INTRODUCTION

The LLNL Livermore Site has volatile organic compound (VOC) contamination of the vadose zone (and in the ground water) as a result of past practices. Up to 10 ppm total VOCs are found in soils (and up to 6 ppm total VOCs are found in ground water) in several areas under the Livermore site. As part of an ongoing monitoring program, VOCs in both soil-air and soil-water are monitored to observe migration patterns and to assess the effects of remediation efforts. SEAMIST sampling systems (by SEA, Santa Fe New Mexico) have been purchased to be used in this effort.

SEAMIST is a flexible, removable, PVC-coated, nylon membrane tube that is used to seal the sides of a borehole and to which sampling devices or other types of instrumentation may be attached. Two basic types of sampling devices available are the gas-port units and the absorbent-pad units. This paper will describe our efforts to determine the recovery rates of TCE from SEAMIST pads over different ranges of sample volume, sample concentration and duration of pad exposure to air.

METHODS AND MATERIALS

Sheets of an absorbent material made of blown polypropylene with cellulose and containing a dry surfactant, were cut into 5 x 6 cm rectangular pads. Each pad was weighed and stored in a clean vial or jar until needed. TCE stock solutions of 0, 50, 100, 200, 300 and 400 ppb were prepared using ultra pure water and were analyzed immediately for their TCE content.

(click image for larger size, 1, 2)

All samples were analyzed for TCE in Soil-Volatile Organic Analysis Vials (SVOAs). SVOAs are glass tubes with Teflon-lined septa and screw-caps on each end. The vial has a sintered glass frit close to one end which is used to hold soil samples in place during analysis. The SVOAs are designed to be used with a Dynatech PTA-30 autosampler, where purge gas or water is injected into the end with the frit, while sample is withdrawn during purging from the other. All SVOAs required for each experiment were labeled and weighed before each experiment was begun.

(click image for larger size, 3, 4)

All experiments were performed with replica sets of three. Controls for each experiment consisted of aliquots of the same volume and concentration of TCE stock solutions used in the experimental groups, added directly to SVOAs containing the pre-weighed SEAMIST pads.

In all experimental groups where one of the variables was handling or air exposure, the pre-weighed pads were placed in screw-cap jars. Aliquots of TCE stock solutions were then added to the pads in these jars and the caps were secured immediately. All jars were then allowed to equilibrate for about 1 hour.

After exposure and prior to analysis, all SVOAs containing pads and liquid were weighed. The final moisture content of the pads was determined by subtracting the dry weight of each SVOA/pad sample from the pre-analysis wet weight of both .

Each SVOA was purged for 20 minutes at 90° C with 40 cc/min. of helium to an OI 4460A purge and trap, using a Dynatech PTA-30 autosampler. Total TCE on the trap was then determined by gas chromatography on an HP 5890 using a J & W Scientific DB624 megabore .53 mm x 3 m column and an ECD detector. The TCE concentration of the sample was calculated by dividing the total TCE in the sample by the pre-analysis water volume in the SVOA. The rate of TCE loss from the experimental groups was determined by dividing the average of all replicates of each treatment by the average of the replicates from the appropriate control.

Experiment 1. Effect of air exposure on TCE recovery from SEAMIST pads.

In this experiment, a 10 mL aliquot of 400 ppb stock solution was placed on each pad in either a jar or an SVOA. The pads containing the experimental treatments were removed from their jars and placed on a plastic membrane were they were then exposed to air for 0, 2, 4, 8,16, 32 and 64 seconds. The exposure time started when the lid on the jar was opened and ending when the pad was removed from the plastic membrane. Handling time was defined as the time it took to transfer the pad from the plastic membrane to the pre-weighed SVOA and to seal the cap. The pads and SVOAs were analyzed for total TCE content within 24 hours of exposure.

Experiment 2. Effect of sample volume on TCE recovery from SEAMIST pads.

