pubs.acs.org/Langmuir © 2009 American Chemical Society
Effects of Surfactant Structure on the Phase Inversion of Emulsions Stabilized by Mixtures of Silica Nanoparticles and Cationic Surfactant Z.-G. Cui,*,† L.-L. Yang,† Y.-Z. Cui,† and B. P. Binks*,‡ †
School of Chemical and Material Engineering, Jiangnan University, 1800 Lihu Road, Wuxi, Jiangsu 214122, P. R. China and ‡Surfactant & Colloid Group, Department of Chemistry, University of Hull, Hull HU6 7RX, U.K. Received September 22, 2009. Revised Manuscript Received November 4, 2009
Silica nanoparticles without any surface modification are not surface active at the toluene-water interface due to their extreme hydrophilicity but can be surface activated in situ by adsorbing cationic surfactant from water. This work investigates the effects of the molecular structure of water-soluble cationic surfactant on the surface activation of the nanoparticles by emulsion characterization, adsorption and zeta potential measurements, dispersion stability experiments, and determination of relevant contact angles. The results show that an adsorbed cationic surfactant monolayer on particle surfaces is responsible for the wettability modification of the particles. In the presence of a trace amount of cationic surfactant, the hydrophobicity of the particles increases, leading to the formation of stable oil-in-water O/W(1) emulsions. At high surfactant concentration (>cmc) the particle surface is retransformed to hydrophilic due to doublelayer or hemimicelle formation, and the concentration of the free surfactant in the aqueous phase is high enough to stabilize emulsions alone. O/W(2) emulsions, probably costabilized by free surfactant and particles, are then formed. The monolayer adsorption seems to be charge-site dependent. Thus, using single-chain trimethylammonium bromide surfactants or a double-head gemini cationic surfactant, the hydrophobicity of the particles achieved is not sufficient to stabilize water-in-oil (W/O) emulsions, and no phase inversion is induced. However, using a double-chain cationic surfactant, the chain density on the particle surfaces endows them with a hydrophobicity high enough to stabilize W/O emulsions, and double phase inversion, O/W(1) f W/O f O/W(2), can then be achieved by increasing the surfactant concentration.
Introduction It is well-known that colloidal particles of suitable wettability can aggregate at liquid-liquid and liquid-vapor interfaces and thus behave as macroemulsion and foam stabilizers, respectively.1-5 The wettability of a particle at an oil-water interface, characterized by the contact angle θow of the particle with the interface measured through the water phase, can be thought of as the particle equivalent of the surfactant hydrophile-lipophile balance (HLB) number, which determines not only the surface activity of the particle but also the type of the emulsion stabilized.4 Particles with θow slightly below or above 90° display very high surface activity and can stabilize oil-in-water (O/W) or water-inoil (W/O) emulsions, respectively. The emulsions and foams formed are in general ultrastable due to the much greater adsorption free energy of the particles compared with the thermal energy kT.3-6 However, particles with very low or very high θow are not very surface active due to their low adsorption free energy. Moreover, using surface active nanoparticles as stabilizer, the phase inversion of aqueous foams has been achieved recently,7 a phenomenon not yet reported in the case of surfactant. *Corresponding authors. E-mail:
[email protected] (Z.-G.C.);
[email protected] (B.P.B.). (1) Ramsden, W. Proc. R. Soc. London 1903, 72, 156. (2) Pickering, S. U. J. Chem. Soc. 1907, 91, 2001. (3) Binks, B. P.; Horozov, T. S. In Colloidal Particles at Liquid Interfaces; Binks, B. P., Horozov, T. S., Eds.; Cambridge University Press: Cambridge, 2006; Chapter 1. (4) Binks, B. P. Curr. Opin. Colloid Interface Sci. 2002, 7, 21. (5) Aveyard, R.; Binks, B. P.; Clint, J. H. Adv. Colloid Interface Sci. 2003, 100-102, 503. (6) Lopetinsky, R. J. G.; Masliyah, J. H.; Xu, Z.-H. In Colloidal Particles at Liquid Interfaces; Binks, B. P., Horozov, T. S., Eds.; Cambridge University Press: Cambridge, 2006; Chapter 6. (7) Binks, B. P.; Murakami, R. Nat. Mater. 2006, 5, 865.
