Vadose Zone Journal 2:168-176 (2003)
© 2003 Soil Science Society of America
SPECIAL SUBMISSIONS: Contaminant Characterization, Transport, and Remediation in Complex Multiphase Systems
Surfactant Effects on Residual Water and Oil Saturations in Porous Media
Jason E. Flaming*,
Robert C. Knox,
David A. Sabatini and
Tohren C. Kibbey
University of Oklahoma, School of Civil and Environmental Engineering and Science, University of Oklahoma, 202 W. Boyd Street, Carson Engineering Center, Rm. 334, Norman, OK 73019-1021
* Corresponding author (jason.flaming{at}tinker.af.mil)
Received 12 November 2002.
 |
ABSTRACT
|
|---|
A series of soil column tests was performed to determine surfactant effects on residual water and oil saturations in porous media. In particular, these tests focused on the impact of submicellar surfactant solutions and the potential application of these low concentration systems to light nonaqueous phase liquid (LNAPL) contamination in the vadose zone. One set of tests involved surfactant flushing in soil-filled columns followed by drainage to residual water saturation and LNAPL injection to determine the subsequent residual LNAPL saturation. Another set of tests involved surfactant application to a soil-filled column already holding residual LNAPL saturation to promote the release of the previously trapped LNAPL. Test results showed surfactant systems could reduce both residual water and oil saturations by up to 50%. In addition, submicellar surfactant systems were equally effective as supramicellar solutions in reducing residual water saturations and potentially more effective at reducing residual oil saturations. Submicellar surfactant applications to a medium-grained (0.85–0.425 mm) soil containing residual LNAPL saturations were effective at releasing up to 50% of the previously trapped residual LNAPL. These applications were less successful in a fine-grained soil as full drainage of water and LNAPL was unachievable due to high capillary pressures. Overall, observations suggest low concentration surfactant solutions may have the ability to release significant amounts of previously trapped LNAPL in the vadose zone, potentially increasing free-product recovery rates and lowering LNAPL saturations to levels more favorable for biodegradation. The decrease in overall saturations (both water and oil) in a contaminated vadose zone could also present an increase in air permeability, thus enhancing other vadose zone treatment technologies such as bioventing or soil vapor extraction.
Abbreviations: CMC, critical micelle concentration • CRA, Canadian river alluvium • DADI, deaired deionized • IFT, interfacial tension • LNAPL, light nonaqueous phase liquid • NAPL, nonaqueous phase liquid • SDBS, sodium dodecylbenzenesulfonate
|
INTRODUCTION
|
|---|
SURFACTANT FLUSHING has emerged as a viable technology for remediation of subsurface LNAPL contamination. Surfactant-enhanced solubilization significantly increases the apparent aqueous concentration of the NAPL via micellar partitioning, thereby reducing the number of pore volumes that must be pumped to extract the NAPL. Surfactant-enhanced mobilization utilizes ultra low interfacial tensions to significantly reduce the capillary forces trapping the NAPL, thereby allowing the oil to be readily extracted as a separate phase. Surfactant-based remediation technologies have been the focus of numerous research activities (Sabatini et al., 1995, 1996) and have recently been applied in numerous field-scale studies to evaluate full-scale applicability (Knox et al., 1997, 1999). The success of these and other surfactant remediation studies have demonstrated that this innovative technology can effectively alleviate groundwater contamination. At the same time, there are ongoing research efforts to further improve the technology.
Two recent field demonstrations of surfactant-enhanced LNAPL mobilization (Surbec-ART Environmental, LLC, personal communication, 1999) produced anecdotal evidence of increased postremediation free-phase mobility; that is, the free-product recovery rates increased fivefold following surfactant flushing. It is hypothesized that a possible mechanism of the enhanced postflush LNAPL recovery could be due to surfactant alterations of the unsaturated soil immediately above the water table (i.e., "smear zone"). It is suspected that surfactant solutions can lower residual water saturations in the vadose zone, creating a drainage effect that leads to the liberation of previously trapped LNAPL. Similar behavior was observed by Chevalier et al. (1994) in two-dimensional sand-tank studies. In addition, studies by Desai et al. (1992) showed that surfactant solutions have a dramatic impact on soil water retention curves, which are commonly used to describe vadose zone behavior. Other studies on surfactant impacts in the vadose zone have shown that surfactant solutions can induce unsaturated flow (Corey, 1994; Karkare and Fort, 1994; Henry et al., 1999). Of particular interest to this research are computer modeling efforts of multiphase systems by Blunt (1997). The effects of media wettability on residual oil saturation were evaluated and showed residual oil saturations can range from 0.05 to 0.60.
