Article pubs.acs.org/est
Colloid Mobilization and Transport during Capillary Fringe Fluctuations Surachet Aramrak,†,‡,∥ Markus Flury,*,‡ James B. Harsh,† and Richard L. Zollars§ †
Department of Crop and Soil Sciences, §The Gene and Linda Voiland School of Chemical Engineering and Bioengineering, Washington State University, Pullman, Washington 99164, United States ‡ Department of Crop and Soil Sciences, Washington State University, Puyallup, Washington 98371, United States S Supporting Information *
ABSTRACT: Capillary fringe fluctuations due to changing water tables lead to displacement of air−water interfaces in soils and sediments. These moving air−water interfaces can mobilize colloids. We visualized colloids interacting with moving air−water interfaces during capillary fringe fluctuations by confocal microscopy. We simulated capillary fringe fluctuations in a glassbead-filled column. We studied four specific conditions: (1) colloids suspended in the aqueous phase, (2) colloids attached to the glass beads in an initially wet porous medium, (3) colloids attached to the glass beads in an initially dry porous medium, and (4) colloids suspended in the aqueous phase with the presence of a static air bubble. Confocal images confirmed that the capillary fringe fluctuations affect colloid transport behavior. Hydrophilic negatively charged colloids initially suspended in the aqueous phase were deposited at the solid−water interface after a drainage passage, but then were removed by subsequent capillary fringe fluctuations. The colloids that were initially attached to the wet or dry glass bead surface were detached by moving air−water interfaces in the capillary fringe. Hydrophilic negatively charged colloids did not attach to static air-bubbles, but hydrophobic negatively charged and hydrophilic positively charged colloids did. Our results demonstrate that capillary fringe fluctuations are an effective means for colloid mobilization.
■
forces exerted on the particles,27,28,35−39 velocities and numbers of air−water interfaces,29,33 surface properties of particles (i.e., charge and wettability),30,34 particle size,31 advancing and receding air− water interfaces,32,33 and particle shape40 have all been found to be relevant for particle detachment by moving air−water interfaces. In previous experiments, we investigated colloid detachment by moving air−water interfaces in a single channel,33,40 and here, we expand on colloid detachment in a porous medium. Our objective was to determine the effect of capillary fringe fluctuations on the behavior of colloids. We hypothesized that a moving air−water interface due to a fluctuating capillary fringe can scour the colloids from the medium surface and carry them along, but only if a colloid−air−water contact line is formed. We further hypothesized that trapped air bubbles can capture and immobilize colloids, but the colloids are being released when the bubbles dissolve or flush out. We tested these hypotheses experimentally by using a glass bead-filled column and fluorescent colloids in combination with confocal microscopy.
INTRODUCTION Colloids can promote the transport of radionuclides,1−4 heavy metals,5 pesticides,6,7 phosphorus,8−10 and animal hormones and veterinary antibiotics.11,12 In addition, viruses, bacteria, protozoa, and spores are all colloids, and their transport in subsurface media is controlled by colloidal mechanisms.13 Many studies of colloid and colloid-facilitated contaminant transport have been conducted under both saturated conditions14 and unsaturated conditions.15,16 The air−water interface and the air−water−solid interface line, that is, the line where the air−water interface intersects the solid phase, have been reported to be retention sites for colloids.17−19 However, only a few authors have focused on the capillary fringe.20−23 The capillary fringe is the zone just above the water table, which is still water-saturated, but has negative capillary pressure.24 The capillary fringe acts as a transition region between vertical unsaturated flow in the vadose zone and horizontal saturated flow in groundwater.25 As a mixing zone, the capillary fringe is expected to affect the transport of colloids that move from the subsurface to groundwater. Knowing the mechanisms of colloid behavior in the capillary fringe increases our fundamental understanding of the capillary fringe system. Fluctuations of the groundwater table lead to moving air− water interfaces as well as to entrapment of air bubbles in the capillary fringe.25,26 Moving air−water interfaces play a major role in colloid mobilization and transport. Colloids deposited on both initially wet solid surfaces27−33 and air-dried surfaces34 can be removed by moving air−water interfaces. Surface tension © 2014 American Chemical Society
■
MATERIALS AND METHODS Experimental Approach. We simulated capillary fringe fluctuations in a porous medium made of a glass column filled with glass beads. The behavior of colloids during the capillary Received: Revised: Accepted: Published: 7272
