Received for review June 19, 2007

Revised manuscript received October 24, 2007

Accepted October 25, 2007

Introduction

Quantum dots (QDs), which vary in size from 1 to 10 nm, are of great interest due to their tunable optical properties (1–3). Although the current market for QDs is small, it is growing rapidly as applications in optical devices and biological analysis are developed (3–6). QDs may ultimately find their way into aquatic environments through manufacturing and commercial use. Modifying the surfaces of QDs with different capping ligands (hydrophobic or hydrophilic) provides stability in biomedical media and water (4–9). Stable, water soluble QDs are often capped with hydrophilic organic acids containing both carboxyl and thio (mercapto) groups, such as mercaptosuccinic acid (9). One end of the capping ligand (thio group) binds with the QD surface, and the other, polar end (carboxyl group) extends into solution. Research has shown that if mercaptoaectic (thioglycolic) acid-capped CdSe QDs are exposed to ultraviolet light, they can release Cd²⁺ and are acutely toxic to cells (10). Mercaptopropionic acid-capped CdTe QDs have also been found to be cytotoxic due to Cd²⁺ released from their cores and/or reactive oxygen species formed via CdTe-triggered photooxidation processes (11). Because of these known risks and potential unknown dangers, the fate of QDs in the environment needs to be examined.

In aquatic environments, the stability of functionalized QDs may be affected by many factors, including pH and ionic strength. Previous research has focused on bare nanomaterials such as hematite nanoparticles and fullerenes (12–14); only a few studies have addressed functionalized nanomaterials (15, 16). But functionalized nanomaterials behave differently than their bare counterparts; for example, Chen et al. found that bare and alginate-coated hematite exhibited different aggregation tendencies in the presence of MgCl₂ and CaCl₂ (15). The fate of functionalized QDs in water treatment processes may also differ from that of bare nanoparticles. Therefore, we investigated the characteristics of functionalized QDs, explored factors affecting their stability in aquatic environments, and examined their removal during potable water treatment.

Thioglycolate capped CdTe QDs were used in this research because they are commercially available and representative of water-stable, functionalized QDs. Our results on the stability of the selected QDs as function of pH, ionic strength, and ionic composition as well as the removal efficiency of QDs from nanopure and tap water by simulated water treatment processes suggest that capping ligands play an important role in determining the fate of QDs in water.

Materials and Methods

Quantum Dots. Water soluble thioglycolate capped CdTe QDs were purchased from American Dye Source Inc., Canada. Thioglycolate (HS-CH₂-COO⁻) capping ligands are bound to the CdTe cores of QDs via thio groups. QDs were received as red stock suspensions with mass concentrations of 0.25 wt Cd % and pH 10 to 11. The QDs ranged from 3.5 to 4.5 nm in size, as reported.

The QD stock suspension was stored under nitrogen to prevent oxidation of the thio groups in the capping ligands. Experimental QD suspensions were prepared by diluting stock suspensions with nanopure water to one-tenth of the original concentration immediately before experiments, except where noted. The diluted suspensions had pH 9, conductivity of approximately 930 µs/cm, and 0.25 g/L Cd from the QDs.

Particle Size Analysis. A dynamic light scattering (DLS) instrument (90 Plus, Brookhaven Instruments Corp., NY) was employed to measure the sizes of QDs in water. New, particle-free polystyrene vials were rinsed with nanopure water before the QD suspension was added. All vials were used only once. Particle size measurements were conducted on the previously mentioned QD suspensions prepared with nanopure water at room temperature (23 ± 2 °C). Both the mean particle sizes (the so-called “z-average diameter”) and particle size distributions of the QDs were obtained by DLS. The nanopure water and all solutions used for QD experiments were verified via DLS analysis to be free of particles.

ζ-Potential Analysis. A ZetaPALS analyzer (Brookhaven Instruments Corp., NY) was employed to measure the electrophoretic mobility of the QDs. Measurements were conducted at 25 °C and the desired ionic strength. ζ-potentials were determined from electrophoretic mobility according to Smoluchowski’s equations (17).

