Layered Double Hydroxides as Hosting Matrices for Storage and Slow Release of Phosphate Analyzed by Stirred-Flow Method

The objective of this work was to synthesize and characterize Layered Double Hydroxides (MgAlLDHs, MgFe-LDHs, and MgAlFe-LDHs) containing phosphate anions and study P release kinetics was conducted with the prepared materials. The release kinetics tests were performed in water by the stirred-flow method to verify the P release profile from LDHs materials. Results showed LDHs have similar characteristics to those of reactive sources of P, such as thermophosphates. The LDHs presented gradual P release and the LDH prepared with the molar ratio of cations Mg:Al = 2:1 showed a more sustained profile of P release compared to other produced materials. Thus, the chemical and structural characteristics of the LDHs make them potential materials for the use and storage of slow P release fertilizer.

Cardoso et al. 6 intercalated 2,4-D; MCPA, and Picloram molecules in release tests performed in leaching columns and bioassays.When LDHs were compared with pure herbicides, leaching test results showed that the release is slow when the molecules are intercalated.For the bioassays, results showed that herbicide molecules intercalated in LDH presented the same pre-emergence control efficiency on Lepidium sativum as the pure herbicide molecules.
Hafez et al. 7 intercalated gibberellic acid (GA) in MgAl-LDH (GA-LDH).The study aimed to verify the slow release of GA in different solutions (pH = 3 and pH = 7) and the degradation of the substance in two soils (sandy and medium texture).The GA-LDH results were compared to pure GA.The authors showed that GA-LDH is more effective than pure GA, since pure GA crystallizes rapidly.After 24 hours of reaction, only 4.5% and 10.4% of GA was released in solutions of pH = 3 and pH = 7.After this period, the hormone becomes insoluble and no longer reacts in the medium.After 24 hours of reaction, the GA-LDH released 80% of GA into the solution at pH = 3 and 60% at pH = 7.In addition, it maintained a slow release until 6 days after the start of the reaction.Soil degradation tests showed that the intercalation increased the GA lifetime in the soil compared to pure GA.The pure GA remained for 6 days in the medium texture soil and 10 days in the sandy soil.After the intercalation, the period of GA in the soil increased to 20 days for the medium texture soil and 28 days for the sandy texture soil.
Silva et al. 14 and Silva et al. 15 synthesized LDHs intercalated with NO 3 -anions, varying the molar ratios of Mg 2+ :Al 3+ and Mg 2+ :Fe 3+ .In both works, the authors concluded LDHs presented protection for intercalated NO 3  -anions with potential uses as matrices for slow release of fertilizers.
Phosphorus (P) is an essential element for food production.P is a limited resource that has been inappropriately exploited, causing high pressure on worldwide reservese 16,17 .In recent years, problems related to P have been addressed in several Materials Research works, which have proposed strategies for dealing with the consequences of the uncontrolled use of this element in agriculture [18][19][20][21][22][23] .P scarcity is one of the biggest challenges of global sustainability in the 21 st century, which directly affects food production and safety systems in the world 24 .Therefore, studies on improving the use of P have increased in recent years.
In many countries, mainly those in tropical regions, soils present low P availability because of the adsorption and fixation of this element by iron and aluminum oxides 25 .In these regions, high P content is usually applied for soil correction rather than plant nutrition [24][25][26] .
The reduced contact between the phosphate ions and the soil mineral phase reduces the adsorption reactions/fixation of P by the soil 27,28 .Thus, an alternative for improving P performance in cultures is the development of sources to provide a physical protection for phosphate ions, in order to slow the release of these ions to plants.This physical protection can be created by the intercalation of phosphate ions in LDHs [29][30][31][32][33] .
Because of the need to develop new technologies to increase the efficiency of phosphate fertilizers, the objectives of this work were as follows: (i) synthesis and characterization of MgAl-LDHs, MgFe-LDHs and MgAlFe-LDHs containing phosphate ions; and (ii) evaluation of P release kinetics in the prepared materials.These LDHs were prepared with different molar ratios of M 2+ and M 3+ cations.The produced materials were characterized by powder X-ray diffraction (PXRD), Fourier transform infrared spectroscopy with accessory of attenuated total reflectance (FTIR-ATR), Raman spectroscopy analyses and specific surface area by adsorption of N 2 (BET).The contents of C, N and H in the LDHs were determined by elementar analysis.For the prepared LDHs, P release kinetics were evaluated in an adaptation method stirred-flow.

