48

into a glass column in three layers to a dry density expected under field
placement conditions. Teflon® adapters and porous stones were attached at the
both ends of the column. A glass fiber filter was placed on the porous stone to
prevent clogging.
The influent was prepared from stock solution diluted in a volumetric flask
containing DI water and stirred for 24 hrs with a magnetic stirrer. The DI water
was purged beforehand with N2 to remove 02, and then mixed with sodium azide
(0.1% by weight) to prevent biological activity. To evaluate losses during
preparation, aliquots of TCE (100-tL) and alachlor and metolachlor (1-pt) were
injected into the GC and analyzed following the methods in Sec. 3.4.2.1. The
variation was ±4% of expected value for TCE and ±5% for alachlor and
metolachlor, which suggests that losses were minimal.
For the constant head tests, solution was introduced into the column from
the influent Teflon® reservoir. The solution was introduced from the top to the
bottom. A constant head drop was maintained by separating the influent and
effluent bags a constant distance. The hydraulic gradient ranged from 4 to 28.
For the constant flow rate tests, solution was introduced into the columns
at a constant flow rate using a peristaltic pump. All contact parts in the pump
were made of Teflon®, except for the tubing, which was Viton®. Silicon tubing
was not used because it exhibited unacceptable losses. A loss of approximately
10% occurred when TOE passed through the Viton® tubes, but a fairly constant