pubs.acs.org/Langmuir © 2009 American Chemical Society
Electrokinetic Potential of Nanoparticles in Reverse AOT Micelles: Photometric Determination and Role in the Processes of Heterocoagulation, Separation, and Concentration Alexander I. Bulavchenko*,† and Pavel S. Popovetsky‡ †
Nikolaev Institute of Inorganic Chemistry, Siberian Branch of the Russian Academy of Sciences, Novosibirsk, 630090 Russia and ‡Novosibirsk State University, Novosibirsk, 630090 Russia Received July 7, 2009. Revised Manuscript Received November 9, 2009
A simple photometric method for determining the electrophoretic mobility of nano- and microparticles in reverse micelles and in solvents with a low dielectric permittivity (2-5) has been developed. The method is based on the use of a thermostatically controlled diaphragm-based optical cell (length 2 cm) with three vertical plane-parallel electrodes (2 3 cm; interelectrode gap, 0.3 cm) placed into a standard photocolorimeter. When an electrostatic field (100-600 V) is applied, the particles begin to move away from the electrode of the same polarity. The path traveled by the particles for a given time (2-30 s) is calculated from the change in the optical density of the solution in the near-electrode zone. The electrophoretic potential of nanoparticles in the model systems, calculated from the values of electrophoretic mobility by Huckel-Onsager theory, varied from 70 (Ag nanoparticles in AOT micelles in decane) to -73 mV (aggregated SiO2 nanoparticles in a decane-chloroform mixture). Calculations by the classical Deryaguin-Landau-Verwey-Overbeek (DLVO) theory determined the contribution of the electrostatic interaction to the stability of the studied systems. We have shown that the surface charge of nanoparticles permits: (1) an electrophoretic concentration of the charged nanoparticles (Ag) with an enrichment factor of up to 104, (2) the separation of nanoparticles with zero (C60) and a high (Ag) electrokinetic potentials, and (3) the formation of electrostatically bound aggregates (Ag-SiO2) through the heterocoagulation of oppositely charged particles.
1. Introduction The surface charge and potential of nanoparticles in different solvents (aqueous,1,2 nonaqueous,3 and supercritical4) are fundamental properties in liquid media. The charge of nanoparticles determines the stability of colloid systems containing them and governs the processes of coagulation5 or heterocoagulation.6,7 The surface charge of nanoparticles permits their use in the construction of nanoelectronic circuits,8 the design of electrophoretic displays,9 and the formation of core-shell nanostructures.7 The surface charge plays a crucial role in the analyticalsignal multiplication10 and favors the binding and identification of DNA and other bioorganic molecules.11,12 *Corresponding author. E-mail:
[email protected]. Phone/Fax: þ7(383) 316-5349.
(1) Velikov, K. P.; Zegers, G. E.; Blaaderen, A. Langmuir 2003, 19, 1384–1389. (2) Vanifatova, N. G.; Spyvakov, B. Y. Russ. Chem. J. 2005, 49, 16–21. (3) Hsu, M. F.; Dufresne, E. R.; Weitz, D. A. Langmuir 2005, 21, 4881–4887. (4) Ryoo, W.; Dickson, J. L.; Dhanuca, V. V.; Webber, S. E.; Bonnecaze, R. T.; Johnston, K. P. Langmuir 2005, 21, 5914–5923. (5) Zhang, Q.; Thompson, M. S.; Caramichael-Baranauskas, A. Y.; Caba, B. L.; Zalich, M. A.; Lin, Y.-N.; Mefford, O. T.; Davis, R. M.; Riffle, J. S. Langmuir 2007, 23, 6927–6934. (6) Castro, R. H. R.; Kodama, P. K.; Gouvea, D.; Muccillo, R. J. Mater. Sci. 2009, 44, 1851–1857. (7) Park, H. S.; Dodbiba, G.; Cao, L. F.; Fujita, T. J. Phys.: Condens. Matter 2008, 20, 204105. (8) Gupta, S.; Zhang, Q.; Russel, T. P. Nano Lett. 2006, 6, 2066–2069. (9) Kim, J.; Garoff, S.; Anderson, J. L.; Schlangen, L. J. M. Langmuir 2005, 21, 10941–10947. (10) Tan, S.; Erol, M.; Attygalle, A.; Du, H.; Sukhishvili, S. Langmuir 2007, 23, 9836–9843. (11) Zanchet, D.; Micheel, C. M.; Parak, W. J.; Gerion, D.; Williams, S. C.; Alivisatos, A. P. J. Phys. Chem. B 2002, 106, 11758–11763. (12) Rezwan, K.; Studart, A. R.; V€or€os, J.; Gauckler, L. J. J. Phys. Chem. B 2005, 109, 14469–14474. (13) Qu, Q.; Zhang, X.; Shen1, M.; Liu, Y.; et al. Electrophoresis 2008, 29, 901– 909.