Determination of Capping Ligand and QD Concentrations. The concentration of thioglycolate capping ligands was determined by measuring the total organic carbon (TOC, TOC-VCPN, Shimadzu, Japan). The TOC content in thioglycolate (C₂H₄O₂S⁻) was about 26% of the total mass. The TOC measurement on the QD stock suspension yielded a total thioglycolate (free plus bound) concentration of 12.99 g/L. With the addition of 0.01 M MgCl₂, the QDs formed aggregates, which were separated by centrifugation at 1300 g for 15 min. Optical observation showed that all red QDs settled at the bottom of the centrifugation tubes, and the supernatant was transparent. DLS measurement of the supernatant confirmed that all QDs settled out of solution. Measurement of the TOC in the supernatant after centrifugation yielded a free thioglycolate concentration of 11.50 g/L. The difference between the concentrations of total and free thioglycolate capping ligands is the bound capping ligand concentration of 1.49 g/L.

The concentration of QD CdTe cores was determined from the Cd mass concentration. The Cd content in QD cores is about 46.8%. In the QD stock suspension, the concentration of Cd from the QDs was 2.47 g/L as measured by graphite furnace atomic absorption spectroscopy (GFAA, Varian SpectrAA 400 Zeeman, Varian Inc., CA) following nitric acid digestion of samples (18). This value was in agreement with the vendor (reported 0.25 wt%). Thus, the mass concentration of QD cores is 5.34 g/L. The concentration of QDs including cores and capping ligands is 6.83 g/L in the stock suspension. Consequently, the concentration of QDs can be determined by measuring Cd using the above method.

Stability of QDs. The stability of QDs in water was investigated under various pH and electrolyte compositions. In a 100 mL beaker, 0.68 g/L QD suspensions were prepared in nanopure water, and their pH values were adjusted using 0.1 M KOH and 0.1 M HCl. KCl was used as a surrogate monovalent electrolyte and added to the suspensions to obtain a variety of concentrations up to 150 meq K⁺/L. The influence of two divalent electrolytes (CaCl₂ and MgCl₂) and a trivalent electrolyte (Al₂(SO₄)₃) at concentrations from 0 to 40 meq/L were also examined. The stability of the QDs was evaluated by measuring their ζ-potentials in the different solutions and by monitoring the changes in their particle sizes over time.

Potentiometic and Calcium Titration for QDs. Potentiometic titrations were conducted on the 0.68 g/L QD suspensions to determine their acidity. To serve as a background electrolyte, 0.01 M NaCl was added to the suspension. Before titration, 50 mL of the QD suspension was purged with argon gas for at least 10 min to remove dissolved CO₂. Then, using 0.1 M HCl solution, the suspension was titrated from pH 9 to pH 2, which is lower than the pK_a (3.67) of the carboxyl groups in the thioglycolate capping ligands (19). A pH 9 control solution (NaOH) without QDs was titrated using the same procedure. By comparing the amounts of HCl required to reduce the pH of the QD suspension and control solution, the acidity of the QD suspension (0.68 g/L) was determined to be 14.1 meq/L, which is close to the total concentration of capping ligands.

Calcium titrations of the 0.68 g/L QD suspension were conducted at pH 5 with 1 M CaCl₂. This pH was selected to reduce the dissolution of CO₂ and to deprotonate all thioglycolate capping ligands. Furthermore, during titration, the suspension was purged with argon gas to prevent the dissolution of CO₂. The concentrations of free (unbound) and bound capping ligands in the QD suspension were 12.63 and 1.633 mM, respectively. Before titration, 0.1 M KCl was added to the suspensions to provide constant ionic strength. At each titration point, the suspension was mixed thoroughly for 2 min, and then the concentration of free calcium ions was measured using a calcium ion selective electrode (Orion 9320, Thermo Orion Corp., MA) with a single junction reference electrode (Orion 9001, Thermo Orion Corp., MA). The electrode was calibrated by titrating a blank KCl solution with the same ionic strength as the samples. The free calcium ion concentration was obtained from the corresponding electrode response (mV). The amount of complexed calcium ions in the samples was the difference between the total added and the free calcium ions. A sodium thioglycolate solution having the same concentration (12.63 mM) as the free capping ligands in the titrated QD suspension was prepared and titrated by CaCl₂ using identical procedures to determine the complexation between calcium ions and free thioglycolate.

