ISSN 00167029, Geochemistry International, 2013, Vol. 51, No. 4, pp. 290–305. © Pleiades Publishing, Ltd., 2013. Original Russian Text © A.Yu. Kudeyarova, 2013, published in Geokhimiya, 2013, Vol. 51, No. 4, pp. 326–343.
Trend and Mechanisms of Transformation of Natural Sorption Barriers in Acid Soils under Phosphate Loading A. Yu. Kudeyarova Institute of Physicochemical and Biological Problems of Soil Science, Russian Academy of Sciences, Institutskaya ul. 2, Pushchino, Moscow oblast, 142290 Russia email:
[email protected] Received March 10, 2011; in final form, November 29, 2011
Abstract—The current status of studies on chemisorption of phosphate anions on soil aluminum(iron)bear ing sorbents has been analyzed. The theoretical possibility of the formation of soluble phosphorus compounds during chemisorption was supported by experiments. It was shown that a decrease in efficiency of aluminum (iron)bearing mineral and humus–mineral sorbents is related to the peculiar transformation of their phos phatized surfaces with formation of soluble metallophosphate anionic complexes. Mechanisms of the decrease in the efficiency of soil sorbents and their destruction are considered. Keywords: chemisorption, phosphate anions, Al, Febearing sorbents, mechanisms of transformation of phosphatized surfaces, soluble metallophosphate complexes DOI: 10.1134/S0016702913040034
INTRODUCTION The effect of anthropogenic redistribution of chem ical elements in nature is now an urgent problem. Min eral phosphate salts are important components of tech nogenic displacements. Application of them (as fertiliz ers) caused a significant increase of phosphorus reserves in soils of humid zones [1–3]. In spite of the sharp decrease in their application since 1990s, sufficiently spacious areas of arable soils remain overphosphatized due to the previous systematic (approximately during 40 years) and progressively increasing phosphate fertili zation. The great amounts of phosphates were introduced with fertilizers in acid arable soils, which are localized mainly in the Northwestern and Far East regions, as well as in some areas of Central Russia [4, 5]. In partic ular, as much as 400–1700* kg Р2О5 /hectare were introduced in the arable soils in the considered regions during 1969–1988 [4]. These data are comparable with natural phosphorus reserves in the top (0–20 cm) layer of acid soils (1500–3500 kg Р2О5/hectare) [6]. At the same time, the total phosphorus content in the micro zones surrounding fertilizer granules may several times exceed natural content [7, 8]. It was shown [7], that the radius of phosphorus dif fusion zone of 2–3 mm fertilizer granule in acid soils (midloamy soddypodzolic and gray forest soils) is usu ally 1.5 cm. The centers of phosphate accumulation in soils have been preserved for long time (26 months of observations). The differences in phosphate contents between layers located at different distances from gran * The conversion factor from oxide to phosphorus and vice versa are as follows: 2.291 P2O5 P 0.436.
ules within these centers were also preserved. This implies the localization of introduced phosphates in soils and the appearance of soil systems with different phosphate loadings on natural sorbents, including humus substances. According to calculations [9], if a fertilizer granule (for instance, ammophos) contains 8 mg Р2О5, the phosphorus to humus carbon (P/C) ratio will be approximately 15–20 in the 0– 0.5 cm layer of acid soil around the granule and decreases to 0.4 cm in the 0.5–1 cm layer. Beyond this center with a radius of 1.5 cm, this ratio is several times of magnitude lower. All localized systems show progressive evolution with formation of increasingly complex structures [10]. Thereby, the heterogeneous distribution of phosphorus in soil indicates the existence of systems with different pace of evolution. To understand the evolution trends of overphosphatized soils, it is required to study processes proceeding in soil at different phosphate loadings, including those occurring in the immediate vicinity of granules. It is generally accepted [11–14] that phosphate anions binding metals (Me) of soil sorbents are strongly retarded on their surface and correspondingly loss their migration ability. However, these concepts are inconsis tent with facts of increasing (since 1970s) migration of Pbearing compounds from fertilized soils [15–17]. An increasing influx of phosphorus compounds in basins caused their eutrophication and, correspondingly, the bloom of water [17–19]. Considerable eutrophication spanned a limnic system of the Russian Northwest [20], which accumulated the majority of overphosphatized soils. Note that in spite of a sharp decrease in soil fertil izing since 1990s, the content of soluble phosphorus
290
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION
compounds actually has increased relative to amounts observed during intense use of phosphate fertilizers [5]. The possible formation of soluble, i.e., migrationcapa ble phosphorus compounds in acid soils in the presence of natural sorption barriers represented by compounds of aluminum and iron having strong chemical affinity to phosphate anions indicates an importance of deepened study of soil processes with participation of phosphates.
influence of all ligands in the internal coordination sphere of the complex due to the redistribution of electron density in it [27, 43]) and the appearance of additional donor O atom, the complex is stabilized due to the involvement of second Me atom in coordi nation [36]: OH Me O P OH O
STATUS OF STUDIES ON CHEMISORPTION OF PHOSPHATE ANIONS ON Al, FeBEARING SOIL SORBENTS The main natural sorption barriers in acid soils are such Al and Fe compounds as hydroxides, oxides, clay minerals, as well as corresponding organic (humic) derivatives. The surface layer of natural sorbents in acid soils is usually positively charged through electrondefi cient Mebearing atomic groups, in which the Me atom is mainly bound to water molecules and hydroxyl ions [21–24]. The formation of these groups was caused by the donor–acceptor interaction at the sorbent–water interface, where the Me atom of sorbent behaves as an acceptor of electrons supplied by donor O of H2O molecules and OH– ions [25–27]. Collectivizing of free (unshared) electron pair of O atom provides the formation of the donor–acceptor (coordination) bond between O atom and Me atom. Formed aqua (hydroxyl) complexes of Al and Fe on the sorbent surface should be considered as primary (native) sorption centers in acid soils. Native aqueous and hydroxyl ligands in aqua(hydroxyl) complexes on the surface of soil sor bents are easily substituted for phosphate anions as more strong Obearing ligands [28]. Since aqualigands are more readily substituted, the ability of acid soils to link phosphates is mainly determined by the content of amorphous Al and Fe hydroxides [29]. During forma tion of coordination bond, the electron density of new (phosphate) ligand is shifted to the Me atom. This leads to the cleavage of old Me–OH2 or Me–OH bonds and release of old ligand into solution. The release of OHligands is proved by the increase of pH values in liquid phase of the sorbent–phosphate solution system [22, 29–35]. The reaction of substitution of native ligands with participation of phosphates leads to the formation of metallophosphate complexes on the surface of soil sor bents [32, 36–41]. Literature data on phosphate sorption by soil sum marized in [28, 42] suggest that concepts of strong bind ing of phosphate anions on the surface of Al and Febearing polymeric sorbents are based on the possible formation of binuclear metallophosphate complexes, in which phosphate anion is bound by two metal atoms. Before their formation, the phosphate group coordi nates to one Me atom of hydroxide polymers. After deprotonation of phosphate ligand (effect of mutual GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
291
Me OH
Me O +
–H
O Me
OH
O P
O
+ H2O.
OH
Binuclear metallophosphate complexes are usually formed on the surface of polymeric sorbents at low Р2О5 concentrations in a liquid phase [22]. With increasing Р2О5 contents, individual fragments are cleaved from polymeric Mebearing molecules [44, 45], which facil itates an access of phosphate anions to Me atoms. The appearance of additional sorption sites is favorable for the coverage of sorbent surface by positively charged and neutral mononuclear metallophosphate complexes such as simple phosphate salts (MеzXy, where Me = Al3+, Fe3+; X = PO34−) [23, 38, 41, 46–50] (see scheme below). Sorption of phosphate anions from solutions with high Р2О5 concentration leads to the formation of neg atively charged metallophosphate complexes on the sorbent surface [22, 28, 35, 51]. This provides intense binding of cations presenting in phosphate solution by sorbents [52–54]. A scheme of sequential transformation of Me3+ bearing sorbent surface under the impact of phosphateanions (at pH ≤ 5.5). The substitution of H2Oligands with formation of positively charged and neutral metallophosphate com plexes (simple phosphate salts): +H2PO4– [Me(H2O)5H2PO4]2+ –H2O +H2PO4– [Me(Н О) (H PO ) ]+; 2 4 2 4 2 –H2O
[Me(H2O)6]3+
[Me(Н2О)4(H2PO4)2]+ [MeH2PO4]2+
–H+
+H2PO4– [Me(Н2О)3(H2PO4)3]0; –H2O
[MeHPO4]+
–H+
[MePO4]0.
