49

influent concentration was maintained during testing. Loss in the tubing was less
than 3% for the alachlor and metolachlor solutions.
A control column test was conducted to assess sorption on the tubes and
glass column. Torpedo sand was used for control tests conducted with TCE.
Silica sand was used for the herbicides. Torpedo sand is classified as poorly
graded sand (SP) in the Unified Soil Classification System (USCS). The
uniformity coefficient and coefficient of curvature are 2.6 and 0.84. Silica sand is
a poorly graded sand (SP) with a uniformity coefficient of 2.0 and coefficient of
curvature of 1.39. The sands were washed with DI water until any visible
impurities were removed, and then were packed in the glass column. The test
solution was then introduced  using a peristaltic pump. The effluent
concentrations of compounds were normalized to the concentrations at the
influent reservoir. The breakthrough curve for TCE is shown in Fig. 3.9 and those
for the herbicides are shown in Fig. 3.10. The porosity was 0.33 and seepage
velocity was 1.02x10-2 cm/s for the column using TCE solution. For the column
using the alachlor solution, the porosity was 0.40 and seepage velocity was
3.08x10-3 cm/s. The porosity was 0.36 and seepage velocity was 2.93x10-3 cm/s
for the column with the metolachlor solution.
Partition  coefficients for TCE, alachlor, and  metolachlor were
backcalculated from the retardation factors, which were obtained by fitting Eq.
3.18 to the data. Discussion of the fitting procedure is in Sec. 3.5.3. A higher
partition coefficient was obtained for alachlor (1.56 L/kg) than for metolachlor