Phosphorus availability in a low pH highly weathered soil as affected by added salts

Disponibilidade de fósforo num solo ácido afetada pela aplicação de sais

PHOSPHORUS AVAILABILITY IN A LOW pH HIGHLY WEATHERED SOIL AS AFFECTED BY ADDED SALTS

DISPONIBILIDADE DE FÓSFORO NUM SOLO ÁCIDO AFETADA PELA APLICAÇÃO DE SAIS

Paulo Roberto Ernani¹ Stanley Arthur Barber²

SUMMARY

Concentration and identity of cations and anions in the soil solution may affect soil P reactions and thus P availability. The magnitude of these reactions was evaluated in this research after application of various salts to a highly weathered low pH soil. Chloride, nitrate, and sulfate salts of Na, NH₄, K, Ca, Mg, Sr, or Cu were added to the soil after addition of 360mg P/kg trying to simulate ion concentrations around granules of fertilizers in the soil. Thirty days later, P was determined in the soil solution (P_li) and on the solid phase (P_si). The soil samples of some treatments were leached with water and three days later a new soil solution was displaced. Separate addition of all salts increased P_li, except NaCl at the lowest rate. The increase of P_li was highly associatcd with amount of native cations displaced to the soil solution by the applied salts. Salt solubility, concentration, and sometimes identity of cation and anion also influenced P_li. Some salts decreased P_si, but this was not correlated with any soil property measured. The effects caused by salts on P_li and P_si disappeared after leaching the soil samples.

Key words: acid soils, phosphorus reactions, ionic strength.

RESUMO

A concentração eletrolítica e o tipo de cations e anions da solução do solo podem afetar as reações do fósforo com possíveis reflexos na disponibilidade de P aos vegetais. Nessa pesquisa quantificou-se o efeito de vários sais nos valores das determinações analíticas que afetam a disponibilidade de fósforo. Sais de nitrato, cloreto e sulfato foram aplicados a amostras de um alfisol ácido após a aplicação de 360mg P/kg, simulando concentrações que ocorrem no solo ao redor de grânulos de fertilizantes. Fósforo lábil (P_si) e P na solução do solo (P_li) foram determinados após 30 dias de incubação, antes e depois de percolar água pelo solo. Todos os sais aumentaram a concentração de P na solução do solo, exceto a menor dose de NaCl. O aumento do P_li foi correlacionado com a quantidade de cations originalmente no solo deslocados para a solução do solo. Solubilidade, concentração, e a espécie dos cations e anions aplicados também exerceram efeito no P_li. Alguns sais diminuíram o P_si porém esse decréscimo não se correlacionou com nenhuma determinação efetuada. A percolação de água eliminou os efeitos ocasionados pelos sais nos valores de fósforo.

Palavras-chave: solos ácidos, força iônica, fósforo.

INTRODUCTION

Research with a mechanistic nutrient uptake model has shown that phosphorus concentration in the soil solution (P_li) is the soil supply variable most correlated with supply of P to plant roots (CHEN & BARBER, 1990). Thus, in order to increase the effectiveness of P applied to the soil it is essential to understand soil reactions that affect P_li. Solution P is normally controlled by adsorption and precipitation reactions. Adsorption of P is affected by the surface potential (CURTIN et al., 1987), by the surface charge (BOWDEN et al., 1980), and by the extent of the diffuse double layer, DDL, (BAR-YOSEF et al., 1988), which, in turn, are affected by the electrolyte concentration of the soil solution (RYDEN et al., 1977; BAR-YOSEF et al., 1988).

Electric surface potential (BOWDEN et al., 1980) and the extent of DDL (GAST, 1977) decrease with an increase of the electrolyte concentration of the soil solution. Surface charge is affected by addition of salts to soils with predominantiy variable charge, and depends on changes of soil pH and formation of inner or outer sphere complexes (SPOSITO, 1984).

