VZJ sign up for etocs
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Divine, C. E.
Right arrow Articles by McCray, J. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Divine, C. E.
Right arrow Articles by McCray, J. E.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Divine, C. E.
Right arrow Articles by McCray, J. E.
Related Collections
Right arrow Laboratory Column Studies
Right arrow Soil Pollution
Published in Vadose Zone Journal 2:382-388 (2003)
© 2003 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

ORIGINAL RESEARCH PAPER

Helium and Neon Groundwater Tracers to Measure Residual DNAPL

Laboratory Investigation

Craig E. Divine*,a, William E. Sanfordb and John E. McCraya

a Department of Geology and Geological Engineering, Colorado School of Mines, Golden, CO 80401
b Department of Geosciences, Colorado State University
* Corresponding author (cdivine{at}mines.edu).

Received 4 September 2002.



   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
A laboratory investigation was conducted to evaluate the applicability of dissolved He and Ne as partitioning tracers for detecting and quantifying nonaqueous phase liquid (NAPL) in the saturated zone. Based on the results of batch experiments, the equilibrium NAPL–water partition coefficients (KN,W) of these tracers for two common dense NAPLs (DNAPLs), tetrachloroethene (PCE) and trichloroethene (TCE), are: KPCE,W = 1.28 and KTCE,W = 2.42 for He, and KPCE,W = 1.84 and KTCE,W = 3.24 for Ne. Tracer partitioning is linear across the range of concentrations tested, and appears to be linear even near aqueous solubility limits of the gases. Multiple partitioning tracer tests (PTTs) were conducted in columns, and residual TCE saturations (STCE) ranging from 4.7 to 10.5% were successfully measured by the tracers. Sensitivity analysis for the column experiments indicates that random tracer-measurement error of up to ±20% had little effect on results; however, accurate characterization of the tail region of the tracer curves is particularly important. Therefore, the low analytical detection limits possible with dissolved He and Ne (4 to 5 orders of magnitude below aqueous solubility) may permit better tracer curve characterization than commonly used alcohol partitioning tracers, and is a notable advantage for these tracers. Due to their high Henry's Law constants, these gases will also partition into trapped air present in the tracer sweep zone. Equations are presented for estimating both trapped air and NAPL saturation for PTTs where three phases are present (water, trapped air, and residual NAPL). The results of this investigation provide a basis for field-scale application of dissolved He and Ne as groundwater partitioning tracers.

Abbreviations: BTC, breakthrough curve • DNAPL, dense nonaqueous phase liquid • NAPL, nonaqueous phase liquid • PCE, tetrachloroethene • PTT, partitioning tracer test • TCE, trichloroethene


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
THE SUCCESSFUL design and evaluation of cost-efficient groundwater remediation strategies depend on accurate characterization of contaminant mass and distribution, especially in source areas (i.e., zones where NAPL exists). However, traditional site characterization methods, such as drill-core analysis, groundwater quality evaluation, and site history review, rarely provide enough information to reliably characterize the total volume and distribution of DNAPL. One promising new technique for characterizing subsurface NAPL is the PTT. The PTT, which is analogous to tracer techniques previously developed by researchers in the oil industry (see Deans, 1971; Zemel, 1995), involves the injection of a suite of conservative and partitioning tracers into one or more wells and the subsequent recovery of the tracers from one or more nearby extraction wells. Tracer concentrations measured from extraction well effluent are used to characterize the tracer breakthrough curves (BTCs) at each extraction well. The mean tracer travel times are determined from analysis of the tracer BTCs.

By definition, the transport of a conservative tracer is unaffected by the presence of NAPL encountered in the swept pore volume. Partitioning tracers are partially soluble in NAPL; consequently, their transport will be retarded relative to the conservative tracer if NAPL is present. For a PTT conducted below the water table where gas-phase partitioning is insignificant, the NAPL pore-space saturation (SN) can be calculated from the magnitude of the observed tracer retardation (R) and the NAPL–water partition coefficient (KN,W) for that specific partitioning tracer by (see Jin, 1995):

[1a]

and

[1b]
where p is the mean travel time of the partitioning tracer and c is the mean travel time of the conservative tracer. The tracer-specific equilibrium KN,W for a particular NAPL is defined as

[2]
where CN and CW represent the tracer concentrations in the NAPL and water, respectively. Typically, KN,W values are determined in the laboratory with batch equilibrium partitioning tests. Partitioning tracer tests have also been used to measure trapped air saturation (SA) or water saturation (SW) in the vadose zone (see Heilweil and Solomon, 2002; Deeds et al., 1999). In these respective cases, water or gas is the mobile fluid and KN,W in Eq. [1a] is replaced by either the dimensionless Henry's Law constant (H) or H-1.

