| Over the years,crude oil spills occurred frequently,resulting in large amounts of toxic organic contaminants like aromatic hydrocarbons seeping down into the soil aquifer.These petroleum pollutants can easily adhere to soil particle surfaces and are highly toxic,easy to spread and difficult to remove,which continuously endanger the ecological safety.Therefore,proposing efficient remediation methods of petroleum contamination in soil-groundwater system is increasingly urgent.Eco-friendly solvent/chemotactic bacterial remediation can improve the bioavailability of petroleum contaminants in soil and make it possible to solve the problem.In this paper,the microfluidics and related image analysis were combined to establish a stable,accurate and realtime microscopic experimental method,which can detect the distribution of oil-phase pollutants and the remediation effect.The influence of different flow rates and heterogenous structures on the remediation effect were analyzed,and the parameters affecting bacterial chemotactic ability were integrated by the dimensional analysis to reveal the remediation mechanism of oil contaminant,which provides the research basis for the development of comprehensive optimal remediation technology for oil contamination.At the first stage of our research,an image processing method was established for the calculation of residual oil content and the semi-quantitative analysis of fingering flow morphology under different conditions of fluid displacement,and an image segmentation technique was used to identify water flow and residual oil.The results showed that this technique could effectively eliminate the influence of the external background and accurately distinguish and calculate the residual oil content and the fractal dimension of the fingering flow.At the second stage,microfluidics was applied to investigate the formation of oil contamination layers in heterogeneous porous media at typical groundwater flow rates.Our experiments demonstrated that heterogeneity and water velocity could significantly affect the formation of contamination layers,and the residual oil in the high permeable zone was lower than that in the low permeable zone in all micromodels.With the increase of flow velocity,the sweep area and fingering flow branches in the model increased,as well as the degree of chaos.Fingering flow morphology and residual oil content corresponded well to fractal dimension,with higher fractal dimension generally representing wider overflow cross-section,higher sweep area and lower residual oil content.In addition,three forms of residual oil were identified in the water flooding experiments: NAPL blocks,NAPL bridges and NAPL cones.At the third stage,the efficiency and mechanism of the remediation of oil contamination by Tween 80 and the green solvent dihydrolevoglucosenone(product name: Cyrene)were analyzed at typical groundwater flow rates.The results revealed that Cyrene could significantly reduce the residual oil content in the models by increasing the viscous force and reducing the interfacial tension,but different ratios of permeabilities affected the magnitude of the residual oil decline.Although Tween 80 could reduce the amount of residual oil in the models by reducing the interfacial tension of oil droplets and decreasing the critical inlet pressure in the low permeable zone,the results showed that the change in interfacial tension and the effect of heterogeneity led to higher residual oil in the low permeable zone than in water flooding in some models.In summary,Cyrene achieved better remediation effect than Tween 80.At the fourth stage,the parameters affecting bacterial chemotactic performance were summarized from literature research,and integrated by machine learning based on dimensional analysis to obtain the expression for the chemotaxis number Ch,which could evaluate the accumulation effect of chemotactic bacteria in the vicinity of contaminants.The chemotactic enrichment was evident when Ch > 1,and insignificant when Ch < 1. |
PDF Full Text Request