Experimental Part
All reagents used in this work presented analytical degree of purity: HPO 4 , and NaOH.The water used in all reactions was distilled, deionized and decarbonated in a Milli-Q ® Integral Water Purification System.

Synthesis of Layered Double Hydroxides
Initially, five pristine LDHs were synthesized with molar ratios ranging between M 2+ and M 3+ cations.All pristine LDHs were intercalated with NO 3 -anions.These materials were named Mg 2 Al-LDH, Mg 3 Al-LDH, Mg 2 Fe-LDH, Mg 3 Fe-LDH, and Mg 2 Fe 0.5 Al 0.5 -LDH.The pristine LDHs were synthesized by the coprecipitation method at pH constant 4 .To prepare the material, 500 mL of a solution containing M 2+ and M 3+ cations in different molar ratios (Table 1) were slowly added drop by drop to 290 mL of a solution containing 1.48 mol L -1 of KNO 3 .During this time, the suspension formed was kept under agitation in an N 2 atmosphere.A solution containing 2.00 mol L -1 of NaOH was added simultaneously to maintain a constant pH (± 0.5).The cations, molar ratio, number of mols and anions used in the synthesis of the pristine LDHs are shown (Table 1).
After the synthesis, the suspension was hydrothermally treated at 65°C during 24 h.After the hydrothermal treatment, the precipitate was filtered and washed with H 2 O.The solid material obtained was freeze-dried for 120 h.After drying, the material was triturated and passed through an 80-mesh sieve to homogenize the particle sizes.
After synthesizing the pristine LDHs, the materials were subjected to an ion exchange reaction in solution for the incorporation of phosphate anions.The anion exchange in solution method was used for the incorporation of phosphate anions.In this method, 2.00 g of LDH precursor was suspended in 200 mL of a solution containing 1.00 mol L -1 of K 2 HPO 4 .The formed suspension was kept under agitation and atmosphere of N 2 for 24 h at 65 o C. At the end of the exchange reaction, the material was filtered again, washed and vacuum-dried in the presence of silica gel at room temperature.After the anion exchange reaction (Table 1), the names of pristine LDHs were changed to Mg 2 Al-LDH-P, Mg 3 Al-LDH-P, Mg 2 Fe-LDH-P, Mg 3 Fe-LDH-P, and Mg 2 Fe 0.5 Al 0.5 -LDH-P.

Characterization of the LDHs
The quantities of C, N, and H in LDHs were determined by elemental analysis.Chemical analyses were done to quantify the total levels of LDH constituent elements.Metals were determined by inductively coupled plasma atomic emission spectroscopy (ICP-OES).P quantities contained in LDHs and in the solution after the kinetics experiments were determined by molecular absorption spectrometry in the UV-Vis region, as described by Braga and Defelipo 34 .
In Brazil, phosphate fertilizers must contain minimum guarantees of P 2 O 5 quantities as determined in H 2 O, 2% citric acid and neutral ammonium citrate (NAC).To compare the various LDHs synthesized and the commercial sources of P, P 2 O 5 in H 2 O, 2% citric acid and NAC+H 2 O were analyzed in accordance with the methodology described in AOAC 35 .
The LDHs were characterized by powder X-ray diffraction (PXRD), Fourier transform infrared spectroscopy with accessory of attenuated total reflectance (FTIR-ATR), analyses of Raman spectroscopy and specific surface area through adsorption of gaseous N 2 (BET).
PXRD analyses were performed with a Shimadzu X-ray diffractometer, XRD 6000, with a copper cathode lamp (λ = 0.154 nm) and a graffiti monochromer to select the copper emission region, with potential source of 40 kV and current of 40 mA.The scanning rate of 0.01° every 10 s was used to determine the parameters of unitary cells of LDHs.The sweeping 2θ ranged from 4 to 70º.
FTIR-ATR analyses were conducted with a Jasco spectrophotometer, FTIR 4100.The spectra were obtained with 216 scans, in a wavenumber range from 4000 to 400 cm -1 and resolution of 4.0 cm -1 .
Raman measurements were performed in a Renishaw inVia confocal Raman microscope.The spectra were obtained through excitation with a near infrared laser line (λ = 785 nm) at 1.5 mW using a 50X standard objective lens.Fluorescence correction (background) was performed by removing the baseline using the baseline correction and smoothing filter functions in the software Origin 8.0.
The determination of the specific surface area (BET) was performed by the adsorption of N 2 gas in the sample using a Quantachrome analyzer, model Nova 2200e.The samples were degassed under vacuum at high temperature to remove H 2 O and other volatile compounds.After degassing, the sample was transferred to the analysis module, where N 2 gas was injected into the interior of the cell to provide adsorption measures of N 2 .