Langmuir 2010, 26(7), 4717–4724
On the other hand, however, various types of nanoparticle currently commercially available, e.g., silica, calcium carbonate, and carbon black, without any surface treatment are not surface active due to either their extreme hydrophilicity or hydrophobicity. A popular route to obtain surface active nanoparticles is to coat them homogeneously. Thus, for example, by surface silylation non-surface-active hydrophilic silica nanoparticles can be modified to have different wettability and thus act as either O/W or W/O emulsion stabilizers.4 The surface treatments, however, can greatly increase the cost of the particles. Another way to make surface active nanoparticles is to prepare “Janus” particles, which have an asymmetric surface chemistry similar to surfactant molecules, e.g., with one part of the surface hydrophilic and the other part hydrophobic.8-11 This route, however, is complicated, and most methods are not designed to produce large quantities. It is also well-known that the wettability of particles can be easily modified in situ in aqueous media via interaction with amphiphilic compounds.12-15 This presents another way to make surface active nanoparticles. In fact, it has been found that emulsion stabilization in the presence of colloidal particles was significantly (8) Binks, B. P.; Fletcher, P. D. I. Langmuir 2001, 17, 4708. (9) Perro, A.; Reculusa, S.; Ravaine, S.; Bourgeat-Lamic, E.; Duguet, E. J. Mater. Chem. 2005, 15, 3745. (10) Cayre, O.; Paunov, V. N.; Velev, O. D. J. Mater. Chem. 2003, 10, 2445. (11) Paunov, V. N.; Cayre, O. Adv. Mater. 2004, 16, 778. (12) Esumi, K.; Ueno, M. In Structure-Performance Relationships in Surfactant, Surfactant Science Series, Vol. 70; Marcel Dekker: New York, 1997; Chapter 11. (13) Grosse, I.; Estel, K. Colloid Polym. Sci. 2000, 278, 1000. (14) Tiberg, F.; Brinck, J.; Grant, L. Curr. Opin. Colloid Interface Sci. 2000, 4, 411. (15) Somasundaran, P.; Huang, L. Adv. Colloid Interface Sci. 2000, 88, 179. (16) Lucassen-Reynders, E. H.; van den Temple, M. J. Phys. Chem. 1963, 67, 731.
Published on Web 12/01/2009
DOI: 10.1021/la903589e
4717
Article
Cui et al.
improved by adding amphiphilic compounds to the systems.16-32 This surface activation of nanoparticles is comparatively less complicated and inexpensive and is therefore practically significant. In recent studies of the in situ surface activation of nanoparticles by interaction with amphiphilic compounds in water, it is noticed that the wettability/surface activity of the given nanoparticles depends not only on the type and concentration of the amphiphilic compounds but also on their molecular structure. For example, by adsorbing the single-chain cationic surfactant cetyltrimethylammonium bromide (CTAB) from aqueous solution, negatively charged hydrophilic silica nanoparticles can be made surface active and stabilize O/W emulsions,27,28 whereas by adsorbing dichain cationic surfactant, a novel double phase inversion, O/W f W/O f O/W, was observed with increasing surfactant concentration.29,32 However, this double phase inversion was also observed for emulsions stabilized by positively charged hydrophilic calcium carbonate nanoparticles after adsorption of the single-chain anionic surfactant sodium dodecyl sulfate (SDS).30 It seems that the wettability modification of the nanoparticles via interaction with amphiphilic compounds depends on both the particle surface properties and the structure of the surfactant. To reveal the influence of surfactant structure, Binks and Rodrigues32 recently examined the double phase inversion of emulsions stabilized by silica nanoparticles induced by a series of dichain cationic surfactants of different chain length. Here in this paper we study the in situ surface activation of hydrophilic silica nanoparticles and double emulsion phase inversion induced by four water-soluble cationic surfactants with either a single chain, a double chain, or two headgroups in order to understand more fully the factors governing the phase inversion.
Experimental Section Materials. The powdered sample of hydrophilic silica nanoparticles, HL-380, with a purity >99.8%, a primary particle diameter varying between 7 and 40 nm, and a BET surface area of 380 ( 30 m2/g, was provided by Wuxi Jinding Longhua Chemical Co. (China). CTAB of 99% purity was purchased from Sinopharm Chemical Reagent Co. Dodecyltrimethylammonium bromide (DTAB) of 99% purity, SDS of 99% purity, and Hyamine 1622 of 98% purity were purchased from Sigma. Didodecyldimethylammonium bromide (di-C12DMAB) of 99% purity was purchased from Xiamen Pioneer Technology Co. (China), received as an aqueous slurry containing 89 wt % di-C12DMAB and ∼10 wt % moisture and used without drying. The gemini cationic surfactant, trimethylene-di(tetradecacyloxyethyldimethylammonium bromide) (17) Tsugita, A.; Takemoto, S.; Mori, K.; Yoneya, T.; Otani, Y. J. Colloid Interface Sci. 1983, 95, 551. (18) Gelot, A.; Friesen, W.; Hamza, H. A. Colloids Surf. 1984, 12, 271. (19) Yan, Y.; Masliyah, J. H. Colloids Surf., A 1993, 95, 551. (20) Lagaly, G.; Reese, M.; Abend, S. Applied Clay Sci. 1999, 14, 83. (21) Zhai, X.; Efrima, S. J. Phys. Chem. 1996, 100, 11019. (22) Tambe, D. E.; Sharma, M. M. J. Colloid Interface Sci. 1993, 157, 244. (23) Midmore, B. R. Colloids Surf., A 1998, 132, 257. (24) Midmore, B. R. J. Colloid Interface Sci. 1999, 213, 352. (25) Midmore, B. R. Colloids Surf., A 1998, 145, 133. (26) Binks, B. P.; Desforges, A.; Duff, D. G. Langmuir 2007, 23, 1098. (27) Binks, B. P.; Rodrigues, J. A.; Frith, W. J. Langmuir 2007, 23, 3626. (28) Lan, Q.; Yang, F.; Zhang, S.; Liu, S.; Xu, J.; Sun, D. Colloids Surf., A 2007, 302, 126. (29) Binks, B. P.; Rodrigues, J. A. Angew. Chem., Int. Ed. 2007, 46, 5389. (30) Cui, Z.-G.; Shi, K.-Z.; Cui, Y.-Z.; Binks, B. P. Colloids Surf., A 2008, 329, 67. (31) Ravera, F.; Ferrari, M.; Liggieri, L.; Loglio, G.; Santini, E.; Zanobini, A. Colloids Surf., A 2008, 323, 99. (32) Binks, B. P.; Rodrigues, J. A. Colloids Surf., A 2009, 345, 195. (33) Wang, Z.-C.; Xu, H.-J.; Bao, X.-Y.; Xu, F.-P.; Pen, Q.-J. J. Southern Yangtze Univ. (Nat. Sci. Ed.) 2005, 4, 306.