Several successful field-scale PTTs conducted for NAPL quantification have been reported to date (see Annable et al., 1998; Mariner et al., 1999; Jawitz et al., 2000; Cain et al., 2000; Meinardus et al., 2002). Aliphatic alcohols have generally been used as partitioning tracers for the majority of PTTs conducted in the saturated zone, although dissolved sulfur hexafluoride has also been used (Nelson and Brusseau, 1996). However, there are some disadvantages to alcohol tracers. For example, Annable et al. (1998) found evidence of concentration-dependent tracer degradation, even during the relatively short duration of the PTT. Slight adsorption of alcohol tracers to natural aquifer material, which may cause tracer retardation unrelated to the presence of NAPL, was reported by Edgar (1997). Wise et al. (1999) show that nonlinear partitioning behavior is observed for some alcohols at high tracer concentrations. Additionally, regulatory concerns may exist regarding the injection of alcohols into the subsurface.

Perhaps the most significant disadvantage associated with alcohol tracers is their relatively high analytical detection limits. Commonly, the range from the injected tracer concentrations (typically near aqueous solubilities) to the analytical detection limits is only two to three orders of magnitude. This limitation can prevent accurate characterization of the tracer BTC, particularly in the tail region of the curve. Tailing is commonly observed in tracer BTCs from both laboratory and field PTTs. This phenomena may be caused by rate-limited partitioning, heterogeneous NAPL architecture (i.e., pools and fingers), and porous media heterogeneity. Most field PTTs are conducted under a forced-gradient where significant tailing of both the conservative and partitioning tracers occurs due to the hydraulics of the injection–extraction well system. Poor characterization of the BTC tail will increase the error in estimates of solute travel times, which will subsequently increase error in estimates of SN.

Noble gases have been successfully used as groundwater tracers in a variety of applications for many years (see Carter et al., 1959; Sugisaki, 1961; Gupta et al., 1994; Sanford et al., 1996; Jardine et al., 1999). Most commonly, dissolved gases have been used to characterize groundwater velocity, flowpaths, and diffusion-related processes. For these applications, the gas tracers are dissolved in the aqueous phase. Helium and Ne, which were used in this study, are nontoxic, chemically inert, nonbiodegradable, do not sorb to soils, and are relatively inexpensive. Additionally, they have low atmospheric concentrations, detection limits four to five orders of magnitude lower than their aqueous solubilities, and can be easily measured on an inexpensive gas chromatograph. Furthermore, long-term source concentrations are relatively easy to maintain (e.g., Sanford and Solomon [1998] conducted a natural-gradient field tracer test where dissolved gas source concentrations were maintained for more than a year). The theoretical aqueous solubilities, natural background concentrations, and Henry's Law constants for He and Ne are summarized in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. Physicochemical properties and analytical detection limits for He and Ne.

 
Helium and Ne have not been previously investigated as NAPL partitioning tracers, and the potential advantages of these tracers discussed above motivated this experimental study. Specifically, this investigation was designed to evaluate the potential for using He and Ne tracers for NAPL quantification in field PTTs. The KN,W values for He and Ne tracers were determined from batch tests for two commonly occurring NAPLs, PCE and TCE. A series of column PTTs with these tracers were then conducted to evaluate tracer performance under dynamic conditions, and column tests results were analyzed through tracer BTC interpretation and sensitivity analysis. The potential performance of these tracers in the presence of trapped air is also discussed, and equations for estimating both SN and SA are presented for PTTs where three phases are present (mobile water, immobile NAPL, and immobile trapped air). The results of this work provide a basis for field application of these tracers.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 
Batch Partitioning Experiments
Batch equilibrium partitioning experiments were performed to determine the PCE–water (KPCE,W) and TCE–water (KTCE,W) equilibrium partition coefficients for He and Ne. Approximately equal volumes of laboratory-grade PCE or TCE (dyed red with about 40 mg L-1 Sudan IV) and He- or Ne-charged water were injected into 10-mL glass syringes. The syringes were then placed in a mechanical shaker for 45 to 60 min. Shaking was brisk but did not create visible emulsions. Data from preliminary tests indicate that partitioning equilibrium during batch tests was reached with <5 min of shaking (note that this batch test equilibrium time is not indicative of in situ PTT equilibrium times). The NAPL and water fractions were then each analyzed for dissolved gas concentration to determine the NAPL–water partition coefficient (Eq. [2]). The TCE batch tests were performed for a 2.5-order-of-magnitude range of dissolved concentrations consistent with the aqueous tracer concentration range measured for the column PTTs (approximately 0.02–5 mL L-1 for TCE). Mass balance calculations were performed to confirm that degassing did not occur during the tests.

