Environ. Sci. Technol. 2009, 43, 5640–5646
Migration of Colloids through Nonfractured Clay-Rich Aquitards THORSTEN N. RESZAT† AND M. JIM HENDRY* Department of Geological Sciences, University of Saskatchewan, Saskatoon, SK, Canada, S7N 5E2
Received January 27, 2009. Revised manuscript received May 6, 2009. Accepted June 3, 2009.
This study examined characteristics and controls on diffusive transport of colloids through nonfractured, clay-rich glacial till. The range in molecular weight (Mw) of the colloids tested (five polymers and three natural dissolved organic carbons) was 910-15 450 Da. Hydrodynamic diameters increased from 1.45 to 6.05 nm, and aqueous diffusion coefficients decreased from 2.6 × 10-10 to 6.3 × 10-11 m2 s-1 with increasing Mw. All colloids were subjected to diffusion testing using undisturbed core samples placed in double reservoir diffusion cells. All colloids decreased in concentration with time in the spiked reservoirs. Concentrations in the receiving reservoirs increased for only the four smallest colloids. The lack of breakthroughs for larger colloids was attributed to straining. Transport modeling using data from colloids exhibiting breakthrough shows effective diffusion coefficients and tortuosity factors decrease from 1.5 × 10-10 to 6.5 × 10-11 m2 s-1 and from 0.6 to 0.3, respectively, with increasing Mw. The effective porosities are slightly less than total porosity (0.31). Our data suggest diffusive transport through clay-rich aquitards is limited to colloids with mean diameters 255 >540 >540 >540
2.6 2.2 2.2 2.1 1.8 1.4 1.3 1.0 6.3
-1
Dei
10-10 10-10 10-10 10-10 10-10 10-10 10-10 10-10 10-11
1.5 9.5 7.0 6.5
2
(m s ) × × × × × × × × ×
-1
τi
nei
10-10 10-11 10-11 10-11
0.6 0.4 0.3 0.3
0.30 0.28 0.27 0.26
2
(m s ) × × × ×
Kdi
-1
(mL g ) 0.1 0.4 0.2 2.9 2.9 3.4 0.9 1.0 1.0
modeled-diff. cells
Rdi
Kdi (mL g-1)
Rdi
2 4 2 24.3 24.3 28.3 8 12 13
0.1 0.3 0.2 0.8a
2 3 3 7.7a
a Calculated/observed from 1.80 nm fraction of SRFA. DH is the theoretically calculated hydrodynamic diameter. B.T. is breakthrough time of colloids into collection reservoir.
FIGURE 3. Results of batch sorption testing for all colloids examined. The symbols represent the measured data and the solid lines represent the best linear fit to the measured data. are consistent with Reszat and Hendry (34) who show an increase in aromaticity and reactivity of organic matter with increased DH. The Rdi values for the DOC are greater than the value of 1.0 determined by Hendry et al. (32), who used till and pore waters from the same research site. The differences may be attributed to differences in experimental designs. In Hendry et al. (32), sorption was estimated using a diffusive sorption experiment, wherein clay was allowed to equilibrate with the soil without agitating the particles as was done in the present batch testing. The greater sorption measured by batch testing than with diffusive-sorption testing may also be attributed to the increased availability of sorption sites in batch testing and/or the use of the more reactive DOC from shallower piezometers in the batch tests. In the latter, the DOC in the shallower piezometers is younger, more aromatic, and contains more functional groups than DOC found at 12 m depth (34) and thus could give rise to increased sorption. Sorption values are discussed further below. Diffusion Cell Testing. The experimental diffusion cell results are plotted as normalized (C/C0) concentrations for both reservoirs and all colloids in Figure 4. The upper set of data, with an initial high concentration that decreases with time, represents the source reservoir. The lower trace, with an initial concentration of zero, represents the collection reservoir. Mean and standard deviation of the DOC diffusion results from the three diffusion cells (Figure 4b) are consistent for all cells and times. This suggests the methods used are reproducible and therefore diffusion tests conducted on individual subcores can be directly compared. Further, the results are consistent with those from an earlier study (32). To ensure that diffusion was the dominant process in the cells, the transport of the naturally occurring conservative tracer, Cl, was monitored via its migration from Reservoir A to B in diffusion cells 1-3 (Table 1 and SI). In these tests, the