4718 DOI: 10.1021/la903589e
Figure 1. Structure of the gemini cationic surfactant II-14-3. (II-14-3), with a purity of 95% was synthesized and purified by one of our researchers33 and has a structure as shown in Figure 1. These surfactants were all used as received. Toluene of 99.5% purity was purchased from Sinopharm Chemical Reagent Co. and columned through neutral alumina before use. All other chemicals for analysis were analytically pure and from various sources. Ultrapure water with a resistance of 18.2 MΩ cm and a pH of 6.74 at 25 °C was purchased from a local microelectronics factory, Wuxi Huapu Co. Ltd. (China).
Methods. a. Dispersing Silica Nanoparticles in Aqueous Solution. The silica powder was weighed into a glass bottle of dimensions 7.5 cm (h) by 2.5 cm (d) followed by adding pure water or aqueous surfactant solution. The particles were then dispersed using an ultrasound probe (JYD-650, Shanghai) of tip diameter 0.6 cm operating at an output of 50 W for 1 min.
b. Measurement of Zeta Potential of Silica Nanoparticle Dispersions. Silica nanoparticles (1 wt %) were dispersed in water
of different pH, and the dispersions were left overnight at 25 °C. The pH was adjusted by adding HCl or NaOH. The zeta potentials were measured using a Malvern Zetasizer 2000 instrument at 25 °C. The zeta potentials were also measured in particle dispersions (1 wt %) as a function of added surfactant. c. Measurement of Surface Tension. The surface tension of aqueous surfactant solutions was measured using the du No€ uy ring method. A platinum ring (radius 0.955 cm) was hung on a hook at the bottom of an electronic balance of 0.1 mg sensitivity. The measuring cup was placed in a glass jacket on a perpendicularly moveable platform. The jacket was circulated by water of 25 °C from a thermostat. The ring was immersed in the solution by raising the platform and was then pulled out by lowering the platform slowly, and the maximum force required to break the liquid meniscus was recorded. The surface tension was calculated by multiplying the measured maximum force F (in mN/m) by the appropriate correction factor. Using this apparatus, the surface tensions of pure water and n-decane at 25 °C were 71.91 and 23.41 mN/m, respectively, with errors less than 0.2%, and compared well with values reported in the literature.34 d. Preparation and Characterization of Emulsions. In this study the volume ratio of water to oil in emulsions was always kept at 1:1. A certain mass of silica nanoparticles was initially ultrasonically dispersed in 7 cm3 of surfactant solution of different concentration followed by addition of 7 cm3 of oil, and the mixture was left overnight and then homogenized at 5000 rpm for 2 min using a XHF-D homogenizer (Ningbo Xinzhi) fitted with a dispersing tool of outer diameter equal to 1.4 cm. Exactly the same results were found when homogenization was performed immediately after mixing particles with surfactant. In the case of emulsions made with particles alone, homogenization was executed immediately after dispersing the particles in pure water. The (34) Cui, Z.-G.; Binks, B. P.; Clint, J. H. Langmuir 2005, 21, 8319.
Langmuir 2010, 26(7), 4717–4724
Cui et al.
Article
particle concentrations are expressed as weight percentage relative to the water phase. The emulsion type was identified by the drop test. After homogenization, a drop of emulsion was added to a small volume of the oil and aqueous phase separately. An emulsion which dispersed in the aqueous phase but not in the oil phase was assessed as water continuous (O/W) and vice versa. The emulsion stability was assessed by determining the water or oil volume separated as a function of time. Microscope images of the emulsion droplets were obtained using a Motic BA400 microscope system (Xiamen Motic Co.). Emulsion droplets (usually diluted with their continuous phase) were placed either in a wedge-shaped gap formed by a coverslip resting on a microscope slide with one edge propped up by a second coverslip or directly on a slide when the droplet size was large.