Figure 1 presents the modeling results of residual oil variability associated with initial water content for strong and weak water-wet conditions. Figure 1 illustrates the amount of trapped oil that will exist in the subsurface as a function of wettability and water saturation. Of key interest is the reduction of residual oil saturation as residual water saturation decreases below 0.4. This behavior shows that, under moderately water-wet conditions, a reduction in residual water saturation (such as surfactant-enhanced water drainage in the vadose zone) can cause a significant reduction in residual oil saturation, hence liberating previously trapped LNAPL and increasing free-product recovery.
For remediation applications, surfactant solutions are typically formulated well above the critical micelle concentration (CMC) and usually exhibit concentrations from 2% weight up to 15% weight. The objective of this study was to assess the viability of utilizing low concentrations of surfactant solutions (concentrations ranging from 0.15 to 0.005% weight) to enhance free-product recovery and lower residual oil saturations. The purpose of evaluating low-level surfactant concentrations is to use the water–air surface tension reductions (which are achieved at the submicellar level) to lower residual water saturations, which, in turn, could lower residual oil saturations. Achieving these saturation reductions with low levels of surfactants could prove to be cost-effective as well as reducing the impacts of introducing chemicals to the subsurface conditions.
|
MATERIALS AND METHODS
|
|---|
A series of soil column studies was performed to determine the effects of submicellar surfactant solutions on residual water and oil saturations in two types of porous media. Surfactant solution surface tensions and interfacial tensions with a model LNAPL were also measured to provide interfacial properties associated with the column tests. Sodium chloride additions were made to selected 500-mL batches of surfactant solutions.
LNAPL, Surfactant, and Media Selection
Hexadecane (C16H34) was selected as the model LNAPL because of its strong hydrophobicity, low volatility, and high interfacial tension with water. It was anticipated that these characteristics would help reduce experimental error by minimizing LNAPL losses to evaporation and dissolution into the water phase. Deaired deionized (DADI) water was used as the pure water phase and as water used for mixing the surfactant solutions. Hexadecane 99% (Sigma-Aldrich, St. Louis, MO) was used for the soil column tests and interfacial tension (IFT) measurements with water and various surfactant solutions. Sodium dodecylbenzenesulfonate (SDBS) was chosen as the surfactant of interest on the basis of experimental data provided in Rouse et al. (1993). Sodium dodecylbenzenesulfonate was selected due to its significant reduction in water–air surface tension at the submicellar level. Rouse also provided surfactant sorption data on Canadian river alluvium (CRA) media. In addition, a soil analysis of the CRA was also provided that shows the CRA to consist of 91% sand, 2% silt, 7% clay, and an organic C content of 0.0024. Given the data provided and the readily available CRA media, the fine-grained CRA was chosen as one media used in the column tests. In addition, a medium-grained 0.85- to 0.425-mm (20–40 mesh) silica sand (Oglebay Norton Industrial Sands, Cleveland, OH) was also obtained for use as another media utilized for comparative purposes in the column tests.
Interfacial Property Tests
The properties of concern measured for each column test were the IFT between the wetting fluid (water or surfactant solution) and NAPL (hexadecane) and the surface tension of the wetting fluid. The IFT measurements were performed on a spinning drop interfacial tensiometer. Surface tension measurements for water and the surfactant solutions of interest were measured by the capillary rise technique (Adamson, 1990).
