Published online 16 August 2005
Published in Vadose Zone J 4:881-884 (2005)
DOI: 10.2136/vzj2004.0159
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
NOTES
Gas-Phase Partitioning Tracer Test Method for Water Content Measurement
Evaluating Efficacy for a Range of Porous-Medium Textures
Sheng Peng and
Mark L. Brusseau*
Soil, Water and Environmental Science Dep., Mark L. Brusseau, Hydrology and Water Resources Dep., 429 Shantz Building, #38, The Univ. of Arizona, Tucson, AZ 85721
* Corresponding author (brusseau{at}ag.arizona.edu)
Received 28 October 2004.
 |
ABSTRACT
|
|---|
Gas-phase partitioning tracer tests were conducted to measure water content for several porous media with different mean particle diameters and particle-size distributions. Methane was used as the nonreactive tracer and difluoromethane was used as the water-partitioning tracer. Water contents determined from the tracer-test results were similar to values measured gravimetrically. These results support the utility of the gas-phase tracer method for measuring water content in a variety of porous media.
Abbreviations: DFM, diflouromethane • U, uniformity coefficients
|
INTRODUCTION
|
|---|
THE GAS-PHASE partitioning tracer test has been proposed as a method to measure water contents in porous media (Brusseau et al., 1997; Deeds et al., 1999; Kim et al., 1999; Nelson et al., 1999; Carlson et al., 2003; Keller and Brusseau, 2003). This method provides measurements for larger volumes of the subsurface in comparison to traditional gravimetric and other methods (e.g., neutron probe, time-domain reflectometry). For the few tests reported to date, water-content estimates obtained from the tracer method have compared favorably to values determined with other methods. However, the method has been tested for a very limited number of systems. In this study, miscible-displacement experiments were conducted using several porous media with different mean particle diameters and particle-size distributions to further examine the efficacy of the method for porous media of various textures.
|
MATERIALS AND METHODS
|
|---|
Gaseous Tracers
Methane was used as the conservative tracer, given its minimal partitioning to water (e.g., relatively large vapor pressure and low aqueous solubility). Diflouromethane (DFM) was used as the water-partitioning tracer. The concentration of the tracer gases was 100.2 ppm (by volume) for methane and 151 ppm for DFM, which correspond to 0.066 mg L–1 and 0.32 mg L–1, respectively (temperature = 25°C). The tracer gases were mixed with nitrogen (N) carrier gas and stored in high-pressure gas cylinders (Airgas, Inc., Radnor, PA; Spectra Gases, Inc., Alpha, NJ). The physicochemical properties of the tracers are listed in Table 1.
Porous Media
Several porous media were used in this study (see Table 2). The accusand and granusil media are commercially available, well-sorted natural sands (Unimin Corp., Lesueur, MN and Cleburne, TX). Vinton soil (sandy, mixed thermic Typic Torrifluvent) and Hayhook soil (coarse loamy, mixed, superactive, thermic Typic Haplocambids) were collected locally in Tucson, AZ and were sieved to remove the >2-mm fraction before use.
The first five porous media listed in Table 2 have relatively narrow particle-size distributions. Conversely, the Hayhook soil has a much wider particle-size distribution. Uniformity coefficients (U) for these porous media, which is an index of particle-size distribution, were obtained through sieve analysis, where U equals d60/d10. A dispersion solution was used to separate silt and clay from the sand particles before sieve analysis of the Hayhook soil. Surface areas were measured with the N2/BET method by Quantachrome Instruments (Boynton Beach, FL) and Micromeritics Corp. (Norcross, GA).
Gas-Phase Tracer Tests
The experiments were conducted using methods described in Constanza-Robinson and Brusseau (2002a). A stainless steel column (Alltech, Deerfield, IL), 25-cm long and internal diameter of 2.2 cm, was used for the experiments. Water contents were established by adding water to the porous medium before packing the column. A specified amount of deionized water was mixed with the porous medium to achieve a target water content. Subsequently, the column was wet-packed in small increments. The bulk density values obtained for the packed columns are reported in Table 2.
