Ruan Yunjie, Tan Hongxin, Luo Guozhi, Wang Chenyao, Che Xuan, Sun Dachuan. Nitrate removal and kinetic model of saline water by sulfur-based autotrophic denitrification[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2009, 25(2): 167-172.
    Citation: Ruan Yunjie, Tan Hongxin, Luo Guozhi, Wang Chenyao, Che Xuan, Sun Dachuan. Nitrate removal and kinetic model of saline water by sulfur-based autotrophic denitrification[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2009, 25(2): 167-172.

    Nitrate removal and kinetic model of saline water by sulfur-based autotrophic denitrification

    • The investigation of nitrate removal in recirculating aquaculture system as a purpose, autotrophic denitrification of nitrate removal and kinetic characteristics for saline synthetic wastewater were carried out in a sulfur packed-bed reactor. The experimental results show that nitrate is effectively removed when the influent NO3--N concentration is in the range of 22.5~368 mg/L. The optimal influent loading rate in order to ensure that the removal efficiency of nitrate was higher than 95% and nitrite accumulation occurred in effluent below 1 mg/L was in the range of 0.052~1.088 kg/(m3·d) when the water temperature was (29±1)℃. The maximum nitrate volumetric removal rate of 1.65 kg/(m3·d) was achieved at an influent loading rate of 2.171 kg/(m3·d) with obvious nitrite accumulation in effluent reaching 28.69 mg/L. The kinetic experimental results show that the NO3--N removal rate in the reactor can be described by a half-order kinetic model for biofilms. It has been found that the half-order reaction rate constant per unit reactor volume K1/2v is 7.84~8.5 mg1/2/(L1/2·h) and it can be successfully applied in the kinetic model for the prediction of effluent NO3--N concentration. The two groups’ predictive values and actual values were analyzed by using SAS 8.0 software for ANOVA analysis. The ANOVA analysis indicates that the corresponding Pr>F values are 0.9732 and 0.8845, which proved that no significant differences exist between the predictive values and actual values of the model.
    • loading

    Catalog

      /

      DownLoad:  Full-Size Img  PowerPoint
      Return
      Return