Determination of Cation Exchange Capacity in Soil by Automatic Kjeldahl Nitrogen Method after Ultrasonic Exchange and Suction Filtration Leaching

When the pH of soil sample was not more than 7.5, 2.00 g of the sample was taken and washed with 1 mol·L^-1 ammonium acetate solution until there was no caking, and the mixture obtained was diluted to 50 mL and exchanged with ultrasound for 6 min. When the pH of soil sample was more than 7.5, 2.00 g of the sample was taken and mixed with 50 mL of 1 mol·L^-1 ammonium chloride solution. The mixture obtained was boiled on electric furnace until the acidity of the water evaporated on the watch glass reached to pH 7 (judging that the ammonia had been boiled out), and 1 mol·L^-1 ammonium acetate solution was added into the above mixture and exchanged with ultrasound for 6 min. After turning on suction filtration, the sample obtained by ultrasound was poured onto the Brinell funnel, the precipitation was washed with 40 mL ethanol, and the washing process was repeated 4 times until no NH₄⁺ was washed out (detected with Nessler reagent). The precipitation, filter paper together with about 1 g of solid magnesium oxide reagent burned at high temperature in advance were placed on the Automatic Kjeldahl nitrogen analyzer with the concentration of hydrochloric acid solution of 0.05 mol·L^-1, the amount of 20 g·L^-1 boric acid absorption solution containing methyl red-bromocresol green indicator of 20 mL and the distillation time of 5.0 min, to automatically complete the digestion, distillation and titration operations. The output result of the instrument was the determined value. As found by results, the detection limit (3.143s) was 0.083 cmol·L^-1. The proposed method was applied to the analysis for the soil active composition analysis reference materials, and RSDs (n=5) of the determined values were less than 5.0%, and the determined values were within the uncertainty ranges of the certified values. Different types of samples from different regions were analyzed, and RSDs (n=5) of the determined values were less than those given by standard method of LY/T 1243-1999, with the relative errors of the determined values in the range of -1.5%-6.6%. The result (0.041 9) of t-test was less than the critical value (t_10,0.95=2.228), and it turned out that no significant difference exited between the two methods.