Kinetics of P Release
An adapted stirred-flow method, described by Strawn and Sparks 36 , was used for the kinetics of P release experiments.
The analyses system has a reaction chamber containing two filters.One of them has a 0.45 µm cellulose membrane and the other has Teflon with 25 mm pores.Both are used to retain the sample within the chamber.The chamber was placed on a magnetic stirrer (100 rpm) and coupled to a peristaltic pump, generating a flow of 1 mL min -1 .A fraction collector was coupled to the chamber outlet.Ultrapure H 2 O was used as extractor.Approximately 2.00 mg of LDH was put into the agitation chamber and the pumped extraction solution was collected in the outlet of the system every 2 min up to 30 min, after that, every 6 min up to 110 min, and then each 10 min up to 150 min after the beginning of the flow.The P extracted from the samples, which was in the fractions, was determined as described by Braga and Defelipo 34 .For this determination, 0.5 mL of the sample collected in the stirred-flow was pipetted into 25 mL plastic cups and added to 4.5 mL of deionized H 2 O. Immediately afterward, 5 mL of a working reagent composed of ammonium molibidate 0.01 mol L -1 , bismuth subcarbonate 0.01 mol L -1 and 1.6 g ascorbic acid were added.After 30 min, following the total blue color development, the samples were read on a UV-VIS spectrometer at a wavelength of 725 nm.The P contents P collected at different times in the stirred-flow assay were used to elaborate the mathematical model that best fitted the P release.The software CurveExpert version 1.4 was used to build the models.
The amounts of C, H, N and P for pristine LDHs and LDHs-P are shown in Table 2.The LDHs of the Mg-Al system presented higher amounts of anions, both NO 3 -and phosphate.It was also possible to verify, for the MgAl-LDHs and MgFe-LDHs systems, that the molar ratio M (2+) /M (3+) increased from 2:1 to 3:1, reducing the amount of intercalated NO 3 -and phosphate.This reduction is related with the charge density of the materials; the lower the molar ratio, the higher the load density of the layer, which increases the number of intercalated anions between inorganic layers 37 .
Results showed that the ionic exchange occurred in the solution and phosphate anions were able to replace part of intercalated NO 3 -anions.The P values obtained after the exchange reaction are close to those reported in the literature.Gillman et al. 38 used MgAl-LDH and different concentrations of phosphate in the exchange reactions and obtained materials with up to 62 mg g -1 of P. Triantafyllidis et al. 39 used MgAl, MgFe, and MgAlFe-LDHs and obtained materials with approximately 80 mg g -1 of P.
For Crepaldi and Valim 37 , one of the main hindrances to ionic exchange reaction in solution is the precursor anion content that remains after the reaction.In agriculture however, the residual N content in the synthesized LDHs after exchange reaction by phosphate is a desirable characteristic.Nitrogen is an essential element for plants; thus, LDH-P can be a fertilizer and a source of N and P for plants.
The contents of total P 2 O 5 , in H 2 O, 2% citric acid and NAC extractors are shown (Table 3).In relation to the P fraction extracted from different extractors (H 2 O, 2% citric acid and NAC), MgAl-LDHs presented the lowest extracted amount for all three extractors.The use of H 2 O as an extractor can determine only the most soluble forms of P. The 2% citric acid solution extracts low solubility forms, such as P-Ca, in addition to the soluble forms.The NAC solution is able to extract low solubility forms linked with Fe and Al 40 .
As expected, all LDHs presented no solubility in H 2 O and from average to high solubility in 2% citric acid and NAC.Such characteristics are very similar to reactive sources such as thermophosphate 41 .Phosphate sources of high H 2 O solubility, such as triple superphosphate, simple superphosphate and mono ammonium phosphate, suffer huge losses of P due to adsorption abilities of tropical soils.Sources with reactive characteristics are a good alternative under these conditions because of the slow-release effect of these materials 42 .It is worth highlighting however that source efficiency is affected by not only reactivity, but also by soil and plant factors 43 .This suggests that LDHs can become a potential source of slow-release P in agriculture, once reactive sources show good agronomic efficiency 44,45 .