The overal effect of changes of electrolyte concentration on adsorption of P varies according to pH (BARROW et al,, 1980; BOLAN et al., 1986; ERNANI & BARBER, 1991) and nature of soil charge (permanent and pH-dependent). ERNANI & BARBER (1991) have shown that addition of KCl decreased P in the soil solution in less weathered soils, and in highly weathered soils with moderate and high pH, but increased P_li in highly weathered soils with low pH. In addition to electrolyte concentration, the identity of cations (LEHR & VAN WESEMAEL, 1952; HELYAR et al., 1976; BARROW & SHAW, 1979; SMILLIE et al., 1987) or anions (LEHR & VAN WESEMAEL, 1952) may also have a significam effect on P_li, particularly in recently fertilized soil zones.

Soils with low pH and predominantly variable charge are normally deficient in P in their natural state. In these soils, supply of P to plant roots occurs mainly from the fertilized sites in which P reactions with the soil may differ from those in the unfertilized soil.

The objective of this research was to determine the effect of various salts on soil determinations that control the supply of P to the plant roots in a low pH highly weathered soil, before and after leaching the soil samples.

MATERIAL AND METHODS

Solutions of various chiorides, nitrates, and sulfates salts were applied at two rates (1.000mg of cation/kg or 25mmol of salt/kg) to duplicate 650g soil samples (oven-dry basis) after addition of 360mg P/kg as monocalcium phosphate solution. A treatment with P but without salts was used as the control. Calcium sulfate was applied in the solid form because its low solubility.

The salts, the respective rates, and the amount of cation and anion added by each treatment are shown in Table 1. The soil samples used carne from the B horizon of Rarden soil (clayed, mixed, mesic, Aquic Hapludalfs), located in the State of Indiana (USA), that had pH = 4.6, exchangeable aluminum = 5.1cmol/kg, anion-exchangeable resin (P_si) = 0.076mmol/l and P in the soil solution (P_li) = 0.00094mmol/l.

After apphcation of P and salts the treated soil samples were incubated at 25°C and -25kPa of water tension for 30 days. Phosphorus, Ca, Mg, K, and Al were determined in the soil solution and on the solid phase after the incubation time. Some treated soil samples were leached with destilled water in a volume equivalent to the total soil porosity, and three days later the soil solution was displaced again from these samples using the same procedure.

Soil solution was displaced with destilled water from 450g of soil packed in vertical columns at a density of 1.1g/cm³. Labile P (P_si) was extracted with anion-exchange resin. Both forms of P (P_si and P_li) were determined by the method described by MURPHY & RILEY (1962). Calcium, Mg, and K were extracted from 2.5g of soil with 50ml of 1.0M ammonium acetate solution buffered at pH 7.0, and Al with 25ml of 1.0M KCl solution. It was used atomic absorption spectrometry for determining Ca and Mg, flame emission for K, and titration with NaOH for Al. Solution Al was determined by a cathecol violet method (MOSQUERA & MOMBIELLA, 1986).

RESULTS AND DISCUSSION

Addition of salts increased the concentration of P in the soil solution, except NaCl at the lowest rate (Table 2). The increase of P_li was associated with the degree applied cations displaced native cations from the soil adsorption sites. Values of P_li, taken as a natural logarithm, were highly correlated with the amounts of K, Ca, Mg, and Al in the soil solution determined after application of the salts (r = 0.84**) (Table 3). When values for the treatments with salts containing sulfate were dropped from the regression analysis, the coefficient of correlation between solution cations and P_li increased to 0.95** (Table 3). The smaller correlation obtained when sulfate salts (CaSO₄.2H₂O and K₂SO₄) were kept in the regression analysis is probably due to formation of ion-pairs among sulfate and metallic cations, which decreases the ability of the accompanying cations to displace cadons from negativo sites (RAO et al., 1968).

The increase of P in the soil soludon (P_li) that occurred after addition of salts can be explained in two ways in this soil. First, by assuming that P_li is controlled to a larger extent by the surface negativo charge than by the surface potential or by precipitation reactions. Thus, the increase of the negativo charge that occurs after addition of salts would restrict P adsorption. Second, by assuming that P is preferentially adsorbed by sites with the more positive potential, as proposed by BARROW & ELLIS (1986). In the second case, P adsorption would be restricted by a decrease of the positive surface potential that occurs after addition of salts. The negativo potential also decreases after sait addition but the change in positive potential would exert the largest effect.