Cl tracer concentrations in both reservoirs yielded consistent results with the concentrations decreasing in reservoir A and increasing in reservoir B (data not presented). The Cl reached steady state conditions by day 350, earlier than the DOC. This showed that the Cl is conservative with a greater De than the DOC. Calculated De values for Cl in each cell were the same in keeping with ref 32. The diffusion traces for the three smallest colloids (1.45 and 1.75 nm polymers, the 1.70 nm DOC clearly show they diffuse through the till; the concentrations in the source reservoirs decrease while, after a time lag, the concentrations in the collection reservoirs increase. The breakthrough time, defined as the time to attain a measurable C/Co of greater than zero in reservoir B, is the transport time of the colloid through the clay core (which had an initial colloid concentration of zero). The breakthrough times for these colloids increase with increasing colloid diameter, from 16 days for the 1.45 nm colloid to 34 days for the 1.75 nm colloid. The inset graphs in Figure 4 present the relative differences in the mean diameter of these colloids before and after the experiments. These data show that there is no appreciable difference in the colloid diameters in the reservoirs with time, indicating no preferential diffusion or sorption of colloid sizes to the matrix media. In contrast, the three larger colloids (3.05, 3.80, and 6.05 nm) exhibited consistent decreases in concentration in reservoir A (50-60% of total over the test period), but no measurable increase in concentrations in reservoir B during the experiment. The inset graphs in Figure 4 for the larger colloids show an increase in the mean colloid diameter remaining reservoir A with time, suggesting a greater percentage of the smaller diameter colloid fractions were removed from the reservoir during the experiment. Although not evident in Figure 4d, the small size fraction (1.80 nm) of the SRFA diffuses through the till while the larger size fraction does not. Figure 5, which is similar to the inset graph in VOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5643
FIGURE 4. Measured (open diamonds) and modeled (solid lines) diffusion traces for individual colloids. The inset graphs provide a relative comparison of the initial size distributions of the colloids in the source reservoir (t ) 0 days; upper trace), the final size distributions in the source reservoir (t ) 255 or 540 days; middle trace), and where data exists, the size distribution in the collection reservoir (t ) 255 or 540 days; lower trace). In the groundwater DOC data plots (b), the experimental data (open diamonds) are the average of three sets of diffusion cell data the associated vertical bars are the standard deviations of these measurements. Figure 4d except the colloid concentrations are normalized to the void peaks, shows the size distributions of SRFA in the two reservoirs of the diffusion cells more clearly. The mean diameter of the SRFA was 2.15 nm in reservoir A at the start of the experiment but only 1.80 nm in reservoir B at the end of the experiment. No increase in concentration is observed in reservoir B for SRHA (2.70 nm). The insert graphs in Figure 4e for SRHA show a marked decrease in the mean diameter of the colloids remaining in reservoir A with time. This is attributed to an increased amount of sorption of the larger size colloids to the till. Using the time-dependent concentration data from both reservoirs, transport modeling was conducted for the three smallest colloids. Statistical best-fit modeling results (Figure 4) yield r2 values of 0.995, 0.996, and 0.998 for the 1.45 nm polymer, the DOC, and the 1.75 nm polymer colloids, respectively. For these data sets, the nei values ranged from 5644
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 15, 2009
FIGURE 5. Distribution of size fractions of SRFA initially in the source reservoir (fraction A) and at the end of the experiment in the collection reservoir (fraction B).