Powder X-Ray Diffraction (PXRD)
The diffractograms of pristine LDHs are shown in Figure 1.These diffractograms show basal peaks (00l), (003) and ( 006 -in LDHs 46,47 .All synthesized materials presented good structural organization and phase purity.LDH-5, with three different cations in the layers, presented poor crystallinity compared to the others. For Mg 3 Al-LDH and Mg 3 Fe-LDH, the materials with the cation ratio of M 2+ :M 3+ = 3:1 showed more intense basal peaks (00l), indicating higher crystallinity.The lower proportion of trivalent cation in the layers increased the crystallinity of the materials.The lower the number of cations M 2+ replaced by M 3+ , the lower the distortion of the layers and consequently the better the arrangement of stacking for the structure 15,48 .-------------------------(mg/g) --------------------------   The basal spacing values found for this phase were 0.92 nm for Mg 2 Al-LDH-P and 0.81 nm for Mg 3 Al-LDH-P.The obtained results for C, N, and P quantifications are presented in Table 2.After the ionic exchange reaction, the LDHs presented significant amounts of N and C. As expected, the presence of N suggests the exchange reaction between nitrate by phosphate anions was incomplete.The presence of CO 3 2-is due to the contamination by atmospheric CO 2 .Even when all reactions (coprecipitation and exchange) are conducted in an inert environment (N 2 ), handling during washing, drying and   analyses exposes LDHs to the air.Due to the high basicity presented by these materials, contact with atmospheric CO 2 causes them to be absorbed and incorporated to LDH in the CO 3 2-form 49 .Thus, there may be co-intercalation of NO 3 -, phosphate, and CO 3 2-anions in LDHs subjected to anionic exchange reactions.
For Mg 2 Fe-LDH-P and Mg 3 Fe-LDH-P, the calculated basal spacing values were 0.79 nm and 0.78 nm.For these LDHs, the amount of P found in relation to nitrogen (Table 2) is significant, which suggests phosphate ions can also be intercalated.For the MgAl-HPO 4 system, Frost et al. 56 found basal spacing values between 0.78 nm and 1.19 nm.For the Mg-Fe system, Triantafyllidis et al. 39 found basal spacing values around 0.77 nm.

Fourier transform infrared spectroscopy with accessory of attenuated total reflectance (FTIR-ATR)
FTIR-ATR spectra for pristine LDHs are presented in Figure 3.All spectra have a wide band in the 3400 cm -1 region.The presence of these bands is attributed to -OH stretching of H 2 O molecules and to hydroxyl groups present in the layers.The spectrum also presents a strong band in the 1763 cm -1 region, referring to C=O bonds, which allows inferring on the presence of CO 3 2-in LDHs.The spectra for the pristine LDHs also show the presence of bands in the 1350 cm -1 region, attributed to vibration modes (ν3) of NO 3 -.In LDH, it is possible to find two types of nitrate, with bands in the 836 cm -1 , 1380 cm -1 and 1763 cm -1 regions that present "free" NO 3 -and in the 825 cm -1 , 840 cm -1 1050 cm -1 and 1425 cm -1 regions that represent "non-free" NO 3 -(48,57 .Figure 4 shows the spectra for LDHs after the exchange reaction.These spectra also present wide bands in the 3400 cm -1 region, characteristic of the presence of -OH groups.The presence of phosphate can be evidenced by the bands in the 1050 cm -1 , 870 cm -1 and 550 cm -1 regions, which correspond to ν3(PO 4 3-); ν1(PO 4 3-) and ν4(PO 4 3-) vibrations 58 .The band in the 1350 cm -1 region can be attributed to the vibration modes of NO 3 -or CO 3 2-.In the literature, several works have reported the overlap of CO 3 2-and NO 3 -bands in the region ranging from 1350 to 1400 cm (-1) in LDHs [59][60][61][62][63] .In the 550 cm -1 region, the bands refer to the signature octahedral structure of brucite-type layers, i.e.Al-O-Mg or Fe-O-Mg.The 710 cm -1 band is specifically attributed to NO 3 -ion, while in the 1050 cm -1 region, it corresponds to the anion intercalated (independent of the type of anion) in LDH like materials [64][65][66][67][68][69] .

