Letter pubs.acs.org/JPCL
Image Charge Effects on the Formation of Pickering Emulsions Hongzhi Wang, Virendra Singh,† and Sven Holger Behrens* School of Chemical & Biomolecular Engineering, Georgia Institute of Technology S Supporting Information *
ABSTRACT: Vigorous mixing of an aqueous particle dispersion with oil usually produces a particle-stabilized emulsion (a “Pickering emulsion”), the longevity of which depends on the particles’ wetting properties. A known exception occurs when particles fail to adsorb to the oil−water interface created during mixing because of a strong repulsion between charges on the particle surface and similar charges on the oil−water interface; in this case, no Pickering emulsion is formed. Here, we present experimental evidence that the rarely considered electrostatic image force can cause a much bigger hindrance to particle adsorption and prevent the formation of Pickering emulsions even when the particle interaction with the interface charge is attractive. A simple theoretical estimate confirms the observed magnitude of this effect and points at an important limitation of Pickering emulsification, a technology with widespread industrial applications and increasing popularity in materials research and development. SECTION: Glasses, Colloids, Polymers, and Soft Matter
P
negative surface charge above pH 2−3 possibly originating from the preferential adsorption of hydroxide ions,17−19 although this explanation remains controversial.20,21 If double layer forces dominated the electrostatic particle−oil interaction, then only anionic particles should be repelled from the oil− water interface and prevented from adsorbing and stabilizing Pickering emulsions.16 To test this dependence on particle charge, we investigated the emulsification properties of 1.9 μm polystyrene (PS) particles and 0.4 μm amidine-functionalized PS particles, chosen for their large and opposite maximum surface charge densities of −197 and +192 mC/m2.22 Figure 1A shows the zeta potential of these particles in aqueous NaCl solution as a function of pH for different fixed ionic strengths. The reported zeta potentials were calculated from experimental electrophoretic mobilities (obtained by phase analysis light scattering, Malvern Zetasizer Nano ZS90) using O’Brien and White’s method.23,24 These data confirm the expected sign of the particle charge and qualitative pH dependence, as well as the progressive screening with increasing ionic strength. At the highest salt content, particles aggregate in the course of the measurements, and thus, caution against a detailed interpretation of the associated results is warranted. For a rough idea of the particles’ wetting properties in a hexadecane−water emulsion, Figure 1B reports the macroscopic contact angle between a planar solid film, cast from particles dissolved in chloroform, and a drop of aqueous NaCl deposited on the solid and then surrounded by the oil. The contact angles are plotted as measured through the water phase; thus, the observed values between 132 and 142° indicate
article-stabilized emulsions, often referred to as Pickering emulsions, have been known for more than a century.1,2 They are encountered in oil recovery and have long been used in the cosmetics and food industries.3 Renewed interest in particle-decorated emulsion droplets has been sparked by novel uses in encapsulation and controlled release,4−6 most recently with the exploitation of stimulus-responsive particles such as emulsifiers7−11 and by the use of Pickering emulsion polymerization for the production of “armored” latex particles.12,13 Pickering emulsions are most commonly prepared by shaking or vigorously mixing an aqueous particle dispersion with an oil phase, which creates a large oil−water interfacial area and promotes the adsorption of particles onto this interface. For particles with nonextreme wetting behavior14 and a size above 10 nm, adsorption dramatically lowers the interfacial free energy, and the resulting adsorption strength has been associated directly with the achievable emulsion stability.15 Stabilization, however, requires that particles reach the interface in the first place, which should not be taken for granted.16 Particles in water tend to carry an electric surface charge, as do “bare” oil−water interfaces,17 allowing for an electrostatic barrier to particle adsorption. So far, theoretical descriptions of this barrier have focused on the electric double layer repulsion between a particle and a like-charged oil−water interface.16 Data presented in this Letter suggest that the additional, often neglected image charge repulsion can in fact dominate the particle−oil interaction and prevent particle adsorption and emulsification, even when the particle and oil are oppositely charged. The electric charge of colloidal particles in water typically stems from dissociable surface groups and can vary in sign and magnitude. For pristine oil droplets in water, by contrast, electrokinetic and electroacoustic experiments indicate a © 2012 American Chemical Society
Received: July 9, 2012 Accepted: October 2, 2012 Published: October 2, 2012 2986
dx.doi.org/10.1021/jz300909z | J. Phys. Chem. Lett. 2012, 3, 2986−2990
The Journal of Physical Chemistry Letters
Letter
Figure 1. (A) Zeta potential of PS−carboxyl (solid markers) and PS− amidine particles (open markers) in aqueous NaCl. (B) Contact angels of a film cast from these particles with a hexadecane−water interface (right). Data represent averages of four repeat measurements of four contact angels.