0.30-0.27 (Table 2) and exhibit no apparent relationship to Mw. These values are consistent with the measured total porosity (nt) for the till of 0.31 ( 0.01 (n ) 10) (30) and with a total connected porosity of 0.30 determined using mercury intrusion porosimetry on a core sample collected 12 m BG at the site (32). The calculated Rdi values ranged from 2 to 3 (Table 2) and were similar to the batch test results of 2-4. The agreement between methods suggests batch testing may be a viable method to determine Rdi values for these colloids. The modeled Dei values decreased from 1.5 × 10-10 m2 s-1 to 7.0 × 10-11 m2 s-1 with increased Mw. Further, the Dei of the DOC (9.5 × 10-11 m2 s-1) compared well with the Dei obtained for the DOC from a previous study at this research site (9 × 10-11 m2 s-1) (32). Tortuosity values were calculated using equation [S2] in the SI section using the experimentally derived Doi and bestfit Die values. Tortuosity values ranged from 0.6-0.3 decreasing with increasing colloidal diameter (Table 2). Equation [S3] (SI) suggests the smaller colloids follow a more direct route though the matrix. Their le, or travel distance, is closer to the straight-line travel distance, l, than for the larger colloids. Tortuosity values are in the ranges suggested in previous studies. Bear (40) cites literature values of 0.8-0.4, whereas Helmke et al. (21) and Perkins and Johnston (20) report a range from 0.01 to 0.4 for nonreactive solutes in saturated porous media. Hypothetical transport modeling of the spiked reservoir data for the three larger colloids was conducted using measured parameters in the same manner as described above. These simulations suggest the length of the experiment (550 days) should have been sufficient for measurable breakthrough to occur in the receiving reservoir. Sorption of the larger colloids to the till matrix should have yielded a maximum of 20% depletion of mass in the spiked reservoir, not the 50-60% observed. Further, the mass of colloids in the source reservoir at the end of these experiments was depleted in lighter colloids with respect to the start of the experiment (Figure 4f-h). This suggests a mechanism other than diffusion and sorption influences the migration of the larger colloids possibly pore throat constrictions, resulting in straining of the larger colloids. Determining the impact of straining on the migration of the colloids would require, among other things, knowledge of the distribution of the colloids through the core sample at the end of the experiments. Several unsuccessful attempts (data not presented) were made to extract and analyze colloids from the cores after the experiments (i.e., squeezing, centrifuging, solid-water extracts) as well as visualize the colloids in core samples (SEM, TEM, and X-ray microtomography). Our experiments show that the 1.70 nm DOC (with a maximum diameter of about 2.0 nm), the 1.45 and 1.75 nm polymers, and the smaller (3.05 nm, the SRHA, and the larger size fraction of SRFA did not diffuse through the till. The lack of diffusion of these colloids was attributed to straining. Based on these data, the effective pore throat diameter for the clay-till studied was determined to be between 2.0 and 2.2 nm. These data and additional data presented by Reszat (43) suggest the diameter of DOC in clay-rich aquitard media may be of value in defining the size of colloids that can be transported in these aquitards. The effective pore throat diameter of the clay-rich till would effectively prevent the migration of most colloids including viruses, bacteria, protozoa, inorganic minerals, most DOC, and large macromolecules (see Introduction). In the case of bacteria, our data imply that the viable bacteria observed in these massive clay-rich aquitards (29) were emplaced at the time of deposition (tens of thousands of years ago) and did
not migrate to these sites postdeposition. The control(s) on colloidal migration through clays warrants additional study.
Acknowledgments This research was funded by research grants from the Natural Sciences and Engineering Council of Canada (NSERC) through the Industrial Research Chair Program and the Saskatchewan Potash Producers Association. The reviews of three anonymous reviewers was appreciated.