e. Measurement of Adsorption Isotherm of Surfactant on Silica Nanoparticles in Water. 0.15 g of silica nanoparticles was weighed into a series of glass vessels of dimensions 7.5 cm (h) by 2.5 cm (d), followed by adding 15 cm3 of aqueous surfactant solution of different concentration. The particles were dispersed using an ultrasound probe for 1 min at 50 W output, and the dispersions were put in a thermostat at 25 ( 0.5 °C for 24 h. After separating the suspended particles by centrifugation (5000 rpm for 30 min), the equilibrium concentration of cationic surfactant in the supernatant was measured (below), and the adsorbed amount of surfactant on silica nanoparticles, Γ, was calculated by the equation Γ ¼
VðC0 -CÞ ðmmol g -1 Þ w
ð1Þ
where V is the volume (cm3) of the solution, C0 and C are the initial and equilibrium concentrations (M) of cationic surfactant in the dispersion, respectively, and w is the weight (g) of the particles. When C > 0.6 mM, the surfactant concentration was determined using a two-phase titration method using SDS as titrant.35 When C < 0.6 mM, the judgment of the end point becomes difficult, and a spectrophotometric method was then used as described below. In a two-phase system composed of bromocresol green as indicator, chloroform as oil phase, and a mixed aqueous solution of cationic surfactant (to be determined), isopropanol, and Na3PO4-Na2HPO4 buffer solution, a blue-green complex soluble in chloroform can be formed by the cationic surfactant and the indicator, and the colored chloroform solution shows a maximum absorbance at 630 nm. In the case of fixed total volume and volume ratio of the two phases, the absorbance increases linearly with the concentration of the cationic surfactant in the aqueous phase. Thus, a linear working equation can be obtained and used for quantitatively determining the trace cationic surfactant in an aqueous solution. In this study the two-phase system was composed of 10 cm3 chloroform, 2.5 cm3 isopropanol, 5 cm3 indicator buffer solution (0.025 g bromocresol green dissolved in 5 cm3 isopropanol plus 25 g NaCl, 10 g Na3PO4 3 12H2O, and 10 g Na2HPO4 3 12H2O dissolved in 400 cm3 water and diluted to a total of 500 cm3), and 50 cm3 aqueous solution of cationic surfactant to be determined. The absorbance of the chloroform phase at 630 nm was measured using a 721 spectrophotometer (model 16C14, Shanghai Precision Scientific Instrument Co.). The linear relationships between absorbance and concentration for DTAB, CTAB, and di-C12DMAB are shown in Figure 2. For II-14-3, the adsorption at very low concentrations was not easy to determine.
f. Measurement of Contact Angles at Oil (Air)-WaterGlass Interface. The contact angles of aqueous surfactant solution on glass in air and under oil were measured using an optical drop shape analyzer (DropMeter-A-100, Ningbo Haishu Maishi (35) Reid, V. W.; Longman, G. F.; Heinerth, E. Tenside 1967, 4, 292.
Langmuir 2010, 26(7), 4717–4724
Figure 2. Linear relationship between absorbance of chloroform phase and initial aqueous surfactant concentration for three surfactants.
Figure 3. SEM image of HL-380 silica nanoparticles dispersed in pure water. Scientific Test Co.). The analyzer was placed in a transparent plastic box in which the temperature was controlled at 25.0 ( 0.5 °C using an air-therm heater (Air-Thermz-ATX, World Precision Instruments). For contact angles in air, a sessile drop of 5 10-3 cm3 was released from a needle of 0.5 mm diameter on a glass slide newly unpacked without any treatment. For contact angles under oil, the slides were cut to small pieces of 2.5 1 cm and were washed with 30% NaOH solution followed by rinsing with pure water and dried in air. A small piece was put in a glass box of 3.5 (L) 1.5 (W) 3.5 (H) cm, and a sessile drop of aqueous phase of 5 10-3 cm3 was released, followed by adding the oil phase using a pipet along the wall of the box until the sessile drop was totally immersed. The image of the sessile drop was recorded, and the contact angle was measured using software. In cases where the measurement using software was inappropriate, the angle was measured directly using a protractor. The angles were measured through the water phase. All experiments were carried out at room temperature (22 ( 2 °C) unless specified.
Results and Discussion a. Emulsions Stabilized by Silica Nanoparticles Alone. According to the supplier, the silica nanoparticles have a primary diameter of between 7 and 40 nm and a BET surface area of 380 m2 g-1. The SEM image from an aqueous dispersion in Figure 3 shows that the primary particle size is in general less than 50 nm, and an actual specific surface area of 345 m2 g-1 was measured using a specific area and pore size analyzer (ST-2000, Beijing Analytical Instrument Co.), in good agreement with that specified. The silica nanoparticles are strongly hydrophilic and can be easily dispersed in water. Figure 4 shows the zeta potential of the particles in water of different pH. It is seen that the particles have DOI: 10.1021/la903589e
4719
Article
Cui et al.
Figure 4. Zeta potentials of 1 wt % silica nanoparticles dispersed in water of different pH. Figure 6. Surface tension-aqueous concentration curves for di-C12DMAB and gemini II-14-3 surfactants at 25 °C.