Removal of QDs by Alum Coagulation. Removal of QDs by water treatment processes was evaluated using jar tests. Various doses (0−2.86 meq Al³⁺/L) of alum (Al₂(SO₄)₃· 16 H₂O) were added to 2 L jars of nanopure or tap water containing 13.66 mg/L QDs. The nanopure water was buffered with 0.01 M NaHCO₃ and had a pH of 8.2 ± 0.2. Tap water (Tempe, AZ) had a pH of 8.1 ± 0.2 and a conductivity of 750−930 µs/cm. The concentrations of Ca²⁺ and Mg²⁺ were measured using inductively coupled plasma atomic emission spectroscopy (ICP-AES, iCAP 6000, Thermo Fisher Scientific Inc., MA). After alum addition, the pH of the suspensions was adjusted with 0.1 M HCl or 0.1 M NaOH to 7.5 ± 0.3. Jar tests were conducted as follows (1): rapid mixing (coagulation) for 1 min at 100 rpm (2), slow mixing (flocculation) for 30 min at 30 rpm, and (3) settling (sedimentation) for 1 h. After sedimentation, supernatants were filtered with 0.45 µm filter paper (Nylaflo Membrane Disc Filter, Pall Corp., NY). The filtered volume was less than that which would be required to form a monolayer of flocs on the filter paper.

Results and Discussion

Particle Sizes. The DLS measurement indicated that not all QDs in water presented as primary nanoparticles. Figure 1 illustrates a bimodal size distribution of QDs in nanopure water. As the vendor reported, the primary QD particle sizes ranged from 3.5 to 4.5 nm. More than 99% of the particles in the total volume had sizes of 3.5−6.5 nm and were primary QD particles. However, about 1% of the particles had sizes of 60−100 nm and were likely to be aggregates of QDs. The z-average diameter measured by DLS was around 80 nm. The z-average diameter is weighted by the intensity of the light scattering, and larger particles scatter more light. Therefore, the z-average diameter for a poly dispersed suspension is close to the sizes of larger particles, which in this QD system range from 60 to 100 nm.

Figure 1. Particle size distribution of QDs in nanopure water. Click to Enlarge

Stability of QDs Under Various pH and Monovalent Cations. Figure 2 shows that in the absence of K⁺, the ζ-potentials of QDs remained at −30 mV within a pH range of 5−12. Thus, red QD suspensions were clear when the pH was greater than 5, and no QD aggregation was observed. However, as the pH decreased from 5, the negative ζ-potentials also decreased. At pH 3 and lower, the ζ-potentials were less negative than −20 mV, and black QD aggregates larger than 2 µm formed.

Figure 2. ζ-potentials of QDs as a function of pH under various KCl concentrations. Click to Enlarge

The pH-based change in the ζ-potentials and stability of QDs appears to be related to the thioglycolate capping ligands. The carboxyl group in thioglycolic acid has a pK _a of 3.67 (19). Accordingly, at pH values higher than 5, more than 95% of the capping ligands were deprotonated to thioglycolate such that the QDs possessed a relatively high negative surface charge and thus remained stable. No obvious increase occurred in the negative ζ-potentials as the pH rose from 5 to 12 because the change in deprotonated thioglycolate was less than 5% within this pH range. In contrast, at pH 4 and lower, more than 70% of the thioglycoate was protonated to form thioglycolic acid, so the negative ζ-potentials of the QDs were reduced, and they became unstable. The formation of QD flocs at pH 3 occurred because almost all thioglycolate was protonated such that the surface potentials of the QDs were low and could not provide enough electrostatic repulsion between the particles to prevent aggregation.

Addition of a monovalent electrolyte (KCl) resulted in an increase in ionic strength and therefore in the compression of the electric double layers (EDLs) of the QDs. At pH 5 and higher, as the KCl concentration increased from 0 to 150 meq K⁺/L, the ζ-potentials of the QDs decreased from −30 to −20 mV (Figures 3), and the EDL thickness was compressed to 0.79 nm. However, EDL compression did not cause the QD aggregation. This result is quite different from that for bare nanoparticles (e.g., hematite and C₆₀), which can be destabilized by EDL compression with a relatively low concentration (<0.01 M) of monovalent electrolyte (12, 14). A plausible explanation is that the capping ligands on the surfaces of the QDs extended into their electric double layers and prevented the QDs from approaching each other.

Figure 3. ζ-potentials of QDs in the presence of different cations at pH 5 and pH 8. Click to Enlarge

Aggregation of QDs in the Presence of Divalent Cations. QD aggregation in the presence of divalent cations was examined in the pH range of most surface water. Figure 3 presents the ζ-potentials of QDs over a range of Ca²⁺ and Mg²⁺ concentrations (0 to 40 meq/L) at both pH 5 and 8. The ζ-potential was around −30 mV for Ca²⁺ concentrations less than 0.5 meq/L and then became less negative as the Ca²⁺ concentration increased. At 10 meq/L Ca²⁺, the ζ-potential was only −10 mV. Mg²⁺ exhibited similar results. However, this value was much less than the zeta potentials of quantum dots (−20 mV) with K⁺ of 150 meq/L. Thus, divalent cations affect the ζ-potentials of QDs more strongly than ionic strength does, which may be related to the charge neutralization of divalent cations.