Supporting Information Available A description and equations for 1D diffusive transport of solutes and tortuosity, details of the batch sorption testing, calculations to determine the concentration, Mw, hydrodynamic diameter (DH) and aqueous diffusion coefficient (Doi) of the colloids from asymmetrical flow field-flow fractionation measurements, and dissolved ion concentrations in porewater used in diffusion experiments.This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Ranville, J. F.; Hendry, M. J.; Reszat, T. N.; Xie, Q.; Honeyman, B. D. Quantifying uranium complexation by groundwater dissolved organic carbon using asymmetrical flow field-flow fractionation. J. Contam. Hydrol. 2007, 91, 233–246. (2) Reszat, T. N.; Hendry, M. J. Complexation of Aqueous Elements by DOC in a Clay Aquitard. Ground Water 2007, 45 (5), 542–553. (3) Matthess, G. Peckdeger, A. Survival and transport of pathogenic bacteria and viruses in groundwater. In Ground Water Quality; McCarty, P. L.; Ed.; John Wiley: New York, 1985; pp 472-482. (4) Thurman, E. M., 1985. Organic Geochemistry of Natural Waters. Martinus Nijhoff/Dr W. Junk Publishers, Dordrecht, 497 pp. (5) Foppen, J. W. A.; Mporokoso, A.; Schijven, J. F. Determining straining of Escherichia coli from breakthrough curves. J. Contam. Hydrol. 2005, 76 (3-4), 191–210. (6) Harvey, R. W.; George, L. H.; Smith, R. L.; Leblanc, D. R. Transport of microspheres and indigenous bacteria through a sandy aquifer - results of natural-gradient and forced-gradient tracer experiments. Environ. Sci. Technol. 1989, 23 (1), 51–56. (7) Harvey, R. W.; Kinner, N. E.; Macdonald, D.; Metge, D. W.; Bunn, A. Role of physical heterogeneity in the interpretation of smallscale laboratory and field observations of bacteria, microbialsized microsphere, and bromide transport through aquifer sediments. Water Resour. Res. 1993, 29 (8), 2713–2721. (8) Ibaraki, M. Colloid-facilitated contaminant transport in discretelyfractured porous media. In Environmental Geology; University of Waterloo: Waterloo, 1994. (9) McKay, L. D.; Gillham, R. W.; Cherry, J. A. Field Experiments in a fractured clay till. 2. Solute and colloid transport. Water Resour. Res. 1993, 29 (12), 3879–3890. (10) Kretzschmar, R.; Robarge, W. P.; Amoozegar, A. Filter Efficiency of three sparolites for natural clay and iron oxide colloids. Environ. Sci. Technol. 1994, 28 (11), 1907–1915. (11) Kretzschmar, R.; Robarge, W. P.; Amoozegar, A. Influence of natural organic matter on colloid transport through saprolite. Water Resour. Res. 1995, 31 (3), 435–445. (12) Cumbie, D. H.; McKay, L. D. Influence of diameter on particle transport in a fractured shale saprolite. J. Contam. Hydrol. 1999, 37 (1-2), 139–157. (13) McCarthy, J. F.; McKay, L. D.; Brunner, D. D. Influence of ionic strength and cation charge on transport of colloidal particles in fractured shale saprolite. Environ. Sci. Technol. 2002, 36, 3735– 3743. (14) McKay, L. D.; Sanford, W. E.; Strong, J. M. Field-scale migration of colloidal tracers in a fractured shale saprolite. Ground Water 2000, 38 (1), 139–147. (15) Bradford, S. A.; Bettahar, M.; Simunek, J.; van Genuchten, M. T. Straining and attachment of colloids in physically heterogeneous porous media. Vadose Zone J. 2004, 3 (2), 384–394. (16) Bradford, S. A.; Simunek, J.; Bettahar, M.; Van Genuchten, M. T.; Yates, S. R. Modeling colloid attachment, straining, and exclusion in saturated porous media. Environ. Sci. Technol. 2003, 37 (10), 2242–2250. (17) Bradford, S. A.; Yates, S. R.; Bettahar, M.; Simunek, J. Physical factors affecting the transport and fate of colloids in saturated porous media. Water Resour. Res. 2002, 38 (12), 1327, DOI: 10.1029/2002WR001340. VOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5645
(18) Moridis, G. J.; Hu, Q.; Wu, Y. S.; Bodvarsson, G. S. Preliminary 3D site-scale studies of radioactive colloid transport in the unsaturated zone at Yucca Mountain, Nevada. J. Contam. Hydrol. 2003, 60 (3-4), 251–286. (19) Grolimund, D.; Elimelech, M.; Borkovec, M.; Barmettler, K.; Kretzschmar, R.; Sticher, H. Transport of in situ mobilized colloidal particles in packed soil columns. Environ. Sci. Technol. 