Figure 5. Photographs of vessels containing toluene-in-water emulsions stabilized solely by silica nanoparticles at different concentrations; from left to right: 0.1, 0.5, 1.0, 1.5, 2.0, and 3.0 wt %, taken (a) 1 h and (b) 1 week after preparation, and (c) an optical micrograph of the emulsion containing 1.0 wt % particles.
an isoelectric point around pH = 3 as is common for most silicas. At pH > 3.2 the particles are negatively charged, and the magnitude of the zeta potential increases with increasing pH. Using toluene as oil and silica nanoparticles as the sole emulsifier, the emulsions are not stable as expected. Figure 5 shows the appearance of the vessels at different particle concentrations and a micrograph of oil droplets at 1 wt % silica. It is seen that after homogenization (a) some O/W emulsions of very large droplet size are formed which quickly cream, and the stability of the emulsions at intermediate particle concentration seems to be the best. After 1 week (b), there is little change except for the lowest particle concentration (0.1 wt %) where complete phase separation occurred. Although the emulsion phase in the other vessels appears to be stable, coalesced oil can be observed from the top of the vessels, and the volume of oil phase increases with particle concentration. The particles are mostly dispersed in the water phase as indicated by the turbidity of the lower layers. All this indicates that the particles are not surface active as a result of 4720 DOI: 10.1021/la903589e
Figure 7. Photographs of toluene-in-water emulsions stabilized solely by cationic surfactant at different initial concentrations, taken 1 week after preparation. (a) DTAB (from left to right): 0.3, 1.0, 3.0, 10, 20, 30 mM. (b) CTAB (from left to right): 0.1, 0.3, 0.6, 1.0, 2.0, 3.0 mM. (c) di-C12DMAB (from left to right): 0.1, 0.3, 0.6, 1.0, 3.0, 10 mM. (d) II-14-3 (from left to right): 0.01, 0.03, 0.06, 0.1, 0.2, 0.3 mM.
their extreme hydrophilicity. The increase in coalescence at high particle concentration may be due to a gradual aggregation of the particles in bulk, with such aggregates being more easily detached from the interface than discrete particles under conditions of low surface activity as here. Langmuir 2010, 26(7), 4717–4724
Cui et al.
Article
Table 1. cmc, Cmin, Cmin(eq), Cmin(eq)/cmc, and Copt for the Four Cationic Surfactants surfactant
DTAB
cmc/mM 16 10 Cmin/mM 9.02 Cmin(eq)/mM 0.56 Cmin(eq)/cmc 0.3 Copt/mM a More correctly cvc.37
CTAB 0.92 2 0.58 0.63 0.3-1.0
di-C12DMAB 0.03 1 0.31 10.3 0.3
a
II-14-3 0.06 0.2 0.088 1.47 0.3-0.6
b. Emulsions Stabilized by Cationic Surfactant Alone. The four cationic surfactants, although with different molecular structure and total hydrocarbon chain length, are all watersoluble. Among them DTAB and CTAB are frequently used in research, whereas di-C12DMAB and II-14-3 are less popular. Figure 6 shows the air-water surface tension, γ, as a function of surfactant concentration for the latter two surfactants, from which the critical micelle concentrations (cmc) are found from the break points to be 0.03 and 0.06 mM, respectively, much lower than those of CTAB (0.92 mM) and DTAB (16 mM).36 For diC12DMAB, the surface tension-concentration relationship is in good agreement with that of a sample from Aldrich with a purity of 98%, and the critical concentration obtained from the break of the curve is really a critical vesicle concentration (cvc) according to Kunitake and Okahata;37 the value we obtain is in good agreement with the value (0.032 mM) reported in the literature.38 Using these cationic surfactants as sole emulsifiers, the emulsions formed are all O/W, but they are unstable to coalescence at low surfactant concentration, as shown in Figure 7. At higher concentrations, although emulsions cream they remain stable to coalescence. In other words, there exists a minimum initial concentration, Cmin, required to stabilize an emulsion to coalescence for each surfactant. The equilibrium concentration at Cmin, denoted Cmin(eq), was measured for the four surfactants, which together with the Cmin and cmc values, are listed in Table 1. It is seen that DTAB, CTAB, and II-14-3 have a Cmin(eq) slightly lower or higher than their cmc’s, whereas di-C12DMAB has a Cmin(eq) much higher than its cmc, probably due to the partitioning of surfactant monomer into the oil phase. c. Emulsions Stabilized by Mixtures of Silica Nanoparticles and Cationic Surfactant. Upon addition of trace amounts of cationic surfactant (C < Cmin), the stability of the emulsions, denoted as O/W(1), containing 1 wt % of silica nanoparticles increases, as shown in Figure 8. For silica/DTAB mixtures (a), the creaming of the emulsions is initially inhibited by increasing the surfactant concentration, and an optimum initial DTAB concentration (Copt) of ca. 0.3 mM is required. However, in the presence of high concentrations of surfactant, especially when C > Cmin., creaming appears again and increases with increasing surfactant concentration. At C < Cmin, with increasing surfactant concentration the lowering of the oil-water interfacial tension results in a reduction of droplet size and an increase of emulsion viscosity, which lead to an increase in the stability to creaming. At high surfactant concentration (C > Cmin), the emulsion droplet size is found to be similar to that in emulsions stabilized solely by surfactant and the emulsions change gradually from gel-like to fluid. The instability to creaming is probably due to both the decrease in continuous phase viscosity and the micelle (36) Rosen, M. J. Surfactant and Interfacial Phenomena, 2nd ed.; Wiley: New York, 1989; pp 125-126. (37) Kunitake, T.; Okahata, Y. J. Am. Chem. Soc. 1977, 99, 3860. (38) Lu, S.; Kunjappu, J. K.; Somasundaran, P.; Zhang, L. Colloids Surf., A 2008, 324, 65. (39) Aronson, M. P. Langmuir 1989, 5, 494.