April 11, 2014 June 3, 2014 June 4, 2014 June 4, 2014 dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
Environmental Science & Technology
Article
fringe fluctuations were visualized by confocal microscopy. To visualize the different phases, we used fluorescent colloids and a fluorescent aqueous solution. This allowed us to distinguish the colloids; the water and air phases, including the air−water interfaces; and the glass beads in real time. We selected four scenarios for the capillary fringe experiments (Figure S2, Supporting Information). The first three scenarios were situations of colloids interacting with a moving air−water interface. These are (a) colloids suspended in the aqueous phase while the aqueous phase imbibes and drains the porous medium to simulate capillary fringe fluctuations; (b) colloids initially attached to the glass beads in a wet porous medium; and (c) colloids initially attached to the glass beads in a dry porous medium. The last scenario (d) mimics a situation when moving colloids interact with a static air−water interface (i.e., a trapped air bubble). Capillary Fringe System. We used a small-diameter (i.d. 1.5 mm) glass column of 7.5 cm length (see Supporting Information and Figure S1 for more details). We prepared and cleaned the column as described previously,33,40 and filled the column with glass beads. The glass beads had a diameter of 400−600 μm (EW-36270-51, Cole-Parmer Instrument Co., IL). These glass beads were soaked overnight in 10% HCl, rinsed with DI water, and air-dried at room temperature before being filled into the channel. We verified the size uniformity of the glass beads by scanning electron microscopy (FEI Quanta 200F, FEI Co., Hillsboro, OR). The porosity of the packed column was determined by measuring the weight difference between a dry column and a water-saturated column. We then used this weight difference to calculate the saturated volumetric water content,
which is equivalent to the porosity of the packed column. The porosity of our system was 0.55 cm3/cm3. This high porosity was due to boundary effects of the glass channel, preventing a close packing along the channel walls. We focused our confocal view during our experiments into the interior of the channel, and therefore, this boundary effect is not expected to be significant. Both ends of the column were connected to Tygon tubes, and one of the tubes was connected to a withdrawing/infusing syringe pump (KDS 210, KD Scientific, Holliston, MA) so that we could introduce and control capillary fringe fluctuations. The other end of the column was left open to an outflow container to allow free passage of air in and out of the channel. The column was then placed horizontally on the platform of a laser scanning confocal microscope (Axiovert 200 M equipped with LSM 510 META, Carl Zeiss Jena GmbH, Germany). See the Supporting Information for more details on the microscopy. Colloids and Liquids. We used hydrophilic carboxylatemodified polystyrene colloids (FluoSpheres, Lot No. 28120W, Molecular Probes Inc., Eugene, OR) with a diameter of 1 μm. The colloids were spherical, fluorescent with an excitation/ emission wavelength of 505/515 (yellow-green), and negatively charged, coming from the same batch used by Aramrak et al.40 For one selected experiment (scenario 4 below), we also used hydrophobic sulfate-modified (FluoSpheres, Lot No. 556340, Molecular Probes Inc., Eugene, OR), and hydrophilic amine-modified colloids (FluoSpheres, Lot No. 1306543, Molecular Probes Inc., Eugene, OR). These additional two colloids were also yellow-green fluorescent as the carboxylatemodified colloids. The properties of the colloids are listed in Table S1 (Supporting Information).
Figure 1. Locations and trajectory of carboxylate-modified colloids from scenario 1: (a) colloids in aqueous phase only (initial condition), (b) colloids deposited at the solid−water interface (SWI), (c) colloids attached to a mobile air−water interface (AWI), (d, e) colloids moving with AWI, and (f) transport back to the bulk liquid phase. Yellow and blue arrows indicate the direction of imbibition and drainage fronts, respectively. A white curve arrow indicates trajectories of the colloid of interest. Figures show subsequent snapshots during imbibition and drainage. 7273
dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
Environmental Science & Technology
Article
