MSc Research

Monitoring and control of in situ bioremediation of soil contaminated with cable oil

Linear alkylbenzene (LAB) cable oil is used to insulate underground electricity transmission cables in England and Wales. This investigation was intended to provide data on the potential for bioremediation of soil and groundwater contaminated with cable oil, and to assess the suitability of Control Cube datalogging and control technology as the basis for a potential bioremediation monitoring and control system.

A model of a cable joint bay was constructed, contaminated with cable oil and inoculated with microorganisms from a previously contaminated site. Dissolved oxygen, pH, and oxidation-reduction potential in recirculated ground water were monitored before and after the addition of a nutrient solution. A microbiological investigation of the soil used both selective and non-selective media. The distribution of cable oil in the model was studied by fluorometry.

It was found that the soil and ground water contained aerobic cable oil-degrading microorganisms (CDMs). Some of these were isolated and tentatively identified as Actinomycetes. Fluorometric investigation revealed that the cable oil was localised to the upper regions of the saturated zone. Conditions in the ground water were shown to be predominately anaerobic, with no evidence of significant removal of cable oil. Therefore, there was scope for the improvement of the rate of cable oil attenuation through biostimulation - manipulation of conditions by the addition of nutrients and appropriate terminal electron acceptors, specifically oxygen, to encourage the growth of known CDMs.

Control Cube and associated sensors and software were shown to have a number of features that would be desirable in a monitoring and control system for an active bioremediation effort. However, it was thought that the complexity of the interactions between the contaminant, the environment and CDMs was such that a simple feedback control system would not be appropriate and a predictive model would be required in order to allow feed-forward control.

Copyright © 2000 Cranfield University.

This thesis was submitted in partial fulfillment of the requirements for the degree of Master of Science in Environmental Diagnostics at Cranfield University.

The project was supervised by Professor Naresh Magan in the Institute of BioScience and Technology, Cranfield University. Industrial supervisors were Dr. Daxaben Patel of National Grid plc. and Mr. David Crellin of Abington Partners.

The full text of the thesis is available from 10.6084/m9.figshare.96876.