Deprotonation of Pligands, formation of negatively charged complexes and salts of metallophosphoric acids: [Me(H2PO4)3]0
–2H+
[Me(HPO4)2]– [Me(PO4)2]3– +Me 2013
[Me(H2PO4)(НРО4)2]2–; –2H+
+, +Me2+
[Me(PO4)2]3;
Me+Ме2+[Me(PO4)2]0.
292
KUDEYAROVA
Substitution of Pligands in negatively charged com plexes: [Me(PO4)2]3–
+Ln– –PO43–
[Me(PO4)L]3–n.
A negative charge is acquired by metallophosphate complexes due to the coordination of Me atom by sev eral phosphate groups (mainly H 2PO 4− ion in acid soil) and their subsequent deprotonation in the internal coordination sphere of the complex. Deprotonated phosphate ligands are linked with Me atom by two O atoms to form negatively charged metallophosphate complex of a cyclic (chelate) structure: –
O O O P Me3+ P HO O O OH O
.
It follows that depending on the phosphate loading, the positively, neutral, and negatively charged metallo phosphate complexes are sequentially formed on the sorbent surface (scheme). The retardation of phosphate sorption on the nega tively charged phosphatized surfaces [53, 55–57] is usu ally explained by the following physical factors: repul sion of phosphate ions from the surface, low diffusion rate of sorbed phosphates inward the solid phase of sor bent, and a decrease in number of accessible sorption sites owing to the coverage of sorbent surface by previ ously formed metallophosphate complexes such as sim ple phosphate salts [11, 12, 48–50, 52, 55, 56]. The pos sible retardation of sorption owing to specific chemical reactions on phosphatized surfaces has not been con sidered yet in special literature. Thus, a review of studies on sorption of phosphate ions revealed that mechanisms of processes proceeding on the negatively charged surfaces of soil sorbents are poorly understood. Negatively charged metallophos phate complexes should be considered as complex anions of metallo(alumino, ferro) phosphoric acids [58–60], i.e. as higher order compounds than anions of common phosphoric acid. Owing to electrostatic attraction of cations of solution by complex anions [52–54], the layer of complex salts containing two (and more) atoms of different Me can be formed on the sor bent surface (scheme). Compounds Me1b [MеX], where Me1 = Ca, K, Na, and other metals; Me = Al3+, Fe3+; X = PO34− (minerals of taranakite, leucophosphite, crandallite, and other groups) can be formed in acid soils due to interaction with phosphate fertilizers [8, 61–63]. Direct instrumental studies [64] revealed such phosphate minerals in overphosphatized soils. They were also identified in the native soils and sediments [8, 65–67]. Complex metallophosphate salts are formed by interaction between phosphate solution and simple
phosphate salts [25]. In turn, complex salts also release free complex anions in it [58–60]. Formulas of some of them are shown in scheme. The ability of phosphorus compounds to be incongruently dissolved in phosphate solution theoretically supports the possibility of forma tion of soluble metallophosphate complexes. Experi mental evidence for this possibility is showing an increase of metal mobility in soils supplemented with phosphates [8, 68, 69]. These data highlight the necessity to study the phos phatebinding ability of sorbents in connection with the changes in composition of solid and liquid phases of the Al(Fe)bearing sorbent–phosphate solution systems. Such an approach may be useful for understanding mechanisms responsible for decreasing efficiency of natural sorption barriers in overphosphatized acid soils. This research has been carried out with the following aims: (1) to substantiate theoretically and experimen tally the possible formation of soluble metallophos phate compounds during impact of phosphate ions on the Al and Febearing sorbents; (2) to study trend of transformations of mineral and humus Al and Fe com pounds binding phosphate ions; (3) to study features and mechanisms of transformation of phosphatized surfaces of Al and Febearing sorbents. EXPERIMENTAL Techniques and Methods Model experiments were carried out using simple phosphate minerals (variscite, metavariscite), gibbsite (aluminum hydroxide), gray forest soil, and its Fe humus compounds as sorbents. Mineral systems. (1) We studied the structure of phosphate salts that were formed at different P/Al ratios in medium. Reaction mixtures (pH 2.55) with P/Al ratios of 1.2, 1.6, and 2.4 were prepared from 0.77 M solutions of NaH2PO4 ⋅ 2H2O and AlCl3 ⋅ 6H2O [70]. Mixtures with precipitated aluminum phosphate salts were kept at 90°С for 14 days. Then, the precipitates were filtered, dried in air, and ground in a jasper mortar. Obtained powders were used to prepare samples (sus pensions precipitated on glass) for identification on an Xray DRON0.5 diffractometer with Cu anode. The obtained product consisting of variscite and meta variscite (simple phosphate minerals) was studied by a similar manner prior and after its treatment with phos phate solution at 90°С for two weeks. (2) Structural modifications caused by phosphate anions in gibbsite (natural sorbent of acid soils) were studied in two systems (with pH 4.01): Al(OH)3–H2O and Al(OH)3–NH4H2PO4–H2O. Gibbsite (in propor tion 1 : 100) was added to deionized water (control) or to NH4H2PO4 solution (1 M). The systems were stirred for two hours, kept for two months at room temperature with daily shaking, and then solid phases were separated by filtration via dense filters. Air dried solid phases
GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION
(powder) were analyzed for Al and P contents using XRF Spectroscan MAKSGV analyzer (Russia). XRD study of solid phases was carried out on a DRON3 (Co anode) diffractometer by T.V. Alekseeva. The difference absorption electron spectra (ES) of liq uid phase of the Al(OH)3–H2O and Al(OH)3– NH4H2PO4–H2O systems were recorded on a Hitachi 557 spectrometer (Japan) using a 20mm cuvette. Water and liquid phase of the control Al(OH)3–H2O system were used, correspondingly, as standards during ES recording. The pH value and Al content were measured in liq uid phase of the gibbsite–phosphate solution system immediately after separation and during storage for three months in closed vessels at room temperature (with an interval of one month) using Optima 5300 DV atomicemission spectrometer (USA). At the end of experiment, glasslike acicular crystals precipitated from liquid phase were studied by XRF (for Al and P content) and Xray diffraction (Cuanode) methods. (3) Binding of phosphates by gibbsite was studied under dynamic conditions using glass columns. Col umns of 3 cmdiameter were loaded with wet gibbsite (in a mixture with pure quartz sand, 1 : 1) up to a height of 6 cm. The mixture was treated with КН2РО4 solution (Р2О5 concentrations of 16, 100, and 138 mg/L). Solu tions were added by 1 mlportions. The columns were equipped with jacks, which made it possible to sustain the flow rate of phosphate solution equal 1.2 cm/min, which provided 5min contact with sorbent. Each 5ml infiltrate fractions were collected in receivers and ana lyzed for pH (with glass electrode) and Р2О5 content (using “molybdenum blue” photocolorimetry [71]). Soil systems (1) Virgin gray forest soil (chemical characteristic is shown in [76]) was treated with KH2PO4 solutions under static regime according to [46]. Solutions containing different amounts of phos phorus (from 0 to 9000 mg Р2О5 /L) with starting pH = 4.5 were added to soil in 1: 5 ratio. The suspensions were shaken for one hour, and kept for 1 day with periodical stirring, and then filtered. The pH values, contents of K and Fe were determined as indicated above. Organic carbon (Corg) was determined in filtrates of soil suspen sions using bichromate oxidation [72]. (2) The impact of ammophos (consisting mainly of NH4H2PO4) on migration in soil was studied using lysi metric columns with sorbents [73]. Powdered ammo phos was added to the virgin gray forest soil in the amount corresponding to 150 mg Р2О5/100 g. In order to determine the influence of nitrogen component of ammophos, corresponding amount of nitrogen in form of NH4NO3 was added in control soil. The 500g samples from control and overphosphatized soils were moistened up to 80% of the moisture capacity and placed in columns, the lower part of which was filled with alternating layers of KU2 cationite in H+form, GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
293
