複雜地下污染場址之精確整治方法應用

本場址含水層( 滯水層)地質屬砂、坋黏土互層，地質滲透性差異可達2 ~ 3個數量級。場址地下水遭受1,2-二氯乙烷及氯乙烯嚴重污染( 達數個至數十ppm)，且污染物已擴散進入低滲透性地層( 坋黏土)，造成傳統重力或加壓整治或灌注技術無法有效處理污染。為處理本場址複雜地下污染，藉由「精確整治」(surgical remediation)方法，透過細部調查建立水文地質及污染團三維特徵概念模型；採用可同時處理高、低滲透性地層之新穎「雙環塞滲透灌注工法」(double packer injection, DPI)將整治藥劑穩定傳輸到特定深度地層，充分發揮整治藥劑之復育功效。有關整治成效評估，則沿垂直地下水流方向評估整體質量通量變化，並結合新穎「跨孔式地電阻影像法」(cross-hole electrical resistivity tomography, CHERT)，以直接採樣及間接地物時間序列(tim-elapse, TL)監測方式評析整治期間藥劑傳輸之變化與整體空間－時間改善成效，並適時動態設計整治系統( 如整治井位置、灌注深度、灌注量等)。本研究證實精確整治方法確能有效掌握細部地下環境特徵，並能即時動態監控整治成效，整治藥劑能同時在高、低滲透性地層中穩定傳輸，符合預期分布方向，並能與污染目標區作有效接觸混合而大幅降低各項污染物濃度。

