Liu Zhibin, Liu Songyu, Cai Yi, Fang Wei
Institute of Geotechnical Engineering, Southeast University, Nanjing, 210096, China,
Environ Sci Pollut Res Int. 2015 Jun;22(11):8216-23. doi: 10.1007/s11356-014-3964-7. Epub 2014 Dec 18.
As the dielectric constant and conductivity of petroleum products are different from those of the pore water in soil, the electrical resistivity characteristics of oil-contaminated soil will be changed by the corresponding oil type and content. The contaminated soil specimens were manually prepared by static compaction method in the laboratory with commercial kaolin clay and diesel oil. The water content and dry density of the first group of soil specimens were controlled at 10 % and 1.58 g/cm(3). Corresponding electrical resistivities of the contaminated specimens were measured at the curing periods of 7, 14, and 28 and 90, 120, and 210 days on a modified oedometer cell with an LCR meter. Then, the electrical resistivity characteristics of diesel oil-contaminated kaolin clay were discussed. In order to realize a resistivity-based oil detection method, the other group of oil-contaminated kaolin clay specimens was also made and tested, but the initial water content, oil content, and dry density were controlled at 018 %, 018 %, 1.30~1.95 g/cm(3), respectively. Based on the test data, a resistivity-based artificial neural network (ANN) was developed. It was found that the electrical resistivity of kaolin clay decreased with the increase of oil content. Moreover, there was a good nonlinear relationship between electrical resistivity and corresponding oil content when the water content and dry density were kept constant. The decreasing velocity of the electrical resistivity of oil-contaminated kaolin clay was higher before the oil content of 12 % than after 12 %, which indicated a transition of the soil from pore water-controlled into oil-controlled electrical resistivity characteristics. Through microstructural analysis, the decrease of electrical resistivity could be explained by the increase of saturation degree together with the collapse of the electrical double layer. Environmental scanning electron microscopy (ESEM) photos indicated that the diesel oil in kaolin clay normally had three kinds of effects including oil filling, coating, and bridging. Finally, a resistivity-based ANN model was established based on the database collected from the experiment data. The performance of the model was proved to be reasonably accepted, which puts forward a possible simple, economic, and effective tool to detect the oil content in contaminated clayey soils just with four basic parameters: wet density, dry density, measured moisture content, and electrical resistivity.
由于石油产品的介电常数和电导率与土壤孔隙水不同，受油污染土壤的电阻率特性会因相应的油类和含量而改变。在实验室中，采用静态压实法，使用商业高岭土和柴油人工制备受污染土壤试样。第一组土壤试样的含水量和干密度分别控制在10%和1.58 g/cm³。在改良的固结仪上，使用LCR仪表在7、14、28、90、120和210天的养护期测量受污染试样的相应电阻率。然后，讨论了柴油污染高岭土的电阻率特性。为了实现基于电阻率的油检测方法，还制作并测试了另一组受油污染的高岭土试样，但初始含水量、含油率和干密度分别控制在018%、018%、1.30~1.95 g/cm³。基于测试数据，开发了基于电阻率的人工神经网络（ANN）。发现高岭土的电阻率随含油率的增加而降低。此外，当含水量和干密度保持恒定时，电阻率与相应含油率之间存在良好的非线性关系。受油污染高岭土的电阻率在含油率为12%之前的下降速度高于12%之后，这表明土壤的电阻率特性从孔隙水控制转变为油控制。通过微观结构分析，电阻率的降低可以用饱和度的增加以及双电层的坍塌来解释。环境扫描电子显微镜（ESEM）照片表明，高岭土中的柴油通常有三种作用，包括油填充、涂层和桥接。最后，基于从实验数据收集的数据库建立了基于电阻率的ANN模型。该模型的性能被证明是合理可接受的，这提出了一种可能简单、经济且有效的工具，仅用湿密度、干密度、测量含水量和电阻率这四个基本参数就能检测受污染粘性土壤中的含油率。