The liquid phase consisted of an aqueous solution of 1 mM CaCl2 and pH 5.5, that is, the same solution we had used previously,40 but here, we further added 0.09 mM (0.01% w/v) of sulforhodamine B dye (Acid Form, laser grade, Dye Content 95%, Lot No. 20223EAV, Sigma-Aldrich Co., MO). This dye is fluorescent with excitation/emission wavelengths of 565/585 nm (red). The CaCl2 and pH were selected to provide conditions of unfavorable attachment of the colloids to the glass bead surfaces, that is, secondary minimum attachment. Experimental Scenarios. Scenario 1: Suspended Colloids and Dynamic Air−Water Interface. In this scenario, we suspended the colloids (carboxylate-modified) in the aqueous phase to simulate a situation in which colloids are initially present in the water itself. We expected that the suspended colloids would attach to the glass beads under unfavorable attachment conditions, but then be detached again by the moving air−water interfaces in the capillary fringe. Colloids were suspended at a concentration of 3.6 × 1011 particles/L in the aqueous solution (1 mM CaCl2, pH 5.5, 0.09 mM sulforhodamine B). This suspension was then infused into the initially dry column at a flow rate of 0.012 mL/min (mean pore water velocity of 74 cm/h). This flow rate is relatively high for the change of the water table level; however, it is less than velocities occurring during Haines jumps. Haines jumps are the sudden and rapid movements of the infiltration or drainage fronts caused by the irregular pore geometry in porous media. When the imbibition front reached the middle of the microscopic viewing area, we stopped the flow, and upon complete flow cessation, we took an image, which represents the initial condition for this scenario. We could simultaneously observe the fluorescent-green colloids, the fluorescent-red aqueous solution, the dark-gray background of the air-phase, and the light-gray background of the glass beads. The flow was then reversed, with the same rate as the previous infusing flow. During the reversal of the flow, we took images as time series to document the continuous events occurring as the colloids were interacting with the air−water and solid−water interfaces. We made three subsequent cycles of this imbibition/drainage sequence. Scenario 2: Wet-Deposited Colloids and Dynamic Air− Water Interface. In this scenario, we deposited the colloids (carboxylate-modified) under unfavorable attachment conditions onto the glass beads in the porous medium prior to the capillary fringe fluctuation experiments. This represents a situation in which colloids are initially deposited in the capillary fringe and are then exposed to moving air−water interfaces as the capillary fringe fluctuates. To deposit the colloids, the glass bead-filled column was connected to Tygon tubing and a peristaltic pump (Ismatec IP4, Glattburg, Switzerland), and the column was preconditioned with a colloid-free solution (1 mM CaCl2, pH 5.5, and 0.09 mM sulforhodamine B). After discarding the outflow of the preconditioning solution, we introduced the colloidal suspension (1 mM CaCl2, pH 5.5, 0.09 mM sulforhodamine B, and colloid concentration of 3.6 × 1011 particles/L) at a flow rate of 0.33 mL/min (mean pore water velocity of 2,036 cm/h) in a circulating loop. The suspension was circulated for 1 h to deposit colloids onto the glass beads, after which the flow was switched to a colloid-free solution, and the column was flushed for 2 h to remove any nonattached colloids.33 The column and the connecting Tygon tubing were kept water-saturated prior to the capillary fringe fluctuation experiments. The water-saturated column was then placed under the confocal microscope, and capillary fringe fluctuations were simulated as described in scenario 1, except that we used a
Figure 2. Detachment of deposited carboxylate-modified colloids from a glass bead surface by air bubbles in scenario 2: (a) initial wet deposition, (b) air bubble formation after a drainage front passage (air bubbles are circled), (c, d) colloids captured and removed by an air bubble at location iii, (e, f) colloids captured and removed by an air bubble at location ii, and (g, h) colloids captured and removed by an air bubble at location i. Blue and yellow arrows indicate the direction of drainage and imbibition fronts, respectively. These figures are in chronological sequence.
colloid-free solution as the aqueous phase. This allowed us to see how the deposited colloids interacted with the moving air−water interfaces. We first drained the column by withdrawing the solution, followed by an infusion, and we completed three cycles. Scenario 3: Dry-Deposited Colloids and Dynamic Air− Water Interface. In this scenario, we deposited colloids (carboxylate-modified) onto the glass beads as in scenario 2; but after deposition, we flushed out the red color of sulforhodamine B with deionized water and dried out the column under a vacuum for 72 h at room temperature (20−22oC). This scenario represents a situation in which groundwater levels increase and the capillary fringe penetrates an initially dry soil or sediment layer. We followed the same procedure of capillary fringe fluctuations as that 7274
dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
Environmental Science & Technology
Article
Figure 3. Interaction between an air bubble and adjacent carboxylate-modified colloids in scenario 2: (a) colloids deposited at the solid−water interface (SWI), (b) colloids attached to the air−water interface (AWI) during air-bubble formation induced by a drainage front, (c, d) colloids collected during a decrease in air bubble size, (e) colloid removal after withdrawal of the air bubble, and (f) a schematic concept of the removal process. The sequence of parts a−e is chronological.