Raman Spectroscopy
Figure 6 shows the Raman spectra for LDHs subjected to anionic exchange reaction.In these spectra, it is possible to see the permanence of bands in the 470 cm -1 and 550 cm -1 regions, corresponding to the lamellar structure of LDHs.Except for LDH-5P, the band is not present in the 710 cm -1 region, referring to NO 3 -.The presence of phosphate is evidenced by the emergence of bands in the region between 940 cm -1 and 960 cm -167 .Slight changes can also be observed in the positions of the bands of the materials subjected to the exchange reaction in relation to pristine LDHs, such as in bands located in the 1050 cm -1 region.These results are in line with those presented by Palmer et al. 70 , which show the form and position of Raman bands depending on the nature and concentration of the intercalated anions in LDHs.

Specific Surface Area (BET method)
Table 4 presents the measurements of specific surface area, pore volume and mean pore diameter for pristine LDHs and LDHs-P.The increased molar ratio between M 2+ /M 3+ cations present in the layers reduced the specific surface area, pore volume and mean pore diameter.For MgAl-LDHs, the reduction was 1.44 times the specific surface area and 1.50 times the pore volume.For MgFe-LDHs, both specific surface area and pore volume were reduced by 1.23.
The specific surface area values found in the literature for LDHs generally range from 50 to 80 m 2 g -1 .Some works however found values of 200 m² g -171,72 .The reduction of porosity and specific surface area with increased molar ratio M 2+ /M 3+ in LDHs was previously reported by Jaynes and Vance 73 , Di Cosimo et al. 74 , You et al. 75 , and Cantrell et al. 52 .The increased molar ratio reduces the charge density of the layers, and consequently, the specific surface area of these materials.For all LDHs, the sorption and desorption N 2 isotherms (presented in supplementary material) are type II, according to IUPAC (International Union of Pure Applied Chemistry), with close hysteresis, typical of mesoporous materials (pore size between 2-50 nm) 76 .7. For all LDHs studied, the curves that represent the profile of P release in solution are represented by two distinct and complementary moments: (i) fast release and (ii) slow release.The initial fast release is linked with phosphate adsorbed in the surface of LDHs crystallites, which are promptly exchanged.Then, the release slows because of the exchange reaction of the intercalated phosphate with the anions present on reaction medium.

