Ball milling pretreatment affects the content of fixed ammonium in soils in response to the content of exchangeable ammonium

Analysis of fixed ammonium (NH ₄⁺) in soils by the most common method requires the subsample to be ground to pass through a 100-mesh sieve. We tested a hypothesis that the ball milling pretreatment for the analysis of fixed NH ₄⁺ may influence the analytical results depending on the content of exchangeable NH ₄⁺. Five paddy soils collected from the surface layer in Japan were treated with ¹⁵N-labeled ammonium sulfate at 0, 100, 500, or 1000 mg N kg ⁻¹, incubated aerobically at 30°C for 7 days, and air-dried. Part of each soil was then ground in a vibration ball mill at 1200 rpm for 10 min as a pretreatment (BMP treatment). The samples with and without the BMP treatment were analyzed for exchangeable NH ₄⁺ and total fixed NH ₄⁺. The samples that received 0 or 1000 mg N kg ⁻¹ were also analyzed for nonexchangeable ¹⁵N (total fixed ¹⁵NH ₄⁺ plus organic ¹⁵N) and hot KCl-extractable organic N. At 0 mg N kg ⁻¹, the BMP treatment significantly decreased total fixed NH ₄⁺ in all soils and significantly increased exchangeable NH ₄⁺ in two of the five soils. The fixation of exchangeable NH ₄⁺ started to occur when its content is greater than around 300 mg N kg ⁻¹, and its ratio to total fixed NH ₄⁺ is greater than around 1.0. At 1000 mg N kg ⁻¹, the BMP treatment significantly decreased exchangeable NH ₄⁺ and significantly increased total fixed NH ₄⁺ in all soils especially two clayey, vermiculitic soils. On the average of all soils applied 1000 mg N kg ⁻¹, the increase in total fixed NH ₄⁺ explained 55% of the decrease in exchangeable NH ₄⁺, and the decrease in exchangeable NH ₄⁺ was almost equal to the increase in nonexchangeable ¹⁵N. The rest of the NH ₄⁺ unaccounted by the fixation was partly due to the transformation of exchangeable NH ₄⁺ to hot KCl-extractable organic N, because the content of hot KCl-extractable organic N was significantly increased by the BMP treatment. In conclusion, the milling treatment prior to the analysis of fixed NH ₄⁺ in soils might cause either underestimation or overestimation depending on the content of exchangeable NH ₄⁺.