1998, 32 (22), 3562–3569. (20) Perkins, T. K.; Johnston, O. C. A review of diffusion and dispersion in porous media. Trans. Soc. Pet. Eng. Aime 1963, 228 (1), 70– 84. (21) Helmke, M. F.; Simpkins, W. W.; Horton, R. Experimental determination of effective diffusion parameters in the matrix of fractured till. Vadose Zone Journal 2005, 3 (3), 1050–1056. (22) Hendry, M. J.; Wassenaar, L. I. Transport and geochemical controls on the distribution of solutes and stable isotopes in a thick clay-rich till aquitard, Canada. Isot. Environ. Health Stud. 2004, 41 (1), 3. (23) Shackelford, C. D. Laboratory diffusion testing for waste disposalsA review. J. Contam. Hydrol. 1991, 7, 177–217. (24) Fredrickson, J. K.; McKinley, J. P.; Bjornstad, B. N.; Long, P. E.; Ringelberg, D. B.; White, D. C.; Krumholz, L. R.; Suflita, J. M.; Colwell, F. S.; Lehman, R. M.; Phelps, T. J. Pore-size constraints on the activity and survival of subsurface bacteria in a late Cretaceous shale-sandstone sequence, northwestern New Mexico. Geomicrobiol. J. 1997, 14 (3), 183–202. (25) Quigley, R. M.; Fernandez, F.; Yanful, E.; Helgason, T.; Margaritis, A. Hydraulic conductivity of contaminated natural clay directly below a domestic landfill. Can. Geotech. J. 1987, 24 (3), 377–383. (26) BoivinJahns, V.; Ruimy, R.; Bianchi, A.; Daumas, S.; Christen, R. Bacterial diversity in a deep-subsurface clay environment. Appl. Environ. Microbiol. 1996, 62 (9), 3405–3412. (27) Hendry, M. J.; Wassenaar, L. I. Implications of the distribution of delta-D in pore waters for groundwater flow and the timing of geologic events in a thick aquitard system. Water Resour. Res. 1999, 35 (6), 1751–1760. (28) Hendry, M. J.; Wassenaar, L. I. Controls on the distribution of major ions in pore waters of a thick surficial aquitard. Water Resour. Res. 2000, 36 (2), 503–513. (29) Lawrence, J. R.; Hendry, M. J.; Wassenaar, L. I.; Germida, J. J.; Wolfaardt, G. M.; Fortin, N.; Greer, C. W. Distribution and biogeochemical importance of bacterial populations in a thick clay-rich aquitard system. Microbiol. Ecol. 2000, 40, 273–291. (30) Shaw, R. J.; Hendry, M. J. Hydrogeology of a thick clay till and Cretaceous clay sequence, Saskatchewan, Canada. Can. Geotech. J. 1998, 35 (6), 1041–1052.
5646
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 15, 2009
(31) Timms, W. A.; Hendry, M. J. Quantifying the impact of cation exchange on long-term solute transport in a clay-rich aquitard. J. Hydrol. 2007, 332, 110–122. (32) Hendry, M. J.; Ranville, J. F.; Boldt-Leppin, B. E. J.; Wassenaar, L. I. Geochemical and transport properties of dissolved organic carbon in a clay-rich aquitard. Water Resour. Res. 2003, 39 (7), 1194–1203. (33) Hendry, M. J.; Wassenaar, L. I. Origin and migration of dissolved organic carbon fractions in a clay-rich aquitard: C- 14 and delta C-13 evidence. Water Resour. Res. 2005, 41, W0 2021, DOI: 10.1029/2004WR003157. (34) Reszat, T. N.; Hendry, M. J. Characterizing dissolved organic carbon using asymmetrical flow field-flow fractionation with on-line UV and DOC detection. Anal. Chem. 2005, 77, 4194– 4200. (35) Wassenaar, L. I.; Hendry, M. J. Improved piezometer construction and sampling techniques to determine pore water chemistry in aquitards. Ground Water 1999, 37 (4), 564–571. (36) Litzen, A.; Wahlund, K.-G. Effects of temperature, carrier composition and sample load in asymmetrical flow field-flow fractionation. J. Chromatogr., A 1991, 548, 393–406. (37) Wahlund, K.-G. Asymmetrical flow-field flow fractionation. In Field-Flow Fractionation Handbook.; Giddings, C. J., Ed.; John Wiley & Sons: New York, 2000; pp 279-294. (38) Wahlund, K.-G.; Giddings, C. J. Properties of an asymmetrical flow field-flow fractionation channel having one permeable wall. Anal. Chem. 1987, 59, 1332–1339. (39) Rowe, R. K., Booker, J. R. Fraser, M. J. POLLUTEv6 User’S Guide; GAEA Environmental Engineering: Windsor, Ontario, Canada, 1997. (40) Bear, J. The Dynamics of Fluids in Porous Media; Dover Publications, Inc.: New York, 1972. (41) Beckett, R.; Jue, Z.; Giddings, J. C. Determination of molecularweight distributions of fulvic and humic acids using flow fieldflow fractionation. Environ. Sci. Technol. 1987, 21 (3), 289–295. (42) Dycus, P. J. M.; Healy, K. D.; Stearman, G. K.; Wells, M. J. M. Diffusion-Coefficients and molecular-weight distributions of humic and fulvic-acids determined by flow field-flow fractionation. Sep. Sci. Technol. 1995, 30 (7-9), 1435–1453. (43) Reszat, T. N. Dissolved Organic Carbon in Aquitard Environments: Properties, Complexation and Transport.Unpublished Ph.D. thesis, Department of Geological Sciences, University of Saskatchewan, Canada, 2007.
ES803580X
