Molecular Environmental Sciences
Dalton Abdala's webpage
Figure 1. P K-edge XANES spectra for an adjacent forest soil and soils treated with manures with different application history and under different land management systems in Paraná State, Brazil.
Long-Term Manure Application Effects on Phosphorus Speciation, Kinetics and Distribution in Highly Weathered Agricultural Soils
ABDALA, Dalton Belchior, SILVA, Ivo Ribeiro, VERGÃœTZ, Leonardus, SPARKS, Donald Lewis
Abstract
Phosphorus (P) K-edge XANES and Fe K-edge EXAFS spectroscopies, along with sequential P chemical fractionation and desorption kinetics experiments, were employed to provide micro- and macro-scale information on the long-term fate of manure application on the reactivity and distribution of P in highly weathered agricultural soils of southern Brazil. Soil test P values ranged from 7.3 up to 16.5 times as much higher than the reference soil. A sharp increase in amorphous Fe and Al amounts were observed as an effect of the consecutive application of manures. Whereas our results showed that the P sorption capacity of some manured soils was not significantly affected, P risk assessment indices indicated that P losses should be expected, likely due to the excessive manure rates applied to the soils. The much higher contents of amorphous Fe and Al (hydr)oxides (55 % and 80 % increase with respect to the reference soil, respectively) in manured soils seem to have counterbalanced the inhibiting effect of soil organic matter on P sorption by creating additional P sorption sites. Accordingly, the newly created P sorbing surfaces were important to prevent an even larger P loss potential. Phosphorus K-edge XANES lent further evidence for the loss of crystallinity and transformation of originally present Fe-P minerals into poorly crystalline ones as an effect of manuring, whereas Fe K-edge EXAFS provided insights into the structural changes underwent in the soils upon manure application and soil management.
Figure 2. P K pre-edge XANES spectra of an adjacent forest soil showing a more intense pre-edge crest, diagnostic for a more crystalline Fe-P minerals, and the lack of pre-edge, diagnostic for amorphous Fe (hydr)oxide minerals, in manure treated soils.
Phosphorus K-edge XANES analysis
Phosphorus K-edge spectra for orthophosphate (Pinorg) reacted with goethite at pH 4.5 or 6.5 in the presence or absence of Calcium is provided in the supporting information session (Figure 5) to illustrate the features seen on P K-edge XANES spectra as an effect of pH and Ca. The P K-edge XANES spectra of all phosphate minerals in this study had absorption edge energies around 2151.5 ± 0.1 eV. Except for the differences in pre- and post-edges between the reference adjacent forest soil and manured soils, no other remarkable features were observed.
The differences in pre- and post-edges of the P K-edge XANES analysis clearly showed the effect that long-term manure application had on XANES K-edge spectra of manured soils as they resembled manure-like XANES K-edge spectra (Figure 1) found elsewhere in the literature. A pre-edge crest around 2148 eV (Figure 2) is characteristic of crystalline Fe phosphates (Hesterberg et al., 1999; Khare et al., 2005; Ingall et al., 2011) and it is more prominent in crystalline than in non-crystalline minerals (Hesterberg et al., 1999; Kruse & Leinweber, 2008). Likewise, the reference soil from Paraná state showed a small pre-edge, indicating that crystalline Fe-containing minerals are originally found in those soils and that the repeated application of manure led to the formation of non-crystalline Fe minerals, which can be denoted by the resulting weak pre-edge in the XANES spectra of manured soils. As for the post-edge, the adjacent forest soil displayed a shoulder at around 2,158 eV, which is also characteristic for crystalline Fe-phosphates (Ingall et al., 2011). This characteristics was completely absent in treated soils, which otherwise, showed a post-edge crest at around 2,167 eV, denoting the effect of manuring on the formation of organo-mineral phases between Fe-bearing minerals and P in organic compounds, most closely resembling a phytic acid spectrum (Sato et al., 2005), and devoid of any features that could indicate the presence of crystalline phosphates (see Ingall et al. (2011) for a comprehensive library of P K-edge XANES mineral standards).
Whereas our chemical fractionation results raise some interrogation as to whether Al-bound P minerals would be another important mineral phase present along with non-crystalline Fe minerals, the validity of this hypothesis would only be at the expense of a virtual absence of pre-edge. It is important to mention, however, that it does not completely rule out the existence of Al-bound P species in the soils, as XANES is a measure of the average environment of all species present in a sample.