of scenario 1 by infusing and withdrawing the aqueous solution, except in this scenario, we used a colloid-free solution instead of a colloidal suspension. Scenario 4: Suspended Colloids and Static Air−Water Interface. We prepared a colloid-free glass-bead channel and infused and withdrew a colloid-free solution into the channel to generate trapped air bubbles. We then applied a steady-state flow in one direction and selected a stationary air bubble to represent a static air−water interface within the field of view in the microscope. Colloids were then introduced into the flow, and confocal images were taken sequentially to visualize the interactions of the colloids with the static air bubble. We selected the confocal image after 14 pore volumes (to allow for ample opportunity for colloids to interact with the air−water interface) of the suspension flow to demonstrate the results. For these experiments, we used three different colloids: (1) carboxylate-modified (hydrophilic surface and negative charge at pH 5.5), (2) sulfate-modified (hydrophobic surface and negative charge at pH 5.5), and (3) amine-modified (hydrophilic surface and positive charge at pH 3) colloids. No sulforhodamine B was used for the experiments of amine-modified colloids. A schematic of experimental scenarios and a confocal image of the initial condition for each scenario are shown in Figures S2 and S3 (Supporting Information). Data Analysis. Confocal images were analyzed to determine the movement and trajectories of the colloids as caused by flowing water and moving air−water interfaces. We were especially interested in determining (1) mobilization of
attached colloids by moving air−water interfaces and immobilization of suspended colloids in the capillary fringe, (2) the interaction between air bubbles and adjacent colloids, and (3) trajectories of colloid motion in the capillary fringe.
■
RESULTS AND DISCUSSION Suspended Colloids and Dynamic Air−Water Interface. Figure 1 shows the locations and trajectories of colloids during capillary fringe imbibition and drainage. The colloids were initially suspended in the aqueous phase (Figure 1a). After drainage, some colloids were deposited at the glass bead surface (Figure 1b). When the following imbibition front passed through, it detached and moved the deposited colloids (Figure 1c,d). Overall, the colloids that were deposited on the glass bead surfaces were transferred back to the liquid phase again (Figure 1f). The results of scenario 1 showed that most of the suspended colloids remained in the aqueous phase; however, we noticed that capillary fringe drainage could cause some suspended colloids to deposit at the glass bead surface and that these colloids could be detached again by the subsequent imbibition front. This is consistent with the reported enhanced colloid detachment from a surface during advancing as compared to receding air−water interfaces.33,40 When the water saturation decreases during the drainage of the capillary fringe, the liquid film thickness becomes thinner than the diameter of the colloids, thereby allowing the air−water interface to pin the colloids at the collector surface. However, during subsequent imbibition, the liquid film increases again and causes the 7275
dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
Environmental Science & Technology
Article
Figure 4. Detachment of initially dry deposited carboxylate-modified colloids after drainage and imbibition of the capillary fringe in scenario 3: (a) initial condition, (b) after an imbibition passage, and (c) after a drainage passage. The bottom row has been modified by highlighting the positions of the colloids with large green dots. Dashed curves indicate the position of the AWI. Yellow and blue arrows indicate the direction of imbibition and drainage fronts, respectively. Red and pink dots denote the colloids that have translocated to the new positions after respective imbibition and drainage events.
dissolution of air in the water phase and by increase in water pressure when the water table rises. Capillary fringe fluctuations are an important mechanism for colloid detachment from stationary solid surfaces and subsequent colloid mobilization. The trapped colloids at the air−water interface of the air bubble can be mobilized only if the air bubble is moving. Colloid removal by moving air−water interfaces has been demonstrated with flow channels and flat surfaces27−34,40 and is also relevant for porous media and capillary fringe fluctuations. We also observed colloids sliding along the surfaces of the glass beads (Figure S5, Supporting Information). These images show that colloids can transfer from one glass bead surface to another. This movement of colloids along the glass bead surface and their transfer from one glass bead to another is consistent with the “skimming” of colloids under unfavorable attachment conditions along collector surfaces.43 The downstream propagation of near-surface colloids can occur in flow-stagnation zones or grain-to-grain contacts aligned with stagnation zones,44 and our observations provide support for such a mechanism. The average velocity of the colloids sliding along the glass bead surfaces is 24 ± 24 μm/s, which is 9 times as slow as the mean pore water velocity (74 cm/h = 206 μm/s). The large variation of the colloid velocity indicates a nonuniform skimming of the colloids along the grain surfaces. Dry-Deposited Colloids and Dynamic Air−Water Interface. In scenario 3, colloids were most likely attached in the primary energy minimum, as we had flushed the columns after deposition with deionized water, which should have removed all colloids in the secondary minimum. Scenario 3
