Column experiments were conducted to determine the retardation of selected Volatile Organic Compounds (VOCs) in ground water in low organic carbon sediments. The retardation factors and other parameters derived in these experiments were meant to provide constraints on our computational model and on our interpretation of data from other experiments, described elsewhere in this final report, to examine parameters such as diffusion, tortuosity coefficients, and mechanical dispersion, that affect

more »

... ntaminant transport. The goal of the project was to further our understanding of the mechanisms that control the transport of contaminants from fine-grained sediments in contaminant source areas into coarse-grained zones where they are advected away to form contaminant plumes. Experiment Design and Operation . . 1 . Column design The sand was packed in a glass chromatographic column approximately 15 cm long and 5 cm in diameter. The end fittings are made of Teflon and are separated from the sand with stainless steel screens. The plumbing from the pumps to the column is Teflon and stainless steel tubing and fittings ( Figure 1 ). . . 2 Column Packing The sand used in the column was Oklahoma No. 1 sand, a fine-grained, high purity quartz sand with no detectable organic matter. Standard U.S. sieves were used to evaluate grain size. A portion of the Oklahoma No. 1 sand was cleaned to remove a layer of clay-sized minerals that coats the grains. The clay accounts for 0.2-0.3% of the original mass of the sand and is composed of illite, kaolinite, and clay-sized quartz. Two columns were used in the experiment: one containing the cleaned sand and the other containing the original sand. To create as homogenous a packing as possible, the column was half filled with water and vibrated as the sand was slowly poured into the column. One of the ends of the column has a fitting that allows the bedding to be screwed in to the top of the sand pack. This fitting was applied hand tight following the pour. After each column packing, one or two air bubbles could be seen against the side of the glass column but these bubbles dissolved away after several days of pumping deionized water through the column. Porosity (0.32) and bulk density (1.7 g/cm 3 ) of the packed column were determined gravimetrically and hydraulic conductivity (4 x 10 -3 cm/s) was measured using a constant head hydraulic conductivity test. . . 3 Column Operation Influent water was delivered to the column from a syringe pump with four 100 mL syringes loaded with the same water. From the syringes, the water was directed to a stainless steel mixing tube and then to the column. Sampling ports were located after the mixing tube and before the column, and at Abstract A dynamic headspace method using in-vial purging previously developed for solid matrices was modified for analyzing volatile organic compounds (VOCs) in low-volume aqueous samples. Using volumes as low as 1.0 mL in these vials was validated by comparing VOC recovery efficiencies with the standard 5.0 mL method from 40-mL vials, measuring precision and accuracy of replicate vials, determining linearity of the calibrations, and checking sample integrity during the holding time. Results indicate that VOC recovery efficiencies from the invial purging method is analogous to the standard method. This method has advantages over conventional methods because much lower sample volumes are required.