Soil Column Tests
Surfactant Applied Before Oil Injection
Soil column tests were conducted to determine residual water-phase and NAPL saturations in porous media that have been subjected to varying levels of submicellar surfactant flooding. Four chromatography columns (4.8-cm i.d., approximately 29 cm long), obtained from Kontes, Inc. (Vineland, NJ) were used for the soil column tests. The exact length for each packed column was determined by measuring the height of soil in the column. The pore volume obtained in the columns was approximately 200 mL. To obtain homogenous soil packing, the columns were filled with water while dry soil was poured slowly and continuously into the column. The column was occasionally tapped to allow the soil to pack while filling. This method was effective at eliminating layers such as clay lenses that develop with the intermittent filling and packing technique. After packing, a minimum of five pore volumes of DADI water were flushed through each column to remove any mobile fine grains and thoroughly wash the soil before proceeding to the residual saturation determinations. All packed columns showed no visual signs of heterogeneity, and during water drainage no areas of soil in the columns showed greater resistance to water drainage (i.e., no water pockets or preferential air pathways were observed). The mass of soil added and column volume were used to determine properties such as bulk density and porosity, and allowed water saturations to be measured gravimetrically (i.e., by weighing the column). The bulk density average of 29 packed columns was 1.64 g mL-1 with a standard deviation of 0.033.
Due to the capillary pressures created by the packed soil in the column, fluids in the soil columns (i.e., water, surfactant solution, and NAPL) would not gravity drain to residual saturation. Therefore, an applied air pressure was used to drain the columns to residual water saturation. Applying pressure to a column to simulate drainage is a common technique used to develop capillary pressure–residual saturations curves (soil water retention curves). However, this technique is commonly a two-phase technique that utilizes a NAPL to displace water in the soil column. To simulate an unsaturated column, humidified air was used to displace the nonresidual water to allow for the determination of subsequent residual NAPL saturation for the column. The applied air pressure technique was evaluated several times to reach desired pressures that would assure full drainage of water while minimizing evaporative losses in the column. To determine the air pressure needed to reach residual water saturation in each soil, air pressure was applied in 25.4-cm (10 inch) H2O increments while visual observations were noted. For each soil, there was a point where an increase in pressure did not create any further water drainage. This occurred at 100 cm H2O for the silica sand, and 200 cm H2O for the CRA. Therefore, as a safety factor, 125 cm H2O and 250 cm H2O were used respectively to create full water drainage in the packed column. This air pressure was applied to the column for 4 h (visual observations showed full drainage was achieved in approximately 3 h) in the water-saturated columns, and near-complete surfactant solution drainage allowed for minimal use of air pressure to achieve residual saturation. The air stream was saturated through a "humidifier" to reduce evaporation during drainage (Fig. 2). The drained water volume was measured to determine the extent of evaporation in the column due to the applied air pressure. Comparisons were made to the water saturation measurements determined by the above-mentioned gravimetric method and showed that evaporative effects were consistent in all water drainage tests. These effects were very small in the surfactant drainage tests as the surfactant solutions experienced near complete drainage and did not require sustained air pressure.
After the column reached residual water saturation, hexadecane was slowly injected up the column at 0.1 mL min-1 until a minimum of two pore volumes of hexadecane was flushed through the entire column. Residual oil saturation was then achieved by flushing approximately two pore volumes of water at 0.1 mL min-1 up the column until all free-phase LNAPL was removed. After all free-phase LNAPL was removed, the water injection rate was dramatically increased to 20 mL min-1 and sustained for five pore volumes of water flushing to assure no further recovery of free-phase LNAPL. The LNAPL saturations were measured volumetrically (i.e., volume added - volume drained) to determine the residual LNAPL saturation in each column. This method of LNAPL saturation determination did pose interferences such as loss of LNAPL to the water phase or sorbed onto the soil, but these effects were assumed to be negligible. For instance, given an average pore volume of 200 mL, the soil columns at residual saturation typically contained 10 to 40 mL of water or surfactant solution, which leads to a negligible amount of LNAPL that could be partitioned into the water phase when considering typical solubility values of NAPL into surfactant solutions. Light nonaqueous phase liquid sorption onto the soil was not determined, and it is noted that for every 2 mL or 1.5 g of LNAPL lost in the soil column there is a saturation error of approximately 1%. The amount of soil added to each column ranged from 800 to 875 g; therefore, each 1% error is equivalent to a LNAPL/soil sorption rate of approximately 1.75 mg LNAPL g-1 soil.