736 DOI: 10.1021/la903583r
Electrophoresis is a direct, reliable, and convenient method for determining the electrical parameters of nanoparticles.13 The determination of the electrokinetic potential of nanoparticles in nonaqueous media3 is of special (continuously increasing) interest because synthesis in reverse micelles is one of the most popular methods of nanoparticle preparation.14 Organic solvents such as saturated hydrocarbons (iso-octane, decane, etc.) serve as a dispersion media for reverse micelles. The available analytical techniques are difficult to apply to organic media with a low dielectric permittivity. For example, one of the main requirements of organic solvents in nonaqueous capillary electrophoresis is a high dielectric permittivity (>30).15 Phase analysis light scattering (PALS) permits the measurement of the electrophoretic mobility in media with a low dielectric permittivity,16 but commonly used commercial instruments determine the ζ potential only of the particles that are about an order of magnitude larger than the minimum size determinable by photon correlation spectroscopy (usually ∼5 compared with 0.5 nm). Photometric detection of the velocity of nanoparticle motion seems to be promising for micellar systems with small (15 nm; hence, κ a e 0.1. Thus, the electrokinetic potential should be calculated by the formula obtained for the case in which the EDL thickness is much greater than the particle size (κ a e 1). Therefore, the electrokinetic potential was calculated by the Huckel-Onsager formula25,26 3 η uel ζ ¼ 2 εε0 E
ð3Þ
(23) Van Der Minne, J. L.; Hermanis, P. H. J. Colloid Sci. 1952, 7, 600–615. (24) Luan, Y.; Xu, G.; Dai, G.; Liang, H. Colloid Polym. Sci. 2003, 282, 110–118. (25) O’Brien, R. W.; White, L. R. J. Chem. Soc., Faraday Trans. II 1978, 74, 1607–1626. (26) Delgado, A. V.; Gonzalez-Caballero, F.; Hunter, R. J.; Koopal, L. K.; Lyklema, J. J. Colloid Interface Sci. 2007, 309, 194–224.
Langmuir 2010, 26(2), 736–742
Bulavchenko and Popovetsky
Article
Figure 3. TEM (JEM-2010) photomicrographs of silver (A) and SiO2 (B) nanoparticles with size distribution functions (constructed by measuring the diameters of 150-200 particles from TEM photomicrographs at different magnifications).
where a1 and a2 are the diameters of the two particles, h is the particle-surface separation, and ζ1 and ζ2 are their ξ potentials (ζi ≈ ji; ji is the surface potential of the particle, as approximated by the measured ξ potential). By neglecting the electromagnetic retardation, the van der Waals interaction is determined from the formula27 FvdW ¼ Figure 4. Typical experimental dependence on the time of the path traveled by silver nanoparticles during their motion from the central electrode.