Langmuir 2010, 26(7), 4717–4724
Figure 8. Photographs of toluene-water emulsions stabilized by 1 wt % silica nanoparticles plus cationic surfactant at different initial concentrations, taken 1 week after preparation. (a) O/W emulsions stabilized by 1 wt % silica plus DTAB (from left to right): 0.01, 0.03, 0.1, 0.3, 1.0, 3.0, 10, 20, 30 mM. (b) O/W emulsions stabilized by 1 wt % silica plus CTAB (from left to right): 0.03, 0.06, 0.1, 0.3, 0.6, 1.0, 3.0, 10, 100 mM. (c) Emulsions stabilized by1 wt % silica plus di-C12DMAB (from left to right): 0.03, 0.1, 0.3, 1.0, 3.0 (O/W), 6.0 (unstable), 10 (W/O), 30 and 60 (O/W) mM. (d) O/W emulsions stabilized by 1 wt % silica plus II-14-3 (from left to right): 0.01, 0.03, 0.06, 0.1, 0.2, 0.3, 0.6, 1.0, 2.0 mM.
depletion between droplets leading to their mutual attraction.39,40 However, neither coalescence of oil droplets nor phase inversion of the emulsions was observed. Similar phenomena were observed for silica/CTAB mixtures (b), where the optimum initial concentration of CTAB to prevent creaming is between 0.3 and 1 mM. Different phenomena, however, were observed for silica/ di-C12DMAB mixtures (c), where double phase inversion was evidenced. At C < Cmin, O/W(1) emulsions are formed, and the stability of the emulsions increases with increasing surfactant concentration until an optimal initial concentration of 0.3 mM. A further increase in surfactant concentration leads to a reduction in emulsion stability, and at an initial concentration of 10 mM, the emulsion inverts to water-in-toluene, W/O, which is retransformed to a toluene-in-water type, denoted as O/W(2), when the initial concentration of di-C12DMAB is g30 mM. Such a double phase inversion, however, was not observed for silica/ II-14-3 gemini surfactant mixtures (d). Instead, the behavior is reminiscent of that for DTAB and CTAB systems, with the optimal initial concentration between 0.3 and 0.6 mM. Optical micrographs of some of these emulsions are shown in Figure 9. d. Aqueous Dispersions of Silica Nanoparticle/Cationic Surfactant Mixtures. It is believed that in the presence of trace amounts of cationic surfactant (C < Cmin) emulsions are mainly stabilized by the partially coated silica nanoparticles, which are (40) Bibette, J.; Roux, D.; Nallet, F. Phys. Rev. Lett. 1990, 65, 2470.
DOI: 10.1021/la903589e
4721
Article
Cui et al.
Figure 10. Zeta potentials of 1 wt % silica nanoparticles dispersed in aqueous solutions of various cationic surfactant.
Figure 9. Optical micrographs of emulsions stabilized by 1 wt % silica nanoparticles plus cationic surfactant at different concentrations, taken 24 h after preparation.
surface activated in situ via the interaction with surfactant in water. Since the silica nanoparticles are negatively charged in water, they adsorb cationic surfactant molecules via a Coulombic interaction, and a monolayer of cationic surfactant molecules with head-on configuration is formed on particle surfaces. Since the negative charges are neutralized by the positively charged surfactant headgroup, the particle surfaces become partly covered by the hydrocarbon chain of the adsorbed surfactant molecules. As a result, the hydrophilicity of the silica nanoparticles decreases; i.e., the wettability of the particles is modified from strongly hydrophilic to partially hydrophobic, encouraging their adsorption at the oil-water interface of emulsion drops. This scenario is supported by measurement of particle zeta potentials, dispersion stability, and adsorption isotherms, shown in Figures 10-12, respectively. Figure 10 shows that the sign of the zeta potential of the silica nanoparticles dispersed in cationic surfactant solutions changes from negative to positive, and the positive values increase with increasing surfactant concentration, implying adsorption of the cationic surfactant molecules on particle surfaces. Although the cmc’s of the four cationic surfactants are quite different, the concentrations at which the zeta potential of the silica particles tends to zero is in a narrow range, from 0.06 to 0.2 mM, indicating that the initial adsorption of the cationic surfactant molecules to the particles surface is via electrostatic interaction, as reported by Binks and Rodrigues.32 In Figure 11 it can be seen that the silica dispersions progress from being stable to flocculated (with sedimentation) to restabilized as a function of the concentration of added surfactant. The initial dispersion of the particles is due to the high negative charge density on their surfaces. Following their neutralization by adsorbed cationic surfactant molecules, the zeta potential of the 4722 DOI: 10.1021/la903589e
Figure 11. Photographs of 1 wt % silica nanoparticles dispersed in aqueous surfactant solutions of different initial concentrations, taken 1 week after preparation. (a) DTAB (from left to right): 0.01, 0.03, 0.1, 0.3, 1.0, 3.0, 10, 20, 30 mM. (b) CTAB (from left to right): 0.01, 0.03, 0.06, 0.1, 0.3, 0.6, 1.0, 3.0 mM. (c) di-C12DMAB (from left to right): 0.01, 0.03, 0.1, 0.3, 1.0, 3.0, 6.0, 10, 20, 30 mM. (d) II-14-3 (from left to right): 0.01, 0.02, 0.03, 0.1, 0.2, 0.3, 1.0, 2.0, 3.0 mM.