Furthermore, organic matter coupled to Fe(III) reduction is a thermodynamically favorable process that may lead to the formation of amorphous Fe(III)/(II)-containing minerals (Anderson, 2004), thus supporting our guesstimate that Fe-bound mineral phases are likely the resulting mineral being formed upon long-term application of manure. In view of the similarities between our work and the research carried out by Sato et al. (2005) (manure types and application history), comparisons will be made when appropriate. Conversely to what was observed by Sato et al. (2005), Ca-P minerals were not observed in our study.
Calcium phosphates usually display a diagnostic shoulder on the high-energy side of the absorption edge (Hesterberg et al., 1999; Kruse & Leinweber; Ingall et al., 2011) and the prominence of this feature seems to be a measure of the Ca abundance in the mineral (Ingall et al., 2011). This feature was virtually absent in all spectra. It is interesting because the manure application history in the work by Sato et al. (2005) is comparable to ours, although the rates may vary. Even though the rates of animal manures applied in our soils were moderate, some of the soils have received manure application for several decades and the accumulated amounts of manures should taken into account. Manures were applied at rates varying from 5 to 10 ton ha-1.
The calcium content in these materials usually ranges from 3 to 4 %, which is equivalent to an application of 150 to 400 kg ha-1 year-1. It is evident that, in the work by Sato et al. (2005), a combination of higher soil pH and soil Ca concentrations and much lower extractable Fe contents provided the ideal conditions for the formation of Ca-P secondary minerals, whereas in our study, the relatively lower pH together with the much higher extractable Fe contents favored the formation of Fe-P secondary minerals.
Figure 3. Fe K-edge EXAFS spectra of a crystalline Fe oxide (hematite), an adjacent forest soil and manured soils. Gray labeled bars (I, II and III) highlight spectral features present in the reference mineral (hematite) and changes in the polyhedral linkages of the soils as they change as an effect of soil management and manure application.
Iron K-edge EXAFS analysis
The effects of manure application and management in the soils affected mineralogy at some extent, as evidenced by Fe K-edge EXAFS analysis (Figure 3), which showed that although minerals resembling hematite characteristics were the predominant minerals in all soils, manured soils presented some traits that were not observed in the reference adjacent soil. It is important to note, though, that using crystalline reference minerals for linear combination as well as for fingerprinting analysis is usually inadequate for describing multi-component mixtures, such as natural soil minerals, as the latter are often poorly-crystalline, non-stoichiometric and variable in structure and composition (O’Day et al., 2004).
Although mindful of the flaws of using a crystalline mineral as our reference compound, using hematite to serve this purpose was nevertheless the most appropriate venue to draw the due comparisons. In doing so, and whether intrinsic or extrinsic to the conditions to which the experiment was conducted, there are at least three aspects to which the changes in soil mineralogy must be accounted. The consecutive application of animal manures over the years seems to have been an effective intervention that led to the appearance of the distinct features observed throughout the soils. Yet, the presence of some characteristics originally not seen in the adjacent forest soil can be understood as a consequence of dehydration upon warmer temperatures to which the agricultural soils are subjected year-round. Additionally, soil inversion performed under conventional agriculture management and occasionally in the Tifton pasture land increases the rates at which soil moisture is lost and organic matter is decomposed, creating, therefore, the conditions for the rearrangement of the hematite package into some of its variations.
Figure 3 shows the Fe K-edge EXAFS spectra of a crystalline Fe oxide (hematite), an adjacent forest soil and manured soils. The gray bars labeled I, II and III highlight spectral features present in the reference mineral, hematite, and changes in the polyhedral linkages of the soils as they change as an effect of soil management and manure application. In region I, manured soils show similar features at low k values, ranging between 5.4 and 5.9, and a lack of lower-end features in the adjacent forest soil is observed. As for region II, manured and adjacent forest soils all showed similar features at same k values around 8.0 and 8.2 not seen in hematite. Interestingly, the feature positioned at the high-end of this region seems to gradually increase in intensity from adjacent forest all the way up to no-tillage.
Even though one can identify the features at high k values, shown in region III, it is evident that they display a rather smoothening, particularly in the forest soil. The lack of information in the literature on structural changes that hematite may undergo mediated by its reaction with organic matter via EXAFS spectroscopy represents a limitation for drawing a more accurate conclusion as to which EXAFS structures and specific linkages present in the polyhedral structure of hematite are affected by reaction with organic ligands. Therefore, it is necessary to perform additional studies in order to identify accurate Fe K-edge EXAFS features of hematite and related structural changes mediated by its reaction with chelating agents present in soil environments.
