Final Report: Microbial Pathogen Removal During Bank Filtration

EPA Grant Number: R829010
Title: Microbial Pathogen Removal During Bank Filtration
Investigators: Ryan, Joseph N. , Harvey, Ronald W. , Elimelech, Menachem
Institution: University of Colorado at Boulder , United States Geological Survey , Yale University
EPA Project Officer: Page, Angela
Project Period: September 1, 2001 through August 31, 2004 (Extended to August 30, 2005)
Project Amount: $506,006
RFA: Drinking Water (2000) RFA Text |  Recipients Lists
Research Category: Drinking Water , Water


Our incomplete understanding of processes and properties affecting pathogenic microbe transport during riverbank filtration is currently limiting our ability to predict the effectiveness of this water treatment option. We propose a series of fundamental experiments designed to better understand the effects of microbe size, physical and geochemical heterogeneity of the porous media, and high pumping rates on the transport of Cryptosporidium parvum oocysts in alluvial valley aquifers used for riverbank filtration. The hypotheses to be tested during this research address (1) the effect of microbe size on transport, (2) the effect of physical heterogeneity of the porous media on transport, (3) the effect of geochemical heterogeneity of the porous media on transport, (4) the effect of microbe release from unfavorable attachment sites, (5) the effect of high pumping rates on microbe release. Our major objective for this research is to develop a model of oocyst transport in porous media that can accommodate the physical and geochemical heterogeneity present in alluvial valley aquifer used for riverbank filtration.

Summary/Accomplishments (Outputs/Outcomes):

Overview. Clean drinking water is one of the most pressing global environmental and health problems of our time. As the world’s growing population puts greater demands on the available supply of high-quality drinking water, water utilities have developed new technologies for treating waters of degraded quality, such as membrane filtration, soil aquifer treatment, and advanced oxidation. But an old method called riverbank filtration is increasingly being used because it is a relatively inexpensive and sustainable means to improve the quality of surface waters (Tufenkji et al., 2002) . Bank filtration provides passive exposure to various processes and produces water that is relatively consistent in quality and easier to treat to higher levels of finished quality.

The complex setting of alluvial valley aquifers makes interpretation of the limited data on microbe transport during bank filtration a daunting task. Aside from quantifying the effectiveness of riverbank filtration in these diverse and heterogeneous hydrogeological settings, water utility engineers and scientists must learn how to predict the removal of pathogenic microbes and other key contaminants in the riverbank environment so that regulators can establish reasonable standards. In the U.S. Environmental Protection Agency’s proposed Long-Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR) , riverbank filtration has been recognized as one way to improve the removal of Cryptosporidium parvum oocysts from surface water. The EPA is proposing to grant additional treatment credit for removal of Cryptosporidium parvum oocysts to water utilities using bank filtration that meet a set of specified design criteria.

The design criteria specified by the EPA in LT2ESWTR are based on conservative estimates drawn from colloid filtration theory and an analysis of microbial monitoring data from existing bank filtration sites. The EPA proposes that horizontal and vertical wells drilled into unconsolidated, granular aquifers would be eligible for 0.5log (68%) removal credit or 1.0 log (90%) removal credit when located at least 25 or 50 ft (7.6 or 15.2 m) from the source, respectively. However, basing removal credit simply on transport distance does not account for the possible effects of various hydrological, geochemical, and biological factors, such as variations in pore water velocity, degree of groundwater dilution, solution chemistry, aquifer media surface characteristics, and inherent heterogeneities in the microbial population.

The objectives of this research were to examine the effects of some of the physical, geochemical, and biological heterogeneities on the transport of Cryptosporidium parvum oocysts. The research was carried out in Boulder, Colorado, as a collaboration between Joseph Ryan (University of Colorado at Boulder) and Ronald Harvey (U.S. Geological Survey) and in New Haven, Connecticut by Menachem Elimelech. The following sections describe the major results of their research.