The effect of the diffuse double layer extent (DDL) on P adsorption was less important than the effects of surface charge or potential because Na was the cation that gave the least increase in P_li. If the extent of DDL have had a significant effect on P adsorption, the increase of Py caused by Na should be greater because Na prometes a thicker DDL relative to other cations, and as DDL becomes ticker it is more difficult for P to approach the adsorption surface.

The small effect of NaCl on P_li shows also that P was adsorbed via ligand exchange reactions because addition of 908mg of chioride per kg^-1 from this sait had no effect on P_li (Tables 1 and 2). In addition, it shows that the capacity of cations to occupy the negative sites at the soil particles was more important than the increase of electric conductivity (EC) of the soil solution caused by each sait because NaCl was one of the salts that increased EC most, despite its small effect on P_li (Tables 1 and 2).

The ability of each sait to increase P_li was associated with identity of cation, and with sait and cation concentration (Table 2). For chioride salts, added at 25.6mmol/kg, the effect on P_li followed the sequence Sr > Ca > NH₄ = K > Na (Table 2); for K salts, added at the same molar concentration, K₂SO₄ that contained twice as much K than KCl or KNO₃ had me greatest effect (Tables 1 and 2); Mg (NO₃)₂ and Ca (NO₃)₂ increased P_li similarly despite Ca being applied at twice that of Mg (Tables 1 and 2). Precipitation of P, however, is more probable with Ca than with Mg (LINDSAY, 1979), and this reaction may counteract the increase of P_li caused by addition of salts.

The effect of the accompanying anion on P_li was inconsistent. The increase of P_li caused by potassium or ammonium salts was similar for chloride, nitrate, or sulfate (Table 2). When nitrate was applied at twice that of chloride (Table 1) there was no difference on P_li for these ammonium salts but there was for these calcium salts (Table 2).

The effect of salts on resin-exchangeable P (P_si) was less pronounced and in opposite direction to that on P_li (Table 2). At a rate of 1.0g of cation/kg of soil, only NH₄Cl, CaCl₂, and CaSO₄ decreased P_si; at 25.6mmol of salt/kg, K₂SO₄, CaSO₄, CaCl₂, and SrCl₂ decreased P_si (Table 2). Calcium sulfate had a small effect on P_li but caused a significant decrease on P_si. On the other hand, some salts that caused a large increase on P_li such as KNO₃, KCl, CuCl₂, NH₄Cl, and NH₄NO₃ did not affect P_si (Table 2). The highest rates of K₂SO₄ and NH₄Cl, however, that had the largest effect on P_li, also caused the greatest decrease of P_si (Table 2). Resin-exchangeable P measured after addition of salts was not significantiy associated with any soil determination (Table 3).

The effects caused by sait application on P_si and on P_li disappeared after leaching the soil samples, even with a small volume of water, equal to the total soil porosity (data not shown).

As discussed by ERNANI & BARBER (1991), the increase of P_li caused by the increase of the electrolyte concentration, despite its temporary nature, would be important at the beginning of the growth season in soils deficient in P.

2Distinguished Prof. of Agronomy, Purdue University, West Lafayette, IN, USA.