EDE10P anionite in OH–form and neutral alumi num oxide. Each sorbent was represented by a 2.5cm layer and separated by interlayers of pure quartz sand. Columns were connected with receiving flasks by rub ber hoses. Within ten days, the soil was daily treated with 50 ml of deionized H2O. The aliquots of infiltrates percolating in receiver were analyzed for carbon content, and, after boiling in an H2SO4 (conc.) and 57% HClO4 mixture (in 3 : 1 ratio) were analyzed for phosphorus and iron con tents. Sorption columns were disassembled and the contents of absorbed chemical elements were deter mined in each sorbent after corresponding treatment. Cationite was treated according to [74], while anionite, using technique [75], and aluminum oxide was treated with 10% HCl at boiling. The Corg content in the alumi num oxide and infiltrates, as well as phosphorus and iron contents in solutions obtained after treatment of sorbents and infiltrates from receivers were determined as was noted above. (3) Changes in properties of humic substances of arable gray forest soil (chemical characteristic is given in [76]) due to overphosphatization were studied during 3year incubation. The overphosphatized (mixed with powdered ammophos in amount of 100 mg Р2О5/100 g soil) and control (with addition of nitro gen in form of NH4NO3) soils were incubated at room temperature and 55% humidity. After seven days, one year, and three years of incubation, the soil samples were dried in air and ground to 1 mm. Humic substances (HS) were extracted from each soil sample using 0.1 N NaOH (soil to solution ratio of 1 : 50). Alkaline extracts were mixed with 1 N H2SO4 (with 3 : 1 alkali to acid volume ratio) in order to precipitate humic acids (HA). Precipitated HA were filtered and dissolved in 0.1 N NaOH. The con tents of Corg as well as Al, Fe, and P (after acid com bustion) were determined in aliquots of starting extracts of HS, alkaline solutions of HA, and acid solutions after separation of HA (fulvic acids, FA). Difference electron spectra (ES) of alkaline extracts of HS from overphosphatized arable soil were recorded on a Specord UVVIS (Germany) spectrophotometer using 2cm cuvette. Difference ES of HA solution from overphosphatized soil was recorded on a Hitachi557 spectrophotometer (Japan) using 0.5mm cuvette. Solutions of HS and HA from the control (Pfree) soil were used as standards. The technique of ES recording and their interpretation were described in detail in [76, 77]. System with ferrohumate complex. In order to obtain ferrohumate complex, chloric iron was added to alka line solution of HA from arable soil in proportion 300 mg Fe per 50 mg C (pH 5.1). Precipitated complex was filtered, washed with Н2О, and dried at room tem perature. The C : Fe ratio in complex was 1 : 3.2. The ferrohumate complex was treated by NaH2PO4 solution 2013
294
KUDEYAROVA
Table 1. Dynamics of changes in the liquid phase of the gibbsite–phosphate solution system (average ± SD, n = 4) Aliquot sampling
pH
Al, mg/100 mL
After separation of solid phase After one month After two months After three months (after separation of crystals)
5.13 ± 0.02
9.0 ± 0.21
5.06 ± 0.01 5.03 ± 0.01 4.65 ± 0.01
7.4 ± 0.18 4.7 ± 0.09 0.3 ± 0.01
at pH 4.04 for three and five weeks. The C : P and Fe : P ratios in the system were 1 : 17.8 and 1 : 5.3, respectively. They reflected conditions in a thin soil layer surround ing a granule of soluble phosphorus fertilizer. Addition of phosphate in a system with a Fehumate complex immediately brought about the brown color of liquid phase and simultaneous formation of white mud and dark brown flakes, which indicates a cleavage of the complex. Exposure of system at room temperature led to its distinct splitting into light brown liquid and two well separating products, the light gray crystalline product and amorphous flakelike darkbrown mass. All phases were analyzed for carbon, iron, and phos phorus contents. Solid dark brown phase was studied using IR spectroscopy on a Specord M80 apparatus (KBr technique). Spectra were interpreted according to [60, 78, 79]. All experimental data presented in the work are aver age values of 3–4 replicate analyses. The reliability of differences between run variants was estimated using dispersion analysis [80]. The differences between variants were taken to be significant if they were more than least significant difference (LSD) at a signifi cance level of 0.05 (95%). Analytical data were pro cessed using [81]. RESULTS AND DISCUSSION Mineral systems. Figure 1 demonstrates XRD pat terns of aluminum phosphate salts formed in reaction media with different P:Al ratios. Aluminum phosphates precipitated from a mixture with P:Al ratio of 1.2 repre sented Xray amorphous mass (Fig. 1, spectrum a). Crystalline aluminum phosphates were obtained from a mixture with P : Al ratio of 1.6 (Fig. 1, spectrum b). Their XRD pattern exhibits peaks at 6.26, 4.50, 3.47, 2.69, and 1.94 Å ascribed to metavariscite and peaks at 5.32, 4.19, 3.88, 3.02, 2.89, and 2.47 Å typical of variscite [82, 83]. Both the minerals have a common chemical formula (AlРО4 ⋅ 2Н2О) but different syng ony. They belong to a class of simple phosphatic salts. XRD pattern of crystalline aluminum phosphates syn thesized at P : Al ratio of 2.4 (Fig. 1, spectrum c), in addition to peaks assigned to the aforementioned two minerals, showed also peaks at 15.23 and 6.9 Å that are
indicative of complex salts of aluminophosphoric acid (in particular, minerals of the taranakite group) [82, 83]. Data presented in Fig. 2 show that XRD pattern of product consisting of simple phosphate salts (variscite and metavariscite) (Fig. 2, spectrum a) changed after its twoweek treatment with phosphate solution (Fig. 2, spectrum b): initial peaks of simple phosphate salts were replaced by peaks typical of complex salts [82]. Obtained results showed sequential formation of amor phous and crystalline simple aluminum phosphates with increasing P : Al ratio in system. In the presence of free phosphate anions, they were transformed into complex salt of aluminophosphoric acid. Figure 3a demonstrates XRD pattern typical of gibbsite [84]. After treatment with phosphate solution (Fig. 3b), many of gibbsite peaks disappeared, while intensity of remained peaks (4.89 and 4.39 Å) sharply decreased, indicating its transformation. The transfor mation product showed peaks at 16.29, 8.03, 7.56, 5.95, 3.85, 3.38, 3.18, and 2.65 Å typical of ammonium taranakite with a formula H6(NH4)3Al5(PO4)8 ⋅ 18Н2О [83]. The Al and P contents in the newly formed prod uct were, respectively, 7.84 and 10.73% (initially, gibb site contained 24% Al). An increase in pH of liquid phase of the gibbsite–phosphate solution system from 4.01 (initial value) to 5.13 testifies that formation of taranakite was preceded by the substitution of OHligands of gibbsite for phosphate ligands. A release of OH groups was accompanied by release of aluminum in a liquid phase of phosphate system (9.0 mg Al/100 ml as compared to 0.1 mg Al/100 ml in the control system with water). The Al release suggests the dissolution of gibbsite during binding of phosphate anions. Difference ES of liquid phase of the phosphate system (as com pared to control system) (Fig. 4) revealed the presence of four aluminum phosphate complexes with absorp tion bands within wavelengths of 230–260, 265–325, 350–410, and 420–450 nm. A threeweek storage of liquid phase of the gibbsite– phosphate solution system resulted in the progressive decrease of pH values and degree of Al detection (Table 1), which is consistent with deprotonation of phosphate ligands and their increasing ability to form chelate complexes. Observations have no revealed any visual changes in a state of liquid phase for approxi mately 2.5 months: solution was absolutely transparent and devoid of solid particles. After a 3month storage period, acicular crystals containing 1.5% Al and 21% P were precipitated from liquid. XRD pattern of the precipitate (Fig. 5) differed from that of NH4H2PO4 of analytical grade (Fig. 5, spectrum a) applied to prepare initial phosphate solu tion for gibbsite treatment. Note that no precipitate was formed in starting solution after a 3month storage. Based on XRD pattern, the product synthesized in a liquid phase of the gibbsite–phosphate solution system (Fig. 5, spectrum b) corresponded to ammonium (Fig. 5, spectrum c) and aluminum (Fig. 5, spectrum d)
GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION
295
4.50
(c)
70
65
60
55
50
4.23
6.91
15.23
6.28
3.48
3.02
2.19 2.28
2.05
1.85
1.66 1.75
1.60
1.53
1.34 1.40 1.44
1.94
2.47
2.69
4.51
Intensity
5.32 6.26
(b)
5.34
4.19 2.89 3.02 3.22 3.47 3.88
2.47
1.94 2.05 2.19 2.27
1.75 1.84
1.66
1.51 1.60
1.48
1.34 1.40
2.69
(а)
45
40
35
30
25
20
15
10
5 2θ CuKα
Fig. 1. XRD patterns of aluminum phosphates synthesized at different P : Al ratios in reaction mixture: (a) P : Al = 1.2; (b) P : Al = 1.6; (c) P : Al = 2.4.
