Studies of the relative importance of iron and aluminium in the sorption of phosphate by some Australian soils

Abstract

Phosphate sorption by 47 acid surface soils was correlated with the amounts of iron and aluminium extracted from them by ammonium oxalate at pH 3.2. The correlation coefficients, when all soils were considered, were highest for sorption and aluminium. For certain groups of soils, iron or iron plus aluminium gave higher coefficients than did aluminium. The relative contributions of iron and aluminium to phosphate sorption could not be accurately assessed from these correlation coefficients because iron and aluminium themselves were positively correlated. Eleven of the forty-seven soils were chosen for determinations of the effect of selective removal of iron and aluminium on their ability to sorb phosphate. Soils usually sorbed less phosphate after various amounts of their iron and aluminium had been extracted chemically. Many of them, however, still sorbed large amounts of phosphate in spite of the removal of reducible iron and acid-soluble aluminium. Soils and haematite extracted with sodium dithionite in acetate at pH 5.0 usually sorbed more phosphate than untreated samples. This was due largely to the incomplete removal of reduced iron and could be prevented by washing with dilute acid. Acid washing after dithionite treatment caused sorption by the soils to fall below the values for the untreated samples and virtually destroyed that of the oxide. It also removed much aluminium from the soils. The amount by which sorption fell below that of the untreated samples could be accounted for largely by this aluminium. It was concluded tentatively that, for most of the soils, phosphate sorption was due to acid-soluble aluminium and to sorption sites which were non-reducible and insoluble in dilute acid. The contribution of reducible iron in the original soils remains in doubt, but it could be a minor one.