TufenkjiElimelechDDM Model

Deposition Patterns of Colloidal Particles in Saturated Porous Media: 

Deposition Patterns of Colloidal Particles in Saturated Porous Media Nathalie Tufenkji Menachem Elimelech Yale University Department of Chemical Engineering Environmental Engineering Program Deviation from Colloid Filtration Theory

Outline : 

Outline Background and Motivation Experimental Investigation Dual Deposition Mode Model Supporting Experiments Conclusions

Background and Motivation : 

Background and Motivation Classical Colloid Filtration Theory

Slide4: 

Background and Motivation Classical Colloid Filtration Theory

Slide5: 

Background and Motivation Classical Colloid Filtration Theory

Slide6: 

Background and Motivation Classical Colloid Filtration Theory

Slide7: 

Background and Motivation Classical Colloid Filtration Theory

Slide8: 

Background and Motivation Deviations from the Classical Filtration Theory

Slide9: 

Background and Motivation Deviations from the Classical Filtration Theory

Experimental Investigation: 

Experimental Investigation Model Colloidal Particles Experimental Conditions C0 5 x 106 – 2 x 107 particles/mL dc 0.33 mm glass beads dp 3 μm CML particles U0 8.3 x 10-5 m/sec  0.37 pH ~ 8 Lc 12.6 cm

Experimental Investigation: 

Experimental Investigation Model Colloidal Particles Experimental Conditions C0 5 x 106 – 2 x 107 particles/mL dc 0.33 mm glass beads dp 3 μm CML particles U0 8.3 x 10-5 m/sec  0.37 pH ~ 8 Lc 12.6 cm

Experimental Investigation: 

Experimental Investigation Model Colloidal Particles

Experimental Investigation: 

Experimental Investigation Model Colloidal Particles

Experimental Investigation: 

Experimental Investigation Model Colloidal Particles

Experimental Investigation: 

Experimental Investigation Interpreting Results Determining the Attachment Efficiency, α where

Experimental Investigation: 

Experimental Investigation Interpreting Results Determining the Attachment Efficiency, α where Tufenkji and Elimelech, ES&T, 2004.

Experimental Investigation: 

Experimental Investigation Interpreting Results Determining the Attachment Efficiency, α where Tufenkji and Elimelech, ES&T, 2004.

Investigating with 3 μm Particles : 

Investigating with 3 μm Particles Influence of Solution Chemistry on Particle Deposition

Investigating with 3 μm Particles: 

Investigating with 3 μm Particles Influence of Solution Chemistry on Particle Deposition S(x) (g particles/g glass beads) CFT: Classic filtration theory

Investigating with 3 μm Particles: 

Investigating with 3 μm Particles Influence of Solution Chemistry on Particle Deposition S(x) (g particles/g glass beads) CFT: Classic filtration theory

Investigating with 3 μm Particles: 

Investigating with 3 μm Particles Influence of Solution Chemistry on Particle Deposition S(x) (g particles/g glass beads) CFT: Classic filtration theory

Investigating with 3 μm Particles: 

Investigating with 3 μm Particles Influence of Solution Chemistry on Particle Deposition S(x) (g particles/g glass beads) CFT: Classic filtration theory

Investigating with 3 μm Particles: 

Investigating with 3 μm Particles Influence of Solution Chemistry on Particle Deposition S(x) (g particles/g glass beads) CFT: Classic filtration theory

Interpreting Results: 

Interpreting Results DLVO Theory of Colloidal Stability DLVO Parameters Zeta potentials used in place of surface potentials Hogg et al (1966) expression used for electrostatic interaction energies Retarded van der Waals interactions from Gregory (1981) Hamaker constant, A 10-20 J

Interpreting Results: 

Interpreting Results DLVO Theory of Colloidal Stability DLVO Parameters Zeta potentials used in place of surface potentials Hogg et al (1966) expression used for electrostatic interaction energies Retarded van der Waals interactions from Gregory (1981) Hamaker constant, A 10-20 J

Interpreting Results: 

Interpreting Results DLVO Theory of Colloidal Stability

Interpreting Results: 

Interpreting Results DLVO Theory of Colloidal Stability Local charge heterogeneities on glass beads provide sites for 1o minimum deposition Metal oxide impurities (Al2O3, CaO, MgO, Na2O, Fe2O3)

Interpreting Results: 

Interpreting Results DLVO Theory of Colloidal Stability Local charge heterogeneities on glass beads provide sites for 1o minimum deposition Metal oxide impurities (Al2O3, CaO, MgO, Na2O, Fe2O3)

Interpreting Results: 

Interpreting Results Multiple Modes of Deposition “fast” deposition in 2o minimum α2min calculated based on model of Hahn & O’Melia (2004)