Surfactant Applied with Oil in Place
Another series of soil column tests was performed to examine each surfactant's ability to reduce residual oil saturations already in place in the soil column. The focus of these experiments was to determine the amount of free-phase LNAPL released after a surfactant solution was applied to the column. The first step was to establish residual oil saturation within media subjected to DADI water flushing and drainage (i.e., baseline residual water saturation). This was followed by upward surfactant flushing at 0.1 mL min-1. This flushing rate was assumed to be slow enough not to cause any reestablishment of LNAPL (e.g., one pore volume flush took more than 30 h to achieve). Approximately 250 mL of surfactant solution was flushed through the column, and no free-phase LNAPL was removed from the column during the flushing procedure. After flushing, the column was allowed to gravity drain for visual observations and volumetric measurements of released LNAPL. After gravity drainage reached equilibrium (approximately 5 d), air pressure was applied to complete full column drainage.
|
RESULTS AND DISCUSSION
|
|---|
Surfactant Screening
The water–air surface tensions vs. surfactant concentration relationships for the surfactant system are shown in Fig. 3. This figure shows that SDBS has the ability to reach a very low surface tension (<35 mN m-1) at submicellar concentrations, thus providing a wide range of surface tensions that could be evaluated for the soil column tests.
Oil–water IFTs for increasing SDBS concentration are shown in Fig. 4. To estimate the CMC using this data, no electrolyte was added to the surfactant solution because of the effects of electrolyte shifting the CMC. The CMC for SDBS has been reported to range from 400 mg L-1 (Acosta, personal communication, 2001) to 1400 mg L-1 (Rouse et al., 1993). The CMC break for the system appeared to occur around 700 mg L-1, which is in good agreement with these literature values. Overall, the interfacial property tests performed on SDBS showed that the surface tensions for submicellar solutions of SDBS range from about 65 mN m-1 down to approximately 32 mN m-1. Similarly, the IFTs between the LNAPL and submicellar solutions of SDBS range from 48 mN m-1 down to approximately 8 mN m-1. This provided a wide range of interfacial activity of the surfactant solutions used in the soil column tests.
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 4. Interfacial tensions (IFT) of sodium dodecylbenzenesulfonate (SDBS)–hexadecane with estimated critical micelle concentration.
|
|
The presence of salt in the surfactant solutions was also desired in some of the column tests since salts are naturally present in the subsurface. Therefore, 100 and 200 mg L-1 SDBS solutions were evaluated with varying salt concentrations. The IFTs between the LNAPL and 100 and 200 mg L-1 SDBS solutions with varying NaCl concentrations are presented in Fig. 5. Sodium chloride was selected over CaCl2 for salt additions (another common electrolyte additive for surfactant remediation applications) because CaCl2 additions to the submicellar SDBS solutions formed a precipitate, whereas NaCl additions did not form a precipitate. The indicated data points (arrows) in Fig. 5 present the surfactant solutions chosen for soil column tests. Both systems were applied in duplicate to both soils (CRA and silica sand) obtained for the study.
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 5. Salt screen for LNAPL–selected sodium dodecylbenzenesulfonate (SDBS) concentrations (arrows indicate systems used in column tests). IFT is interfacial tension.
|
|
Surfactant Effects on Residual Water Saturation
Figure 6 depicts the water residual saturation data, normalized with respect to their media-associated baseline residual saturation (0.12 for the silica sand, 0.2 for CRA). The average residual water saturations determined from six column tests with DADI water were used as the baseline residual saturations for each media. Figure 6 shows that the 0.85- to 0.425-mm (20–40 mesh) silica sand may have experienced greater residual water saturation reductions than the CRA media at submicellar surfactant concentrations (e.g., see surface tensions of 55 mN m-1). One possible explanation for why residual water saturation reductions in the CRA media were less responsive in this range could be the presence of clay minerals in the CRA. It is possible that the surface tension reductions were not sufficient enough to overcome the water layer surrounding these clay particles, thus having a lesser impact on water saturation reductions in the CRA media. In addition, the surfactant effects on residual water saturations in the silica sand are achieved at very low surfactant concentrations (i.e., 100 mg L-1 SDBS applied to the silica sand resulted in a reduction of approximately 35%). The SDBS solutions with surface tension values <50 mN m-1 appeared to be equally effective in both soils. Visual observations also suggest dramatic impacts on water saturations are apparent at the submicellar level. All surfactant systems containing 100 mg L-1 SDBS or more resulted in complete gravity drainage of the water phase in the silica sand; that is, subsequent pressure application could not produce any more water drainage than what occurred during gravity drainage. In contrast, DADI water could only gravity drain the silica sand to approximately 25% water saturation with about 5 to 8 cm of saturated soil at the bottom of the column.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 6. Normalized residual water saturations vs. surfactant surface tension for the silica sand and Canadian river alluvium (CRA).
