is left on the surface when the sampling device is
ppb solution of chloroform for one hour. They
removed from the well, and any highly hydro-
found that 8090% of the desorption they ob-
phobic contaminants that may have adsorbed to
served occurred within the first 510 minutes,
the surface.
and that the amount desorbed after one hour con-
In contrast, several studies (Barcelona et al.
stituted less than 10% of the total chloroform
1985, Reynolds and Gillham 1985, Gillham and
sorbed. The results of this study indicate that con-
O'Hannesin 1990, Parker et al. 1990, Reynolds et
taminants at or near the surface desorb rapidly
al. 1990, Parker and Ranney 1996a, b) have shown
and that release of the remainder of the absorbed
that permeable materials, such as polymeric well
contaminants is slower, most likely because it is
casings and the tubings used in groundwater
controlled by diffusion. All three studies found
sampling pumps, sorb substantial quantities of
that the compounds present in the greatest con-
some organic contaminants from aqueous solu-
centrations following desorption were not the
tions. Most of these studies agree that these or-
same compounds that had been sorbed the most
ganic compounds have diffused into the polymer
rapidly or to the greatest extent. Parker et al.
matrix, i.e., absorption has occurred. The rate and
(1990) noted that it was the smaller molecules
extent of sorption varies between analytes for a
that were more readily desorbed and attributed
given material, and varies considerably between
this to the fact that diffusion is more rapid for
polymers for a given analyte. Generally, flexible
smaller molecules (Berens and Hopfenberg 1982,
materials tend to be much more sorptive, i.e.,
Reynolds et al. 1990).
both the rate of sorption and the extent of sorp-
tion are greater (Barcelona et al. 1985, Gillham
Studies that demonstrate the factors
and O'Hannesin 1990, Reynolds et al. 1990, Park-
that affect decontamination efficiency
er and Ranney 1996a). For example, Gillham and
While not many studies have examined the ef-
O'Hannesin (1990) found that after only 10 min-
ficiency of the various decontamination proto-
utes, sorption of low ppm levels of benzene by
cols, three studies that demonstrate the impact
flexible PVC was approximately 35%, while loss-
the type of contaminants, level of contamination,
es were less than 1% for solutions exposed to rig-
and materials being decontaminated can have on
id PVC.
decontamination effectiveness are summarized
For polymers that are exposed to very low
below.
activities (i.e., trace levels) of organic solutes, dif-
Devlin (1987) found that polyethylene tubing
fusion of the organic molecule in the polymer is
was harder to decontaminate than Teflon tubing
considered to be concentration independent, and
(the actual type was not specified). These tubings
at slightly higher activities diffusion is consid-
had been exposed to ppb levels of a suite of VOCs
ered to be concentration dependent (Berens 1985,
and were decontaminated by pumping deionized
Jenkins et al. 1986, Holsen 1988). However, when
water through them.
glassy amorphous polymers (e.g., rigid PVC) are
Schuh et al. (1993) compared the effectiveness
subjected to even higher concentrations (i.e., ap-
of distilled water rinses for removing seven pesti-
proaching the compound's aqueous solubility) or
cides (bromoxynil, diclofop-methyl, dimethoate,
to the neat chemical, and the chemical is a solvent
2-methyl-4-chlorophenoxyacetic acid [MCPA],
or swelling agent of the polymer, then diffusion
methyl parathion, propiconazole, and trifluralin)
will increase several orders of magnitude (Berens
from a PVC bailer. They selected a one-minute
1985, Jenkins et al. 1986, Holsen 1988).
contact time to represent the time required to take
Parker (1995) proposed that decontaminating
a bailed sample in a shallow well. With one ex-
permeable materials should involve more than
ception (dimethoate), they correlated the number
removing surface contaminants if desorption of
of rinses required with the analyte's solubility in
water and its Kow value. They observed that the
absorbed contaminants is significant. Unfortu-
most hydrophilic contaminants (i.e., with a solu-
nately, there have been only a few studies (Miller
bility of >500 mg/L or log Kow <2.3) were re-
1982, Barcelona et al. 1985, Parker et al. 1990) that
moved from the bailer with no residual carryover
have examined desorption of organic contami-
with just one rinse, while the most hydrophobic
nants from polymeric materials. Barcelona et al.
analytes (i.e., with a solubility of <3 mg/L or log
(1985) followed the kinetics of desorption of chlo-
Kow ~4.6, 4.7) had residual carryover after six
roform from five tubing materials (polypropy-
rinses.
lene, polyethylene, PVC, silicone rubber, and
Fink et al. (1989) decontaminated a stainless
PTFE) that had been previously exposed to a 100-
3
Previous Page