colloids to detach and transfer back to the bulk aqueous phase. This pinning in thin water films and subsequent detachment due to imbibition has been described by others.41,42 Wet-Deposited Colloids and Dynamic Air−Water Interface. In this scenario, we first deposited the carboxylatemodified colloids onto the glass beads under unfavorable attachment conditions. Figure 2 illustrates the overall detachment of wet-deposited colloids in the capillary fringe in a chronological sequence. The sequence of images shows the colloids (a) that were initially deposited; (b) that were attached to the air−water interface of air bubbles at locations i, ii, and iii during the drainage event; and (c−h) that were removed by air bubbles during the following imbibition event. These confocal images show that (1) air intruded during drainage, (2) the air was expelled again during the imbibition front, and (3) the deposited colloids were removed from the capillary fringe only if they appeared to be attached to the air−water interfaces. The detachment mechanisms of colloids by an air bubble are further illustrated in Figure 3. The colloids previously deposited at the solid−water interface (a) were readily detached from the soil-water interface and attached to a moving air−water interface during air bubble formation (b). These colloids were captured (c), moved along with the moving air−water interface as the air bubble size decreased (d), and were removed by the withdrawing air− water interface of the air bubble (e). The schematic concept (f) shows the overall colloid removal process. These results demonstrate the effect of the formation/deformation of airbubbles caused by the capillary fringe fluctuations. Reduction of air-bubble size during imbibition can happen because of 7276
dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
Environmental Science & Technology
Article
shows that colloids were readily mobilized by the moving air− water interface during imbibition and drainage (Figure 4). We observed (by visual counting) that one cycle of the capillary fringe fluctuation (imbibition and drainage) caused more than 50% of the deposited colloids to detach within the fluctuation region (Figure 4a vs c). The advancing (imbibition) interface caused the deposited colloids to mobilize (see red dots in Figure 4b). These translocated colloids were then removed from the capillary fringe by the subsequent receding (drainage) interface. Previous reports30,31,33,34,40 showed that a moving air−water interface is effective in colloid mobilization from flat surfaces, even when attached in the primary energy minimum,28,34 and our results demonstrate that this also applies to the capillary fringe. Although some colloids were mobilized by the air−water interface, other colloids did not move or detach (Figure 4d). We offer two explanations for this observation: (1) colloids attached under favorable conditions are less likely detached from a surface than are colloids attached under unfavorable conditions, so there is a better chance that the colloids remain attached,28,34 and (2) colloids are not detached if no air−water solid-interface line forms. It is likely that some colloids were located in regions, for example, local depressions, where the air−water interface could not form a contact line, and therefore, no capillary force would act on these colloids (Figure S6, Supporting Information). In addition, surface roughness can increase the strength of colloid attachment relative to a flat surface,45,46 thereby making them less likely to be removed by a moving air−water interface. Suspended Colloids and Static Air−Water Interface. In scenarios 1, 2, and 3, we showed colloid mobilization by a dynamic air−water interface, whereas in scenario 4, we report on suspended colloids interacting with a static air−water interface. Figure 5 shows three types of suspended colloids interacting with a static air bubble after 14 pore volumes of suspension flow. In the first case (Figure 5a), the suspended colloids did not attach to the static air bubble. Most of the colloids bypassed the static air−water interface when they approached the air bubble, and no colloid attachment to the airbubble was visible. We noticed the first colloid deposition on the glass beads after six pore volumes of flow. In the second case (Figure 5b), we also observed bypass occurring near the air bubble interface, but after 3 pore volumes of the suspension flow, colloids started to attach to the air bubble. The collection of colloids at the air bubble gradually increased until the flow was stopped at 14 pore volumes. The colloids from cases a and b have negative charges and are therefore electrostatically repelled from the negatively charged air-bubble.47 The sulfatemodified colloids, however, had less negative electrophoretic mobility than the carboxylate-modified colloids and will therefore experience less repulsive interactions and more attractive hydrophobic interaction with the air bubble. Our results agree with those of others48,49 who studied particle interactions with a static air bubble using atomic force microscopy. They reported on a long-range repulsive force between a hydrophilic particle and an air bubble, but no repulsion experienced by a hydrophobic particle. The hydrophobicity promotes the particle to penetrate the air bubble and form a three-phase contact line.50 In the third case (Figure 5c), to maintain the colloid’s positive charge, no sulforhodamine B was used. The results show (1) pronounced colloid deposition on the glass beads, (2) small amounts of colloid attachment to the static air bubble, and (3) bypass of colloids approaching the air bubble. Colloids were readily deposited onto the glass beads within the first pore
Figure 5. Colloid interactions with a static air bubble: (a) hydrophilic and negatively charged colloids, (b) hydrophobic and negatively charged colloids, and (c) hydrophilic and positively charged colloids. The left portion is a confocal view, and the right portion is a schematic view. The bright green dots with and without arrow indicate the mobile and immobile colloids, respectively. Curved blue arrows indicate the direction of colloids when bypass of air-bubble occurred.
volume of the flow. This is because of the electrostatic attraction between the positively charged colloids and the negatively charged glass beads. Nevertheless, we could see only a few examples of such electrostatic attraction between the colloids and the air bubble. Although the zeta potential of the amine-modified colloids was highly positive at 40.4 mV (Table S1, Supporting Information), the attraction between the colloids and the air bubble apparently was much less than we expected considering their opposite charges. We still observed repulsive interactions indicated by colloids bypassing the air bubble, similar as seen in the case of carboxylate-modified colloids. Our results suggest that the colloid surface wettability is more important than the electrostatic charges in controlling how the suspended colloids interact with the static air water interface in a porous medium.