salts of pyrophosphoric acid (Н4Р2О7). According to [82], peaks observed in XRD patterns in Figs. 5b, 5c, and 5d are typical of salts of all condensed phosphoric (polyphosphoric) acids with a general formula (Hn+2PnO3n+1). Obtained data indicate that binding of common orthophosphate anions by gibbsite was accompanied by the formation of soluble aluminum pyrophosphate anionic complexes passing into liquid phase. One can suggest that the formation of these complexes resulted from reactions in the internal coordination sphere of initially forming orthophosphate complex anions of [Me3+(PO4)2]3– type. Due to a shift of electron density in such complexes to the unbound oxygen atoms, P atom of ligand obtains effective positive charge and becomes a center of coordination of additional phos GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
phate anion, i.e. a center of condensation of phosphate groups [85]. Pyro and polyphosphate complex anions [Me3+(P2O7)2]5– and [Me3+(P3O10)2]7– have higher negative charge than their orthophosphate analog [Me3+(PO4)2]3–. Cation deficiency for compensation of negative charge leads to passing of complex ions into liquid phase. The higher the charge of complex anion, the more its ability to be released in liquid phase [58, 60]. It was shown [86] that polyphosphates more intensely dissolved metallic compounds in soil than orthophosphates. Correspondingly, polyphosphate metallic complexes have higher ability to migrate in aqueous solutions. Data presented in Fig. 6 show that binding of phos phate anions during free percolation of phosphate solu tions via gibbsitefilled columns occurred in two stages. 2013
KUDEYAROVA 4.25
296
4.30
4.98
7.81
Intensity
6.40
5.39
4.56 4.82
3.50 3.92 3.52 25
20
15
(b)
1.94
3.12
3.20
3.05
2.91
2.48
2.20
1.95
2.71
(а)
50
45
40
35
30
10 2θ CuKα
Fig. 2. XRD patterns of aluminum phosphates: (a) mixture of variscite (shaded peaks) with metavariscite; (b) the same, after treatment with phosphate solution.
The first stage involved a complete elimination of phos phates from percolating solutions, whereas the second stage was marked by a sharp decrease in sorption effi ciency: the Р2О5 content in infiltrates was higher than that in percolating solution and tended to increase with each next portion. This implies the release of phos phates adsorbed at the first stage. Phosphates were released during passage of phosphate solutions of diverse concentrations via gibbsite columns; an increase in concentrations led to the earlier beginning and inten sification of this process. Obtained data indicate that the incorporation of phosphate groups in the structure of surface layer of sor bent owing to coordination by positively charged alumi num atoms (first sorption stage) results in the formation of phosphatized layer, which is dissolved under the impact of free phosphate anions (potential ligands) sup plied from the outside. The above considered data (Figs. 4, 5) suggest that dissolution was related to the formation of negatively charged aluminum complexes with phosphate, including polyphosphate ligands. The formation of aluminum polyphosphate complexes indi cates that chemisorption of phosphate anions is not limited by their coordination to the metal atoms. If con centration of phosphates in solution is high enough, they may be coordinated to the positively polarized P atoms of ligands of aluminum phosphate complexes. Soil systems. The study of liquid phase of the virgin gray forest soil–phosphate solution system showed that its variations strongly depends on phosphate loading (Fig. 7). Three intervals of Р2О5 concentrations in start ing solution corresponding to different composition of liquid phase in the soil–phosphate solution system were revealed. The first interval included up to 10 mg
Р2О5/L, the second interval, from 10 to 300 mg Р2О5/L, and the third interval, more than 300 mg Р2О5/L. Phosphate loadings within the first interval led to the pH increase of liquid phase of soil systems relative to pH 4.5 in control (Pfree) system (Fig. 7a), which is consis tent with substitution of original soil OHligands during coordination of phosphate ligands. Phosphate loadings of the second and third intervals provided pH decrease of liquid phase of soil systems (Fig. 7a) owing to the deprotonation of ligands in met allophosphate complexes. Acidification was accompa nied by increase of amount of potassium bound by soil from phosphate solution (Fig. 7b). Binding of potas sium indicates that phosphatized surface of soil sorbents acquired negative charge owing to increasing dentation of phosphate ligands. Stages of acquisition of negative charge by surface of Febearing sorbents corresponded to the Fe release into liquid phase (Fig. 7c), which was sharply intensified with phosphate loading of soil. The iron release proves the formation of soluble iron phosphate anionic com plexes and, respectively, the destruction of phos phatized surface of Febearing soil sorbents. The for mation of soluble iron complexes was followed by an increase of organic carbon in liquid phase with further increase of phosphate impact on soil (Fig. 7c). It can be assumed that liquid anionic iron complexes as strong nucleophilic reagents catalyzed destruction of humic macroligands of soil sorbent complexes. As was shown by Table 2, the products of Pinduced destruction of soil iron–humus compounds passing into liquid phase were involved in migration with infil tration waters. It is seen that the iron was removed by
GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION
16
24
32
1.756
2.171
2.459 2.395
3.360 3.199
5.395
4.395
4.889
(а)
16.289
40
48
56
64
48
56
64
(b)
8
16
24
32 40 2θ CoKα
2.277
2.649
7.560
5.952 5.313 4.691 4.395 4.040 3.848 3.612 3.381 3.180 2.968 2.858
4.866
8.036
Intensity
8
297
Fig. 3. XRD pattern of gibbsite: (a) system with H2O, (b) system with phosphate solution.
water from overphosphatized soil more rapidly than from control soil, and together with phosphorus or with carbon and phosphorus. The majority of phosphorus migrated with mineral iron anionic complexes. This is supported by its combination with iron found on EDE 10P anionite. Soluble iron complexes with C and P bearing ligands were absorbed by neutral aluminum oxide, which suggests their chelate structure. As was mentioned above, the coordination of a new ligand causes a redistribution of electron density (charges), which enabled interaction between inner sphere ligands. The interaction between ligands may result in a change of their reactivity, providing possibil ity of coordination of solution species to atoms of com plex ligands [27]. As was shown above (Fig. 5), P atoms GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
of phosphate ligands of aluminum complexes are capa ble of interacting with free phosphate anions in solution to form polyphosphate ligands. The distribution of electron density is particularly important in studying mechanisms of reactions with participation of organic molecules [85, 87]. The changes in electron system of structurally modified organic macromolecules can be identified using differ ential (difference) electron spectroscopy [88]. This method was used to study the dynamics of structural modifications of molecules of humic substances (HS) of overphosphatized arable soil (Fig. 8). Difference ES showed that molecules of HS from overphosphatized soil contained modified fragments, the electron system of which sharply changed with time. In particular, after 2013
298
KUDEYAROVA
absorption respectively, within 210–230 and 250– 280 nm wavelengths [88].
D 0.100 0.075 2
0.050 1 0.025 0 200
250
300
350
400
nm
Fig. 4. Difference electron absorption spectra of liquid phase of systems: (1) gibbsite–H2O (H2O as standard), (2) gibbsite–NH4H2PO4–H2O (gibbsite–H2O system as standard).
