3.2.1. Leachate pH values
The pH values of the leachates were monitored to relate its variations with possible reactions caused by the dissolution of fertilizers into the soil solution (
Figure 1). The changes in pH values when used deionized water may have been influenced by the characteristic of factors such as soil pH, microorganisms’ activity etc., rather than by the dissolution of the applied P sources. The control treatment pH was close to 5.0 at the beginning, increasing gradually until day 8 (6.8), and presented variations until the end of the incubation between 6.1 and 7.4. This same trend was also seen for all treatments, although with different amplitudes (
Figure 1A).
The pH variation observed initially in the treatment with S+MAP and MAP may have occurred due to the solubilization of MAP (NH
4H
2PO
4) and the formation of ammonium bicarbonate, a salt with high pH in solution [
42]. However, it was expected that pH would decrease over time because of the dissolution of ammoniacal phosphate fertilizers, what would lead to the nitrification process of NH
4+ that generates free acidity [
43]. However, it is worth noting that the columns remained in water holding field capacity for most of the time, which may have disfavored nitrification because of the lack of oxygen. Nevertheless, the average pH value of MAP (6.4) and S+MAP (6.6) were the only ones lower than the control (6.8).
For the other treatments the average pH stabilized after 7-8 days just above the neutral value (SSC = 7.4; SCP = 7.3; S+ASD and ASD = 7.0). The SSC was the treatment with the greatest pH range (4.9 - 8.0), however a neutral/slightly alkaline pH prevailed for most of the observed time. Reports of pH increase in acidic soils due to the application of SSC are common in literature, staying in the range from neutral to slightly alkaline, due to the ammonification process that consumes H
+ and generates OH
-, however, subsequent ammonia nitrification is common, reducing it to nitrate and releasing H
+, thus acidifying the soil soon after [
44]. As can be observed, the pH values of the leachate did not decrease even at the end of the trial, which may be related to the low mineralization of the organic compost in the analyzed period, on the other hand the constant leaching may have caused the removal of basic cations present in the compost, which resulted in a higher pH value of the leachate.
Under 2% citric acid leaching test there was smaller variations in pH values of the leachates over the time. This is because of the buffering capacity promoted by the percolated solution itself in the soil. The exception occurred in the first three days of leaching, where leachates from the SSC and ASD treatments were higher in the beginning, being 3.9 and 4.4 respectively, while the other treatments presented pH close to the control, 2.8 (
Figure 2B). This may be due to the rapid dissolution of these sources in the citric acid solution, immediately increasing its pH. The pH (CaCl
2) of the SSC = 6.4 (
Table S1) is a result of the composting of SS with plant material, which starts with a pH close to neutral and decreases, due to the production of organic acids and CO
2, until stabilizing due to the buffering action of humic substances formed [
45,
46]. The ASD, on the other hand, is a product of the mixture of phosphate with alkaline compounds (CaNaPO
4) with solubilization favored by the acidification of the medium [
22]. However, from the sixth day and so on these two treatments stabilized the pH to values similar to other treatments (~2.5).
It is also noted that there were peaks in pH values of the SSC (3.1) and S+ASD (3.0) leachates between the 8th and the 13th day, in contrast to what occurred with the raw material of these powdered and non-pelletized sources, SSC and ASD. This observed trend reinforces the idea that the elements present in the pellet are protected and therefore solubilization is slower [
16].
3.2.2. Fertilizer P solubilization dynamics
There was greater solubilization of P from recycled fertilizers when leached with 2% citric acid than with water (
Figure 2), which was expected since most P in these sources is poorly water-soluble and almost totally soluble in citric acid (
Table 1). When leached by deionized water treatments with MAP presented high P solubility, with similarity behavior between MAP and S+MAP, since both solubilized more than 50% of the P in 5 days (
Figure 2A). The total leached P (
Figure 2C) was highest in MAP (84 mg column
-1), followed by the organomineral S+MAP (73 mg column
-1), while the other sources solubilized little during the incubation period, not exceeding 20% of the applied dose (100 mg column
-1). This shows that the mixture of the low water-solubility SSC with a high soluble inorganic fertilizer resulted in a product that is able to release P into the soil within a short time after application.
