This site will look much better in a browser that supports web standards, but it is accessible to any browser or Internet device.

![]() |
Ben's research focuses on the use of hollow fiber membranes for the remediation of contaminated groundwater. The membrane material was polyethylene. This material's hydrophobic nature allows for it to be wetted by hydrocarbons, and to exclude water. The hollow fibers also have the advantage of a high surface area to volume ratio; many fibers can be placed within a small, confined space. These characteristics make them of possible value for retrieval of light non-aqueous phase liquids from the subsurface. These liquids can be present as a layer of fluid floating atop the groundwater table, particularly near refineries, leaking underground storage tanks, and spill sites.
![]() |
The recovery device required the manifolding of ~1500 hollow fibers into a polyethylene cylinder. This cylinder was attached to a pump, and then laid onto a hydrocarbon slick that was simulated in the lab. The slick was a layer of diesel fuel, composed of relatively few suspended solids, that floated atop a layer of distilled water. Tests using this device showed improving recovery with increasing trans-membrane pressure. However, this recovery, when scaled to results from single-fiber studies, was only about 35% of the expected recovery. This trend was observed when the fibers were folded upon themselves, and when they were flat. This led us to conclude that fiber-fiber interactions do not impact LNAPL recovery. Instead, problems in the construction process likely decreased the amount of fluid that could be recovered.
![]() |
Single-fiber studies were used to characterize the permeability of the membranes, so that the observed vs. expected recovery of the full-scale device could be evaluated. These studies were also used to characterize the effects of membrane fouling when fluids from the subsurface at a contaminated refinery were recovered. This site fluid possessed a large number of suspended solids, which fouled the membrane surface. Such fouling was combated with the use of backpulsing - a periodic reversal of trans-membrane pressure which extricated particles from the membrane surface. The relative importance of backpulse amplitude, duration, and frequency were explored, and backpulse amplitude was found to have the greatest effect upon membrane cleaning. Our results showed that backpulsing likely removed particles that constricted the membrane pores, and those that formed a cake layer. However, it was unable to remove large particles that completely blocked the pores.
![]() |
The single-fiber study results were scaled to the full-size recovery device, and so the applicability of the device in the field was explored. Because some fouling still occurs despite the use of backpulsing, recovery may not be performed indefinitely. Instead, periodic retrieval of the device from the recovery well will be required. The hollow fibers must then be cleaned or replaced. The most effective method of membrane cleaning for this particular situation is the topic of current research.