Figure 4a shows the changes in particle size for QDs in the presence of Ca²⁺ at pH 5. Based on the aggregation kinetics, the QD aggregation at low Ca²⁺ concentrations (no more than 1 meq/L Ca²⁺) occur over a relatively long observation time. It was found that the QDs remained stable for 40 min when the Ca²⁺ concentration was 0.5 meq/L. A concentration of 1 meq/L Ca²⁺ was required to induce QD aggregation. Even then, 30 min were needed for the average size of QD aggregates to become larger than 1 µm. However, at higher Ca²⁺ concentrations (2 meq/L or higher), the QDs aggregated very quickly, and the z-average diameter of QD aggregates increased to more than 3 µm after only 5 min. Figure 4b provides examples of particle size distributions at different Ca²⁺ concentrations. In the presence of Mg²⁺, QDs exhibited a similar aggregation tendency, but the threshold destabilization concentration was 2 meq/L Mg²⁺.

Figure 4. Aggregation of QDs in the presence of Ca2+ at pH 5. (a) Z-average diameters; (b) Particle size distributions after 30 min of aggregation. Click to Enlarge

The destabilization mechanism of divalent cations (Ca²⁺ or Mg²⁺) may involve the formation of complexes with the thioglycolate capping ligands on the QD surface that neutralize the negative surface charge of the QDs. The experimental results on ζ-potentials support this mechanism. In addition, QDs with a ζ-potential of −26 mV aggregated at 1 meq/L Ca²⁺. However, QDs in the presence of KCl at pH 5 were stable with ζ-potentials as low as −20 mV. This also indicates that complexes with capping ligands may bridge QDs to form aggregates. Thus, divalent cation complexation constants for capping ligands can be used to quantify QD aggregation.

Calcium Complexation with QDs Capping Ligands. Calcium complexation is thought to occur via Ca²⁺ attaching to the carboxyl groups in the capping ligands. A Ca²⁺ ion may bond with either one (monodentate) or two (bidentate) capping ligand sites. In the studied QD suspension, Ca²⁺ complexation occurs only with deprotonated thioglycolate capping ligands, but these can be free (unbound) or bound (see equations in Table 1). Protonations of either are considered the same (Table 1).

The Ca²⁺ complexation constants were determined by calcium titration. Figure 5 shows the calcium complexation data for thioglycolate solution and the QD suspension at pH 5. Considering the potential difference between the calcium complexation constants of free and bound capping ligands, the calcium complexation constants of bound capping ligands were determined through two steps. First, the calcium complexation constants of free capping ligands were calculated using calcium complexation data from thioglycolate solution (containing only free capping ligands). By fitting the calculated results with the experimental data for the thioglycolate solution, the calcium monodentate (K₁) and bidentate (K₂) complexation constants for free thioglycolate capping ligands are found to be 30 (log K₁ = 1.48) and 977 (log K₂ = 2.99), respectively. Then, based on the K₁ and K₂ values and the results of the calcium titration of the QD suspension, the complexed free capping ligands in the QD suspension can be determined. The total complexed capping ligands (free and bound) in the QD suspension are obtained from the calcium titration results. The difference between the total complexed capping ligands and the free complexed capping ligands is the complexed bound capping ligands. From this data, the calcium monodentate (K₃) and bidentate (K₄) complexation constants for the bound capping ligands can be calculated to be 245 (log K₃ = 2.39) and 4.79 × 10⁴ (log K₄ = 4.68), respectively. These results show that binding to the QD surface through their thio groups enhances the ability of capping ligands to bind with Ca²⁺. The high complexation constants for the bound capping ligands suggest that even a low concentration of Ca²⁺ could result in the formation of calcium complexes with QD capping ligands, which supports the results of previous aggregation experiments. The calculation for complexed capping ligand species showed that when QDs aggregated after the addition of 1 meq/L Ca²⁺ at pH 5, about 9% of the bound capping ligands complexed with Ca²⁺ by monodentate and bidentate binding; this is approximately 1% of the total capping ligands in the QD suspension (see SI Figure S1).