sevenday incubation, the difference ES of HS (Fig. 8, curve 1) was characterized by increasing intensity of electron transitions (relative to HS of control soil), mainly in the visible part of spectrum (wavelength more than 350 nm). If organic molecule contains metal atoms, the changes in visible region are usually assigned to the transformation of electron system in their inner sphere owing to the incorporation of a new ligand [87, 89]. Curve 1 (Fig. 8) presumably reflected a stage of coordination of phosphate ligands to the metal atoms of humic molecules, when electron density is shifted from ligand to metal. The difference ES (Fig. 8, curves 2, 3) recorded after one and threeyear incubation of overphosphatized soil revealed that electron density in Pmodified fragments of humic molecules was shifted towards organic ligands, which follows from increasing absorption of chro mophore groups in UVregion. The most pronounced change in this region was noted in curve 3, which char acterizes electron state of humic molecules of overphos phatized soil after threeyear incubation. The presence of a broad absorption band at 215–300 nm is presum ably related to the intramolecular interactions of chro mophore groups with C=C– and C=Obonds, which
A change in spectrum marks Pinitiated structural transformations in humus microligands, which, in turn, affected the extractability of carbon and metals (Al,Fe) from humic substances of overphosphatized soil with 0.1 N NaOH after threeyear incubation (Fig. 9a). The distribution of chemical elements passing into the alka line extract between fractions of humic acids (HA) and fulvic acids (FA) also changed after soil incubation (Figs. 9b, 9c), In particular, after 3year incubation of overphosphatized soil, carbon of extracts tended to be accumulated in HA, while phosphorus and metals were partitioned in FA (Fig. 9c). The transition of majority of P and Me in FA fraction was preceded by accumulation of these elements in HA (after oneyear incubation) (Fig. 9b). This implies sequential structural transforma tion of Pmodified humic acids, which caused a release of P and Mebearing atomic groups in acid solution (FA fraction). Obtained data indicate that acidbasic properties of HA of overphosphatized soil changed within a period from 1 to 3 year incubation. Since a change in acid–basic properties of ligands are thought to be related to their reorganization [27, 90], one may suggest that this period marks a considerable change of humic acids owing to the involvement of Pbearing groups in their structure. Difference ES of humic acids of overphosphatized soil after threeyear incubation (Fig. 10) revealed the partial overlapping of four absorp tion bands with maximums at 205, 220, 250, and 310 nm in UV region, which were not observed in spec trum of control HA. The presence of additional absorp tion bands (besides two bands assigned to the main groups with C=C and C=O bonds) implies the appear ance of Pbearing chromophore groups in humic acid. These groups, as seen from the spectrum, affected the electron state of main chromophores. The strongest influence of Pbearing groups was noted within the region of 230–290 nm (broad and intense band), where chromophores with double heteroatom bonds (for instance, C=O and P=O) have absorption bands [88]. Absorption in 290–340 nm ascribable to the metal bearing groups in organic macroligand [89, 90] makes it
Table 2. Effect of phosphates on the removal of Fe, C, and P from soil with water (mg/kg soil) Including absorbed Variants of the experiment
Control (soil without phosphorus) Soil with introduced P* LSD0.05
Fe
C
P
Al2O3
EDE10P anionite
Fe
C
P
Fe
P
5.2
38.4
1.0
1.4
31.0
0.2
0.4
0.8
13.6 0.2
88.9 2.6
69.9 0.4
4.0 0.2
67.6 3.0
9.0 0.2
5.8 0.2
60.1 0.4
Note: * Addition of 65.4 mg P per 100 g soil. GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION
299
2.659
(а)
3.770
1.475
1.684
2.019
1.473
1.603
1.778
2.010 2.014 2.014
(c)
3.089
0.6 0.4
3.770
0.2
1.0
1.778 1.678 1.605 1.539
0.8
2.659
5.372
1.0
2.380
0.2
3.770
Intensity
0.4
2.380
0.6
3.089
5.574 5.372
0.8
2.374
1.0
(b)
2.659
0.2
2.959
0.4
5.985
12.109
0.6
3.786
5.405
0.8
3.110
1.0
(d)
0.8
2
8
16
24
32 40 2θ CuKα
1.684
0.2
2.674
5.405 4.396 4.171
0.4
3.089
0.6
48
56
64
Fig. 5. XRD patterns of phosphate salts: (a) NH4H2PO4 of analytical grade; (b) precipitate in a liquid phase of the gibbsite–phos phate solution system; (c) (NH4)4P2O7 of analytical grade; (d) Al4(P2O7)3 of analytical grade. GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
300
KUDEYAROVA 3
(а)
450
(b)
pH
5.5 5.0 4.5 4.0
175
ΔK, mg/100 g soil
2 125
300
100
150
75 50
ΔC, Fe, mg/100 g soil
P2O5 in infiltrate, mg/L
150
1
25
15 10 Fractions
5
20
Fig. 6. P2O5 concentration in infiltrate fractions after pass ing of phosphate solutions (1, 2, 3) via columns with gibb site: (1) 16, (2) 100, (3) 138 mg P2O5/L.
O M
P O
O P,
which may contribute in a general bond system and thus affect the shape of spectrum [90]. This assump tion is supported by data in Table 3, which demon strate that HA of overphosphatized soil after 3year incubation have higher contents of iron, aluminum, and phosphorus than HA of control soil, which fol lows from the lower C : (Al + Fe) and C : P ratios. The metallophosphate anions forming during coordina tion of simple phosphate anions to the atoms of triva lent metals and passing into solution could then inter act with atoms of humic macroligands. It is known [91] that complex anions as stronger nucleophilic reagents relative to simple phosphate anions may be easily coordinated to positively polarized (electro philic) C atoms of organic molecules. System with IronHumate Complex. Mechanisms responsible for the transformation of humic macroli gand in phosphate solution were studied in model experiment with synthesized ironhumate complex. Obtained results showed that the treatment of iron humate complex with phosphate solution led to its destruction with release of soluble and two solid prod ucts, whose compositions are listed in Table 4. It is seen that both solid products represented iron complexes with phosphorus and carbonbearing ligands. The
Fe C
30 20 10
10
possible to suggest that Pmodified humic acids contain newly formed metalphosphate fragments such as O
(c)
40
25
100
300 750 1500 3000 P2O5, mg/L
Fig. 7. Diagram showing variations of parameters of liquid phase of soil systems versus different loading of KH2PO4 (pH 4.5): (a) pH, (b) binding of potassium from KH2PO4 solution by soil; (c) release of iron and carbon from soil into phosphate solution (Δ is an increase of K, Fe, C amounts relative to control experiment (without P), mg/100 g soil).
phosphorus content in the ligands of two solid com plexes is approximately two and seven times higher than the carbon content. At the same time, around 45% car bon and 7% iron contained in initial ironhumate com plex were found in the liquid phase of ironcomplex– phosphate solution system. Thus, the coordination of phosphate anions to iron atoms was accompanied by release of both ironphosphate complexes, and organic D 0.6
3
0.4 0.2 0 200
2 1 250
300 350 400
500 600 nm
Fig. 8. Difference electron absorption spectra of Pmodi fied fragments of molecules of humus substances of soil at different stages of transformation (the solution of humus substances from soil of corresponding control (Pfree) variant was used as standard): (1) seven days, (2) one year; (3) three years of incubation.
GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION
GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
800
(a)
700 600 500 400 300 200 100 0 600
(b)
500 mg/100 g soil
ligand groups of humic macroligand into a liquid phase (similar data were obtained in experiments with soil, Table 2, Fig. 7). Figure 11 shows the IR spectra of starting iron humate complex (spectrum a), as well as dark brown (with higher carbon content) product of its decomposi tion in phosphate solution. The decomposition product after three (spectrum b) and fiveweek (spectrum c) interaction with phosphate solution revealed strong dif ferences. The presence of intense band at 1052 cm–1 in the spectrum of threeweek product may indicate that phosphate groups (P–O bond absorption corresponds to a band at 538 cm–1) are coordinated to positively polarized C atoms of polar C–O bonds in alcoholic, phenol, polysaccharide, and other structural compo nents of humic macroligand to form new groups –C–O–P, where effective positive charge is concen trated on P atom [60, 78]. It should be noted that IR spectrum (b) of threeweek product preserved absorption bands ascribed to stretching vibrations of carbonyl groups coordinated to metal atoms, as well as OHgroups, which existed in starting complex (region 3320–1402 cm–1) [78]. However, IR spectrum of fiveweek product (Fig. 11c) revealed strong changes in the indicated regions, as well as in the lower frequency region. The formation of additional bands within 1100–1000 cm–1 corresponding to vibrations of P–O–C groups [60] points to the appearance of organic substituents at posi tively polarized P atom of previously formed C–O–P group. A weak absorption at 910 cm–1 assigned to pyro phosphate (P–O–P) group [78] presumably indicates also binding of phosphate group to P atom. According to obtained data, organic and phosphate ligand groups that are present in liquid phase may be built in solid complex, adding to electrophilic P atoms of modified humic microligand. It is known [87] that interaction with electrophilic P atom leads to the formation of compounds with P=O group, which corresponds to the presence of two bands ascribed to stretching vibrations of this group (free P=O at 1250 cm–1 and P=O participating in the formation of additional bonds at 1170 cm–1) in IR spectrum [60, 78]. A wide band at 2392 cm–1 indicates the formation of strong intramolecular hydrogen bond between P=O group of formed ester fragment and OHgroup [78, 79]. It may be ascribed also to P=Ogroup, the oxygen atom of which is coordinated by metal atom [60, 79]. The dis appearance in fiveweek product (Fig. 11c) of absorp tion bands at 1630 and 1396 cm–1 observed in spectrum of three week product (Fig. 11b) and the appearance of a broad band at 1664 cm–1 correspond to the formation of intramolecular (chelate) cycles containing not only P=O, but also C=O groups of humic microli gand [60, 78]. Obtained data show a successive transformation of Pmodified humic macroligand under the influence of
301
400 300 200 100 0 300
(c)
250 200 150 100 50 0 Without P P, 7 days P, 1 year Incubation period 1
2
P, 3 years
3
Fig. 9. Dynamics of C, Me, and P extractability from over phosphatized soil by 0.1 N NaOH and distribution between fractions of humic and fulvic acids (mg/100 g soil): (1) C, (2) Me (Al + Fe), (3) P; (a) content in extract, (b) in humic acids, (c) in fulvic acids.
