11
with a polyethylene mesh covering to protect the bag from abrasion. Stainless
steel or Teflon-coated stainless steel lines are recommended (Vroblesky 2001).
These devices were developed by and have been well studied by Vroblesky and
co-workers (Vroblesky and Hyde 1997, Vroblesky 2001, Vroblesky and Camp-
bell 2001). According to USGS guidance (Vroblesky 2001), they are suitable for
sampling most VOCs. However, they are not suitable for sampling inorganic ions
and have limited applicability for non-volatile organic contaminants and even
some VOCs (e.g., ketones, ethers, and alcohols) (Vroblesky 2001, Vroblesky
and Campbell 2001). Vroblesky and Campbell (2001) found concentrations of
ketones, ethers, and alcohols tended to be lower in the PDB samplers than in
control samples.
Equilibration times for these devices in a well vary considerably with water
temperature. For example, at 21C, equilibration times (in closed vessels) for a
number of VOCs (benzene, cis-1,2-dichloroethene, PCE, TCE, toluene) were 48
hours. However, at 10C, equilibration times ranged from 52 hours for TCE and
PCE to seven days for other compounds (vinyl chloride, 1,1,1-trichloroethane,
1,2-dichloroethane, and 1,1-dichloroethane) (Vroblesky 2001, Vroblesky and
Campbell 2001). The equilibration time of a closed system can be calculated
using Fick's second law of diffusion, given by Crank (1975).
These devices are becoming popular because they are relatively easy to
deploy, relatively inexpensive ( per bag), and reportedly delineate
stratification of contaminants well. Because they are disposable, they do not
require decontamination and they reportedly eliminate or substantially reduce the
amount of purge water required, thereby further reducing cost. These devices
presumably would work well in formations with low permeability where it is
difficult to obtain samples using other types of sampling devices. The pore size
(~10 angstroms) of the polyethylene does not allow sediment to pass and this
One of the disadvantages of using this type of device is that it takes several
days to reach equilibrium, and therefore the sample reflects concentrations that
have been in the well over that time period rather than the particular instance
when the sample was taken. In addition, long-term deployment may result in
biofilm development, which can slow transport across the film (Vroblesky 2001).