At the start of each experiment, N was injected into the column to establish steady-state flow. A tracer pulse was then injected of sufficient volume for the effluent concentration to reach relative concentration of one. The flow was then switched back to N to elute the tracer. Flow was generated using high-pressure gas cylinders. A digital flow meter was connected to the system to monitor gas flow, which was set to a flow rate of 20 mL min–1 (equivalent to a Darcy velocity of approximately 5.3 cm min–1). Nitrogen and tracer gases were passed through humidifier flasks to reduce potential water stripping. Effluent from the column was ported to an in-line gas chromatograph (with flame ionization detection) for analysis. Output from the gas chromatograph was routed through a voltmeter to a computer for automated data collection. Duplicate experiments were conducted for each tracer/porous medium/water content combination.
The column was weighed both before and after each experiment to monitor for possible water loss. Following the experiments, the porous medium was unpacked from the column and 10 samples were collected along the column length. These 10 subsamples, along with the remaining material, were dried at 106°C for 24 h to determine gravimetric water contents, which were converted to volumetric water contents using the measured soil bulk density. The water contents referred to in the remainder of the paper are volumetric water contents. The measured water contents indicated there was no significant change in water content during the experiments, and that water-content distributions along the columns were uniform. The average values of these water contents are compared to the water contents obtained with the gas-phase tracer method.
Data Analysis
Temporal moment analysis of the tracer breakthrough curves was conducted to obtain retardation factors for DFM, which were determined from the quotient of the mean travel times of DFM and methane. Values for water content were then determined using Eq. [1], which represents a mass-balance distribution of tracer among pertinent phases:
 | [1] |
where R is retardation factor, 
w is volumetric water content, a is volumetric gas content, and KH is dimensionless Henry's Law constant. Equation [1] is developed with the assumption that retention of the tracers by the solid phase and the air–water interface is negligible. Water-content values can be determined with equation one using the tracer-test derived R values and known values of KH and porosity (or a).
Before each set of tracer tests, experiments were conducted to determine the effective dead volume of the apparatus for both tracers. For these experiments, the packed column was replaced with a connector of negligible volume. The average dead-volume travel time for both methane and DFM is 0.44 s. This indicates there is no preferential retention of DFM by the apparatus. The dead-volume travel time was subtracted from the total travel times measured in the column experiments.
|
RESULTS AND DISCUSSION
|
|---|
Representative breakthrough curves for methane and DFM transport in the packed columns are shown in Fig. 1
. These breakthrough curves are relatively symmetrical with sharp arrival and elution fronts, indicating ideal transport. Similar results were observed for all experiments, with the exception of the highest water-content experiment for Hayhook soil. For this experiment, the breakthrough curve exhibited early breakthrough and tailing, indicative of non-ideal transport (see Fig. 2)
. Similar behavior has been observed previously, and is attributed to rate-limited mass transfer associated with solute diffusion within the bulk water (e.g., Popovicova and Brusseau, 1998; Costanza-Robinson and Brusseau, 2002b). Specifically, under certain conditions (e.g., larger water contents), mass transfer of tracer within the immobile water phase may become rate limited with respect to the residence time associated with gas flow, resulting in observed non-ideal transport behavior. As expected, retardation of DFM increased with increasing water content for all porous media.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 1. Representative breakthrough curves for tracer tests; granusil 70 to 100 at water contents of 0.069 and 0.126.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 2. Breakthrough curves for difluoromethane transport in Hayhook soil at water contents of 0.015 and 0.136.
|
|
Volumetric water contents obtained from the tracer-test results are compared to the gravimetrically measured volumetric water contents in Fig. 3
. The water contents determined with the tracer method are similar to the independently measured values. The average uncertainty associated with the tracer-test-derived water contents is calculated to be approximately 1% based on the duplicate results for all experiments. The accuracy (tracer-derived value divided by gravimetric-derived value) for all measurements is 104 ± 8%.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 3. Tracer-test derived volumetric water content vs. gravimetric-derived volumetric water content for all porous media.
|
|
The only system for which there was a significant difference between tracer-test and gravimetric derived values is the highest water-content experiment for Hayhook soil, where the tracer-test-derived water content is approximately 82% of the gravimetric value. This reduced accuracy is attributed to tracer mass-transfer and accessibility constraints associated with higher levels of water saturation. Similar degrees of reduced accuracy at higher water contents have been observed in prior studies of the tracer method (Carlson et al., 2003; Keller and Brusseau, 2003).