Interpreting Results: 

Interpreting Results Multiple Modes of Deposition “fast” deposition in 2o minimum “slow” deposition in 1o minimum unfavorable conditions

Interpreting Results: 

Interpreting Results Multiple Modes of Deposition “fast” deposition in 2o minimum “slow” deposition in 1o minimum “fast” deposition in 1o minimum unfavorable conditions favorable condition

Interpreting Results: 

Interpreting Results Dual Deposition Mode Model Bimodal distribution in particle deposition rate coefficient, k

Interpreting Results: 

Interpreting Results Dual Deposition Mode Model Bimodal distribution in particle deposition rate coefficient, k calculations based on transport-limited deposition rate

Interpreting Results: 

Interpreting Results Dual Deposition Mode Model Bimodal distribution in particle deposition rate coefficient, k calculations based on transport-limited deposition rate

Interpreting Results: 

Interpreting Results Dual Deposition Mode Model Bimodal distribution in particle deposition rate coefficient, k calculations based on transport-limited deposition rate

Interpreting Results: 

Interpreting Results Comparison of CFT with Dual Deposition Mode Model S(x) (g particles/g glass beads)

Interpreting Results: 

Interpreting Results Comparison of CFT with Dual Deposition Mode Model C/C0, EXPT = 0.95 C/C0, DDMM = 0.95 S(x) (g particles/g glass beads)

Interpreting Results: 

Interpreting Results Comparison of CFT with Dual Deposition Mode Model C/C0, EXPT = 0.95 C/C0, DDMM = 0.95 C/C0, EXPT = 0.91 C/C0, DDMM = 0.94 S(x) (g particles/g glass beads)

Interpreting Results: 

Interpreting Results Comparison of CFT with Dual Deposition Mode Model C/C0, EXPT = 0.95 C/C0, DDMM = 0.95 C/C0, EXPT = 0.64 C/C0, DDMM = 0.65 C/C0, EXPT = 0.91 C/C0, DDMM = 0.94 S(x) (g particles/g glass beads)

Interpreting Results: 

Interpreting Results Comparison of CFT with Dual Deposition Mode Model C/C0, EXPT = 0.95 C/C0, DDMM = 0.95 C/C0, EXPT = 0.64 C/C0, DDMM = 0.65 C/C0, EXPT = 0.91 C/C0, DDMM = 0.94 C/C0, EXPT = 0.41 C/C0, DDMM = 0.41 S(x) (g particles/g glass beads)

Slide41: 

Changes in ionic strength and pH alter DLVO profiles Elimination of 2o minimum will result in release of deposited particles High pH promotes release of particles deposited on metal oxide impurities Supporting Experimentation Elution of Retained Particles

Slide42: 

Supporting Experimentation Elution of Retained Particles 10 mM IS + 3 μm colloids 10 mM IS

Slide43: 

Supporting Experimentation Elution of Retained Particles 10 mM IS + 3 μm colloids 10 mM IS

Slide44: 

Supporting Experimentation Elution of Retained Particles 0.1 mM KHCO3 10 mM IS + 3 μm colloids 10 mM IS

Slide45: 

Supporting Experimentation Elution of Retained Particles 0.1 mM KHCO3 10 mM IS + 3 μm colloids 1 mM KOH 10 mM IS

Slide46: 

Supporting Experimentation Elution of Retained Particles 10 mM IS 100 mM IS 30 mM IS

Slide47: 

Supporting Experimentation Elution of Retained Particles 10 mM IS 100 mM IS 30 mM IS frel = Nrel/Ndep= 0.13

Slide48: 

Supporting Experimentation Elution of Retained Particles 10 mM IS 100 mM IS 30 mM IS frel = Nrel/Ndep= 0.13 frel = 0.30 frel = 0.62

Slide49: 

Comparison of Model Predictions and Experimental Results a Determined by nonlinear regression b Predicted by DDM model. b

Slide50: 

Comparison of Model Predictions and Experimental Results a Determined by nonlinear regression b Predicted by DDM model. b

Slide51: 

Comparison of Model Predictions and Experimental Results a Determined by nonlinear regression b Predicted by DDM model. b

Conclusions : 

Conclusions In the presence of repulsive EDL interactions, particle deposition rates deviate significantly from CFT A DDM model is presented which considers the combined influence of “fast” and “slow” deposition Agreement between experimental results and model calculations suggests that the deviation from CFT is controlled by the concurrent existence of favorable and unfavorable interactions

Acknowledgements : 

Acknowledgements Natural Sciences and Engineering Research Council of Canada (NSERC) National Science Foundation (NSF) US EPA