Kinetic and thermodynamic effects of moisture content and temperature on the ammonia volatilization of soil fertilized with urea

The traditional qualitative analysis of the individual factors on the kinetic and thermodynamic parameters cannot sufficiently reveal the mechanism underlying ammonia volatilization in soil. This study aimed to determine the effects of temperature, moisture content, and their interaction on the kinetic and thermodynamic parameters, which revealed the key control mechanism underlying ammonia volatilization, modified the traditional Arrhenius model, and established a quantitative prediction model of cumulative NH3-N loss (CNL). Laboratory culture experiments were conducted under different temperatures (T) (15°C, 20°C, 25°C and 35°C) and moisture contents (θ) (60%, 80%, and 100% field capacities). Soil ammonia volatilization was also measured every 2 d. Results showed that the effects of individual factors and their interaction on the values of reaction rate (KN), Activation free energy (ΔG), and activation entropy (ΔS) followed the descending order of T>θ>T×θ, whereas those of activation enthalpy (ΔH) and activation degree (lgN) followed the descending order of θ>T>T×θ. The interaction showed significant effect on KN value and insignificant effect on all the thermodynamic parameters. The effects of water and temperature were mainly observed during the preparatory stage and the most critical transition state stage of the chemical reaction, respectively. Given that ΔH > 0, ΔG > 0, and ΔS > 0, ammonia volatilization is found to be an endothermic reaction controlled by enthalpy. The new KN(T)-2 model with the determination coefficient (R2) of 0.999 was more accurate than the traditional Arrhenius model with the R2 of 0.936. The new NH3(T, θ) model with the mean absolute percentage error (MAPE) of 4.17% was more accurate than the traditional NH3(T) model with the MAPE of 7.11%. These results supplemented the control mechanism underlying ammonia volatilization in soil fertilized with urea and improved the prediction accuracy of CNL.