a rather hydrophobic surface. A trend of decreasing contact angle with increasing pH for the PS−carboxyl surface is apparent from Figure 1B, and there is an opposite trend for the PS−amidine surface, showing that an increase in the (absolute) surface charge density slightly reduces the hydrophobicity, as one might expect. In the emulsification tests central to our study, we mixed aqueous 1 wt % dispersions of the particles described above25 with an equal volume of hexadecane (Sigma-Aldrich, ReagentPlus grade, stained with oil-soluble red dye Sudan III) using a rotor-stator homogenizer (IKA Ultra-Turrax T10) for 30 s at 11000 rpm. Depending on the pH and salt (NaCl) content of the dispersion, we obtained either an oil-in-water (o/w) emulsion, a water-in-oil (w/o) emulsion, or no emulsion at all (Figure 2). In the cases where an emulsion with some short-term stability was formed, creaming (o/w) or sedimentation (w/o) occurred over time and revealed the emulsion type, which was confirmed by conductivity and drop tests. The observed Pickering emulsions differed widely in stability. In the cases with poor short-term stability, indicated by the shaded area in Figure 2B, droplet coalescence proceeded fast until only a few large drops remained stabilized in an otherwise phaseseparated system, whereas the most stable emulsions showed no significant coalescence for at least several weeks. We did not investigate in detail the long-term emulsion stability but focused instead on the question of whether a Pickering emulsion was ever formed. In the “No Emulsion” regime at high particle charge and weak screening (Figure 2, ■), mixing resulted in systems undergoing rapid coalescence, with no significant retardation by the presence of particles, until complete phase separation into an aqueous particle dispersion and a clear oil phase was achieved in a matter of seconds. We take this complete failure of particles to stabilize any droplets as an indication that particle adsorption at the oil− water interface is severely restricted. To test this notion further, we have prepared a dilute dispersion of our particles in heavy water, capped this dispersion with hexadecane, and observed the particles’ buoyancy-driven accumulation in or near the macroscopic interface. Subsequent replacement of the heavy water by regular water triggered the sedimentation and thereby the disappearance from the interfacial region of any nonadsorbed particles. Only under conditions of high particle charge and weak screening did the vast majority of particles initially accumulated at the interface turn out to be nonadsorbed. These experiments, which are described in detail in the Supporting Information, thus confirmed the hypothesized
Figure 2. Outcome of surfactant-free emulsification of hexadecane with aqueous dispersions of anionic (A) and cationic (B) particles. Solid symbols mark experiments, and open symbols mark theoretical estimates of whether particle adsorption to the interface is suppressed (□) or not (◇).
correlation between the previously described “failure to emulsify” and a strong suppression of particle adsorption. One might wonder whether this suppression stems from the particle interaction with the interface itself or from the interaction with particles already adsorbed at an interfacial coverage too low to stabilize droplets. If it was the repulsion by adsorbed particles that prevented further adsorption, then coalescing droplets should nonetheless reach good coverage, as illustrated in Figure 3. We therefore believe that the particles
Figure 3. Schematic of limited particle adsorption and droplet coalescence. Coalescence reduces the total interfacial area available to the adsorbed particles until the droplet coverage is sufficient for stabilization. 2987
dx.doi.org/10.1021/jz300909z | J. Phys. Chem. Lett. 2012, 3, 2986−2990
The Journal of Physical Chemistry Letters
Letter
are repelled by the interface itself. Moreover this repulsion likely has an electrostatic origin because its strength appears to depend on particle charge and screening. For the negatively charged carboxyl particles, the electric double layer repulsion from the similarly charged oil−water interface17 would be the obvious culprit.16 If this was the only force preventing Pickering emulsification, then the positively charged amidine particles, which are attracted to the oil by double layer forces, should encounter no such difficulty. Our experiments (Figure 2B) suggest otherwise. While emulsions are formed at high salt content, there is also a regime (shaded area in Figure 2) in which short-term stability is very poor and only a few large drops (of oil, water, or both) remain stabilized near the macrophase boundary, suggesting limited particle adsorption. More importantly, we again encounter a regime at high and weakly screened particle charge where particle adsorption to the liquid interface appears to be suppressed altogether.26 We propose that this suppression is caused by the repulsion of particles from their electrostatic “image” on the oil side of the liquid interface. It is known that an electric charge near an interface between dielectric media of different permittivity sets up a polarization field in which the charge experiences a force pointing toward the more polarizable medium.27 Near a large flat interface, the force on a charge q takes the same form as the electrostatic interaction with an “image charge” of magnitude qimage = q
ε1 − ε2 ε1 + ε2
max(FvdW + FEDL + Fimage) > Fmix
Details of these calculations are given in the Supporting Information. Figure 4 shows the result for the estimated forces
Figure 4. Force between a positively charged particle (with surface potential ψp = 78 mV) and a negatively charged interface (ψoil = −30 mV).