|
|
Although none of the surfactant systems could completely gravity drain out of the CRA media, their effects on gravity drainage were still pronounced. Deaired, deionized water alone (no surfactant) was not able to gravity drain out of the CRA-filled column. However, surfactant solutions were able to promote significant gravity drainage, leaving only approximately 5 cm of saturated soil remaining in the column. The CRA media exhibited a 50% reduction in water saturation, but only with the lower surface tension (i.e., higher surfactant concentration) systems.
Surfactant Effects on Subsequent Residual Oil Saturations
The residual oil saturations determined from the water- and surfactant-drained soil columns are presented in Fig. 7 and 8. The initial water saturations are plotted because they had a direct effect on the subsequent oil saturation. The initial water saturation values include water-phase evaporation during pressure draining of the water phase. In general, the lower initial water saturations resulted in lower residual oil saturations for the submicellar solutions.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 7. Surfactant effects on residual water and oil saturations in 0.85- to 0.425-mm (20–40 mesh) silica sand. DADI is deaired, deionized water.
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 8. Surfactant effects on residual water and oil saturations in Canadian river alluvium (CRA) media. DADI is deaired, deionized water.
|
|
Surfactant impacts on residual water and oil saturations were apparent at submicellar SDBS concentrations. It appears that the surfactant systems of 100 and 400 mg L-1 SDBS with no salt added performed best; these systems lowered residual water and oil saturations by 40 to 50% in both media tested.
Overall, the submicellar SDBS solutions performed better at reducing residual water and oil saturations than supramicellar solutions. Furthermore, the salt additions to the submicellar SDBS solutions did not seem to help achieve these reductions. In fact, the systems with salt appeared to have less ability to achieve lower residual oil saturations than the submicellar solutions with no salt added. This could be due to media wettability changes or phase behavior between the LNAPL and the surfactant solutions. Phase behavior studies showed that the surfactant solutions containing no salt produced two clear separate phases, whereas the solutions containing salt produced a gel-like substance after mixing with the LNAPL.
The supramicellar SDBS solution of 1500 mg L-1 produced similar residual water saturation changes as the submicellar solutions with analogous surface tension values (Fig. 7 and 8). However, this solution did not reduce residual oil saturations in either soil. Given the reduction in water saturations of the supramicellar columns, one would expect a proportional decrease in residual oil saturations. However, the residual oil saturations are at or above the baseline value established by the DADI water. The higher residual oil saturations for the supramicellar solutions could be attributable to media wettability changes after surfactant application and drainage or increases in LNAPL sorption on the media. Further evaluation of these effects should be studied to determine the impacts of these changes. It should be noted that in the case of the reported LNAPL saturation after flushing an SDBS solution of 1500 mg L-1 through the CRA, the residual LNAPL saturation increased by approximately 0.30, which would correlate to a sorption interference of approximately 45 g of LNAPL, or 52.5 mg LNAPL sorbed g-1 CRA soil.