■
IMPLICATIONS Our results demonstrate that the capillary fringe is an important zone for colloid mobilization and transport. Colloid mobilization occurs predominantly during imbibition, when advancing air− water interfaces move over soil, sediment, or rock surfaces. During drainage, it is less likely that colloids are being detached 7277
dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
Environmental Science & Technology
Article
(5) Gao, B.; Dong, Y.; Luo, Y.; Ma, L. Q. Colloid deposition and release in soils and their association with heavy metals. Crit. Rev. Environ. Sci. Technol. 2011, 41, 336−372. (6) Sprague, L. A.; Herman, J. S.; Hornberger, G. M.; Mills, A. L. Atrazine adsorption and colloid-facilitated transport through the unsaturated zone. J. Environ. Qual. 2000, 29, 1632−1641. (7) Gjettermann, B.; Petersen, C. T.; Koch, C. B.; Spliid, N. H.; Grøn, C.; Baun, D. L.; Styczen, M. Particle-facilitated pesticide leaching from differently structured soil monoliths. J. Environ. Qual. 2009, 38, 2382−2392. (8) Heckrath, G.; Brookes, P. C.; Poulton, P. R.; Goulding, K. W. T. Phosphorus leaching from soils containing different phosphorus concentrations in the Broadbalk experiment. J. Environ. Qual. 1995, 24, 904−910. (9) de Jonge, L. W.; Moldrup, P.; Rubek, G. H.; Schelde, K.; Djurhuus, J. Particle leaching and particle-facilitated transport of phosphorus at field scale. Vadose Zone J. 2004, 3, 462−470. (10) Vendelboe, A. L.; Moldrup, P.; Heckrath, G.; Jin, Y.; de Jonge, L. W. Colloid and phosphorus leaching from undisturbed soil cores samples along a natural clay gradient. Soil Sci. 2011, 176, 399−406. (11) Steiner, L. D.; Bidwell, V. J.; Di, H. J.; Cameron, K. C.; Northcott, G. L. Transport and modeling of estrogenic hormones in a dairy farm effluent through undisturbed soil lysimeters. Environ. Sci. Technol. 2010, 44, 2341−2347. (12) Zou, Y.; Zheng, W. Modeling manure colloid-facilitated transport of the weakly hydrophobic antibiotic florfenicol in saturated soil columns. Environ. Sci. Technol. 2013, 47, 5185−5192. (13) Sen, T. K. Processes in pathogenic biocolloidal contaminants transport in saturated and unsaturated porous media: a review. Water Air Soil Pollut. 2011, 216, 239−256. (14) Ryan, J. N.; Elimelech, M. Colloid mobilization and transport in groundwater. Colloids Surf. Physicochem. Eng. Aspects 1996, 107, 1−56. (15) Bradford, S. A.; Torkzaban, S. Colloid transport and retention in unsaturated porous media: a review of interface-, collecter-, and porescale processes and models. Vadose Zone J. 2008, 7, 667−681. (16) Flury, M.; Qiu, H. Modeling colloid-facilitated contaminant transport in the vadose zone. Vadose Zone J. 2008, 7, 682−697. (17) Wan, J. M.; Wilson, J. L. Colloid transport in unsaturated porous media. Water Resour. Res. 1994, 30, 857−864. (18) Crist, J. T.; McCarthy, J. F.; Zevi, Y.; Baveye, P. C.; Troop, J. A.; Steenhuis, T. S. Pore-scale visualization of colloid transport and retention in partially saturated porous media. Vadose Zone J. 2004, 3, 444−450. (19) Zevi, Y.; Gao, B.; Zhang, W.; Morales, V. L.; Cakmak, M. E.; Medrano, E. A.; Sang, W. J.; Steenhuis, T. S. Colloid retention at the meniscus−wall contact line in an open microchannel. Water Res. 2012, 46, 295−306. (20) Weisbrod, N.; Niemet, M. R.; Selker, J. S. Light transmission technique for the evaluation of colloidal transport and dynamics in porous media. Environ. Sci. Technol. 2003, 37, 3694−3700. (21) Bridge, J. W.; Banwart, S. A.; Heathwaite, A. L. High-resolution measurement of pore saturation and colloid removal efficiency in quartz sand using fluorescence imaging. Environ. Sci. Technol. 2007, 41, 8288−8294. (22) Cheng, T.; Saiers, J. E. Mobilization and transport of in situ colloids during drainage and imbibition of partially saturated sediments. Water Resour. Res. 2009, 45, W08414 DOI: 10.1029/ 2008WR007494. (23) Jost, D.; Winter, J.; Gallert, C. Distribution of aerobic motile and non-motile bacteria within the capillary fringe of silica sand. Water Res. 2010, 44, 1279−1287. (24) SSSA Glossary of soil science terms; SSSA, Madison, WI, on-line at http://www.soils.org/sssagloss/; accessed in December, 2013, 2007. (25) Haberer, C. M.; Rolle, M.; Cripka, O. A.; Grathwohl, P. Oxygen transfer in a fluctuating capillary fringe. Vadose Zone J. 2012, 11, 10.2136/vzj2011.0056. (26) Zhang, M. H.; Geng, S.; Ustin, S. L. Quantifying the agricultural landscape and assessing spatio-temporal patterns of precipitation and groundwater use. Landscape Ecol. 1998, 13, 37−53.