at a pH and ionic strength where experiments suggest that particle adsorption to the liquid interface is suppressed. Indeed, the height of the estimated force barrier (Figure 4) is 14 times larger than the hydrodynamic mixing force promoting adsorption. Similar theoretical estimates were carried out for other solution conditions where experimental data for the zeta potential of oil droplets in NaCl solution were available.17 Calculations suggesting the suppression of particle adsorption and Pickering emulsification are marked with open squares in Figure 2, and theoretical results favoring particle adsorption are marked with open diamonds. Given the crude approximations made in the theoretical estimates, their detailed agreement with experiments seems fortuitous, but it does support our proposition that image forces can have the right order of magnitude to impede Pickering emulsification. Consistently with this interpretation, we found that a significant increase in the hydrodynamic driving force, realized experimentally by increasing the rotor frequency almost 3-fold to 30 000 rpm, achieved emulsification in select cases (PS− carboxyl, pH 6, ionic strength 1 mM; PS−amidine, pH 5, ionic strength 1 mM) where weaker mixing failed to produce a Pickering emulsion. By contrast, simply increasing the mixing time by a factor of 2 or 3 did not change the observations reported in Figure 2. We also carried out emulsification experiments with 1-octanol (ACS reagent, Sigma-Aldrich), which has negligibly low water miscibility, like hexadecane, but a higher dielectric constant (10.3) and should therefore give rise to a weaker image force. In emulsification experiments analogous to the ones depicted in Figure 2A and B, but with octanol replacing hexadecane as the oil phase, we found that mixing always resulted in the formation of w/o Pickering emulsions regardless of the pH and ionic strength in the aqueous phase. While theoretical studies have long hinted at the importance of image forces for the interaction of particles with small ions28 and even with liquid interfaces,29 considerations of image forces in Pickering emulsions have so far focused on the image charge
(1)
located in the position of a mirror image across the interface. Here, ε1 is the dielectric constant of the medium hosting the real charge, and ε2 is that of the charge-free medium. For a charge in water (ε1 = 78) facing a flat surface of hexadecane (ε2 = 2), the image charge has the same sign and almost the same magnitude as the real charge, and just like the interaction with a second real charge, the charge−image interaction can be screened by salt ions. Similarly, a charged colloidal particle near a low-curvature oil−water interface is repelled by its image, but is this repulsion strong enough to prevent particle adsorption? In contrast to diffusive adsorption in a stagnant fluid, which involves thermal activation over any interaction energy barrier, adsorption under turbulent mixing (high Peclet number) is governed by the competition of the barrier force with the hydrodynamic force pushing the particle toward the interface.16 For a particle of radius a near a droplet of radius R ≫ a, this driving force of mixing can be estimated as16 Fmix ≈ a 2ρc ε 2/3R2/3
(3)
(2)
where ρc is the density of the continuous liquid and ε the rate of energy dissipation per unit mass. With ρc = 103 kg m−3 for water, ε ≈ 105 J kg−1 s−1 (typical for lab-scale rotor-stator mixers),16 and an estimated droplet radius of 20 μm, we obtain Fmix ≈ 1 nN for the carboxyl particles and 64 pN for the smaller amidine particles. This force has to be compared with the interaction force between the particle and the interface, which we write as the sum of the van der Waals force FvdW, the electric double layer force FEDL, and an image force Fimage evaluated like a double layer force with an image particle at twice the distance of the interface. We predict that particle adsorption is suppressed if and only if 2988
dx.doi.org/10.1021/jz300909z | J. Phys. Chem. Lett. 2012, 3, 2986−2990
The Journal of Physical Chemistry Letters
Letter
attraction for particles in the oil phase.30,31 To our best knowledge, this Letter reports the first experimental evidence of image charge repulsion preventing the formation of Pickering emulsions. We expect this very direct effect of particle charge on emulsification to be no less important for emulsion technology than that the widely acknowledged indirect effects via the wetting and pair interaction of interfacially adsorbed particles.
