摘要
建立了原位臭氧化土壤修复的扩散-反应模型,该模型可以预测被处理污染物在土壤中的时空分布及臭氧的穿透曲线。运用序贯求解策略,由有限差分法获得了该模型的数值解,并根据一维土柱实验对其进行验证。土柱实验中菲含量为17mgkg-1,臭氧进口浓度为9.4mgL-1,气量为50mLmin-1。模拟土壤由90%的硅砂和10%的高岭土混合而成,土壤水分含量为12%。由土柱实验分别测定不同时刻菲在土柱中不同位置的残留量以及臭氧的穿透曲线。结果表明,在大多数时间里,模型对菲的时空分布拟合较好,但实验后期拟合效果不理想,而且根据模型求出的臭氧出口浓度也较实测值提前到达穿透点。这是由于模型中仅考虑了土壤中臭氧和污染物菲之间的反应,没有考虑臭氧和菲的中间降解产物之间的反应,因此需要对土壤中的臭氧化反应机理进行更深入的研究。
A microscopic diffusion-reaction model was developed to simulate in-situ ozonation for the remediation of contaminated soil, i.e., to predict the temporal and spatial distribution of target contaminant in the subsurface and the ozone breakthrough curve. The sequential strategy was employed to obtain the numerical solution of the model using finite difference method. One-dimensional column tests were conducted to verify the model. The column was packed with simulated soils that were spiked with 17 mgkg-1 phenanthrene. Ozone gas with the concentration of 9.4 mgL-1 passed through the column at the flow rate of 50 mLmin-1. The simulated soil was composed of 90% silica sand and 10% kaolinite, and the soil moisture content was 12%. The residual phenanthrene on the soil at different depths of the column and the effluent gaseous ozone concentration was determined respectively at fixed time intervals. Compared the experimental data with the simulated values, it was found that the mathematical model fitted data well during most time of the experiment, but the simulation became worse near the end of the experiment. In addition, the calculated breakthrough point was earlier than the measured value. This is due to the fact that only reaction of ozone with parent compound, rather than the reactions of ozone with parent compound and its intermediates, was considered when modeling. Further work should be done on the mechanism and kinetics of ozonation in the soil.
出处
《高校化学工程学报》
EI
CAS
CSCD
北大核心
2003年第4期466-470,共5页
Journal of Chemical Engineering of Chinese Universities
基金
国家留学基金管理委员会资助项目(97842039)
国家自然科学基金项目(20107005)
关键词
原位臭氧化
土壤修复
菲:模型
in-situ ozonation
soil remediation
phenanthrene
modeling