Figure 5. Calcium complexation for the QD suspension and thioglycolate solution at pH 5. Click to Enlarge

Aggregation of QDs in the Presence of a Trivalent Cation. Similar to the divalent cations, the trivalent cation (Al³⁺) also decreased the negative ζ-potentials of the QDs and caused them to aggregate. Within the Al³⁺ concentration range of 0−40 meq/L, however, the aggregation tendency and changes in ζ-potentials were quite different at pH 5 and pH 8. As shown in Figure 3, at pH 5, the ζ-potentials of QDs became less negative as the Al³⁺ concentration increased, with aggregation occurring at 0.3 meq/L Al³⁺. QD aggregates larger than 1 µm formed within the observed time (30 min) in the presence of 0.6 meq/L Al³⁺ (the solubility of Al³⁺ is 1.5 meq/L at pH 5). At pH 8, however, the negative ζ-potentials of the QDs became more negative upon addition of Al³⁺, and at 3.0 meq/L Al³⁺, the ζ-potential reached −37 mV (Figure 3). Further increases in Al³⁺ concentrations decreased the negative ζ-potentials, however. Although the ζ-potential only decreased to −35 mV after the addition of 4.5 meq/L Al³⁺ (the solubility of Al³⁺ is 0.38 meq/L at pH 8), QD aggregates larger than 1 µm formed within 30 min.

The discrepancy in QD aggregation with Al³⁺ between pH 5 and 8 may be related to the complexation mechanism of aluminum with capping ligands. In water, Al³⁺ gives rise to hydrolytic species and presents as Al³⁺(H₂O)_n[(OH)_6-n]^n-6. Complexation with capping ligands may occur through carboxylic groups replacing either the original water molecules or the OH⁻ groups on the hydrated aluminum species. This “ligand exchange” is expressed as follows:

Removal of QDs by Alum Coagulation. Figure 6 shows the amounts of QDs remaining in water after treatment with different alum dosages. In buffered nanopure water without alum, the z-average diameter of the QDs remained around 60 nm, and less than 1% of the total mass of QDs was removed by sedimentation and 0.45 µm filtration. The addition of alum led to QD aggregation, but when the alum dosage was 0.95 meq/L Al³⁺ or lower, the aggregates only had z-average diameters of 550 nm, and sedimentation removed less than 12% of total mass (0.6 mg /L Cd from QDs). At alum dosages of 1.43 meq/L Al³⁺, settleable QD flocs larger than 2 µm formed, and sedimentation removal efficiencies were more than 90%. The high alum dosage required to remove QDs is related to their destabilization mechanism in the presence of Al³⁺. As discussed in the previous section, because the dominant hydrated Al³⁺ species is Al(OH)₄(H₂O)₂⁻ at pH 7.5, settleable QD flocs formed only by sweep floc. Aluminum complexation with the capping ligands consumed Al³⁺, however, and therefore the alum dosage needed to produce amorphous Al(OH)₃ precipitates was higher than the solubility of Al³⁺ (0.25 meq/L at pH 7.5). Further treatment by 0.45 µm filtration exhibited the potential to remove additional QD flocs. At a 0.95 meq/L Al³⁺ alum dosage, 0.45 µm filtration improved the removal efficiency of QDs after sedimentation from 12% to 93%.

Figure 6. Removal of QDs (13.66 mg/L) in water by alum coagulation. Click to Enlarge

The removal of QDs differed in tap and buffered nanopure water. The concentrations of Mg²⁺ and Ca²⁺ in tap water were 1.52 and 2.18 meq/L, respectively. Due to the presence of these divalent cations, QDs formed settleable flocs larger than 2 µm in tap water even without alum. Removal efficiency by sedimentation was about 70% by mass. The addition of alum did not increase removal efficiencies significantly. With an alum dosage up to 2.86 meq Al³⁺/L, the QDs removed by sedimentation were still no more than 80% of the total mass. With 0.45 µm filtration, QD removal only increased to about 85% regardless of the addition of alum.

Supporting Information Available

The species diagram of capping ligands (free and bound) under experimental conditions. This material is available free of charge via the Internet at http://pubs.acs.org.

* Corresponding author phone: (480) 965-3272 ; fax: (480) 965-0557; e-mail: yschen@asu.edu.