Effect of Physical Heterogeneity on Oocyst Transport. The transport and filtration behavior of Cryptosporidium parvum oocysts in columns packed with quartz sand was systematically examined under repulsive electrostatic conditions by Tufenkji et al. (2004) . They observed that an increase in solution ionic strength resulted in greater oocyst deposition rates despite theoretical predictions of a significant electrostatic energy barrier to deposition. Relatively high deposition rates obtained with both oocysts and polystyrene latex particles of comparable size at low ionic strength (1 mM) suggest that straining may play a key role in oocyst removal. Supporting experiments conducted with latex particles of varying sizes, under very low ionic strength conditions where physicochemical filtration is negligible, clearly indicated that physical straining is an important capture mechanism. The results of this study indicate that irregularity of sand grain shape (verified by SEM imaging) contributes considerably to the straining potential of the porous medium. Hence, both straining and physicochemical filtration are expected to control the removal of oocysts in settings typical of riverbank filtration, soil infiltration, and slow sand filtration. Because classic colloid filtration theory does not account for removal by straining, these observations have important implications with respect to predictions of oocyst transport.

In work still being prepared for publication, Abudalo et al. (in preparation, a) examined the removal of oocysts in columns filled with pure, well-rounded quartz sand ranging in size from 0.10 to 2.18 mm to determine the threshold for straining. Initial column experiments conducted at pH 7 and ionic strength of 10-4 M, conditions which were expected to result in oocyst removal only by straining, revealed significant oocyst removal even in the grains of the largest size. To further reduce the likelihood of oocyst removal by mechanisms other than straining, the 10-4 sodium chloride solution was replaced with high-purity water (estimated ionic strength of 2.3´10-6M based on equilibrium with atmospheric carbon dioxide) . Oocyst removal decreased by about 6-8% for all grain sizes, but removal was still significant even in the largest grains. To reduce the rate of collisions leading to physicochemical filtration, the flow rate was increased from 2.0 to 14.6md-1. The increase in flow rate resulted in no significant oocyst removal in grains larger than 0.92 mm, but oocyst removal only be straining in grains of 0.92mm still seemed unlikely. They hypothesized that a significant fraction of oocyst removal was the result of secondary minimum deposition. Tests of the release of oocysts from the sand grains supported this hypothesis because nearly equal fractions of oocysts were released by successive suspension of the grains in the background solution. The application of mechanical energy in the form of vibrations was applied to the column during oocyst deposition and release phases. The vibrations were expected to prevent secondary minimum deposition. The vibrations resulted in further reductions in the deposition of oocysts, finally resulting in an estimate of the straining threshold at a ratio of oocyst diameter to grain diameter of 0.0080.

Effect of Geochemical Heterogeneity on Oocyst Transport. A growing body of experimental evidence suggests that the deposition behavior of microbial particles is inconsistent with the classical colloid filtration theory. Tufenkji and Elimelech (2004a) conducted a set of well-controlled laboratory-scale column deposition experiments with uniform model particles and collectors to obtain insight into the mechanisms that give rise to the diverging deposition behavior of microorganisms. Both the fluid-phase effluent particle concentration and the profile of retained particles were systematically measured over a broad range of physicochemical conditions. The results indicated that the concurrent existence of both favorable and unfavorable colloidal interactions causes significant deviation from the colloid filtration theory in the presence of repulsive Derjaguin-Landau-Verwey-Overbeek (DLVO) interactions. A dual deposition mode model was developed to account for the combined influence of “fast” and “slow” particle deposition. This model was shown to adequately describe both the spatial distribution of particles in the packed bed and the suspended particle concentration at the column effluent. The need for dual deposition modes was attributed to two possible phenomena, surface charge heterogeneity and secondary minimum deposition. Both of these phenomena could give rise to fast, or favorable, deposition.