■
(9) Ngai, T.; Behrens, S. H.; Auweter, H. Novel Emulsions Stabilized by pH and Temperature Sensitive Microgels. Chem. Commun. 2005, 331−333. (10) San Miguel, A.; Scrimgeour, J.; Curtis, J. E.; Behrens, S. H. Smart Colloidosomes With a Dissolution Trigger. Soft Matter 2010, 6, 3163−3166. (11) Liu, T. T.; Seiffert, S.; Thiele, J.; Abate, A. R.; Weitz, D. A.; Richtering, W. Non-Coalescence of Oppositely Charged Droplets in pH-Sensitive Emulsions. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 384− 389. (12) Cauvin, S.; Colver, P. J.; Bon, S. A. F. Pickering Stabilized Miniemulsion Polymerization: Preparation of Clay Armored Latexes. Macromolecules 2005, 38, 7887−7889. (13) Fujii, S.; Okada, M.; Nishimura, T.; Maeda, H.; Sugimoto, T.; Hamasaki, H.; Furuzono, T.; Nakamura, Y. Hydroxyapatite-Armored Poly(epsilon-caprolactone) Microspheres and Hydroxyapatite Microcapsules Fabricated via a Pickering Emulsion Route. J. Colloid Interface Sci. 2012, 374, 1−8. (14) Nonextreme wetting refers to particles forming an equilibrium three-phase contact angle between 30 and 150° with the liquid interface. (15) Aveyard, R.; Binks, B. P.; Clint, J. H. Emulsions Stabilised Solely by Colloidal Particles. Adv. Colloid Interface Sci. 2003, 100, 503−546. (16) Tcholakova, S.; Denkov, N. D.; Lips, A. Comparison of Solid Particles, Globular Proteins and Surfactants as Emulsifiers. Phys. Chem. Chem. Phys. 2008, 10, 1608−1627. (17) Marinova, K. G.; Alargova, R. G.; Denkov, N. D.; Velev, O. D.; Petsev, D. N.; Ivanov, I. B.; Borwankar, R. P. Charging of Oil-Water Interfaces Due to Spontaneous Adsorption of Hydroxyl Ions. Langmuir 1996, 12, 2045−2051. (18) Zangi, R.; Engberts, J. B. F. N. Physisorption of Hydroxide Ions from Aqueous Solution to a Hydrophobic Surface. J. Am. Chem. Soc. 2005, 127, 2272−2276. (19) Creux, P.; Lachaise, J.; Graciaa, A.; Beattie, J. K.; Djerdjev, A. M. Strong Specific Hydroxide Ion Binding at the Pristine Oil/Water and Air/Water Interfaces. J. Phys. Chem. B 2009, 113, 14146−14150. (20) Knecht, V.; Levine, Z. A.; Vernier, P. T. Electrophoresis of Neutral Oil in Water. J. Colloid Interface Sci. 2010, 352, 223−231. (21) Vacha, R.; Rick, S. W.; Jungwirth, P.; de Beer, A. G. F.; de Aguiar, H. B.; Samson, J. S.; Roke, S. The Orientation and Charge of Water at the Hydrophobic Oil Droplet−Water Interface. J. Am. Chem. Soc. 2011, 133, 10204−10210. (22) Molecular Probes, product nr. C37278 and A37474. The cited maximum surface charge densities were provided by the producer and obtained via conductometric titration. (23) O’Brien, R. W.; White, L. R. Electrophoretic Mobility of a Spherical Colloidal Particle. J. Chem. Soc., Faraday Trans. 2 1978, 74, 1607−1626. (24) Mangelsdorf, C. S.; White, L. R. Effects of Stern-Layer Conductance on Electrokinetic Transport Properties of Colloidal Particles. J. Chem. Soc., Faraday Trans. 