where uel is the particle velocity, η is the dynamic viscosity, and E is the electric-field strength. The calculated electrokinetic potentials for different systems are listed in Table 1. Thus, the proposed photometric method for determining the electrophoretic mobility of particles in media with a low dielectric permittivity covers a wider range of particle sizes (including the smallest ones, from 1 nm and smaller) as compared to PALS. The method is simple and does not require expensive equipment. Later, we estimate the contribution of the electrostatic interaction due to the surface charge of nanoparticles to the stability of the studied systems and consider the prospects for using the obtained results. 3.3. DLVO Analysis of the Stability of Dispersed Systems with Nanoparticles. According to DLVO theory,27 the stability of dispersed systems is determined by the total energy of pair particle interaction, F, F ¼ Felec þ FvdW þ Fosm
ð4Þ
where Felec is the energy of the electrostatic interaction due to the overlapping of electrical double layers, FvdW is the energy of the van der Waals interaction, and Fosm is the energy due to the osmotic pressure arising from the overlapping of the adsorption layers of nanoparticles. In the general case, the electrostatic interactions of spherical particles of different sizes at κ a < 5 are calculated as28 Felec ¼ πε0 ε
a1 a2 ½ðζ þ ζ2 Þ2 lnð1 þ e -Kh Þ a1 þ a2 þ h 1 þ ðζ1 -ζ2 Þ2 lnð1 -e -Kh Þ
ð5Þ
(27) Derjagin, B. V.; Churaev, N. V., Muller V. M. Surface Force; Nauka: Moscow, 1987. (28) Sun, J.; Velamakanni, B. V.; Gerberich, W. W.; Francis, L. F. J. Colloid Interface Sci. 2004, 280, 387–399.
Langmuir 2010, 26(2), 736–742
-A132 2a1 a2 2a1 a2 þ 6 h2 þ 2a1 h þ 2a2 h h2 þ 2a1 h þ 2a2 h þ 4a1 a2 !# h2 þ 2a1 h þ 2a2 h ð6Þ þ ln 2 h þ 2a1 h þ 2a2 h þ 4a1 a2
where A132 is the Hamaker constant, with indices 1 and 2 marking the particles and index 3 marking the solvent. A132 is calculated from pffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffi A132 ¼ ð A11 - A33 Þ ð A22 - A33 Þ
ð7Þ
The interaction of the surface layers of nanoparticles is estimated from the formula29 Fosm ¼
4πakB T 2 1 h -χ l - ; l < h < 2l φ υsolv 2 2 "
φ ¼ 0:9
#
3la2 ða þ lÞ
3
-a3
;χ ¼
ð8Þ
ν3 ðδ3 -δ2 Þ2 RT
where vsolv is the molecular volume of the solvent, φ is the volume fraction occupied by AOT, χ is the Flory-Huggins interaction parameter, l is the chain length of the AOT, δi represents the Hildebrand solubility parameters, and ν3 is the molar volume of the solvent. When the particles approach each other at a distance smaller than 2l, the carbon chains of the AOT molecules interpenetrate, resulting in a local increase in their concentration and an osmotic pressure. The latter leads to the spontaneous flow of solvent into the interpenetrated volume, forcing the particles apart. In the latter case, we used a simple dependence for similar particles because of the lack of experimental data for some parameters. (For the calculations, we used the values of (29) Herrera, A. P.; Resto, O.; Briano, J. G.; Rinaldi, C. Nanotechnology 2005, 16, 618–625. (30) Chen, K. L.; Elimelech, M. Langmuir 2006, 22, 10994–11001. (31) Kruglyakov, P. M.; Rovin, Yu. G. Physical Chemistry of Black Hydrocarbon Films; Nauka: Moscow, 1978.
DOI: 10.1021/la903583r
739
Article
Bulavchenko and Popovetsky
Figure 5. Changes in Vt (voltage at the photocolorimeter electron amplifier) (A) and recalculated lt (B) at six different electrode voltages (as shown; three measurement runs at each voltage). Table 1. Determined Electrokinetic Potentials of Nano- and Microparticles (P = 0.95)
nanoparticle
Figure 6. Velocity of silver nanoparticles (dependence 1, AOTdecane system) and SiO2 microparticles (dependence 2, decanechloroform system) as a function of the field strength.
electrophoretic mobility 1010 (m2/(V s))
surfactantsolventb
ζ potential (mV)
Ag AOT-decane 9.0 ( 0.2 70 ( 2 Au AOT-decane 1.6 ( 0.2 13 ( 2 AOT-decane -2.2 ( 0.1 -17 ( 1 Ag2CrO4 CHCl3-decane (1:1) 0 0 fullerene C60 22 ( 5 80 ( 10 Ag CHCl3-decane (1:1) CHCl3-decane (1:1) -20 ( 2 -73 ( 8 SiO2a a Microparticle. b Dielectric permittivity and viscosity used for the calculations were, respectively, 2.0 and 0.87 mPa/s for the system AOT-decane and 3.4 and 0.70 mPa/s for the system CHCl3decane (1:1).