1. Alivisatos, A. P. Semiconductor clusters, nanocrystals, and quantum dots. Science. 1996, 217, 933–937.

2. Kenneth, J. K. Nanoscale Materials in Chemistry; John Wiley and Sons: New York, 2001.

3. Alivisatos, A. P. Perspectives on the physical chemistry of semiconductor nanocrystals. J. Phys. Chem. 1996, 100, 13226–13239. [Full Text - ACS] [ChemPort] [CrossRef]

4. Colvin, V. L.; Schlamp, M. C.; Alivisatos, A. P. Light-emitting diodes made from cadmium selenide nanocrystals and a semiconducting polymer. Nature 1994, 370, 354–357. [ChemPort] [CrossRef]

5. Chan, W. C. W.; Nie, S. M. Quantum dots bioconjugates for ultrasensitive nonisotopic detection. Science. 1998, 281, 2016–2018. [ChemPort] [Medline] [CrossRef]

6. Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Semiconductor nanocrystals as fluorescent biological labels. Science. 1998, 281, 2013. [ChemPort] [Medline] [CrossRef]

7. Murray, C. B.; Noms, D. J.; Bawendi, M. G. Synthesis and characterization of nearly monodisperse CdE (E = S, Se, Te) semiconductor nanocrystallites. J. Am. Chem. Soc. 1993, 115, 8706–8715. [Full Text - ACS] [ChemPort] [CrossRef]

8. Peng, Z. A.; Peng, X. Formation of high-quality CdTe, CdSe, and CdS nanocrystals using CdO as precursor. J. Am. Chem. Soc. 2001, 123, 183–184. [Full Text - ACS] [ChemPort] [Medline] [CrossRef]

9. Zhivko, Z.; Rumiana, B.; Hideki, O.; Rajan, J.; Yusuke, I.; Yoshinobu, B. Uncoated, broad fluorescent, and size-homogeneous CdSe quantum dots for bioanalyses. Anal. Chem. 2006, 78, 321–330. [Full Text - ACS] [Medline] [CrossRef]

10. Cho, J. S.; Maysinger, D.; Jain, M.; Rö der, B.; Hackbarth, S.; Winnik, F. M. Long-term exposure to CdTe quantum dots causes functional impairments in live cells. Langmuir. 2007, 23, 1974–1980. [Full Text - ACS] [ChemPort] [Medline] [CrossRef]

11. Derfus, A. M.; Chan, W. C. W.; Bhatia, S. N. Probing the cytotoxicity of semiconductor quantum dots. Nano Lett. 2004, 4 (1), 11–18. [Full Text - ACS] [ChemPort] [CrossRef]

12. Schudel, M.; Behren, S. H.; Holthoff, H.; Kretzschmar, R.; Borkovec, M. Absolute aggregation rate constants of hematite particles in suspensions: a comparison of two different surface morphologies. J. Colloid Interface Sci. 1997, 196, 241–253. [ChemPort] [Medline] [CrossRef]

13. Zhang, J.; Buffle, J. Kinetics of hematite aggregation by polyacrylic acid: importance of charge neutralization. J. Colloid Interface Sci. 1995, 174, 500–509. [ChemPort] [CrossRef]

14. Brant, J.; Lecoanet, H.; Wiesner, M. R. Aggregation and deposition characteristics of fullerene nanoparticles in aqueous systems. J. Nanopart. Res. 2005, 7, 545–553. [ChemPort] [CrossRef]

15. Chen, K. L.; Mylon, S. E.; Elimelech, M. Aggregation kinetics of alginate-coated hematite nanoparticles in monovalent and divalent electrolytes. Environ. Sci. Technol. 2006, 40, 1516–1523. [Full Text - ACS] [ChemPort] [Medline] [CrossRef]

16. Lecoanet, H. F.; Wiesner, M. R. Velocity effects on fullerene and oxide nanoparticle deposition in porous media. Environ. Sci. Technol. 2004, 38, 4377–4382. [Full Text - ACS] [ChemPort] [Medline] [CrossRef]

17. Hiemenz, P. C.; Rajagopalan R. Principles of Colloid and Surface Chemistry; Marcel Dekker: New York, 1997.

18. American Public Health Association; American Water Works Association; Water Environment Federation. Standard Methods for the Examination of Water and Wastewater; APHA: Washington, DC, 1998.

19. Edsal, J. T.; Wyman, J. Biophysical Chemistry; Academic Press: New York, 1958.

---

Table 1. Protonation Constants and Calcium Binding Constants of Capping Ligandsa
aqueous reaction Log K constants for free (unbound) thioglycolate capping ligands   3.67b 1.48c 2.99c constants for bound thioglycolate capping ligands   3.67d 2.39c 4.68c
a I = 0.1 M; activity coefficients are calculated using the Davis equation. b Ref 19. c This study. d Assumed to be the same as free (unbound) thioglycolate.

---