Since MAP is an acidulated phosphate fertilizer with high water-solubility, it is rapidly available in the soil solution [
3], the expected benefit of mixing such a fertilizer with a less soluble organic one is to better synchronize the availability of P with crop demand. The P from the mineral fertilizer is immediately available in the solution, supplying the demand in the early growth stages of the crop, while the P from the organic source is solubilized slowly, as the organic compounds are mineralized, meeting plants needs in later stages and potentially even being available for subsequent crops [
16].
Most of the recycled sources evaluated here presented very small solubility in water (
Figure 2B) but were highly soluble in citric acid (
Figure 2D), where more than 85% of the applied P was leached, with the exception of the organomineral S+ASD that solubilized only 42% of the applied P. If only the water-solubility of recycled products are considered for their use as fertilizers, the agronomic potential of most would be underestimated. For example, Raniro et al. [
47] observed that the poorly water-soluble ASD promoted dry biomass production of sugarcane equivalent to triple superphosphate, a high water-soluble P source. For this reason, considering the solubility of fertilizers in citric acid (organic acid synthesized and exuded by plants roots) may decrease the chance of misinterpretations related to low water-solubility of P and availability of P to crops. Furthermore, to ensure a complete picture of the agronomic potential of alternative sources, this type of study should be related to plant experiments.
Although the total P leached in 2% citric acid was not significantly different between the recycled sources and MAP, with the exception of S+ASD, the P solubilization rate was distinct. SCP and ASD showed less pronounced curves of solubilized P, indicating slower and more gradual dissolution. SSC, S+MAP and MAP had most of their P solubilized in the first days, while S+ASD solubilized P only from the 14th day and forward.
The S+ASD treatment solubilized less P than its individual components (SCP and ASD), indicating that the pelletized mixture left the product more recalcitrant and impaired the release of P from the pellet. It was also found that the physical form modified the dissolution dynamics of the fertilizer in soil, as SSC (powder) compared to its pelletized form (SCP) solubilized more P in water and had faster initial solubilization in citric acid, which is expected due to the pellet having less specific surface area, which reduces the reactivity of the material and causes slower nutrient release [
48].
Soluble phosphate fertilizers rapidly release P into the soil solution, which can be taken up by plants, shift to less labile forms through specific and nonspecific adsorption reactions with the surface of minerals or precipitate with cations ([
49]. In weathered soils, where Fe and Al oxides are predominant, sorption reactions are more rapid and intense, becoming practically irreversible as P remains in the soil [
7,
50]. Therefore, part of the P applied via fertilizer can be immobilized in the soil, decreasing the efficiency of fertilization by reducing P bioavailability in later stages of crop development.
In order to meet the P demand of the plant during its cycle, slow release fertilizers are alternatives for supplying P in the long term, capable of making P available in a more synchronized manner with plant development, and thus avoiding environmental losses. In the initial phases, the plant does not yet have a well-developed root system, so if most of the P applied is made available immediately, little will be absorbed and most of it may be lost by surface runoff, leaching and/or erosion processes, or else reduce its lability in the soil and bioavailability at later plants stages[
11,
33].
3.2.3. Soil lability
Most of the remaining P in the soil from the water leached columns was found in the labile and moderately labile pools in the 0-1 cm layer (
Figure 3). This is indicative that these fertilizers solubilized P very slowly or even not solubilized, making the nutrient being accumulated at the application site. The slow release of P from fertilizers is one way to avoid losses to the environment (which decreases fertilizer efficiency) either by immobilization in the soil through sorption, making P less available to plants [
43], or by leaching [
51].
The ASD and S+ASD were the sources that increased total soil surface P the most: 9,778 and 1,479 mg kg
-1 respectively, with expressive amount in the labile fraction extracted with bicarbonate (
Table S4). The labile P pool is the one in osmotic equilibrium with the soil solution, that is, it is more easily replaced either by mineralization of organic P or desorption of inorganic P [
5,
47], what demonstrates that these fertilizers are capable of increasing available soil P as they solubilize over several days after application. Despite the high soil P concentration with S+ASD, only 18.3% of the applied P was recovered, indicating that most of it remained in the pellet (
Table 4).