Water samples were analyzed for dissolved He and Ne by headspace analysis with a gas chromatograph (Shimadzu GC-8A, Shimadzu Corp., Kyoto, Japan) equipped with a thermal conductivity detector and a molecular sieve column (5Å, 60/80 mesh, with N2 carrier gas). Natural background concentrations of He and Ne are approximately an order of magnitude lower than the quantitation limits achieved in this study; however, instrument responses for background He and Ne were observed in laboratory water. It is likely that minor modifications to the analytical set-up (using Ar carrier gas, lowering column velocities), could permit quantitation of these low background concentrations. The estimated analytical limits and precision achieved for this study and by Sanford et al. (1996) are provided in Table 1.

Column Partitioning Tracer Tests
Ten column PTTs (including two with no NAPL) were conducted in 30-cm long, 5-cm diameter, glass liquid-chromatography columns packed with a well-sorted medium Ottawa sand (660–925 µm, 20–30 mesh). Column pore volumes ranged from 205 to 230 mL. During column construction, 5-cm lifts of sand were poured into the column, which was partially filled with deaired water. During each lift, approximately equal volumes of dyed NAPL (TCE) were introduced into the water-saturated sand with a syringe equipped with a 20-cm-long needle. The sand was mechanically mixed with a stirring rod and then tamped after each lift to create a relatively homogeneous NAPL distribution and prevent the formation of continuous NAPL zones. Based on visual observation, the diameters of individual NAPL zones were several millimeters or smaller. Columns were constructed with TCE saturation (STCE) values ranging from 4.7 to 10.5%. To minimize the possibility of continuous or mobile NAPL, experiments were not conducted with STCE values >10.5%. Immediately before tracer tests, the columns were flushed with several pore volumes of deaired water.

Helium and Ne were dissolved into the tracer reservoir water by methods similar to those of Wilson and Mackay (1993). Dissolved gas concentrations in the tracer reservoir ranged from 38 to 98% of the aqueous solubility. Bromide (as CaBr2) was used as the conservative tracer, and tracer reservoir concentrations ranged from 44 to 80 mg L-1. The columns were oriented vertically, and upward water flow was maintained with a peristaltic pump. Column influent samples were collected to determine the exact tracer source concentrations, and effluent samples were analyzed to characterize the tracer BTCs.

Water samples were analyzed for He and Ne by the methods described in the previous section, and for Br- with an ion-selective electrode and a single junction reference electrode. The estimated precision for Br- analysis was ± 1.5%, and the lower detection limit was approximately 0.5 mg L-1. For additional information on the experimental methods, see Divine (2000).

Breakthrough Curve Analysis
Temporal moments can provide meaningful information regarding solute transport for one-dimensional miscible-displacement systems. For example, the first temporal moment represents the mean solute travel time (), and the second and third temporal moments express the degree of spreading and asymmetry about the mean, respectively. When the tracer input is constant for a finite period of time (ts), the mean solute travel time is given by

[3]
where t is the measurement time and C(t) is the tracer concentration with time at the extraction well (i.e., the tracer BTC). Equations [1a] and [1b] are then used to calculate SN.