from stationary surfaces, but colloids can skim along the surfaces of stationary particles, and previously mobilized and suspended colloids can be translocated deeper into the saturated zone. It is unlikely that stationary air−water interfaces (for example, trapped air bubbles) will capture suspended colloids if the colloids are hydrophilic and negatively charged. Only hydrophobic or positively charged colloids can attach to stationary airbubbles. Although our experimental system, consisting of glass beads and monodisperse spherical colloids, is a simple model of the capillary fringe, we expect that the observed mechanisms hold in a real subsurface system and are even more pronounced because irregularly and angular-shaped colloids will more readily interact with moving air−water interfaces. Capillary fringe fluctuations are particularly pronounced for perched water tables near the soil surface and close to rivers where periodic stage changes cause groundwater tables to oscillate. Colloidal and colloid-associated contaminants present in or moving into the capillary fringe will be affected by the unique mechanisms operating in this region.
■
ASSOCIATED CONTENT
S Supporting Information *
Details on experimental setup, colloid properties, confocal microscopy, figures on experimental setup and additional confocal images. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 1-253-445-4522. E-mail: fl
[email protected]. Present Address ∥
Presently at: Department of Soil Science, Faculty of Agriculture, Kasetsart University, Bangkok-10900, Thailand. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS S.A. was financially supported by the Ananda Mahidol Foundation, under the Royal Patronage of H.M. the King, Bhumibol Adulyadej, Thailand. This material is based upon work supported by the U.S. Department of Energy, Office of Science (BER), under Award No. DE-FG02-08ER64660. We thank the WSU Franceschi Microscopy Center for access to their facility and Chris Davitt for help with the use of the confocal microscope. Funding was further provided by the Washington State University Agricultural Research Center through Projects 0267 and 0152.
■
REFERENCES
(1) Kersting, A. B.; Efurd, D. W.; Finnegan, D. L.; Rokop, D. J.; Smith, D. K.; Thompson, J. L. Migration of plutonium in ground water at the Nevada Test Site. Nature 1999, 397, 56−59. (2) Novikov, A. P.; Kalmykow, S. N.; Utsunomyia, S.; Ewing, R. C.; Horreard, F.; Merkulov, A.; Clark, S. B.; Tkachev, V.; Myasoedov, B. F. Colloid transport of plutonium in the far-field of the Mayak production association, Russia. Science 2006, 314, 638−641. (3) Cheng, T.; Saiers, J. E. Colloid-facilitated transport of cesium in vadose-zone sediments: the importance of flow transients. Environ. Sci. Technol. 2010, 44, 7443−7449. (4) Liu, Z.; Flury, M.; Zhang, Z. F.; Harsh, J. B.; Gee, G. W.; Strickland, C. E.; Clayton, R. E. Transport of Europium colloids in vadose zone lysimeters at the semi-arid Hanford site. Environ. Sci. Technol. 2013, 47, 2153−2160. 7278
dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
Environmental Science & Technology
Article
colloids from rough collector surfaces. Colloids Surf. Physicochem. Eng. Aspects 2012, 410, 98−110. (47) Takahashi, M. ζ Potential of microbubbles in aqueous solutions: electrical properties of the gas−water interface. J. Phys. Chem. B 2005, 109, 21858−21864. (48) Preuss, M.; Butt, H. Direct measurement of particle−bubble interactions in aqueous electrolyte: dependence on surfactant. Langmuir 1998, 14, 3164−3174. (49) Ren, S.; Masliyah, J.; Xu, Z. Studying bitumen−bubble interactions using atomic force microscopy. Colloids Surf. Physicochem. Eng. Aspects 2014, 444, 165−172. (50) Fielden, M. L.; Hayes, R. A.; Ralston, J. Surface and capillary forces affecting air bubble−particle interactions in aqueous electrolyte. Langmuir 1996, 12, 3721−3727.
