Getting errant gas out of soil can be a guessing game for the environmental service sector, but new research at the School of Engineering, University of Guelph could give clean-up crews an important tool. This is critical as gasoline contamination of subsurface soils is a widespread problem around the world, as an estimated 25 per cent of existing gas stations have leaks in their storage tanks systems. These cost billions of dollars to remediate.

So, Prof. Richard G. Zytner, School of Engineering, is leading an effort to increase the efficiency and predictability of remediation for these sites. The collaborative research team is developing a model into which information for each site can be inputted, to evaluate how long it may take to clean-up a site. Currently remediation consultants are operating mostly blind, and it is difficult for them to tell their clients how long clean-up will take or what the cost will be. By developing the proposed model, remediation consultants will have a tool to answer these questions and help optimize the design process and improve efficiency.

The most common technique for cleaning gasoline- (hydrocarbon-) contaminated sites is the use of soil vapour extraction (SVE), where powerful blowers pump air from the soil to remove the hydrocarbon molecules in the vapour form. Considerable experience has been developed over the past 10-15 years with this approach. However, system design and operation is governed, almost exclusively, by qualitative experience. Unfortunately, optimal performance is not often achieved. Sub-optimal performance translates into higher final concentrations in the soil and substantially higher cleanup costs. Depending on the structure of the soil, molecules of gas can be trapped in the small spaces between soil particles, such as silt and clay. As such, the air may not flow through well enough to carry them out. The presence of water also impacts the process.

Compounding the problem is that currently the optimal location for the extraction wells and optimal operating flowrates cannot be determined. Also, at contaminated sites a phenomena known as ‘tailing’ is the norm. Tailing occurs when the removal concentration falls far below the optimal saturation level in the extracting air. It is recognized to occur due to the large scale spatial variability of real sites and due to small scale effects of water/gasoline distribution in the soil’s porous structure. Severe tailing increases operating costs and can lead to final contamination levels that exceed cleanup standards.

Bioventing is considered a possible option to provide final polishing of the site on a cost effective basis. Bioventing, through the supply of sufficient oxygen and the addition of nutrients, stimulates indigenous hydrocarbon degraders to break down the contaminants left by SVE. The literature reports that many bioventing sites are nutrient limited, especially in terms of nitrogen and phosphorous. However, the carbon-nitrogen-phosphorous (CNP) ratios presented in many papers vary widely, from 100:10:1 to 1000:10:1. In addition, there has been no consensus reached regarding the ideal form of nitrogen to supply. Some authors found that supplying ammonia-nitrogen was more efficient than supplying nitrate-nitrogen because ammonia is in the reduced form required by most hydrocarbon degraders. However, bioventing systems supplied with ammonia-nitrogen were also observed to suffer a substantial decrease in pH, resulting in critically reduced biodegradation rates. Very few authors have addressed the effects of nitrogen source on biodegradation rates under bioventing conditions.

An additional challenge is knowing when to convert from SVE to bioventing. Unfortunately, it is currently it is not possible to determine the optimal transition point between SVE operation to remove the bulk and bioventing to achieve clean-up standards.

Accordingly, the objectives of this project are to quantify tailing factors for a comprehensive range of conditions and determine the associated scale-up factors, develop the ability to model bioventing performance and to determine the optimal transition point and operating strategies, and develop the ability to model the performance of heterogeneous sites with emphasis on optimal system design.

Determining mathematically-based guidelines for the operation of remediation systems is difficult because of the complexity of contaminated sites. Most remediation technologies, such as SVE, were developed through trial and error in the field and laboratory scale experiments. Unfortunately, real-life sites are much less uniform and controlled and have many factors affecting system performance. However, the School of Engineering project integrates this through the co-operation of Cushman-Ball Environmental Ltd., an engineering consulting company located in Windsor, ON. With the cooperation of Cushman-Ball Environmental Ltd. and their clients, results obtained from laboratory experiments are compared to results from the field.

Additional soil cores and ground penetrating radar will be used to map the soil subsurface, compositional analysis of the effluent air will be completed, more frequent soil sampling and the operating conditions will be varied during the course of cleanup. This level of data resolution for a real site coupled with good data on how the system responds to operating changes will provide the foundation for validating a quantitative model with the ability to handle the complexity of real sites and the complexity of real gasoline. Laboratory scale experiments will measure the extraction mass transfer parameters and biodegradation kinetic rate coefficients under the various operating conditions that each element of the heterogeneous soil is anticipated to experience. An additional challenge in stimulating bioremediation at field locations is the effective delivery of nutrients to the contaminated region. Comparison of field and laboratory tests will be conducted to test the effectiveness of various techniques and evaluate important scale-up factors. The ability to effectively predict SVE/Bioventing system performance could lower the cost of cleanup and improve environmental performance worldwide.

The research is done in a collaborative mode. In addition to staff engineers at Cushman-Ball Environmental Ltd., there is input from co-investigators Prof. Warren H. Stiver, School of Engineering and Prof. Hung Lee, Environmental Biology. Currently there are three students working on the project, with an additional four graduate students scheduled to join the research team shortly. In addition there are a number of summer and co-op research placements involved in the project.

The funding from this research comes from the Natural Sciences and Engineering Research Council (NSERC) of Canada, through their Strategic Grant Initiative and Cushman-Ball Environmental Ltd., the industrial sponsor. Additional funding for bioventing has been obtained from Centre for Research in Earth and Space Technology (CRESTech) a Province of Ontario supported initiative.
 

Top