Kinetics of P Release
The P profiles release and cumulative release rates for Mg 2 Al-LDH-P and Mg 3 Al-LDH-P are shown in Figure 7. Mg 2 Al-LDH-P showed better adjustment to the Hoerl model, as did Mg 3 Al-LDH-P to Vapor Pressure model.The adjustments were done using the equations obtained for each model.The turning point was also estimated, where the release rate of P becomes descendant and slower.In Mg 2 Al-LDH-P, this change occurs at 22 min, while for Mg 3 Al-LDH-P, this point is reached more rapidly, at 12 min.The cumulative rate of the two LDHs shows that Mg 2 Al-LDH-P releases P more slowly than Mg 3 Al-LDH-P.After 150 min of flow, Mg 2 Al-LDH-P released approximately 60% of its total P content, while Mg 3 Al-LDH-P released around 83% for the same amount of time.
Figure 8 shows the release profiles of P and cumulative release rates for Mg 2 Fe-LDH-P and Mg 3 Fe-LDH-P.The release profiles of P for Mg 2 Fe-LDH-P and Mg 3 Fe-LDH-P were adjusted for the Vapor Pressure and Hoerl modified   The Hoerl and Vapor Pressure models are variations of the exponential model.These models are extremely flexible and can be used in cases where other non-linear models fail.Due to the complexity of the medium and the number of variables involved, the Hoerl model has been used in studies to evaluate plants nutrition experiments with sufficient accuracy 77 .
Comparatively, among the five LDHs-P studied, the Mg 2 Al-LDH-P obtained better results from its release profile and cumulative release rate.The highest amount of P release in all LDHs-P occurred by exchange reactions.Once LDHs-P are in contact with water, they dissociate part of the hydroxyls from the layers 15     . The O 3 2-formed in suspension provides the exchange with phosphate intercalated in LDHs.
For Silva et al. 15 , the differences in release rates are controlled by factors such as layer components, charge density of the layers, and size of crystallites.The different compositions and M 2+ /M 3+ molar ratios presented in this work also evidenced these characteristics.Mg 2 Fe 0.5 Al 0.5 -LDH-P with three different cations in the layers promoted the formation of a material with higher structural disorder and lower crystallinity, as evidenced by PXRD measurements.The low crystallinity presented by this material provided a fast release rate compared to the other LDHs-P with better structural organization.
MgAl-LDHs-P released less P than the compounds with Fe 3+ , which demonstrates that Al 3+ retained phosphate in its structure with a stronger connection and thus lower possibility of displacing this anion.For the MgFe-LDHs-P, the amount of P initially released (fast moment) was higher than the amounts released in the samples containing Al 3+ .This relationship suggests LDHs synthesized with Al 3+ ions have better P retention capacity and that the release occurs gradually.
As suggested by Silva et al. 15 , in addition to the physical protection of anions provided by intercalation, LDHs also contain a chemical protection called the "interlayer ionic barrier".Both are responsible for the gradual release of intercalated anions.These authors conducted a slow release study of NO 3 -intercalated in MgAl-LDHs and MgFe-LDHs.A solution of NaHCO 3 was used for the release test.According to the authors, the exchange of NO 3 -intercalated in LDHs by the CO 3 2-of the solution starts in the edges of the layers, where the newly exchanged CO 3 2-anions form a barrier that hinders the exit of NO 3 -located in the internal region of the layers and prevents its fast release.
The results obtained in this work demonstrate that LDHs are materials that can resolve problems related with P losses in tropical soils, due to the protection and gradual release of the phosphate ions they contain.With regard to P recycling, LDHs are also potential materials because they can incorporate phosphate anions from aqueous solutions and then be used in the soil as sources of P.There are several reports in the literature about the use of LDH for P removal from sewage sludge, in residual water and even in ocean water 32,[78][79][80][81] .

Conclusions
The results presented in this work showed that by using the co-precipitation method followed by the ionic exchange method in solution it was possible to synthesize LDH containing phosphate ions in different systems, with different types of cations and different molar ratios (M 2+ :M 3+ ) between them (Mg 2 Al, Mg 3 Al, Mg 2 Fe, Mg 3 Fe and MgFeAl).In addition to significant amounts of P, the ionic exchange reaction produced materials with amounts of N and C.Among the LDHs obtained, Mg 2 Al-LDH-P presented the highest amount of P. The contents of P 2 O 5 , extracted with H 2 O, 2% citric acid and CNA + H 2 O extractors, showed that LDHs-P have similar characteristics to reactive sources of P, such as commercial thermophosphates.In studies on the kinetics of P release, LDHs-P showed a gradual P release profile and Mg 2 Al-LDH-P presented a slower release than the other materials.Likewise, LDHs presented structural and chemical characteristics that make them potential materials for use as storage matrices for slow P release.New studies and tests with plants and soils (bioassays) are being developed by our research group to try to validate this new LDH technology for use as a source of P. The development of a unique material able to solve all worldwide problems related to P in agriculture is not an easy task and may even be unfeasible.LDHs have characteristics however that make them potential materials for solving one or more aspects of this problem.
), which indicate the presence of lamellar materials, such as hydrotalcite.Using Bragg's equation (n•λ = 2•d•sin θ) and the average values of 2θ of the basal peaks (00l) of each LDH, it was possible to calculate the basal spacing d (spacing between two adjacent layers) of these materials.The obtained d values were 0.89; 0.88; 0.79; 0.79, and 0.81 nm for Mg 2 Al-LDH, Mg 3 Al-LDH, Mg 2 Fe-LDH, Mg 3 Fe-LDH, and Mg 2 Fe 0.5 Al 0.5 -LDH, respectively.These values concur with those reported in the literature for intercalation of NO 3