(27) Leenaars, A. F. M.; O’Brien, S. B. G. Particle removal from silicon substrates using surface tension forces. Philips J. Res. 1989, 44, 183−209. (28) Noordmans, J.; Wit, P. J.; van der Mei, H. C.; Busscher, H. J. Detachment of polystyrene particles from collector surfaces by surface tension forces induced by air-bubble passage through a parallel plate flow chamber. J. Adhes. Sci. Technol. 1997, 11, 957−969. (29) Gomez-Suarez, C.; Noordmans, J.; van der Mei, H. C.; Busscher, H. J. Removal of colloidal particles from quartz collector surfaces as simulated by the passage of liquid−air interfaces. Langmuir 1999, 15, 5123−5127. (30) Gomez-Suarez, C.; van der Mei, H. C.; Busscher, H. J. Air bubble-induced detachment of positively and negatively charged polystyrene particles from collector surfaces in a parallel-plate flow chamber. J. Adhes. Sci. Technol. 2000, 14, 1527−1537. (31) Gomez-Suarez, C.; Noordmans, J.; van der Mei, H. C.; Busscher, H. J. Air bubble-induced detachment of polystyrene particles with different sizes from collector surfaces in a parallel plate flow chamber. Colloids Surf. 2001, 186, 211−219. (32) Lazouskaya, V.; Wang, L.-P.; Gao, H.; Shi, X.; Crymmek, K.; Jin, Y. Pore-scale investigation of colloid retention and mobilization in the presence of a moving air−water interface. Vadose Zone J. 2011, 10, 1250−1260. (33) Aramrak, S.; Flury, M.; Harsh, J. B. Detachment of deposited colloids by advancing and receding air-water interfaces. Langmuir 2011, 27, 9985−9993. (34) Sharma, P.; Flury, M.; Zhou, J. Detachment of colloids from a solid surface by a moving air-water interface. J. Colloid Interface Sci. 2008, 326, 143−150. (35) van Nierop, E. A.; Stijnman, M. A.; Hilgenfeldt, S. Shapeinduced capillary interactions of colloidal particles. Europhys. Lett. 2005, 72, 671−677. (36) Lehle, H.; Noruzifar, E.; Oettel, M. Ellipsoidal particles at fluid interfaces. Eur. Phys. J. E: Soft Matter Biol. Phys. 2008, 26, 151−160. (37) Shang, J.; Flury, M.; Deng, Y. Force measurements between particles and the air−water interface: Implications for particle mobilization in unsaturated porous media. Water Resour. Res. 2009, 45, W06420 DOI: 10.1029/2008WR007384. (38) Danov, K. D.; Kralchevsky, P. A. Capillary forces between particles at a liquid interface: General theoretical approach and interactions between capillary multipoles. Adv. Colloid Interface Sci. 2010, 154, 91−103. (39) Chatterjee, N.; Lapin, S.; Flury, M. Capillary forces between sediment particles and an air-water interface. Environ. Sci. Technol. 2012, 46, 4411−4418. (40) Aramrak, S.; Flury, M.; Harsh, J. B.; Zollars, R. L.; Davis, H. P. Does colloid shape affect detachment of colloids by a moving air− water interface? Langmuir 2013, 29, 5770−5780. (41) Wan, J. M.; Tokunaga, T. K. Film straining of colloids in unsaturated porous media: conceptual model and experimental testing. Environ. Sci. Technol. 1997, 31, 2413−2420. (42) Shang, J.; Flury, M.; Chen, G.; Zhuang, J. Impact of flow rate, water content, and capillary forces on in situ colloid mobilization during infiltration in unsaturated sediments. Water Resour. Res. 2008, 44, W06411 DOI: 10.1029/2007WR006516. (43) Ma, H.; Pazmino, E. F.; Johnson, W. P. Surface heterogeneity on hemisphere-in-cell model yields all experimentally observed non straining colloid retention mechanisms in porous media in the presence of energy barriers. Langmuir 2011, 27, 14982−14994. (44) Johnson, W. P.; Hilpert, M. Upscaling colloid transport and retention under unfavorable conditions: Linking mass transfer to pore and grain topology. Water Resour. Res. 2013, 49, 10.1002/wrcr.20433. (45) Shen, C.; Wang, F.; Li, B.; Jin, Y.; Wang, L. P.; Huang, Y. Application of DLVO energy map to evaluate interactions between spherical colloids and rough surfaces. Langmuir 2012, 28, 14681− 14692. (46) Shen, C.; Lazouskaya, V.; Zhang, H.; Wang, F.; Li, B.; Jin, Y.; Huang, Y. Theoretical and experimental investigation of detachment of 7279
dx.doi.org/10.1021/es501797y | Environ. Sci. Technol. 2014, 48, 7272−7279