Most of the soil P with ASD was found in the moderately labile pool (7.774,7 mg kg
-1) extracted by HCl (
Table S2), usually associated with desorption of Ca-bound P forms [
52]. The formation of P-Ca under ASD may have been favoured due to the high concentration of basic cations in this source. At high P and Ca concentrations, it may precipitate forming poorly water-soluble minerals, which can solubilize in acidic conditions such as that of the rhizosphere [
53]. Furthermore, high moderately labile P values in the 0-1 cm layer under ASD could have partially come from fertilizer that was not completely dissolved, being sampled along with the soil, as supported by Nanzer et al. [
54], who reported that the P in different thermochemically treated ashes were mostly in the forms of chloroapatite or hydroxyapatite, and 90% could be extracted by HCl.
The moderately labile pool plays an important role as a source of P to the soil, as it keeps the solution P levels in equilibrium when the supplement of available P via fertilizer is insufficient to sustain plant growth. In fact, all less labile forms of P can function as a buffer in the soil when the P exported by crops exceeds the P content in solution. Intermediate labile P fractions can switch to more labile forms in order to maintain equilibrium, however it may not solubilize as quickly as required by some crops [
5,
51,
53].
Under low available P conditions, plants may adopt some strategies to access more stable forms in the soil, such as increasing root system, association with microorganisms, release of phosphatase enzymes, and increasing the exudation of organic acids in rhizosphere [
55]. Almeida et al. [
56] found that tropical grass species such as Urochloa ruziziensis release more organic acids (citric, malic and lactic) in the rhizosphere under deficiency conditions than in P sufficiency. For this reason, plants that exude more organic acids are more likely to benefit from phosphate fertilization from slow-release sources. This was evident in the research of Talboys et al. [
11], where the species that exudes more organic acids (Fagopyrum esculentum), was able to absorb more P than the species that exudes less (Triticum aestivum).
As for the recovery of P applied via fertilizers, in the treatments with S+MAP and MAP the amount of P remaining in the soil smaller than the others, because much of the solubilized P was leached out of the soil columns carried by water percolation. However, even so, the amount of P remaining in the soil was similar to SSC and SCP sources. In the latter two there was almost no dissolution of the fertilizers, as can be seen when calculated the P balance (
Table 4), where only 29% (SSC) and 20% (SCP) of the applied P was recovered in the leachate or in the soil. The non-recovered part was supposedly kept in fertilizer pellets yet. Although much of the P in SSC is inorganic in nature, it does not mean that they are in readily available forms [
9]. According to O'Connor et al. [
57], high concentrations of Fe and Al (10-30 g kg-1) in biosolids can reduce P bioavailability, as may have occurred to SSC and SCP, and their low P solubilization is consequence of the high Fe concentration (13 g kg-1) found (
Table S1).
Regarding to the balance of P under 2% citric acid, most of the applied P leached out for all P sources (>85%), so there was minimal P remaining from fertilizer (<15%) in the soil (
Table 4). Overall, the distribution of the soil P pools did not show large variations. The difference was small between the sources in the moderately labile and non-labile pools, but in the labile pool the source ASD promoted more P in the 6-10 cm layer (
Figure 3), indicating that there was movement of P in the soil profile when in contact with citric acid. The presence of P far from the application site indicates that there was P displacement in the soil, but perhaps with longer duration of the experiment, this labile P would move deeply in the soil profile until it completely leaves the column system.
The fact that the sources SCC, SCP, ASD and S+ASD solubilized P in citric acid, even though they are slightly soluble in water, indicates that they are capable of supplying P to plants in such environments with low P availability. Plants use some mechanisms to mobilize and absorb the recalcitrant P, by increasing the rooting system, excrete acid phosphatases, associate with mycorrhizal fungi and exudate organic acids, including citric acid [
58]. For these reasons, in addition to fertilizer solubility, plant and soil types are important factors to consider in managing soil nutrients with recycled sources. Based on the results presented, it can be inferred that sources that are poorly soluble in water, such as SCC, SCP, ASD and C+ASD, would be more suitable for the cultivation of crops with a longer cycle, with the solubilization of ASD and S+ASD being favoured in acidic environments, while SSC and SSP will depend on OM mineralization. On the contrary, OMF formed by S+MAP, which mixes sources of high and low solubility of P, has the potential to serve short-cycle crops, as it supplies P immediately and also gradually in the soil. Although these inferences can be made only after knowing the solubilization dynamics of the sources, their potential as a fertilizer must be more investigated in studies when plants are actively growing and extracting nutrients from soil solution.