Direct hot water from a steam cleaner jet
Sample handling and analysis
inside the standpipe to clean the pump
Samples were collected in glass vials (care was
exterior, filling the standpipe. (We used a
taken to eliminate any bubbles or headspace) and
pressure washer rather than a steam
sealed with Teflon-lined plastic caps. For each
cleaner.)
sample that was taken, there were three replicate
2) Using the bladder pump, pump the hot
vials. These vials were stored in a refrigerator
water in the standpipe through the pump
until it was time to analyze the samples. At that
system. Remove the pump, discard the re-
time, an aliquot was transferred to a glass, 1.8-mL
maining water, and replace the pump in
autosampler vial using a glass Pasteur pipet.
standpipe.
Analytical determinations were performed
3) Add the detergent solution to the stand-
using reversed-phase high-performance liquid
pipe using the pressure washer. Using the
chromatography (RP-HPLC). A modular system
bladder pump, pump the wash solution (5
was employed consisting of a Spectra Physics
SP8875 autosampler with a 100-L injection loop,
gal. or ~5 system volumes) through the
pump system. (No mention is made as to
a Spectra Physics SP8810 isocratic pump, a Spec-
temperature; we used hot detergent solu-
tra Physics SP8490 variable-wavelength UV de-
tion.)
tector set at 215 nm, and a Hewlett Packard 3396
4) Remove pump from standpipe, drain re-
series II digital integrator. Separations were ob-
tained on a 25-cm 0.46-cm (5 m) LC-18 column
maining detergent solution, rinse the
standpipe with water from the pressure
(Supelco) and eluted with 2.0 mL/minute of 65/
washer, and fill standpipe with water. (No
35 (V/V) methanol/water. The detector response
mention is made as to temperature; we used
was obtained from the digital integrator operat-
hot water.)
ing in the peak height mode.
5) Pump (5 gal.) ambient temperature tap wa-
A primary TCE standard was made by weigh-
ter through the pump.
ing neat TCE into methanol in a 100-mL glass
6) Remove pump from standpipe and drain
volumetric flask, giving a final concentration of
remaining water from system.
3000 mg/L. This standard was kept in the freezer.
7) Place the pump in a large standpipe filled
Each sampling day, the primary standard was
with DI water, and pump DI water through
diluted in methanol to give a working standard
the system.
(30 mg/L). A series of water standards was made
Samples were collected after one, three, and six
from the working standard by serial dilution us-
system volumes of DI water were pumped
ing glass pipets and glass volumetric flasks. These
through the system.
standards ranged in concentration from 3.0 to
Sixth TCE study--Effect of a wash and oven drying
0.006 mg/L. The method detection limit (MDL)
on pump exposed to high concentrations. The EPA
was obtained according to the EPA protocol de-
protocol used in the previous study was also fol-
scribed in the Federal Register (1984).
lowed in this study. In addition, a hot air drying
step was added. After the rinse water was drained
Pesticide study
from the system, the pump was disconnected from
the tubing and any remaining water was drained.
Experimental methods
For 24 hours, hot air was passed through the
The pesticide solution contained four analytes:
pump and tubing, each separately. The entry tem-
lindane, heptachlor, aldrin, and dieldrin. These
perature was 70C. The temperature of the exit air
compounds were selected because they varied
was approximately 45C from the pump and 25C
in their hydrophobicity. The pesticide solution
from the tubing (135 ft). DI water was run through
was made by adding the pesticides directly to
the bladder pump (without any tubing) and
deionized water in four 5-gal. glass bottles and
samples were taken after 1, 3, 6, and 9 gal. of
stirring for two weeks on a magnetic stirrer.
water were flushed through the pump. A gallon
When the solutions were poured into the
of water represents 11.5 pump volumes. The tub-
standpipe, the solution was filtered through a
ing was connected to the laboratory DI water and
polymeric fabric to remove any undissolved par-
samples were taken after 1, 3, 6, and 9 gal. were
ticles of analyte.
flushed through the tubing. A gallon of water
Bailer study. The bailer was washed as described
represents 1.3 tubing volumes.
in the first TCE study.
7
Previous Page