Table 2. Parameters Used for the Calculation of the Interaction Energy decanedecane chloroform SiO2 Aii (Hamaker constant, 10-20 J) a (diameter of particles, nm) ζ (zeta potential, mV)
Figure 7. Dependence of the specific electrical conductivity of micellar AOT-decane solutions on the volume content of the aqueous pseudophase.
parameters given in ref 29). This interaction was introduced to estimate the depth of the potential well, which is determined by the X coordinate cutting off the van der Waals and Coulomb interactions. Varying the parameters in eq 8 over a wide range of values changes the Fosm(h) dependence, but the position of the potential well remains practically the same. The constants of all interactions used for the calculations are listed in Table 2. Although the obtained estimates are rough, a comparison of these data with the thermal motion energy (3/2kBT) allows a reliable evaluation of the stability of the produced systems.5,6 In the Ag-Ag systems, different solvents show both electrostatic repulsion and van der Waals attraction, with the latter prevailing at small distances. As a result, a weak minimum appears (∼0.7kBT for the decane solution and ∼0.2kBT for the decane-chloroform mixture, Figure 8A). The systems are stable. In the system C60-C60 (Figure 8B), there is no electrostatic repulsion; the energy of the van der Waals interaction is negligible (2 10-6kBT) because of the small size of the nanoparticles and the small value of the Hamaker constant. As a result, the dispersed system with fullerene is stable. In the system Ag-C60, the van der Waals interaction is predominant but the energy of the van der Waals interaction is 740 DOI: 10.1021/la903583r
31
6.89
31
8.46
C60 30
Ag
27
10
7.5
5427
27 -73
0.7 0
4.8 80
lower than the energy of thermal motion (0.15kBT, Figure 8C) because of the small size of C60 particles and the small values of the Hamaker constant. The system is stable. In the Ag-SiO2 system, the interaction is actually not between two particles but between a silver nanoparticle and an aggregate of SiO2 nanoparticles. However, this fact can be neglected because a SiO2 nanoparticle is much larger than a Ag nanoparticle. Because of the opposite charges of the particles, the electrostatic interaction is negative and the depth of the potential well (7kBT) considerably exceeds the energy of thermal motion. Note that the electrostatic attraction exceeds the van der Waals interaction by more than an order of magnitude. As a result, the system is unstable. The interparticle interaction leads to heterocoagulation (Figure 8D). Note that for all systems the variation in the Hamaker constants and the thickness of the adsorption layer over a wide range of values did not lead to quantitative changes. 3.4. Concentration of Nanoparticles of Metals (Stable Systems). The stability of the systems and the presence (or absence) of the surface charge allows a series of colloid-chemical processes to be carried out in these systems. The high ζ potential of silver nanoparticles allows their electrophoretic concentration in AOT micelles in decane in a horizontal cell (Figure 1B) by the scheme proposed for gold.20 Electrophoresis was performed for 15 min at a field strength of 1 kV/cm. The initial concentration of silver in the organic phase Langmuir 2010, 26(2), 736–742
Bulavchenko and Popovetsky
Article
Figure 8. Dependence of the energy of interparticle interaction on the surface-surface distance for different systems in a decane-chloroform medium: Ag-Ag (A), C60-C60 (B), Ag-C60 (C), and Ag-SiO2 (D).