Tufenkji and Elimelech (2005a) further investigated the mechanisms and causes of deviation from the classical colloid filtration theory (CFT) in the presence of repulsive Derjaguin-Landau-Verwey-Overbeek (DLVO) interactions. The deposition behavior of uniform polystyrene latex colloids in columns packed with spherical soda-lime glass beads was systematically examined over a broad range of physicochemical conditions, whereby both the fluid-phase effluent particle concentration and the profile of retained particles were measured. Experiments conducted with particles of three different sizes in a simple (1:1) electrolyte solution revealed the controlling influence of secondary minimum deposition on the deviation from CFT. In a second series of experiments, sodium dodecyl sulfate (SDS) was added to the background electrolyte solution with the intent of masking near-neutrally charged regions of particle and collector surfaces. These results indicated that the addition of a small amount of anionic surfactant is sufficient to reduce the influence of certain surface charge heterogeneities on the deviation from CFT. To verify the validity of CFT in the absence of surface charge heterogeneities, a third set of experiments was conducted using solutions of high pH to mask the influence of metal oxide impurities on glass bead surfaces. The results demonstrate that both secondary minimum deposition and surface charge heterogeneities contribute significantly to the deviation from CFT generally observed in colloid deposition studies. It is further shown that agreement with CFT is obtained even in the presence of an energy barrier (i.e., repulsive colloidal interactions) , suggesting that it is not the general existence of repulsive conditions which causes deviation but rather the combined occurrence of “fast” and “slow” particle deposition.

Tufenkji and Elimelech (2005b) examined the deposition of Cryptosporidium parvum oocysts under conditions expected to lead to dual modes of deposition. Spatial distributions of oocysts in columns packed with uniform glass-bead collectors were measured over a broad range of physicochemical conditions. Oocyst deposition behavior was shown to deviate from predictions based on classical CFT in the presence of repulsive (unfavorable) colloidal interactions. Specifically, CFT tended to predict greater removal of oocysts (less transport) than that observed in controlled laboratory experiments. Comparison of oocyst retention with results obtained using polystyrene latex particles of similar size suggests that mechanisms controlling particle deposition are the same in both systems. At a given ionic strength, the deposition of oocysts is generally greater than that of the microspheres; however, this discrepancy is partly attributable to large differences in oocyst and microsphere zeta potentials. A dual deposition mode model was applied which considers the combined influence of “fast” and “slow” oocyst deposition due to the concurrent existence of favorable and unfavorable oocyst-collector interactions. Model simulations of retained oocyst profiles and suspended oocyst concentration at the column effluent are consistent with experimental data. Because classic CFT does not account for the effect of dual mode deposition (i.e., simultaneous “fast” and “slow” oocyst deposition) , these observations have important implications for predictions of oocyst transport in subsurface environments, where repulsive electrostatic interactions predominate. Supporting elution experiments further suggest that specific surface interactions between oocyst wall macromolecules and the glass bead collectors could retard or even completely inhibit oocyst release upon perturbation in solution chemistry.

To test the effect of geochemical heterogeneity on microorganism transport in saturated porous media, Abudalo et al. (2005) measured the removal of two microorganisms, the bacteriophage PRD1 and oocysts of the protozoan parasite Cryptosporidium parvum, in flow-through columns of quartz sand coated by different amounts of ferric oxyhydroxide. The experiments were conducted over ranges of ferric oxyhydroxide coating fraction from 0 to 0.12 for PRD1 and from 0 to 0.32 for the oocysts at pH 5.6-5.8 and 10-4 M ionic strength. To determine the effect of pH on the transport of the oocysts, experiments were also conducted over a pH range of 5.7-10.0 at a coating fraction of 0.04. Collision (attachment) efficiencies increased from 0.0071 to 0.13 for PRD1 and from 0.059 to 0.75 for the oocysts as the fraction of ferric oxyhydroxide coated quartz sand increased. Increasing the pH from 5.7 to 10.0 resulted in a decrease in the oocyst collision efficiency as the pH exceeded the expected point of zero charge of the ferric oxyhydroxide coatings. The collision efficiencies correlated very well with the fraction of quartz sand coated by the ferric oxyhydroxide for PRD1 but not as well for the oocysts.