Figure 9 displays the SDBS impacts on residual water and oil saturations in both soils. Trend lines have been added to illustrate the impacts of increasing SDBS concentrations. The direction provided by the arrows illustrates increasing surfactant concentration and is intended for discussion purposes. It should be noted that changes in salt concentrations could alter this behavior. When applying submicellar SDBS solutions to the porous media, resultant residual water and oil saturations tend to decrease at a similar rate until reaching a minimum around the surfactant's CMC. This portion of the trend appears to be similar to the results of Blunt (1997). Residual saturation reductions of up to 50% can be achieved at the submicellar level, suggesting that submicellar surfactant solutions could be effective in substantially reducing oil saturations in the vadose zone. However, SDBS additions beyond the CMC do not appear to lower residual water saturations any further than what was achieved by the submicellar solutions. It is likely that this behavior is due to the inability to further reduce the surface tension of the solution beyond the value achieved at the CMC. Although this suggested behavior is based on one test in each soil and further evaluation of supramicellar surfactant solutions should be considered, it is believed that interferences were minimal and both tests adequately represent the effects of the supramicellar SDBS solutions. This trend also suggests that SDBS additions beyond the CMC can actually increase residual oil saturations. It is likely that at these concentrations, significant sorption of SDBS to the media impacts the media wettability, creating oil-wet conditions subject to high residual oil saturations. Groundwater table fluctuations during surfactant remediation could produce this undesirable effect, suggesting that supramicellar surfactant migration or application in the vadose zone could be detrimental to product recovery efforts.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 9. Trend of sodium dodecylbenzenesulfonate (SDBS) impacts on residual saturations in silica sand and Canadian river alluvium (CRA).
|
|
Application of Surfactant with Oil in Place
Table 1 shows the reductions in oil saturations for soil columns flushed with submicellar SDBS solutions. The data in Table 1 show that flushing with submicellar SDBS solutions followed by gravity drainage and applied air pressure drainage was able to reduce the amount of oil held in the silica sand down to oil saturations similar to those achieved when applying the surfactant before the oil was in place. Gravity drainage after surfactant application stimulated the release of both residual water and NAPL. Air pressure was applied to the column to achieve full drainage and collect previously trapped NAPL.
View this table:
[in this window]
[in a new window]
|
Table 1. Surfactant (sodium dodecylbenzenesulfonate [SDBS]) effects on oil saturations (surfactant applied after oil in place) in silica sand and Canadian river alluvium (CRA).
|
|
However, SDBS application was less successful in the CRA media. Gravity drainage did not release any NAPL, and applied pressures were unable to fully drain the CRA-filled columns. It is possible that this poor behavior was due to the presence of oil in combination with the high capillary pressures associated with the CRA media. The bottom portion of the column near the outlet always appeared to be above residual liquid (both oil and water) saturations, whereas the upper portion of the column experienced full drainage and appeared to contain less NAPL than initially held. Further research on this phenomenon (e.g., a sand-tank study that might create a vadose zone that is less restrictive than a column) should be considered before evaluating this effect in fine-grained soils.
|
CONCLUSIONS
|
|---|
The results presented here show that low-level surfactant solutions can lower both residual water and oil saturations, and these effects could occur in the vadose zone at existing surfactant flushing sites. In addition, low-level surfactant solution applications to a contaminated vadose zone could be an effective strategy for promoting the release of previously trapped water and LNAPL, which could increase free-product recovery rates while only introducing small amounts of surfactants into the subsurface. Residual water and oil saturations were lowered by up to 50% from their baseline values in the porous media chosen for these studies. The surfactant-induced saturation reductions became apparent using surfactant solutions as low as 100 mg L-1 SDBS (two to four orders of magnitude lower than concentrations used in surfactant remediation field demonstrations). It is suspected that this effect is due to the lowering of the surface tension of the residual water in the pore space. As discussed in Blunt (1997), more than two-thirds of residual water retained in the pore space can be trapped in the pore throats, whereas less than one-third is positioned in the corners of the pore space. It is therefore anticipated that much of the residual water trapped in the pore throats is being drained when the surface tension of the fluid is no longer sufficient to hold water in that particular pore space. This pore throat drainage can dramatically decrease water saturations, as well as increase permeability and lower NAPL saturations when maintaining water-wet conditions. Sodium chloride additions up to 0.2% appeared to have little effect on residual saturations.