Recebido para publicação em 19.07.94. Aprovado em 12.04.95

• BARROW, N.J., SHAW, T.C. Effects of ionic strength and nature of the cation on desorption of phosphate from soil. Journal of Soil Science, Oxford, v. 30, p. 53-65, 1979.
• BARROU, N.J., BOWDEN, J.W., POSNER, A.M. et al. Describing the effects of electrolyte on adsorption of phosphate by a variable charge surface. Australian Journal of Soil Research, v. 18, p. 395-404, 1980.
• BARROW, N.J., ELLIS, A.S. Testing a mechanistic model. The points of zero sait effect for phosphate retention, for zinc relention and for acid/alkali titration of a sofl. Journal of Soil Science, Oxford, v. 37, p 303-310, 1985.
• BAR-YOSEF, B., KAFKAFI, U., ROSENBERG, R. et al. Phosphorus adsorption by kaolmite and montmorillonite: I. Effect of time, ionic strength, and pH. Soil Science Sodety of América Journal, Madison, v. 52, p. 1580-1585, 1988.
• BOLAN, N.S., SYERS, J.K., TILLMAN, R.W. Ionic strength effects on surface charge and adsorption of phosphate and sulfate by soik. Journal of Soil Science, Oxford, 37: p. 379-388, 1986.
• BOWDEN, J.W., POSNER, A. M., QUIRK, J. P. Adsorption and charging phenomena in variable charge soils. In: THENG, B.K.G. (ed). Seils with variable charge Palmerston North, New Zealand, 1980.
• CHEN, J.H., BARBER, S.A. Effect of soil pH on P and K uptake by maize as evaluated with a mechanistic uptake model. Soil Science Soriety of America Journal, Madison, v. 54, p. 1032-1036, 1990.
• CURTIN, D., SYERS, J.K., SMILLE, G.W. The importance of exchangeable cations and resin-sink characteristics in the release of soil phosphorus. Journal of Soil Science, Oxford, v. 38, p. 711-716, 1987.
• ERNANI, PH., BARBER, S.A. Predicted soil phosphorus uptake as affected by banding potassium chloride with phosphorus. Soil Science Society of América Journal, Madison, v. 55, p. 534-538, 1991.
• GAST, R.G. Surface and colloid chemistry. In: DINAUER, R.D. (ed). Minerais in soil environments ASA, CSSA, & SSSA, Madison, WI, 1977.
• HELYAR, K.R., MUNNS, D.N., BURAU, R.G. Adsorption of phosphate by gibbsite. I. Effects of neutral chloride salts of calcium, magnesium, sodium, and potassium. Journal of Soil Science, Oxford, v. 27, p. 307-314, 1976.
• LEHR, J.J., VAN WESEMAEL, J.C. The influence of neutral salts on the solubility of soil phosphate with special reference to the effect of the nitrates of sodium and calcium. Journal of Soil Science, Oxford, v. 3, p. 125-135, 1952.
• LINDSAY, W.L. Chemical equilibria in soils New Yoik: Wfley-Interscience, 1979.
• MOSQUERA, A., MOMBDELLA, F. Comparison of three methods for the determination of aluminum in an unbuferred sait extract. Communications in Soil Science and Plant Analysis, Monticello, v. 17, p. 97-113, 1986.
• MURPHY, J., RDLEY, J.D. A modifiedd single solution naethod for the determination of phosphate in natural waters. Analytica Chemica Acta, Amsterdam, v. 27, p. 31-36, 1962.
• RAO, T.S., PAGE, A.L., COLEMAN, N.T. The influence of ionic strength and ion-pair formation between alkaline earth metals and sulfate on Nadivalent cation-exchange equlibria. Soil Science Sodety of America Proceedings, Madison, v. 32, p. 639-643, 1968.
• RYDEN, J.C., SYERS, J.K., McLAUGHLIN, R. Effects of ionic strength on chemisorption and potential-determining sorption of phosphate by soils. Journal of Soil Science, Oxford, v. 28, p. 62-71, 1977.
• SMILLIE, G.W., CURTIN, D., SYERS, J.K. Influence of exchangeable calcium on phosphate retention by weakly acid soils. Soil Science Society of America Journal, Madison, v. 51, p. 1169-1172, 1987.
• SPOSITO, G. The surface chemistry of soils New York: Oxford University Press, 1984.

1

PhD, Professor da Faculdade de Agronomia da Universidade do Estado de Santa Catarina (UDESC), Caixa Postal 281, Lages-SC, Brasil. Pesquisador do CNPq. Author for correspondence.

Publication Dates

• Publication in this collection
  20 Oct 2009
• Date of issue
  1995

History

• Accepted
  12 Apr 1995
• Received
  19 July 1994