Formatted Title
Passive Membrane Sampler for Assessing VOCs Contamination in Unsaturated and Saturated Media
Background/Objectives
The heterogeneity of soil matrices can create random and complex migration pathways of subsurface dense non aqueous phase liquid (DNAPL), making it challenging to find and assess the contaminant source. A semipermeable membrane, e.g., polydimethylsiloxane (PDMS), has been used as a medium in passive samplers, allowing volatile organic compounds (VOCs) to transfer through the membrane while preventing other compounds from passing through. Therefore, equilibrium concentration is achieved based on the diffusive concentration gradient mechanism of the target VOCs. In this study, a PDMS dialysis passive sampler with PDMS as an interface between the two phases (i.e., aqueous vs. aqueous phases or aqueous vs. vapor phases) was designed with possible advantages, including quick assembly of the sampler, to minimize losses of VOC from the dialysis cell, the sample to be obtained from the dialysis vial by inserting a needle through the PDMS membrane and extracting the required volume without opening the vial and therefore eliminating the possibility of VOC loss from off-gas, repeated uses of the sampler, and reproduction of soil pore vapor concentration by employing the physics of Henry’s law. The designed sampler was tested for its potential applicability in both aqueous and gaseous phases.
Approach/Activities
Laboratory-scale batch and three-dimensional sand tank experiments were conducted under controlled conditions. The passive sampler was constructed with PTFE material, with dimensions of 100 mm in length, 23 mm O.D., 11 mm I.D., and PTFE caps screwed on both ends to seal the sampler. Two PDMS membranes were inserted between the PTFE cap and the sampler body on both ends. The inside of the sampler was filled with RO water (5 mL) with zero headspace. Aqueous or gaseous trichlorethylene (TCE) (a model VOC compound) concentration gradient causes TCE molecules to transfer through the PDMS membrane until equilibrium concentrations are attained. Once equilibrium concentration is reached, sampling is complete and PTFE end bolts are screwed into both sides of the sampler to prevent aqueous contaminants from volatilization. Initial experiments were conducted to test the functionality of the constructed sampler (e.g., testing for water leakage or PDMS adsorption). Two sets of sampler aqueous and vapor equilibrium batch experiments were conducted. The sampler aqueous equilibrium tests for TCE in the aqueous phase were conducted by placing samplers in each of a series of 1 L bottles containing TCE at two concentration levels and allowing equilibration for different periods. In the TCE vapor phase sampling equilibrium test, the passive sampler was placed in a reactor containing TCE vapor with the sampler containing RO water as in the aqueous phase test. Various equilibrium periods were evaluated. A three-dimensional (3-D) sandbox was employed to simulate field unsaturated aquifer for assessing TCE vapor plume distribution. The measured TCE concentrations were interpolated with the software extensive 3-D modeling tool to create a TCE vapor distribution model.
Results/Lessons Learned
The PDMS passive sampler was constructed for this study and tested for its potential application to detect VOCs in the subsurface. It was determined that, an equilibrium time period of >10 d was required in the aqueous phase when the TCE concentration ranged between 3 and 25 mg/L, while an equilibrium time period of >12 d was necessary for TCE sampling in the vapor phase. To provide greater confidence that the passive sampler measures TCE accurately, an extended equilibrium time of 14 d is suggested. In the three-dimensional sandbox experiment, 3-D visualization was performed by subdividing the 3-D mapping in horizontal and vertical slices along three geometrical axes. The data obtained from the passive sampling process provided a more reliable simulation of the 3-D TCE spatial distribution than the AS process. The passive sampler developed in this study appears to be an effective and more accurate option, than active sampling, for assessing TCE vapor in the unsaturated zone.