Temporal moments were estimated by direct numerical integration of the BTC (Direct Integration method) and from infinite data distributions generated from solute transport models fitted to the observed BTC (Model Fitting method). Often, experimental BTC data is sparse, truncated, and noisy, which increases the error of moment estimates. Because a fitted model can interpolate between measurements and extrapolate beyond measurements in the BTC tail region, the Model Fitting method can minimize the effects of poor BTC characterization and early test termination. However, the accuracy of the estimated moments is directly related to the degree to which the fitted model accurately reproduces and represents the actual BTC. For this analysis, the fitted models were used primarily to improve temporal moment estimates, and as pointed out by Helms (1997), the only requirement for a fitted model in this application is that it sufficiently reproduces the actual tracer BTC. That is, the model need not be physically meaningful to improve temporal moment estimation (e.g., this is the case with the "exponential extrapolation" tail fitting method [Jin, 1995]). Of course, if the fitted model is physically appropriate for the system, it may also be used to evaluate other transport processes and the fitted model parameters may have physical significance.

The unknown parameters in the transport models were estimated using CXTFIT 2.1, a parameter estimation program that uses a nonlinear least-squares optimization approach to estimate transport parameters (Toride et al., 1995). A commonly used one-dimensional advection–dispersion equilibrium model was fitted to the Br- BTC. The greater tailing observed for the partitioning tracer BTCs was poorly fit with the equilibrium model and was assumed to be caused by rate-limited partitioning. Therefore, a physical two-region (i.e., "dual porosity") nonequilibrium solute transport model was used to estimate temporal moments for the partitioning tracers. As noted above, the primary purpose of fitting a model was to improve temporal moment estimates; however, the nonequilibrium model may also have some physical significance for partitioning tracer transport. Conceptually, the water-saturated zone may be represented by a flowing region, and the component of the NAPL zone where partitioning is rate limited may be represented by a stagnant region. Solute transfer between the two regions is assumed to be diffusion controlled. It should be noted that fitted model parameters related to the degree of nonequilibrium transport were completely unconstrained, and CXTFIT was free to fit the BTC data with parameter values equivalent to the equilibrium model if these values yielded the best data fit. Presentation of the governing equations, appropriate boundary conditions, and solutions for these models are presented in Toride et al. (1995).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND CONCLUSIONS
 REFERENCES
 
Batch Tests
The results of the experimental isotherms for the batch tests are shown in Fig. 1. Based on these results, dissolved gas partitioning appears to be linear for the range of measurements, including equilibrium aqueous concentrations up to about 35% of the aqueous solubilities. Nonlinear partitioning isotherms have been observed for some alcohol tracers when concentrations are relatively high (10–30% of the aqueous solubility or higher) (Wise et al., 1999). This occurs because as the tracer mole fraction in the NAPL increases, the tracer activity coefficient in the NAPL decreases. This can result in much greater proportional NAPL partitioning at high tracer concentrations (i.e., partitioning is nonlinear). However, Wise (1999) shows that this phenomenon is generally only significant when tracer mole fractions in the NAPL are fairly high (greater than ~0.05–0.2). Since the maximum mole fraction aqueous concentrations achievable (i.e., the aqueous solubilities) for He and Ne are several orders of magnitude lower than this (7.00 x 10-6 and 8.15 x 10-6 at 20°C and 0.101 MPa [1 atm] for He and Ne, respectively [Lide, 1993]), the maximum tracer mole fraction in the NAPL due to partitioning will also be very low. Therefore, it is reasonable to assume that tracer activity coefficients in the NAPL do not decrease significantly even when aqueous concentrations of He and Ne are near solubility. Consequently, the partitioning behavior of these gases is expected to be linear over the practical application range, including concentrations near the aqueous solubilities, and the experimental isotherms clearly support this.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 1. Helium and Ne batch equilibrium partitioning isotherms: (a) He tracer with trichloroethene (TCE) nonaqueous phase liquid (NAPL); (b) Ne tracer with TCE NAPL; (c) He tracer with tetrachloroethene (PCE) NAPL; (d) Ne tracer with PCE NAPL. Isotherms are plotted on log-log scale to more clearly show partitioning behavior for the entire measurement range. Confidence bars indicate analytical instrument precision.

