Search

Summary Phosphate was adsorbed at a constant pH, on allophane clay samples which had been pre-treated with selenite to remove adsorbed silicate. The concomitant release of selenium, silicon, and hydroxyl ion on phosphate adsorption was also measured. The results are compared with those obtained from a similar experiment conducted using clay which had not been treated and they substantiate the earlier conclusion that phosphate displaces structural silicon of clays. The amount displaced increases parabolically with increasing levels of phosphate adsorbed over the concentration range studied.

Summary The two main motivations for studying equations which describe adsorption of phosphate are to understand the processes involved, and to summarize many results by a few numbers. If more than one hypothesis about the process is tenable, appropriate statistical procedures should be used both to choose between them and to obtain the best summary of the results with a given equation. Equations most likely to be successful in describing adsorption are those in which the affinity for adsorption decreases as the amount of adsorption increases. This effect is inherent in the process of phosphate adsorption and occurs because specific adsorption of anions increases the negative charge on the adsorbing surface. It is included in the complex equations of Bowden, and in simpler fashion in the Freundlich equation. Because the simpler equation can only approximately describe the true situation, a perfect fit over a wide range of concentrations should not be expected. Equations for which the affinity for adsorption is constant — such as the Langmuir equation — are not consistent with our knowledge of the adsorption process. This difficulty is not avoided by the multi-surface Langmuir equations: such equations may also be difficult to justify on statistical grounds.

Summary The amounts of hydroxyl ions released from hydrous alumina during phosphate adsorption were measured at constant pH values and at different phosphate concentrations using an automatic titrator. The hydroxyl ions released were plotted against phosphate adsorbed and the equations that gave best fit to the data were differentiated. The variation in the differentials (dOH/dP) with increasing phosphate adsorption was consistent with the theory, proposed below, on the type of sites responsible for phosphate adsorption, by synthetic hydrous alumina containing bayerite and pseudoboehmite, at different concentrations. At low phosphate adsorption which corresponded to low phosphorus concentration, adsorption took place mainly by displacing aquo groups (Al-H2O). With increase in phosphorus concentration hydroxo groups (Al-OH) became the predominant sites of adsorption. At still higher concentrations, where phosphate adsorption displacing aquo and hydroxo groups was nearly completed, the hydroxyl bridges linking aluminium atoms (Al-OH-Al, ol groups) were broken. This created new sites which were responsible for additional adsorption of phosphate at these concentrations.

Summary Measurements of equilibrium phosphate concentrations, adsorbed phosphate and isotopically exchangeable phosphate (Et) were made on suspensions of acid soils at initial P concentrations ranging from 1–1000 μM and pH values between 4.2 and 6.8. At low P concentrations (1–100 μM), the affinity of the adsorption reaction was greatest at pH 5.2–5.5. This result, which was inconsistent with the ligand-exchange hypothesis of Kingston et al. (1967, 1972, 1974), was interpreted in terms of the formation of basic aluminium phosphates of varying composition that depended on the OH/P ratio in solution. At the highest P concentrations (1000 μM), when the affinity of adsorption was least at pH 5.2–5.5, it appeared that basic calcium phosphates were precipitating at pH values > 5.5. In suspensions containing 1–100 μM P initially, values of Et did not change consonantly with values of adsorbed P over the pH range 4.8 to 5.5. It is suggested that the Et values were underestimated due to the presence of labelled complexes in solution which were detected by scintillation counting but not by the analysis for orthophos-phate. Evidence from solution studies (White et al., 1976) suggested that as much as one-third of the phosphate could be complexed with aluminium in solutions between pH 4.7 and 5.4 at P/A1 mole ratios of 0.075 to 7.5. It was apparent that the interpretation of the physical significance of E values in acid soils is complex and requires careful appraisal of the experimental techniques used.

Summary Soils from Denmark and Tanzania were extracted with ammonium acetate (controls), EDTA to dissolve amorphous iron oxides, and dithionite-EDTA (DE) to dissolve crystalline iron oxides. The phosphate adsorption capacities of the extracted soils were taken as the maximum quantity of phosphate adsorbed computed from the Langmuir equation. The decreases in the phosphate adsorption capacity following EDTA extraction and DE extraction were attributed to the removal of iron oxides. Close correlations (P

Summary For forty-one soils (pH > 5·0) from southern England and eastern Australia, the Langmuir equation was an excellent model for describing P adsorption from solutions < 103 M P, if it was assumed that adsorption occurs on two types of surface of contrasting bonding energies. For most of these soils, which were relatively undersaturated with P, this equation may be written as

Summary The effects were measured of varying the solution: soil ratio, and the vigour of shaking on the rate of adsorption of phosphate. One soil appeared to break down with vigorous shaking, prior to adding phosphate, especially at low solution: soil ratio. As a result, subsequent phosphate adsorption was faster. With gentle shaking, there was no effect of solution: soil ratio with the soil. Another soil was more stable and there were only small differences due to vigour of shaking and no effect of solution: soil ratio with any method of shaking. The change in phosphate concentration was not proportional to the reciprocal of time as reported by Ryden and Syers (1975) and Hope and Syers (1976). Hence the validity of extrapolating from short sections of reciprocal-time graphs in order to estimate equilibrium is questioned.

Summary Phosphate adsorption data for 24 soils showed that although the Langmuir equation gave a good fit, there was significant curvature. Of the alternative equations examined, that of Freundlich provided a better description than that of Langmuir in 22 soils, but the best fit, accounting for more than 99·8 per cent of the variation in phosphate adsorption, was given by introducing a square-root term: where y= adsorbed phosphate, c= equilibrium concentration of phosphate in solution, B, A, and D are constants. The phosphate adsorption maximum, obtained from I/B, gave values in the range 325 to 1427 μg P/g soil which were 1·39 to 2·40 times those obtained from the Langmuir equation. The inclusion of a square-root term in the Langmuir equation could be taken to imply that the soil will adsorb a little phosphate firmly, a slightly greater amount of phosphate less firmly, and so on until a limiting value is reached when all the components of the phosphate adsorption system are saturated.