizers provide greater release of phosphorus, compared to organomineral formulations, after incubation for 60 days in sandy soil using extractor Melich-1 and Melich-3 (Silva et al,2).010).
Applying 50% of the dose of organomineral fertilizers Vieira et al. (2020) saw an increase of at least 11% in the production of fresh and dry matter compared to mineral fertilization.The same authors observed an increase in residual phosphorus in the soil for the highest dose of organomineral fertilizer applied in two carried out in Oxisol (Latossolo) and in the second cycle in Entisol (Neosolo).

CONCLUSIONS
Formulations containing combination components provided higher phosphorus levels when compared to the application of the isolated forms.
Reactive natural phosphate contains cadmium content (41.07 mg kg -1 ), which suggests critical assessment of the applied dose.
The tarts do not contain restrictive toxic elements and contain considerable amounts of nitrogen, and can be used in the composition of fertilizers.

Figure 1 -
Figure 1-Samples of castor, crambe, sunflower and jatropha pies in crude form (A, B, C and D), and after grinding (E, F, G and H).Source: Prepared by the authors.
Figure E), jatropha pie (TPM) with FNR (Figure G), TM + SS (Figure H), CT with SS (Figure I), TG + SS (Figure J) and TPM with SS (Figure (K) except in the initial evaluation periods, provided higher values of P extracted with bicarbonate (P-bic).
Figure HIJK) favored higher P levels in soil compared to the single application (Figure (B).Functions adjusted for the content of each formulation evaluated at times 15, 45, 75 and 105 days in clay texture samples are presented in the Table .

Figure 1 .
Figure 1.Grouping for formulations containing reactive natural phosphate (RNF) as a function of the quantified phosphorus content in the soil.Source: Prepared by the authors.

Figure 2 .
Figure 2. Grouping for formulations containing single superphosphate (SS) as a function of the phosphorus content quantified in the soil.Source: Prepared by the authors.

Figure 3 .
Figure 3. Grouping for formulations containing reactive natural phosphate (NRF) and simple superphosphate (SS) as a function of the quantified phosphorus content in soil.Source: Prepared by the authors.

Table 1Identification
and composition of formulations (treatments) used in the selection of promising mixtures for phosphorus supply.Source: Prepared by the authors.

Table 2 -
Chemical attributes of the soils employed in the conduct of the assessment.Source: Prepared by the authors.

Table 4 -
Models adjusted for bicarbonate extractable phosphorus (mg dm -3 ) assessed as a function of incubation time of treatments in clay textured soil (argisol).

Table 3 .
Models adjusted for bicarbonate extractable phosphorus (mg dm -3 ) assessed against the incubation time of treatments in sandy soil (Planosols).