 
Since the batch test data and the analysis presented above indicate that partitioning is linear for the practical application range, KN,W values were determined by averaging all individual batch test results for each series, and confidence intervals were estimated with a two-sided t statistic. By this analysis, equilibrium partition coefficients and 95% confidence intervals are KPCE,W = 1.28 (±10%) and KTCE,W = 2.42 (±10%) for He, and KPCE,W = 1.84 (±10%) and KTCE,W = 3.24 (±9%) for Ne (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Summary of batch partitioning experiments.{dagger}

 
Column Tests
Tracer BTCs for several column PTTs are presented in Fig. 2. As shown in these figures, the fitted transport models adequately reproduced observed BTCs; therefore, they were considered appropriate for temporal moment estimation. Average absolute STCE estimation error was similar for both BTC analysis techniques (14% for Direct Integration and 13% for Model Fitting). However, the Direct Integration method underestimated STCE overall (average error = -9%), while the Model Fitting method generally overestimated STCE (average error = +8%). A plot of actual vs. calculated STCE for the column PTTs is provided as Fig. 3.



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 2. Tracer breakthrough curves for columns constructed with (a) STCE = 5.2%, (b) 8.3%, and (c) 9.7%. The results of two separate tests conducted on the same column are superimposed for the bottom figure.

 


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 3. Comparison of actual versus calculated STCE values for the column partitioning tracer tests.

 
As mentioned above, physically appropriate models can also provide meaningful solute transport information in addition to improving temporal moment estimates. For example, application of the equilibrium model to nonreactive solute transport in columns is well established, and this model yielded the best fit for all Br- BTCs and for dissolved gas BTCs when STCE = 0. Therefore, the fitted parameters (velocity and dispersivity) applied to the Br- data are believed to be physically meaningful. For all PTTs where NAPL was present, the nonequilibrium model provided the best fit for the partitioning tracer BTCs. Furthermore, the fitted parameters indicative of nonequilibrium transport were strongly correlated with the magnitude of STCE (i.e., at higher STCE values, the degree of nonequilibrium transport indicated by these parameters increased). These results suggest that the nonequilibrium model and associated transport parameters may have physical significance for this application; however, further evaluation and interpretation of this model and these experimental results are necessary.

The ability of the analytical equipment to achieve the detection limits mentioned above for these gases was demonstrated through analysis of low-concentration standards. Since the primary purpose of the column tests was to evaluate the general suitability of these gases as groundwater partitioning tracers, column PTTs were generally terminated for logistical considerations once the normalized effluent partitioning tracer concentrations reached about 0.001 to 0.03. A sensitivity analysis was performed to further understand the significance of this BTC truncation, as well as the effect of sampling frequency, tracer-measurement error, and moment-estimation technique upon SN estimation.

A synthetic finite BTC similar to the experimental BTCs was created with the nonequilibrium model. This synthetic BTC was then systematically modified by introducing random measurement error (via a random number generator), reducing measurement frequency, and truncating the tail. Analysis of the modified synthetic BTCs indicates that random measurement error of up to ±20% had little effect on estimation of SN. However, sparse measurement in the tail region (less frequent than every 0.2 pore volumes) and truncation (test termination when the normalized effluent concentration reaches 0.02) can introduce significant error, especially for the Direct Integration method. This indicates that for these column tests a low relative tracer detection limit and frequent sampling are more important than good measurement precision. Furthermore, extending the time until test termination might have reduced SN estimation errors for these tests and is probably an important consideration for field PTTs.

In the sensitivity analysis, the Direct Integration method consistently underestimated tracer mass recovery and SN, while the Model Fitting method frequently overestimated SN. This trend generally agrees with the results of the experimental column PTTs, where the Model Fitting method overestimated SN for six of the eight experiments. Most likely, the Direct Integration method tends to underestimate SN because of integration error and truncation in the tail region. The Model Fitting method may tend to overestimate SN because CXTFIT's least-squares fitting scheme may be less sensitive to (i.e., underweight) late-time low concentration values in the tail region of the BTC. This is supported by the fact that the synthetic BTCs used in the sensitivity analysis were initially created with the nonequilibrium model.

Three-Phase Partitioning
In theory, the PTT technique is appropriate for measurement of both DNAPL and light NAPL (LNAPL). However, because of their high Henry's Law constants, dissolved He and Ne will be retarded by trapped air as a result of gas-phase partitioning. Trapped air may be present near the water table in unconfined aquifers because of seasonal water level changes or variable pumping rates from extraction wells. Of course, dissolved gases could be used to quantify trapped air saturation (SA) for systems without NAPL by Eq. [1a] (where the Henry's Law constant is substituted for KN,W). For systems with both residual NAPL and trapped air, both NAPL- and gas-phase partitioning by dissolved gases must be accounted for.