Figure 9. Enrichment factor as a function of the initial Ag concentration; solubilization capacity, 1 vol %. -3
-4
þ
varied from 10 to 10 mol/L (the reduction of Ag was performed in the system with a concentration of 10-3 mol/L; the lower concentrations were obtained by diluting the initial solution20). The determined concentration coefficients (the ratio of the silver concentration in the cathode concentrate to the initial silver concentration) were rather high (up to 104, Figure 9). The electrophoretic-concentration technique can be applied to determine metals in diluted aqueous solutions (in combination with liquid-liquid (micellar) extraction).32 It can also be used to obtain an electrophoretic hydrophobic concentrate of metal nanoparticles (g1 mol/L) as a source of production of nanomaterials. 3.5. Separation of Nanoparticles (Stable Systems). We performed electrophoresis on a model system consisting of nanoparticles with a high ζ potential (Ag) and a practically zero ζ potential (C60). The mixture decane þ chloroform (1:1) was used as a solvent. Silver nanoparticles were introduced into the fullerene solution as a cathode concentrate prepared by the above-described scheme of electrophoretic concentration. The initial concentrations of the mixture components were as follows: Ag, 1.5 10-3 mol/L and C60, 1.5 10-4 mol/L. After the electrophoretic separation (the conditions were the same as during the silver concentration), the fullerene particles remained in the solution in the interelectrode region and the silver nanoparticles were deposited on the cathode as a liquid concentrate (Figure 10). Thus, the fullerene concentration in the solution (32) Bulavchenko, A. I.; Arymbaeva, A. T.; Bulavchenko, O. A.; Tatarchuk, V. V.; Petrova, N. I. Russ. J. Phys. Chem. 2006, 80, 1980–1984.
Langmuir 2010, 26(2), 736–742
Figure 10. Absorption spectra of the initial mixture C60 þ Ag, cathode concentrate of Ag (with an 800-fold dilution), and fullerene solution from the interelectrode space after the separation, with decane-chloroform (1:1) used as a solvent.
Figure 11. TEM (JEM-2010) photomicrographs of Ag (small dark points)-SiO2 aggregated nanoparticles.
remained virtually unchanged, and the silver concentration decreased by more than 2 orders of magnitude. (The residual concentrations were estimated from the absorption spectra in Figure 10) This approach can be used for the separation of nanoparticles or the purification of nanoparticles from the starting reagents or the side products of the reaction.33 (33) Hu, H.; Yu, A.; Kim, E.; Zhao, B.; Itkis, M. E.; Bekyarova, E.; Haddon, R. S. J. Phys. Chem. B 2005, 109, 11520–11524.
DOI: 10.1021/la903583r
741
Article
3.6. Formation of Nanoparticle Aggregates through Heterocoagulation (Unstable System). We showed the possibility of the deposition of positively charged silver nanoparticles on negatively charged SiO2 (and Al2O3) particles through heterocoagulation in a decane-chloroform mixture (1:1). An electrophoretic concentrate with Ag nanoparticles was added to a SiO2 suspension on agitation. After the deposition of nanoparticles (deposition time ∼30 min), the precipitate was thoroughly washed with the used solvent several times; the organic phase was separated by centrifugation (10 min, 5000 rev/min). The micellar phase after deposition and after the washing solutions were used was colorless; the electron beam did not detach silver nanoparticles from the SiO2 surface during the observation and photography. This points to their rather strong fixation on the surface. The powder color varied from white to yellow and dark brown depending on the silver concentration, which indicates that the nanoparticles on oxide supports are not bound to each other by ohmic contacts (Figure 11). Such systems can be used for catalysis and sorption.
742 DOI: 10.1021/la903583r
Bulavchenko and Popovetsky
4. Conclusions A simple, accessible method has been developed to determine the electrophoretic mobility of metal nanoparticles and their poorly soluble compounds in micellar systems and coarse dispersions in solvents with a low dielectric permittivity (2-5). The method allows the determination of the electrophoretic mobility of nano- and microparticles over a wide range of sizes, including the smallest nanoparticles (g0.7 nm). The surface potential of the particles allowed the electrophoretic concentration of silver nanoparticles with high enrichment factors (up to 104) in aggregation-stable systems and the separation of nanoparticles with strongly differing charges. In unstable systems, the formation of electrostatically bound aggregates is possible. Acknowledgment. This work was supported by the Russian Foundation for Basic Research (grants 05-03-32308 and 09-0300511).
Langmuir 2010, 26(2), 736–742
