Multi-Objective Optimization of Engineered Injection and Extraction to Remediate Sorbing Contaminants in Homogeneous and Heterogeneous Aquifers

Abstract:

Ex situ remediation of sorbing contaminants in groundwater aquifers is especially difficult due to the tendency of sorbing contaminants to remain attached to the soil matrix. Consequently, in situ remediation, which typically involves injecting a treatment chemical into the aquifer to degrade the contaminant, is a more effective option. To enhance contaminant degradation during in situ remediation, a sequence of injections and extractions of clean water can be performed to increase the contact of treatment chemical and contaminant to enable more reaction. This technique is known as Engineered Injection and Extraction (EIE). In prior work, EIE was simulated for contaminants with varying sorption properties and reaction rates using two heuristically-developed sequences of injections and extractions. These EIE sequences achieved nearly complete contaminant degradation for weakly-sorbing contaminants with fast reaction rates; however, the sequences were much less effective for strongly-sorbing contaminants with slow reaction rates. In this work, we use multi-objective evolutionary algorithms to optimize the design of EIE sequences to determine sequences that achieve high amount of contaminant degradation for strongly-sorbing contaminants with slow reaction rates. We consider both homogeneous and heterogeneous aquifers.