Groundwater Nitrate Removal by Zero-Valent Iron Permeable Reactive Barrier

• Main Authors: Hsu, Chia-Ping, 徐嘉彬
• Other Authors: Chen, Shiao-Shing
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/59266737515942053622

碩士 === 國立臺北科技大學 === 環境規劃與管理研究所 === 92 === Laboratory scale column tests were conducted to obtain the field design parameters and possible operation problems after field installation. For the nitrate contaminant water made in the laboratory, nitrate reduction by zero-valent iron was found affected by the coating or metal oxide on the ZVI surface. Without pretreatment of ZVI, nitrate removal rate was only 30%. Surface coating can be removed by strong reduction agent (NaBH4) or acid washing, but acid washing was shown better efficiency in this study. In the column tests, 1 mg ZVI can only remove 2.14 mg nitrate and the used ZVI was only 3.42% according to the stoichiometric ratio, and which is an evidence of surface reaction for the ZVI. The reduction of ZVI becomes iron oxides coating on the ZVI surface. In the column test, iron color was grayish in the beginning (pure ZVI) and became deep black after the ZVI was used. Effluent soluble iron concentration was below 1 mg/L. ESCA analysis was conducted for the surface species of the ZVI and Fe2O3 was found on the coated surface. Effluent pH was directly related to the initial nitrate concentration and removal rate and effluent ORP was decreased during the experiment. Therefore, pH is suggested as an indicator for nitrate removal efficiency for field operation. Increasing flux leads to increasing nitrate removal percentage. For the same flux, hydraulic resident time (HRT) and removal rate had no relationship, but higher HRT would increase operation time. The problem for field operation (using actual groundwater) was the clogging in the column. The reason for clogging was due to mineral precipitates after the increased pH and gas production. The mineral precipitates include Fe2O3, FeCO3, CaCO3. ZVI mixed with sand in the column can solve the clogging, but operation time was decreased. Transfer zone in the column were proportionately related to flow rate and flux. For pretreated ZVI by acid wash (5 r.pm with 157 cm3/cm2day), transfer zone were from1.85 to 3.03 cm and real hydraulic retention time were from 10 to 16 minutes. For determination of reaction constants, Thomas Equation was applicable, where R square was above 0.8 and reaction constants were between 0.0240 and 0.0553 (L/mg-day). Column tests were scaled up using two methods including Kinetic Approach and Scale-Up Approach. The differences were small comparing the calculated to the actual breakthrough volume and time using Kinetic Approach. For calculation of PRB installation, W/A (mass of zero-valent iron/cross-sectional plume) was below 1 kg/m2, and iron well was below 16 cm. It was low comparing to the literature value and it was due to that the ZVI used in this study was pretreated.