The solubilization of phosphates in the presence of organic acids

Loess based soils of West Tennessee are insensitive to the Mehlich I soil P test. A more accurate assessment of plant available P may be obtained using organic acids produced in the rhizosphere. Two such organic acids, citric acid and 2-ketogluconic acid, have been shown to effectively solubilize soil P. The objectives of this study were to: 1) develop extraction procedures that utilize citric and 2-ketogluconic acid and that maximize P solubilization; 2) compare extraction capabilities of the organic acid extraction procedures to those of the Mehlich III and Olsen bicarbonate methods; 3) correlate organic acid extractable P to P solid phase speciation; 4) correlate corn (Zea mays L.) yield and ear leaf P concentration to organic acid extractable P, Mehlich III-P, Olsen-P, and to the inorganic P fractions of the soil solids; and 5) investigate the mechanism of soil P solubilization by 2-ketogluconic acid. Two extraction procedures were found to maximize P solubility from a loess based soil: 1) 10 g soil sample extracted with 25 ml of a 10 mM citric acid, pH 3.5 solution for 3 hrs (OEl), and 2) 5 g soil sample extracted with 25 ml of a 5 mM citric acid + 5 mM 2-ketogluconic acid, pH 3.5 solution for 6 hrs (0E2). Soil samples were obtained from a study that investigated the use of swine manure as a P fertilizer source (at Ames Plantation Experiment Station) and from a study that investigated the long-term influence of tillage practice and fertilizer P rates on com production (at Milan Experiment Station). Both Experiment Stations are located in West Tennessee. In the Ames soil, OEl was significantly correlated with Mehlich III-P (r = 0.84***), Olsen-P (r = 0.50*), and Al-P (r = 0.85***), but OE2 was only correlated with Mehlich III-P (r = 0.74***) and Al-P (r = 0.74***). In the Milan soil, OEl and OE2 were significantly correlated with Mehlich III-P, Olsen-P, Al-P, and Fe-P. None of the available P indices or inorganic P fractions were significantly correlated with plant P from the Ames or Milan studies. In the Milan soil, com yield was significantly correlated with OEl-P (r = 0.68**), OE2-P (r = 0.68**), Mehlich III-P (r = 0.70**), Olsen-P (0.68*), Al-P (0.67*), and Fe-P (r = 0.64*). All of the P indices were poorly correlated with yield in the Ames soil, perhaps due to the relative abundance of P in the organic P phase. Therefore, all four extractants inadequately assessed plant available P in the Ames and Milan soils. The ability of 2-ketogluconic acid to solubilize soil P may be attributed to the formation of soluble Al³⁺ and Fe³⁺ complexes and to competition with phosphates for adsorption sites. The former mechanism was investigated by examining the equilibrium solubility of gibbsite as a function of 2-ketogluconic acid concentration, pH, and ionic strength. The solubility results suggest the formation of [Al(C₆H₉O₇)²⁺], [Al(OH)(C₆H₉O₇)⁺], [Al(OH)₂(C₆H₉O₇)⁰], and [Al(OH)₂(C₆H₉O₇)^-], with log K values of 3.0, -1.1, -5.2, and - 10.6, respectively. The formation of these soluble complexes would promote the dissolution of aluminum phosphates (and possible iron phosphates), thereby releasing the soil P into the extracting solution.