The transport of microbes in bank filtration environments is also affected by the presence of dissolved organic matter. Abudalo et al. (in preparation, b) examined the effect of dissolved organic matter on oocyst transport in a ferric oxyhydroxide-coated quartz sand. The columns, filled with 4%-coated sand, were pre-equilibrated with a well-characterized hydrophobic acid fraction of dissolved organic matter collected from the Florida Everglades. Oocyst removal in the column decreased as the concentration of the dissolved organic matter increased from 0 to 20mgL-1. Microelectrophoresis measurements showed that the dissolved organic matter did not significantly affect the zeta potential of the oocysts. Streaming potential measurements demonstrated that the organic matter affected oocyst transport by adsorbing to the ferric oxyhydroxide coating.

The combined effect of physical and geochemical heterogeneities on oocyst transport was explored by Abe et al. (in preparation) . They filled an intermediate-scale aquifer tank of about 4.3m length, 1.0m height, and 0.1m width with homogeneous cells of quartz sand fractions of different grain size and ferric oxyhydroxide coating fraction. Each cell was 20cm length and 5cm height. The background solution in the tank was pH 5.7 and 10-4M sodium chloride. The oocysts and fluorescent polystyrene microspheres of 1.0, 2.9, and 4.6mm diameter were injected as a pulse at the upstream end of the tank, and their transport was monitored by sampling at vertical arrays at six distances (10cm, 20cm, 40cm, 80cm, 1.6m, and 3.2m) and the tank downstream effluent. The breakthrough of the oocysts and microspheres was coincident with the breakthrough of a bromide tracer. The retention of oocysts in the column was about five times greater than the retention of the microspheres, which was attributed to the greater negative zeta potentials of the microspheres. The transport of the oocysts was simulated with an advection-dispersion transport model with first-order irreversible attachment of the oocysts.

Effect of Biological Heterogeneity on Oocyst Transport. The transport and fate of microbial particles in subsurface environments is controlled by their capture (filtration) by sediment grains. Typically, filtration models used to describe microbe removal in porous media predict exponential decrease in microbial particle concentration with travel distance. However, a growing body of laboratory-scale column experiments suggests that the retained microbial particle profiles decay non-exponentially. Tufenkji et al. (2003) attributed the observed behavior to the heterogeneity in the interactions between microbial particles and sediment grains, most likely due to the inherent variability in the microbial particles. They incorporated this factor into classical colloid filtration theory by inclusion of a distribution in the deposition rate coefficient. They showed that certain distributions of the deposition rate coefficient (i.e., log-normal, bimodal, and power-law distributions) give rise to non-exponential deposition patterns. Comparisons of model predictions to experimental data indicate that the observed non-exponential deposition behavior of bacteria and virus particles may be attributed to a broad range (i.e., a power-law distribution) of microbial deposition rates. Other mechanisms, such as particle release and blocking by previously deposited microbial particles, are also shown to be potential sources of deviation from the classical filtration theory. Our results further suggest that monitoring fluid-phase particle concentration is insufficient for accurate characterization of the deposition and transport behavior of microbial particles in saturated porous media. Rather, the shape of the microbial particle retention profile is shown to be a key indicator of the mechanisms controlling microbial deposition and transport.

Colloid Filtration Theory Improvement. Tufenkji and Elimelech (2004b) developed a new equation for predicting the single-collector contact efficiency (η) in physicochemical particle filtration in saturated porous media. The correlation equation was developed assuming that the overall single-collector efficiency can be calculated as the sum of the contributions of the individual transport mechanisms: Brownian diffusion, interception, and gravitational sedimentation. To obtain the correlation equation, they regressed the dimensionless parameters governing particle deposition against the theoretical value of the single-collector efficiency over a broad range of parameter values. Rigorous numerical solution of the convective-diffusion equation with hydrodynamic interactions and universal van der Waals attractive forces fully incorporated provided the theoretical single-collector efficiencies. The resulting equation overcomes the limitations of current approaches and shows remarkable agreement with exact theoretical predictions of the single-collector efficiency over a wide range of conditions commonly encountered in natural and engineered aquatic systems. Furthermore, experimental data are in much closer agreement with predictions based on the new correlation equation compared to other available expressions.