particles falls to zero and flocculation occurs. Comparing Figure 10 with Figure 11, it is found that the flocculation generally coincides with zero or low zeta potentials. For the three singlehead cationic surfactants, flocculation starts at between ca. 0.06 and 0.1 mM, but for the double-head cationic surfactant (II-14-3), flocculation starts at a lower concentration, 0.02 mM, reflecting that the adsorption is driven by electrostatic interaction, since each II-14-3 molecule carries two positive charges and is more efficient in neutralizing the negative charges on particle surfaces. However, upon increasing the surfactant concentration further, although the zeta potentials become more positive, flocculation still occurs. This suggests that the particle surface becomes Langmuir 2010, 26(7), 4717–4724
Cui et al.
progressively covered by a monolayer of hydrocarbon chains, and it is the chain-chain interaction between neighboring particles that results in flocculation. Since CTAB molecules have longer hydrocarbon chains than DTAB molecules, di-C12DMAB molecules contain two dodecyl chains, and one II-14-3 molecule contains two chains and two heads, the chain-chain interaction follows the order di-C12DMAB > CTAB > II-14-3 > DTAB, which is well reflected by the photographs in Figure 11. The redispersion of the systems is due to the formation of a second adsorbed layer of surfactant on particle surfaces now exposing positively charged headgroups to the aqueous phase,27,28 rendering particles hydrophilic again. Evidence for surfactant adsorption is provided by the adsorption isotherms displayed in Figure 12. e. Adsorption of Cationic Surfactants on Silica Nanoparticles in Water. As mentioned in the Introduction, the type and stability of Pickering emulsions depend significantly on the wettability of the particles, usually characterized by the contact angle, θow, of the particles at the oil-water interface. In the presence of trace amounts of cationic surfactant, adsorption of the surfactant neutralizes the negative charge on particle surfaces, rendering particles more hydrophobic. The emulsion stability thus increases with increasing surfactant concentration until Copt, where the particle wettability is optimum to stabilize O/W(1) emulsions. For the four cationic surfactants, Copt (as initial concentrations) is around 0.3 mM, much smaller than Cmin for DTAB, CTAB, and di-C12DMAB but close to Cmin for II-14-3. Since almost no water phase is separated at Copt, it is difficult to measure the corresponding equilibrium concentration in the aqueous phase. But considering the adsorption of surfactant at the oil-water interface and its potential partitioning into the oil phase, the equilibrium concentrations in the aqueous phase of the emulsion systems must be lower than that in aqueous dispersions for a given initial concentration. It is thus reasonable to predict that Copt corresponds to the condition of zero potential of the particles. The in situ modification of particle wettability by the adsorption of single-chain cationic surfactants such as DTAB and CTAB, however, is not sufficient to induce emulsion phase inversion. From the adsorption data, it is found that the adsorbed amount of DTAB and CTAB at Cmin(eq) are 4.13 10-2 and 2.66 10-2 mmol/g, respectively, corresponding to a chain area of 13.87 and 21.54 nm2. In previous work,30 we established the double phase inversion of emulsions stabilized by positively charged calcium carbonate nanoparticles via adsorption of anionic SDS molecules, where the chain area of the adsorbed SDS on particle surfaces corresponding to O/W(1) f (W/O) phase inversion was much lower at 7.8 nm2. The low coverage of hydrocarbon chains on the silica nanoparticles following adsorption of a single chain cationic surfactant does not raise the hydrophobicity of the particles sufficiently to induce phase inversion. On the other hand, using the double-chain surfactant di-C12DMAB, we found that the adsorbed amount at its Cmin(eq) is 4.00 10-2 mmol/g, corresponding to a chain area of 7.16 nm2, which is sufficient to induce O/W(1) f W/O phase inversion. Binks et al.27 and Lan et al.28 have recently reported that the adsorption of CTAB on silica nanoparticles cannot induce O/W(1) f W/O phase inversion. In contrast, Binks and Rodrigues29,32 achieved O/W(1) f W/O inversion by using a series of double-chain cationic surfactants, di-CnDMAB (n = 8-12), where it is shown that the dichain surfactant molecules pack more closely on particle surfaces than the corresponding single-chain surfactant of the same chain length at saturated adsorption. Similar adsorption trends have been reported by Lu et al.38 Langmuir 2010, 26(7), 4717–4724
Article
Figure 12. Adsorption isotherms of cationic surfactants on silica nanoparticles in water at 25 °C.
Figure 13. (a) Contact angles of drops of aqueous surfactant solution on hydrophilic glass slide in air and (b) under toluene versus initial surfactant concentration.
