Influence of Aluminum Hydroxide on Potassium Release from Two Colombian Soils

Abstract

The effect of sesquioxides on the mechanisms of chemical reactions that govern the transformation between exchangeable potassium (Kex) and non‐exchangeable K (Knex) was studied on acid tropical soils from Colombia: Caribia with predominantly 2∶1 clay minerals and High Terrace with predominantly 1∶1 clay minerals and sesquioxides. Illite and vermiculite are the main clay minerals in Caribia followed by kaolinite, gibbsite, and plagioclase, and kaolinite is the major clay mineral in High Terrace followed by hydroxyl‐Al interlayered vermiculite, quartz, and pyrophyllite. The soils have 1.8 and 0.5% of K₂O, respectively. They were used either untreated or prepared by adding AlCl₃ and NaOH, which produced aluminum hydroxide. The soils were percolated continuously with 10 mM NH₄OAc at pH 7.0 and 10 mM CaCl₂ at pH 5.8 for 120 h at 6 mL h⁻¹ to examine the release of Kex and Knex. In the untreated soils, NH₄ ⁺ and Ca²⁺ released the same amounts of Kex from Caribia, whereas NH₄ ⁺ released about twice as much Kex as Ca²⁺ from High Terrace. This study proposes that the small ionic size of NH₄ ⁺ (0.54 nm) enables it to enter more easily into the K sites at the broken edges of the kaolinite where Ca²⁺ (0.96 nm) cannot have access. As expected for a soil dominated by 2∶1 clay minerals, Ca²⁺ caused Knex to be released from Caribia with no release by NH₄ ⁺. No Knex was released by either ion from High Terrace. After treatment with aluminum hydroxide, K release from the exchangeable fraction was reduced in Caribia due to the blocking of the exchange sites but release of Knex was not affected. The treatment increased the amount of Kex released from the High Terrace soil and the release of Knex remained negligible although with Ca²⁺ the distinction between Kex and Knex was unclear. The increase in Kex was attributed to the initially acidic conditions produced by adding AlCl₃ which may have dissolved interlayered aluminum hydroxide from the vermiculite present, thus exposing trapped K as exchangeable K. The subsequent precipitation of aluminum hydroxide when NaOH was added did not interfere with the release of this K, and so was probably formed mostly on the surface of the dominant kaolinite. Measurement of availability of K by standard methods using NH₄ salts could result in overestimates in High Terrace and this may be a more general shortcoming of the methods in kaolinitic soils.

Keywords:

• K release
• Colombian soils
• sesquioxides
• clay minerals
• ion exchange

Acknowledgments

Dr. Rodriguez is grateful to the Colombian Government (COLCIENCIAS) for financial support and to CORPOICA for leave to study for a PhD degree and for supplying the soil samples.