D 0.15 0.10 0.05 0 220
260
300 340 nm
380
420
Fig. 10. Difference electron spectrum of humic acid from overphosphatized soil after threeyear incubation (humic acid of control Pfree soil as standard). 2013
302
KUDEYAROVA
Table 3. Effect of phosphates on the content of carbon, metals (Fe, Al) and phosphorus in the humic acids of arable soil C
Fe
Al
P
Variants of the experiment
C/(Fe+Al)
C/P
mg/100 g soil Control (soil without phos phorus) Soil with introduced P* LSD0.05
349
15.9
12.5
2.1
12.3
166
552 26
24.2 3.1
31.4 2.8
5.4 0.2
9.9 n.d.
102 n.d.
Notes: * Addition of 43.6 mg P per 100 g soil.
Table 4. Composition of iron humate complex and products of its transformation in phosphate solution, % (average ± SD, n = 3) Complex
C
Starting Pmodified: dark brown light gray
Fe
21.1 ± 1.1
66.7 ± 2.3
0.25 ± 0.02
5.1 ± 0.3 2.2 ± 0.2
15.8 ± 0.9 30.1 ± 1.6
11.8 ± 0.27 13.9 ± 0.31
organic ligand groups released during phosphate coor dination.
836
1402
Intracomplex metalbearing fragments with strong hydrogen bonds have a cyclic structure and are inter preted as components of a general conjugated bond sys tem in molecules [78, 87, 88, 90]. This presumably was reflected in a shape of difference spectral curve of humic acids from overphosphatized soil (Fig. 10). One can suggest that an increase in absorption intensity of chro mophores with C=O bonds and hypsochromic widen ing of corresponding band were caused by adjacent HO–P=O group (resonance structure) having brightly expressed ability to form both hydrogen bond and com plex with metal [60, 78, 79]. Resonance effect presum ably was responsible for the change in acidbasic prop erties of humic acids of overphosphatized soil with increase of incubation period up to three years (Fig. 9c).
680
432
1056
3320 2920 2848
1610
(а)
538
1052
1396 1272
2848 2920
3252
1630
(b) Transmission, %
1800 1500 1200 900
520 420
1250 1170 1050 910
1664 2392
2852
3440
(c)
3530
P
600 сm–1
Fig. 11. IR spectra: (a) iron humate complex, (b) the same, after threeweek treatment by phosphate solution; (c) the same, after fiveweek treatment.
compounds that are present in liquid phase. The forma tion of new ligand groups with P=O bond indicates that the mechanism of transformation was based on the interaction between positively polarized P atom and free potential ligands that are present in liquid phase and contain (besides phosphate) metallophosphate and
The release of P and Mebearing fragments into fraction of fulvic acids after threeyear incubation (Fig. 9c) is well consistent with possible formation of (Me)HO–P=O groups typical of acid organic phos phate esters in Pmodified humic molecules. According to [79], such esters are decomposed under the effect of mineral acids and salts with release of phosphate groups. The destruction mechanism is related to reac tions at positively polarized P atom with participation of anions of acids and salts. Interaction between P atom and phosphate anions (as was shown above) was respon sible for the destruction of phosphated surface of gibb site with release of pyrophosphate aluminum com plexes into solution.
GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION
CONCLUSIONS Results of our studies showed that chemisorption of phosphate anions on aluminum (iron)bearing sorbents of acid soil is not limited, as is generally accepted, by coordination to positively charged metal atoms with formation of solid metallophosphate complexes, in which phosphorus is protected from involvement in migration. Such complexes may be formed only on the primary sorption centers represented by native posi tively charged aqua(hydroxyl) complexes of metals. This (first) stage of chemisorption should be considered as preparatory to subsequent destructing of phos phatized negatively charged surfaces of sorbents under the effect of potential ligands of liquid phase, which in addition to phosphate ions, include metallophosphate and organic groups released during phosphate coordi nation. It was shown that the destruction of phos phatized mineral and humus–mineral sorbents of acid soil is related to the coordination of newly formed anionic metallophosphate complexes to the positively polarized P atoms. Such reactions on a phosphatized gibbsite surface represented by ammonium salt of alu minophosphoric acid, taranakite, provided the forma tion of soluble aluminum pyrophosphate complexes. As for phosphatized humusmineral sorbents, reactions at P atom of humic macroligand determined its trans formation resulting in the formation of new groups with P=O bonds, which promoted release of phosphate frag ments in acid medium. According to obtained data, a decrease of the effi ciency of sorption barriers in acid overphosphatized soils is explained by the formation of soluble Pbearing products of destructive transformation of aluminum and iron compounds fulfilling function of main sorp tion barriers in acid soils. ACKNOWLEDGMENTS We are grateful to T.V. Alekseeva, N.V. Perfilova, V.E. Ostroumov, and P.I. Kalinin for help in investiga tions. The work was supported by the Program no. 4 of Basic Research of Russian Academy of Sciences. REFERENCES 1. V. N. Pereverzev and E. A. Koshleva, “Fractional Com position of Phosphates of Cultivated Podzolic Soils of the Murmansk District,” Agrokhimiya, No. 5, 10–20 (1990). 2. V. N. Efimov and A. I. Ivanov, “Hidden Degradation of Well Cultivated Soils of Russia,” Agrokhimiya, No. 6, 5– 10 (2001). 3. E. Barberis, F. A. Marsan, R. Scalenghe, A. Lammers, U. Schwertmann, A. C. Edwards, R. Maguire, M. J. Wil son, A. Delgado, and J. Torrent, “European Soils Over fertilized with Phosphorus. I. Basic Properties,” Fertil. Res. 45 (3), 199–207 (1996). GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
303
4. A. Yu. Kudeyarova, Phosphatogenic Transformation of Soils (Nauka, Moscow, 1995) [in Russian]. 5. S. A. Shafran, “Dynamics of Soil Fertility versus Appli cation of Fertilizers,” Agrokhimiya, No. 1, 9–17 (2004). 6. Soils of Northeast Europe and their Fertility (Nauka, Len ingrad, 1989), [in Russian]. 7. A. Yu. Kudeyarova, “Problem of Size and Timing of Phosphorus Accumulation Zones in Soils,” Agro khimiya, No. 2, 32–37 (1979). 8. A. Yu. Kudeyarova, Pedogeochemistry of Fertilizer Ortho and Polyphosphates (Nauka, Moscow, 1993) [in Rus sian]. 9. A. Yu. Kudeyarova, “Methodology of the Study of Humic Substances: Theoretical and SpatialTemporal Conceptual Aspects,” Euras. Soil Sci. 37 (Suppl. 1), 85– 92 (2004). 10. I. Yu. Vinokurov, Evolution of Soil Ecosystems: Chemical Pollution, SelfRegulation, SelfOrganization, and Stabil ity (Yurkniga, Moscow, 2007) [in Russian]. 11. S. N. Ivanov and T. F. Stolyarova, “Role of Iron and Alu minum Sesquioxides in the Adsorption of Phosphate Ions by SodPodzol Soils,” Pochvovedenie, No. 7, 72– 78 (1972). 12. N. L. Afonina, R. Del’gado, O. A. Agafonov, and O. G. Us’yarov, “Sorption of Phosphate Ions by Iron Clay Mineral Complexes,” Pochvovedenie, No. 12, 36– 42 (1983). 13. I. S. Kaurichev, A. S. Pel’ttser, T. A. Anisimova, and V. K. Kukushkin, “Sorption and Isotope Exchange of Phosphate Ions in the Gleized and NonGleized Sod Podzol Soil,” Izv. Timiryaz. Sel’khoz. Akad., No. 2, 55– 63 (1989). 14. R. E. Eleshev, A. L. Ivanov, and M. Shakhdzhakhan, “Phosphate Sorption by Individual Soil Components,” Agrokhimiya, No. 11, 32–39 (1992). 15. E. Ahti, “FertilizerInduced Leaching of Phosphorus and Potassium from Peatlands Drained for Forestry,” Commun. Inst. Forest. Fenn. 111, 1–20 (1983). 16. A. Yu. Kudeyarova and V. N. Bashkin, “Study of Land scape–Agrogeochemical Balance of Nutrients in Agri cultural Regions. 1. Phosphorus,” Water, Air and Soil Pollut. 21, 87–95 (1984). 17. I. S. KoplanDiks, G. V. Nazarov, and V. K. Kuznetsov, Role of Mineral Fertilizers in the Eutrophication of Conti nental Waters (Nauka, Leningrad, 1985) [in Russian]. 18. Evolution of Phosphorus Cycle and Eutrophication of Nat ural Waters (Nauka, Leningrad, 1988), p. 205 [in Rus sian]. 19. M. J. S. Bellemakers and M. Maessen, “Effects of Alka linity and External Sulphate and Phosphorus Load on Water Chemistry in Enclosures in an Eutrophic Shallow Lake,” Water, Air, and Soil Pollut. 101 (1–4), 3–13 (1998). 20. N. A. Petrova, I. V. Iofina, L. L. Kapustina, T. P. Kulish, T. N. Petrova, and G. F. Raspletina, “Anthropogenic Eutrophication of Lake Ladoga: Stages of Ecosystem Transformation, 1975–2004,” Ekol. Khimiya 14 (2005). 21. K. R. Helyar, D. N. Munns, and R. G. Burau, “Adsorp tion of Phosphate by Gibbsite: II. Formation of a Surface Complex Involving Divalent Cations,” J. Soil Sci. 27 (3), 315–323 (1976). 22. R. L. Parfitt, A. R. Fraser, J. D. Russell, and V. C. Far mer, “Adsorption on Hydrous Oxides. II. Oxalate, Ben 2013
304
23. 24.