Mariner et al. (1999) presented equations describing tracer retardation due to partitioning into both NAPL and residual water for a vadose zone PTT (in this case, the tracer is transported in the gas phase). Using a parallel analysis, tracer retardation due to partitioning into both trapped air and residual NAPL where water is the mobile phase is described by

[4]

To demonstrate the usefulness of Eq. [4], we briefly present results for one laboratory PTT where an air bubble was discovered in the column during column deconstruction (results for this test are not included in Fig. 3). If the effects of the air bubble are ignored (i.e., assuming SA = 0), the anticipated retardation of the He tracer is 1.28. However, the actual retardation determined from the observed BTC was notably higher: 1.47 by Direct Integration and 1.51 by Model Fitting. Based on a visual estimation of the bubble volume (0.2–0.6 mL) and the known STCE (10.3%), the predicted retardation range calculated by Eq. [4] is 1.38 to 1.58, which agrees with the observed retardation. The air bubble volume calculated directly from the observed retardation is 0.5 mL.

If a minimum of two partitioning tracers with differing H and KN,W values and a conservative tracer (a tracer that does not partition into NAPL or air) are used in the PTT, both SN and SA can be quantified by solving a system of three equations including Eq. [4] (describing two different partitioning tracers) and the following saturation constraint:

[5]

In this case, SN and SA are calculated explicitly by

[6a]

[6b]
where the parameters associated with the specific multiphase partitioning tracers i and j are indicated by the respective superscripts (i.e., Hi indicates the Henry's Law constant for tracer i). More general forms of these equations can be solved for a system of three partitioning tracers (i.e., a conservative tracer is unnecessary). However, in practice it is convenient to include a conservative tracer in the tracer suite so that the velocity of the mobile fluid is measured directly, and so the difference between tracer travel times is maximized. Furthermore, there is typically minimal error associated with the H and KN,W values for conservative tracers, since they are strictly defined as tracers that do not measurably partition into the immobile phases.


   SUMMARY AND CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
REFERENCES
 
Dissolved He and Ne exhibit several potential advantages over alcohol partitioning tracers. They are nontoxic, inert, nonbiodegradable, relatively inexpensive, and have low analytical detection limits. Additionally, unlike some alcohol tracers, equilibrium partitioning appears to be linear near the aqueous solubilities. This is important because standard PTT analysis methods are predicated on the assumption of linear partitioning. For field-scale PTTs, Jin (1995) recommended selection of partitioning tracers that will exhibit retardation of 1.2 to 4.0 to maximize partitioning tracer separation without substantially increasing the length of the tracer test. Therefore, He and Ne may be best suited as partitioning tracers when SN values are relatively high. If trapped air is present in the aquifer, application of these tracers may be more complicated due to both NAPL- and gas-phase partitioning. However, it may be possible to estimate both residual NAPL and trapped air saturation in these cases if the appropriate tracer suite is selected. On the basis of the results from this study, we recommend application of these tracers in a field-scale PTT, preferably in conjunction with alcohol partitioning tracers that have demonstrated previous success.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Vadose Zone JHome page
J. C. Seaman, P. M. Bertsch, M. Wilson, J. Singer, F. Majs, and S. A. Aburime
Tracer Migration in a Radially Divergent Flow Field: Longitudinal Dispersivity and Anionic Tracer Retardation
Vadose Zone J., May 17, 2007; 6(2): 373 - 386.
[Abstract] [Full Text] [PDF]

Home page
 Environmental and Engineering GeoscienceHome page
C. GULER and G. D. THYNE
Statistical Clustering of Major Solutes: Use as a Tracer for Evaluating Interbasin Groundwater Flow into Indian Wells Valley, California
Environmental and Engineering Geoscience, February 1, 2006; 12(1): 53 - 65.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Divine, C. E.
Right arrow Articles by McCray, J. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Divine, C. E.
Right arrow Articles by McCray, J. E.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Divine, C. E.
Right arrow Articles by McCray, J. E.
Related Collections
Right arrow Laboratory Column Studies
Right arrow Soil Pollution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome