pH-Responsive Aqueous Foams Stabilized by Ionizable Latex Particles

Langmuir 2007, 23, 8691-8694

8691

pH-Responsive Aqueous Foams Stabilized by Ionizable Latex Particles Bernard P. Binks,*,† Ryo Murakami,† Steven P. Armes,*,‡ Syuji Fujii,§ and Andreas Schmid‡ Surfactant & Colloid Group, Department of Chemistry, UniVersity of Hull, Hull HU6 7RX, U.K., Department of Chemistry, UniVersity of Sheffield, Sheffield S3 7HF, U.K., and Department of Applied Chemistry, Osaka Institute of Technology, 5-16-1 Ohmiya, Asahi-ku, Osaka 535-8585, Japan ReceiVed February 14, 2007. In Final Form: July 3, 2007 We have designed a type of colloidal particle whose surface characteristics are sensitive to the pH of the aqueous phase in which they are dispersed. Particles of polystyrene latex stabilized by poly(acrylic acid) can act as stabilizers of aqueous foams by adsorbing at the air-water surface. Foams can be prepared and stabilized only at pH values below the isoelectric point where particles are either uncharged and flocculated or acquire a positive charge. At high pH where particles are anionic, no foam forms. This influence of pH on foamability and stability applies to both pHdependent and pH-responsive systems.

Introduction Foams are of widespread importance and are used as intermediates and end products in a diverse array of industrial sectors, including mineral separation by ore flotation, food manufacturing, cosmetic formulations, and personal care products, as well as in the synthesis of porous advanced materials.1-4 It is well known that ionic and nonionic surfactants or polymeric stabilizers (including proteins) can stabilize foams,3 but it has also long been recognized that foams can be stabilized solely by colloidal particles.5 In most of the foams studied to date, solid particles have been used in combination with surfactants.2 It is only recently, however, that their precise role has been elucidated in surfactant-free systems.6-10 The adsorption of particles at an air-water surface depends critically on the hydrophobicity of the particle, which can be quantified by the contact angle, θ (measured through water).1 The angle θ increases with the hydrophobicity of the particle. It has been shown that if the particles are adsorbed they are strongly held at the fluid interface. The energy ∆G required to remove a spherical particle of radius r from an air-water surface is at a maximum when θ ) 90°. For the particles used here for which r ) 0.4 µm, ∆G is several orders of magnitude greater than the thermal energy. Hence, the particles are effectively irreversibly bound, in marked contrast to surfactant molecules that adsorb and desorb reversibly, and foams and emulsions * To whom correspondence should be addressed. E-mail: b.p.binks@ hull.ac.uk. [email protected]. † University of Hull. ‡ University of Sheffield. § Osaka Institute of Technology. (1) Binks, B. P., Horozov, T. S., Eds. Colloidal Particles at Liquid Interfaces; Cambridge University Press: Cambridge, U.K., 2006. (2) Pugh, R. J. AdV. Colloid Interface Sci. 1996, 64, 67. (3) Sadoc, J. F., Rivier, N., Eds. Foams and Emulsions; NATO ASI Series E; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1999; Vol. 354. (4) (a) Carn, F.; Saadaoui, H.; Masse´, P.; Ravaine, S.; Julian-Lopez, B.; Sanchez, C.; Deleuze, H.; Talham, D. R.; Backov, R. Langmuir 2006, 22, 5469. (b) Shirtcliffe, N. J.; McHale, G.; Newton, M. I.; Perry, C. C. Langmuir 2003, 19, 5626. (5) Ramsden, W. Proc. R. Soc. London 1903, 72, 156. (6) Sun, Y. Q.; Gao, T. Metall. Mater. Trans. A 2002, 33, 3285. (7) Alargova, R. G.; Warhadpande, D. S.; Paunov, V. N.; Velev, O. D. Langmuir 2004, 20, 10371. (8) Binks, B. P.; Horozov, T. S. Angew. Chem., Int. Ed. 2005, 44, 3722. (9) Fujii, S.; Ryan, A. J.; Armes, S. P. J. Am. Chem. Soc. 2006, 128, 7883. (10) Binks, B. P.; Murakami, R. Nat. Mater. 2006, 5, 865.

stabilized by particles of appropriate wettability at the air-water or oil-water interfaces exhibit excellent long-term stability.1 However, the destruction of foams and emulsions (defoaming and demulsification) is often required in practical applications. There are a few examples of in situ demulsification. It has been shown that the stability of emulsions prepared using stimulusresponsive latex particles can be controlled by adjusting, for example, the pH. The contact angle of the particles and consequently the energy required to remove particles from the oil-water interface are modified in this way, leading to desorption. Thus, adjusting the dispersion pH led to complete demulsification of oil-in-water emulsions when using lightly cross-linked nanocomposite microgel particles11 or shell cross-linked micelles12 as pH-responsive particulate emulsifiers. However, stimulus-responsive particle-stabilized foams have not been reported. Herein, we describe the synthesis of polystyrene (PS) latex particles coated with a pH-responsive polymer and the behavior of aqueous foams stabilized by them. Depending on the dispersion pH, the resulting foams can be defoamed in situ. We link this effect to the charge properties of the particles themselves. Experimental Section Materials. Styrene (Sigma-Aldrich) was treated with basic alumina in order to remove inhibitor and was then stored at -25 °C before use. Poly(acrylic acid) (PAA, M h w ) 250 000) was purchased from Fluka Chemicals. 2,2′-Azobis(isobutyronitrile) (AIBN) was used as a free radical initiator. HPLC-grade ethanol was purchased from Fisher Scientific. NaCl, NaOH (AnalaR, BDH), and HCl (analytical grade, Fisher Scientific) were used to adjust the salt concentration and pH of aqueous dispersions. All water used in this study was passed through a Milli-Q reagent water system and had a surface tension of 71.9 mN m-1 at 25 °C. Synthesis of PAA-Stabilized PS Latex Particles. Dispersion polymerization of styrene was performed in the presence of PAA homopolymer as a polymeric stabilizer in batch mode at 70 °C using an AIBN initiator. PAA (1.2 g, 12.0 w/w % based on styrene) was added to an ethanol-water mixture (68.5 g ethanol, 20 g water) in a three-necked 100 mL flask fitted with a reflux condenser and a magnetic stir bar. This reaction mixture was vigorously stirred at (11) Binks, B. P.; Murakami, R.; Armes, S. P.; Fujii, S. Langmuir 2006, 22, 2050. (12) Fujii, S.; Cai, Y.; Weaver, J. V. M.; Armes, S. P. J. Am. Chem. Soc. 2005, 127, 7304.

10.1021/la700444a CCC: $37.00 © 2007 American Chemical Society Published on Web 07/21/2007

8692 Langmuir, Vol. 23, No. 17, 2007 70 °C until the PAA had dissolved completely and was then degassed using a nitrogen purge. Polymerization commenced after the injection of a mixture of styrene and AIBN (0.168 g of AIBN in 10 g of styrene) into the reaction vessel, and the reaction was allowed to proceed for 48 h with continuous stirring at 250 rpm under a nitrogen atmosphere. The conversion of styrene was almost 100%, as determined by gravimetry. Successive centrifugation-redispersion cycles (6000 rpm for 30 min) were used to purify this latex, with each supernatant being decanted and replaced with deionized water. The extent of purification was assessed by measuring the surface tension of the supernatant. Foam Preparation. Sodium chloride was added to a 2.6 wt % aqueous dispersion of PAA-stabilized PS latex particles to obtain a 0.1 M NaCl solution. The natural pH of these dispersions was 4.2; the pH was adjusted as required by adding concentrated aqueous solutions of either HCl or NaOH. The aqueous particle dispersion (1.00 g) was placed in a glass vessel (2 mL) with a screw cap and shaken using an IKA MS1 mini-shaker for 30 s at 2500 rpm. Foams were prepared and stored at room temperature. At least two separate preparations were made for most of the pH values. Characterization of Foams and Aqueous Dispersions. Photographs of vials containing samples were taken with an IXY digital 800 IS camera (Canon). Small samples of the dispersion/foam were placed in a hemeocytometer cell (Weber Scientific) and observed with a Nikon Labophot microscope fitted with a QICAM 12-bit Mono Fast 1394 camera (QImaging). All images were processed with Image-Pro Plus version 5.1 software (Media Cybernetics). Scanning electron microscopy (SEM) images were obtained using a Zeiss EVO 60 instrument and a coating of gold. Zeta potentials were calculated from the measured electrophoretic mobilities determined using a Malvern Nano ZS ZEN3600 instrument at 20 °C. Three samples were used at each pH. The volume-weighted average particle diameter of a 0.03 wt % aqueous dispersion in the presence of 0.1 M NaCl was determined by light diffraction using a Malvern Mastersizer 2000 at different pH values. Measurements were made on two fresh samples.

Letters

Figure 1. (a) SEM image of dried PAA-stabilized PS latex particles. The scale bar is 4 µm. (b-d) Optical micrographs of aqueous dispersions of these particles (2.6 wt %, 0.1 M aqueous NaCl) at (b) pH 5.0, (c) pH 3.5, and (d) pH 2.5. The scale bar is 50 µm in each case.

Results and Discussion The polystyrene latex particles used in this study were prepared by dispersion polymerization in an ethanol-water mixture using poly(acrylic acid) as a steric stabilizer.13 Prior to purification, this milky-white dispersion contained excess PAA, and no foam was observed after vigorous agitation. After one centrifugationredispersion (CR) cycle to remove the nonadsorbed stabilizer and ethanol, some foam was formed that remained stable for at least 24 h. After the second and third CR cycles, highly stable foams were obtained. After another three CR cycles, the surface tension of the supernatant was 63.4 mN m-1, implying that nonadsorbed PAA stabilizer still remained in the aqueous phase. After the fourth CR cycle and ultrafiltration, no foam was formed, and the surface tension of the supernatant was 72.1 mN m-1 at pH around 2, being virtually that of pure water at 25 °C. This confirms that the concentration of non-adsorbed PAA in the aqueous phase is negligible. Figure 1a is an SEM image of the dried PAA-stabilized PS particles, which confirms their nearly monodisperse spherical morphology. The image allows an estimation of the numberaverage particle diameter to be calculated as 0.80 ( 0.09 µm. Optical microscopy images of aqueous dispersions of these particles in the presence of NaCl at different pH values are given in Figure 1b-d. At and above pH 4.2, the particles are discrete. However, flocculated particles were observed at and below pH 3.5, with the extent of flocculation increasing with decreasing pH. Furthermore, the aqueous dispersions at and below pH 3.5 showed signs of sedimentation at the bottom of the vessel.

The zeta potentials and the volume-weighted average diameters of the particles, D4/3, determined by light diffraction are plotted against the pH of the aqueous dispersion in Figure 2. The apparent diameters above pH 5.0 are constant (0.87 ( 0.01 µm) and close to the number-average diameter from SEM analysis, reflecting the fact that the particles within this pH range are well dispersed and discrete. At around pH 4.5, the D4/3 value increases sharply, becoming almost constant (3.78 ( 0.16 µm) at and below pH 3.5. This apparent increase is due to flocculation of the particles at low pH as seen earlier. The measured size relates to an average flocculation dimension. The zeta potential values at high salt concentration (filled circles) are negative at high pH owing to ionized carboxyl groups of the PAA stabilizer on the surfaces of the particles. On lowering the pH, the particles become less negatively charged and then acquire cationic character at low pH, with the isoelectric point being located at pH 3.5. The origin of the cationic nature of the particles at low pH has not been clarified, but this may be due to the preferential adsorption of sodium ions within the PAA surface layer, as reported earlier for micrometer-sized aluminum hydroxide sols.14 This is more likely

(13) Lovell, P. A., El-Aasser, M. S., Eds. Emulsion Polymerization and Emulsion Polymers; Wiley: New York, 1997.

(14) Rowlands, W. N.; O’Brien, R. W.; Hunter, R. J.; Patrick, V. J. Colloid Interface Sci. 1997, 188, 325.

Figure 2. Zeta potential (0.1 wt % particles, b) and volume-weighted average diameter (0.03 wt % particles, 4) of PAA-stabilized PS latex particles in 0.1 M aqueous NaCl versus pH of the aqueous dispersion. Also shown are the zeta potentials of particles in 0.001 M KCl (O).

Letters

Langmuir, Vol. 23, No. 17, 2007 8693

Figure 4. Photographs illustrating the pH-responsive behavior of foam prepared using PAA-stabilized PS latex particles (2.6 wt %, 0.1 M aqueous NaCl). (a) At pH 2.5 after 24 h, (b) immediately after pH adjustment from 2.5 to 5 with gentle inversion, and (c) at pH 5 after 24 h.

Figure 3. pH-dependent behavior of foams prepared in batch mode using PAA-stabilized PS latex particles (2.6 wt %, 0.1 M aqueous NaCl). (a) Photograph of vessels taken 1 week after shaking an aqueous dispersion of particles at different pH values (given). (b) Height of the foam layer versus pH of the aqueous dispersion recorded at different times: immediately after preparation (2), after 1 week (0), and after 1 month (O). (c) Optical microscopy images of foams prepared at pH 2.5 immediately after preparation at two magnifications. Scale bars are 400 µm (left) and 50 µm (right).

than the adsorption of hydrogen ions because the particles do not become cationic at low pH in the absence of salt. By contrast, the zeta potentials of the particles dispersed in background electrolyte (10-3 M KCl, open circles) remain negative at all pH values, and the inflection point suggests a pKa for the poly(acrylic acid) layer around pH 4.5, in agreement with a reliable literature value.15 On shaking aqueous dispersions of PAA-stabilized PS latex particles (2.6 wt %) in background electrolyte in glass vessels, no stable foam was formed at any pH. Reasonably stable foams, however, were obtained below a certain critical pH of the aqueous dispersion in the presence of a higher concentration of salt (0.1 M). This is shown in Figure 3a 1 week after preparation. At high pH (e.g., 6.0, 7.6), no foam can be prepared. At intermediate pH (4.2, 5.0), a small number of bubbles are formed, but these disappeared within 1 month. However, reasonably stable foams are obtained at low pH (at or below 3.5). In these cases, the foam partially drains, and the aqueous phase separates into a supernatant and a sedimented phase, reminiscent of the phase separation of dispersions before aeration. The height of the foam layer, defined as the distance between the foam/dispersion boundary and the three-phase contact line of air, glass, and foam, is plotted against the pH of the aqueous dispersion in Figure 3b, immediately after preparation and for some time afterward. The height gradually (15) Kirby, G. H.; Harris, D. A.; Li, Q.; Lewis, J. A. Key Eng. Mater. 2004, 264-268, 161.

decreases with time because of water drainage and the partial coalescence of air bubbles. Although future experiments would ideally concentrate on the kinetics and precise mechanism of the drainage from such foams, this is not needed in the present context in which the main point arising is the dramatic influence of dispersion pH on foamability. The foam is composed of nearly spherical bubbles ranging in size from several micrometers to millimeters, as shown in the optical microscopy images presented in Figure 3c. It appears that stable foams are formed only under conditions where particles are flocculated in the bulk; see Figure 1. In addition, the cationic surface charge of the particles at low pH may assist their adsorption onto air-water surfaces through electrostatic attraction because pristine air-water surfaces possess a negative charge.16 The dramatic influence of pH on the foamability of these aqueous particle dispersions is similar to the effect of pH on emulsions stabilized by carboxyl-coated latex particles.17 Here, hydrophilic, charged carboxylate particles stabilized oil-in-water emulsions at high pH, whereas uncharged, hydrophobic particles at low pH preferred to stabilize waterin-oil emulsions. The addition of salt not only increases the foamability of particle dispersions8 but also enhances the formation and stability of air bubbles in water stabilized by charged silica particles.18 It was shown that the hydrophobicity of the particle surfaces increased with salt concentration, enabling such particles to be well held at the air-water surfaces. On the basis of our findings for the pH-dependent foams prepared in batch mode discussed above, we investigated the possibility of inducing the collapse of a stable foam by subsequent pH adjustment. In other words, is the foam pH-responsive? Thus, a foam was prepared at pH 2.5 and allowed to stand for 24 h (Figure 4a). Then, a small volume of concentrated aqueous NaOH was added to the vessel along with gentle inversion in order to increase the pH of the dispersion to 5. Relatively large (millimeter) bubbles disappeared, but smaller bubbles still remained after this pH adjustment (Figure 4b). However, after 24 h most of these smaller bubbles subsequently disappeared (Figure 4c). Alternatively, foam prepared at pH 2.5 was completely destroyed immediately after increasing the pH to 5 with vigorous shaking. It is suggested that the initially adsorbed latex particles become negatively charged in situ, encouraging their detachment from the air-water surfaces and ultimately leading to the breakdown of the foam. (16) Karraker, K. A.; Radke, C. J. AdV. Colloid Interface Sci. 2002, 96, 231. (17) Binks, B. P.; Rodrigues, J. A. Angew. Chem., Int. Ed. 2005, 44, 441. (18) Kostakis, T.; Ettelaie, R.; Murray, B. S. Langmuir 2006, 22, 1273.

8694 Langmuir, Vol. 23, No. 17, 2007

Letters

Conclusions

potential applications in food manufacturing, cosmetic formulations, and personal care products.

Cationic, flocculated poly(acrylic acid)-coated latex particles can stabilize aqueous foam at low pH. No foam is possible at high pH when the particles are colloidally stable and negatively charged. Defoaming can be simply achieved by increasing the pH of the system after formation. The encapsulation of air bubbles in water using such stimulus-responsive latex particles may have

Acknowledgment. This work was funded by the EPSRC, U.K. (GR/S69283 and GR/S69276). We thank Mr. A. Sinclair, University of Hull, for the SEM analysis. LA700444A