To evaluate the influence of the method of water addition on tracer-test results, two experiments were conducted wherein water was imbibed into a column packed with air-dried porous media. This method was used for the granusil 70 to 100 porous medium. The results for these experiments are listed in Table 3. The tracer-derived water contents are approximately 92% of the gravimetric-derived values. This suggests that the method of water addition does not appear to significantly affect the tracer-test results for the two methods used herein. The overall results indicate that the gas-phase partitioning tracer method provides reasonable water contents for a range of porous-media textures and a relatively wide range of water contents for the laboratory system employed in this study.
|
ACKNOWLEDGMENTS
|
|---|
This research was supported by funding provided by the USDA National Research Initiative Program. We thank Dr. Janick Artiola for his assistance, and the reviewers for their useful suggestions.
|
REFERENCES
|
|---|
- Brusseau, M.L., J. Popovicova, and J.A.K. Silva. 1997. Characterizing gas-water interfacial and bulk-water partitioning for gas-phase transport of organic contaminants in unsaturated porous media. Environ. Sci. Technol. 31:1645–1649.[CrossRef]
- Carlson, T.D., M.S. Costanza, J. Keller, P.J. Wierenga, and M.L. Brusseau. 2003. Intermediate-scale tests of the gas-phase partitioning tracer method for measuring soil-water content. Soil Sci. Soc. Am. J. 67:483–486.[Abstract/Free Full Text]
- Costanza-Robinson, M.S. 2001. Elucidation of retention processes governing the transport of volatile organic compounds in unsaturated soil systems. Ph.D. The Univ. of Arizona, Tucson.
- Costanza-Robinson, M.S., and M.L. Brusseau. 2002a. Air-water interfacial areas in unsaturated soils: Evaluation of interfacial domains. Water Resour. Res. 38:13-1–13-17.
- Constanza-Robinson, M.S., and M.L. Brusseau. 2002b. Gas phase advection and dispersion in unsaturated porous media. Water Resour. Res. 38:7-1–7-10.
- Deeds, N.E., D.C. McKinney, G.A. Pope, and G.A. Whitley. 1999. Difluoromethane as partitioning tracer to estimate vadose water saturations. J. Environ. Eng. (NY) July:630–633.
- Daubert, T.E., and R.P. Danner. 1989. Physical and thermodynamic properties of pure chemicals: Data compilation. Hemisphere Publ. Corp., New York.
- Keller, J.M., and M.L. Brusseau. 2003. In-situ characterization of soil-water content using gas-phase partitioning tracer tests: Field-scale evaluation. Environ. Sci. Technol. 37:3141–3144.[Medline]
- Kim, H., P.S.C. Rao, and M.D. Annable. 1999. Gaseous tracer technique for estimating air-water interfacial areas and interface mobility. Soil Sci. Soc. Am. J. 63:1554–1560.[Abstract/Free Full Text]
- Nelson, N.T., M.L. Brusseau, T.D. Carlson, M.S. Costanza, M.H. Young, G.R. Johnson, and P.J. Wierenga. 1999. A gas-phase partitioning tracer method for in situ measurement of soil-water content. Water Resour. Res. 35:3699–3707.[CrossRef]
- Popovicova, J., and M.L. Brusseau. 1998. Contaminant mass transfer during gas-phase transport in unsaturated porous media. Water Resour. Res. 34:83–92.
TOP
ABSTRACT
INTRODUCTION
