Reactive Membrane Barriers for Containment of Subsurface Contamination
The overall goal of this project was to develop reactive membrane barriers--a new and flexible technique to contain and stabilize subsurface contaminants. Polymer membranes will leak once a contaminant is able to diffuse through the membrane. By incorporating a reactive material in the polymer, however, the contaminant is degraded or immobilized within the membrane. These processes increase the time for contaminants to breakthrough the barrier (i.e. the lag time) and can dramatically extend barrier lifetimes. In this work, reactive barrier membranes containing zero-valent iron (Fe{sup 0}) or crystalline silicotitanate (CST) were developed to prevent the migration of chlorinated solvents and cesium-137, respectively. These studies were complemented by the development of models quantifying the leakage/kill time of reactive membranes and describing the behavior of products produced via the reactions within the membranes. First, poly(vinyl alcohol) (PVA) membranes containing Fe{sup 0} and CST were prepared and tested. Although PVA is not useful in practical applications, it allows experiments to be performed rapidly and the results to be compared to theory. For copper ions (Cu{sup 2+}) and carbon tetrachloride, the barrier was effective, increasing the time to breakthrough over 300 times. Even better performance was expected, and the percentage of the iron used in the reaction with the contaminants was determined. For cesium, the CST laden membranes increased lag times more than 30 times, and performed better than theoretical predictions. A modified theory was developed for ion exchangers in reactive membranes to explain this result. With the PVA membranes, the effect of a groundwater matrix on barrier performance was tested. Using Hanford groundwater, the performance of Fe{sup 0} barriers decreased compared to solutions containing a pH buffer and high levels of chloride (both of which promote iron reactivity). For the CST bearing membrane, performance improved by a factor of three when groundwater was used in place of deionized water. The performance of high density polyethylene (HDPE) membranes containing Fe{sup 0} was then evaluating using carbon tetrachloride as the target contaminant. Only with a hydrophilic additive (glycerol), was the iron able to extend lag times. Lag times were increased by a factor of 15, but only 2-3% of the iron was used, likely due to formation of oxide precipitates on the iron surface, which slowed the reaction. With thicker membranes and lower carbon tetrachloride concentrations, it is expected that performance will improve. Previous models for reactive membranes were also extended. The lag time is a measurement of when the barrier is breached, but contaminants do slowly leak through prior to the lag time. Thus, two parameters, the leakage and the kill time, were developed to determine when a certain amount of pollutant has escaped (the kill time) or when a given exposure (concentration x time) occurs (the leakage). Finally, a model was developed to explain the behavior of mobile reaction products in reactive barrier membranes. Although the goal of the technology is to avoid such products, it is important to be able to predict how these products will behave. Interestingly, calculations show that for any mobile reaction products, one half of the mass will diffuse into the containment area and one half will escape, assuming that the volumes of the containment area and the surrounding environment are much larger than the barrier membrane. These parameters/models will aid in the effective design of barrier membranes.
- Research Organization:
- University of Minnesota, Minneapolis, MN
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- FG07-02ER63509
- OSTI ID:
- 899852
- Report Number(s):
- DOE/ER/63509; TRN: US0702567
- Country of Publication:
- United States
- Language:
- English
Similar Records
Environmental Management Science Program (EMSP) Annual Report
Fundamental studies of the removal of contaminants from ground and waste waters via reduction by zero-valent metals. Annual progress report, September 1, 1996--August 31, 1997