The synthesized Mg 2
Fe 0.5 Al 0.5 -LDH showed lower crystallinity when compared with the LDHs synthesized with two different cations.This result is highlighted by the lower intensity of the basal peaks in the diffractogram of Mg 2 Fe 0.5 Al 0.5 -LDH.The presence of three different types of cations in the layers results in lower structural organization when compared with LDHs synthesized with two different types of cations.

Figure 2
presents the diffractograms of pristine LDHs subjected to anionic exchange reaction in solution.Mg 2 Al-LDH-P and Mg 3 Al-LDH-P presented significant changes in the diffraction patterns, compared to pristine Mg 2 Al-LDH and Mg 3 Al-LDH.Two phases are found for these materials.In the first phase, the basal peaks are described in the diffractogram as ○(003) and ○(006) and the basal spacing values found for this phase were 0.62 nm for Mg 2 Al-LDH-P and 0.64 nm for Mg 3 Al-LDH-P.In the second phase, the basal peaks are marked in the diffractogram as •(003), •(006), and •(009).

Figure 5
Figure 5 presents the Raman spectra for pristine LDHs.The spectra show bands in the 470 cm -1 region, which are attributed to the vibration of connections Al-O-Al and Fe- LDHs were subjected to release tests of P in H 2 O.The pH values of the suspensions before the stirred-flow experiments were Mg 2 Al-LDH-P = 7.85; Mg 3 Al-LDH-P = 8.15; Mg 2 Fe-LDH-P = 8.37; Mg 3 Fe-LDH-P = 9.01, and Mg 2 Fe 0.5 Al 0.5 -LDH = 8.81.The cumulative rate releases and

Figure 7 .
Figure 7. P release in solution of Mg 2 Al-LDH-P and Mg 3 Al-LDH-P in relation to the time.(a) release profile, (b) cumulative rate release.

Figure 8 .
Figure 8. P release in Mg 2 Fe-LDH-P and Mg 3 Fe-LDH-P solutions in relation to time.(a) release profile, (b) cumulative release rate.

Figure 9
presents the release profile of P and the cumulative release rate for Mg 2 Fe 0.5 Al 0.5 -LDH-P.Mg 2 Fe 0.5 Al 0.5 -LDH-P had the same release profile with two distinct and complementary moments, as well as the previous ones.The Vapor Pressure model best represents the release profile of this LDH.It took Mg 2 Fe 0.5 Al 0.5 -LDH-P 16 min to reach the maximum point of release.Compared to the other materials however, Mg 2 Fe 0.5 Al 0.5 -LDH-P has a faster P release after entering the phase of descending release.The cumulative release rate shows that only Mg 2 Fe 0.5 Al 0.5 -LDH-Pcompletely released its full amount of P. The cumulative release reached approximately 100% after 130 min of flow.

Figure 9 .
Figure 9. P release in Mg 2 Fe 0.5 Al 0.5 -LDH solution in relation to time.(a) release profile.(b) cumulative release rate.

Table 1 .
List of cations, anions, molar ratio of cations and number of mols used for pristine LDHs Al 3+ and Fe 3+ ).Layered Double Hydroxides as Hosting Matrices for Storage and Slow Release of Phosphate Analyzed by Stirred-Flow Method **Solution prepared in 250 mL of H 2 O. ***0.0625mols of each trivalent cation were added (

Table 3 .
Contents of total P 2 O 5 , in water, 2% citric acid and neutral ammonium citrate (NAC)

Table 4 .
Specific surface area and pore volume of different LDHs