25.
26. 27. 28.
29. 30.
31. 32.
33. 34.
35.
36.
37. 38. 39.
KUDEYAROVA zoate and Phosphate on Gibbsite,” J. Soil Sci. 28 (1), 40–47 (1977). A. J. Herbillon, “Introduction to the Surface Charge Properties of Iron Oxides and Oxidic Soils,” in Iron in Soils and Clay Minerals (Reidel, Dordrecht, 1988). U. K. Saha, S. Hiradate, and K. Inoue, “Retention of Phosphate by Hydroxyaluminosilicate and Hydroxy aluminum–Montmorillonite Complexes,” Soil Sci. Soc. Am. J. 62 (4), 922–929 (1998). D. H. Bailar and J. C. Busch, “A General Review of Coordination Compounds,” in The Chemistry of Coor dination Compounds (Reinhold, New York, 1958), pp. 7–86. F. Basolo and R. Pearson, R., Mechanisms of Inorganic Reactions. A Study of Metal Complexes in Solution, (Wiley, New York, 1967). N. A. Kostromina, V. N. Kumok, and N. A. Skorik, Chemistry of Coordination Compounds (Vyssh. shkola, Moscow, 1990) [in Russian]. A. Yu. Kudeyarova, “Ligand Exchange as the Main Mechanism of Phosphate Ion Adsorption on the Surface of Soil Solid Phases,” Agrokhimiya, No. 5, 125–137 (1991). S. Kuo, “Concurrent Sorption of Phosphate and Zinc, Cadmium, or Calcium by a Hydrous Ferric Oxide,” Soil Sci. Soc. Am. J. 50 (6), 1412–1419 (1986). W. Stumm, R. Kummert, and L. Sigg, “A Ligand Exchange Model for the Adsorption of Inorganic and Organic Ligands at Hydrous Oxide Interfaces,” Croat. Chem. Acta 53, 291–312 (1980). C. J. B. Mott, “Anion and Ligand Exchange,” in The Chemistry of Soil Processes (Wileys, Chichester–New York, 1981). J. R. McLaughlin, J. C. Ryden, and J. K. Syers, “Sorp tion of Inorganic Phosphate by Iron and Aluminum Containing Components,” J. Soil Sci. 32, 365–377 (1981). J. B. Harsh and H. E. Doner, “The Nature and Stability of Aluminum Hydroxide Precipitated on Wyoming Montmorillonite,” Geoderma 36 (1), 45–56 (1985). M. Okazaki, K. Sakaidani, T. Saigusa, and N. Sakaida, “Ligand Exchange of Oxyanions on Synthetic Hydrated Oxides of Iron and Aluminum,” Soil Sci. Plant Nutr. 35 (3), 337–346 (1989). B. Manning and S. Goldberg, “Modeling Competitive Adsorption of Arsenate with Phosphate and Molybdate on Oxide Minerals,” Soil Sci. Soc. Am. J. 60 (1), 121– 131 (1996). S. Goldberg and G. Sposito, “On the Mechanism of Specific Phosphate Adsorption by Hydroxylated Min eral Surfaces: A Review,” Commun. Soil Sci. Plant Anal. 16 (8), 801–821 (1985). J. Gerke, “Phosphate Adsorption by Humic/FeOxide Mixtures Aged at pH 4 and 7 and Poorly Ordered Fe Oxide,” Geoderma 59 (1–4), 279–288 (1993). G. Yuan and L. M. Lavkulich, “Phosphate Sorption in Relation To Extractable Iron and Aluminum in Spodo sols,” Soil Sci. Soc. Am. J. 58 (2), 343–346 (1994). M. P. F. Fontes and S. B. Weed, “Phosphate Adsorption by Clays from Brazilian Oxisols, Relationships with Spe cific Surface Area and Mineralogy,” Geoderma 72 (1– 2), 37–51 (1996).
40. E. OwusuBennoah, C. Szilas, H. C. B. Hanson, and O. K. Borggaard, “Phosphate Sorption in Relation to Aluminum and Iron Oxides of Oxisols from Ghana,” Commun. Soil Sci. Plant Anal. 28 (9–10), 685–697 (1997). 41. M. Arias, J. Da SilvaCarballal, L. GarciaRio, J. Mejuto, and A. Nunez “Retention of Phosphorus by Iron and AluminumOxidesCoated Quartz Particles,” J. Colloid Interface Sci. 295, 65–70 (2006). 42. T. A. Sokolova and S. Ya. Trofimov, Sorption Properties of Soils. Adsorption. Cation Exchange (Universitetskaya kniga, Moscow, 2009) [in Russian]. 43. A. A. Grinberg, Introduction to the Chemistry of Complex Compounds (Khimiya, Leningrad, 1971). 44. S. S. Rajan and K. W. Perrott, “Phosphate Adsorption by Synthetic Amorphous Aluminosilicates,” J. Soil Sci. 26 (3), 257–266 (1975). 45. V. E. Berkheiser, J. J. Street, P. S. Rao, and T. L. Yuan, “Partitioning of Inorganic Orthophosphate in Soil Water Systems,” CRC Crit. Rev. Environ. Control 10 (3), 179–224 (1980). 46. R. E. White and A. W. Taylor, “Reactions of Soluble Phosphate with Acid Soils: the Interpretation of Adsorp tionDesorption Isotherms,” J. Soil Sci. 28 (2), 314–328 (1977). 47. W. H. Van Riemsdijk, L. J. M. Boumans, and F. A. M. De Haan, “Phosphorus Sorption by Soils: I. A. Model of Phosphate Reaction with Metal Oxides in Soils,” Soil Sci. Soc. Am. J. 48, 537–541 (1984). 48. M. Nanzyo, “Diffuse Reflectance Infrared Spectra of Phosphate Sorbed on Alumina Gel,” J. Soil Sci. 35 (1), 63–69 (1984). 49. R. R. Villapando and D. A. Graetz, “Phosphorus Sorp tion and Desorption Properties of the Spodic Horizon from Selected Florida Spodosols,” Soil Sci. Soc. Am. J. 65, 331–339 (2001). 50. J. A. Veith and G. Sposito, “Reactions of Aluminosili cates, Aluminum Hydrous Oxides, and Aluminum Oxide with oPhosphate: The Formation of XRay Amorphous Analogs of Variscite and Montebrasite,” Soil Sci. Soc. Am. J. 41 (5), 870–876 (1977). 51. M. Laverdiere, “The Effects of Phosphate Additions on the Charge Properties of a Podzolic B Horizon,” Can. J. Soil Sci. 62, 519–525 (1982). 52. J. C. Ryden, J. K. Syers, and J. R. McLaughlin, “Effects of Ionic Strength on Chemisorption and Potential Determining Sorption of Phosphate by Soils,” J. Soil Sci. 28 (1), 62–71 (1977). 53. J. C. Ryden, J. R. McLaughlin, and J. K. Syers, “Mech anisms of Phosphate Sorption by Soils and Hydrous Fer ric Oxide Gel,” J. Soil Sci. 28 (1), 72–92 (1977). 54. W. H. Van Riemsdijk and J. Lyklema, “Reaction of Phos phate with Gibbsite (Al(OH)3) Beyond the Adsorption Maximum,” J. Colloid Interface Sci. 76 (1), 55–66 (1980). 55. N. S. Bolan, N. J. Barrow, and A. M. Posner, “Describ ing the Effect of Time on Sorption of Phosphate by Iron and Aluminum Hydroxides,” J. Soil Sci. 36 (2), 187– 197 (1985). 56. B. A. Linquist, P. W. Singleton, R. S. Yost, and K. G. Cassman, “Aggregate Size Effects on the Sorption and Release of Phosphorus in an Ultisol,” Soil Sci. Soc. Am. J. 61, 160–166 (1997).
GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
2013
TREND AND MECHANISMS OF TRANSFORMATION OF NATURAL SORPTION 57. J. Torrent, V. Barron, and U. Schwertmann, “Phosphate Adsorption and Desorption by Goethites Differing in Crystal Morphology,” Soil Sci. Soc. Am. J. 54 (4), 1007– 1012 (1990). 58. J. R. Van Wazer, Phosphorus and its Compounds (Inter science, New York, 1958). 59. L. M. Lapina and I. A. Grishina, “Solubility of Alu moammonium Phosphates in the Ammonium Phos phate Solution,” in Mineral Fertilizers and Sulfuric Acid (NIUIF, Moscow, 1973), No. 221 [in Russian]. 60. D. Corbridge, Phosphorus. An Outline of Its Chemistry, Biochemistry, and Technology, (Elsevier, Amsterdam, 1980). 61. J. D. Beaton, J. M. Bryan, J. C. Russell, and R. C. Speer, “Reaction of Several Clay Minerals and Gibbsite with Monoammonium and Monopotassium Phosphate,” Nature 201 (4920), 739–740 (1964). 62. A. Yu. Kudeyarova and A. I. Trubin, “Study of the Inter action Products of Double Superphosphate with Sod– Podzolic Soil,” Pochvovedenie, No. 11, 108–118 (1976). 63. A. Yu. Kudeyarova, “Participation of Phosphorus Fertil izers in Phosphorus Cycle,” in Ecological Problems of Fertilizer Application (Nauka, Moscow, 1984), pp. 61– 124 [in Russian]. 64. G. M. Pierzynski, “The Chemistry and Mineralogy of Phosphorus in Excessively Fertilized Soils,” Crit. Rev. Environ. Control 21 (3), 4 (1991). 65. M. J. Wilson and D. C. Bain, “Occurrence of Leucoph osphite in a Soil from Elephant Island, British Antarctic Territory,” Am. Mineral. 61, 1027–1028 (1976). 66. R. H. Qureshi, D. A. Jenkins, and R. I. Davies, “Elec tron Probe Microanalytical Studies of Phosphorus Dis tribution within Soil Fabric,” Soil Sci. Soc. Am. J. 42 (5), 698–703 (1978). 67. A. V. Savenko, G. N. Baturin, and S. V. Golubev, “Cran dallite Solubility in Aqueous Solutions,” Geochem. Int. 43 (11), 1135–1137 (2005). 68. A. Yu. Kudeyarova and M. Z. Kvaratskheliya, “Influence of Fertilizer Phosphates on the Solubility of Ca, Mg, Mn, Zn, Al and Fe Compounds of Gray Forest Soil,” Pochvovedenie, No. 8, 26–34 (1989). 69. M. K. Zhang, Z. L. He, D. V. Calvert, and P. J. Stoffella, “Colloidal Iron Oxide Transport in Sandy Soil Induced by Excessive Phosphorus Application,” Soil Sci. 168 (9), 617–626 (2003). 70. A. Yu. Kudeyarova and A. I. Trubin, “Synthesis of Alu minum Phosphate Minerals Depending on pH and Phosphorus and Aluminum Contents,” Pochvovedenie, No. 5, 137–141 (1978). 71. J. Murphy and J. P. Riley, “A Modified Single Solution Method for Determination of Phosphate in Natural Waters,” Anal. Chim. Acta 27, 31–36 (1962). GEOCHEMISTRY INTERNATIONAL
Vol. 51
No. 4
305
72. E. V. Arinushkina, Manual on Chemical Analysis of Soils (Mosk. Gos. Univ., Moscow, 1970). 73. I. S. Kaurichev and E. M. Nozdrunova, “Measurement of Migration of Some Compounds in Soil Using Lysim etric Chromatography,” Pochvovedenie, No. 12, 30–35 (1960). 74. O. Samuel’son, IonExchange Separation in Analytical Chemistry (Khimiya, Moscow, 1966) [in Russian]. 75. A. A. Lur’e, “Determination of Phosphorus in Anion ites, Plant Materials, and Soils with Calcination Using Chloric Acid,” Agrokhimiya, No. 11, 80–86 (1964). 76. A. Yu. Kudeyarova, “Use of Electron Spectroscopy to Reveal Structural Differences between Humus Acids from Virgin and Arable Gray Forest Soils,” Eur. Soil Sci ence, 41 (9), 953–964 (2008). 77. A. Yu. Kudeyarova, “Electron Spectroscopy as Method for Studying the Influence of Anions of Mineral Acids and Fertilizers on the Structural Characteristics of Humic Acids,” Agrokhimiya, No. 11, 71–84 (2007). 78. L. J. Bellami, InfraRed Spectra of Complex Molecules, (Inostrannaya Literatura, Moscow, 1963). 79. E. E. Nifant’ev, Chemistry of Organophosphorous Com pounds (MGU, Moscow, 1971) [in Russian]. 80. B. A. Dospekhov, Method of Field Experiment (Kolos, Moscow, 1965) [in Russian]. 81. P. F. Rokitskii, Biological Statistics (Vyssh. Shkola, Minsk, 1967) [in Russian]. 82. J. R. Lehr, E. H. Brown, A. W. Frazier, J. P. Smith, and R. D. Thrasher, Crystallographic Properties of Fertilizer Compounds (National Fertilizer Development Center, Muscle Shoals, 1967). 83. E. K. Vasil’ev, G. M. Kashaeva, and Z. F. Ushapovskaya, Xray Detector of Minerals (Nauka, Moscow, 1974). 84. V. I. Mikheev, XRay Determination of Minerals (Gos geoltekhizdat, Moscow, 1957) [in Russian]. 85. K. Ingol’d, Theoretical Principles of Organic Chemistry (Mir, Moscow, 1973) [in Russian]. 86. A. Yu. Kudeyarova and M. Z. Kvaratskheliya, “Contri bution of Inorganic Polyphosphates to Migration of Ele ments in a Grey Forest Soil, U.S.S.R.,” Geoderma 34, 251–259 (1984). 87. R. F. Hudson, Structure and Mechanism in Organic Phosphorus Chemistry (Academic Press, London, 1965). 88. O. V. Sverdlova, Electronic Spectra in Organic Chemistry (Khimiya, Leningrad, 1985). 89. R. Brested and V. Kuli, “Physical Methods in Coordina tion Chemistry,” in The Chemistry of Coordination Com pounds (Reinhold, New York, 1958), pp. 474–525. 90. S. B. Savvin and E. L. Kuzin, Electronic Spectra and Structure of Organic Reagents (Nauka, Moscow, 1974). 91. A. N. Nesmeyanov, Studies in Organic Chemistry (Nauka, Moscow, 1971) [in Russian]. 2013