Table 3. Sampling times during sorption studies.
First study
10-ft tubing
50-ft tubing
100-ft tubing
Second study Third study
10 min
X
X
X
X
X
30 min
X
X
X
X
X
1 hr
X
X
X
X
X
4 hr
X
X
X
X
X
8 hr
X
X
X
X
X
13 hr
X
1 day (24 hr)
X
X
X
X
X
2 days (48 hr)
X
X
X
X
X
3 days (72 hr)
X
X
X
X
X
4 days (96 hr)
X
X
X
6 days (144 hr)
X
7 days (168 hr)
X
X
X
8 days (192 hr)
X
X
9 days (216 hr)
X
10 days (240 hr)
X
11 days (264 hr)
X
16 days (384 hr)
X
The times that the samples were collected in
using RP-HPLC. A modular system was
these three studies are given in Table 3. For each
employed consisting of a Dynatech LC241 auto-
sampler with a 100-L injection loop, a Spectra
tubing material and sampling time, three sam-
ples were collected by filling 1.8-mL HPLC auto-
Physics SP8810 isocratic pump, a Spectra Physics
sampler vials directly from the tubing outflow.
To determine the initial concentration of TCE in
nm, and a Hewlett Packard 3396 series II digital
integrator. Separations were obtained on a 25-
the well water for a given time, three control
0.46-cm (5 m) LC18 column (Supelco) and
samples were obtained by collecting water from
the three-port valve (Fig. 3). The samples were
eluted with 1.5 mL/min of 65/35 (V/V) metha-
analyzed immediately after collection.
nol/water. The detector response was obtained
from the digital integrator operating in the peak
height mode. A primary TCE standard was made
Experimental design for the
by weighing neat TCE into methanol in a 100-mL
desorption/leaching study
Three of the tubing materials from the sorp-
volumetric flask, giving a final concentration of
tion/leaching study were selected for this study
3,000 mg/L. This standard was kept in the freezer.
(PVDF, LDPE, and PP1). Of the polymeric tubings
Each day a series of water standards was made
tested previously (Parker and Ranney 1996),
from the primary standard by serial dilution us-
these materials ranged from the least sorptive to
ing pipets and volumetric flasks. These standards
highly sorptive. These tubings were contaminat-
ranged in concentration from 3.00.006 mg/L.
ed by pumping contaminated well water
The Method Detection Limit (MDL) for TCE
through them at a flow rate of 1 L/min for seven
(0.0026 mg/L) was determined by using the pro-
days (168 hr). The needle valves in the distribu-
tocol described in the Federal Register (1984).
tion system were closed to stop flow into the tub-
ings. The distribution system (Fig. 4) was then
Data analysis
disconnected from the well's pump and the com-
For each time, analysis of variance (ANOVA)
plete system was moved from the well house to
tests were conducted to determine whether the
the laboratory. The distribution system was then
tubing had any significant effect (at the 95% confi-
connected to the laboratory's DI water source,
dence level) on the TCE concentrations when
and DI water was pumped through each tubing
compared with control values. When significant
at a flow rate of 100 mL/min. Samples were col-
differences were found, Fisher's Protected Least
lected after 10 min, 30 min, 1 hr, 4 hr, 8 hr, 24 hr,
Significant Difference Test was performed to de-
48 hr, 72 hr (three days), and 96 hr (four days).
termine which tubing materials were significant-
ly different from the controls and each other.
The mean normalized concentrations of TCE
Analyses
Analytical determinations were performed
were obtained by taking the mean concentration
7
Previous Page