It is interesting to note that when the gemini surfactant II-14-3 is used, the adsorbed amount at its Cmin(eq) is 1.27 10-2 mmol/g, much smaller than that of the other three surfactants, corresponding to a chain area of 22.56 nm2. Such a chain area similar to that of CTAB is of course not sufficient to induce phase inversion. Figure 12 indicates that the adsorption of the gemini surfactant II-14-3 is in general similar to that of CTAB, which is in agreement with that obtained by Sakai et al.,41 who showed that the adsorbed amount at saturation of C12TAB on montmorillonite clay surfaces was the same as the gemini surfactant 12-2-12. The quantity Cmin is a useful and important parameter since below it the emulsions, denoted as O/W(1), are stabilized by particles, whereas beyond it the emulsions can be stabilized solely by surfactant. In the case of surfactant concentrations beyond Cmin, it is reasonable to expect that the emulsions, denoted as O/W(2), are costabilized by particles and surfactant molecules. However, with a further increase in surfactant concentration, a (41) Sakai, K.; Nakajima, E.; Takamatsu, Y.; Sharma, S. C.; Torigoe, K.; Yoshimura, T.; Esumi, K.; Sakai, H.; Abe, M. J. Oleo Sci. 2008, 57, 423.
DOI: 10.1021/la903589e
4723
Article
surfactant bilayer forms on particle surfaces transforming them to hydrophilic again. This, along with the oil-water interfacial tension being reduced,28 may inhibit the adsorption of the particles to the emulsion drop interfaces, thus reducing the stability of the O/W(2) emulsions. f. Contact Angle of Aqueous Phase on Glass Slide in Air and Under Oil. It is well-known that the contact angle, θow, of the particles at the oil-water interface is crucial in determining the type and stability of Pickering emulsions. The in situ modification of the silica particle surface by adsorption of cationic surfactant can after all be reflected by changes in θow. Since the direct measurement of θow for nanoparticles is difficult, we measured the contact angle of aqueous surfactant solutions on negatively charged glass slides in air and under oil. The fresh slides without any treatment are partially hydrophilic, as reflected by a contact angle of 37.4 ( 0.2° for pure water in air. Figure 13a shows how the equilibrium contact angle of the surfactant solutions in air varies with concentration for three surfactants of different structure. It is seen that the contact angles increase initially and then decrease with increasing surfactant concentration, reflecting the transition from hydrophilic to hydrophobic to hydrophilic, accompanying monolayer adsorption followed by bilayer formation. The maximum contact angles for di-C12DMAB, II-14-3, and CTAB are 76.8 ( 1.6°, 66.2 ( 1.3°, and 57.1 ( 1.4°, respectively. As expected, di-C12 DMAB exhibits the highest maximum angle, consistent with its highest hydrocarbon density in a monolayer and strongest chain-chain interactions. When the as received slides were treated with 30% NaOH solution and rinsed in pure water, they become more hydrophilic as reflected in a lower contact angle of 28.2 ( 3.0° for pure water in air. Figure 13b displays the equilibrium contact angles of the surfactant solutions on the treated glass slide under toluene, where the aqueous drops touched the glass slide in air first. The same maximum appears with CTAB and II-14-3
4724 DOI: 10.1021/la903589e
Cui et al.
reaching 74.6 ( 2.8° and 96.2 ( 13.1°, respectively. However, di-C12DMAB gives a maximum of 120.1 ( 16.9°, much higher than 90°, and is therefore able to induce emulsion inversion from O/W(1) to W/O. It will be noticed that the surfactant concentration for maximum contact angle (10-4 M) is much lower than that required to invert O/W(1) to W/O emulsions (between 3 and 10 10-3 M) for di-C12DMAB. This is partly due to differences between silica particle surfaces and the surfaces of glass slides but mainly due to the significant loss of surfactant to both drop and particle surfaces of high total surface area in the emulsion experiments compared to minimal loss in the contact angle determinations.
Conclusions 1. At pH ≈ 6.7, unmodified silica nanoparticles are not good emulsion stabilizers of water-toluene mixtures due to their extreme hydrophilicity. The nanoparticles, however, can be surface activated in situ by adsorption of cationic surfactant from water. 2. The formation of a monolayer of cationic surfactant molecules on particle surfaces renders them more hydrophobic. 3. Using single-chain cationic surfactants such as DTAB or CTAB or the gemini surfactant II-14-3, the increase of the particle hydrophobicity by monolayer adsorption leads to the formation of stable O/W(1) emulsions but is not sufficient to induce phase inversion to W/O emulsions. 4. Using the double-chain cationic surfactant di-C12DMAB, the adsorption density of hydrocarbon chains is increased, and the subsequent hydrophobicity of the particles is high enough to induce O/W(1) f W/O phase inversion. 5. For di-C12DMAB, at high surfactant concentration (.cmc) particle surfaces are retransformed to hydrophilic due to surfactant bilayer formation, and the concentration of free surfactant in the aqueous phase is high enough to stabilize O/W(2) emulsions alone.
Langmuir 2010, 26(7), 4717–4724