關鍵字

低滲透性地層；精確整治；雙環塞滲透灌注工法；地物時間序列監測

並列摘要

Because of geological heterogeneity, it is easy to misestimate the distribution of pollutants and incorrectly predict their transportation during investigations and groundwater contamination remediation. Furthermore, remediation medicaments or biological species cannot be effectively transported to polluted areas with complex geology, especially in low-permeability strata. Therefore, it is necessary to develop Surgical Remediation (SR) for some sites with complex heterogeneity. The SR uses effective investigation techniques to obtain underground information, such as 3-D pollutant distribution, geological distribution, and biochemical information, and applies better remediation transportation techniques to address geological heterogeneity. For example, as a result of such efforts remediation medicament would spread more widely and thus better remediate pollutants in the low-permeability strata. We chose a groundwater pollution site located in an industrial zone in northern Taiwan to test SR. The testing area (15 m in length and width) is an aquitard which is between the first and the second aquifers. The aquitard is at the depth of 17 ~ 32 m underground. The results of multi-depth passivediffusion bag sampling and micro-purge sampling (samples are collected from 9 ~ 12 different depths of each monitoring well: 17.5, 19.5, 21.5, 23.5, 24, 25, 25.5, 27.5, 29.5, 30, 31, 31.5 m), show that the 1,2-DCA and VC of several monitoring wells are higher than 10 mg/L (highest 1,2-DCA is 58.9 mg/L; highest VC is 14.2 mg/L). Concentrations of pollutants differ by depth. Pollutant concentrations seem comparatively high in deep areas (25 m), and thus it is suggest that increased the amount of injections to 25 m in depth in order to enhance performance. According to the results of the analysis, pollutants are distributed over both high-permeability strata (sand) and low-permeability strata (silty clay), and thus the traditional gravity and compressive remediation methods are invalid. We therefore introduce the concept of SR, using Multidepth Pollutant Sampling Analysis, Bacteria Flora Analysis, Multi-depth Slug Tests, Well Log Analysis, Multi-depth Flow Velocity, and Direction of Single Well Test, to evaluate 3-D hydrogeology characteristics and the space-time variation of pollutants. The aquitard consists of two high-permeability layers inside and three low-permeability layers outside. The southwest side of the testing zone contains the highest K value which shows the preferential paths of the aquitard. Injection pressure and flow are designed according to the values of hydraulic conductivity, concentration of pollutants, and depth of pollution. Double Packer Injection (DPI) is conducted with multi-depth medicaments (Japanese patented bio-stimulation products, EcoClean®/EcoClean-E®, which can degrade high-concentration chlorinated contaminants effectively was chosen) in the specific strata. An automated monitoring system developed by this study's research team is used to confirm the association between the pressure in each injection depth and flow changes. Since DPI would not change the original volume and structure of strata, it is not possible that pollution areas expand from artificial cracks. DPI can be used with medicaments in the treatment of different depths (1 m is divided into two to three sections) up from several to ten atmospheric pressure injections, allowing the medicament -- with a limited range horizontal transfer injection -- to easily break through the surrounding biofilm or injection well's chemical precipitation barrier. After the DPI injection period, the results from long-term water level monitoring wells show that the injection process did not change the test area northeast to the southwest direction of the groundwater flow field, and at the same time there was no overflow of the remediation medicament to the surface. Instead the remediation medicament was stably transmitted to the surrounding formations. Finally, we evaluate the variation of mass flux in accordance with the groundwater flow direction and perpendicular direction and directly or indirectly assess the variation of remediation medicament using Cross-Hole Electrical Resistivity Tomography (CHERT). A time lapse (TL) method is adopted to monitor the distribution of electrical characteristics underground. CHERT-TL is also an efficient approach to evaluate remediation effectiveness in remediation or post-remediation sites. When the lapse of time is short, CHERT-TL is similar to real-time monitoring. Therefore, before and after injection, as well as during different stages of the injection batches, this study performed CHERT-TL to examine the advantages of detecting the preferential paths, or the potential transmission bottleneck areas, dynamically adjusting the next stage of the next batch or injection position, depth, injection depth, and injection pressure design, in order to achieve more SR. The main compounds examined by the monitoring wells (1,2-DCA, VC), in the first stage after treatment with DPI for approximately one and a half months, showed declines in each well. 1,2-DCA at various depths declined up to 95% and the declines in VC were also up to 90%. The contamination of some of the monitoring wells from groundwater exceeding the second class pollution control standards fells hundreds of thousands of times: after the second phase of DPI injection for about two months, for 1,2-DCA at each depth of each well and VC, the rate of decline was also up to 90%, and the results for the majority of the pollution monitoring wells were lower than the concentration of groundwater pollution control standards; after the third stage of DPI injection for about two months and then at the three months re-inspection stage, at the depth of each of the monitoring wells the major compounds of concern (1,2- DCA, VC) were still well below the detection limit. Dehalococcoides sp. monitoring wells found that the microbial population numbered from 5.09 ~ 2.65 × 10^2 gene copies/mL, and rose sharply to 10^6 gene copies/mL. It was also found that bacteria-containing functional gene vcrA microorganisms were present, and at greater depths each of the monitoring wells had a high measured ethylene concentration. The results from another multi-depth mass flux assessment showed that from each monitoring well, after three stages of the injection medicament, the pollutants mass flux -- from two vertical cross-sectional views of the direction of flow in the monitoring wells -- were significantly reduced. This indicates the injection medicament did indeed transfer contamination to the target area, and that when evenly distributed in the test area in the underground environment, as intended by this biological treatment, mass flux can cause a steady decline in the contaminants measured by the monitoring wells: 1,2-DCA (initially about 9.6 × 10^(-3) ~ 6.1 × 10^(-1) g/day) and VC (initial about 1.6 × 10^(-3) ~ 1.9 × 10^(-1) g/day) after the injections were reduced over five to seven months by about 10^(-5) g/day. Overall, the three-stage injections of groundwater pollution and the CHERT-TL monitoring results during injection -- with dynamic adjustment of the position of the next phase of injection, injection depth, and injection pressure to effectively improve the transmission efficiency of the medicament -- can significantly enhance the effectiveness of this type of bioremediation. From five to seven months after the first injection, the high concentrations of contaminants in the test area groundwater were reduced to pollution control standards, and then to levels below the detection limit, and this shows this process described in this work can result in significant improvements. This research proves that SR can effectively investigate the underground characteristics of a contaminated site, and remediation medicament can be efficiently transported within high-permeability strata and low-permeability strata to fit in with the expected direction, High-concentration pollution in groundwater can be effectively addressed five to seven months after injections. 1,2-DCA and VC have been lower than the detection limit and did not exhibit increasing trend in three months.

並列關鍵字

low-permeability layer ； surgical remediation ； double packer injection ； time-lapse geophysical monitoring

國際替代計量

複雜地下污染場址之精確整治方法應用

全文下載

主題瀏覽