In this experiment, pads for the experimental treatments were placed in jars and pads for the corresponding controls were placed in SVOAs. Aliquots of 2.5, 5.0 and 10.0 mL of 400 ppb TCE stock solution were added to SEAMIST pads in both the control and experimental groups. The total exposure time for the experimental group was 30 seconds. The pads and SVOAs were analyzed for total TCE content within 24 hours of exposure.

Experiment 3. Effect of sample concentration on TCE recovery from SEAMIST pads.

5 mL aliquots of 0, 50, 100, 200, 300 and 400 ppb TCE stock solutions were added to SEAMIST pads in both jars and pre-weighed SVOAs. Each pad from the experimental group was exposed to air for a total of 30 seconds, including handling time, before being sealed in a pre-weighed SVOA. The pads and SVOAs were analyzed for total TCE content within 24 hours of exposure.

RESULTS

As may be seen in Table 1, the total time of air exposure, up to 64 seconds, did not seem to have much affect on the recovery rate of TCE from SEAMIST pads. Of far greater consequence was the act of transferring the pads from one vessel to another. The average handling time for this operation seems to be about 12 seconds, but can range from 4.5 seconds to 18 seconds, depending upon the operator. There was no TCE detected in any of the 0 ppb solutions, and none of these data are reported in the following tables.

Table 1. Effect of time exposure of SEAMIST pads on the recovery of TCE.

Exposure	Total	Handling	TCE	Standard	Percent
Treatment	time	time	(µg/L)	deviation	recovery
(Seconds)	(Seconds)	(Seconds)		(µg/L)
Control	-	-	364	14	100
0	11	11	267	16	73
2	15	13	263	11	72
4	16	12	257	12	71
8	20	12	245	9	67
16	28	12	258	18	71
32	44	12	274	15	75
64	75	11	246	7	67

Table 2 suggests that the TCE recovery rate from SEAMIST pads is constant , about 67%, when the sample volume on each pad is in the range of 2.5 to 10 mL. This is fortunate, as we expect different pad moisture contents to be associated with different depths and strata in the SEAMIST boreholes. It is not likely that we will see sample volumes of 10 mL or greater per pad, because the pad is quit saturated at this volume. When 2.5 mL were added to a pad, dry patches were evidenced on the pad after 1 hour. This suggests that much more liquid could be absorbed.

Table 2. Effect of sample volume on the recovery TCE from SEAMIST pads.

	Control		30 Seconds Exposed
	TCE	Standard	TCE	Standard	Percent
Treatment	µg/L	deviation	(µg/L)	deviation	recovery
mL		(µg/L)		(µg/L)
2.5	497	1	328	8	66
5.0	428	3	287	2	67
10.0	389	8	261	9	67

Table 3. Effect of sample concentration on the recovery of TCE from SEAMIST

	Control		30 Seconds Exposed
Treatment	TCE	Standard	TCE	Standard	Percent
Concentration	µg/L	deviation	(µg/L)	deviation	recovery
(ppb)		(µg/L)		(µg/L)
50	60	1	41	1	68
100	103	5	67	2	65
200	193	3	133	4	69
300	332	2	219	6	66
400	428	2	287	2	67

The results in table 3 indicate that there is no correlation between the TCE concentration in pad-liquid and the recovery rate of TCE from those pads when sample volumes are constant. Again, this is quite reassuring because we expect to see a range of TCE concentration in pore-water between 0 ppb and 400 ppb.

CONCLUSIONS

Results of this study suggest that recovery of TCE from SEAMISTpads is not affected by sample concentration or sample volume. A mean loss of TCE from the pads of 31.2% with a standard deviation of 2.9% was observed when these pads were transferred to the SVOAs used in analysis of VOCs. The conclusion drawn from these studies is that reasonably good estimates of TCE concentrations in soil pore-water may be obtained using the SEAMIST system, providing that adequate control procedures are used.

Work performed under the auspices of the U.S. Department of Energy by The Lawrence Livermore National Laboratory under contract W-7405-Eng-48.