1990, 86, 2859−2870. (25) Changing the particle concentration by a factor of 2 in either direction did not change the qualitative outcome of emulsification experiments. (26) The stabilization of water−continuous 1:1 emulsions by hydrophobic particles is noteworthy too,15 but the discussion is beyond the scope of this Letter. (27) Israelachvili, J. N. Intermolecular and Surface Forces, 3rd ed.; Elsevier: Amsterdam, The Netherlands, 2011. (28) Messina, R. Image Charges in Spherical Geometry: Application to Colloidal Systems. J. Chem. Phys. 2002, 117, 11062−11074. (29) von Grünberg, H. H.; Mbamala, E. C. Charged Colloids Near Interfaces. J. Phys.: Condens. Matter 2001, 13, 4801−4834. (30) Danov, K. D.; Kralchevsky, P. A.; Ananthapadmanabhan, K. P.; Lips, A. Particle−Interface Interaction Across a Nonpolar Medium in Relation to the Production of Particle-Stabilized Emulsions. Langmuir 2006, 22, 106−115. (31) Leunissen, M. E.; van Blaaderen, A.; Hollingsworth, A. D.; Sullivan, M. T.; Chaikin, P. M. Electrostatics at the Oil−Water
ASSOCIATED CONTENT
S Supporting Information *
Calculation of the interaction force barrier and microscopy study of particle adsorption to a macroscopic oil−water interface. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Address: Georgia Institute of Technology, School of Chemical & Biomolecular Engineering, 311 Ferst Drive NW, Atlanta, GA 30332-0100, U.S.A. Fax: +1 404 894 2866. Tel: +1 404 894 3166. E-mail:
[email protected]. Present Address †
The George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, 771 Ferst Drive NW, Atlanta, GA 30332.
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We gratefully acknowledge financial support by the National Science Foundation (CBET-1134398) and fellowship support for H.W. by the Institute of Paper Science & Technology.
■
REFERENCES
(1) Ramsden, W. Separation of Solids in the Surface-Layers of Solutions and ‘Suspensions’ (Observations on Surface-Membranes, Bubbles, Emulsions, and Mechanical Coagulation). Preliminary Account. Proc. R. Soc. London 1903, 72, 156−164. (2) Pickering, S. U. Emulsions. J. Chem. Soc. 1907, 91, 2001−2021. (3) Hunter, T. N.; Pugh, R. J.; Franks, G. V.; Jameson, G. J. The Role of Particles in Stabilising Foams and Emulsions. Adv. Colloid Interface Sci. 2008, 137, 57−81. (4) Velev, O. D.; Furusawa, K.; Nagayama, K. Assembly of Latex Particles by Using Emulsion Droplets as Templates. 1. Microstructured Hollow Spheres. Langmuir 1996, 12, 2374−2384. (5) Dinsmore, A. D.; Hsu, M. F.; Nikolaides, M. G.; Marquez, M.; Bausch, A. R.; Weitz, D. A. Colloidosomes: Selectively Permeable Capsules Composed of Colloidal Particles. Science 2002, 298, 1006− 1009. (6) Shilpi, S.; Jain, A.; Gupta, Y.; Jain, S. K. Colloidosomes: An Emerging Vesicular System in Drug Delivery. Crit. Rev. Ther. Drug Carrier Syst. 2007, 24, 361−391. (7) San Miguel, A.; Behrens, S. H. Permeability Control in StimulusResponsive Colloidosomes. Soft Matter 2011, 7, 1948−1956. (8) Amalvy, J. I.; Armes, S. P.; Binks, B. P.; Rodrigues, J. A.; Unali, G. F. Use of Sterically-Stabilised Polystyrene Latex Particles as a pHResponsive Particulate Emulsifier to Prepare Surfactant-Free Oil-inWater Emulsions. Chem. Commun. 2003, 1826−1827. 2989
dx.doi.org/10.1021/jz300909z | J. Phys. Chem. Lett. 2012, 3, 2986−2990
The Journal of Physical Chemistry Letters
Letter
Interface, Stability, and Order in Emulsions and Colloids. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 2585−2590.
2990
dx.doi.org/10.1021/jz300909z | J. Phys. Chem. Lett. 2012, 3, 2986−2990