Application of surfactant solutions after the oil was in place provided mixed results. These attempts at replicating field conditions in the columns were somewhat hindered due to capillary effects that made it difficult to fully drain the column. The CRA tests were not successful in lowering oil saturations using this technique. It is believed that this tight soil produced very high capillary forces that could not be overcome in the column to stimulate full drainage. However, this might not be the case in an in situ pore space network that isn't subject to the restrictions of the column (e.g., the small diameter of the column could create wall restrictions).
In contrast, this surfactant procedure appeared to be effective in releasing trapped residual oil in the silica sand. Both tests with this media released significant amounts of previously trapped residual oil, lowering oil saturations in the column by approximately 50%. This is a significant reduction in residual values and suggests further study is needed to observe what this effect could have in pore spaces that contain water and oil well above residual values. These areas of contamination would be targeted because the surfactant-enhanced drainage of this area could enhance free-product recovery and dramatically reduce oil saturations in the vadose zone. These effects could provide an environment more favorable for bioremediation (i.e., lower residual oil saturations and higher interfacial area of the remaining oil) or other vadose zone remediation technologies, such as bioventing or soil vapor extraction (e.g., lower residual fluids could increase air permeabilities in the vadose zone). However, as with any free-product recovery system, media wettability is an important parameter in achieving lower oil saturations in the subsurface and must also be evaluated in further studies for this application.
|
REFERENCES
|
|---|
- Adamson, A.W. 1990. Physical chemistry of surfaces. 4th ed. John Wiley & Sons, New York.
- Blunt, M.J. 1997. Pore level modeling of the effects of wettability. SPE J. 2:494–510.
- Chevalier, L.A., R.B. Wallace, D.C. Wiggert, and M.D. Shabana. 1996. 2-D experimental investigation of surfactant mobilization of light nonaqueous phase liquid. In Non-Aqueous Phase Liquids (NAPLS) in Subsurface Environment: Assessment and Remediation, Conference Proceedings. ASCE National Convention. Washington, DC. 12–14 Nov. 1996. ASCE, Reston, VA.
- Corey, A.T. 1994. Mechanics of immiscible fluids in porous media. Water Resource Publications, Highland Ranch, CO.
- Desai, F.N., A.H. Demond, and K.F. Hayes. 1992. Influence of surfactant sorption on capillary pressure–saturation relationships. p.133–148. In D.A. Sabatini and R.C. Knox (ed.) Transport and remediation of subsurface contaminants: Colloidal, interfacial, and surfactant phenomena. American Chemical Society, Washington, DC.
- Henry, E.J., J.E. Smith, and A.W. Warrick. 1999. Solubility effects on surfactant induced unsaturated flow through porous media. J. Hydrol. (Amsterdam) 223:164–174.
- Karkare, M.V., and T. Fort. 1994. Flow in ‘unsaturated’ porous media due to water-insoluble surfactants: Role of momentum transfer from a spreading monolayer. Langmuir 10:3701–3704.
- Knox, R.C., D.A. Sabatini, J.H. Harwell, R.E. Brown, C.C. West, F. Blaha, and C. Griffin. 1997. Surfactant remediation field demonstration using a vertical circulation well. Ground Water 35:948–953.
- Knox, R.C., B.J. Shau, D.A. Sabatini, and J.H. Harwell. 1999. Field Demonstration Studies of Surfactant-Enhanced Solubilization and Mobilization at Hill Air Force Base, Utah. In M.L. Brusseau et al. (ed.) Innovative subsurface remediation: Field testing of physical, chemical, and characterization technologies. ACS Symposium Series 725. American Chemical Society, Washington, DC.
- Rouse, J.D., D.A. Sabatini, and J.H. Harwell. 1993. Minimizing surfactant losses using twin-head anionic surfactants in subsurface remediation. Environ. Sci. Technol. 27:2072–2078.
- Sabatini, D.A., R.C. Knox, and J.H. Harwell (ed.) 1995. Surfactant enhanced subsurface remediation: Emerging technologies. ACS Symposium Series 594. American Chemical Society, Washington, DC.
- Sabatini, D.A., R.C. Knox, and J.H. Harwell. 1996. Surfactant-enhanced DNAPL remediation: Surfactant selection, hydraulic efficiency, and economic factors. USEPA, EPA/600/S-96/002.
TOP
ABSTRACT
INTRODUCTION