Journal Articles on this Report : 9 Displayed | Download in RIS Format

Other project views: All 31 publications 9 publications in selected types All 9 journal articles
Type Citation Project Document Sources
Journal Article Abudalo RA, Bogatsu YG, Ryan JN, Harvey RW, Metge DW, Elimelech M. Effect of ferric oxyhydroxide grain coatings on the transport of bacteriophage PRD1 and Cryptosporidium parvum oocysts in saturated porous media. Environmental Science and Technology 2005;39(17):6412-6419. R829010 (Final)
  • Abstract from PubMed
  • Full-text: Yale-PDF
  • Journal Article Abudalo RA, Ryan JN, Harvey RW, Metge DW, Landkamer L. Influence of organic matter on the transport of Cryptosporidium parvum oocysts in a ferric oxyhydroxide-coated quartz sand saturated porus medium. Water Research 2010;44(4):1104-1113. R829010 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct-Full Text HTML
  • Other: Science Direct-PDF
  • Journal Article Tufenkji N, Ryan JN, Elimelech M. The promise of bank filtration. Environmental Science and Technology 2002;36(21):422A-428A. R829010 (Final)
  • Abstract from PubMed
  • Journal Article Tufenkji N, Redman JA, Elimelech M. Interpreting deposition patterns of microbial particles in laboratory-scale column experiments. Environmental Science and Technology 2003;37(3):616-623. R829010 (Final)
  • Abstract from PubMed
  • Full-text: McGill University-PDF
  • Journal Article Tufenkji N, Elimelech M. Deviation from the classical colloid filtration theory in the presence of repulsive DLVO interactions. Langmuir 2004;20(25):10818-10828. R829010 (Final)
  • Abstract from PubMed
  • Full-text: Yale-PDF
  • Abstract: ACS Publications-Abstract
  • Journal Article Tufenkji N, Elimelech M. Correlation equation for predicting single-collector efficiency in physicochemical filtration in saturated porous media. Environmental Science and Technology 2004;38(2):529-536. R829010 (Final)
  • Abstract from PubMed
  • Journal Article Tufenkji N, Miller GF, Ryan JN, Harvey RW, Elimelech M. Transport of Cryptosporidium oocysts in porous media: role of straining and physicochemical filtration. Environmental Science and Technology 2004;38(22):5932-5938. R829010 (Final)
  • Abstract from PubMed
  • Journal Article Tufenkji N, Elimelech M. Breakdown of colloid filtration theory: role of the secondary energy minimum and surface charge heterogeneities. Langmuir 2005;21(3):841-852. R829010 (Final)
  • Abstract from PubMed
  • Abstract: ACS Publications-Abstract
  • Journal Article Tufenkji N, Elimelech M. Spatial distributions of Cryptosporidium oocysts in porous media: evidence for dual mode deposition. Environmental Science and Technology 2005;39(10):3620-3629. R829010 (Final)
  • Abstract from PubMed
  • Full-text: Yale-PDF
  • Supplemental Keywords:

    groundwater, pathogen, virus, protozoa, Cryptosporidium parvum, oocyst, transport, bank filtration,, RFA, Health, Scientific Discipline, Water, Environmental Chemistry, Health Risk Assessment, Risk Assessments, Environmental Microbiology, Environmental Monitoring, Drinking Water, ecological risk assessment, fate and transport, microbial contamination, monitoring, pathogens, bacteria, exposure and effects, exposure, viruses, other - risk assessment, modeling, human exposure, treatment, cryptosporidium , microbial risk management, groundwater contamination, drinking water contaminants, Giardia, drinking water treatment, exposure assessment, groundwater, human health risk

    Relevant Websites:

    Joseph Ryan: Exit
    Menachem Elimelech: Exit
    Nathalie Tufenkji: Exit

    Progress and Final Reports:

    Original Abstract
  • 2002 Progress Report
  • 2003 Progress Report
  • 2004